Anda di halaman 1dari 75

Seminars in Fetal & Neonatal Medicine

Amsterdam

Boston

London

New York

Oxford

Paris

Philadelphia

San Diego

St. Louis

Aims and Scope


Seminars in Fetal & Neonatal Medicine (formerly Seminars in Neonatology) is a bi-monthly journal which publishes topic-based issues, including current Hot Topics on the latest advances in fetal and neonatal medicine. The change in title relates to the growing interest amongst obstetricians, midwives and fetal medicine specialists. The Journal commissions review-based content covering current clinical opinion on the care and treatment of the neonate and draws on the necessary specialist knowledge, including that of the respiratory physician, the infectious disease physician, the surgeon, as well as the paediatrician and obstetrician. Each topic-based issue is edited by an authority in their field and contains 810 articles. Current and forthcoming events can be viewed on the Internet at: http://www.elsevier.com/locate/siny Seminars in Fetal & Neonatal Medicine provides: coverage of major developments in neonatal care; value to practising neonatologists, consultant and trainee paediatricians, obstetricians, midwives and fetal medicine specialists wishing to extend their knowledge in this field; up-to-date information in an attractive and relevant format.

Editorial Board
Editor-in-Chief
Professor M I Levene
University of Leeds School of Medicine D Floor, Clarendon Wing The General Infirmary at Leeds Leeds LS2 9NS, UK

Associate Editors
M Blennow, Huddinge, Sweden L Cornette, Brugge, Belgium D J Field, Leicester, UK I Laing, Edinburgh, UK l, Lund, Sweden a K Mars D Peebles, London, UK S Sinha, Middlesbrough, UK A M Weindling, Liverpool, UK

Advisory Board
F A Chervenak, USA S M Donn, USA N Evans, Australia V Fellman, Sweden N N Finer, USA P C NG, Hong Kong J M Perlman, USA E Saliba, France M Vento, Alicante L de Vries, The Netherlands

Seminars in Fetal & Neonatal Medicine 14 (2009) 331

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Editorial

Prologue: Advances in Bronchopulmonary Dysplasia

It has been 42 years since our rst published report of bronchopulmonary dysplasia (BPD)1; it is still a problem for premature infants. The original goal of using mechanical ventilation to treat premature infants with respiratory distress syndrome and respiratory failure was to decrease the signicant mortality. During the ensuing decades, a decrease in mortality has indeed occurred. Once recognized, it was hoped that a reduction of supplemental oxygen concentrations and ventilatory pressure would eliminate or decrease the incidence of BPD. This has, for the most part, been achieved in the 33 week gestational age infants originally described. Advances in neonatal care and respiratory therapy since 1967 have resulted in the successful ventilation of increasingly more immature infants. As a result the original radiographic picture and pathology have been modied. The understanding of the growth and development of the extremely premature lung and the genetic, intrauterine, biochemical, and infectious factors that inuence the susceptibility and severity of BPD has also advanced. This has led to improvements in prevention strategies and both ventilatory and non-ventilatory treatment of BPD. Unfortunately, it has not eliminated BPD or led to a more precise diagnosis. The chest radiographic changes are much more subtle and are seldom used in the diagnosis. Infants are now delivered while their lungs are in the late canalicular to early saccular stage of development. It has yet to be determined whether there is an oxygen concentration or ventilator pressure below which these treatments are non-injurious. Prevention of premature delivery remains an elusive and persistent problem. The long-term pulmonary function of premature infants with or without BPD surviving beyond adolescence to older adulthood is also unknown. Whereas some of the original premature

infants with BPD displayed persistent pulmonary dysfunction as adolescents, it is not known if these changes regressed, persisted, or exacerbated in older adulthood. These original infants were less premature than todays neonatal intensive care unit population. They also were treated with higher concentrations of supplemental oxygen, and higher ventilator pressures than are currently used, and prior to the widespread use of antenatal corticosteroids and surfactant therapy. The persistence of BPD in more immature premature infants indicates the need for long-term clinical and pulmonary function testing. Physicians who deal with these patients as adults will need to be aware of their possible late pulmonary disability. Knowledge of the evolving pulmonary function of infants born very prematurely might best be obtained if pediatric and adult oriented pulmonologists combine their interest and talents. We should not forget, however, that the continuing morbidity from BPD is the result of the successful application of modern neonatal intensive care to very premature infants to improve their survival.

Reference
1. Northway Jr WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med 1967;276:35768.

William H. Northway Jr. Stanford University, Lucile Packard Childrens Hospital, Palo Alto, California, USA E-mail address: northway@stanford.edu

1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.008

Seminars in Fetal & Neonatal Medicine 14 (2009) 332

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Editorial

Despite the myriad of advances in neonatal intensive care in the more than 40 years since Prof. Northway and colleagues rst coined the term bronchopulmonary dysplasia to describe the aftermath of neonatal mechanical ventilation, the incidence of chronic lung disease has not appreciably changed. Approximately 3040% of infants weighing less than 1500 g at birth sustain chronic lung disease, even with the advent of antenatal corticosteroid treatment, surfactant replacement therapy, and sophisticated techniques for both non-invasive and invasive mechanical ventilation. What has changed, however, is the demographic composition of affected infants, many of whom received only mild or modest respiratory support, suggesting that chronic lung disease may now reect an alteration in lung development and growth. This issue of Seminars in Fetal & Neonatal Medicine examines recent advances in the understanding and management of bronchopulmonary dysplasia. We are most grateful to our distinguished group of contributors, headed by Professor Northway, himself, who kindly wrote the Prologue for this issue. Our rst article was written by Professor Philip, who puts another historical context to chronic lung disease. His interest in the subject dates back to his seminal paper published in 1975. Next, Professor Greenough presents a novel perspective on the potential prenatal factors which might predispose an infant to the development of chronic lung disease. Professor Merritt and colleagues provide an interesting commentary on whether the new BPD is actually different from the old BPD and examine the challenges that are ahead. Dr. Van

Marter subsequently re-examines the epidemiology of chronic lung disease and evaluates its multifactorial etiology. Dr. Laughon and colleagues critically evaluate the evidence for preventive strategies. Our own chapter on ventilation follows, where we summarize both lung protective strategies and ventilatory management of BPD. Professor Wiswell and Dr. Tin examine in detail many of the medical myths that surround BPD and present the evidence to date. Prof. Chiswick has written a thoughtful and provocative piece on the ethics of managing end-stage BPD. Finally, Professor Doyle and colleagues examine the long term outcomes of affected infants. We hope that the readers will realise that BPD is a multi-faceted disorder and not merely the effect of ventilator-induced lung injury. The genetic, cellular, developmental, and nutritional milieu of the fetus and newborn are all contributing factors and deserve to be thoroughly investigated. If this issue is a stimulus for such, it will be an unqualied success. Steven M. Donn* C S Mott Childrens Hospital, Div. of Neonatal-Perinatal Medicine, Dept. of Pediatrics, Ann Arbor, Michigan, USA Corresponding author. E-mail address: smdonnmd@med.umich.edu (S.M. Donn) Sunil K. Sinha Middlesbrough, UK

1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.006

Seminars in Fetal & Neonatal Medicine 14 (2009) 333338

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Chronic lung disease of prematurity: A short history


Alistair G.S. Philip
Stanford University School of Medicine, Division of Neonatal and Developmental Medicine, Suite 315, 750 Welch Road, Palo Alto, CA 94304, USA

s u m m a r y
Keywords: Bronchopulmonary dysplasia Chronic lung disease Genetic predisposition Prematurity WilsonMikity syndrome

Chronic lung disease of prematurity (CLD) is commonly considered to be a consequence of assisted ventilation. However, prior to the description in 1967 of bronchopulmonary dysplasia (BPD), following ventilator therapy for respiratory distress syndrome, WilsonMikity syndrome (WMS) had been described in very preterm infants on minimal oxygen supplementation. In the 1970s and 1980s, many infants treated with assisted ventilation required prolonged mechanical ventilation after developing radiographic features of coarse inltrates, severe hyperination, and microcystic changes, associated with hypercarbemia and the need for increased inspired oxygen concentrations. Some infants died and showed evidence of pulmonary brosis, obstructive bronchiolitis, and dysplastic change. The role of supplemental oxygen, positive pressure ventilation, and the immaturity of the lung have long been considered important in the etiology of CLD/BPD. More recently, the role of inammation (particularly antenatal exposure to cytokines) and individual susceptibility (genetic predisposition) have assumed greater etiologic importance. The historical setting into which corticosteroid treatment for BPD was introduced is also discussed. After the licensing of exogenous surfactant to treat RDS in the early 1990s and more widespread use of prenatal corticosteroids in the mid-1990s, severe BPD became an unusual event. Gradually, the diagnosis of CLD, still often referred to as BPD, was based on an oxygen requirement at 36 weeks postmenstrual age. However, it is not clear that this new BPD is substantially different from WMS. It is difcult to make prognostications about long-term lung function of these infants based on oxygen requirement at 36 weeks, since supplemental oxygen is frequently used unnecessarily. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Prior to the introduction of assisted (mechanical) ventilation, comparatively few very low birth weight (VLBW) infants (<1500 g) survived. Neonatal mortality for infants with birth weights <1000 g (ELBW) exceeded 90%.1 This birth weight is roughly equivalent to a gestational age of 28 weeks. Although infants born at <28 weeks of gestation have immaturity of many organ systems, the primary problem in the 1950s was respiratory immaturity giving rise to respiratory distress syndrome (RDS) or hyaline membrane disease (HMD), as it was originally called. The primary problem producing RDS/HMD is a lack of surfactant. Early attempts in the mid-1960s to rescue infants with RDS using mechanical ventilation were not always successful. This resulted from a variety of factors; ventilators were not designed to be used in neonates. Many advances in the general care of very preterm neonates have contributed to the successful management of RDS today. Improved thermoregulation, better uid and electrolyte management, and parenteral nutrition have all made

a signicant impact. However, the present discussion will focus primarily on the respiratory system. 2. Management prior to use of mechanical ventilation Looking back from todays vantage point, it is difcult to appreciate how little was offered to infants <28 weeks of gestation in the 1950s and early 1960s. Although babies were placed in incubators, body temperatures were often kept quite low, until after 1958, when Silverman et al.,2 showed that increased environmental temperatures could improve survival. Feeding such infants was difcult and because of this they were frequently starved for two to three days. With the introduction of special needles, intravenous feeding became more routine, and Usher reported decreased mortality from RDS using a combination of early intravenous glucose and sodium bicarbonate.3 A chronic pulmonary syndrome was reported by Wilson and Mikity in 1960 in ve very small premature infants who survived.4 Their report was during an era when assisted ventilation was not used in the preterm infant, although mechanical ventilation had been used in some larger infants with tetanus.5 An additional 29 cases of WilsonMikity syndrome (WMS) were reported from the

E-mail address: aphilip@stanford.edu 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.07.013

334

A.G.S. Philip / Seminars in Fetal & Neonatal Medicine 14 (2009) 333338 Table 1 Comparative features of bronchopulmonary dysplasia (BPD) and WilsonMikity syndrome (WMS). Feature Antecedents Respiratory distress syndrome Positive pressure ventilation Increased O2 concentration Extreme prematurity Age at onset (days) Biochemistry Decreased pH Increased pCO2 Chest radiograph ndings Bubbly/cystic lung Hyperination/at diaphragms White out Pathology Exudative change Pulmonary brosis Dysplastic change Obstructive bronchiolitis BPD 47 (early) (early) WMS / 1014 / / /

same medical center in 1969,6 and many other cases were reported from around the world.7 WMS was observed in very preterm infants approximately two weeks after birth. For the most part, these were infants who initially had little or no oxygen requirement, but later developed the need for increasing supplemental oxygen concentrations of up to 40% to prevent cyanosis. Radiographically, microcystic changes, with some degree of hyperination and attening of the diaphragms were seen. Some infants recovered spontaneously over weeks to months, but others died and demonstrated hyper-aeration and reduced alveolar septa at postmortem examination. A few had evidence of pulmonary brosis and obstructive bronchiolitis. 3. Oxygen toxicity in adults, animal models and in vitro The terms pulmonary oxygen toxicity or pulmonary poisoning were used at about this time, both in animals and in humans.8 Pathologic ndings were described in adult humans treated with oxygen and articial (mechanical) ventilation.9 The additive roles of oxygen and assisted ventilation were noted in lambs10 and pulmonary oxygen toxicity was reported in monkeys.11 Somewhat later, ndings similar to those in WMS were noted when lung histology was evaluated in extremely premature baboons exposed to carefully controlled assisted ventilation and oxygen administration.12 Loss of muco-ciliary function was also noted when cultured human neonatal respiratory epithelium was exposed to 80% oxygen for 4896 h in vitro.13 4. First description of bronchopulmonary dysplasia Soon after the introduction of mechanical ventilation to manage RDS in the mid-1960s, reports began to appear of radiographic and pathologic abnormalities that seemed to result from exposure to high concentrations of oxygen using mechanical ventilation. The rst description of bronchopulmonary dysplasia (BPD) is generally attributed to Northway et al. in 1967.14 They used the term bronchopulmonary dysplasia to describe ndings of pulmonary disease following respirator therapy of hyaline membrane disease. However, almost identical ndings were described in the same year by Hawker et al. from London, noting infants who died after respirator treatment for severe hyaline membrane disease.15 Northway et al. believed that the critical factor appeared to be exposure to an inspired oxygen concentration >80% for longer than 150 h.14 Hawker et al. linked the lung ndings to >60% oxygen for more than 5 days (120 h).15 In a review of pulmonary oxygen toxicity in 1969, Nelson stated that there seemed to be general agreement that oxygen partial pressures of less than 0.6 atmospheres are not toxic, over any time span.16 On the other hand, Pusey et al. (also in 1969) suggested that neither HMD nor concentrations of oxygen >80% were needed to produce pulmonary broplasia, the key feature of BPD, and that assisted ventilation was more important.17 A comparison of the principal features of BPD (in the late 1960s) and WMS is provided in Table 1. 5. Negative pressure respirators (ventilators) Although the introduction of mechanical ventilation in the 1960s was predominantly concerned with positive pressure machines, some investigators, including Stahlman and Stern, believed that there might be advantages in using negative pressure devices. They reported that it was unusual to observe pulmonary oxygen toxicity when negative pressure respirators were used, even with prolonged exposure to high oxygen concentrations.18

, rare; /, sometimes; , quite common; , common; , very common.

This apparent advantage of the negative pressure respirator did not gain general acceptance. The negative pressure device available at the time looked a little like a modied incubator, based on the iron lung devices used to manage severe cases of poliomyelitis in older children and adults (Fig. 1). The most difcult part about using the machines was to create an adequate seal at the neck to generate adequate negative pressure around the baby, without disrupting venous return from the brain. It was also difcult to use in infants <1500 g, although in centers where negative pressure ventilation was used frequently, it was claimed that infants as small as 1200 g could be managed effectively. Another problem was that the negative pressure needed to ventilate the lungs could cause the whole baby to be thrown around. Because of this size limitation, preterm infants with severe RDS were increasingly managed with positive pressure ventilation. Consequently, the negative pressure ventilator fell into disuse. 6. Oxygen plus pressure plus time During the early 1970s, I learned to use mechanical ventilation, initially with volume-limited ventilators such as the Engstrom or the Bourns devices, although I developed limited experience with the negative pressure device. Because of the reports of pulmonary

Fig. 1. The negative pressure respirator being used to ventilate a larger neonate, demonstrating the neck seal.

A.G.S. Philip / Seminars in Fetal & Neonatal Medicine 14 (2009) 333338

335

oxygen toxicity, we attempted to minimize exposure of preterm infants to high oxygen concentrations. Nevertheless, we encountered many infants treated with positive pressure who developed radiographic features of BPD. It seemed that the problem could not simply be duration of exposure to high oxygen concentrations. Two other factors seemed to be important: the immaturity of the lung, and the role of the endotracheal tube. Since negative pressure ventilation was rarely associated with chronic lung disease, it seemed possible that the endotracheal tube allowed pressure to be applied more directly to the immature lung (later referred to as barotrauma). In a series of ten infants who developed BPD that I reported in 1975,19 only two required >80% oxygen for more than 24 h, and only one required 60% oxygen for more than 100 h. I suggested that the etiology of BPD might be a complicated interaction between oxygen plus pressure plus time and that immature lungs when exposed to oxygen concentrations >40% for as little as three days, via positive pressure ventilation, might develop BPD. It also seemed possible that WMS might form one end of a spectrum, where the very immature lung might react adversely to even minor increases in oxygen exposure (since they should not have been exposed to any). I also suggested that there might be individual susceptibility.19 Much evidence now supports these hypotheses. 7. National Institutes of Health (NIH) consensus conference In the late 1970s, there was increasing attention paid to the development of lung injury in preterm infants, most of whom had received assisted ventilation. The NIH convened a consensus conference in late 1978. The results were published in 1979.20 Northway was the opening speaker. Some of those in attendance were inclined to use the term chronic lung disease of prematurity (CLD), but the majority favored the term bronchopulmonary dysplasia (BPD). This seemed somewhat unusual to me, since the term RDS (a clinical denition) had been favored over HMD (a pathological denition) at an earlier consensus conference. BPD, like HMD, suggests knowledge of histopathology, whereas CLD expresses what is happening clinically. There was considerable discussion of possible etiologic factors, the clinical and radiographic ndings and diagnostic criteria.20 Key features of BPD/CLD were prolonged oxygen dependency and associated radiographic features of hyperination, usually accompanied by increased pCO2. Not infrequently, in the 1970s and 1980s, prolonged assisted ventilation was required and hospital stays of many months were common. Severe cases of BPD frequently resulted in death from cor pulmonale. 8. Oxygen radical disease Although the etiology of BPD is likely multifactorial, the concept of pulmonary oxygen toxicity received additional support in the 1980s. In a series of papers, Saugstad promoted the concept that oxygen free radicals could result in pulmonary damage.21,22 He had earlier documented the importance of hypoxanthine and xanthine oxidase in asphyxial insult.23 He extended these observations and proposed that several problems frequently encountered in very preterm infants might be the result of a common pathway. He proposed that BPD, retinopathy of prematurity (ROP), necrotizing enterocolitis (NEC), patent ductus arteriosus (PDA), periventricular leukomalacia (PVL), and possibly intraventricular hemorrhage (IVH) might all be part of an entity he called oxygen radical disease of the newborn.22 The initial proposal was that a burst of oxygen free radicals was generated in the re-oxygenation phase following an acute hypoxic insult, which could cause widespread injury. This idea has continued to be supported, rather than refuted, in recent

years. Attempts are now made to minimize the exposure of preterm infants to high oxygen saturations. 9. Chorioamnionitis, cytokines and chemokines Before considering the role that exogenous surfactant has played in decreasing the severity of BPD, other etiologic factors in the production of BPD/CLD emerged in the late 1970s and 1980s. Excessive uid administration was implicated, although this may have been because it resulted in exposure to higher oxygen concentrations.24 Cytomegalovirus (CMV) infection was also implicated in late-onset chronic lung disease25 in an era before CMV screening of blood for transfusion became routine. The possibility that chorioamnionitis might predispose to BPD seemed unlikely when it was rst reported, but the role of inammation gained increasing acceptance as a variety of cytokines and chemokines were reported to be circulating following chorioamnionitis. One of the rst reports came from Japan in 1983.26 Elevated levels of immunoglobulin M (IgM) were noted in low birth weight infants with chronic respiratory insufciency. Subsequently, the same group reported higher IgM levels in infants evaluated within 72 h of birth who went on to develop WMS.27 Somewhat surprisingly, much lower levels were seen in those who developed BPD or unexplained chronic lung disease. Over the next decade, lung inammation was increasingly recognized, and the role of cytokines was reported in evolving BPD.28,29 By the turn of the millennium, inammation was rmly established in the etiology of BPD30,31 and currently seems to be even more entrenched as an antenatal precursor to lung injury and the development of BPD.32,33 10. Status of BPD in the 1980s During the 1970s, neonatologists became more adept at managing mechanical ventilation, and survival of VLBW infants improved. This improvement was partly the result of the learning curve as increasing numbers of people learned to manage VLBW infants successfully. This was, in part, attributed to the development of ventilators specically designed to be used in infants, and partly from other improvements (e.g. parenteral nutrition). The possibility that undernutrition was a contributing factor in the pathogenesis of BPD was proposed.34 In 1980, the rst report of successful treatment of RDS with exogenous surfactant came from Fujiwara et al. in Japan.35 Within a few years, there was an explosion of interest in exogenous surfactants, which were evaluated at the end of the decade in multiple randomized controlled studies around the world, with spectacular success. By 1991, both Exosurf (colfosceril, a synthetic surfactant) and Survanta (beractant, a bovine-extract surfactant) had been approved by the Federal Drug Administration. Although the introduction of exogenous surfactant did not eliminate the development of BPD, it substantially altered its severity, so that by 1999 Jobe referred to chronic lung disease of preterm infants observed at that time as the new BPD.36 During the 1980s, our ability to keep babies alive with successful mechanical ventilation came at considerable expense (Fig. 2). This was an era of chronic ventilator dependence, with some infants being ventilated for several months. Whereas many were eventually able to be weaned, some were pulmonary cripples and others died of cor pulmonale after several months in the neonatal intensive care unit (NICU). We tried to remain optimistic about the future for these preterm infants by reminding ourselves (and parents) that lung growth continues for many months after birth (for as long as three years), and we tried to minimize the duration of exposure to ventilation as much as possible.

336

A.G.S. Philip / Seminars in Fetal & Neonatal Medicine 14 (2009) 333338

Fig. 3. An infant on nasal cannula oxygen supplementation being assessed as an outpatient with a transcutaneous oxygen monitor placed on the back.

Fig. 2. An example of the chest radiograph ndings of bronchopulmonary dysplasia commonly observed in the 1970s and 1980s, demonstrating coarse inltrates, hyperination and microcystic changes. (Courtesy of Steven M. Donn, MD.).

11. Prevention and therapeutic strategies Prior to surfactant therapy, there were few prevention strategies. Antioxidants were promoted, but initial enthusiasm for vitamin E and superoxide dismutase were tempered by negative results.37,38 On the other hand, vitamin A was more convincing39 and seems to have a place in standard care of most VLBW infants. More recently, a large, prospective, randomized study of caffeine to prevent apnea documented a signicant reduction in CLD/BPD.40 In 1972, reduction in the incidence and severity of RDS was reported by Liggins and Howie using antenatal betamethasone.41 Although used by many, widespread acceptance of this strategy did not occur for more than 20 years.42 However, maternal antenatal glucocorticoid administration was shown to reduce the risk of BPD in 1990.43 Since cor pulmonale was frequently observed, diuretics and digoxin were often used, but the most impressive improvement was noted with postnatal corticosteroids (see below). In order to decrease hospital stay, some infants were discharged home on nasal cannula oxygen (Fig. 3). Initially, this was facilitated by transcutaneous oxygen monitoring44 and later by pulse oximetry.45 Another suggestion was to use early continuous positive airway pressure, as the incidence of CLD was much lower in one center adopting this strategy compared with seven that did not.46 This approach has gained increased support in recent years, although it may be difcult to separate it from the effects of early surfactant administration.47

neurodevelopmental delay.48 It seems important to establish the circumstances into which this treatment was introduced. Some infants who developed CLD/BPD required assisted ventilation for months. Against this backdrop, it is a little easier to understand why steroid use had considerable appeal. Steroids certainly seemed to shorten the duration of mechanical ventilation and the link between mechanical ventilation and CLD/BPD appeared to be well-established. My own experience may have contributed to more widespread use of corticosteroids. In 1974, I presented a paper at the Western Society for Pediatric Research documenting pulmonary improvement and rapid weaning from assisted ventilation in ten infants with severe BPD.49 In the rst two patients, steroids had been used for other reasons (cerebral edema in one and laryngeal edema at attempted extubation in the other), with serendipitous pulmonary improvement. Although I prepared a manuscript describing the results, it was not published because reviewers were concerned about possible long-term consequences, based on animal data. This was not something to which I had given much thought. My use of steroids for CLD/BPD was subsequently limited to late treatment in severe cases with prolonged requirements of assisted ventilation. In the ensuing years, there was considerable discussion. We all wrestled with how to minimize the long-term effects of assisted ventilation. The rst randomized controlled study of corticosteroids for BPD was published in 1983 by Mammel et al.50 and a second study appeared in 1985.51 Protocols for both of these studies had comparatively short courses of steroids. The most inuential study appeared in 1989, with earlier administration and a prolonged weaning course.52 After this, corticosteroid use became more popular, until concerns about longterm sequelae were expressed.48 Exogenous surfactant was being evaluated at about the same time that this report appeared.52 Any decrease in severity of CLD/BPD may have been attributed, by some, more to steroids than to surfactant, because there was a tendency to use steroids earlier than before. Even when surfactant was being used more routinely, it was still difcult to wean VLBW infants from assisted ventilation and even short courses of steroids (seven days) seemed to be helpful in facilitating extubation and decreasing the duration of mechanical ventilation and the incidence of CLD.53

12. Introduction of corticosteroids for BPD There has been increasing concern that using corticosteroids to prevent BPD may be creating other problems, particularly

13. Surfactant protein deciency states One group of infants who developed CLD that was difcult to understand was term infants who appeared to have RDS.

A.G.S. Philip / Seminars in Fetal & Neonatal Medicine 14 (2009) 333338

337

Frequently, it was believed that mistakes had been made in determining gestational age. In recent years, surfactant proteins have been categorized, and many of these infants probably had a specic surfactant protein (SP) deciency, usually SP-B or SP-C.54 There is also evidence that there are associations between genetic variants of surfactant proteins and BPD.55 14. Individual susceptibility While debate continued about the relative importance of oxygen or assisted ventilation in the etiology of CLD/BPD,20 at or soon after birth, it was difcult to predict which infants would go on to develop CLD/BPD. Thus, it seemed possible that there might be some factor or factors that contributed to individual susceptibility. One of the rst associations, reported in 1982, was between CLD and HLA (human lymphocyte antigen) phenotypes.56 In the ensuing 25 years, the idea that there may be a genetic predisposition to develop CLD/BPD has become generally accepted.57,58 Further details can be found in Chapter 3 of this issue. 15. The new BPD With the widespread use of surfactant, in conjunction with greater use of antenatal betamethasone, following another NIH consensus development conference in 1994,42 the outcome for most preterm infants was altered substantially. Not only did survival improve for each gestational age, but the severity of residual lung damage also decreased.59 Today, we almost never encounter the severe hyperination of the lungs with attened diaphragms and associated severe hypercarbemia seen in the 1970s and 1980s. The concern now is whether or not an infant with continuing oxygen requirement does or does not have CLD/BPD. In many NICUs, the nursing staff became very reliant on pulse oximetery monitors, introduced in the mid to late 1980s.60 They would frequently adjust inspired oxygen concentration to maintain SaO2 at 98 to 100%, in order to decrease the amount of apnea and reduce the frequency of alarms. However, early supplemental oxygen may have contributed to the development of chronic lung disease because of oxygen free radicals. Recent years have seen changes in delivery room and NICU management to limit exposure to supplemental oxygen.61 There has been a move away from using 100% oxygen as the standard method of delivery room resuscitation, since many infants respond to room air and may even do better. Oxygen blenders have become more common in delivery rooms. The radiographic features of old BPD are now seen infrequently and rarely do babies die of BPD, so that the pathologic features are almost never seen. Jobe coined the term the new BPD in 1999.36 A second National Institute of Child Health and Human Development (NICHD) consensus conference followed.62 The new denition of BPD became a supplemental oxygen need at 36 weeks postmenstrual age. However, such an oxygen requirement may not be real. The NICHD neonatal network centers demonstrated that many babies who required oxygen, according to the nursing staff, were able to maintain SaO2> 90% on room air.63 Of 1598 infants with birth weights 5011249 g, 560 (35%) had clinical BPD (oxygen use at 36 weeks), but only 398 (25%) had physiologic BPD (SaO2< 90% in room air).63 16. Final thoughts The late Joan Hodgman (who collaborated with Mikity in the 1960s) questioned whether or not the new BPD was indeed another name for WMS.7 Even more recently, Lefkowitz and Rosenberg pointed out that it is now unfeasible to directly examine the relationship between the histopathologic disease and its

long-term pulmonary outcome.64 It might be possible if one were willing to perform lung biopsies on these infants (and some were done in the past), but few would recommend this now. As they state, The term bronchopulmonary dysplasia (BPD) is an overused catchall for all aspects of chronic lung disease in the neonatal population.64 They noted that the main reason for labeling an infant as having BPD was to predict long-term pulmonary outcome, but also observed that Shennan et al.65 had suggested that many infants with a functional abnormality as neonates had normal long-term outcomes. They further added . it is important to recognize that from Shennan et al. forward, the clinical characteristic of some degree of supplemental oxygen use at 36 weeks post-menstrual age was considered simply a possible screening test for abnormal long-term pulmonary outcome, not a marker of a particular histopathology.64 Because neonatology has changed substantially in the past 40 years, CLD may be one entity (a clinical description) and BPD another (a histopathologic description). It still seems probable that WMS and BPD are parts of a continuum of CLD of prematurity. Conict of interest statement None declared. Funding sources None. References
1. Behrman RE, Babson GS, Lassel R. Fetal and neonatal mortality in white middle class infants: mortality risks by gestational age and weight. Am J Dis Child 1971;121:4869. 2. Silverman WA, Fertig JW, Berger AP. The inuence of the thermal environment upon the survival of newly born premature infants. Pediatrics 1958;22:87686. 3. Usher R. Reduction of mortality from respiratory distress syndrome of prematurity with early administration of intravenous glucose and sodium bicarbonate. Pediatrics 1963;32:96675. 4. Wilson MJ, Mikity VG. A new form of respiratory disease in premature infants. Am J Dis Child 1960;99:48999. 5. Sykes MK. Intermittent positive pressure respiration in tetanus neonatorum. Anesthesia 1960, Oct;15:40110. 6. Hodgman JE, Mikity VG, Tatter D, Cleland RS. Chronic respiratory distress in the premature infant: WilsonMikity syndrome. Pediatrics 1969;44:17995. 7. Hodgman JE. Relationship between WilsonMikity syndrome and the new bronchopulmonary dysplasia. Pediatrics 2003;112:14145. 8. Giamonna ST, Kerner D, Bondurant S. Effect of oxygen breathing at atmospheric pressure on pulmonary surfactant. J Appl Physiol 1965;20:8558. 9. Nash G, Blennerhassett JB, Pontoppidan H. Pulmonary lesions associated with oxygen therapy and articial ventilation. N Eng J Med 1967;276:36874. 10. de Lemos R, Wolfsdorf J, Nachman R, et al. Lung injury from oxygen in lambs: the role of articial ventilation. Anesthesiology 1969;30:60918. 11. Kaplan HP, Robinson FR, Kapanci Y, Weibel ER. Pathogenesis and reversibility of the pulmonary lesions of oxygen toxicity in monkeys. 1: clinical and light microscopic studies. Lab Invest 1969;20:94100. 12. Coalson JJ, Winter VT, Siler-Khodr T, Yoder BA. Neonatal chronic lung disease in extremely immature baboons. Am J Respir Crit Care Med 1999;160:133346. 13. Boat TF, Kleinerman JI, Fanaroff AA, Matthews LW. Toxic effects of oxygen on cultured human neonatal respiratory epithelium. Pediatr Res 1973;7:60715. 14. Northway Jr WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline membrane disease: bronchopulmonary dysplasia. N Eng J Med 1967;276:35768. 15. Hawker JM, Reynolds EOR, Taghizadeh A. Pulmonary surface tension and pathological changes in infants dying after respirator treatment for severe hyaline membrane disease. Lancet 1967;ii:757. 16. Nelson NM. Chronology, morphology and physiology of pulmonary oxygen toxicity. In: Lucey JF, editor. Problems of neonatal intensive care units. Report of the 59th Ross Conference on Pediatric Research. Columbus, Ohio: Ross Laboratories; 1969. p. 44. 17. Pusey VA, MacPherson RI, Chernick V. Pulmonary broplasia following prolonged articial ventilation of newborn infants. Can Med Assoc J 1969;100:4517. 18. Swyer PR. Symposium on articial ventilation. Summary of conference proceedings. Biol Neonate 1970;16:1915. 19. Philip AGS. Oxygen plus pressure plus time: the etiology of bronchopulmonary dysplasia. Pediatrics 1975;55:4550. 20. Workshop on bronchopulmonary dysplasia. J Pediatr 1979;95(5, part 2): 815920.

338

A.G.S. Philip / Seminars in Fetal & Neonatal Medicine 14 (2009) 333338 43. van Marter LJ, Leviton A, Kuban KCK, et al. Maternal glucocorticoid therapy and reduced risk of bronchopulmonary dysplasia. Peditatrics 1990;86: 3316. 44. Philip AGS, Peabody JL, Lucey JF. Transcutaneous pO2 monitoring in the home management of bronchopulmonary dysplasia. Pediatrics 1978;61:6557. 45. Hudak BB, Allen MC, Hudak ML, Loughlin GM. Home oxygen therapy for chronic lung disease in extremely low birth-weight infants. Am J Dis Child 1989;143:35760. 46. Avery ME, Tooley WH, Keller JE, et al. Is chronic lung disease in low birth weight infants preventable? A survey of eight centers. Pediatrics 1987;79: 2630. 47. Patel D, Greenough A. Does nasal CPAP reduce bronchopulmonary dysplasia (BPD)? Acta Paediatr 2008;97:13147. 48. Stark AR. Risks and benets of post-natal corticosteroids. NeoReviews 2005;6:e99103. 49. Philip AGS. Treatment of bronchopulmonary dysplasia with corticosteroids. Clin Res 1974;22:242A. 50. Mammel MC, Green TP, Johnson DE, Thompson TR. Controlled trial of dexamethasone therapy in infants with bronchopulmonary dysplasia. Lancet 1983;i:13568. 51. Avery GB, Fletcher AB, Kaplan M, Brudno DS. Controlled trial of dexamethasone in respirator-dependent infants with bronchopulmonary dysplasia. Pediatrics 1985;75:10611. 52. Cummings JJ, DEugenio DB, Gross SJ. A controlled trial of dexamethasone in preterm infants at high risk for bronchopulmonary dysplasia. N Engl J Med 1989;320:150510. 53. Durand M, Sardesai S, McEvoy C. Effects of early dexamethasone therapy on pulmonary mechanics and chronic lung disease in very low birth weight infants: a randomized controlled trial. Pediatrics 1995;95:58490. 54. Gower WA, Wert SE, Nogee LM. Inherited surfactant disorders. NeoReviews 2008;9:e45867. 55. Pavlovic J, Papagaroufalis C, Xanthou M, et al. Genetic variants of surfactant proteins A, B, C and D in bronchopulmonary dysplasia. Dis Markers 2006;22:27791. 56. Clark DA, Pincus LG, Oliphant M, Hubbell C, Oates RP, Davey FR. HLA-A2 and chronic lung disease in neonates. J Am Med Assoc 1982;248:18689. 57. Bhandari V, Gruen JR. The genomics of bronchopulmonary dysplasia. NeoReviews 2007;8:e33644. 58. Abman SH, Mourani PM, Sontag M. Bronchopulmonary dysplasia: a genetic disease. Pediatrics 2008;122:6589. 59. Horbar JD, Badger GJ, Carpenter JH, et al. Trends in mortality and morbidity for very low birth weight infants, 19911999. Pediatrics 2002;110:14351. 60. Hay Jr WW, Thilo E, Curlander JB. Pulse oximetry in neonatal medicine. Clin Perinatol 1991;18:44172. 61. Saugstad OD. Room air resuscitation two decades of neonatal research. Early Hum Dev 2005;81:1116. 62. Jobe AH, Bancalari E. NICHD/NIH Workshop summary: bronchopulmonary dysplasia. Am J Respir Crit Care Med 2001;163:17239. 63. Walsh MC, Yao Q, Gettner P, et al. NICHD Neonatal Research Network: impact of a physiologic denition on bronchopulmonary dysplasia rates. Pediatrics 2004;114:130511. 64. Lefkowitz W, Rosenberg SH. Bronchopulmonary dysplasia: pathway from disease to long-term outcome. J Perinatol 2008;28:83740. 65. Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirement in the neonatal period. Pediatrics 1988;82:52732.

21. Saugstad OD. Oxygen radicals and pulmonary damage. Pediatr Pulmonol 1985;1:16775. 22. Saugstad OD. Oxygen toxicity in the neonatal period. Acta Pediatr Scand 1990;79:88192. 23. Saugstad OD, Gluck L. Plasma hypoxanthine levels in newborn infants: a specic indicator of hypoxia. J Perinat Med 1982;10:26672. 24. Brown ER, Stark A, Sosenko I, Lawson EE, Avery ME. Bronchopulmonary dysplasia: possible relationship to pulmonary edema. J Pediatr 1978;92:9824. 25. Ballard RA, Drew WL, Hufnagle KG, Riedel PA. Acquired cytomegalovirus infection in preterm infants. Am J Dis Child 1979;133:4825. 26. Fujimura M, Takeuchi T, Ando M, et al. Elevated immunoglobulin M levels in low birth-weight neonates with chronic respiratory insufciency. Early Hum Dev 1983;9:2733. 27. Fujimura M, Takeuchi T, Kitajima H, Nakajima M. Chorioamnionitis and serum immunoglobulin M in WilsonMikity syndrome. Arch Dis Child 1989;64: 137983. 28. Watterberg KL, Carmichael DF, Gerdes JS, Werner S, Backstrom C, Murphy S. Secretory leukocyte protease inhibitor and lung inammation in developing BPD. J Pediatr 1994;125:2649. 29. Yoon BH, Romero R, Jun JK, et al. Amniotic uid cytokines (interleukin-6, tumor necrosis factor-alpha, interleukin-1-beta and interleukin-8) and the risk for development of bronchopulmonary dysplasia. Am J Obstet Gynecol 1997;177:82530. 30. Lyon A. Chronic lung disease of prematurity: the role of intra-uterine infection. Eur J Pediatr 2000;159:798802. 31. Speer CP. Inammation and bronchopulmonary dysplasia. Semin Neonatol 2003;8:2938. 32. Kramer BW. Antenatal inammation and lung injury: prenatal origin of neonatal disease. J Perinatol 2008;28(Suppl. 1):S217. 33. Bose CL, Dammann CF, Laughon MM. Bronchopulmonary dysplasia and inammatory biomarkers in the premature neonate. Arch Dis Child Fetal Neonatal Ed 2008;93:F45561. 34. Frank L, Sosenko IR. Undernutrition as a major contributing factor in the pathogenesis of bronchopulmonary dysplasia. Am Rev Respir Dis 1988;138: 7259. 35. Fujiwara T, Maeta H, Chida S, Morita T, Watabe Y, Abe T. Articial surfactant therapy in hyaline membrane disease. Lancet 1980;i:559. 36. Jobe AH. The new BPD: an arrest of lung development. Pediatr Res 1999;66:6413. 37. Saldanha RI, Cepeda EE, Poland RL. The effect of vitamin E prophylaxis on the incidence and severity of bronchopulmonary dysplasia. J Pediatr 1982;101: 8993. 38. Thomas W, Speer CP. Non-ventilator strategies for prevention and treatment of bronchopulmonary dysplasiawhat is the evidence? Neonatology 2008;94:1509. 39. Shenai JP, Kennedy KA, Chytil F, Stahlman MT. Clinical trial of vitamin A supplementation in infants susceptible to bronchopulmonary dysplasia. J Pediatr 1987;111:26977. 40. Schmidt B, Roberts R, Miller D, Kirpalani H. Evidence-based neonatal drug therapy for prevention of bronchopulmonary dysplasia in very low birthweight infants. Neonatology 2008;93:2847. 41. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972;50:51525. 42. McCarthy MUS. Recommendations for antenatal corticosteroids. Lancet 1994;343:726.

Seminars in Fetal & Neonatal Medicine 14 (2009) 339344

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Prenatal factors in the development of chronic lung disease


Anne Greenough*
Division of Asthma, Allergy and Lung Biology, Kings College London School of Medicine, London, UK

s u m m a r y
Keywords: Bronchopulmonary dysplasia Cytokines Glucocorticoids Infection Prematurity

Chronic lung disease (CLD), dened as chronic oxygen dependency, is a common outcome of neonatal intensive care. It occurs most frequently in infants born very prematurely, but also in infants born at term who had severe lung disease and those with abnormal antenatal lung growth due particularly to reduction in fetal breathing movements, amniotic uid volume or intrathoracic space. There are, however, other causes and the importance of antenatal infection/inammation regarding impairment of antenatal lung growth is increasingly recognised. Affected infants can suffer chronic respiratory morbidity including an excess of respiratory symptoms and lung function abnormalities even in adulthood. Antenatal interventions directed at improving lung growth are available, but require testing inappropriately designed trials with pulmonary function at follow-up as an outcome. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Chronic lung disease (CLD), dened as chronic oxygen dependency, is a common adverse outcome of neonatal intensive care. It occurs most frequently in infants born very prematurely; more than 40% of one series of infants born before 29 weeks of gestation were affected.1 Such infants are now usually described as having bronchopulmonary dysplasia (BPD). Infants born at term, however, can remain chronically oxygen dependent, particularly if they have had severe lung disease as evidenced by a requirement for extracorporeal membrane oxygenation (ECMO).2 The other major group of infants who can suffer chronic oxygen dependency are those with abnormal antenatal lung growth (Box 1). The aims of this review are to emphasise the importance of CLD by briey describing the associated chronic respiratory morbidity and then to discuss the evidence as to whether certain prenatal factors predispose to the development of CLD and if there are effective antenatal interventions.

BPD. At preschool age, 28% of a BPD cohort coughed and 7% wheezed more than once a week.3 Prematurely born infants, even without BPD, are more at risk of being symptomatic at follow-up than children born at term. In 78-year-olds, 30% of BPD children and 24% of prematurely born children without BPD were wheezing, whereas only 7% of term controls were so affected.4 This adverse respiratory outcome, particularly in those who had BPD, can persist into adulthood. In a follow-up study in The Netherlands, the prevalence of doctor-diagnosed asthma was signicantly higher in 19-year-olds born prior to 32 weeks of gestational age than in agematched controls5; the females not the males had BPD, and were more likely to wheeze without a cold (35% vs 13%) and be short of breath on exercise (43% vs 16%).

2.2. Lung function abnormalities Prematurely born infants, particularly those with wheezing at follow-up, have evidence of airways obstruction (a raised airways resistance and gas trapping) in the rst 2 years after birth.6 Even at school age, particularly in those with ongoing recurrent respiratory symptoms, evidence of poor airway growth persists.7 A strong correlation was demonstrated between the maximum ow at functional residual capacity at 2 years of age and forced expiratory volume in one second at school age, suggesting persistent airow limitation at least in some patients with BPD.7 Adolescents who had BPD have evidence of airways obstruction and hyper-reactivity, with an increased responsiveness to histamine8 and apparently asymptomatic BPD patients have been demonstrated to desaturate on exercise.9 It is important to emphasise, however, that those adolescents received intensive care many years before and that the

2. Chronic respiratory morbidity 2.1. Respiratory symptoms Recurrent respiratory symptoms requiring treatment are common in prematurely born children, particularly those who had

* Newborn Unit, 4th Floor Golden Jubilee Wing, Kings College Hospital, Denmark Hill, London SE5 9RS, UK. Tel.: 44 20 3299 3037; fax: 44 20 3299 8284. E-mail address: anne.greenough@kcl.ac.uk 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.001

340

A. Greenough / Seminars in Fetal & Neonatal Medicine 14 (2009) 339344

long-term pulmonary function of those currently receiving intensive care is not known. Such infants are described as suffering from new BPD and, although they have less inammation and brosis, they have an arrest in acinar development resulting in fewer and larger alveoli and reduction in the number of arteries.10 Whether such patients have appropriate catch-up lung growth requires careful investigation, particularly in infants with BPD, as there is evidence that their small airway function can decline over the rst year.11 There is a spectrum of severity in survivors with abnormal lung growth; some require many months of supplementary oxygen and recurrent respiratory symptoms are common, whereas those mildly affected may have only a raised respiratory rate in infancy.12 Those patients who had surgically repaired congenital diaphragmatic hernia in infancy have lung function abnormalities in adolescence, including reduction in forced expiratory volume, airways obstruction and increased airways hyper-responsiveness.13 In addition, perfusion to the ipsilateral lung is decreased,14 reecting a persistent reduction in the number of branches or generations of pulmonary arteries. Some studies have suggested that these abnormalities are not associated with persisting symptoms, but others have reported reduced exercise tolerance, with only 50% having enough stamina to take part in sport.15

alveolar airspaces.23 Decreased CTGF expression inhibits vascular development. 3.1.2. Antenatal glucocorticoids It has been suggested that antenatal glucocorticoids are the rst hit taken by the fetal lung, which primes the lung for more ventilation-induced injury.24 The impact of antenatal endotoxin and betamethasone on the structure and function of preterm sheep lungs has been compared.25 Both treatments led to thinning of the alveolar walls, but the average alveolar volume increased by about 20% and the total alveolar number decreased by almost 30%. The impact of antenatal betamethasone on fetal inammation has been investigated in a sheep model.26 Ewes were treated at 108110 days of gestation (term being 150 days) with intra-amniotic endotoxin, intramuscular betamethasone, both or saline (control). At ve days, the lambs who had received the combined intervention had increased alveolar neutrophils and proinammatory cytokine mRNA expression, hence it was hypothesised that glucocorticoids impair the ability of the preterm lung to downregulate endotoxininduced inammation.26 3.2. Reduction in amniotic uid volume 3.2.1. Renal abnormalities Reduction in amniotic uid is associated with bladder outlet obstruction, bilateral renal dysplasia/hypoplasia and multicystic kidneys.27 Affected infants may be further predisposed to pulmonary hypoplasia by reduced renal proline production or thoracic compression. 3.2.1.1. Antenatal interventions. Antenatal interventions to try to prevent pulmonary hypoplasia in infants with renal anomalies include relieving the oligohydramnios by amnioinfusion or by bypassing the urinary tract obstruction. A review of 169 cases of vesico-amniotic shunting (placing a pigtail shunt between the fetal bladder and amniotic cavity under ultrasound guidance) demonstrated only 47% perinatal survival and 45% shunt-related complications.28 Oligohydramnios present before shunt replacement (56% mortality) and failure to restore amniotic uid volume (100% mortality) were signs of a poor prognosis.28 An alternative approach has been to undertake laser therapy during fetal cystoscopy to disrupt posterior urethral valves, but this may result in damage to the nearby bowel. The impact of neither intervention on pulmonary development has been determined. 3.2.2. Oligohydramnios The timing of onset of oligohydramnios in pregnancies complicated by premature, prelabour rupture of the membranes (PPROM) is critical; pulmonary hypoplasia only occurs if membrane rupture is prior to 26 weeks of gestation.29 Abnormal lung development, however, is not an invariable consequence of early-onset oligohydramnios; 23% of one cohort with membrane rupture prior to 20 weeks of gestation had no signs of pulmonary hypoplasia.30 3.2.2.1. Antenatal interventions. Serial amnio-infusion has been studied in pregnancies complicated by oligohydramnios resulting from PPROM. Unfortunately, the infused uid was retained only in 30% of one series of patients and in the remaining 70% oligohydramnios recurred within at least 48 h of the procedure.31 3.2.3. Invasive antenatal diagnostic procedures Several groups have reported an excess of lung function abnormalities in the neonatal period and early infancy following rst32 or second33 trimester amniocentesis. In addition, in a randomised trial the occurrence of neonatal RDS and pneumonia was

3. Prenatal factors predisposing to CLD 3.1. Immaturity In the past, prematurely born infants who developed CLD frequently had had severe respiratory failure requiring high pressure mechanical ventilation with high concentrations of supplementary oxygen. Nowadays, CLD can occur in very prematurely born infants who initially had minimal or even no signs of lung disease,16 so-called new BPD. It has been suggested that in such infants abnormal vascular development may lead to the abnormalities of lung growth. It has been proposed that the new BPD may be a maldevelopment sequence resulting from interference/ interruption of normal development signalling for terminal maturation and alveolarisation of the lungs of very preterm infants.17 3.1.1. Antenatal infection/inammation Certain evidence suggests that CLD may be more common if there has been antenatal infection/inammation. Chorioamnionitis reduces the incidence of respiratory distress syndrome (RDS), but has been shown to be associated with an increase in CLD.18 In another series,19 however, histological chorioamnionitis was only associated with an increased risk of BPD if the infant subsequently developed postnatal infection or required mechanical ventilation for more than 7 days. It was therefore suggested that antenatal infection and/or inammation is protective, unless there is postnatal sepsis or prolonged ventilation, and a multi-hit model for BPD development was proposed.19 A strong joint effect of prematurity and chorioamnionitis was demonstrated on early childhood wheezing in the Boston Birth cohort (n 1096), who were followed from birth to a mean age of 2.2 years.20 Results from animal models demonstrate mechanisms by which antenatal infection/inammation may result in BPD. For example, in preterm fetal lambs, a single dose of intra-amniotic endotoxin given before preterm delivery at 125 days of gestation resulted in an increased expression of mRNA for transforming growth factor (TGF)-b1 and a reduction in the expression of connective tissue growth factor (CTGF).21 TGF-b1 is a regulator of lung development, angiogenesis and airway remodelling22; transfer of TGF-b1 genes by an adenovirus vector to newborn rat pups resulted in enlarged

A. Greenough / Seminars in Fetal & Neonatal Medicine 14 (2009) 339344

341

doubled in infants whose mother had undergone amniocentesis at a mean gestation of 16 weeks.34 Infants whose mothers who had undergone rst trimester amniocentesis had more neonatal unit admissions35 and were more likely to be symptomatic at followup36 compared not only with controls, but also with infants whose mothers had had rst trimester chorion villous sampling. Those data36 suggest that removal of even a small amount of amniotic uid at a critical stage during pregnancy can adversely affect lung growth. 3.3. Reduction in fetal breathing movements Selective destruction of the upper cervical cord between the lower medulla and the level of the phrenic nucleus results in cessation of fetal breathing movements and arrested lung growth and development.37 Cessation or reduction of fetal breathing movements may be responsible for the abnormal lung growth seen in such conditions as WerdigHoffman Disease and myotonic dystrophy. It has been suggested that the persistent absence of fetal breathing movements in pregnancies complicated by oligohydramnios due to premature and prolonged rupture of the membranes is a poor prognostic sign.38 Infants with anterior abdominal wall defects (AWD) can have abnormal lung growth. Stillbirths with exomphalos have been demonstrated to have small chests and infants with giant exomphalos reduced chest wall widths and lung areas on chest radiography. At follow-up, infants with either gastroschisis or exomphalos had signicantly lower lung volumes than controls and ve of the 13 had lung volumes below the normal range.39 Fetuses with AWD have a reduction in viscera in the upper part of the abdominal cavity and hence an inadequate framework for chest wall development. The low intraabdominal pressure experienced by such patients could result in impaired diaphragmatic development. Indeed, in the neonatal period, the trans-diaphragmatic pressure generated by magnetic stimulation of the phrenic nerves has been demonstrated to be signicantly lower in infants with gastroschisis than controls.40 Those data40 suggest that the abnormal lung growth seen in certain AWD infants may be a reduction in fetal breathing activity. 3.4. Reduction in intrathoracic space This can occur because of a small chest, particularly asphyxiating thoracic dystrophy, malformations of the lung [e.g. cystic adenomatoid malformation (CCAM) or lung cysts], pleural effusions and congenital diaphragmatic hernia (CDH). Fetal pleural effusions which spontaneously resolve have a good prognosis, but they can progress to non-immune hydrops because of impaired venous return and congestive cardiac failure due to compression. Fatal pulmonary hypoplasia also occurs in fetuses with rhesus isoimmunisation and results from chronic compression of the lungs by fetal ascites and pleural effusions. There may also be a direct immune-mediated injury affecting lung growth.41 Follow-up infants who had rhesus isoimmunisation demonstrated that those who had lower lung volumes had the rst fetal blood sampling and intrauterine transfusion at a signicantly earlier gestation.42 3.4.1. Antenatal interventions 3.4.1.1. Thoraco-amniotic shunting. Thoraco-amniotic shunting was initially used to decompress a large cyst in a fetus with a CCAM; subsequently further affected fetuses have been so treated, with 70% survival in fetuses with macrocystic CCAM in one series.43 This intervention, however, is inappropriate for fetuses with solid or multicystic CCAMs. Pleural effusions can also be treated by thoracoamniotic shunting with a pigtail catheter being placed under local anaesthetic into the effusion, which is then drained into the

amniotic cavity. This can achieve reversal of the associated hydrops and chronic drainage with fetal lung expansion facilitating resuscitation at birth. Placement of thoraco-amniotic shunts was associated with 57% survival in fetuses with hydrops in one series; all 31 survivors of 54 treated fetuses had chylothorax.44 Follow-up lung function studies45 demonstrated that the majority of infants who had had thoraco-amniotic shunting had lung volumes within the normal range, but the procedure was usually performed in the third trimester and hence probably too late to inuence lung growth. The UK National Institute for Health and Clinical Excellence (NICE)46 supports insertion of pleura-amniotic shunts with appropriate patient selection and counselling. An alternative strategy to manage fetal pleural effusions is to cause pleurodesis by an intrapleural injection of OK-432. OK-432 is derived from a low virulence Su strain of type 3 Group A Streptococcus pyrogens. There are case reports of successful treatment.47 Rusticos review of the literature suggests that in-utero intervention (repeated thoracocentesis, intrauterine shunting or pleurodesis) improves the chances of survival in fetuses with persistent effusions48; nevertheless, randomised trials are required to determine if long-term respiratory outcome is improved. 3.4.1.2. In-utero surgery. Open resection via maternal hysterotomy has been associated with 5060% survival rates for fetuses with CCAM and associated hydrops.49 In animal models of CDH, pulmonary hypoplasia was reversed following repair of surgically created diaphragmatic defects; direct repair was performed by maternal hysterotomy and subsequent fetal thoracotomy.50 In a prospective trial in infants, however, no benet over standard postnatal therapy was demonstrated51; fetal surgery was particularly inappropriate if there was liver herniation, as liver reduction resulted in kinking of the umbilical vein compromising venous return.51 Clinical observations of distended, hyperplastic lungs in cases of congenital high airway obstruction syndrome (CHAOS) prompted investigation of temporary tracheal occlusion in animal models as a potential method of preventing lung hypoplasia. Prevention of lung uid egress causes lung tissue stretch, promoting airway and pulmonary vessel growth.52 Initial attempts at tracheal occlusion were via hysterotomy, but with poor survival rates; the limited data available demonstrated that, although open fetal tracheal occlusion was associated with increased lung growth as evidenced by an increase in lung weight, there was no improvement in the lung parenchymal lung structure or reduction in muscularisation of the pulmonary arteries.53 Fetoscopic intervention was then used which involved fetal neck dissection and temporary occlusive titanium clips; unfortunately this was associated with complications including vocal cord paresis.54 As a consequence, the technique has now been modied such that the endoscopic placement of a balloon is used to temporarily obstruct the trachea (FETO). The balloon is retrieved by fetal tracheoscopy at 34 weeks or is punctured during ultrasound guidance. Ultrasound examinations after FETO have demonstrated an increase in the echogenicity of the lungs within 48 h.55 Survival to discharge was 50% following FETO in a multicentre study compared to 8% in eligible contemporary controls,56 but the cases were not randomised. Lung function of infants entered into a randomised trial of tracheal occlusion by an external clip or a balloon demonstrated only modest improvements in neonatal pulmonary function, but only 20 infants were studied.57 Complications of FETO include PPROM. The efcacy of FETO needs to be studied in a randomised trial including long term pulmonary outcomes. 3.4.2. Prediction of pulmonary hypoplasia Antenatal lung:head circumference ratio (LHR) has been used as a prognostic indicator for CDH outcome. In a retrospective

342

A. Greenough / Seminars in Fetal & Neonatal Medicine 14 (2009) 339344

alveolar surface area and altered airway morphology and function.65 Those effects could be explained by the suppressant effect of nicotine on fetal breathing. There have been at least 22 reports of the effects of antenatal smoking on infant and childhood respiratory function. The results are not consistent, but most of the larger studies demonstrate an adverse effect on airway development (Table 1). There is no evidence that antenatal smoking increases bronchial hyper-reactivity and, despite the data from animal models, there is no effect on lung volume other than that expected from the reduction in overall somatic growth.66 3.6.2. Sudden infant death syndrome (SIDS) Infants whose mothers smoke during pregnancy are at a two- to four-fold increased risk of SIDS compared with infants of nonsmoking mothers. A possible explanation for the association is that the infants have neurodevelopmental abnormalities of the control of ventilation; prenatal nicotine exposure results in cell death in the brainstem of animal models and nicotine is a powerful vasoconstrictor and a neuroteratogen. Term newborns of smoking mothers compared with those of non-smoking mothers have been demonstrated to have a dampened respiratory response to added dead space when examined in the maternity unit prior to discharge.67 Those results67 are compatible with dampened chemoreceptor function in infants exposed in utero to cigarette smoking.

Fig. 1. Image generated by three-dimensional ultrasound using virtual organ computer aided analysis. Left: sagittal view. Right: constructed by taking six views at 30 to each other.

multicentre review, LHR measurements and position of the liver were obtained in 134 cases of left-sided CDH between 24 and 28 weeks of gestation.58 When the LHR was <0.6, there were no survivors irrespective of the position of the liver. The presence of liver herniation and a low LHR (<1) also predicted a poor prognosis.58 However, a systematic review59 of the literature concluded that the prognostic use of LHR in fetal CDH lacks a sufcient evidence base and that there has been overlapping and methodologically heterogeneous reporting. Alternative methods of assessment include three-dimensional ultrasound, particularly using virtual organ computer-aided analysis (VOCAL) (Fig. 1), for which reference ranges have been established,60 or magnetic resonance imaging.61 3.5. Malnutrition In-utero growth restriction and being born small for gestational age (SGA) is a risk factor for respiratory morbidity in childhood; both wheezing and respiratory infections are reported more commonly in children who were SGA. Lung function abnormalities also occur more frequently. Prematurely born SGA infants had signicantly higher airway resistance in infancy than AGA controls62 and a signicant correlation between birth weight (adjusted for gestational age) and lung function has been reported in children aged 511 years.63 In addition, diminished airway function was highlighted in adults born of lower birth weight.64 3.6. Maternal smoking 3.6.1. Effect on lung function Antenatal nicotine exposure results in pulmonary hypoplasia in a variety of animal models. For example, exposure of the fetal monkey to nicotine infusion from day 26 to day 134 of pregnancy resulted in the offspring having enlarged air spaces, reduction in

Box 1. Causes of abnormal antenatal lung growth  Reduction in amniotic uid volume Fetal renal anomalies, e.g. Potters syndrome Prolonged rupture of the membranes Uteroplacental insufciency Amniocentesis  Reduction in fetal breathing movements Congenital myotonic dystrophy Spinal muscular atrophy Phrenic nerve agenesis Cervical cord lesions Anterior abdominal wall defects  Reduction in intrathoracic space Pleural effusions Congenital diaphragmatic hernia Congenital cystic adenomatoid malformation Congenital lung cysts Small chest syndromes (de Jeune syndrome)  Genetic Trisomies 21 and 18  Malnutrition Vitamin A deciency Uteroplacental insufciency  Maternal smoking

Practice points  Abnormal antenatal lung growth is common and there is a spectrum of severity. Clinicians should have a high index of suspicion and consider such infants for measures to prevent further respiratory morbidity, e.g. advice regarding parental smoking.  Women who are at high risk of very premature delivery should be investigated, treated for infection and appropriately treated, if possible before antenatal steroids are given.

Table 1 Antenatal smoking exposure and lung function in infants (modied from Milner et al.66) Positive Lung volume (FRC) Compliance Resistance Maximum expiratory ow at FRC FRC, functional residual capacity. Data are expressed as the number of studies. 1 2 1 6 Negative 7 3 4 4

A. Greenough / Seminars in Fetal & Neonatal Medicine 14 (2009) 339344

343

Research directions  The long-term respiratory outcome of children who have had new BPD needs careful examination; in particular, do such children have catch-up growth?  All antenatal interventions, both invasive diagnostic techniques and therapies, should be assessed in randomised trials with infant pulmonary function at followup as an outcome.

Acknowledgements I thank Mrs D. Gibbons for secretarial assistance. Conict of interest statement None declared. Funding sources None. References
1. Johnson AH, Peacock JL, Greenough A, et al. United Kingdom Oscillation Study Group. High frequency oscillatory ventilation for the prevention of chronic lung disease of prematurity. N Engl J Med 2002;347:63342. 2. Kornhauser MS, Cullen JA, Baumgart S, McKee LJ, Gross GW, Spitzer AR. Risk factors for bronchopulmonary dysplasia after extracorporeal membrane oxygenation. Arch Pediatr Adolesc Med 1994;148:8205. 3. Greenough A, Alexander J, Burgess S, et al. Health care utilisation of prematurely born, preschool children related to hospitalisation for RSV infection. Arch Dis Child 2004;89:6738. 4. Gross SJ, Iannuzzi DM, Kveselis DA, Anbar RD. Effect of preterm birth on pulmonary function at school age: a prospective controlled study. J Pediatr 1998;133:18892. 5. Vrijlandt EJ, Gerritsen J, Marike Boezen H, Duiverman EJ, the Dutch POPS-19 Collaborative Study Group. Gender differences in respiratory symptoms in 19 year old adults born preterm. Respir Res 2005;6:117. 6. Broughton S, Thomas MR, Marston L, et al. Very prematurely born infants wheezing at follow up lung function and risk factors. Arch Dis Child 2007;92:77680. 7. Filippone M, Sartor M, Zachello F, Baraldi E. Flow limitation in infants with bronchopulmonary dysplasia and respiratory function at school age. Lancet 2003;361:7534. 8. Koumbourlis AC, Motoyama EK, Mutich RL, Mallory GB, Walczak SA, Fertal K. Longitudinal follow up of lung function from childhood to adolescence in prematurely born patients with neonatal chronic lung disease. Pediatr Pulmonol 1996;21:2834. 9. Santuz P, Baraldi E, Zaramella P, Filippone M, Zacchello F. Factors limiting exercise performance in long term survivors of bronchopulmonary dysplasia. Am J Respir Crit Care Med 1995;152:12849. 10. Husain AN, Siddiqui NH, Stocker JT. Pathology of arrested acinar development in postsurfactant bronchopulmonary dysplasia. Hum Pathol 1998;29: 7107. 11. Hofhuis W, Huysman MW, van der Wiel EC, et al. Worsening of Vmax/FRC in infants with chronic lung disease in the rst year of life: a more favourable outcome after high frequency oscillation ventilation. Am J Respir Crit Care Med 2002;166:153943. 12. Aiton NR, Fox GF, Hannam S, Stern CM, Milner AD. Pulmonary hypoplasia presenting as persistent tachypnoea in the rst few months of life. Br Med J 1996;312:114950. 13. Ijsselstijn H, Tibboel D, Hop WJ, Molenaar JC, de Jongste JC. Long term pulmonary sequelae in children with congenital diaphragmatic hernia. Am J Respir Crit Care Med 1997;155:17480. 14. Delepoulle F, Martinot A, Leclerc F, et al. Long term outcome of congenital diaphragmatic hernia. A study of 17 patients. Arch Fr Pediatr 1991;48: 7037. 15. Zaccara A, Turchetta A, Calzolari A, et al. Maximal oxygen consumption and stress performance in children operated on for congenital diaphragmatic hernia. J Pediatr Surg 1996;31:10925. 16. Rojas MA, Gonzalez A, Bancalari E, et al. Changing trends in the epidemiology and pathogenesis of chronic lung disease. J Pediatr 1995;126:60510. 17. Jobe AH. The new BPD: an arrest of lung development. Pediatr Res 1999;46: 6413. 18. Watterberg KL, Demers LM, Scott SM, et al. Chorioamnionitis and early lung inammation in infants in whom bronchopulmonary dysplasia develops. Pediatrics 1996;97:2105.

19. Van Marter LJ, Ammann O, Allred EN, et al. Chorioamnionitis, mechanical ventilation and postnatal sepsis as modulators of chronic lung disease in preterm infants. J Pediatr 2002;140:1716. 20. Kumar R, Yu Y, Story RE, et al. Prematurity, chorioamnionitis, and the development of recurrent wheezing: a prospective birth cohort study. J Allergy Clin Immunol 2008;121:87884. 21. Kunzmann S, Speer C, Jobe AH, et al. Antenatal inammation induced TGF-b 1 but suppressed CTGF in preterm lungs. Am J Physiol Lung Cell Mol Physiol 2007;292:L22331. 22. Bartram U, Speer CP. The role of transforming growth factor-b in lung development and disease. Chest 2004;125:75465. 23. Gauldie J, Galt T, Bonniaud P, et al. Transfer of the active form of transforming growth factor-b 1 gene to newborn rat lung induces changes consistent with bronchopulmonary dysplasia. Am J Pathol 2003;163:257584. 24. Jobe AH. Antenatal factors and the development of bronchopulmonary dysplasia. Semin Neonatol 2003;8:917. 25. Willet KE, Jobe AH, Ikegami M, Newnham J, Brennan S, Sly PD. Antenatal endotoxin and glucocorticoid effects on lung morphometry in preterm lambs. Pediatr Res 2000;48:7828. 26. Kallapur SG, Kramer BW, Moss TJ, et al. Maternal glucocorticoids increase endotoxin-induced lung inammation in preterm lambs. Am J Physiol Lung Cell Mol Physiol 2003;284:L63342. 27. May C, Greenough A. Pulmonary hypoplasia and congenital renal abnormalities. Arch Med Sci 2006;2:14. 28. Quintero RA, Johnson MP, Romero R, et al. In utero percutaneous cystoscopy in the mangement of fetal lower obstructive uropathy. Lancet 1995;346:53740. 29. Nimrod C, Nicholson S, Davies D, Harder J, Dodd G, Sauve R. Pulmonary hypoplasia testing in clinical obstetrics. Am J Obstet Gynecol 1988;158:27780. 30. Blott M, Greenough A. Neonatal outcome after prolonged rupture of the membranes starting in the second trimester. Arch Dis Child 1988;63: 114650. 31. Locatelli A, Vergani P, Di Pirro G, Doria V, Bif A, Ghidini A. Role of amnioinfusion in the management of premature rupture of the membranes at <26 weeks gestation. Am J Obstet Gynecol 2000;183:87882. ksel B, Greenough A, Naik S, Nicolaides KH. Perinatal lung function and 32. Yu invasive antenatal procedures. Thorax 1997;52:1814. 33. Milner AD, Hosykyns EW, Hopkin IE. The effects of mid-trimester amniocentesis on lung function in the neonatal period. Eur J Pediatr 1992;151:45860. 34. Tabor A, Philip J, Madsen M, Bang J, Obel EB, Nrgaard-Pedersen B. Randomised controlled trial of genetic amniocentesis in 4606 low risk women. Lancet 1986;1:128793. 35. Greenough A, Yuksel B, Naik S, Nicolaides KH. Invasive antenatal procedures and requirement for NICU admission. Eur J Pediatr 1997;156:5502. 36. Greenough A, Yuksel B, Naik S, Cheeseman P, Nicolaides KH. First trimester invasive procedures: effects on symptom status and lung volume in very young children. Pediatr Pulmonol 1997;24:41522. 37. Wigglesworth JS, Winston RM, Bartlett K. Inuence of the central nervous system on fetal lung development. Experimental study. Arch Dis Child 1977;52:9657. 38. Blott M, Greenough A, Nicolaides KH. Fetal breathing movements in pregnancies complicated by premature membrane rupture in the second trimester. Early Hum Dev 1990;21:418. 39. Thompson PJ, Greenough A, Dykes E, Nicolaides KH. Impaired respiratory function in infants with anterior abdominal wall defects. J Pediatr Surg 1993;28:6646. 40. Dimitriou G, Greenough A, Kavvadia V, et al. Diaphragmatic function in infants with surgically corrected anomalies. Pediatr Res 2003;54:5028. 41. Chamberlain D, Hislop A, Hey E, Reid L. Pulmonary hypoplasia in babies with severe rhesus isoimmunisation: a quantitative study. J Pathol 1977;122:4352. 42. Greenough A, Yuksel B, Nicolaides KH. Abnormalities of lung volume at follow up after antenatal rhesus iso-immunisation. Acta Paediatr 1994;83:498500. 43. Douglas Wilson R, Baxter J, Johnson MP, et al. Thoracoamniotic shunts: fetal treatment of pleural effusions and congenital cystic adenomatoid malformations. Fetal Diagn Ther 2004;19:41320. 44. Picone O, Benachi A, Mandelbrot L, Ruano R, Dumez Y, Dommergues M. Thoracoamniotic shunting for fetal pleural effusions with hydrops. Am J Obstet Gynecol 2004;191:204750. 45. Thompson PJ, Greenough A, Nicolaides KH. Respiratory function in infancy following pleuro-amniotic shunting. Fetal Diagn Ther 1993;8:7983. 46. NICE guidelines: insertion of pleuroamniotic shunt for fetal pleural effusion. Available from: http://guideance, nice.org.uk/ipg190/guidance/pdf/English; 2006 [accessed September 2006]. 47. Okawa T, Takano Y, Fujimori K, Yanagida K, Sato A. A new fetal therapy for chylothorax: pleurodesis with OK-432. Ultrasound Obstet Gynecol 2001;18: 3767. 48. Rustico MA, Lanna M, Coviello D, Smoleniec J, Nicolini U. Fetal pleural effusion. Prenat Diagn 2007;27:7939. 49. Adzick NS, Harrison MR, Crombleholme TM, Flake AW, Howell LJ. Fetal lung lesions: management and outcome. Am J Obstet Gynecol 1998;179:8849. 50. Marechal M, Gillerot Y, Chef R. Lhypoplasia pulmonaire. A propos dune observation chez des jumeaux. J Gynecol Obstet Biol Reprod 1984;13: 897902. 51. Harrison MR, Langer JC, Adzick NS, et al. Correction of congenital diaphragmatic hernia in utero, V: initial clinical experience. J Pediatr Surg 1990;25:4755.

344

A. Greenough / Seminars in Fetal & Neonatal Medicine 14 (2009) 339344 A systematic review and meta-analysis. Ultrasound Obstet Gynecol 2007;30: 897906. Jani J, Peralta CF, Van Schoubroeck D, Deprest J, Nicolaides KH. Relationship between lung to head ratio and lung volume in normal fetuses with diaphragmatic hernia. Ultrasound Obstet Gynecol 2006;27:54550. Cannie M, Jani J, Meersschaert J, et al. Prenatal prediction of survival in isolated diaphragmatic hernia using observed to expected total fetal lung volume determined by magnetic resonance imaging based on either gestational age or fetal body volume. Ultrasound Obstet Gynecol 2008;32:6339. Greenough A, Yuksel B, Cheeseman P. Effect of in utero growth retardation on lung function at follow up of prematurely born infants. Eur Respir J 2004;24:7313. Rona R. Effect of prematurity and intrauterine growth on respiratory health and lung function in childhood. Br Med J 1993;306:81720. Barker DJP. Relation of birthweight and childhood respiratory infection to adult lung function and death from chronic obstructive airway disease. Br Med J 1991;303:6715. Sekhon HS, Keller JA, Benowitz NL, Spindel ER. Prenatal nicotine exposure alters pulmonary function in newborn rhesus monkeys. Am J Respir Crit Care Med 2001;164:98994. Milner AD, Rao H, Greenough A. The effects of antenatal smoking on lung function and respiratory symptoms in infants and children. Early Hum Dev 2007;83:70711. Bhat RY, Broughton S, Khetriwal B, et al. Dampened ventilatory response to added deadspace in newborns of smoking mothers. Arch Dis Child Fetal Neonatal Ed 2005;90:F3169.

52. Bratu I, Flageole H, Laberge JM, Chen MF, Piedboeuf B. Pulmonary structural maturation and pulmonary artery remodeling after reversible fetal ovine tracheal occlusion in diaphragmatic hernia. J Pediatr Surg 2001;36:73944. 53. Danzer E, Davey MG, Kreiger PA, et al. Fetal tracheal occlusion for severe congenital diaphragmatic hernia in humans: a morphometric study of lung parenchyma and muscularization of pulmonary arterioles. J Pediatr Surg 2008;43:176775. 54. Harrison MR, Keller RL, Hawgood SB, et al. A randomized trial of fetal endoscopic tracheal occlusion for severe fetal congenital diaphragmatic hernia. N Engl J Med 2003;349:191624. 55. Deprest J, Gratacos E, Nicolaides KH. Fetoscopic tracheal occlusion (FETO) for severe congenital diaphragmatic hernia: evolution of a technique and preliminary results. Ultrasound Obstet Gynecol 2004;24:1216. 56. Deprest J, Jani J, Gratacos E, et al. Fetal intervention for congenital diaphragmatic hernia: the European experience. Semin Perinatol 2005;29:94103. 57. Keller RL, Hawgood S, Neuhaus JM, et al. Infant pulmonary function in a randomized trial of fetal tracheal occlusion for severe congenital diaphragmatic hernia. Pediatr Res 2004;56:81825. 58. Laudy JA, Van Gucht M, Van Dooren MF, Wladimiroff JW, Tibboel D. Congenital diaphragmatic hernia: an evaluation of the prognostic value of the lung to head ratio and other prenatal parameters. Prenat Diagn 2003;23:6349. 59. Baath ME, Jesudason EC, Losty PD. How useful is the lung to head ratio in predicting outcome in the fetus with congenital diaphragmatic hernia?

60.

61.

62. 63. 64. 65.

66. 67.

Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

The new bronchopulmonary dysplasia: challenges and commentary


T. Allen Merritt a, *, Douglas D. Deming a, Bruce R. Boynton b
a b

Loma Linda University, School of Medicine, Department of Pediatrics, Loma Linda, California, USA Bureau of Medicine and Surgery, United States Navy, Washington, DC, USA

s u m m a r y
Keywords: Bronchopulmonary dysplasia Chorioamniotitis Fetal inammatory response Respiratory distress syndrome Surfactant deciency

Lung development is orchestrated by highly integrated morphogenic programs of interrelated patterns of gene and protein expression. Injury to the developing lung in the canalicular and saccular phases of lung development alters subsequent alveolar and vascular development resulting in simplied alveolar structures, dysmorphic capillary conguration, variable interstitial cellularity and broproliferation that are characteristic of the new bronchopulmonary dysplasia (BPD). Fetal and neonatal infection, abnormal stretch of the developing airways and alveoli, altered expression of surfactant proteins (or genetically altered proteins), polymorphisms of genes encoding for vascular endothelial growth factors, and reactive oxygen species result in imparied gas exchange in the developing lung. However, the new BPD represents only one form of neonatal chronic lung disease and the consistent use of both the physiologic denition and severity scale would provide greater accuracy in determining the impact of the disease currently dened by its treatment. Our present labelling of the clinical state of oxygen supplementation and/or ventilatory support at 36 weeks postmenstrual age and the histopathologic severity of alveolar arrest and vascular simplication may not always be predictive of the degree of altered lung development and thus longer-term pulmonary function evaluations are needed to determine the impact of this disorder in specic infants. The proposed role of novel molecular therapies, and the combined effects of currently established therapies, as well as exogenous surfactant and inhaled nitric oxide or repetitive surfactant dosing, on the severity and incidence of new BPD hold considerable promise for reducing the long-term pulmonary moribidity among infants delivered prematurely. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Forty-two years ago Northway et al. initially described bronchopulmonary dysplasia (BPD) as an evolving radiographic pattern of lung injury among a group of moderately premature infants in the late saccular stage of lung development, who had primarily been treated with pressure-limited time-cycled ventilators and high levels of supplemental oxygen for prolonged intervals.1 Over the past four decades, dramatic changes in maternal care, including nearly universal use of antenatal steroids for mothers <34 weeks of gestation, introduction of surfactant therapy in the late 1980s, use of a variety of gentler ventilation strategies, and continuous monitoring of oxygen saturation have been associated with a marked increase in survival of very preterm infants predisposed to signicant pulmonary morbidities including BPD. However, exogenous surfactant therapy with animal-derived surfactants and

* Corresponding author. Division of Neonatology, Loma Linda University, 11175 Campus Street, Suite 11121, Loma Linda, CA 92354, USA. Tel.: 1 909 558 7448; fax: 1 909 558 0298. E-mail address: tamerritt@llu.edu (T.A. Merritt). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.009

more gentle ventilation strategies, including nasal continuous positive airway pressure (nCPAP) support for infants 2528 weeks gestational age, have not signicantly lowered the incidence of BPD at 36 weeks postconceptional age. By early 2009, only vitamin A supplementation, caffeine treatment, and intratracheal instillations of a mixture of budesonide and beractant among infants in the canalicular and early saccular phases of lung development have been associated with signicant reductions in BPD,24 while a Breathsavers quality collaborative demonstrated a 10.2% reduction in BPD at 16 centers.5 However, advances in neonatal care have created a different form of BPD that generally is dened less by radiographic pattern than by requirement for ongoing ventilation or supplemental oxygen treatment at 36 or 40 weeks corrected age (depending on gestational age at birth). This new form of BPD is associated with a disruption of lung organogenesis and impaired pulmonary function during the rst years of life. This new form of BPD has primarily been based upon autopsy studies of 14 very preterm infants who died at various intervals after birth, and on a baboon model of lung injury that is devoid of the human intrauterine environment and frequent inammatory

346

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

changes associated with human premature birth. These histologic studies suggest that the new form of BPD represents a disruption of lung organogenesis, and specically an arrest of alveolar septation and vascular development in the distal lung.6,7 The new BPD has become the most frequent chronic lung disease in infancy.8 However, prior to the denition of BPD suggested by the US National Institutes of Health, other forms of atypical chronic lung disease occurring after a course of respiratory distress syndrome (RDS), or later without preceding acute lung disease, have been identied.9 The extent to which these atypical forms of chronic lung disease overlap has not been rigorously evaluated. A novel classication of diffuse lung disease in infants and young children prepared by the Childrens Interstial Lung Disease (ChILD) Research Cooperative is based on the pathologic material and clinical data from 187 infants undergoing lung biopsies in 11

centers in 2007. This system describes acinar dysplasia characterized by lung growth arrest in the pseudoglandular or early canalicular phase, and congenital alveolar dysplasia characterized by growth arrest in the late canalicular/early saccular phase of lung development under a general pathologic category of growth abnormalities reecting decient alveolarization10 (Fig. 1). It is noteworthy that the new BPD in is not mentioned in this classication. In the National Institute for Child Health and Human Development (NICHD) neonatal network, BPD, dened as supplemental oxygen requirement at 36 weeks postmenstrual age, had an incidence of 52% in infants with birth weights 501750 g, 34% among infants with birth weights of 7511000 g, 15% among those 10011200 g at birth, and 7% in infants born between 1201 and 1500 g.11 However, a recent longitudinal evaluation of 1656

Fig. 1. Proposed classication of diffuse lung disease in children by the Childrens Interstial Lung Disease (ChILD) Research Cooperative. The study cohort was composed of patients under the age of 2 years who has a diagnostic lung biopsy during a 5-year period. The clinicalpathologic classication scheme is detailed with numbers of cases and specic entities identied within each category. Note that 22 cases were excluded, and cases with focal lesions, lobectomies, segmental resections, transbronchial biopsies, and needle biopsies are excluded. Reproduced with permission from Deutsch et al.10

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

347

surviving infants born at 2329 weeks of gestation from 2001 to 2006, reported an increase in this chronic lung disease from 47.8% to 57.8% (Fig. 2). This increase may reect increased survival of more extremely preterm infants, but the report also found that the use of surfactant during these years was decreasing as more infants were managed by non-invasive ventilation strategies.12 Whether the new BPD represents a specic new disease entity of extremely preterm infants with lung injury and disrupted repair during critical phases of lung organogenesis, or whether it is a group of entities associated with complex epigenetic,

environmental (especially pre- and postnatal infections), inammatory-mediated dysregulation of lung maturation, and/or other factors within a milieu of rapidly evolving therapies within the neonatal intensive care unit (NICU) remains unresolved. The orchestration of lung development by nely integrated and regulated networks of transcriptional factors, growth factors, matrix components, and physical factors (primarily airway stretch) results in morphologic maturation and vascular development associated with the alveolarcapillary interposition required for extrauterine gas exchange. Alveolar epithelial cells undergo

Fig. 2. Incidence of chronic lung disease stratied according to gestational age and surfactant use 20012006. Note that during this time-period, overall use of surfactant decreased from 67% to 59.9% while the chronic lung disease increased from 47.8% to 57.8%. A total of 598 infants did not receive surfactant, and 1120 infants were treated with surfactant(s). There was no difference in survival between the groups over this time-span. Adapted with permission from Chong et al.12

348

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

marked biochemical, morphological, and functional changes in the latter at 1214 weeks of gestation. Such alveolarization is heralded by the formation of secondary crests and alveolar formation after 36 weeks of gestation. Early in development lung cell lineage is marked by the expression of thyroid transcription factor I (Tit), and genetic studies have shown that Titf1 is critical for the development of distal lung progenitors.13 Alveolar saccules are lined with type I cells which, along with pulmonary capillaries, form an extensive and effective gas exchange area. Alveolar type II cells compose 510%of the alveolar surface and are critical for production of surfactant phospholipids and proteins required for lung stability and immunologic protection at birth. Transcriptional networks inuence both sacculation and alveolarization by coordinating differentiation of alveolar epithelial cells critical for surfactant synthesis and secretion. Sacculation, alveolarization, and vasculogenesis of the peripheral lung are dependent on many of the transcription factors involved in lung maturation (e.g. Tit). These same transcription factors inuence the expression of factors [Fox, vascular endothelial growth factor (VEGF)] that orchestrate pulmonary vasculogenesis by regulating a number of genes that inuence mesenchymal differentiation and blood vessel formation.1417 An interplay of these and other transcriptional factors occurs in the maturing lung, and their expression is altered during the transdifferentiation and proliferative stages of lung development when multiple forms of lung injury occur.18 This article will review mechanisms of lung injury in the preterm infant, call for greater precision regarding the diagnosis of new BPD, a disorder dened by its treatment, and suggest that it be differentiated from other chronic lung diseases that affect preterm infants. Because nearly all diseases of the newborn have a basis in genetics, and since there is strong evidence that the new BPD is heritable, the genetics will be briey reviewed. The role of an altered intrauterine environment, and how the neonatologists choice of treatments at birth inuences its outcome, including factors that may contribute to its severity, will also be discussed. 2. Epigenetics of thenew BPD Model-tting analyses in preterm monozygotic and dizygotic twins have conrmed that susceptibility to BPD is highly heritable when controlling for shared and non-shared environmental effects.19 Male infants are at increased likelihood of severe acute lung disease and BPD.20 However, to ascertain the genetic susceptibility and to examine the linkage between specic genes and BPD, transmission disequilibrium testing has found that SP-B intron 4 deletion (i4del) signicantly increased the risk of BPD in a Finnish cohort21 and also among 140 German preterm infants and 58 term infants.22 Several other candidate genes have been explored by Hallman et al. for possibly contributing to increased risk for BPD,23 and SP-B i4del consistently increases BPD risk for the presenting twin. Further, it is proposed that an interaction exists between fetuses exposed to the intrauterine cervix, who are more prone to ascending infection than the non-presenting twin during gestation.24 Similar effects of birth order have also been reported by these workers in allelic association studies of SP-A and SP-B genes and RDS.25 Intron regions of the genome have regulatory roles in gene expression, and thus predispose infants with deletions to produce less SP-B initially or in response to injury. Specic insertion/deletion polymorphisms have been associated with altered mRNA and protein levels. In SP-B i4del several transcription factor binding sites are lost compared to the wild type SP-B intron 4 allele. These sites are related to the inammatory cascade, and regulators of surfactant protein expression essential for normal lung morphogenesis and alveolarization.25,26 Thus, SP-B i4del may

dysregulate SP-B expression and lower levels of SP-B, which would predispose infants to continued respiratory failure. These lower levels of SP-B in BPD have been reported from airway secretions of baboons developing the new BPD.27 Studies have also examined the association between BPD and single nucleotide polymorphisms for gene expression of tumour necrosis factor a (TNFa), interleukin1b (IL-1b), transforming growth factor b (TGFb), and monocyte chemoattractant protein-1. Data from these studies suggest that these single nucleotide polymorphisms do not play a signicant role in determining risk for BPD, although one study reported that the adenine allele of TNFa-238 may reduce the incidence of BPD.28 Polymorphisms of genes coding for VEGF reveal that infants carrying the -460T allele had increased risk of BPD by 9% [95% condence interval (CI): 214%] above baseline risk accounting for gender, gestational age, and length of supplemental oxygen therapy.29 Given that shared genetic and environmental factors (and their interaction) account for 6580% of enhanced BPD risk (accounting for institution of treatment), knowledge of unique genetic risks may permit more tailored therapies for the infant at highest risk. Polymorphisms in pro-/anti-inammatory cytokines, their receptors, proteases and their inhibitors, and tissue antioxidant enzymes will likely be identied that may contribute to the occurrence and severity of BPD in an individual infant. A metaanalysis of studies involving 804 preterm infants did not nd a signicant association of TNFa (-308A) polymorphism and risk for BPD.30 Multiple lines of evidence suggest that the new BPD likely represents a polygenetic disorder.31 Gene pathways differentially expressed during the development of premature infants are primarily those related to oxidative phosphorylation, mitochondrial energy metabolism, and repair of DNA. Analyzing umbilical cord RNA of 54 infants at <28 weeks of gestation using gene-chip microarrays, Cohen et al. found that histone acetyltransferase binding activity, chromatin remodeling pathways, and pathways regulating cell growth were among the differentially expressed pathways in premature infants developing the new BPD.32 These ndings are of particular relevance because non-invasive nasal CPAP or high frequency jet ventilation of preterm lambs preserves histone acetylation pathways, whereas mechanical ventilation does not. However, use of histone deacetylase inhibitors (such as valproic acid or trichostatin A) during mechanical ventilation may protect the developing lung from altered gene expression accompanying mechanical ventilation.33 3. Chorioamnionitis and fetal systemic response to inammation Chorioamnionitis is the single most important cause of preterm birth, and severe chorioamnionitis is seen most frequently in preterm deliveries before 30 weeks of gestation.34 Fetal systemic response to inammation35 is associated with the older form of BPD,36 intraventricular hemorrhage, cystic periventricular leukomalacia,37 and cerebral palsy.38 Watterberg36 reported that ventilated preterm infants exposed to histologically conrmed chorioamnionitis had a lower incidence of RDS, but higher rates of BPD than did infants not exposed to chorioamnionitis. On the other hand, Van Marter et al.39 found a lower incidence of BPD among infants delivered with chorioamnionitis, except among those with conrmed sepsis or those who required ventilation for 7 or more days. Using rigorously histologically dened chorioamnionitis and chronic lung disease in very low birth weight infants, Redline et al. found no increased risk of BPD.40 Similarly, after an extensive evaluation of cord blood immunoproteins and placental pathology, Kaukola et al.41 found no association between chorioamnionitis and the new BPD. However, in animal models, intrauterine endotoxin exposure leads to disturbed alveolar growth and vascularization

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

349

coincident with the inammatory challenge.42 Within 14 days after preterm birth, inammatory biomarkers (chemokines, adhesion molecules, pro- and anti-inammatory cytokines, proteases and their inactivated inhibitors, and growth factors) have complex

interactions that alter subsequent lung maturation.4345 These have been elegantly summarized by Bose et al. (Fig. 3).46 Goldenberg et al. recently reported 23% of infants born between 23 and 32 weeks of gestation have umbilical blood cultures positive

Fig. 3. Critical steps, and associated mediators, in lung inammation, injury, and remodeling that result in bronchopulmonary dysplasia. Reproduced with permission from Bose et al.43

350

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

for genital mycoplasmas (Ureaplasma urealyticum and Mycoplasma hominis).47 Intrauterine infection, histologic chorioamnionitis, and cord blood IL-6 levels were elevated among infants exposed to the fetal inammatory response syndrome with a highly signicant increased risk in these infants of developing BPD.48 There is abundant evidence in fetal sheep that in the presence of chorioamnionitis there is recruitment of neutrophils into the neonatal lung, expression of proinammatory mediators, such as endotoxin and IL-1a, IL-1b, lung microvascular injury, decreased vascular endothelial growth factor, reduced nitric oxide synthase activity, smooth muscle proliferation, and an arrest in alveolar septation.49 Recently, cord blood IL-8 was found to predict risk for BPD. Its concentration on day 1 predicted severity of illness at 36 weeks postmenstural age among infants with chorioamnionitis, although some infants with elevated IL-8 in the cord blood did not have histologically conrmed chorioamnionitis. It is noteworthy that IL-19 was specically elevated among infants with chorioamnionitis. Despite the rapid fall in elevated fetal cytokine levels by 24 h after birth, plasma levels of cytokines consistently predicted risk for BPD, suggesting that downregulation of cytokine production is compromised in the population of very preterm infants predisposed to develop BPD.50 Whether the chronic lung disease developed by infants exposed to chorioamnionitis represents the new BPD has been questioned. Hodgman51 reported the relationship between extremely preterm infants delivered with chorioamnionitis and WilsonMikity disease, a chronic interstitial disease of premature infants described in 1960. This progressive neonatal lung disease occurs in the absence of an early severe insult, is relatively common, and is often called atypical BPD. A recent report documents its onset between 6 and 28 days, a male predominance, and a mortality rate of 11% among nine infants who, after brief postdelivery resuscitation and/or a brief period of nasal CPAP, developed signs of respiratory distress and cyanosis in the absence of infection, cystic brosis, or immune disorders. Although their initial chest radiographs were normal or with mild granularity, all progressed to cysts, areas of scarring, atelectasis, and hyperination typical of BPD.52 Thus, chronic lung disease among very preterm infants is not simply synonymous with ventilatorinduced damage associated with high concentrations of supplemental oxygen, but the nal result of a range of complex insults including intrauterine inammation, postnatal infection, lung immaturity, the effects of the patent ductus arteriosus, resuscitation maneuvers, and ventilator-associated injury.53 4. Volutrauma, atelectotrauma, and ventilator-induced lung injury Lung distension with mechanical ventilation is related to the magnitude and duration of cyclic over-distension or volutrauma of the ventilated airways and their collapse at end-expiration. Mechanical stretch of the immature lung distorts the cells and extracellular matrix leading to alterations of stretch-responsive genes with inuence downstream on the expression of growth factors and inammatory mediators.54,55 Another consequence of mechanical ventilator-associated lung injury is inammation within the airways and the developing alveoli. Ventilator-induced lung injury was explored in fetal lambs (110 days, midlate canalicular stage) by ventilating in utero while maintaining uteroplacento-fetal circulation, thus obviating the need for neonatal resuscitation or supplemental oxygen exposure. In-utero ventilation for 16 h with replacement of lung liquid resulted in more simplied distal airways, and hypercellularity (proliferation of interstitial broblasts), enlarged alveolar sacs with thickened septal walls, and hemorrhage. Secondary crest density within the saccules of the immature lung was reduced, indicative of alveolar

simplication, and was worse among fetuses undergoing ventilation for 612 h. Elastin and collagen deposition was increased in the saccular walls, and myobroblasts (reected by a smooth muscle actin staining) showed differentiation consistent with brotic changes in fetuses undergoing 12 h of in-utero ventilation followed by replacement of lung uid, and maintained in-utero for 7 additional days. Septal crest densities in the in-utero-ventilated fetuses had reabsorption of pre-existing secondary septal crests (rather than reduction in their new formation), suggesting that ventilation induced de-differentiation rather than an arrest of alveolar formation and saccular simplication.56 To reduce cyclic stress to the delicate preterm lung and to ameliorate lung injury, the use of CPAP has been advocated for both resuscitation and early management.57 In preterm lambs, 2 h of intermittent mandatory ventilation resulted in almost a sevenfold increase in neutrophils and a doubling of hydrogen peroxide in alveolar washes compared to similar preterm lambs, who were intubated but managed by CPAP alone.58 However, with this model and with PaCO2 levels maintained between 50 and 55 mmHg for 6 h, inammatory mediators in lung washes were similar.59 In related experiments, preterm lambs exposed to gentle ventilation or CPAP and given intratracheal lipopolysaccharides (LPS), cytokine mRNA was increased in both lung and liver in a similar manner in both lambs, suggesting similar lung injury with either CPAP or conventional ventilation.60 In many centers, the initial care has now focused on the use of CPAP and avoidance of conventional ventilation as less invasive and injurious to the immature lung. A large randomized trial reported early benets of CPAP relative to conventional ventilation despite a signicant threefold increase in pneumothorax rates; however, no signicant differences in death or BPD at 36 weeks post conception were noted. Furthermore, among infants randomized to CPAP alone, 46% required intubation during the rst 5 days because of severe respiratory failure.61 Using high frequency nasal ventilation contrasted to intermittent mandatory ventilation or continuous distending pressure, Reyburn et al.61 found that oxygenation and ventilation goals were achieved with lower airway pressures and less supplemental oxygen in nasally ventilated preterm sheep (130132 days); however, these fetal animals required resuscitation using endotracheal intubation and surfactant administration prior to high frequency nasal ventilation. Morphologic analysis of the lungs revealed distal airspace wall thickness, density of secondary septa, and improved radial alveolar counts among those lambs receiving distending pressures of about 2 cmH2O contrasted to those receiving intermittent mandatory ventilation and 13 cmH2O. Although these lambs were nasally ventilated for only 3 days, lung morphology documented apoptotic mesenchymal cells, thinning of the distal airspaces with normal septal formation, and near normal lung histology. Unfortunately, the effects of prolonged nasal high frequency ventilation have not been reported in either animal models or preterm human infants. These ndings suggest important opportunities regarding non-invasive ventilation strategies, and new potential therapies focused upon the use of inhibitors of the histone deacetylation and chromatin remodeling pathways to prevent dysregulation of lung organogenesis. The concept of lung stretch is now well-founded in multiple experimental models of neonatal treatments that inate the immature lung causing stretch on immature airways and respiratory saccules followed by deation. Parathyroid hormone-related protein (PTHrP) is a highly conserved, stretch-regulated protein that has been identied as mediating lung development. Knock-out mice for PHTrP fail to develop alveoli.62 PTHrP is expressed in the endoderm and binds to mesoderm to upregulate the local inammatory response, and thus acts to integrate surfactant synthesis

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

351

and alveolar capillary perfusion.63 Under the inuence of sonic hedgehog, the endoderm expresses PTHrP and its receptor in the mesenchyme. PTHrP binds to its receptor in the mesenchyme and activates the protein kinase A pathway to downregulate the Wingless/int pathway through the PPARg (a nuclear transcription factor), and its downstream regulatory genes prompt adipocyte differentiation. Lipobroblasts also release leptin, promoting alveolar type II cell phospholipids and surfactant protein synthesis, which provide protection against reactive oxygen species (ROS), and further promote alveolar type II cell proliferation. Alveolar stretch, infection, or ROS are suspected of shifting lipobroblast to myobroblast transdifferentiation into brotic cells, which disrupt alveolar growth.64 Tracheal aspirate levels of PTHrP during the rst week of ventilation in very low birth weight infants were inversely correlated to prolonged ventilation and the subsequent development of BPD.65 Cerny et al.66 proposed that a PPARg agonist, rosaglitazone, will reduce hyperoxic, stretch-related lung injury and reduce myobroblast formation. This PPARg agonist has been shown to reduce hyperoxia-induced neonatal lung injury in the rat; however, to date there have been no reports showing that this agent (or others) reduces either disordered lung maturation in animal models or the new BPD in infants. 5. Disruption of vasoculogenesis Among infants succumbing to the new BPD, there is a reduction in both lung vasculature and simplication of the alveoli. There are two processes of vascular growth: vasculogenesis and angiogenesis. Vasculogenesis is the de-novo formation of vessels from precursor cells (angioblasts and endothelial precursors). Angiogenesis is the formation of vessels by extending new vessels from existing vessels. In the most widely accepted model, it is currently thought that the large pulmonary vessels form by angiogenesis, and the pulmonary capillary bed forms by vasculogenesis (Fig. 4).6769 There is a further process of fusion that is necessary to connect the proximal and distal vessels.67,68 In de Mellos model, there is central angiogenesis and distal vasculogenesis, where the capillary bed is created from hematopoietic lakes in the lung mesenchyme.68 Hall et al. have suggested that the vasculogenesis of the distal vessels originates from endothelial cell precursors.70,71 Both de Mello and Hall suggest that central pulmonary venous development occurs by angiogenesis and precedes that of arterial development.67,71 Recently, Parera has proposed a different mechanism, where both proximal and distal vessels develop by angiogenesis.72

Vascular development in the lung is under complex but tight control. There are several specic regulators of vessel formation. The endothelial-specic regulators include the ligand/receptor pairs of: VEGF/VEGF receptor (VEGFR), angiopoietin/TIE, notch/ jagged.73 It is known that VEGF-A is a requirement for embryonic vascular development. VEGF is released by respiratory epithelial cells and enhances migration, proliferation, and differentiation of adjacent endothelial cells through the VEGFRs Flt-1 and Flk-1.73 VEGF stimulates the formation of mesoderm and the aggregation of angioblasts. Angiopoietins and their major receptor, Tie-2, promote vascular integrity.74 Defects in the notch-2 signaling pathway have been shown to cause abnormal lung microvasculature development.75 A complete review of lung vascular development is beyond the scope of this paper; several extensive reviews of the mechanisms of lung vascular development have been recently published.73,76,77 VEGF pathways are also important in the development of normal alveolarization.78,79 Inhibition of VEGF during alveolar development leads to decreased alveolarization (Fig. 5).80 Giving VEGFR inhibitors to the developing rat not only inhibits vascular growth, it reduces alveolar septation with a decrease in the nal alveolar number.8082 However, giving neonatal rats that have been exposed to VEGFR inhibitors inhaled nitric oxide improves lung growth.83 VEGF regulation remains essential even through adult life. Adult rats chronically treated with VEGFR inhibition84 and lung-targeted VEGF inactivation85 develop emphysema. In infants born prematurely, acute lung injury impairs the growth, structure, and function of the developing pulmonary circulation.73,86 Preterm infants with BPD have pulmonary hypertension and abnormal pulmonary vasoreactivity.87,88 Infants with BPD have also been shown to have dysregulated VEGF/VEGFR signaling.89,90 Several mechanisms have been shown to inhibit angiogenesis and thus potentially alveologenesis. The elevated oxygen levels of extrauterine life lead to downregulation of the VEGF/VEGFR signaling pathway with inhibition of vascular growth and alveolarization.91,92 Endoglin, part of the transforming growth factor (TGF)-b receptor complex, has been implicated in the dysangiogenesis that is seen in mechanically ventilated preterm infants. In autopsy specimens of human infants who have short-term mechanical ventilation, endoglin levels have been shown to be signicantly elevated, while VEGF and Flt-1 levels are markedly lower when compared to control non-ventilated infant specimens.93 A potential mechanism of microvascular dysangiogenesis

Fig. 4. Three models of pulmonary vascular development. Model 1: central angiogenesis with distal vasculogenesis with the capillaries arising from vascular lakes. Model 2: central angiogenesis with distal vasculogenesis with the capillaries arising from endothelial cells. Model 3: central and distal angiogenesis. HL, Hematopoietic Lakes; ECP, Endothelial Precursor Cells. Reproduced with permission from Parera et al.72

352

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

Fig. 5. Barium arteriograms showing the effects of SU-5416 (vascular endothelial growth factor receptor inhibitor) and inhaled nitric oxide (iNO) on lung vascular perfusion in infant rats. SU-5416 reduced the perfusion of small pulmonary arteries. iNO treatment enhanced vascular density after SU-5416 treatment. Reproduced with permission from Tang et al.83

is a shift in the regulatory control from VEGF and angiopoietin-1 to endoglin. Endostatin is a potent angiogenesis inhibitor that inhibits endothelial cell proliferation and migration and also induces endothelial cell apoptosis. Endostatin has been shown to antagonize the effects of VEGF-A and other pro-angiogenic factors.94 Endostatin has been suggested to play a physiologic role in the development of the human lung; in addition, a disturbance in the equilibrium between endostatin and VEGF-A interferes with the development of the lung.95 Conversely, up-regulation of the VEGF/VEGFR signaling pathway has been shown to mature lung architecture even in the presence of lung injury. Kunig et al. have shown in a neonatal rat hyperoxic lung injury model that the delivery of human recombinant VEGF enhances vessel growth and alveolarization.96 Inhaled nitric oxide has been shown in ovine lung cell culture and eNOSdecient mice to improve both angiogenesis and normal lung structure.97,98 These ndings suggest that a reinstitution of the disruption of the VEGF/VEGFR signaling pathway may allow more normal development of preterm lungs. 6. Toxic effects of oxygen on the developing lung The intuition of Northway et al. that oxygen toxicity contributed to the genesis of BPD has been fully conrmed by subsequent research.99 ROS, including superoxide anion, singlet oxygen, hydroxyl radical, and peroxide are produced by both the cellular metabolism of molecular oxygen and the activation of neutrophils and macrophages exposed to inammatory mediators. ROS cause oxidative stress by apoptosis and the oxidation of lipids, proteins, and DNA,100 and may also function as signaling molecules that mediate biological responses.101 The premature infant is particularly susceptible to oxidant stress because of the high concentrations of supplemental oxygen in respiratory support, the relative deciency of their immature antioxidant systems, and the high prevalence of inammatory conditions, such as chorioamnionitis, in the perinatal period. In addition, preterm macrophages may have an enhanced cytokine inammatory response to oxygen exposure. Alveolar macrophages from preterm rabbits show an increase in IL1-b and IL-8 mRNA when exposed to 95% oxygen, whereas term rabbit macrophages do not.102 The infants described by Northway et al. suffered from surfactant-decient RDS, and were treated with high concentrations of supplemental oxygen (80100%) and prolonged positive pressure ventilation. The pathologic markers of old BPD inammation, brosis, and airway smooth muscle hypertrophy are rarely seen

today. However, the extremely premature infants who now survive have abnormalities of alveologenesis and vasculogenesis, manifested by chronic radiographic abnormalities (hazy lung) and prolonged dependence upon supplemental oxygen, although they are treated with oxygen concentrations that have not been associated with lung injury in more mature infants. These observations have led some investigators to question the role of oxygen toxicity as a culprit in the new BPD.8,103 However, there is accumulating evidence that ROS can cause the pathological changes observed in infants with new BPD. These pathological changes depend less upon oxidative injury to structural elements followed by aberrant healing, and more upon the disruption of pulmonary organogenesis. Oxidative stress upsets a complex orchestration of developmental events, mediated by a bewildering array of genes, growth factors, receptors and inammatory mediators. The extremely premature infants who comprise those at risk for the new BPD are in the late canalicular and early saccular stages of lung development, occurring between about 23 and 30 weeks of postmenstrual age. During this period, the developing lung undergoes major alterations of both the pulmonary epithelium, which branches from the terminal bronchioles to form canaliculi and acini, and the surrounding mesenchyme, which supports an invasion of capillaries surrounding the developing airways. Towards the end of this period the airways divide by primary septation to form saccules, the precursors of alveoli. Hyperoxia inhibits branching morphogenesis in both animal models and cell culture. Preterm baboons ventilated for 7 days with 100% oxygen had 50% fewer alveoli compared to animals ventilated with lower concentrations of oxygen.104 Even modest hyperoxia (35% and above) for 48 h impaired branching morphogenesis and growth of lung explants from 12 day gestation mouse fetuses. This effect was ameliorated by several varieties of superoxide dismutase but not by other antioxidants, suggesting that superoxide anion was responsible.105 Late-gestation lung 1 (LGL1), a glucocorticoidinduced, developmentally regulated gene required for branching morphogenesis, and perhaps for alveolar septation, was downregulated in newborn rats exposed to either 60% oxygen for 2 weeks or 95% oxygen for 1 week. Animals exposed to hyperoxia for only 1 week and recovered in air over a 3 week period had a normalization of LGL 1 levels.106 Vascular endothelial growth factors (VEGF) are required for vasculogenesis and may also promote alveolar development. Lungs of 140 day gestation fetal baboons exposed to 100% oxygen for 10 days showed oxidant DNA damage, an increase in p53 transcription factor, and decreased VEGF expression. The p53 transcription factor

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

353

is known to mediate cell cycle arrest and endothelial cell death.107 Branching morphogenesis may be inhibited even by oxygen tensions considered normal, since the lung develops in a milieu that is hypoxic by extrauterine standards, with oxygen tensions in the umbilical artery that are about 20 torr (2.66 kPa) in late fetal life. Interestingly, explants of primitive mouse lung buds demonstrated greater branching when cultured in 3% vs 20% oxygen.108 The expression of hypoxia-inducible factor, tenascin-C, and VEGF, both of which are important for angiogenesis and branching morphogenesis, are inhibited by normoxia.108110 Hyperoxic disruption of alveolarization is also associated with alterations in growth factor receptors. Newborn mice exposed to 85% oxygen showed fewer alveolar secondary crests, larger alveoli, and greater septal wall thickness than did mice raised in 21% oxygen. Postnatal increases in FGFR-3, FGFR-4, and FGFR-7 were observed in room air mice but not in mice exposed to hyperoxia. This suggests that alterations in these factors may contribute to abnormal lung development.111 The deleterious effects of ROS include stimulation of factors responsible for inhibitory signaling. Newborn rats, exposed to 85% oxygen for 10 days, showed decreased alveologenesis and disrupted vasculogenesis. TGF-b levels were increased with levels that correlated to the degree of lung injury. Pre-treatment with anti-TGF antisera partially prevented lung injury, increased alveologenesis, improved alveolar septal elastin organization and improved the development of the microvasculature.112 Hyperoxia may also affect morphogenesis through its chemotactic effect on neutrophils. Neutrophils and macrophages are known to release superoxide anion and H2O2 at sites of inammation with resultant tissue damage.113 However, they can also affect morphogenesis. Newborn rats exposed to 60% oxygen develop a heterogeneous lung injury, with areas of decreased alveologenesis interspersed with areas of interstitial thickening. This injury is associated with an increase in lung neutrophil content and neutrophil chemokines. Blocking neutrophil inux with selective chemokine receptor antagonists reduced the production of ROS and improved alveologenesis.114 What are the implications of these studies for the care of extremely premature infants who require respiratory support using supplementary oxygen? All supplemental oxygen, even levels within the normoxic range are potentially injurious to the development of the canalicular lung. Selection of the ideal level of oxygen saturation is problematic. The STOP-ROP clinical trial demonstrated that infants cared for in an elevated oxygen saturation range had higher rates of pneumonia or chronic lung disease than those managed with lower oxygen saturations, and those with underlying lung disease had a signicantly greater need for supplemental oxygen treatment, longer hospitalization, and more diuretic treatments.115 Similarly, the BOOST Trial compared infants <30 weeks of gestation between targeted oxygen saturation ranges of 9194% vs 9598% and found no developmental advantages in those exposed to higher oxygen saturations and a signicantly higher rate of dependence on supplemental oxygen at 36 weeks postmenstrual age.116 The concentration of oxygen that is safe for a particular gestational age is currently unknown.117 The measurement of PaO2 and oxygen saturation, which currently guides clinical practice, is insufcient to assess either oxygen delivery to tissues or oxygen exposure risk to the respiratory epithelium. Finally, our increased understanding of oxidant stress and antioxidant defenses, or even the supplementation of superoxide dismutase,118 has not resulted in new therapies that have been proven to reduce the incidence of BPD. As profoundly stated by Tin, Oxygen must have been given to more infants than any other medicinal product in the last 60 years. Despite that, we still know very little about how much infants actually need, or how much it is wise to give. Given that we have also known for nearly 50 years that it is easy to damage

the eyes of preterm infants by giving too much oxygen, especially in the rst weeks of life, the depth of our ignorance is really, quite embarrassing.119 We contend that this ignorance also extends to the developing lungs of very preterm infants and that our treatments, as demonstrated in at least two clinical trials, have contributed to the numbers of infants with new BPD. 7. Conclusion There are several current challenges for BPD researchers and neonatologists. Is there evidence to support the use of the physiologic oxygen challenge at 36 weeks postmenstrual age (for infants <32 weeks of gestation) or >28 days but <56 days (those delivered at 32 weeks of gestation) to correctly identify those infants at highest risk for adverse pulmonary sequelae after NICU discharge? Further, does the severity scale of BPD using the National Institutes of Health classication have any relationship to degree of impairment in alveolar and vascular development regardless of cause, or should other markers of lung injury, degree of inammation, tissue repair, or lung volume be used prior to a severity that implies eventual pulmonary impairment? At this point these data are lacking, yet there are multiple biomarkers that might be used to assist the clinician in determining the extent of impairment in lung development that translates into impaired lung function. However, the physiologic oxygen challenge at 36 weeks does appear to account for variability in oxygen use in many NICUs, and therefore its use needs to be applied by neonatologists and adopted by major national or international databases of infant outcomes as this will reduce variation between centers. Emphasis must be placed on greater precision in differentiating new BPD from other neonatal lung disorders, although it is unlikely that lung biopsies are justied in the majority of infants. Advanced imaging modalities may provide the opportunity for greater diagnostic specicity and permit a prediction of lung function. Gene microarrays to determine specic pro- and anti-inammatory pathway regulators, or abnormalities in other pathways discussed in this article, or specic gene products closer to term gestation among very preterm infants, may be helpful. However, the current practice of labeling as BPD any infant who is requiring supplemental oxygen at 36 weeks as synonymous with an arrest of alveolar and vascular development or as having histopathologic ndings identical with those described in the pathology of the new BPD is currently unwarranted, as these two ndings may not be causally linked as recently suggested.120 Greater insight is needed into the pathogenesis of the new BPD (Fig. 6). If the primary problem is arrest of alveologenesis and vasculogenesis, and if that developmental disruption impairs gas exchange, why do some infants develop a requirement for supplemental oxygen during the perinatal period when they had none at birth? Greater understanding of the gas exchange abnormality in these infants may guide clinical management. Their mild hypoxemia is probably caused by ventilationperfusion mismatch, although it is possible that diffusion limitation is also contributory. Infants with accompanying hypercarbia may have increased dead space ventilation (ventilation of unperfused alveoli). Analytic techniques, such as multiple inert gas elimination and the diffusing capacity for carbon monoxide, have been adapted for small animals and can be used to sort out these possibilities. Techniques to measure tissue oxygenation in order to minimize needless oxygen exposure should guide clinical use of supplemental oxygen. Measurement of arterial oxygen saturation is insufcient for this purpose. Although the fetus resides in a relative hypoxemia, tissue oxygenation may be greater than we suppose. Neonates have increased oxygen-carrying capacity because of their relatively large red cell mass, fetal hemoglobin (P50 19 torr), and high mean

354

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

Fig. 6. Multiple factors contributing to the epigenetics of the new bronchopulmonary dysplasia associated with lung development, and gestational and postnatal developmental periods. FIO2, fraction of inspired oxygen; VT, tidal volume; CPAP, continuous positive airway pressure; PEEP, positive end-expiratory pressure; NICU, neonatal intensive care unit; IMV, invasive mechanical ventilation; HIFI, High Frequency Ventilation in Premature Infants study; VEGF, vascular endothelial growth factor; FEV1, forced expiratory volume in 1 s.

corpuscular hemoglobin. It may be that the premature newborn can maintain adequate tissue oxygenation with less supplemental oxygen than is currently used, and individualization of the use of oxygen from the moment of birth to NICU discharge may reduce the

consequences of undue oxygen exposure. Combining the measurement of non-invasive oxygen saturation and hemoglobin provides determination of continuous oxygen content and delivery to tissues that may guide more prudent use of this drug.

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357

355

Although conicting data exist regarding the precise mechanism(s) of lung protection, it is likely that signicantly fewer infants require mechanical ventilation and thus can be managed with noninvasive ventilation after initial surfactant administration. The combination of inhaled nitric oxide and non-invasive ventilation may be effective in preserving pulmonary and vascular development. New pharmacologic therapies that preserve intricate pathways modulating alveolar and vascular development are currently being tested in animal models. Promising therapies including inhibitors of histone deacetylation and agonists of peroxisome proliferator-activated receptor g pathways, such as prostaglandin J2 or rosiglitazone, may offer new pharmacologic interventions that preserve pulmonary and vascular morphogenesis. Recent reports have supported the trend for reduction in BPD when using surfactant as an adjunct to CPAP therapy or the extended use of a peptide-derived surfactant administered to infants requiring assisted ventilation. Ongoing studies evaluating the effect of inhaled nitric oxide and surfactant and repeated dosing of surfactant are eagerly awaited. Clearly, the morbidities of prematurity could be ameliorated if the causes of premature labor in individual pregnancies were better understood, and effective therapies to prolong gestation without compromising fetal well-being would be a major breakthrough in preventing chronic lung disease. In addition, universal access to prenatal care and improvements in prevention of prematurity would result in fewer infants at risk for new BPD.

Practice points  The new BPD occurs among preterm infants delivered in the late canalicular or early saccular stages of lung development. Lung injury during fetal and early neonatal life alters the highly integrated morphogenic program of lung development. These injuries include fetal and neonatal infection, neonatal ventilation and surfactant deciency, oxygen therapy, reactive oxygen species, nutritional deciencies, and perhaps unknown mechanisms that dysregulate subsequent lung development.  Greater precision in understanding the molecular mechanims of lung injury offer novel approaches to therapies in the future, whereas current therapies offer only a modest impact on the incidence and severity of this chronic lung disease.  Use of the NIH denitions and clinical severity of BPD and physiologic diagnosis of BPD should be precise and the impact of new biomarkers or therapies judged not only on the requirement of oxygen and/or ventilation at 36 weeks postmenstrual age, but on longer-term pulmonary function in affected infants.

Conict of interest statement None declared. Funding sources None. References


1. Northway Jr WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med 1967;276:35768. 2. Tyson JE, Wright LL, Oh W, et al. Vitamin A supplementation for extremelylow-birth-weight infants. National Institute of Child Health and Human Development Neonatal Research Network. N Engl J Med 1999;340:19628.

3. Schmidt B, Roberts RS, Davis P, et al. Caffeine therapy for apnea of prematurity. N Engl J Med 2006;354:211221. 4. Yeh TF, Lin HC, Chang CH, et al. Early intratracheal instillation of budesonide using surfactant as a vehicle to prevent chronic lung disease in preterm infants: a pilot study. Pediatrics 2008;121:e13108. 5. Payne NR, LaCorte M, Karna P, et al. Reduction of bronchopulmonary dysplasia after participation in the Breathsavers Group of the Vermont Oxford Network Neonatal Intensive Care Quality Improvement Collaborative. Pediatrics 2006;118(Suppl. 2):S737. 6. Coalson JJ. Pathology of new bronchopulmonary dysplasia. Semin Neonatol 2003;8:7381. 7. Husain AN, Siddiqui NH, Stocker JT. Pathology of arrested acinar development in postsurfactant bronchopulmonary dysplasia. Hum Pathol 1998;29:7107. 8. Jobe AJ. The new BPD: an arrest of lung development. Pediatr Res 1999;46:6413. 9. Charafeddine L, DAngio CT, Phelps DL. Atypical chronic lung disease patterns in neonates. Pediatrics 1999;103(4 Pt 1):75965. 10. Deutsch GH, Young LR, Deterding RR, et al. Diffuse lung disease in young children: application of a novel classication scheme. Am J Respir Crit Care Med 2007;176:11208. 11. Ehrenkranz RA, Walsh MC, Vohr BR, et al. Validation of the National Institutes of Health consensus denition of bronchopulmonary dysplasia. Pediatrics 2005;116:135360. 12. Chong E, Greenspan J, Kirkby S, Culhane J, Dysart K. Changing use of surfactant over 6 years and its relationship to chronic lung disease. Pediatrics 2008;122:e91721. 13. Maeda Y, Dave V, Whitsett JA. Transcriptional control of lung morphogenesis. Physiol Rev 2007;87:21944. 14. Kalinichenko VV, Lim L, Stolz DB, et al. Defects in pulmonary vasculature and perinatal lung hemorrhage in mice heterozygous null for the Forkhead Box f1 transcription factor. Dev Biol 2001;235:489506. 15. DeFelice M, Silberschmidt D, DiLauro R, et al. TTF-1 phosphorylation is required for peripheral lung morphogenesis, perinatal survival, and tissuespecic gene expression. J Biol Chem 2003;278:3557483. 16. Maniscalco WM, Watkins RH, Pryhuber GS, Bhatt A, Shea C, Huyck H. Angiogenic factors and alveolar vasculature: development and alterations by injury in very premature baboons. Am J Physiol Lung Cell Mol Physiol 2002;282:L81123. 17. De Paepe ME, Mao Q, Powell J, et al. Growth of pulmonary microvasculature in ventilated preterm infants. Am J Respir Crit Care Med 2006;173:20411. 18. Park KS, Wells JM, Zorn AM, et al. Transdifferentiation of ciliated cells during repair of the respiratory epithelium. Am J Respir Cell Mol Biol 2006;34:1517. 19. Lavoie PM, Pham C, Jang KL. Heritability of bronchopulmonary dysplasia, dened according to the consensus statement of the National Institutes of Health. Pediatrics 2008;122:47985. 20. Henderson-Smart DJ, Hutchinson JL, Donoghue DA, Evans NJ, Simpson JM, Wright I. Prenatal predictors of chronic lung disease in very preterm infants. Arch Dis Child Fetal Neonatal Ed 2006;91:F405. 21. In:Haataja R, Rova, M, Marttila R, et al. (editors). TDT analysis of surfactant protein B gene intron 4 deletion variants in Finnish preterm infants with bronchopulmonary dysplasia. ATS 2005 International Conference, May 2025, 2005, San Diego, CA. Proceedings of the American Thoracic Society. 22. Makri V, Hospes B, Stoll-Becker S, Borkhardt A, Gortner L. Polymorphisms of surfactant protein B encoding gene: modiers of the course of neonatal respiratory distress syndrome? Eur J Pediatr 2002;161:6048. 23. Hallman M, Marttila R, Pertile R, Ojaniemi M, Haataja R. Genes and environment in common neonatal lung disease. Neonatology 2007;91:298302. 24. Rova M, Haataja R, Marttila R, Ollikainen V, Tammela O, Hallman M. Data mining and multiparameter analysis of lung surfactant protein genes in bronchopulmonary dysplasia. Hum Mol Genet 2004;13:1095104. 25. Marttila R, Haataja R, Ramet M, Lofgren J, Hallman M. Surfactant protein B polymorphism and respiratory distress syndrome in premature twins. Hum Genet 2003;112:1823. 26. Margana RK, Boggaram V. Functional analysis of surfactant protein B (SP-B) promoter. Sp1, Sp3, TTF-1, and HNF-3alpha transcription factors are necessary for lung cell-specic activation of SP-B gene transcription. J Biol Chem 1997;272:308390. 27. Ballard PL, Gonzales LW, Godinez RI, et al. Surfactant composition and function in a primate model of infant chronic lung disease: effects of inhaled nitric oxide. Pediatr Res 2006;59:15762. 28. Kazzi SN, Kim UO, Quasney MW, Buhimschi I. Polymorphism of tumor necrosis factor-alpha and risk and severity of bronchopulmonary dysplasia among very low birth weight infants. Pediatrics 2004;114:e2438. 29. Przemko K. Miroslaw BM, Zoa M. Tomasz T, Magdalena L, Jacek PJ, Genetic risk factors of bronchopulmonary dysplasia Pediatr Res 2008;64(6):6828. 30. Chauhan M, Bombell S, McGuire W. Tumour necrosis factor (308A) polymorphism in very preterm infants with broncho-pulmonary dysplasia: meta-analysis. Arch Dis Child Fetal Neonatal Ed 2009;94:F2579. 31. Bhandari V, Bizzarro MJ, Shetty A, et al. Familial and genetic susceptibility to major neonatal morbidities in preterm twins. Pediatrics 2006;117:19016. 32. Cohen J, Van Marter LJ, Sun Y, Allred E, Leviton A, Kohane IS. Perturbation of gene expression of the chromatin remodeling pathway in premature newborns at risk for bronchopulmonary dysplasia. Genome Biol 2007;8:R210.

356

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357 62. Torday JS, Rehan VK. Deconvoluting lung evolution using functional/ comparative genomics. Am J Respir Cell Mol Biol 2004;31:812. 63. Rehan V, Torday J. Hyperoxia augments pulmonary lipobroblast-to-myobroblast transdifferentiation. Cell Biochem Biophys 2003;38:23950. 64. Torday JS, Rehan VK. Up-regulation of fetal rat lung parathyroid hormonerelated protein gene regulatory network down-regulates the Sonic Hedgehog/ Wnt/betacatenin gene regulatory network. Pediatr Res 2006;60:3828. 65. Rehan VK, Torday JS. Lower parathyroid hormone-related protein content of tracheal aspirates in very low birth weight infants who develop bronchopulmonary dysplasia. Pediatr Res 2006;60:21620. 66. Cerny L, Torday JS, Rehan VK. Prevention and treatment of bronchopulmonary dysplasia: contemporary status and future outlook. Lung 2008;186:7589. 67. deMello DE, Reid LM. Embryonic and early fetal development of human lung vasculature and its functional implications. Pediatr Dev Pathol 2000;3:43949. 68. deMello DE, Sawyer D, Galvin N, Reid LM. Early fetal development of lung vasculature. Am J Respir Cell Mol Biol 1997;16:56881. 69. Hislop AA. Airway and blood vessel interaction during lung development. J Anat 2002;201:32534. 70. Hall SM, Hislop AA, Pierce CM, Haworth SG. Prenatal origins of human intrapulmonary arteries: formation and smooth muscle maturation. Am J Respir Cell Mol Biol 2000;23:194203. 71. Hall SM, Hislop AA, Haworth SG. Origin, differentiation, and maturation of human pulmonary veins. Am J Respir Cell Mol Biol 2002;26:33340. 72. Parera MC, van Dooren M, van Kempen M, et al. Distal angiogenesis: a new concept for lung vascular morphogenesis. Am J Physiol Lung Cell Mol Physiol 2005;288:L1419. 73. Stenmark KR, Abman SH. Lung vascular development: implications for the pathogenesis of bronchopulmonary dysplasia. Annu Rev Physiol 2005;67:62361. 74. Suri C, Jones PF, Patan S, et al. Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell 1996;87:117180. 75. Kalinichenko VV, Gusarova GA, Kim IM, et al. Foxf1 haploinsufciency reduces Notch-2 signaling during mouse lung development. Am J Physiol Lung Cell Mol Physiol 2004;286:L52130. 76. Thebaud B. Angiogenesis in lung development, injury and repair: implications for chronic lung disease of prematurity. Neonatology 2007;91:2917. 77. Bourbon J, Boucherat O, Chailley-Heu B, Delacourt C. Control mechanisms of lung alveolar development and their disorders in bronchopulmonary dysplasia. Pediatr Res 2005;57(5 Pt 2):38R46. 78. Thebaud B, Abman SH. Bronchopulmonary dysplasia: where have all the vessels gone? Roles of angiogenic growth factors in chronic lung disease. Am J Respir Crit Care Med 2007;175:97885. 79. Akeson AL, Cameron JE, Le Cras TD, Whitsett JA, Greenberg JM. Vascular endothelial growth factor-A induces prenatal neovascularization and alters bronchial development in mice. Pediatr Res 2005;57:828. 80. Jakkula M, Le Cras TD, Gebb S, et al. Inhibition of angiogenesis decreases alveolarization in the developing rat lung. Am J Physiol Lung Cell Mol Physiol 2000;279:L6007. 81. Le Cras TD, Markham NE, Tuder RM, Voelkel NF, Abman SH. Treatment of newborn rats with a VEGF receptor inhibitor causes pulmonary hypertension and abnormal lung structure. Am J Physiol Lung Cell Mol Physiol 2002;283:L55562. 82. McGrath-Morrow S, Cho C, Molls R, et al. VEGF receptor 2 blockade leads to renal cyst formation in mice. Kidney Int 2006;69:17418. 83. Tang JR, Markham NE, Lin YJ, et al. Inhaled nitric oxide attenuates pulmonary hypertension and improves lung growth in infant rats after neonatal treatment with a VEGF receptor inhibitor. Am J Physiol Lung Cell Mol Physiol 2004;287:L34451. 84. Kasahara Y, Tuder RM, Taraseviciene-Stewart L, et al. Inhibition of VEGF receptors causes lung cell apoptosis and emphysema. J Clin Invest 2000;106:13119. 85. Tang K, Rossiter HB, Wagner PD, Breen EC. Lung-targeted VEGF inactivation leads to an emphysema phenotype in mice. J Appl Physiol 2004;97:155966 [discussion 49]. 86. DAngio CT, Maniscalco WM. The role of vascular growth factors in hyperoxiainduced injury to the developing lung. Front Biosci 2002;7:d160923. 87. Abman SH, Wolfe RR, Accurso FJ, Koops BL, Bowman CM, Wiggins Jr JW. Pulmonary vascular response to oxygen in infants with severe bronchopulmonary dysplasia. Pediatrics 1985;75:804. 88. Mourani PM, Ivy DD, Gao D, Abman SH. Pulmonary vascular effects of inhaled nitric oxide and oxygen tension in bronchopulmonary dysplasia. Am J Respir Crit Care Med 2004;170:100613. 89. Bhatt AJ, Pryhuber GS, Huyck H, Watkins RH, Metlay LA, Maniscalco WM. Disrupted pulmonary vasculature and decreased vascular endothelial growth factor, Flt-1, and TIE-2 in human infants dying with bronchopulmonary dysplasia. Am J Respir Crit Care Med 2001;164(10 Pt 1):197180. 90. Lassus P, Turanlahti M, Heikkila P, et al. Pulmonary vascular endothelial growth factor and Flt-1 in fetuses, in acute and chronic lung disease, and in persistent pulmonary hypertension of the newborn. Am J Respir Crit Care Med 2001;164(10 Pt 1):19817. 91. Klekamp JG, Jarzecka K, Perkett EA. Exposure to hyperoxia decreases the expression of vascular endothelial growth factor and its receptors in adult rat lungs. Am J Pathol 1999;154:82331. 92. Maniscalco WM, Watkins RH, DAngio CT, Ryan RM. Hyperoxic injury decreases alveolar epithelial cell expression of vascular endothelial growth factor (VEGF) in neonatal rabbit lung. Am J Respir Cell Mol Biol 1997;16:55767.

33. Albertine K, Amundsen S, Metcalfe D, Wint A, et al. Histone acetylation in the lung is affected by ventilation mode in preterm lambs. Publication 3060.2. Pediatric Academic Societies Annual Meeting, Honolulu, 3 May 2008. 34. Bernirschke K. Abnormalities of the human placenta. NeoReviews [online] 2006;6. 35. Gomez R, Romero R, Ghezzi F, Yoon BH, Mazor M, Berry SM. The fetal inammatory response syndrome. Am J Obstet Gynecol 1998;179:194202. 36. Watterberg KL, Demers LM, Scott SM, Murphy S. Chorioamnionitis and early lung inammation in infants in whom bronchopulmonary dysplasia develops. Pediatrics 1996;97:2105. 37. Yoon BH, Jun JK, Romero R, et al. Amniotic uid inammatory cytokines (interleukin-6, interleukin-1beta, and tumor necrosis factor-alpha), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol 1997;177:1926. 38. Nelson KB, Dambrosia JM, Grether JK, Phillips TM. Neonatal cytokines and coagulation factors in children with cerebral palsy. Ann Neurol 1998;44:66575. 39. Van Marter LJ, Dammann O, Allred EN, et al. Chorioamnionitis, mechanical ventilation, and postnatal sepsis as modulators of chronic lung disease in preterm infants. J Pediatr 2002;140:1716. 40. Redline RW, Wilson-Costello D, Hack M. Placental and other perinatal risk factors for chronic lung disease in very low birth weight infants. Pediatr Res 2002;52:7139. 41. Kaukola T, Tuimala J, Herva R, Kingsmore S, Hallman M. Cord immunoproteins as predictors of respiratory outcome in preterm infants. Am J Obstet Gynecol 2009;200(1):100.e18. 42. Moss TJ, Newnham JP, Willett KE, Kramer BW, Jobe AH, Ikegami M. Early gestational intra-amniotic endotoxin: lung function, surfactant, and morphometry. Am J Respir Crit Care Med 2002;165:80511. 43. Bose CL, Dammann CE, Laughon MM. Bronchopulmonary dysplasia and inammatory biomarkers in the premature neonate. Arch Dis Child Fetal Neonatal Ed 2008;93:F45561. 44. Merritt TA, Cochrane CG, Holcomb K, et al. Elastase and alpha 1-proteinase inhibitor activity in tracheal aspirates during respiratory distress syndrome. Role of inammation in the pathogenesis of bronchopulmonary dysplasia. J Clin Invest 1983;72:65666. 45. Groneck P, Gotze-Speer B, Oppermann M, Eiffert H, Speer CP. Association of pulmonary inammation and increased microvascular permeability during the development of bronchopulmonary dysplasia: a sequential analysis of inammatory mediators in respiratory uids of high-risk preterm neonates. Pediatrics 1994;93:7128. 46. Bry K, Lappalainen U, Hallman M. Intraamniotic interleukin-1 accelerates surfactant protein synthesis in fetal rabbits and improves lung stability after premature birth. J Clin Invest 1997;99:29929. 47. Goldenberg RL, Andrews WW, Goepfert AR, et al. The Alabama Preterm Birth Study: umbilical cord blood Ureaplasma urealyticum and Mycoplasma hominis cultures in very preterm newborn infants. Am J Obstet Gynecol 2008;198. 43e15. 48. Kundsin RB, Driscoll SG, Monson RR, Yeh C, Biano SA, Cochran WD. Association of Ureaplasma urealyticum in the placenta with perinatal morbidity and mortality. N Engl J Med 1984;310:9415. 49. Kallapur SG, Bachurski CJ, Le Cras TD, Joshi SN, Ikegami M, Jobe AH. Vascular changes after intra-amniotic endotoxin in preterm lamb lungs. Am J Physiol Lung Cell Mol Physiol 2004;287:L117885. 50. Paananen R, Husa AK, Vuolteenaho R, Herva R, Kaukola T, Hallman M. Blood cytokines during the perinatal period in very preterm infants: relationship of inammatory response and bronchopulmonary dysplasia. J Pediatr 2009;154(1):3943.e3. 51. Hodgman JE. Relationship between WilsonMikity syndrome and the new bronchopulmonary dysplasia. Pediatrics 2003;112(6 Pt 1):14145. 52. Hoepker A, Seear M, Petrocheilou A, et al. WilsonMikity syndrome: updated diagnostic criteria based on nine cases and a review of the literature. Pediatr Pulmonol 2008;43:100412. 53. Chess PR, DAngio CT, Pryhuber GS, Maniscalco WM. Pathogenesis of bronchopulmonary dysplasia. Semin Perinatol 2006;30:1718. 54. Harding R, Hooper SB. Regulation of lung expansion and lung growth before birth. J Appl Physiol 1996;81:20924. 55. Allison BJ, Crossley KJ, Flecknoe SJ, et al. Ventilation of the very immature lung in utero induces injury and BPD-like changes in lung structure in fetal sheep. Pediatr Res; 2008 Jun 11. 56. Ammari A, Suri M, Milisavljevic V, et al. Variables associated with the early failure of nasal CPAP in very low birth weight infants. J Pediatr 2005;147:3417. 57. Jobe AH, Kramer BW, Moss TJ, Newnham JP, Ikegami M. Decreased indicators of lung injury with continuous positive expiratory pressure in preterm lambs. Pediatr Res 2002;52:38792. 58. Pillow JJ, Hillman N, Moss TJ, et al. Bubble continuous positive airway pressure enhances lung volume and gas exchange in preterm lambs. Am J Respir Crit Care Med 2007;176:639. 59. Polglase GR, Hillman NH, Ball MK, et al. Lung and systemic inammation in preterm lambs on CPAP or conventional ventilation. Pediatr Res; 2008 Aug 13. 60. Morley CJ, Davis PG, Doyle LW, Brion LP, Hascoet JM, Carlin JB. Nasal CPAP or intubation at birth for very preterm infants. N Engl J Med 2008;358:7008. 61. Reyburn B, Li M, Metcalfe DB, et al. Nasal ventilation alters mesenchymal cell turnover and improves alveolarization in preterm lambs. Am J Respir Crit Care Med 2008;178:40718.

T.A. Merritt et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 345357 93. De Paepe ME, Patel C, Tsai A, Gundavarapu S, Mao Q. Endoglin (CD105) upregulation in pulmonary microvasculature of ventilated preterm infants. Am J Respir Crit Care Med 2008;178:1807. 94. Abdollahi A, Hahnfeldt P, Maercker C, et al. Endostatins antiangiogenic signaling network. Mol Cell 2004;13:64963. 95. Janer J, Andersson S, Haglund C, Lassus P. Pulmonary endostatin perinatally and in lung injury of the newborn infant. Pediatrics 2007;119:e2416. 96. Kunig AM, Balasubramaniam V, Markham NE, et al. Recombinant human VEGF treatment enhances alveolarization after hyperoxic lung injury in neonatal rats. Am J Physiol Lung Cell Mol Physiol 2005;289:L52935. 97. Balasubramaniam V, Maxey AM, Fouty BW, Abman SH. Nitric oxide augments fetal pulmonary artery endothelial cell angiogenesis in vitro. Am J Physiol Lung Cell Mol Physiol 2006;290:L11116. 98. Balasubramaniam V, Maxey AM, Morgan DB, Markham NE, Abman SH. Inhaled NO restores lung structure in eNOS-decient mice recovering from neonatal hypoxia. Am J Physiol Lung Cell Mol Physiol 2006;291:L11927. 99. Northway Jr WH. Bronchopulmonary dysplasia: twenty-ve years later. Pediatrics 1992;89(5 Pt 1):96973. 100. Albertine KH, Plopper CG. DNA oxidation or apoptosis: will the real culprit of DNA damage in hyperoxic lung injury please stand up? Am J Respir Cell Mol Biol 2002;26:3813. 101. Koli K, Myllarniemi M, Keski-Oja J, Kinnula VL. Transforming growth factorbeta activation in the lung: focus on brosis and reactive oxygen species. Antioxid Redox Signal 2008;10:33342. 102. Rozycki HJ, Comber PG, Huff TF. Cytokines and oxygen radicals after hyperoxia in preterm and term alveolar macrophages. Am J Physiol Lung Cell Mol Physiol 2002;282:L12228. 103. Welty SE. Is oxidant stress in the causal pathway to bronchopulmonary dysplasia? NeoReviews [online] 2000;1. 104. Coalson JJ, Winter V, deLemos RA. Decreased alveolarization in baboon survivors with bronchopulmonary dysplasia. Am J Respir Crit Care Med 1995;152:6406. 105. Wilborn AM, Evers LB, Canada AT. Oxygen toxicity to the developing lung of the mouse: role of reactive oxygen species. Pediatr Res 1996;40:22532. 106. Nadeau K, Jankov RP, Tanswell AK, Sweezey NB, Kaplan F. Lgl1 is suppressed in oxygen toxicity animal models of bronchopulmonary dysplasia and normalizes during recovery in air. Pediatr Res 2006;59:38995.

357

107. Maniscalco WM, Watkins RH, Roper JM, Staversky R, OReilly MA. Hyperoxic ventilated premature baboons have increased p53, oxidant DNA damage and decreased VEGF expression. Pediatr Res 2005;58:54956. 108. van Tuyl M, Liu J, Wang J, Kuliszewski M, Tibboel D, Post M. Role of oxygen and vascular development in epithelial branching morphogenesis of the developing mouse lung. Am J Physiol Lung Cell Mol Physiol 2005;288:L16778. 109. Groenman F, Rutter M, Caniggia I, Tibboel D, Post M. Hypoxia-inducible factors in the rst trimester human lung. J Histochem Cytochem 2007;55:35563. 110. Gebb SA, Fox K, Vaughn J, McKean D, Jones PL. Fetal oxygen tension promotes tenascin-C-dependent lung branching morphogenesis. Dev Dyn 2005;234:110. 111. Park MS, Rieger-Fackeldey E, Schanbacher BL, et al. Altered expressions of broblast growth factor receptors and alveolarization in neonatal mice exposed to 85% oxygen. Pediatr Res 2007;62:6527. 112. Nakanishi H, Sugiura T, Streisand JB, Lonning SM, Roberts Jr JD. TGF-betaneutralizing antibodies improve pulmonary alveologenesis and vasculogenesis in the injured newborn lung. Am J Physiol Lung Cell Mol Physiol 2007;293:L15161. 113. Ward PA. Oxygen radicals, cytokines, adhesion molecules, and lung injury. Environ Health Perspect 1994;102(Suppl. 10):136. 114. Yi M, Jankov RP, Belcastro R, et al. Opposing effects of 60% oxygen and neutrophil inux on alveologenesis in the neonatal rat. Am J Respir Crit Care Med 2004;170:118896. 115. Hay Jr WW, Bell EF. Oxygen therapy, oxygen toxicity, and the STOP-ROP trial. Pediatrics 2000;105:4245. 116. Askie LM, Henderson-Smart DJ, Irwig L, Simpson JM. Oxygen-saturation targets and outcomes in extremely preterm infants. N Engl J Med 2003;349:95967. 117. Thomas W, Speer CP. Nonventilatory strategies for prevention and treatment of bronchopulmonary dysplasia what is the evidence? Neonatology 2008;94:1509. 118. Davis JM, Parad RB, Michele T, Allred E, Price A, Rosenfeld W. Pulmonary outcome at 1 year corrected age in premature infants treated at birth with recombinant human CuZn superoxide dismutase. Pediatrics 2003;111:46976. 119. Tin W. Oxygen therpay: 50 years of uncertainty. Pediatrics 2002;110:6156. 120. Lefkowitz W, Rosenberg SH. Bronchopulmonary dysplasia: pathway from disease to long-term outcome. J Perinatol 2008;28:83740.

Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Ventilatory management and bronchopulmonary dysplasia in preterm infants


Samir Gupta a, Sunil K. Sinha b, *, Steven M. Donn c
a

Department of Neonatal Paediatrics, University Hospital of North Tees, Stockton-on-Tees, UK Department of Neonatal Paediatrics, James Cook University Hospital, University of Durham, Middlesbrough, UK c Department of Pediatrics, Division of NeonatalPerinatal Medicine, C.S. Mott Childrens Hospital, University of Michigan Health System, Ann Arbor, Michigan, USA
b

s u m m a r y
Keywords: Bronchopulmonary dysplasia Mechanical ventilation Prematurity Respiratory distress syndrome

Improvements in antenatal and neonatal care have resulted in increased survival of very preterm infants. However, the incidence of bronchopulmonary dysplasia (BPD) has not changed, probably as a consequence of a demographic shift. The underlying pathophysiology of BPD appears to differ for the current population of preterm infants compared to that described by Northway et al., and management strategies should be targeted to limit ventilator-induced lung injury. Non-invasive respiratory support techniques are currently under evaluation, but results of the trials have thus far failed to show a reduction in BPD. This review will focus upon various ventilation modalities for preventing and managing bronchopulmonary dysplasia. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Classic bronchopulmonary dysplasia (BPD) described by Northway et al. in 1967 was seen in relatively mature babies, who were ventilated at high pressures, with high fractional concentration of inspired oxygen, resulting in lung overination, cystic emphysema, and brosis.1 By contrast, BPD is now seen primarily in very preterm newborns weighing <1000 g, who are born at 2426 weeks of gestation. The histopathologic lesions of the old BPD have now been replaced by a new BPD with the large, simplied alveolar structures, variable interstitial cellularity and/or broproliferation.2 The clinical picture is also different. Todays premature babies developing BPD may initially have modest ventilatory and oxygen needs and demonstrate a different radiographic picture, which shows diffuse haziness and a ne, lacy pattern (Fig. 1). These differences can be better understood with the realisation that the lungs of infants born at 2426 weeks of gestation are in the early phases of lung development (canalicular and saccular stages), and that alveolar and capillary development are inhibited following exposure to the noxious effects of mechanical ventilation, oxygen exposure, and inammation. These differences become important in planning ventilatory strategies to prevent or treat BPD in preterm infants.

2. Can ventilatory strategies prevent BPD? BPD is multifactorial and an understanding of the pulmonary injury sequence may enable the development of strategies to circumvent this cascade. There are two prominent pathways leading to BPD (Fig. 2). The rst is intrinsic and is related to a developmental arrest of the lung resulting in diminished alveolarisation. This results in inadequate surface area for gas exchange and impaired lung function, requiring the initiation of chronic ventilation and subsequent ventilator-induced lung injury (VILI). Other factors which may injure immature lungs include surfactant deciency and the need to deal with a very compliant chest wall. Extrinsic factors also inhibit alveolarisation and lung growth, and they include intrauterine cytokine exposure, antenatal and postnatal glucocorticoid treatment, insufcient nutrition, lung and systemic infections, and exposure to high concentrations of oxygen. 2.1. Understanding ventilator-induced lung injury Although a key component in the pulmonary injury sequence, VILI itself is multifactorial (Box 1). Volutrauma refers to injury related to overdistension or stretching of the lung units (alveoli and smaller airways) by delivering too much gas (tidal volume). Atelectotrauma refers to the damage caused by insufcient tidal volumes (recruitment and derecruitment of alveoli) associated with repetitive opening and closing of lung units. Biotrauma is a collective term to describe the adverse effects of infection and inammation. Rheotrauma refers to damage caused by inappropriate airway ow.

* Corresponding author. Address: Department of Neonatal Paediatrics, James Cook University Hospital, University of Durham, Middlesbrough, TS4 3BW UK. Tel./fax: 44 (0) 1642 854874. E-mail address: Sunil.sinha@stees.nhs.uk (S.K. Sinha). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.011

368

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

Fig. 1. Radiographic picture of bronchopulmonary dysplasia (BPD). (a) Old BPD: areas of atelectasis and hyperination. (b) New BPD: diffuse opacication of lung elds.

If the ow is excessive, it may cause inadvertent positive end-expiratory pressure (PEEP), turbulence and ineffective gas exchange, and lung overination. On the other hand, if the ow is inadequate, it may lead to air hunger (ow starvation) and increased work of breathing. An understanding of the pathophysiology of VILI is essential to formulating ventilatory strategies aimed at reducing or preventing BPD in very small infants.3 2.2. Non-invasive forms of respiratory support There has been a recent upsurge in the use of non-invasive respiratory support, using continuous positive airway pressure (CPAP) or nasal intermittent positive pressure ventilation (NIPPV). CPAP is a form of distending pressure provided through a nasal interface. NIPPV provides intermittent positive pressure ventilation in addition to continuous distending pressure. The proponents of non-invasive ventilation claim that the absence of an endotracheal tube reduces the risk of trauma to the airways, reduces the risk of infection, and causes less acute and chronic lung damage.4 However, the scientic evidence for the efcacy and safety of both

CPAP and NIPPV may vary depending upon whether they are used as a primary treatment for respiratory failure, or as an adjunct following extubation. The term non-invasive is also a misnomer, as both methods still deliver pressure that is supraphysiologic. 2.3. Continuous positive airway pressure CPAP supports the breathing of premature infants in a number of ways and can be delivered by a variety of devices and interfaces. CPAP devices can be categorised by the ow characteristics into continuous ow systems, such as bubble CPAP and ventilatorderived CPAP, and variable ow systems such as the infant ow driver and Benveniste (gas jet) valve CPAP. Two randomised trials have addressed the question of whether CPAP commenced soon after birth reduces mortality and morbidity of very preterm infants.5,6 Meta-analysis of these trials showed no difference in the rates of death, BPD, subsequent endotracheal intubation, or intraventricular haemorrhage.7 In a recent randomised controlled trial, 610 infants born at 2528 weeks of gestation were randomised to receive either CPAP or intubation within 5 min of delivery. Infants

Fig. 2. The effect of ventilator-induced lung injury and other factors on lung development, and their relationship to chronic lung disease. Reproduced with permission from Attar and Donn.3

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

369

Box 1. Components of ventilator-induced lung injury      Barotrauma Volutrauma Atelectotrauma Biotrauma Rheotrauma

were eligible if they were breathing spontaneously but still required respiratory support. Babies randomised to CPAP had a signicantly lower rate of the combined outcome, death or need for supplementary oxygen at 28 days (OR: 0.63; 0.460.88), but this difference disappeared at 36 weeks postmenstrual age. Moreover, babies who were intubated in the delivery room had a signicantly lower rate of pneumothorax than those randomised to CPAP (3% vs 9.1%, P 0.001).8 Thus, the evidence so far suggests that although CPAP may be an acceptable alternative to endotracheal intubation in some low birth weight babies, there is no convincing evidence that it reduces the incidence of BPD. The challenge remains to identify strategies that preserve the benet of CPAP and reduce the rate of complications such as pneumothorax. The benets of continuous CPAP with prophylactic exogenous surfactant treatment have been explored, but methodological constraints of these studies preclude any meaningful conclusion. Two large studies are currently under way [Surfactant Positive Airway Pressure and Pulse Oximetry Trial (SUPPORT), and National Institute of Child and Human Health Neonatal Research Network trials] comparing nasal CPAP, mechanical ventilation, and surfactant administration followed by early extubation to nasal CPAP in extremely premature babies. The results of these ongoing trials may help to shed some light on this controversy. At the present time, we advocate that standard therapy should be the administration of surfactant followed by mechanical ventilation for babies <28 weeks of gestation. The claim that non-invasive ventilation protects against BPD has also been challenged in a recent animal study by Polglase et al., who failed to demonstrate any advantages of non-invasive respiratory support over mechanical ventilation.9 In extremely preterm babies it seems advisable to administer early surfactant and attempt to limit VILI by aggressively weaning mechanical support (described below). 2.4. Nasal intermittent positive pressure ventilation Despite increasing experience, 4560% of extremely low birth weight infants fail early CPAP therapy for treatment of respiratory distress syndrome (RDS) and require subsequent intubation and ventilation. Even among VLBW infants extubated to CPAP, 2540% still require reintubation.10,11 Efforts to improve these failure rates have prompted the use of NIPPV. The mechanism of action of NIPPV, however, remains uncertain. It is not clear whether mechanical inations during NIPPV are actually transmitted to the lungs.10 Synchronising NIPPV inations with the infants own breath is theoretically advantageous. Non-synchronised NIPPV may deliver high pressure during spontaneous exhalation, increasing the risk of elevated upper airway pressure and subsequent pneumothorax. No trials, however, have compared synchronised with non-synchronised NIPPV, or examined the accuracy of synchronisation devices in NIPPV. Moreover, there is no consensus regarding the appropriate pressure levels, rate, or inspiratory time during NIPPV. In a recent survey, widely varying settings were utilised during NIPPV: PIP 720 cmH2O; PEEP 49 cmH2O; Ti 0.30.5 s; rate 1060 per min.12 A number of trials have compared NIPPV to CPAP following extubation and have shown a signicant reduction in extubation

failure using NIPPV. Other investigations compared NIPPV as a primary treatment of RDS and showed reduction in BPD rates with NIPPV (2%), compared to NCPAP (17%) (P 0.03).13 However, the sample sizes are small and results should be interpreted with caution. In summary, although non-invasive ventilation is being increasingly used both as a primary strategy for respiratory support or to facilitate extubation, there is no conclusive evidence that it prevents BPD.14 The use of any form of distending pressure is likely to be associated with an increased risk of pneumothorax15 and none of the reported trials of non-invasive respiratory support conrm any benecial effects on long-term pulmonary and neurodevelopmental outcomes. 2.5. Post-extubation CPAP and NIPPV Whereas there is no consensus regarding the benets of CPAP and NIPPV as a primary modality to treat RDS, the scientic evidence for their efcacy in facilitating extubation and reducing the duration of mechanical ventilation is well established.11 However, certain conditions apply. For example, the effectiveness of CPAP depends upon a number of factors, such as the level of pressure used (benecial with pressure 5 cmH2O), duration of prior mechanical ventilation (effective if duration is 14 days), and the nasal interface (short binasal prongs provide the lowest resistance). The CPAP devices also inuence performance. This was recently evaluated in a controlled trial, which showed bubble CPAP to have some advantage over the infant ow driver in relation to some short-term outcome measures.16 NIPPV, on the whole, seem to be better in preventing extubation failure compared with CPAP but requires further evaluation. Moreover, despite the conrmed advantages of CPAP and NIPPV in reducing the duration of mechanical ventilation, none of the meta-analyses or individual ransomised controlled trials has shown that they reduce BPD.11 2.6. Conventional mechanical ventilation (CMV) For nearly three decades, CMV was the only method of ventilating preterm infants. Using time-cycled, pressure-limited ventilation (TCPLV) in an intermittent mandatory ventilation mode, this form of ventilation was relatively easy to use and resulted in consistent delivery of pressure, although delivered tidal volumes varied according to lung mechanics. The judicious use of sophisticated respiratory support has been the centre of attention for the past two decades. With the implementation of microprocessorbased technology, ventilators can now be adjusted to suit the individual needs of the most complex patients. The availability of real-time graphic monitoring on ventilators enables clinicians to visualise respiratory support on a breath-to-breath basis. Despite the advances in technology, there is no clear evidence for an optimal ventilator strategy that crosses different gestational ages or types of pulmonary diseases. However, recent clinical investigation suggests that we may be on the right track. 2.7. Volume-targeted ventilation Volume-targeted modalities are relatively new to neonatal intensive care and represent a departure from traditional TCPLV by focusing on tidal volume and allowing pressure to vary. Although there are only a few published randomised controlled clinical trials, thus far the evidence is highly encouraging. The consistency of tidal volume delivery during volume-controlled ventilation (VCV) in the face of varying lung compliance and the auto-weaning of airway pressure may be clinically advantageous, especially in conditions in which lung compliance can change rapidly, such as after surfactant administration. A meta-analysis of volume-targeted ventilation

370

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

reported a reduction in the total duration of ventilation, severe intraventricular haemorrhage and pneumothorax, and a strong trend towards a reduction in the incidence BPD.17 Stability of tidal volume delivery may account for this, especially in extremely low birth weight infants, who are most at risk for sustaining complications associated with mechanical ventilation.18 Volume-targeted ventilation can be provided in two ways. Volume-controlled ventilation (VCV) targets the delivery of a clinician-set tidal volume. This is delivered irrespective of how much pressure is required (although for safety reasons this can be limited). Hybrid forms, such as Volume Guarantee, Pressure Regulated Volume Control, and Volume Assured Pressure Support, are essentially pressure-targeted modalities but aim to deliver a tidal volume within a set range using computer-controlled feedback and in some cases a breath-averaging technique. We recently reported the results of a randomised controlled trial of VCV in a population of preterm babies with RDS, born between 24 and 31 weeks of gestation and weighing 6001500 g. There were trends towards faster weaning, reduction in the duration of ventilation, decreased BPD, and improved survival favouring VCV.19 As part of the original study design, a longer-term follow-up of these babies was conducted to examine clinically relevant outcomes beyond discharge. The primary aim of the follow-up study was to assess survival and respiratory and gross neurodevelopmental status at 2 years corrected age. The respiratory outcomes of interest were the occurrence and frequency of cough and wheeze, the use of inhaled medications for respiratory dysfunction, and hospital readmissions for respiratory illnesses. The study showed that the trend towards less BPD reported at 36 weeks PMA in the original trial had a clinical correlate at follow-up, with fewer children requiring pharmacotherapy.20 2.8. Patient-triggered ventilation One of the major problems with intermittent mandatory ventilation (IMV) is patient-ventilator dyssynchrony. Babies can exhale against incoming pressure, may exhibit inefcient gas exchange, gas trapping and air leaks, and the need for higher levels of support.21 During the 1990s, the capability of detecting spontaneous respiratory activity led to the development of patient-triggered ventilation (PTV) in neonatal respiratory care. Modes of ventilation utilising PTV include synchronised intermittent mandatory ventilation (SIMV), assist/control (A/C) ventilation, and pressure support ventilation (PSV). Transducers detect some measure of spontaneous respiratory effort, such as changes in airway ow, and respond with a mechanical breath timed to begin almost immediately. Early clinical trials suggested a trend towards a reduction in BPD, but these studies were underpowered. One large, multicentre open trial did not show any differences in the incidence of BPD, but methodological and design aws, and signicant user inexperience, limit the interpretation of the ndings.22 Although short-term benets of PTV have been demonstrated,23 further studies are required to address its impact on BPD. 2.9. Synchronised intermittent mandatory ventilation In SIMV, the clinician selects a mandatory rate of ventilation. Each time the baby is scheduled to receive a breath, the ventilator waits briey for the infant to initiate a breath. If the baby breathes within this timing window, a mechanical breath will be initiated synchronously with the spontaneous breath. If the baby fails to breathe, the ventilator will cycle at the set rate. In between mechanical breaths, spontaneous breathing is supported, only by PEEP, resulting in a higher work of breathing than with either A/C or PSV.24

2.10. Assist/control ventilation In A/C all spontaneous breaths that exceed the trigger threshold result in the delivery of a mechanical breath that is synchronised to the onset of inspiration. If the baby is apnoeic or fails to exceed the trigger threshold, the ventilator will cycle at the control (back-up) rate. Management of infants ventilated by A/C is different from that of IMV or SIMV. Provided that the patient is breathing above the control rate, reductions in the control rate will have no effect. Weaning is accomplished by decreasing the peak inspiratory pressure rst, which should be benecial in reducing both barotrauma and volutrauma.25 However, reduction in BPD has not been uniformly demonstrated.

2.11. Pressure support ventilation PSV is primarily a weaning mode of ventilation in which spontaneous breaths receive an inspiratory pressure boost to decrease the imposed work of breathing and unload the respiratory musculature. A spontaneous breath triggers the delivery of a timeand pressure-limited breath, which is ow-cycled (see below). Breaths may be fully (provide a full tidal volume) or partially supported. PSV is usually applied in conjunction with SIMV, although it may be utilised alone if the baby has reliable respiratory drive and can easily exceed the trigger threshold. Another feature of PSV is variable inspiratory ow, which is proportional to patient effort, and helps to overcome increased resistance.26 A cross-over trial by Gupta et al. reported that PSV increases total minute ventilation and stabilises breathing in proportion to the applied level of pressure support.27 This may be advantageous and provide a useful ventilation strategy during weaning in preterm infants to allow a more consistent transfer of the work of breathing from the ventilator to the infant.28

2.12. Flow-cycling Cycling refers to the mechanism responsible for transitioning from inspiration to expiration and from expiration to inspiration. In the past, this has been accomplished by time; inspiration ended when the exhalation valve opened after a preset interval. The new technology offers ow-cycling, where inspiration ends when inspiratory ow decreases to a certain percentage of peak inspiratory ow. This is interpreted by the ventilator as a sign that the baby is about to end his own inspiratory effort and thus functions as an expiratory trigger. The advantages of ow cycling are that the baby is able to set the inspiratory time (a breath starts and ends at the babys preference) and, at rapid rates, an inverse inspiratory:expiratory ratio cannot occur. The mechanical breath will be terminated before the baby can initiate another breath, rather than persisting for a xed time interval, as would occur with timecycling (Fig. 3).26 Flow-cycling is incorporated in PSV. It allows the baby to have control over the ventilator rate and inspiratory time, and it also enables full synchronisation of the respiratory cycle.

2.13. High frequency ventilation First introduced into neonatal practice in the early 1980s, high frequency ventilation (HFV) uses extremely small tidal volumes delivered at rapid rates to effect gas exchange at lower alveolar pressures than CMV. There are two primary forms of HFV: high frequency jet ventilation (HFJV), and high frequency oscillatory ventilation (HFOV), as well as hybrid forms.

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

371

RDS.32 Two recently completed trials found differing results. The HIFO trial of Courtney et al. found a very slight reduction in BPD in babies receiving HFOV, although this could have reected the fact that the control group was managed in SIMV, which is not the ideal mode for acutely ill infants.33 The UKOS study found no difference in the incidence of BPD among infants receiving HFOV or conventional ventilation.34 Currently, there is insufcient evidence to recommend HFOV.35 2.14. Permissive hypercapnia The rationale for permissive hypercapnia, using a low lung volume strategy, is that it may decrease volutrauma and lung injury, lessen the duration of mechanical ventilation, reduce alveolar ventilation and the complications of hypocapnia (especially reduced cerebral blood ow), and increase oxygen unloading at the tissue level (Bohr effect).36 The strategy of ventilating infants at a higher PaCO2 level, termed permissive hypercapnia, is based upon the retrospective observation of Kraybill et al. in 1989. In a multicentre analysis of 235 preterm infants, these investigators demonstrated a higher incidence of BPD in the infants with the lowest PaCO2 on the second and fourth postnatal days.36 Two prospective, randomised, controlled trials of permissive hypercapnia have been conducted.37 Although both studies demonstrated a statistically signicant reduction in the duration of ventilation and other secondary outcome measures, the incidence of BPD did not differ. 2.15. Monitoring of ventilated infants Advances in biomedical engineering have also substantially changed the way in which babies receiving mechanical ventilation can be monitored. Intermittent chest radiographs and blood gas sampling have been replaced by continuous monitoring of pulmonary mechanics and pulse oximetry or transcutaneous oxygen/carbon dioxide tensions. Real-time pulmonary mechanics monitoring has become a valuable adjunct to CMV.38 The same sensor technology utilised for PTV enables measurement of changes in airway pressure or ow, which can be converted to a volume signal. Determination of

Fig. 3. Pulmonary graphic scalar record showing gas trapping. The inspiration commences before the expiratory ow could reach the baseline, leading to gas trapping and inadvertent positive end-expiratory pressure.

2.13.1. High frequency jet ventilation HFJV uses rates of 240660 breaths/min, and inspiratory times are typically about 0.02 s. It is used in tandem with a conventional ventilator, which provides PEEP and optional sigh breaths. High velocity pulsations are injected into the proximal airway using a special connector. Ventilator adjustments are similar to those of CMV. The amplitude is set by adjusting peak pressure and PEEP.29 2.13.2. High frequency oscillatory ventilation HFOV differs from HFJV in that even smaller tidal volumes are used at rates of 815 Hz. Mean airway pressure is used as a continuous distending pressure to inate the lung to a static volume, and oscillations around this mean are used to effect gas exchange. Adjustments for oxygenation (via mean airway pressure) and ventilation (via amplitude) can be done independently of one another.30 Moreover, HFOV uses active exhalation, whereby gas is actively withdrawn from the lung during expiration. A number of clinical studies of high frequency ventilation have reported long-term outcomes, although BPD was not the primary objective in most cases. Keszler et al. reported a reduced incidence of BPD and the need for home oxygen in infants treated with HFJV compared with CMV for uncomplicated RDS.31 A number of recent clinical trials compared HFOV with CMV in preterm infants with

Fig. 4. Time-cycling vs ow-cycling. Note that in time-cycling there is a prolongation of a no ow state (shaded area) at the end of inspiration until the exhalation valve opens. With ow-cycling, inspiration ends when inspiratory ow has decreased to a small percentage of peak inspiratory ow. Note how the breath cycles directly into expiration at this point (arrow).

372

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

Table 1 Respiratory management options for bronchopulmonary dysplasia. Issue Increased resistance Strategy Variable inspiratory ow (pressure control or pressure support ventilation). Consider using higher PEEP to stent collapsing airways. Use largest comfortable ETT. Use higher PIP/PEEP. Consider high frequency ventilation. Avoid hyperination. Monitor tidal volumes. Consider volume-targeted ventilation Consider pharmacotherapy. Liberalisation of arterial oxygen tension. Re-evaluate gas exchange capability; allow higher carbon dioxide tension if pH is not deranged Try to optimise caloric and uid intake, but avoid excess carbohydrate or fat Comments Seen with airway injury and hyper-reactivity

Decreased compliance

Reects parenchymal lung injury

Heterogeneous lung disease Pulmonary hypertension Hypercapnia Nutrition

Pressure-targeted ow patterns may contribute to ventilation/perfusion mismatch Limited evidence; follow echocardiogram. Do not attempt to achieve normal gas exchange Non-nitrogen calories will contribute to carbon dioxide production and respiratory load

PEEP, positive end-expiratory pressure; ETT, endotracheal tube; PIP, peak inspiratory pressure.

tidal volume, minute ventilation, and breath-to-breath displays of ow, pressure, and volume waveforms are now standard techniques. Many devices also enable real-time assessment of pulmonary mechanics through displays of pressure-volume and ow-volume loops, and calculation of resistance and compliance. Detection of potentially adverse situations, such as gas trapping or hyperination, is now possible before these dangers become clinically apparent (Fig. 4).39 Pulmonary graphics are critical in the appropriate application of some of the newer ventilatory techniques, such as PSV. Tidal volume monitoring allows adjustments that can provide optimal lung expansion and selection of the optimal PEEP. Spontaneous minute ventilation monitoring has been shown to be useful in assessing the readiness for extubation in mechanically ventilated infants.40 Pulmonary mechanics monitoring permits the clinician to objectively assess the effect of pharmacologic agents with a narrow therapeutic index, such as corticosteroids, bronchodilators, or diuretics. Continuous pulse oximetry or blood gas analysis, using either indwelling intravascular sensors or transcutaneous electrodes, transforms ventilator management from an intermittent to a dynamic process. Maintaining oxygenation and ventilation within a desired range should help to minimise the deleterious effects of both oxygen toxicity and positive pressure. Techniques that adapt the mechanical ventilatory support and supplemental oxygen to the changing needs of the preterm infants are being developed in order to improve the stability of gas exchange, to minimise respiratory support.41 3. Ventilatory strategies for infants with established BPD The management of infants with established BPD has not been studied adequately, and the role of various ventilatory strategies for infants with the disorder is not clear. A suggested guideline for dealing with different lung issues is shown in Table 1. A number of other strategies can also be employed to reduce the duration of mechanical ventilation with a hope that earlier extubation might ameliorate the pulmonary injury sequence and thus reduce the severity of BPD. The scientic evidence in favour of such interventions varies; however, clinicians should always assess the level of evidence and risk:benet ratio before implementing them in routine practice. 3.1. Physiologic principles The infant with ventilator-dependent BPD presents many challenges to the clinician. Chronic respiratory insufciency may result in aberrant gas exchange, manifest by severe hypercarbia, with marked renal compensation. The inexperienced clinician often

attempts to make the baby conform to physiologic blood gases, and in doing so may inadvertently overventilate the baby and unwittingly contribute to VILI. It is perhaps wiser to view the chest radiograph in the context of, What can I expect from these lungs? and adjust ones expectations of blood gases accordingly. Alteration in lung mechanics is another key feature of BPD. Reactive airways may result in increased pulmonary resistance. This may be treated empirically by adjustments in ventilator strategy, such as increasing PEEP and/or airway ow, or by using a modality with variable inspiratory ow, such as pressure control or pressure support compared with traditional xed ow TCPLV. Lung compliance may also be abnormal. Ventilating the lung at an appropriate functional residual capacity, where incremental changes in pressure recruit the most lung volume, is often a challenge. Assessing compliance at different levels of PEEP may be helpful in this regard. Alterations in resistance and compliance will alter the respiratory time constant, and sufcient expiratory time to avoid gas trapping and inadvertent PEEP must be provided. Decisions about the best mode or modality to ventilate a baby with BPD are best made on an individual basis. Theoretically, if lung disease is homogeneous, a modality such as pressure control might be advantageous, whereas heterogeneous lung disease should respond better to VCV because peak pressure and peak volume delivery occur at the end of inspiration, helping to make gas delivery more uniform throughout the lung. This will require clinical investigation. Periodic assessment of cardiac function, looking for evidence of pulmonary hypertension, may also guide ones use of supplemental oxygen and establishment of appropriate ranges for oxygen saturation, although the optimal limits are yet to be established. All changes in strategy should be objectively assessed for response. 4. Conclusion BPD is a recognised sequel of preterm birth. With improving survival of infants at lower gestational ages, the prevalence is on the rise. Pathological features of BPD include alveolar maldevelopment, with or without areas of pulmonary brosis. Assisted ventilation, infection/inammation, oxygen administration, and uid overload are major identied risk factors in the evolution of BPD. The prevention of BPD needs a multifocal approach by decreasing VILI through utilisation of newer ventilatory strategies such as volume-controlled ventilation, and the use of non-invasive forms of respiratory support in selected preterm babies. Many other therapies are still investigational and potentially dangerous and need further evidence before they can be routinely recommended. Determining the optimal strategy for infants with ventilator-

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

373

dependent BPD is a work in progress, but a necessary one in view of the number of affected infants.

Practice points  Various modalities of ventilation are available, but efforts should be made to limit ventilator-induced lung injury by limiting the duration of ventilation and avoiding excessive volume and pressure.  Non-invasive respiratory support techniques are currently under investigation in extremely preterm infants and, until further data are available, conventional mechanical ventilation should be the primary strategy for these infants.  Volume-targeted ventilation appears to decrease the incidence of BPD and limiting volutrauma seems to be a logical approach.

Research directions  Evaluation of non-invasive respiratory support techniques.  Evaluation of pressure support ventilation for early and subsequent management of RDS.  Large, adequately powered randomised clinical trials to dene the roles of various devices (such as high frequency ventilation), modalities (such as patient-triggered or volume-targeted ventilation), and strategies (such as permissive hypercapnea) in limiting lung injury.

Conict of interest statement None declared. Funding sources None. References


1. Northway Jr WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med 1967;276:35768. 2. Coalson JJ. Pathology of new bronchopulmonary dysplasia. Semin Neonatol 2003;8:7381. 3. Attar MA, Donn SM. Mechanisms of ventilator-induced lung injury in premature infants. Semin Neonatol 2002;7:35360. 4. Morley C, Davis P. Continuous positive airway pressure: current controversies. Curr Opin Pediatr 2004;16:1415. 5. Verder H, Albertsen P, Ebbesen F, et al. Nasal continuous positive airway pressure and early surfactant therapy for respiratory distress syndrome in newborns of less than 30 weeks gestation. Pediatrics 1999;103:E24. 6. Verder H, Robertson B, Greisen G, et al. Surfactant therapy and nasal continuous positive airway pressure for newborns with respiratory distress syndrome. DanishSwedish Multicenter Study Group. N Engl J Med 1994;331:10515. 7. Ho JJ, Subramaniam P, Henderson-Smart DJ, Davis PG. Continuous distending airway pressure for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2000 (3):CD002271. 8. Morley CJ, Collaborators CT. Nasal CPAP or ventilation for very preterm infants at birth. A randomised controlled trial. The COIN Trial. E-PAS2007:61:6090.1, 2007. 9. Polglase GR, Hillman NH, Ball MK, et al. Lung and systemic inammation in preterm lambs on continuous positive airway pressure or conventional ventilation. Pediatr Res 2009;65:6771. 10. Davis PG, Morley CJ, Owen LS. Non-invasive respiratory support of preterm neonates with respiratory distress: continuous positive airway pressure and nasal intermittent positive pressure ventilation. Semin Fetal Neonatal Med 2009;14:1420.

11. De Paoli AG, Davis PG, Faber B, Morley CJ. Devices and pressure sources for administration of nasal continuous positive airway pressure (NCPAP) in preterm neonates. Cochrane Database Syst Rev; 2008 (1):CD002977. 12. Owen LS, Morley CJ, Davis PG. Neonatal nasal intermittent positive pressure ventilation: a survey of practice in England. Arch Dis Child Fetal Neonatal Ed 2008;93:F14850. 13. Kugelman A, Feferkorn I, Riskin A, Chistyakov I, Kaufman B, Bader D. Nasal intermittent mandatory ventilation versus nasal continuous positive airway pressure for respiratory distress syndrome: a randomized, controlled, prospective study. J Pediatr 2007;150:5216. 526 e1. 14. Donn SM, Sinha SK. Invasive and noninvasive neonatal mechanical ventilation. Respir Care 2003;48:42639. discussion 43941. 15. Ho JJ, Subramaniam P, Henderson-Smart DJ, Davis PG. Continuous distending pressure for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2002 (2):CD002271. 16. Gupta S, Sinha SK, Tin W, Donn SM. A randomized controlled trial of postextubation bubble continuous positive airway pressure versus Infant Flow Driver continuous positive airway pressure in preterm infants with respiratory distress syndrome. J Pediatr 2009;154:64550. 17. McCallion N, Davis PG, Morley CJ. Volume-targeted versus pressure-limited ventilation in the neonate. Cochrane Database Syst Rev; 2005 (3):CD003666. 18. Sinha SK, Donn SM. Volume-controlled ventilation. Variations on a theme. Clin Perinatol 2001;28:54760. vi. 19. Singh J, Sinha SK, Clarke P, Byrne S, Donn SM. Mechanical ventilation of very low birth weight infants: is volume or pressure a better target variable? J Pediatr 2006;149:30813. 20. Singh J, Sinha SK, Alsop E, Gupta S, Mishra A, Donn SM. Long term follow-up of very low birthweight infants from a neonatal volume versus pressure mechanical ventilation trial. Arch Dis Child Fetal Neonatal Ed 2009;94:F3602. 21. Donn SM, Sinha SK. Can mechanical ventilation strategies reduce chronic lung disease? Semin Neonatol 2003;8:4418. 22. Donn SM, Greenough A, Sinha SK. Patient triggered ventilation. Arch Dis Child Fetal Neonatal Ed 2000;83:F2256. 23. Greenough A, Dimitriou G, Prendergast M, Milner AD. Synchronized mechanical ventilation for respiratory support in newborn infants. Cochrane Database Syst Rev; 2008 (1):CD000456. 24. Donn SM, Sinha SK. Newer techniques of mechanical ventilation: an overview. Semin Neonatol 2002;7:4017. 25. Donn SM, Sinha SK. Newer modes of mechanical ventilation for the neonate. Curr Opin Pediatr 2001;13:99103. 26. Sinha SK, Donn SM. Newer forms of conventional ventilation for preterm newborns. Acta Paediatr 2008;97:133843. 27. Gupta S, Sinha SK, Donn SM. The effect of two levels of pressure support ventilation on tidal volume delivery and minute ventilation in preterm infants. Arch Dis Child Fetal Neonatal Ed 2009;94:F803. 28. Patel DS, Rafferty GF, Lee S, Hannam S, Greenough A. Work of breathing during SIMV with and without pressure support. Arch Dis Child 2009;94:4346. 29. Joshi VH, Bhuta T. Rescue high frequency jet ventilation versus conventional ventilation for severe pulmonary dysfunction in preterm infants. Cochrane Database Syst Rev; 2006 (1):CD000437. 30. Ventre KM, Arnold JH. High frequency oscillatory ventilation in acute respiratory failure. Paediatr Respir Rev 2004;5:32332. 31. Keszler M, Modanlou HD, Brudno DS, et al. Multicenter controlled clinical trial of high-frequency jet ventilation in preterm infants with uncomplicated respiratory distress syndrome. Pediatrics 1997;100:5939. 32. Henderson-Smart DJ, Bhuta T, Cools F, Offringa M. Elective high frequency oscillatory ventilation versus conventional ventilation for acute pulmonary dysfunction in preterm infants. Cochrane Database Syst Rev; 2003 (4):CD000104. 33. Courtney SE, Durand DJ, Asselin JM, Hudak ML, Aschner JL, Shoemaker CT. High-frequency oscillatory ventilation versus conventional mechanical ventilation for very-low-birth-weight infants. N Engl J Med 2002;347:64352. 34. Johnson AH, Peacock JL, Greenough A, et al. High-frequency oscillatory ventilation for the prevention of chronic lung disease of prematurity. N Engl J Med 2002;347:63342. 35. Marlow N, Greenough A, Peacock JL, et al. Randomised trial of high frequency oscillatory ventilation or conventional ventilation in babies of gestational age 28 weeks or less: respiratory and neurological outcomes at 2 years. Arch Dis Child Fetal Neonatal Ed 2006;91:F3206. 36. Kraybill EN, Runyan DK, Bose CL, Khan JH. Risk factors for chronic lung disease in infants with birth weights of 751 to 1000 grams. J Pediatr 1989;115:11520. 37. Woodgate PG, Davies MW. Permissive hypercapnia for the prevention of morbidity and mortality in mechanically ventilated newborn infants. Cochrane Database Syst Rev; 2001 (2):CD002061. 38. Sinha SK, Nicks JJ, Donn SM. Graphic analysis of pulmonary mechanics in neonates receiving assisted ventilation. Arch Dis Child Fetal Neonatal Ed 1996;75:F2138. 39. Sinha SK, Gupta S, Donn SM. Immediate respiratory management of the preterm infant. Semin Fetal Neonatal Med 2008;13:249. 40. Gillespie LM, White SD, Sinha SK, Donn SM. Usefulness of the minute ventilation test in predicting successful extubation in newborn infants: a randomized controlled trial. J Perinatol 2003;23:2057. 41. Claure N, Bancalari E. Automated respiratory support in newborn infants. Semin Fetal Neonatal Med 2009;14:3541.

Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Ventilatory management and bronchopulmonary dysplasia in preterm infants


Samir Gupta a, Sunil K. Sinha b, *, Steven M. Donn c
a

Department of Neonatal Paediatrics, University Hospital of North Tees, Stockton-on-Tees, UK Department of Neonatal Paediatrics, James Cook University Hospital, University of Durham, Middlesbrough, UK c Department of Pediatrics, Division of NeonatalPerinatal Medicine, C.S. Mott Childrens Hospital, University of Michigan Health System, Ann Arbor, Michigan, USA
b

s u m m a r y
Keywords: Bronchopulmonary dysplasia Mechanical ventilation Prematurity Respiratory distress syndrome

Improvements in antenatal and neonatal care have resulted in increased survival of very preterm infants. However, the incidence of bronchopulmonary dysplasia (BPD) has not changed, probably as a consequence of a demographic shift. The underlying pathophysiology of BPD appears to differ for the current population of preterm infants compared to that described by Northway et al., and management strategies should be targeted to limit ventilator-induced lung injury. Non-invasive respiratory support techniques are currently under evaluation, but results of the trials have thus far failed to show a reduction in BPD. This review will focus upon various ventilation modalities for preventing and managing bronchopulmonary dysplasia. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Classic bronchopulmonary dysplasia (BPD) described by Northway et al. in 1967 was seen in relatively mature babies, who were ventilated at high pressures, with high fractional concentration of inspired oxygen, resulting in lung overination, cystic emphysema, and brosis.1 By contrast, BPD is now seen primarily in very preterm newborns weighing <1000 g, who are born at 2426 weeks of gestation. The histopathologic lesions of the old BPD have now been replaced by a new BPD with the large, simplied alveolar structures, variable interstitial cellularity and/or broproliferation.2 The clinical picture is also different. Todays premature babies developing BPD may initially have modest ventilatory and oxygen needs and demonstrate a different radiographic picture, which shows diffuse haziness and a ne, lacy pattern (Fig. 1). These differences can be better understood with the realisation that the lungs of infants born at 2426 weeks of gestation are in the early phases of lung development (canalicular and saccular stages), and that alveolar and capillary development are inhibited following exposure to the noxious effects of mechanical ventilation, oxygen exposure, and inammation. These differences become important in planning ventilatory strategies to prevent or treat BPD in preterm infants.

2. Can ventilatory strategies prevent BPD? BPD is multifactorial and an understanding of the pulmonary injury sequence may enable the development of strategies to circumvent this cascade. There are two prominent pathways leading to BPD (Fig. 2). The rst is intrinsic and is related to a developmental arrest of the lung resulting in diminished alveolarisation. This results in inadequate surface area for gas exchange and impaired lung function, requiring the initiation of chronic ventilation and subsequent ventilator-induced lung injury (VILI). Other factors which may injure immature lungs include surfactant deciency and the need to deal with a very compliant chest wall. Extrinsic factors also inhibit alveolarisation and lung growth, and they include intrauterine cytokine exposure, antenatal and postnatal glucocorticoid treatment, insufcient nutrition, lung and systemic infections, and exposure to high concentrations of oxygen. 2.1. Understanding ventilator-induced lung injury Although a key component in the pulmonary injury sequence, VILI itself is multifactorial (Box 1). Volutrauma refers to injury related to overdistension or stretching of the lung units (alveoli and smaller airways) by delivering too much gas (tidal volume). Atelectotrauma refers to the damage caused by insufcient tidal volumes (recruitment and derecruitment of alveoli) associated with repetitive opening and closing of lung units. Biotrauma is a collective term to describe the adverse effects of infection and inammation. Rheotrauma refers to damage caused by inappropriate airway ow.

* Corresponding author. Address: Department of Neonatal Paediatrics, James Cook University Hospital, University of Durham, Middlesbrough, TS4 3BW UK. Tel./fax: 44 (0) 1642 854874. E-mail address: Sunil.sinha@stees.nhs.uk (S.K. Sinha). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.011

368

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

Fig. 1. Radiographic picture of bronchopulmonary dysplasia (BPD). (a) Old BPD: areas of atelectasis and hyperination. (b) New BPD: diffuse opacication of lung elds.

If the ow is excessive, it may cause inadvertent positive end-expiratory pressure (PEEP), turbulence and ineffective gas exchange, and lung overination. On the other hand, if the ow is inadequate, it may lead to air hunger (ow starvation) and increased work of breathing. An understanding of the pathophysiology of VILI is essential to formulating ventilatory strategies aimed at reducing or preventing BPD in very small infants.3 2.2. Non-invasive forms of respiratory support There has been a recent upsurge in the use of non-invasive respiratory support, using continuous positive airway pressure (CPAP) or nasal intermittent positive pressure ventilation (NIPPV). CPAP is a form of distending pressure provided through a nasal interface. NIPPV provides intermittent positive pressure ventilation in addition to continuous distending pressure. The proponents of non-invasive ventilation claim that the absence of an endotracheal tube reduces the risk of trauma to the airways, reduces the risk of infection, and causes less acute and chronic lung damage.4 However, the scientic evidence for the efcacy and safety of both

CPAP and NIPPV may vary depending upon whether they are used as a primary treatment for respiratory failure, or as an adjunct following extubation. The term non-invasive is also a misnomer, as both methods still deliver pressure that is supraphysiologic. 2.3. Continuous positive airway pressure CPAP supports the breathing of premature infants in a number of ways and can be delivered by a variety of devices and interfaces. CPAP devices can be categorised by the ow characteristics into continuous ow systems, such as bubble CPAP and ventilatorderived CPAP, and variable ow systems such as the infant ow driver and Benveniste (gas jet) valve CPAP. Two randomised trials have addressed the question of whether CPAP commenced soon after birth reduces mortality and morbidity of very preterm infants.5,6 Meta-analysis of these trials showed no difference in the rates of death, BPD, subsequent endotracheal intubation, or intraventricular haemorrhage.7 In a recent randomised controlled trial, 610 infants born at 2528 weeks of gestation were randomised to receive either CPAP or intubation within 5 min of delivery. Infants

Fig. 2. The effect of ventilator-induced lung injury and other factors on lung development, and their relationship to chronic lung disease. Reproduced with permission from Attar and Donn.3

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

369

Box 1. Components of ventilator-induced lung injury      Barotrauma Volutrauma Atelectotrauma Biotrauma Rheotrauma

were eligible if they were breathing spontaneously but still required respiratory support. Babies randomised to CPAP had a signicantly lower rate of the combined outcome, death or need for supplementary oxygen at 28 days (OR: 0.63; 0.460.88), but this difference disappeared at 36 weeks postmenstrual age. Moreover, babies who were intubated in the delivery room had a signicantly lower rate of pneumothorax than those randomised to CPAP (3% vs 9.1%, P 0.001).8 Thus, the evidence so far suggests that although CPAP may be an acceptable alternative to endotracheal intubation in some low birth weight babies, there is no convincing evidence that it reduces the incidence of BPD. The challenge remains to identify strategies that preserve the benet of CPAP and reduce the rate of complications such as pneumothorax. The benets of continuous CPAP with prophylactic exogenous surfactant treatment have been explored, but methodological constraints of these studies preclude any meaningful conclusion. Two large studies are currently under way [Surfactant Positive Airway Pressure and Pulse Oximetry Trial (SUPPORT), and National Institute of Child and Human Health Neonatal Research Network trials] comparing nasal CPAP, mechanical ventilation, and surfactant administration followed by early extubation to nasal CPAP in extremely premature babies. The results of these ongoing trials may help to shed some light on this controversy. At the present time, we advocate that standard therapy should be the administration of surfactant followed by mechanical ventilation for babies <28 weeks of gestation. The claim that non-invasive ventilation protects against BPD has also been challenged in a recent animal study by Polglase et al., who failed to demonstrate any advantages of non-invasive respiratory support over mechanical ventilation.9 In extremely preterm babies it seems advisable to administer early surfactant and attempt to limit VILI by aggressively weaning mechanical support (described below). 2.4. Nasal intermittent positive pressure ventilation Despite increasing experience, 4560% of extremely low birth weight infants fail early CPAP therapy for treatment of respiratory distress syndrome (RDS) and require subsequent intubation and ventilation. Even among VLBW infants extubated to CPAP, 2540% still require reintubation.10,11 Efforts to improve these failure rates have prompted the use of NIPPV. The mechanism of action of NIPPV, however, remains uncertain. It is not clear whether mechanical inations during NIPPV are actually transmitted to the lungs.10 Synchronising NIPPV inations with the infants own breath is theoretically advantageous. Non-synchronised NIPPV may deliver high pressure during spontaneous exhalation, increasing the risk of elevated upper airway pressure and subsequent pneumothorax. No trials, however, have compared synchronised with non-synchronised NIPPV, or examined the accuracy of synchronisation devices in NIPPV. Moreover, there is no consensus regarding the appropriate pressure levels, rate, or inspiratory time during NIPPV. In a recent survey, widely varying settings were utilised during NIPPV: PIP 720 cmH2O; PEEP 49 cmH2O; Ti 0.30.5 s; rate 1060 per min.12 A number of trials have compared NIPPV to CPAP following extubation and have shown a signicant reduction in extubation

failure using NIPPV. Other investigations compared NIPPV as a primary treatment of RDS and showed reduction in BPD rates with NIPPV (2%), compared to NCPAP (17%) (P 0.03).13 However, the sample sizes are small and results should be interpreted with caution. In summary, although non-invasive ventilation is being increasingly used both as a primary strategy for respiratory support or to facilitate extubation, there is no conclusive evidence that it prevents BPD.14 The use of any form of distending pressure is likely to be associated with an increased risk of pneumothorax15 and none of the reported trials of non-invasive respiratory support conrm any benecial effects on long-term pulmonary and neurodevelopmental outcomes. 2.5. Post-extubation CPAP and NIPPV Whereas there is no consensus regarding the benets of CPAP and NIPPV as a primary modality to treat RDS, the scientic evidence for their efcacy in facilitating extubation and reducing the duration of mechanical ventilation is well established.11 However, certain conditions apply. For example, the effectiveness of CPAP depends upon a number of factors, such as the level of pressure used (benecial with pressure 5 cmH2O), duration of prior mechanical ventilation (effective if duration is 14 days), and the nasal interface (short binasal prongs provide the lowest resistance). The CPAP devices also inuence performance. This was recently evaluated in a controlled trial, which showed bubble CPAP to have some advantage over the infant ow driver in relation to some short-term outcome measures.16 NIPPV, on the whole, seem to be better in preventing extubation failure compared with CPAP but requires further evaluation. Moreover, despite the conrmed advantages of CPAP and NIPPV in reducing the duration of mechanical ventilation, none of the meta-analyses or individual ransomised controlled trials has shown that they reduce BPD.11 2.6. Conventional mechanical ventilation (CMV) For nearly three decades, CMV was the only method of ventilating preterm infants. Using time-cycled, pressure-limited ventilation (TCPLV) in an intermittent mandatory ventilation mode, this form of ventilation was relatively easy to use and resulted in consistent delivery of pressure, although delivered tidal volumes varied according to lung mechanics. The judicious use of sophisticated respiratory support has been the centre of attention for the past two decades. With the implementation of microprocessorbased technology, ventilators can now be adjusted to suit the individual needs of the most complex patients. The availability of real-time graphic monitoring on ventilators enables clinicians to visualise respiratory support on a breath-to-breath basis. Despite the advances in technology, there is no clear evidence for an optimal ventilator strategy that crosses different gestational ages or types of pulmonary diseases. However, recent clinical investigation suggests that we may be on the right track. 2.7. Volume-targeted ventilation Volume-targeted modalities are relatively new to neonatal intensive care and represent a departure from traditional TCPLV by focusing on tidal volume and allowing pressure to vary. Although there are only a few published randomised controlled clinical trials, thus far the evidence is highly encouraging. The consistency of tidal volume delivery during volume-controlled ventilation (VCV) in the face of varying lung compliance and the auto-weaning of airway pressure may be clinically advantageous, especially in conditions in which lung compliance can change rapidly, such as after surfactant administration. A meta-analysis of volume-targeted ventilation

370

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

reported a reduction in the total duration of ventilation, severe intraventricular haemorrhage and pneumothorax, and a strong trend towards a reduction in the incidence BPD.17 Stability of tidal volume delivery may account for this, especially in extremely low birth weight infants, who are most at risk for sustaining complications associated with mechanical ventilation.18 Volume-targeted ventilation can be provided in two ways. Volume-controlled ventilation (VCV) targets the delivery of a clinician-set tidal volume. This is delivered irrespective of how much pressure is required (although for safety reasons this can be limited). Hybrid forms, such as Volume Guarantee, Pressure Regulated Volume Control, and Volume Assured Pressure Support, are essentially pressure-targeted modalities but aim to deliver a tidal volume within a set range using computer-controlled feedback and in some cases a breath-averaging technique. We recently reported the results of a randomised controlled trial of VCV in a population of preterm babies with RDS, born between 24 and 31 weeks of gestation and weighing 6001500 g. There were trends towards faster weaning, reduction in the duration of ventilation, decreased BPD, and improved survival favouring VCV.19 As part of the original study design, a longer-term follow-up of these babies was conducted to examine clinically relevant outcomes beyond discharge. The primary aim of the follow-up study was to assess survival and respiratory and gross neurodevelopmental status at 2 years corrected age. The respiratory outcomes of interest were the occurrence and frequency of cough and wheeze, the use of inhaled medications for respiratory dysfunction, and hospital readmissions for respiratory illnesses. The study showed that the trend towards less BPD reported at 36 weeks PMA in the original trial had a clinical correlate at follow-up, with fewer children requiring pharmacotherapy.20 2.8. Patient-triggered ventilation One of the major problems with intermittent mandatory ventilation (IMV) is patient-ventilator dyssynchrony. Babies can exhale against incoming pressure, may exhibit inefcient gas exchange, gas trapping and air leaks, and the need for higher levels of support.21 During the 1990s, the capability of detecting spontaneous respiratory activity led to the development of patient-triggered ventilation (PTV) in neonatal respiratory care. Modes of ventilation utilising PTV include synchronised intermittent mandatory ventilation (SIMV), assist/control (A/C) ventilation, and pressure support ventilation (PSV). Transducers detect some measure of spontaneous respiratory effort, such as changes in airway ow, and respond with a mechanical breath timed to begin almost immediately. Early clinical trials suggested a trend towards a reduction in BPD, but these studies were underpowered. One large, multicentre open trial did not show any differences in the incidence of BPD, but methodological and design aws, and signicant user inexperience, limit the interpretation of the ndings.22 Although short-term benets of PTV have been demonstrated,23 further studies are required to address its impact on BPD. 2.9. Synchronised intermittent mandatory ventilation In SIMV, the clinician selects a mandatory rate of ventilation. Each time the baby is scheduled to receive a breath, the ventilator waits briey for the infant to initiate a breath. If the baby breathes within this timing window, a mechanical breath will be initiated synchronously with the spontaneous breath. If the baby fails to breathe, the ventilator will cycle at the set rate. In between mechanical breaths, spontaneous breathing is supported, only by PEEP, resulting in a higher work of breathing than with either A/C or PSV.24

2.10. Assist/control ventilation In A/C all spontaneous breaths that exceed the trigger threshold result in the delivery of a mechanical breath that is synchronised to the onset of inspiration. If the baby is apnoeic or fails to exceed the trigger threshold, the ventilator will cycle at the control (back-up) rate. Management of infants ventilated by A/C is different from that of IMV or SIMV. Provided that the patient is breathing above the control rate, reductions in the control rate will have no effect. Weaning is accomplished by decreasing the peak inspiratory pressure rst, which should be benecial in reducing both barotrauma and volutrauma.25 However, reduction in BPD has not been uniformly demonstrated.

2.11. Pressure support ventilation PSV is primarily a weaning mode of ventilation in which spontaneous breaths receive an inspiratory pressure boost to decrease the imposed work of breathing and unload the respiratory musculature. A spontaneous breath triggers the delivery of a timeand pressure-limited breath, which is ow-cycled (see below). Breaths may be fully (provide a full tidal volume) or partially supported. PSV is usually applied in conjunction with SIMV, although it may be utilised alone if the baby has reliable respiratory drive and can easily exceed the trigger threshold. Another feature of PSV is variable inspiratory ow, which is proportional to patient effort, and helps to overcome increased resistance.26 A cross-over trial by Gupta et al. reported that PSV increases total minute ventilation and stabilises breathing in proportion to the applied level of pressure support.27 This may be advantageous and provide a useful ventilation strategy during weaning in preterm infants to allow a more consistent transfer of the work of breathing from the ventilator to the infant.28

2.12. Flow-cycling Cycling refers to the mechanism responsible for transitioning from inspiration to expiration and from expiration to inspiration. In the past, this has been accomplished by time; inspiration ended when the exhalation valve opened after a preset interval. The new technology offers ow-cycling, where inspiration ends when inspiratory ow decreases to a certain percentage of peak inspiratory ow. This is interpreted by the ventilator as a sign that the baby is about to end his own inspiratory effort and thus functions as an expiratory trigger. The advantages of ow cycling are that the baby is able to set the inspiratory time (a breath starts and ends at the babys preference) and, at rapid rates, an inverse inspiratory:expiratory ratio cannot occur. The mechanical breath will be terminated before the baby can initiate another breath, rather than persisting for a xed time interval, as would occur with timecycling (Fig. 3).26 Flow-cycling is incorporated in PSV. It allows the baby to have control over the ventilator rate and inspiratory time, and it also enables full synchronisation of the respiratory cycle.

2.13. High frequency ventilation First introduced into neonatal practice in the early 1980s, high frequency ventilation (HFV) uses extremely small tidal volumes delivered at rapid rates to effect gas exchange at lower alveolar pressures than CMV. There are two primary forms of HFV: high frequency jet ventilation (HFJV), and high frequency oscillatory ventilation (HFOV), as well as hybrid forms.

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

371

RDS.32 Two recently completed trials found differing results. The HIFO trial of Courtney et al. found a very slight reduction in BPD in babies receiving HFOV, although this could have reected the fact that the control group was managed in SIMV, which is not the ideal mode for acutely ill infants.33 The UKOS study found no difference in the incidence of BPD among infants receiving HFOV or conventional ventilation.34 Currently, there is insufcient evidence to recommend HFOV.35 2.14. Permissive hypercapnia The rationale for permissive hypercapnia, using a low lung volume strategy, is that it may decrease volutrauma and lung injury, lessen the duration of mechanical ventilation, reduce alveolar ventilation and the complications of hypocapnia (especially reduced cerebral blood ow), and increase oxygen unloading at the tissue level (Bohr effect).36 The strategy of ventilating infants at a higher PaCO2 level, termed permissive hypercapnia, is based upon the retrospective observation of Kraybill et al. in 1989. In a multicentre analysis of 235 preterm infants, these investigators demonstrated a higher incidence of BPD in the infants with the lowest PaCO2 on the second and fourth postnatal days.36 Two prospective, randomised, controlled trials of permissive hypercapnia have been conducted.37 Although both studies demonstrated a statistically signicant reduction in the duration of ventilation and other secondary outcome measures, the incidence of BPD did not differ. 2.15. Monitoring of ventilated infants Advances in biomedical engineering have also substantially changed the way in which babies receiving mechanical ventilation can be monitored. Intermittent chest radiographs and blood gas sampling have been replaced by continuous monitoring of pulmonary mechanics and pulse oximetry or transcutaneous oxygen/carbon dioxide tensions. Real-time pulmonary mechanics monitoring has become a valuable adjunct to CMV.38 The same sensor technology utilised for PTV enables measurement of changes in airway pressure or ow, which can be converted to a volume signal. Determination of

Fig. 3. Pulmonary graphic scalar record showing gas trapping. The inspiration commences before the expiratory ow could reach the baseline, leading to gas trapping and inadvertent positive end-expiratory pressure.

2.13.1. High frequency jet ventilation HFJV uses rates of 240660 breaths/min, and inspiratory times are typically about 0.02 s. It is used in tandem with a conventional ventilator, which provides PEEP and optional sigh breaths. High velocity pulsations are injected into the proximal airway using a special connector. Ventilator adjustments are similar to those of CMV. The amplitude is set by adjusting peak pressure and PEEP.29 2.13.2. High frequency oscillatory ventilation HFOV differs from HFJV in that even smaller tidal volumes are used at rates of 815 Hz. Mean airway pressure is used as a continuous distending pressure to inate the lung to a static volume, and oscillations around this mean are used to effect gas exchange. Adjustments for oxygenation (via mean airway pressure) and ventilation (via amplitude) can be done independently of one another.30 Moreover, HFOV uses active exhalation, whereby gas is actively withdrawn from the lung during expiration. A number of clinical studies of high frequency ventilation have reported long-term outcomes, although BPD was not the primary objective in most cases. Keszler et al. reported a reduced incidence of BPD and the need for home oxygen in infants treated with HFJV compared with CMV for uncomplicated RDS.31 A number of recent clinical trials compared HFOV with CMV in preterm infants with

Fig. 4. Time-cycling vs ow-cycling. Note that in time-cycling there is a prolongation of a no ow state (shaded area) at the end of inspiration until the exhalation valve opens. With ow-cycling, inspiration ends when inspiratory ow has decreased to a small percentage of peak inspiratory ow. Note how the breath cycles directly into expiration at this point (arrow).

372

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

Table 1 Respiratory management options for bronchopulmonary dysplasia. Issue Increased resistance Strategy Variable inspiratory ow (pressure control or pressure support ventilation). Consider using higher PEEP to stent collapsing airways. Use largest comfortable ETT. Use higher PIP/PEEP. Consider high frequency ventilation. Avoid hyperination. Monitor tidal volumes. Consider volume-targeted ventilation Consider pharmacotherapy. Liberalisation of arterial oxygen tension. Re-evaluate gas exchange capability; allow higher carbon dioxide tension if pH is not deranged Try to optimise caloric and uid intake, but avoid excess carbohydrate or fat Comments Seen with airway injury and hyper-reactivity

Decreased compliance

Reects parenchymal lung injury

Heterogeneous lung disease Pulmonary hypertension Hypercapnia Nutrition

Pressure-targeted ow patterns may contribute to ventilation/perfusion mismatch Limited evidence; follow echocardiogram. Do not attempt to achieve normal gas exchange Non-nitrogen calories will contribute to carbon dioxide production and respiratory load

PEEP, positive end-expiratory pressure; ETT, endotracheal tube; PIP, peak inspiratory pressure.

tidal volume, minute ventilation, and breath-to-breath displays of ow, pressure, and volume waveforms are now standard techniques. Many devices also enable real-time assessment of pulmonary mechanics through displays of pressure-volume and ow-volume loops, and calculation of resistance and compliance. Detection of potentially adverse situations, such as gas trapping or hyperination, is now possible before these dangers become clinically apparent (Fig. 4).39 Pulmonary graphics are critical in the appropriate application of some of the newer ventilatory techniques, such as PSV. Tidal volume monitoring allows adjustments that can provide optimal lung expansion and selection of the optimal PEEP. Spontaneous minute ventilation monitoring has been shown to be useful in assessing the readiness for extubation in mechanically ventilated infants.40 Pulmonary mechanics monitoring permits the clinician to objectively assess the effect of pharmacologic agents with a narrow therapeutic index, such as corticosteroids, bronchodilators, or diuretics. Continuous pulse oximetry or blood gas analysis, using either indwelling intravascular sensors or transcutaneous electrodes, transforms ventilator management from an intermittent to a dynamic process. Maintaining oxygenation and ventilation within a desired range should help to minimise the deleterious effects of both oxygen toxicity and positive pressure. Techniques that adapt the mechanical ventilatory support and supplemental oxygen to the changing needs of the preterm infants are being developed in order to improve the stability of gas exchange, to minimise respiratory support.41 3. Ventilatory strategies for infants with established BPD The management of infants with established BPD has not been studied adequately, and the role of various ventilatory strategies for infants with the disorder is not clear. A suggested guideline for dealing with different lung issues is shown in Table 1. A number of other strategies can also be employed to reduce the duration of mechanical ventilation with a hope that earlier extubation might ameliorate the pulmonary injury sequence and thus reduce the severity of BPD. The scientic evidence in favour of such interventions varies; however, clinicians should always assess the level of evidence and risk:benet ratio before implementing them in routine practice. 3.1. Physiologic principles The infant with ventilator-dependent BPD presents many challenges to the clinician. Chronic respiratory insufciency may result in aberrant gas exchange, manifest by severe hypercarbia, with marked renal compensation. The inexperienced clinician often

attempts to make the baby conform to physiologic blood gases, and in doing so may inadvertently overventilate the baby and unwittingly contribute to VILI. It is perhaps wiser to view the chest radiograph in the context of, What can I expect from these lungs? and adjust ones expectations of blood gases accordingly. Alteration in lung mechanics is another key feature of BPD. Reactive airways may result in increased pulmonary resistance. This may be treated empirically by adjustments in ventilator strategy, such as increasing PEEP and/or airway ow, or by using a modality with variable inspiratory ow, such as pressure control or pressure support compared with traditional xed ow TCPLV. Lung compliance may also be abnormal. Ventilating the lung at an appropriate functional residual capacity, where incremental changes in pressure recruit the most lung volume, is often a challenge. Assessing compliance at different levels of PEEP may be helpful in this regard. Alterations in resistance and compliance will alter the respiratory time constant, and sufcient expiratory time to avoid gas trapping and inadvertent PEEP must be provided. Decisions about the best mode or modality to ventilate a baby with BPD are best made on an individual basis. Theoretically, if lung disease is homogeneous, a modality such as pressure control might be advantageous, whereas heterogeneous lung disease should respond better to VCV because peak pressure and peak volume delivery occur at the end of inspiration, helping to make gas delivery more uniform throughout the lung. This will require clinical investigation. Periodic assessment of cardiac function, looking for evidence of pulmonary hypertension, may also guide ones use of supplemental oxygen and establishment of appropriate ranges for oxygen saturation, although the optimal limits are yet to be established. All changes in strategy should be objectively assessed for response. 4. Conclusion BPD is a recognised sequel of preterm birth. With improving survival of infants at lower gestational ages, the prevalence is on the rise. Pathological features of BPD include alveolar maldevelopment, with or without areas of pulmonary brosis. Assisted ventilation, infection/inammation, oxygen administration, and uid overload are major identied risk factors in the evolution of BPD. The prevention of BPD needs a multifocal approach by decreasing VILI through utilisation of newer ventilatory strategies such as volume-controlled ventilation, and the use of non-invasive forms of respiratory support in selected preterm babies. Many other therapies are still investigational and potentially dangerous and need further evidence before they can be routinely recommended. Determining the optimal strategy for infants with ventilator-

S. Gupta et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 367373

373

dependent BPD is a work in progress, but a necessary one in view of the number of affected infants.

Practice points  Various modalities of ventilation are available, but efforts should be made to limit ventilator-induced lung injury by limiting the duration of ventilation and avoiding excessive volume and pressure.  Non-invasive respiratory support techniques are currently under investigation in extremely preterm infants and, until further data are available, conventional mechanical ventilation should be the primary strategy for these infants.  Volume-targeted ventilation appears to decrease the incidence of BPD and limiting volutrauma seems to be a logical approach.

Research directions  Evaluation of non-invasive respiratory support techniques.  Evaluation of pressure support ventilation for early and subsequent management of RDS.  Large, adequately powered randomised clinical trials to dene the roles of various devices (such as high frequency ventilation), modalities (such as patient-triggered or volume-targeted ventilation), and strategies (such as permissive hypercapnea) in limiting lung injury.

Conict of interest statement None declared. Funding sources None. References


1. Northway Jr WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med 1967;276:35768. 2. Coalson JJ. Pathology of new bronchopulmonary dysplasia. Semin Neonatol 2003;8:7381. 3. Attar MA, Donn SM. Mechanisms of ventilator-induced lung injury in premature infants. Semin Neonatol 2002;7:35360. 4. Morley C, Davis P. Continuous positive airway pressure: current controversies. Curr Opin Pediatr 2004;16:1415. 5. Verder H, Albertsen P, Ebbesen F, et al. Nasal continuous positive airway pressure and early surfactant therapy for respiratory distress syndrome in newborns of less than 30 weeks gestation. Pediatrics 1999;103:E24. 6. Verder H, Robertson B, Greisen G, et al. Surfactant therapy and nasal continuous positive airway pressure for newborns with respiratory distress syndrome. DanishSwedish Multicenter Study Group. N Engl J Med 1994;331:10515. 7. Ho JJ, Subramaniam P, Henderson-Smart DJ, Davis PG. Continuous distending airway pressure for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2000 (3):CD002271. 8. Morley CJ, Collaborators CT. Nasal CPAP or ventilation for very preterm infants at birth. A randomised controlled trial. The COIN Trial. E-PAS2007:61:6090.1, 2007. 9. Polglase GR, Hillman NH, Ball MK, et al. Lung and systemic inammation in preterm lambs on continuous positive airway pressure or conventional ventilation. Pediatr Res 2009;65:6771. 10. Davis PG, Morley CJ, Owen LS. Non-invasive respiratory support of preterm neonates with respiratory distress: continuous positive airway pressure and nasal intermittent positive pressure ventilation. Semin Fetal Neonatal Med 2009;14:1420.

11. De Paoli AG, Davis PG, Faber B, Morley CJ. Devices and pressure sources for administration of nasal continuous positive airway pressure (NCPAP) in preterm neonates. Cochrane Database Syst Rev; 2008 (1):CD002977. 12. Owen LS, Morley CJ, Davis PG. Neonatal nasal intermittent positive pressure ventilation: a survey of practice in England. Arch Dis Child Fetal Neonatal Ed 2008;93:F14850. 13. Kugelman A, Feferkorn I, Riskin A, Chistyakov I, Kaufman B, Bader D. Nasal intermittent mandatory ventilation versus nasal continuous positive airway pressure for respiratory distress syndrome: a randomized, controlled, prospective study. J Pediatr 2007;150:5216. 526 e1. 14. Donn SM, Sinha SK. Invasive and noninvasive neonatal mechanical ventilation. Respir Care 2003;48:42639. discussion 43941. 15. Ho JJ, Subramaniam P, Henderson-Smart DJ, Davis PG. Continuous distending pressure for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2002 (2):CD002271. 16. Gupta S, Sinha SK, Tin W, Donn SM. A randomized controlled trial of postextubation bubble continuous positive airway pressure versus Infant Flow Driver continuous positive airway pressure in preterm infants with respiratory distress syndrome. J Pediatr 2009;154:64550. 17. McCallion N, Davis PG, Morley CJ. Volume-targeted versus pressure-limited ventilation in the neonate. Cochrane Database Syst Rev; 2005 (3):CD003666. 18. Sinha SK, Donn SM. Volume-controlled ventilation. Variations on a theme. Clin Perinatol 2001;28:54760. vi. 19. Singh J, Sinha SK, Clarke P, Byrne S, Donn SM. Mechanical ventilation of very low birth weight infants: is volume or pressure a better target variable? J Pediatr 2006;149:30813. 20. Singh J, Sinha SK, Alsop E, Gupta S, Mishra A, Donn SM. Long term follow-up of very low birthweight infants from a neonatal volume versus pressure mechanical ventilation trial. Arch Dis Child Fetal Neonatal Ed 2009;94:F3602. 21. Donn SM, Sinha SK. Can mechanical ventilation strategies reduce chronic lung disease? Semin Neonatol 2003;8:4418. 22. Donn SM, Greenough A, Sinha SK. Patient triggered ventilation. Arch Dis Child Fetal Neonatal Ed 2000;83:F2256. 23. Greenough A, Dimitriou G, Prendergast M, Milner AD. Synchronized mechanical ventilation for respiratory support in newborn infants. Cochrane Database Syst Rev; 2008 (1):CD000456. 24. Donn SM, Sinha SK. Newer techniques of mechanical ventilation: an overview. Semin Neonatol 2002;7:4017. 25. Donn SM, Sinha SK. Newer modes of mechanical ventilation for the neonate. Curr Opin Pediatr 2001;13:99103. 26. Sinha SK, Donn SM. Newer forms of conventional ventilation for preterm newborns. Acta Paediatr 2008;97:133843. 27. Gupta S, Sinha SK, Donn SM. The effect of two levels of pressure support ventilation on tidal volume delivery and minute ventilation in preterm infants. Arch Dis Child Fetal Neonatal Ed 2009;94:F803. 28. Patel DS, Rafferty GF, Lee S, Hannam S, Greenough A. Work of breathing during SIMV with and without pressure support. Arch Dis Child 2009;94:4346. 29. Joshi VH, Bhuta T. Rescue high frequency jet ventilation versus conventional ventilation for severe pulmonary dysfunction in preterm infants. Cochrane Database Syst Rev; 2006 (1):CD000437. 30. Ventre KM, Arnold JH. High frequency oscillatory ventilation in acute respiratory failure. Paediatr Respir Rev 2004;5:32332. 31. Keszler M, Modanlou HD, Brudno DS, et al. Multicenter controlled clinical trial of high-frequency jet ventilation in preterm infants with uncomplicated respiratory distress syndrome. Pediatrics 1997;100:5939. 32. Henderson-Smart DJ, Bhuta T, Cools F, Offringa M. Elective high frequency oscillatory ventilation versus conventional ventilation for acute pulmonary dysfunction in preterm infants. Cochrane Database Syst Rev; 2003 (4):CD000104. 33. Courtney SE, Durand DJ, Asselin JM, Hudak ML, Aschner JL, Shoemaker CT. High-frequency oscillatory ventilation versus conventional mechanical ventilation for very-low-birth-weight infants. N Engl J Med 2002;347:64352. 34. Johnson AH, Peacock JL, Greenough A, et al. High-frequency oscillatory ventilation for the prevention of chronic lung disease of prematurity. N Engl J Med 2002;347:63342. 35. Marlow N, Greenough A, Peacock JL, et al. Randomised trial of high frequency oscillatory ventilation or conventional ventilation in babies of gestational age 28 weeks or less: respiratory and neurological outcomes at 2 years. Arch Dis Child Fetal Neonatal Ed 2006;91:F3206. 36. Kraybill EN, Runyan DK, Bose CL, Khan JH. Risk factors for chronic lung disease in infants with birth weights of 751 to 1000 grams. J Pediatr 1989;115:11520. 37. Woodgate PG, Davies MW. Permissive hypercapnia for the prevention of morbidity and mortality in mechanically ventilated newborn infants. Cochrane Database Syst Rev; 2001 (2):CD002061. 38. Sinha SK, Nicks JJ, Donn SM. Graphic analysis of pulmonary mechanics in neonates receiving assisted ventilation. Arch Dis Child Fetal Neonatal Ed 1996;75:F2138. 39. Sinha SK, Gupta S, Donn SM. Immediate respiratory management of the preterm infant. Semin Fetal Neonatal Med 2008;13:249. 40. Gillespie LM, White SD, Sinha SK, Donn SM. Usefulness of the minute ventilation test in predicting successful extubation in newborn infants: a randomized controlled trial. J Perinatol 2003;23:2057. 41. Claure N, Bancalari E. Automated respiratory support in newborn infants. Semin Fetal Neonatal Med 2009;14:3541.

Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Prevention of bronchopulmonary dysplasia


Matthew M. Laughon a, *, P. Brian Smith b, Carl Bose a
a b

University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA Duke University, Durham, North Carolina, USA

s u m m a r y
Keywords: Bronchopulmonary dysplasia Caffeine Corticosteroids Prevention Prophylaxis Vitamin A

Considerable effort has been devoted to the development of strategies to reduce the incidence of bronchopulmonary dysplasia (BPD), including use of medications, nutritional therapies, and respiratory care practices. Unfortunately, most of these strategies have not been successful. To date, the only two treatments developed specically to prevent BPD whose efcacy is supported by evidence from randomized, controlled trials are the parenteral administration of vitamin A and corticosteroids. Two other therapies, the use of caffeine for the treatment of apnea of prematurity and aggressive phototherapy for the treatment of hyperbilirubinemia, were evaluated for the improvement of other outcomes and found to reduce BPD. Cohort studies suggest that the use of continuous positive airway pressure as a strategy for avoiding mechanical ventilation might also be benecial. Other therapies reduce lung injury in animal models but do not appear to reduce BPD in humans. The benets of the efcacious therapies have been modest, with an absolute risk reduction in the 711% range. Further preventive strategies are needed to reduce the burden of this disease. However, each will need to be tested in randomized, controlled trials, and the expectations of new therapies should be modest reductions of the incidence of the disease. 2009 Elsevier Ltd. All rights reserved.

1. Introduction After the development of bronchopulmonary dysplasia (BPD), medical care usually focuses on utilizing respiratory therapies that minimize further lung injury and optimizing nutrition and growth, with varying degrees of success. Therefore, because of the serious, long-term health consequences of BPD, considerable effort has been devoted to the prevention of the disease. In this seminar, we examine the evidence of strategies that have been tested to prevent BPD. BPD most often begins with early lung injury from respiratory distress syndrome (RDS). Treatments that prevent or reduce the severity of RDS (e.g. surfactant replacement therapy) were developed with the hope that they would reduce the likelihood of developing BPD. Morbidities of prematurity, most notably persistent patency of the ductus arteriosus (PDA), may contribute to BPD. Prevention or treatment of PDA have been advocated as strategies to prevent BPD. Various uid and nutritional regimens have been utilized during the early neonatal period because of their potential to modify the likelihood of developing BPD. Finally, treatment of

diseases or problems not related directly to the lungs (e.g. apnea of prematurity) may have an impact on vulnerability to BPD. We review evidence primarily from randomized controlled trials (RCTs), or the combined results of trials in meta-analyses, using the Cochrane Database of Systematic Reviews as the main source of information. Although meta-analyses, including the Cochrane reviews, suffer from a number of limitations, we believe that they provide the highest level of evidence.1,2 2. Treatments of early morbidities of prematurity 2.1. Surfactant therapy Following the development of exogenous surfactant preparations for the prevention and treatment of RDS, there were expectations that the use of these products would decrease the incidence of BPD. Surfactant therapy decreases mortality, particularly when used very soon after birth as a prophylactic strategy (compared with treatment after established RDS) and also reduces the incidence of lung injury as evidenced by a reduction in pneumothorax.3 However, surfactant therapy does not reduce the incidence of BPD among survivors.46 Since the original surfactant trials, two additional strategies of surfactant therapy have been reported. The rst method is to administer surfactant during a brief period of intubation without

* Corresponding author. CB# 7596, 4th Floor, UNC Hospitals, Chapel Hill, NC 27599-7596, USA. Tel.: 1 919 966 5063; fax: 1 919 966 3034. E-mail address: matt_laughon@med.unc.edu (M.M. Laughon). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.002

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

375

obligatory mechanical ventilation (i.e. intubation, delivery of surfactant, and immediate extubation).7 Benets of this strategy have been reported in three trials.810 However, a statistically signicant reduction in BPD was not observed in any. In a Cochrane review, early surfactant treatment during a brief period of ventilation compared with later selective surfactant therapy in ventilated infants resulted in a reduced incidence of air leak and BPD [relative risk (RR): 0.51; 95% condence interval (CI): 0.260.99], dened as oxygen therapy at 28 days.11 However, the more common denition of BPD, oxygen therapy at 36 weeks postmenstrual age (PMA), could not be evaluated in this meta-analysis because it was not reported in all trials. In a more recent trial, not included in the Cochrane review, very early surfactant therapy administered in the delivery room without obligatory ventilation was demonstrated to reduce the risk of mechanical ventilation.12 This study was not powered to examine BPD at 36 weeks PMA, and there was no statistically signicant improvement in BPD. Another strategy is to administer surfactant beyond the rst postnatal week to infants in whom a secondary surfactant dysfunction may be present,13 and whose risk for BPD is high. In an RCT of surfactant therapy using this strategy, infants assigned to one of two doses of surfactant had a lower mean fraction of inspired oxygen (33%) at 24 h after dosing, compared with infants assigned to the placebo group (39%).14 However, there was no difference in the incidence of mortality or BPD between groups. Surfactant replacement therapy is a standard of care for infants at risk for RDS or with established disease based upon reductions in mortality and air leak. However, current evidence is insufcient to support alternate early surfactant treatment strategies (e.g. treatment without obligatory mechanical ventilation) or treatment beyond the immediate neonatal period for the prevention of BPD.

ibuprofen, is as effective at promoting ductal closure, but, like indomethacin, does not appear to reduce mortality or the likelihood of BPD.25 Although there have been no recent controlled trials comparing outcomes following ligation compared with either placebo or medical therapy, a recent post-hoc analysis of an RCT of prophylactic PDA ligation26 found that 48% of the infants allocated to prophylactic PDA ligation developed BPD compared with 21% in the control group.27 This lack of demonstrable impact of ductal closure on the incidence of BPD may have resulted, in part, from limitations imposed by study designs. Under the assumption that closure of the PDA is benecial, virtually all clinical trials in the modern era have focused on the most expeditious way in which to close a PDA. None has addressed the more fundamental question of whether closing the PDA improves outcome. Those trials that have included control groups treated with a placebo have permitted treatment of a PDA that persisted after reaching a dened study endpoint, often within days after enrollment. This study design has resulted in high rates of treatment in the placebo group (usually in the range of 40%) and has diminished the likelihood of identifying the effect of PDA closure on other outcomes. Additionally, the ability to detect adverse effects of treatment is equally compromised. Evidence does not support the prevention or treatment of PDAs, with the goal of closure, for the prevention of BPD.

2.3. Caffeine for apnea of prematurity Caffeine is a methylxanthine used to treat apnea of prematurity. In a large RCT, the impact of treatment with caffeine on neurodevelopmental outcome was tested in infants with birth weights from 500 to 1250 g. Infants assigned to receive caffeine had a lower incidence of neurodevelopmental impairment, most notably cerebral palsy (4.4% in the caffeine group vs 7.3% in the placebo group) and cognitive delay (33.8% vs 38.3%).28 One of the secondary outcomes of the trial was BPD. Infants in the caffeine group had an incidence of BPD of 36% compared with 47% of infants in the placebo group [adjusted odds ratio (OR): 0.63; CI: 0.520.76; P < 0.001]. The mechanism by which caffeine reduced the incidence of BPD is uncertain. Positive airway pressure was discontinued one week earlier in the infants receiving caffeine.29 Therefore, it is possible that caffeine reduced the exposure to mechanical ventilation, thereby reducing ventilator-induced lung injury. Evidence supports the use of caffeine for the treatment of apnea of prematurity in extremely low gestational age newborns, with the probability that a secondary benet will be a reduction in BPD.

2.2. Closure of the patent ductus arteriosus Persistent PDA following preterm birth is common, particularly among the least mature infants, and is associated with neonatal morbidities, including BPD.15,16 Shunting of blood from the systemic to pulmonary circulation may have an effect on the mechanical properties of the lung, such as a decrease in lung compliance,17,18 and these changes may provoke the need to increase ventilatory support. This cascade of events may increase the likelihood of BPD by increasing ventilator-induced lung injury.19 This hypothesis is supported by a strong association between the presence of a PDA and BPD. Although a causal relationship between a PDA and BPD has not been established, many clinicians attempt to close PDA under the assumption that early closure decreases the likelihood of BPD and other morbidities.20 The Cochrane reviews include two strategies to promote closure of PDA: prophylactic treatment with cyclo-oxygenase (COX) inhibitors of infants at risk for PDA and treatment with COX inhibitors of infants with asymptomatic PDA identied by echocardiography.21,22 The review of prophylactic treatment with the COX inhibitor indomethacin suggests that there are several short-term benets of this therapy, including a reduction in the incidence of symptomatic PDA, serious intraventricular hemorrhage, and surgical ligation of PDAs. However, despite improvements in these outcomes, the incidence of BPD is not reduced.23 Treatment with indomethacin of infants with asymptomatic PDA appears to result in similar benets.22 There have been no reports of large, controlled trials of indomethacin for the treatment of symptomatic PDA conducted in the postsurfactant era. However, a comprehensive report of older trials suggests a similar lack of reduction of BPD using this strategy.24 Another COX inhibitor,

2.4. Aggressive phototherapy Bilirubin is an efcient scavenger of oxygen and peroxyl radicals.30 Because oxidative injury appears to be critical in the pathogenesis of BPD,31 bilirubin may have a protective effect in the lung and decrease the likelihood of developing BPD. This possibility is supported by observations from a single-center retrospective study in which infants with moderate and severe BPD had lower total bilirubin levels on the seventh postnatal day compared with infants who did not develop BPD.32 However, in this study, infants with moderate and severe BPD had lower birth weights, and adjustments were not made for these differences or for exposure to phototherapy in the analyses. More denitive evidence about the potential inuence of bilirubin and phototherapy on the development of BPD is available from a study in a large cohort of extremely

376

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

low birth weight (ELBW; <1000 g birth weight) infants designed to examine the effect of phototherapy on death and neurodevelopment. In this study, infants were randomized either to aggressive phototherapy (starting at bilirubin levels of 5 mg/dL for all infants) or to conservative phototherapy (bilirubin levels were allowed to rise to 8 mg/dL for infants <750 g birth weight and 10 mg/dL for infants 7511000 g birth weight).33 There was no difference in mortality between the infants assigned to aggressive phototherapy (21%) compared with those assigned to conservative phototherapy (20%). Although BPD was not the primary outcome of the study, its rates were reported as a secondary outcome. Contrary to expectations based on the hypothesized role of bilirubin in oxidative injury, the rate of BPD was signicantly lower in the aggressive phototherapy group compared with the conservative phototherapy group (41% vs 48%; OR: 0.86; 95% CI: 0.780.96). Evidence supports the use of aggressive phototherapy in infants with birth weights <1000 g, with the probability that a secondary benet will be a reduction in BPD.

3. Anti-inammatory therapies 3.1. Systemic corticosteroids: dexamethasone and hydrocortisone Corticosteroids affects pulmonary function and lung disease through several mechanisms. Fetal exposure causes increased surfactant synthesis and lung epithelial differentiation. In animals, early postnatal exposure causes increased surfactant synthesis. However, their role in modulating lung inammation is the primary mechanism through which they modify the likelihood of developing BPD.34 Corticosteroids decrease recruitment of polymorphonuclear leukocytes to the lung, and reduce the production of prostaglandins, leukotrienes, elastase and other inammatory mediators, and decrease vascular permeability and pulmonary edema formation.34 Corticosteroids may also modulate repair after lung injury by reducing bronectin production, and subsequent brosis, and by increasing retinol concentrations (see Vitamin A, below). The use of corticosteroids for the prevention or treatment of BPD has been examined in numerous clinical trials over the past 25 years. Most have demonstrated short-term improvements in pulmonary function. Although these studies are not easily combined in meta-analyses because of variability in study design, two Cochrane reviews provide information about the effect of treatment with systemic steroids on the incidence of BPD. The reviews differ based on the timing of treatment. Early treatment was dened as beginning before eight postnatal days35; late treatment was dened as beginning after 7 days postnatal age.36 The review of early treatment examined 28 trials (20 used dexamethasone, eight used hydrocortisone) that enrolled 3740 infants. Benets of early treatment included decreased risks of BPD at both 28 days and 36 weeks PMA (RR: 0.79; 95% CI: 0.710.88) and death or BPD at 28 days and at 36 weeks PMA (RR: 0.89; 95% CI: 0.840.95). There were no differences in the incidence of mortality. Twelve trials examined the neurodevelopmental impact of corticosteroids and demonstrated an increase in the risk of cerebral palsy (RR: 1.45; 95% CI: 1.061.98). The interpretation of neurodevelopmental outcomes in these studies is challenging because of varying quality of assessment across studies. However, impairment of motor function appears to be a risk associated with early treatment. The review of late treatment examined 19 trials that enrolled 1345 infants. Benets of late treatment included decreased risks of BPD at both 28 days and 36 weeks PMA (RR: 0.72; 95% CI: 0.610.85) and death or BPD at 28 days and 36 weeks PMA

(RR: 0.72; 95% CI: 0.630.82). There was an increased risk of hyperglycemia, hypertension, and hypertrophic cardiomyopathy among treated infants. There were no increases in the risk of blindness, deafness, major neurosensory disability, or cerebral palsy. However, there was an increased risk of abnormal neurological examination. The implications of this nding are unclear in the absence of an increase in neurosensory impairment or cerebral palsy. Given the concerns regarding the adverse effects of dexamethasone, hydrocortisone has been examined separately as an alternative therapy to prevent BPD. Eight of the trials reviewed in the Cochrane meta-analysis on early corticosteroids used hydrocortisone.35 Hydrocortisone had little effect on the incidence of BPD at 28 days or 36 weeks PMA (RR: 0.96; 95% CI: 0.821.12) and there was an increased risk of gastrointestinal perforations (RR: 2.02; 95% CI: 1.133.59). Early hydrocortisone was not associated with neurodevelopmental impairment. Although the risk of corticosteroid treatment for the prevention of BPD is not justied in all premature infants, there might be some populations in which the benets outweigh harm. Doyle et al. suggest that the balance between benet and risk may be modied by baseline risk for BPD.37 In a meta-regression analysis of previous RCTs of corticosteroids, infants at high risk of developing BPD treated with corticosteroids were shown to be more likely to survive without neurodevelopmental impairment. This study suggested that a strategy of treatment would have net benet if the risk for BPD of a population reached a certain threshold. Presumably, the decrease in the risk of cerebral palsy associated with BPD, by virtue of the benecial effects of steroids in preventing BPD, outweighs the increase in CP caused by steroids. Evidence does not support the use of corticosteroids for the prevention of BPD in populations of premature infants at relatively low risk of the disease. Caution in the use of corticosteroids has been advised by the American Academy of Pediatrics which recommends that systemic corticosteroids not be used outside the context of a clinical trial.38 However, there might be populations with a high baseline risk in which a reduction in BPD would be accompanied by improved survival without neurodevelopmental impairment. Unfortunately, at this time, we do not have a mechanism to identify these populations. Therefore, any strategies for the use of corticosteroids for the prevention of BPD are not supported by evidence.

3.2. Inhaled corticosteroids In addition to systemic therapy, corticosteroids may also be administered via an inhaler or, more commonly, via nebulization. Eleven trials were analyzed in a Cochrane review examining the effect of early inhaled steroids beginning at <2 weeks postnatal age in ventilator-dependent infants on the incidence of BPD.39 There was no signicant effect of inhaled steroids on either BPD (RR: 0.97; 95% CI: 0.621.52) or mortality. The use of inhaled steroids for the prevention of BPD in infants not receiving mechanical ventilation has not been evaluated. Evidence does not support the use of inhaled corticosteroids for the prevention of BPD.

3.3. Inhaled nitric oxide Inhaled NO (iNO) improves oxygenation and reduces the need for extracorporeal membrane oxygenation in term and near-term

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

377

infants with respiratory failure from a variety of causes, including meconium aspiration syndrome, sepsis, and idiopathic persistent pulmonary hypertension.40 Previous studies suggest that iNO can also improve gas exchange in premature newborns with hypoxemia caused by the RDS or persistent pulmonary hypertension.41 iNO may also have anti-inammatory properties. In animal studies, iNO decreased early lung inammation and oxidant stress,42 and improved lung structure in models of BPD.43 These ndings have stimulated the investigation of the potential benet of iNO in reducing the risk or severity of BPD in premature infants. Interpreting the results of RCTs that examined the efcacy of iNO in the prevention of BPD of these trials is challenging because of variability in study design, most notably variable postnatal age of enrollment and baseline risk of BPD. Eleven RCTs of iNO therapy in preterm infants were included in a Cochrane review.44 To account for variation in study design, the authors divided the trials into three categories sorted by entry criteria: (1) enrollment during the rst 3 postnatal days based on oxygenation criteria; (2) enrollment of all intubated infants;45,46 and (3) enrollment after the third postnatal day based on increased risk of BPD.47,48 Only analyses of these subgroups were performed. Trials of early treatment of infants enrolled based on oxygenation criteria demonstrated no signicant effect of iNO on mortality or BPD. Studies of iNO in intubated infants demonstrated a marginally signicant reduction in the combined outcome of death or BPD (RR: 0.91; 95% CI: 0.840.99). Late treatment based on the risk of BPD demonstrated no reduction in death or BPD. However, one of the larger trials which enrolled intubated infants during the second postnatal week demonstrated a reduction in BPD (50.7% in the iNO group compared with 56.9% in the placebo group) but not death (5.4% vs 6.3%).48 Evidence does not support a clear role for iNO for the prevention of BPD. It is possible that certain populations dened by postnatal age and baseline risk might benet, but identifying these populations is not possible at this time. 3.4. Superoxide dismutase Superoxide dismutase (SOD) is an enzyme that catalyzes the free oxygen radical superoxide (O 2) into oxygen and hydrogen peroxide. Superoxide is toxic and promotes oxidative damage in the lungs.49 In a piglet model of acute lung injury caused by hyperoxia and barotrauma, intratracheal administration of SOD improved lung compliance, and decreased neutrophil chemotactic activity, total cell counts, and elastase activity recovered from tracheal aspirates.49 SOD has been investigated in two RCTs to assess its effect in preventing BPD and there was no reduction.50,51 However, 1-year follow-up in one trial demonstrated that SOD might reduce the use of treatments with medications such as inhaled corticosteroids and/or albuterol used to treat asthma.52 Evidence does not support the use of SOD for the prevention of BPD. 3.5. Glutathione Lung inammation during mechanical ventilation, particularly with coincidental exposure to high fractions of inspired oxygen, results in the generation of reactive oxygen species and proinammatory cytokines.53 These mediators may damage cells of the lungs and can impair pulmonary development. Glutathione is an antioxidant that may inhibit some of these changes.

Concentrations of glutathione are relatively decient in preterm infants, and further deciency may occur because of limited availability of its precursor amino acid, cysteine.54 Supplementation of glutathione has been suggested as a strategy for minimizing the effects of oxidative injury. Because glutathione does not readily cross cell membranes and cysteine is unstable in solution, investigators have examined the ability of Nacetylcysteine, a precursor of cysteine, to prevent BPD. In a multicenter RCT of 391 ELBW infants, no difference in BPD or death was observed between infants randomized to receive N-acetylcysteine compared with placebo (51% vs 49%).55 Levels of cysteine and glutathione were actually lower among infants in the treatment group on days 3 and 7 of the study. This nding might have resulted from the early administration of parenteral nutrition supplemented with cysteine to both groups of infants. Evidence does not support treatment with glutathione or its precursors for the prevention of BPD.

4. Other medications 4.1. Vitamin A Vitamin A derivatives dene a group of fat-soluble compounds called retinoids. These compounds appear to play an important role in lung disease because they are critical in the regulation and promotion of growth and differentiation of lung epithelial cells, particularly during repair after lung injury.56 In animals, deciency of vitamin A results in abnormal lung growth following prolonged exposure to hyperoxia.57 Preterm infants have low vitamin A levels at birth, and low levels of vitamin A are associated with an increased risk of BPD.58 Therefore, vitamin A supplementation was developed as a strategy to prevent BPD. In a Cochrane review, eight studies that examined the efcacy of vitamin A supplementation to prevent BPD were identied. In ELBW infants, supplementation reduced the incidence of BPD, dened as receipt of oxygen at 36 weeks PMA (RR: 0.87; CI: 0.770.98), with a number needed to treat of 13 (95% CI: 7100).59 In the largest study in the review, among infants with birth weight <1000 g, death or BPD was lower in the vitamin A group compared with placebo group (55% vs 62%; RR: 0.89; 95 CI: 0.800.99).60 Neurodevelopment at 1822 months of age was not different between groups.61 Evidence supports the use of vitamin A supplementation in infants with birth weights <1000 g and meeting other eligibility requirements of the RCTs for the prevention of BPD. 4.2. H2 blockers: cimetidine Cimetidine is an H2 blocker that also inhibits cytochrome P450. Because upregulation of cytochrome P450 may be important during lung injury, its inhibition may be protective. However, an RCT evaluating a 10-day infusion of cimetidine beginning in the rst 24 h after birth in infants with birth weights <1250 g demonstrated no reduction in death or BPD at 36 weeks PMA.62 No signicant differences were identied for other earlier markers of pulmonary disease. The study was stopped after enrollment of 84 patients because of a signicant increase in death or severe intraventricular hemorrhage among infants receiving cimetidine. Evidence does not support treatment with cimetidine for the prevention of BPD.

378

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

4.3. Treatment of ureaplasma with macrolides Although there is an association between colonization of the respiratory tract with Ureaplasma and the development of BPD, a causal link between colonization and the disease remains uncertain. In addition, treatments that both eradicate colonization and reduce BPD remain elusive.63 For example, treatment of infants colonized with Ureaplasma with erythromycin does not appear to reduce the incidence of BPD.64 Treatment with other macrolides, such as azithromycin, have been investigated because these drugs also have anti-inammatory properties. However, in a small, single-center pilot study, azithromycin therapy reduced the incidence of treatment with dexamethasone, but did not reduce the incidence of BPD.65 Evidence does not support treatment of colonization of the respiratory tract with Ureaplasma sp. for the prevention of BPD. However, because of the strong association between colonization and BPD, strategies to prevent or treat colonization warrant further investigation.

have yielded mixed results.77,78 Studies of the impact on the incidence of BPD of two other methods of high frequency ventilation, high frequency jet ventilation and the high frequency ow interrupter, are inconclusive.71,7981 Current evidence does not support the routine use of HFV for the prevention of BPD. 5.3. Continuous positive airway pressure The use of nasal continuous positive airway pressure (CPAP) appears to be a successful strategy for avoiding the need for mechanical ventilation in some infants, with the presumptive benet of decreasing the risk of BPD.82,83 In an observational cohort study, an association between the early use of nasal CPAP and decreased risk of BPD was reported.84 However, to date, a reduction in BPD has not been demonstrated in RCTs of early CPAP compared with conventional ventilation. For example, in an RCT of 104 infants born at <28 weeks of gestation, infants receiving nasal CPAP in the delivery room had a rate of BPD (29.4%), comparable with those treated with mechanical ventilation (27.9%), and more infants treated with CPAP died (27% vs 13%).85 More recently, a larger RCT compared the initiation of nasal CPAP with intubation and mechanical ventilation in the delivery room among 610 infants between 25 and 28 weeks of gestation at birth.82 Although there was a reduction in the combined outcome of death or oxygen requirement at 28 days of age among infants treated with early CPAP compared with those who were ventilated (OR: 0.63; 95% CI; 0.460.88), no difference was observed at 36 weeks PMA. These studies suggest that the benets reported in observational studies associated with the early CPAP as a strategy for avoiding mechanical ventilation might be a marker for other center-specic care practices that have not been elucidated to date. A large trial in the National Institute of Child Health and Human Development Neonatal Research Network comparing the use of nasal CPAP with intubation and surfactant therapy has completed enrollment.86 The results of this trial may help to answer the unresolved questions about this potentially benecial strategy. Current evidence from RCTs does not support the routine use of CPAP to avoid mechanical ventilation and to prevent BPD. However, the evidence in support of this strategy from contemporary observational studies is compelling. It is seems likely that the judicious use of CPAP immediately after birth, in lieu of mechanical ventilation or after a brief period of ventilation, in carefully selected patient populations minimizes the likelihood of developing BPD. This potential benet may be augmented by a brief period of ventilation for surfactant administration (see Surfactant therapy above). However, the method for selecting the population is not certain at this time.

5. Ventilatory strategies 5.1. Permissive hypercapnea Mechanical ventilation, although necessary for survival for many preterm infants, injures lung tissue and is a risk factor for the development of BPD.66 When retrospective studies indicated that higher levels of carbon dioxide in the rst few days of life were associated with lower risk of BPD,67,68 several investigators prospectively studied the practice of permissive hypercapnea.6971 The largest of these trials randomized 220 infants to a target PaCO2 >52 mmHg (minimal ventilation) or <48 mmHg (routine care).70 No difference in the incidence of BPD was observed, possibly because the separation of PaCO2 between groups was only 4.0 1.3 mmHg. A similar nding of a small difference in PaCO2 was demonstrated in a subsequent study.69 A larger difference in PaCO2 might have demonstrated a difference in outcome. Although the practice of permissive hypercapnea has been widely adopted, its contribution to reducing the overall burden of BPD is likely to be small if benets exists because of the pervasive practice and success of avoidance of ventilation. Evidence is inconclusive that permissive hypercapnea is effective for the prevention of BPD. 5.2. High frequency ventilation Many clinicians use some form of high frequency ventilation (HFV), in lieu of conventional mechanical (tidal) ventilation (CMV), in selected infants under the assumption that HFV is less likely to induce lung injury. The mechanisms by which HFV might reduce lung injury, thereby decreasing the incidence of BPD, include: (1) reducing regional lung overination; (2) minimizing volutrauma; (3) minimizing pressure changes at the alveolar level, and (4) lowering oxygen requirements.7274 The initial large RCT comparing one form of HFV, high frequency oscillatory ventilation (HFOV), with CMV was conducted in the pre-surfactant era.75 This study found no difference in death or BPD between the HFOV and CMV groups, but was criticized because of the manner in which HFOV was employed. A Cochrane review that includes 11 studies of comparing HFOV with CMV demonstrates a small decrease in the relative risk of death or BPD at 36 weeks in infants treated with HFOV (RR: 0.90; 95% CI: 0.830.97).76 Subsequent studies of HFOV

6. Nutritional and uid therapies 6.1. Nutritional support The quality and quantity of energy substrates provided to premature infants might play a role in the development of BPD. Premature infants may be vulnerable to the effects of nutritional deprivation because they are born with lower glycogen stores and body fat compared with term infants. Poor caloric intake during respiratory illness may result in respiratory muscle fatigue and a longer duration of mechanical ventilation.87 In one casecontrol study, infants who developed BPD had signicantly lower mean energy intakes than matched controls.88 In an unmasked RCT, 125

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

379

infants <1500 g birth weight were allocated either to standard parenteral nutrition (maximum of 2 g/kg/day lipid, 2.5 mg/kg/day amino acids, 10% dextrose solution, standard enteral feeding schedule) or to aggressive nutrition (maximum of 3.5 g/kg/day of lipids, 3.5 g/kg/day amino acids, 15% dextrose solution, aggressive enteral feeding schedule). Infants in the aggressive nutrition cohort received more caloric intake and experienced faster growth. However, pulmonary outcomes including duration of mechanical ventilation, development of BPD (oxygen requirement at 28 days of life), and need for supplemental oxygen at term were similar between groups.89 Preterm infants are born with smaller fat and protein stores compared with infants born at term. Lipids are normally administered as part of total parenteral nutrition in the rst few postnatal days. Polyunsaturated fatty acids can act as an antioxidant. Thus, in addition to providing additional calories for growth, lipids are high in polyunsaturated fatty acids and might protect the lung from oxygen toxicity.90 In a study examining the potential benets of early lipid supplementation, 183 ELBW infants were randomized to lipid administration either beginning in the rst 12 h of life or after the 7th day of life, and the primary outcome was death or BPD.91 The study was stopped because of a signicant increase in mortality among the smallest infants (birth weights 600800 g) in the early lipid group. There was no difference between groups in the entire cohort in the combined outcome of death or BPD. There does not appear to be a nutritional strategy that decreases the likelihood of developing BPD. A prudent approach would be to avoid malnutrition. 6.2. Fluid restriction and diuretic therapy Excessive lung water may interfere with respiration by impairing lung mechanics, and thus increasing the need for oxygen and ventilatory support. By these mechanisms, excessive lung water may play a role in the development of BPD. Extremely premature infants have a relatively high percentage of body water that is largely in the extracellular uid compartment.92 The normal weight loss that occurs in the rst few days after birth may not occur if inappropriately large volumes of uid are administered. Fluid restriction during this critical period of time, therefore, may help prevent BPD. This possibility was examined in a cohort of 1382 ELBW infants. Higher uid intake and lack of weight loss during the rst postnatal week, even after adjustment for factors known to predict BPD, were associated with a higher risk of death and BPD at 36 weeks PMA.93 A Cochrane review examined restricted versus liberal water intake for preventing morbidity and mortality in preterm infants and included ve trials.94 The incidence of PDA and necrotizing enterocolitis was lower in the restricted water intake groups. However, the risk of BPD was no different between groups dened by water intake (RR: 0.85; 95% CI: 0.631.14). Because sodium supplementation favours expansion of the extracellular uid compartment, early administration of sodium may also adversely affect risk of BPD. Forty-six infants were randomized either to sodium supplementation after 6% weight loss following birth (delayed supplementation) or to supplementation beginning on the second day of life (early supplementation).95 Although there was a lower proportion of infants receiving oxygen at 28 days in the delayed sodium supplementation group, by 36 weeks PMA, there was no difference in death or BPD in the early compared with the delayed group (54% vs 41%; P 0.38). Because pulmonary edema appears to be a prominent feature of BPD, diuretics are very commonly used to improve lung mechanics in infants with established disease. There is far less experience with the use of diuretics early in life for the prevention of BPD. Two Cochrane reviews examine the use of loop diuretics and diuretics

acting on the distal renal tubule.96,97 They include small studies of single dose or short-course therapy and only report short-term outcomes including: extubation rates, change in lung compliance, and change in fraction of inspired oxygen.98,99 The potential of these drugs to prevent BPD cannot be determined from these studies. An appropriate fractional loss of weight immediately after birth, compared with persistence of excessive extracellular uid, appears to decrease the likelihood of BPD. Strategies for the administration of free water and sodium during the rst week of life should include this goal. Evidence does not support the use of diuretics to further decrease exrtacellular uid for the prevention of BPD.

7. Quality improvement techniques There has been considerable recent interest in using quality improvement or implementation science methodologies to reduce morbidities associated with prematurity, including BPD. This interest has been stimulated by the observation that there is great variability in outcomes such as BPD among centers, even after adjustment for confounding risk factors (e.g. birth weight distribution). The assumption is that variability in practice accounts for the differences in outcome, and that uniform application of best practices could improve outcomes in poorly performing centers. The success of these techniques in reducing BPD has been reported from single centers and uses a beforeafter study design. That is, the incidence of BPD is measured, an intervention or panel of interventions is implemented, and the incidence of BPD is measured again.100,101 Using a similar, beforeafter design to compare outcomes in 2001 vs 2003, 19 hospitals in the Vermont Oxford Network reported a higher incidence of survival without BPD after initiation of a package of potentially better practices (63.4% vs 53.6%; OR: 1.86; CI: 1.412.46).102 Hospitals used conventional ventilation, postnatal steroids, and supplemental oxygen less frequently, increased the use of nasal CPAP, and reduced the median time to rst surfactant dose from 22 to 10 min after delivery after initiation of quality improvement techniques. In a study using a cluster, randomized, controlled design, seven intervention centers received quality improvement training that included potentially better practices. The outcomes after this intervention were compared with outcomes in seven control centers.38 Intervention centers decreased the time of administration of the rst surfactant dose (from 51 to 31 min), increased the use of nasal CPAP on the rst day (from 16.9% to 24.2%), and decreased the duration of mechanical ventilation in the rst week of life (from 4.0 2.7 to 3.5 2.8 days). However, despite successful implementation of quality improvement techniques and changes in these short-term outcomes, the rate of BPD was similar between the intervention and the control centers (38.5% vs 36.1%). Incorporation of quality improvement methodology into clinical practice has vast potential for improving outcomes and will unquestionably impact outcomes in neonatal medicine. However, the paucity of evidenced-based practices for the prevention of BPD makes the development of bundles of best practices, and therefore the application of quality improvement methodology, difcult. Whether the utilization of these strategies can reduce the incidence of BPD is as yet uncertain.

8. Summary Various treatments have been used to prevent BPD, but only two pharmacological therapies, vitamin A supplementation and

380

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382

treatment with systemic corticosteroids, have proven efcacy based on RCTs. Although not used specically for the prevention of BPD, the treatment of apnea of prematurity with caffeine and the use of aggressive phototherapy in ELBW infants are also associated with reductions in BPD. Individually, these treatments reduce risk by 711%. Unfortunately, treatment with systemic corticosteroids (particularly dexamethasone), although effective in reducing the risk of BPD, is associated with an increased rate of neurodevelopmental impairment. The benets of treatment with systemic corticosteroids may outweigh the risks in infants with high baseline risk of BPD. The use of CPAP in selected populations of infants, in lieu of mechanical ventilation, may also decrease the incidence of BPD, but further data are needed to dene this population. The use of quality improvement methodologies has potential for reducing BPD, but will rely upon high quality evidence from RCTs that support bundles of best practices. Because of the economic impact and long-term consequences of BPD, new preventive therapies are desirable. Future trials that test these therapies should incorporate strategies to systematically quantify the risk of BPD prior to enrollment. To date, a method for predicting BPD with sufciently high sensitivity and specicity has not been available. A new tool using clinical and demographic variables appears promising.103 Because of the heterogeneity of the causal pathways that lead to the development of BPD, it is unlikely that a single preventive strategy will have a major impact on its reduction. Rather, several strategies, with the expectation that each will contribute to a modest reduction in BPD, will need to be tested in RCTs.

References
1. McGuire W, Fowlie P, Soll R. What has the Cochrane collaboration ever done for newborn infants? Arch Dis Child Fetal Neonatal Ed 2009. [Epub ahead of print]. 2. Gambrill E. Evidence-based clinical practice, [corrected] evidence-based medicine and the Cochrane collaboration. J Behav Ther Exp Psychiatry 1999;30:114. 3. Soll RF, Morley CJ. Prophylactic versus selective use of surfactant for preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2000. CD000510. 4. Soll RF. Synthetic surfactant for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2000. CD001149. 5. Soll RF. Prophylactic synthetic surfactant for preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2000. CD001079. 6. Soll RF. Prophylactic natural surfactant extract for preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2000. CD000511. 7. Stevens TP, Harrington EW, Blennow M, Soll RF. Early surfactant administration with brief ventilation vs selective surfactant and continued mechanical ventilation for preterm infants with or at risk for respiratory distress syndrome. Cochrane Database Syst Rev; 2007. CD003063. 8. Verder H, Albertsen P, Ebbesen F, et al. Nasal continuous positive airway pressure and early surfactant therapy for respiratory distress syndrome in newborns of less than 30 weeks gestation. Pediatrics 1999;103:E24. 9. Verder H, Robertson B, Greisen G, et al. Surfactant therapy and nasal continuous positive airway pressure for newborns with respiratory distress syndrome. DanishSwedish Multicenter Study Group. N Engl J Med 1994;331:10515. 10. Reininger A, Khalak R, Kendig JW, et al. Surfactant administration by transient intubation in infants 29 to 35 weeks gestation with respiratory distress syndrome decreases the likelihood of later mechanical ventilation: a randomized controlled trial. J Perinatol 2005;25:7038. 11. Stevens TP, Blennow M, Soll RF. Early surfactant administration with brief ventilation vs selective surfactant and continued mechanical ventilation for preterm infants with or at risk for RDS. Cochrane Database Syst Rev; 2002. CD003063. 12. Rojas MA, Lozano JM, Rojas MX, et al. Very early surfactant without mandatory ventilation in premature infants treated with early continuous positive airway pressure: a randomized, controlled trial. Pediatrics 2009;123:13742. 13. Merrill JD, Ballard RA, Cnaan A, et al. Dysfunction of pulmonary surfactant in chronically ventilated premature infants. Pediatr Res 2004;56:91826. 14. Laughon M, Bose C, Moya F, et al. A pilot randomized, controlled trial of later treatment with a peptide-containing, synthetic surfactant for the prevention of bronchopulmonary dysplasia. Pediatrics 2009;123:8996. 15. Marshall DD, Kotelchuck M, Young TE, et al. Risk factors for chronic lung disease in the surfactant era: a North Carolina population-based study of very low birth weight infants. North Carolina Neonatologists Association. Pediatrics 1999;104:134550. 16. Rojas MA, Gonzalez A, Bancalari E, et al. Changing trends in the epidemiology and pathogenesis of neonatal chronic lung disease. J Pediatr 1995;126:60510. 17. Stefano JL, Abbasi S, Pearlman SA, et al. Closure of the ductus arteriosus with indomethacin in ventilated neonates with respiratory distress syndrome. Effects of pulmonary compliance and ventilation. Am Rev Respir Dis 1991;143:2369. 18. Yeh TF, Luken JA, Thalji A, et al. Intravenous indomethacin therapy in premature infants with persistent ductus arteriosus a double-blind controlled study. J Pediatr 1981;98:13745. 19. Van Marter LJ, Pagano M, Allred EN, et al. Rate of bronchopulmonary dysplasia as a function of neonatal intensive care practices. J Pediatr 1992;120:93846. 20. Clyman RI. Recommendations for the postnatal use of indomethacin: an analysis of four separate treatment strategies. J Pediatr 1996;128:6017. 21. Fowlie PW. Intravenous indomethacin for preventing mortality and morbidity in very low birth weight infants. Cochrane Database Syst Rev; 2000. CD000174. 22. Cooke L, Steer P, Woodgate P. Indomethacin for asymptomatic patent ductus arteriosus in preterm infants. Cochrane Database Syst Rev; 2003. CD003745. 23. Schmidt B, Davis P, Moddemann D, et al. Long-term effects of indomethacin prophylaxis in extremely-low-birth-weight infants. N Engl J Med 2001;344: 196672. 24. Knight DB. The treatment of patent ductus arteriosus in preterm infants. A review and overview of randomized trials. Semin Neonatol 2001;6:6373. 25. Ohlsson A, Walia R, Shah S. Ibuprofen for the treatment of patent ductus arteriosus in preterm and/or low birth weight infants. Cochrane Database Syst Rev; 2008. CD003481. 26. Cassady G, Crouse DT, Kirklin JW, et al. A randomized, controlled trial of very early prophylactic ligation of the ductus arteriosus in babies who weighed 1000 g or less at birth. N Engl J Med 1989;320:15116. 27. Clyman R, Cassady G, Kirklin JK, et al. The role of patent ductus arteriosus ligation in bronchopulmonary dysplasia: reexamining a randomized controlled trial. J Pediatr 2009;154:8736. 28. Schmidt B, Roberts RS, Davis P, et al. Long-term effects of caffeine therapy for apnea of prematurity. N Engl J Med 2007;357:1893902. 29. Schmidt B, Roberts RS, Davis P, et al. Caffeine therapy for apnea of prematurity. N Engl J Med 2006;354:211221. 30. Stocker R, Glazer AN, Ames BN. Antioxidant activity of albumin-bound bilirubin. Proc Natl Acad Sci U S A 1987;84:591822.

Practice points  BPD is an important morbidity associated with premature birth.  The prevention strategies with the highest quality evidence with most favorable benet/risk ratio include vitamin A and caffeine.  Corticosteroids reduce the incidence of BPD, but increase the risk of abnormal neurologic examination.

Research directions  A simple, clinically relevant, predictive model that objectively assesses the risk of BPD needs to be developed.  Well-powered trials of surfactant therapy with brief ventilation, and later surfactant therapy with the primary endpoint of CLD, are needed.  An RCT of systemic corticosteroids versus placebo among patients at high risk of bronchopulmonary dysplasia, with appropriate neurodevelopmental followup, is needed.

Conict of interest statement None declared. Funding sources None.

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382 31. Fardy CH, Silverman M. Antioxidants in neonatal lung disease. Arch Dis Child Fetal Neonatal Ed 1995;73:F1127. 32. Schrod L, von Stockhausen HB. Serum bilirubin and development of chronic lung disease in very low birth weight infants. Eur J Pediatr 1999;158: 16970. 33. Morris BH, Oh W, Tyson JE, et al. Aggressive vs conservative phototherapy for infants with extremely low birth weight. N Engl J Med 2008;359:188596. 34. Bancalari E. Corticosteroids and neonatal chronic lung disease. Eur J Pediatr 1998;157(Suppl. 1):S317. 35. Halliday HL, Ehrenkranz RA, Doyle LW. Early (<8 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev; 2009. CD001146. 36. Halliday HL, Ehrenkranz RA, Doyle LW. Late (>7 days) postnatal corticosteroids for chronic lung disease in preterm infants. Cochrane Database Syst Rev; 2009. CD001145. 37. Doyle LW, Halliday HL, Ehrenkranz RA, et al. Impact of postnatal systemic corticosteroids on mortality and cerebral palsy in preterm infants: effect modication by risk for chronic lung disease. Pediatrics 2005;115:65561. 38. Walsh M, Laptook A, Kazzi SN, et al. A cluster-randomized trial of benchmarking and multimodal quality improvement to improve rates of survival free of bronchopulmonary dysplasia for infants with birth weights of less than 1250 grams. Pediatrics 2007;119:87690. 39. Shah V, Ohlsson A, Halliday HL, Dunn MS. Early administration of inhaled corticosteroids for preventing chronic lung disease in ventilated very low birth weight preterm neonates. Cochrane Database Syst Rev; 2007. CD001969. 40. Finer NN, Barrington KJ. Nitric oxide for respiratory failure in infants born at or near term. Cochrane Database Syst Rev; 2006. CD000399. 41. Abman SH, Kinsella JP, Schaffer MS, Wilkening RB. Inhaled nitric oxide in the management of a premature newborn with severe respiratory distress and pulmonary hypertension. Pediatrics 1993;92:6069. 42. Kinsella JP, Ivy DD, Abman SH. Inhaled nitric oxide improves gas exchange and lowers pulmonary vascular resistance in severe experimental hyaline membrane disease. Pediatr Res 1994;36:4028. 43. McCurnin DC, Pierce RA, Chang LY, et al. Inhaled NO improves early pulmonary function and modies lung growth and elastin deposition in a baboon model of neonatal chronic lung disease. Am J Physiol Lung Cell Mol Physiol 2005;288:L4509. 44. Barrington KJ, Finer NN. Inhaled nitric oxide for respiratory failure in preterm infants. Cochrane Database Syst Rev; 2007. CD000509. 45. Schreiber MD, Gin-Mestan K, Marks JD, et al. Inhaled nitric oxide in premature infants with the respiratory distress syndrome. N Engl J Med 2003;349:2099107. 46. Kinsella JP, Cutter GR, Walsh WF, et al. Early inhaled nitric oxide therapy in premature newborns with respiratory failure. N Engl J Med 2006;355:35464. 47. Subhedar NV, Ryan SW, Shaw NJ. Open randomised controlled trial of inhaled nitric oxide and early dexamethasone in high risk preterm infants. Arch Dis Child Fetal Neonatal Ed 1997;77:F18590. 48. Ballard RA, Truog WE, Cnaan A, et al. Inhaled nitric oxide in preterm infants undergoing mechanical ventilation. N Engl J Med 2006;355:34353. 49. Davis JM, Rosenfeld WN, Sanders RJ, Gonenne A. Prophylactic effects of recombinant human superoxide dismutase in neonatal lung injury. J Appl Physiol 1993;74:223441. 50. Rosenfeld W, Evans H, Concepcion L, et al. Prevention of bronchopulmonary dysplasia by administration of bovine superoxide dismutase in preterm infants with respiratory distress syndrome. J Pediatr 1984;105:7815. 51. Davis JM, Rosenfeld WN, Richter SE, et al. Safety and pharmacokinetics of multiple doses of recombinant human CuZn superoxide dismutase administered intratracheally to premature neonates with respiratory distress syndrome. Pediatrics 1997;100:2430. 52. Davis JM, Parad RB, Michele T, et al. Pulmonary outcome at 1 year corrected age in premature infants treated at birth with recombinant human CuZn superoxide dismutase. Pediatrics 2003;111:46976. 53. Speer CP. New insights into the pathogenesis of pulmonary inammation in preterm infants. Biol Neonate 2001;79:2059. 54. Ahola T, Levonen AL, Fellman V, Lapatto R. Thiol metabolism in preterm infants during the rst week of life. Scand J Clin Lab Invest 2004;64:64958. 55. Ahola T, Lapatto R, Raivio KO, et al. N-Acetylcysteine does not prevent bronchopulmonary dysplasia in immature infants: a randomized controlled trial. J Pediatr 2003;143:7139. 56. Veness-Meehan KA. Effects of retinol deciency and hyperoxia on collagen gene expression in rat lung. Exp Lung Res 1997;23:56981. 57. Veness-Meehan KA, Bottone Jr FG, Stiles AD. Effects of retinoic acid on airspace development and lung collagen in hyperoxia-exposed newborn rats. Pediatr Res 2000;48:43444. 58. Kennedy KA. Epidemiology of acute and chronic lung injury. Semin Perinatol 1993;17:24752. 59. Darlow BA, Graham PJ. Vitamin A supplementation to prevent mortality and short and long-term morbidity in very low birthweight infants. Cochrane Database Syst Rev; 2007. CD000501. 60. Tyson JE, Wright LL, Oh W, et al. Vitamin A supplementation for extremelylow-birth-weight infants. National Institute of Child Health and Human Development Neonatal Research Network. N Engl J Med 1999;340:19628. 61. Ambalavanan N, Tyson JE, Kennedy KA, et al. Vitamin A supplementation for extremely low birth weight infants: outcome at 18 to 22 months. Pediatrics 2005;115:e24954.

381

62. Cotton RB, Hazinski TA, Morrow JD, et al. Cimetidine does not prevent lung injury in newborn premature infants. Pediatr Res 2006;59:795800. 63. Schelonka RL, Katz B, Waites KB, Benjamin Jr DK. Critical appraisal of the role of Ureaplasma in the development of bronchopulmonary dysplasia with metaanalytic techniques. Pediatr Infect Dis J 2005;24:10339. 64. Mabanta CG, Pryhuber GS, Weinberg GA, Phelps DL. Erythromycin for the prevention of chronic lung disease in intubated preterm infants at risk for, or colonized or infected with Ureaplasma urealyticum. Cochrane Database Syst Rev; 2003. CD003744. 65. Ballard HO, Anstead MI, Shook LA. Azithromycin in the extremely low birth weight infant for the prevention of bronchopulmonary dysplasia: a pilot study. Respir Res 2007;8:41. 66. Jobe AH, Ikegami M. Mechanisms initiating lung injury in the preterm. Early Hum Dev 1998;53:8194. 67. Kraybill EN, Runyan DK, Bose CL, Khan JH. Risk factors for chronic lung disease in infants with birth weights of 751 to 1000 grams. J Pediatr 1989;115:11520. 68. Garland JS, Buck RK, Allred EN, Leviton A. Hypocarbia before surfactant therapy appears to increase bronchopulmonary dysplasia risk in infants with respiratory distress syndrome. Arch Pediatr Adolesc Med 1995;149:61722. 69. Mariani G, Cifuentes J, Carlo WA. Randomized trial of permissive hypercapnia in preterm infants. Pediatrics 1999;104:10828. 70. Carlo WA, Stark AR, Wright LL, et al. Minimal ventilation to prevent bronchopulmonary dysplasia in extremely-low-birth-weight infants. J Pediatr 2002;141:3704. 71. Thome U, Kossel H, Lipowsky G, et al. Randomized comparison of highfrequency ventilation with high-rate intermittent positive pressure ventilation in preterm infants with respiratory failure. J Pediatr 1999;135:3946. 72. Clark RH, Gerstmann DR, Null Jr DM, deLemos RA. Prospective randomized comparison of high-frequency oscillatory and conventional ventilation in respiratory distress syndrome. Pediatrics 1992;89:512. 73. Ogawa Y, Miyasaka K, Kawano T, et al. A multicenter randomized trial of high frequency oscillatory ventilation as compared with conventional mechanical ventilation in preterm infants with respiratory failure. Early Hum Dev 1993;32:110. 74. Frantz 3rd ID, Close RH. Alveolar pressure swings during high frequency ventilation in rabbits. Pediatr Res 1985;19:1626. 75. High-frequency oscillatory ventilation compared with conventional mechanical ventilation in the treatment of respiratory failure in preterm infants. The HIFI Study Group. N Engl J Med 1989;320:8893. 76. Henderson-Smart DJ, Cools F, Bhuta T, Offringa M. Elective high frequency oscillatory ventilation versus conventional ventilation for acute pulmonary dysfunction in preterm infants. Cochrane Database Syst Rev; 2007. CD000104. 77. Courtney SE, Durand DJ, Asselin JM, et al. High-frequency oscillatory ventilation versus conventional mechanical ventilation for very-low-birth-weight infants. N Engl J Med 2002;347:64352. 78. Johnson AH, Peacock JL, Greenough A, et al. High-frequency oscillatory ventilation for the prevention of chronic lung disease of prematurity. N Engl J Med 2002;347:63342. 79. Keszler M, Modanlou HD, Brudno DS, et al. Multicenter controlled clinical trial of high-frequency jet ventilation in preterm infants with uncomplicated respiratory distress syndrome. Pediatrics 1997;100:5939. 80. Wiswell TE, Graziani LJ, Kornhauser MS, et al. High-frequency jet ventilation in the early management of respiratory distress syndrome is associated with a greater risk for adverse outcomes. Pediatrics 1996;98:103543. 81. Craft AP, Bhandari V, Finer NN. The sy- study: a randomized prospective trial of synchronized intermittent mandatory ventilation versus a high-frequency ow interrupter in infants less than 1000 g. J Perinatol 2003;23:149. 82. Morley CJ, Davis PG, Doyle LW, et al. Nasal CPAP or intubation at birth for very preterm infants. N Engl J Med 2008;358:7008. 83. Davis PG, Henderson-Smart DJ. Nasal continuous positive airways pressure immediately after extubation for preventing morbidity in preterm infants. Cochrane Database Syst Rev; 2003. CD000143. 84. Van Marter LJ, Allred EN, Pagano M, et al. Do clinical markers of barotrauma and oxygen toxicity explain interhospital variation in rates of chronic lung disease? The Neonatology Committee for the Developmental Network. Pediatrics 2000;105:1194201. 85. Finer NN, Carlo WA, Duara S, et al. Delivery room continuous positive airway pressure/positive end-expiratory pressure in extremely low birth weight infants: a feasibility trial. Pediatrics 2004;114:6517. 86. Surfactant Positive Airway Pressure and Pulse Oximetry Trial (SUPPORT). Available at: http://clinicaltrials.gov/ct2/show/NCT00233324. Accessed August 31, 2009. 87. Frank L, Sosenko IR. Undernutrition as a major contributing factor in the pathogenesis of bronchopulmonary dysplasia. Am Rev Respir Dis 1988;138:7259. 88. Wilson DC, McClure G, Halliday HL, et al. Nutrition and bronchopulmonary dysplasia. Arch Dis Child 1991;66:378. 89. Wilson DC, Cairns P, Halliday HL, et al. Randomised controlled trial of an aggressive nutritional regimen in sick very low birthweight infants. Arch Dis Child Fetal Neonatal Ed 1997;77:F411. 90. Sosenko IR, Innis SM, Frank L. Polyunsaturated fatty acids and protection of newborn rats from oxygen toxicity. J Pediatr 1988;112:6307. 91. Sosenko IR, Rodriguez-Pierce M, Bancalari E. Effect of early initiation of intravenous lipid administration on the incidence and severity of chronic lung disease in premature infants. J Pediatr 1993;123:97582.

382

M.M. Laughon et al. / Seminars in Fetal & Neonatal Medicine 14 (2009) 374382 99. Hoffman DJ, Gerdes JS, Abbasi S. Pulmonary function and electrolyte balance following spironolactone treatment in preterm infants with chronic lung disease: a double-blind, placebo-controlled, randomized trial. J Perinatol 2000;20:415. 100. Kaempf JW, Campbell B, Sklar RS, et al. Implementing potentially better practices to improve neonatal outcomes after reducing postnatal dexamethasone use in infants born between 501 and 1250 grams. Pediatrics 2003;111:e53441. 101. Birenbaum HJ, Dentry A, Cirelli J, et al. Reduction in the incidence of chronic lung disease in very low birth weight infants: results of a quality improvement process in a tertiary level neonatal intensive care unit. Pediatrics 2009; 123:4450. 102. Payne NR, LaCorte M, Karna P, et al. Reduction of bronchopulmonary dysplasia after participation in the Breathsavers Group of the Vermont Oxford Network Neonatal Intensive Care Quality Improvement Collaborative. Pediatrics 2006;118(Suppl. 2):S737. 103. Laughon MM, Langer J, Wilson-Costello D, et al. Predictive models for bronchopulmonary dysplasia by postnatal day in premature infants. ePAS; 2009:3450.7.

92. Friis-Hansen B. Body water compartments in children: changes during growth and related changes in body composition. Pediatrics 1961;28:16981. 93. Oh W, Poindexter BB, Perritt R, et al. Association between uid intake and weight loss during the rst ten days of life and risk of bronchopulmonary dysplasia in extremely low birth weight infants. J Pediatr 2005;147:78690. 94. Bell EF, Acarregui MJ. Restricted versus liberal water intake for preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2008. CD000503. 95. Hartnoll G, Betremieux P, Modi N. Randomised controlled trial of postnatal sodium supplementation on oxygen dependency and body weight in 2530 week gestational age infants. Arch Dis Child Fetal Neonatal Ed 2000;82:F1923. 96. Brion LP, Primhak RA, Ambrosio-Perez I. Diuretics acting on the distal renal tubule for preterm infants with (or developing) chronic lung disease. Cochrane Database Syst Rev; 2002. CD001817. 97. Brion LP, Primhak RA. Intravenous or enteral loop diuretics for preterm infants with (or developing) chronic lung disease. Cochrane Database Syst Rev; 2002. CD001453. 98. Segar JL, Chemtob S, Bell EF. Changes in body water compartments with diuretic therapy in infants with chronic lung disease. Early Hum Dev 1997;48:99107.

Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Drug therapies in bronchopulmonary dysplasia: debunking the myths


Win Tin a, *, Thomas E. Wiswell b
a b

James Cook University Hospital, Marton Road, Middlesbrough TS4 3BW, UK Center for Neonatal Care, Florida Hospital Orlando, 2718, North Orange Avenue, Suite B, Orlando, FL 32804, USA

s u m m a r y
Keywords: Bronchopulmonary dysplasia Chronic lung disease Drug therapies Preterm Respiratory distress syndrome

Bronchopulmonary dysplasia (BPD), also known as chronic lung disease (CLD), is one of the most challenging complications in premature infants. The incidence of BPD has been increasing over the past two decades in parallel with an improvement in the survival of this population. Furthermore, the clinical characteristics and the natural history of infants affected by BPD have changed considerably, and newer denitions to clarify the term BPD have also evolved since its rst description more than four decades ago. Several drug therapies have also evolved, either to manage these infants respiratory distress syndrome with an aim to prevent BPD or to manage the established condition. Although there is good evidence to support the routine use of some therapies, many other therapies currently used in relation to BPD remain individual- or institution-specic, depending on beliefs and myths that we have adopted. In this article, we discuss the importance of dening BPD more objectively and the support or lack thereof for the drug therapies used in relation to BPD. 2009 Elsevier Ltd. All rights reserved.

1. Introduction The term bronchopulmonary dysplasia (BPD) was rst described in 1967 by Northway et al.,1 and despite the ongoing researches to improve the neonatal respiratory care, BPD continues to impose a considerable risk for mortality and long-term morbidity. Several devices and strategies have been developed to provide optimum respiratory support to the newborn infants, and several drug therapies have also been used in an attempt to prevent BPD or to mitigate the course of the established condition. Although there is considerable evidence to support the routine use of some therapies, most therapies remain individual- or institutionspecic, based on anecdotes, evidence of short-term (but not clinically meaningful) benets, beliefs, and myths. There is also a considerable variation in clinical practices to dene and establish the diagnosis of BPD,2 upon which the infants receive therapies. In this context, we have tried to provide the available evidence for drug therapies and hope that the clinicians will nd this useful in appraising their own current practices.

2.1. Prenatal corticosteroid therapy The single course of glucocorticoids (dexamethasone or betamethasone) given to a mother who is in preterm labour to accelerate the maturation of surfactant system in the fetal lung is known to be safe, and reduces the mortality [relative risk (RR): 0.6; 95% condence interval (CI): 0.480.75] and the risk of respiratory distress syndrome (RDS) (0.64; 0.560.72), but there is no evidence that this intervention reduces the risk of BPD (1.38; 0.902.11), probably because of an increase in survival.3 The use of multiple courses of prenatal corticosteroids signicantly increased the risk of BPD (3.01; 1.545.88), but the systematic review comparing the use of multiple courses versus a single course did not show any differences in the risk of BPD (1.01; 0.631.65).3 The available evidence strongly supports the use of a single course of prenatal corticosteroids given to women at risk of preterm labour as good practice, but no more than one course should be used routinely outside of a clinical trial.4 2.2. Surfactant replacement therapy

2. Prevention of BPD As with all the clinical conditions, the ideal management strategy would be to prevent BPD; once established, the treatment for BPD remains mostly supportive.
* Corresponding author. Tel.: 44 1642 854834; fax: 44 1642 854874. E-mail address: win.tin@stees.nhs.uk (W. Tin). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.003

The routine use of surfactant therapy in preterm infants was introduced into neonatal clinical practice about 20 years ago, and the ongoing developments of synthetic as well as natural surfactants, along with their rigorous controlled trial evaluations have been one of the major achievements in the eld of neonatology. More than 20 000 newborn infants have been randomised into controlled trials. Systematic review of different approaches of

384

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

surfactant therapy510 showed that it signicantly reduces mortality, but does not reduce the risk of BPD in survivors (Table 1). This may also result from the increased survival of preterm infants. Although a meta-analysis of 10 controlled trials showed that the use of natural compared with synthetic surfactant reduces the risk of mortality,5 two recently published controlled trials suggested that the newer generation of protein-containing synthetic surfactant, lucinactant (not yet available commercially), may have some advantages over synthetic surfactants that contain only phospholipid, and that it is as safe and efcacious as natural surfactants.1012 The immediate respiratory management of preterm infants, whether to intubate them and give early surfactant therapy or to manage them with a non-invasive form of respiratory support (hence not using early surfactant), is currently the subject of controversy and debate, but all the available evidence supports the use of prophylactic surfactant therapy for all intubated preterm infants.

2.3. Caffeine therapy Systematic review of ve trials with a total of 192 infants concluded that methylxanthines reduced the frequency of apnoea and the use of mechanical ventilation in the rst 27 days of therapy. The same review also cautioned about the lack of knowledge of its long-term effects.13 Caffeine (usually caffeine citrate) is now the methylxanthine therapy of choice, since routine drug monitoring is not necessary because of its wider therapeutic margin.14 The use of caffeine for apnoea of prematurity was rst reported in 1977,15 and despite the widespread use of this therapy over the past three decades, there were no reliable data on its short-term safety, or on any long-term effects including safety until recently.16 Some small observational studies with follow-up information did not show any long-term adverse effects of methylxanthine therapy,1719 but one large follow-up study of more than 400 very low birth weight infants reported that theophylline administration was signicantly associated with an increased risk of cerebral palsy (CP).20 Furthermore, reports from some observational studies suggested that caffeine reduces cerebral and intestinal blood ow velocity in preterm infants,21 resulting in the perception that it may increase the risk of necrotising enterocolitis (NEC).22

The multicentre, placebo-controlled, randomised trial of caffeine, also known as the Caffeine for Apnoea of Prematurity (CAP) Trial was conducted to determine whether survival at 18 months (corrected age) without disability is improved if apnoea of prematurity is managed without methylxanthines in infants with birth weights of 5001250 g. The trial design was pragmatic, and infants were enrolled in the rst 10 days of life if they were considered suitable candidates for methylxanthine therapy. They received either caffeine citrate (20 mg/kg of initial loading dose, followed by 510 mg/kg/day) or the equivalent volume of placebo. The trial enrolled 2006 infants between 1999 and 2004, and collaborating clinicians managed to keep the use of open-labelled methylxanthines to <10%. Caffeine therapy signicantly reduced the risk of BPD, dened as a need for supplemental oxygen at 36 weeks postmenstrual age (PMA), 36% in caffeine group vs 47% in placebo group [odds ratio (OR): 0.63; 95% CI: 0.520.76; P < 0.001]. Supplemental oxygen therapy, continuous positive airway pressure (CPAP) and mechanical ventilation were discontinued 1 week earlier with the use of caffeine compared to placebo. Unexpectedly, caffeine therapy was also found to be associated with a signicant reduction in the risk of persistent patency of the ductus arteriosus (PDA) requiring pharmacological or surgical closure. There was a temporary reduction in weight gain during the rst three weeks of caffeine therapy, but there was no adverse effect on growth at the time of hospital discharge. This large study also provides the reassurance that caffeine therapy does not increase the risk of death, nor of neonatal complications including NEC or ultrasonographic evidence of brain injuries.23 Adequate data for an analysis of the primary outcome measure were available for 93% of the subjects, and the high quality follow-up data from the CAP trial also showed that caffeine therapy improves the rate of survival without neurodevelopmental disability at corrected age of 1821 months (death or disability 40.2% in the caffeine group vs 46.2% in the placebo group; OR: 0.77; 95% CI: 0.640.93; P 0.008). Treatment with caffeine also significantly reduced the incidence of CP (4.4% vs 7.3%; P 0.009) and of cognitive delay (33.8% vs 38.3%; P 0.04).24 The post-hoc analysis to explore the likely mechanism as to why caffeine therapy resulted in better neurodevelopmental outcome suggested that earlier discontinuation of positive airway pressure in infants who received caffeine, compared with placebo, was the most important

Table 1 Summary of different strategies of surfactant replacement therapy. Strategies of surfactant therapy Prophylactic synthetic surfactant vs control Reference 5 Outcomes Mortality Death or BPD at 28 days BPD at 28 days Mortality Death or BPD at 28 days BPD at 28 days Mortality Death or BPD at 28 days BPD at 28 days Mortality Death or BPD at 28 days BPD at 28 days Death or BPD at 36 weeks PMA BPD at 36 weeks PMA Mortality Death or BPD at 28 days BPD at 28 days Death or BPD at 36 weeks PMA BPD at 36 weeks PMA No. of studies 7 4 4 7 7 7 7 8 8 10 5 8 4 5 2 2 2 2 2 No. of patients 1500 1018 1086 932 932 932 2613 2816 2816 4588 3044 3515 2565 3179 1028 1028 1028 1028 1028 RR (95% CI) 0.70 (0.580.85) 0.89 (0.770.03) 1.06 (0.831.36) 0.60 (0.440.83) 0.84 (0.750.93) 0.93 (0.801.07) 0.61 (0.480.77) 0.85 (0.760.95) 0.96 (0.821.12) 0.86 0.95 1.02 0.98 1.01 0.79 0.99 1.0 0.96 0.99 (0.760.98) (0.901.01) (0.931.11) (0.901.06) (0.901.12) (0.611.02) (0.881.11) (0.891.12) (0.821.12) (0.841.18)

Prophylactic natural surfactant vs control

Prophylactic vs selective use of surfactant

Natural vs synthetic surfactant

Protein containing synthetic surfactant vs natural surfactant

10

RR, relative risk; CI, condence interval; BPD, bronchopulmonary dysplasia; PMA, postmenstrual age.

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

385

intermediate variable. This mechanism was supported by further analyses of subgroups: in terms of death or major disability, infants receiving CPAP or mechanical ventilation by endotracheal tube appeared to have gained more benet from caffeine therapy, compared with those infants who were not requiring any positive pressure ventilation.25 The CAP trial is one of the best examples of using the gold standard, the randomised controlled trial, to examine the benet and long-term safety of a commonly used therapy and to debunk longstanding myths and beliefs. 2.4. Oxygen therapy Supplemental oxygen is probably the most commonly used drug in preterm infants with RDS and BPD along with other therapeutic adjuncts. The ultimate goal of oxygen therapy is to achieve adequate tissue oxygenation, but without creating oxygen toxicity and oxidative stress. Direct exposure to high concentrations of oxygen can damage the pulmonary epithelium, and this has been recognised as an important cause of BPD since its rst description by Northway et al.1 There is increasing evidence that oxidative stress is implicated in the development of BPD, reviewed extensively by Saugstad.26 Most observational studies published since 2001 have suggested that in comparison with the liberal approach of accepting high oxygen saturation values, the restrictive approach of using low oxygen saturation values was associated with lower incidences of BPD and retinopathy of prematurity (ROP), without an increased risk of mortality.2730 In contrast, an observational study by Poets et al., involving 891 babies of <30 weeks gestation and admitted to two neonatal units, using different oxygen saturation limits (8092% vs 9297%) found that the incidence of ROP was signicantly higher in the unit that used a lower alarm limit but allowed wider variation in saturation (13% vs 6%).31 There is no sufcient evidence to date to suggest the optimal level in preterm infants who receive oxygen therapy in the early neonatal period.32,33 An international collaborative effort has been mounted since 2003 to conduct large, multicentre, masked, randomised trials to answer the question, What oxygen saturation level should be targeted in very premature infants? Five trials [Surfactant Positive Airway Pressure and Pulse Oximetry Trial (SUPPORT), Benets of Oxygen Saturation Targeting (BOOST)-2 Australia, BOOST-2 New Zealand, Canadian Oxygen Trial (COT) and BOOST-2 UK] have been launched. All these trials address the principal research question, Does varying the concentration of inspired oxygen to maintain a low oxygen saturation range of 8589% versus a high range of 9195% in babies <28 weeks gestation from the day of birth until 36 weeks PMA affect the incidence of (1) death or severe neurosensory disability at a corrected age of 18 months to 2 years, (2) retinal surgery for ROP, (3) BPD (the need for supplemental oxygen therapy or respiratory support at 36 weeks postmenstrual age), (4) PDA and NEC, and (5) poor growth at 36 weeks PMA and at 2 years? There has already been a prospective agreement to combine the individual patient data from all the trials in order to increase the ability to detect much smaller differences in the primary outcome, and this controlled trial strategy of prospective meta-analysis is likely to be established for the rst time in neonatal medicine. 2.5. Inhaled nitric oxide (iNO) therapy Nitric oxide is a potent vasodilator. Inhaled nitric oxide can produce selective pulmonary vasodilation without lowering systemic blood pressure. The gas has a very short half-life (24 s) in the body,14 and this forms the physiological rationale for its use in

newborn infants with pulmonary hypertension. Animal studies also suggest that iNO reduces lung inammation,34 improves surfactant function,35 and promotes lung growth36; hence, iNO may offer therapeutic benets in reducing the risk of developing BPD in preterm infants. Inhaled NO therapy improved oxygenation and reduced the risk of death or the requirement for treatment with extracorporeal membrane oxygenation (ECMO) in term or late preterm infants (34 weeks gestation) with respiratory failure.37 However, the role of iNO in preterm infants with hypoxaemic respiratory failure, both in terms of its efcacy and safety, is less clear and remains controversial.38 Early trials of iNO in preterm infants showed no benet in reducing the risk of death or BPD.3941 A double-blind RCT by Kinsella et al. (1999), looking at the efcacy of low dose iNO (5 ppm) on 80 preterm infants with severe hypoxaemic respiratory failure, showed no differences in the primary outcome measure of survival to discharge, although there was an observed short-term improvement in oxygenation among treated infants.39 Another multicentre RCT carried out in the UK and the Republic of Ireland to evaluate the efcacy of iNO in preterm infants (<34 weeks gestation) with severe respiratory failure, showed that there was no evidence of an effect of iNO on the primary outcome of death or severe disability at 1 year corrected age, death or supplemental oxygen need on the expected date of delivery, or death or supplemental oxygen at 36 weeks PMA. Furthermore, this study also showed that mean total costs at 1 year corrected age were signicantly higher in a group of infants allocated to receive iNO, partly because of the very high cost of the gas, but mainly because of the difference in initial hospitalisation costs.42 The main drawback of this trial was its recruitment of only 108 subjects despite the planned sample size of 200 infants. A large multicentre study by Van Meurs et al. also showed that iNO has no impact on the risk of death or BPD in preterm infants with birth weight <1500 g, and with severe respiratory failure. In fact, this study suggested an increased risk of death with iNO therapy among infants with a birth weight <1000 g.43 However, a single-centre study by Schreiber et al. demonstrated that iNO reduced the risk of death and BPD in preterm infants with mild or moderate RDS from 67% to 49%. Treated infants also had a reduced risk of sonographically apparent brain injuries, and, more relevant, had a better neurodevelopmental outcome compared with the control group.44 Nonetheless, it is important to note that the rates of BPD and brain injuries were higher than expected in the control group of infants who received oxygen as a placebo. Two large RCTs reported in 2006 have hinted at the potential benets of iNO in preterm infants with a birth weight between 500 and 1250 g. Mean gestational age of infants in both studies was the same, but the age at which iNO was initiated, the dose of iNO, and the duration of intervention were considerably different. In the trial by Kinsella et al. (2006), infants who were receiving mechanical ventilation and who were <48 h old were randomly assigned to receive either 5 ppm of iNO or nitrogen as a placebo. This trial did not show a reduction in the overall incidence of BPD by early iNO therapy, with the exception of the pre-stratied group of infants with a birth weight between 1000 and 1250 g (16% of the study group).45 In contrast, the trial by Ballard et al. enrolled infants who were ventilator dependent (and also infants with a birth weight between 500 and 799 g, who were receiving CPAP) at 721 days. Treated infants received iNO (20 ppm) for 4896 h, with the dose reduced at weekly intervals. This large trial showed that iNO improved the survival without CLD at 36 weeks PMA, but this was limited to infants who were 714 days of age at randomisation. The complications of prematurity, including ultrasonographic evidence of brain injuries, were not different between the two groups.46

386

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

Although these two studies suggest that iNO may be of benet in preterm infants who are less critically ill, and the ndings on brain injuries are reassuring, preliminary data from the most recently completed large European Collaborative Trial show that early use of low dose iNO (5 ppm) in preterm infants (240 to 286 weeks gestation) with mild to moderate respiratory failure (needing FIO2 of 0.30.5) has no impact on survival free of BPD.47 Long-term follow-up information from this European Trial will be available in 2011, but all the evidence available to date fails to support the routine use of iNO therapy in preterm infants. 2.6. Early postnatal corticosteroid therapy 2.6.1. Early systemic corticosteroid therapy Systemic corticosteroids have been used to prevent or treat BPD in preterm infants. These therapies have been given early (<96 h of age)48 and moderately early (714 days of age).49 with the aim to minimise the risk of developing BPD, or delayed (>3 weeks of age)50 as treatment for established BPD (see below). The short-term efcacy of dexamethasone in preventing BPD was reported about two decades ago by Cummings et al.,51 and the early or moderately early use of a high dose,42 day-tapering course of dexamethasone became a common practice during the 1990s. Both early and moderately early systemic corticosteroids signicantly reduced the incidence of BPD at either 28 days of age or 36 weeks PMA. Moderately early treatment was associated with the reduction in mortality through 28 days of age, but not thereafter, and there was no reduction in mortality following early treatment. Unfortunately, more adverse effects including hypertension, hyperglycaemia, gastrointestinal bleeding, hypertrophic cardiomyopathy, and infection were observed following both early and moderately early treatments.48,49 More importantly, in the trials that have reported the long-term follow-up information, signicantly higher adverse neurodevelopmental outcomes (developmental delay, CP, abnormal neurological examinations) were seen following early systemic corticosteroid therapy.48,5254 There was no evidence of an increase in adverse neurodevelopmental outcomes following moderately early treatment with systemic cortcosteroids, but the follow-up data were limited.49 In view of the concerns about the long-term adverse effect on neurodevelopment, and short term side-effects, the Committee on the Fetus and Newborn of the American Academy of Pediatrics and Canadian Pediatric Society, as well as the European Association of Perinatal Medicine, recommended against the routine use of corticosteroids to prevent or treat BPD.55,56 More recently, a published double-blind RCT that recruited 70 ventilator-dependent infants of <28 weeks gestation or <1000 g to receive either low dose dexamethasone or placebo after the rst week of life, showed that this treatment shortened the duration of intubation, without any short-term complications,57 and that there was no obvious long-term with respect to neurosensory outcome, blood pressure, or hospital readmissions.58 Although the study was originally designed to detect a 10% difference in the rate of survival without major neurosensory disability with an adequate power, it was stopped with <10% of the target sample size because of very poor recruitment. The effect of low dose dexamethasone in the long term, therefore, cannot be conclusive. 2.6.2. Early inhaled corticosteroid therapy Early use of inhaled glucocorticoids (budesonide, beclomethasone, uticasone) is an attractive therapeutic option because of its potential of gaining benets in preventing BPD with much lower systemic side-effects. Although some observational studies suggested its benecial effect on chemotactic activity and

inammatory mediators,59,60 and a small pilot trial suggested a trend towards a reduction in the duration of mechanical ventilation and BPD,61 meta-analysis of seven trials of early inhaled coticosteroids showed no differences in mortality, BPD, or the combined outcome of death and BPD.62 Systematic review of inhaled versus systemic corticosteroids in preventing BPD showed that the duration of mechanical ventilation and the need for supplemental oxygen therapy were signicantly longer in the inhaled steroid group compared with the systemic group, but there was no difference in the incidence of BPD.63 2.7. Antioxidant therapies Preterm infants are more likely to be exposed to oxidative stress and are decient in endogenous antioxidant systems26; therefore, a number of antioxidant therapies have been used to prevent BPD. 2.7.1. Vitamin A therapy Vitamin A is involved in the regulation and promotion of growth and differentiation of multiple cells. It also maintains the integrity of the epithelial cells of the respiratory tract.14 Preterm infants are relatively decient in vitamin A, and this has been shown to be associated with BPD. A large RCT of 807 infants with a birth weight <1000 g demonstrated that a large dose of intramuscular vitamin A, given three times a week for 4 weeks from birth, reduced the risk of CLD (RR: 0.89, 95% CI: 0.80.99).64 A trial of oral vitamin A therapy daily for 4 weeks in a similar population of infants failed to detect any benet.65 Meta-analysis of eight trials suggests that vitamin A supplementation reduces the risk of BPD at 36 weeks PMA (0.87; 0.770.98).66 Neurodevelopmental assessment of 85% of surviving infants in the largest trial of vitamin A showed no differences in outcome between supplementation and placebo groups at 1822 months corrected age.64 2.7.2. Vitamin E therapy Vitamin E consists of several types of tocopherols with antioxidant activity. As a scavenger of free radicals, it could potentially limit the processes that lead to BPD. Vitamin E has been extensively studied in preterm infants with the expectation that it could help prevent oxidative stress-related damage to multiple organs, but a systematic review stated that this therapy did not reduce the incidence of BPD.67 2.7.3. Superoxide dismutase therapy There has been hope that the antioxidant superoxide dismutase (SOD) could ameliorate oxygen free radical injury to the premature lungs. However, two small RCTs carried out to assess the effect of SOD in preventing BPD failed to detect any benet of this therapy,68 and therefore the use of SOD in preterm infants cannot be recommended. 2.8. Bronchodilators 2.8.1. Beta-receptor agonist therapy Bronchodilators, such as beta-receptor agonists, are potentially attractive therapies in preventing BPD as bronchial hyperreactivity and responsiveness may be more common in this population, and also because assisted ventilation may aggravate the response. One double-blind RCT using inhaled salbutamol to prevent BPD found no evidence that this therapy reduced mortality, oxygen dependency at 28 days, duration of ventilation, or duration of oxygen supplementation in preterm infants.69 Several studies that assessed the immediate changes in pulmonary mechanics (compliance, resistance) following the use of salbutamol have shown variable results: improvement, no change, or even worsening in these

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

387

respiratory parameters.70,71 As there is no RCT to date with meaningful clinical outcomes following beta-receptor agonist therapy, its routine use in prevention or treatment of BPD cannot be justied. 2.8.2. Anticholinergic therapy Ipratropium bromide, a muscarinic antagonist, has some effect on bronchodilation and has been used to treat preterm infants with a high risk of BPD. The data available are limited to case reports following short-term use of this therapy,7173 and the results are mixed. Some demonstrate positive and some indicate either no or negative effects. Without the supportive evidence from any RCT, this therapy cannot be recommended for the management of BPD. 2.9. Diuretic therapy Lung oedema and PDA may complicate RDS in preterm infants. These, along with renal insufciency, are the main reasons that furosemide is used in the early neonatal period. Prolonged use increases renal loss of sodium and potassium and also urinary calcium excretion and the risk of renal calcium deposition. Early use of furosemide is associated with an increased incidence of PDA in preterm infants, as it stimulates the production of prostaglandin E2 from the kidneys. Repeated use of this therapy in preterm infants increases the risk of ototoxicity, and it can also enhance the risk of aminoglycoside ototoxicity. Bumetanide is another loop diuretic with a similar mechanism of action, but more potent and probably less ototoxic than furosemide. However, it can also cause signicant urinary losses of sodium, chloride, and calcium, and its neonatal use has not been fully evaluated.14 A systematic review of the use of furosemide for RDS in preterm infants showed that early use of furosemide had no effect on meaningful outcomes, including duration of mechanical ventilation and oxygen supplementation, length of hospitalisation, and more relevantly on mortality and chronic lung disease.74 All six trials included in this review were carried out before the era of prenatal corticosteroids and surfactant replacement therapy, and there is no current evidence to support the routine use of furosemide or any other diuretics in preterm infants with RDS. 2.10. Other therapies for preventing BPD 2.10.1. Indomethacin and ibuprofen PDA is one of the risk factors for BPD in preterm infants. The cyclo-oxygenase inhibitors indomethacin and ibuprofen are the two most commonly used therapies in preventing or treating PDA. It is reasonable to assume that a treatment regimen that prevents PDA should reduce the risk of BPD in this population. The largest RCT to assess the outcomes following the early prophylactic use of indomethacin in very low birth weight infants, the International Trial of Indomethacin Prophylaxis in Preterms (TIPP),75 showed a signicant reduction in the incidence of PDA. However, there was no reduction in the incidence of BPD, a nding similar to all other RCTs that have assessed this therapy.76 Ibuprofen has been shown to be equally effective at PDA closure and is gaining in popularity over the past decade as it does not cause reduction in regional blood ow to the brain, kidneys, or gut.14 It has also been assessed in RCTs with results similar to those seen with indomethacin; neither reduction nor even a trend towards reduction in BPD was observed.77 2.10.2. Inositol This is an essential nutrient that may play an important role in preventing BPD, as it promotes maturation of the surfactant system. It can be given intravenously, orally, and also in a form of

supplementation in enteral feeds. Meta-analysis of the two RCTs that reported outcomes of death and BPD showed signicant reduction in death (RR: 0.48, 95% CL: 0.280.80), as well as death or BPD (0.56; 0.420.77). Furthermore, inositol therapy also reduced the risk of any stage of ROP (0.53; 0.290.97), without increasing the risk of neonatal outcomes such as NEC and sepsis.78 These conclusions are based on a limited number of infants, but form the basis to justify a much larger RCT with long-term outcomes of this low cost, simple nutritional intervention. 2.10.3. Antibiotic therapy The effects of colonisation or infection with Ureaplasma urealyticum on the risk and severity of BPD remain controversial. However, available evidence from two small controlled studies does not support the use of routine testing for this infection in preterm infants or the prophylactic treatment with erythromycin in this high risk population.79 3. Treatment of established BPD 3.1. Dening BPD Since Bancalari et al. proposed a denition of BPD 30 years ago, based on clinical characteristics (the need for oxygen on 28 of the rst 28 days with chest radiographic changes), different denitions of BPD have evolved.80 The National Institutes of Health Consensus in 2001 helps the clinician to dene BPD with more information about its severity.81 However, the denition of BPD and the decision to initiate therapies is largely based on whether the infant needs oxygen therapy, and clinicians have widely divergent practices regarding oxygen saturation targets. Dening established BPD is still subjective and opinion-based. It is important for assessment of oxygen dependency to be as objective and as uniform as possible by using a physiological assessment,82 and for clinicians to carefully scrutinise their diagnosis of BPD in preterm infants before subjecting them to any treatment. 3.2. Oxygen therapy The aim of oxygen therapy in preterm infants with established BPD has traditionally been to maintain pulse oximetry in the range that is normal for a term baby (9497%), believing that this is the optimal range to promote growth and development, to reduce the risk of worsening pulmonary hypertension, and to ameliorate sleep pattern abnormalities. Two RCTs have been conducted to see whether it is better to aim for high oxygen saturation in preterm infants who are more than a few weeks old. The American STOP-ROP trial83 which recruited 649 babies with a mean gestation of 25.4 weeks showed that keeping functional oxygen saturation >95% slightly reduced the number of infants with pre-threshold retinopathy who went on to develop disease severe enough to require retinal surgery. However, benet was only seen in those without evidence of plus disease (dilated and tortuous vessels in at least two quadrants of the posterior pole) at recruitment. More unexpectedly, targeting higher oxygen saturations signicantly increased the number of infants who remained in the hospital, in supplemental oxygen, and who were receiving diuretic therapy at 50 weeks PMA. Signicant pulmonary deterioration after recruitment (13.2% vs 8.5%) was seen in those with worse than average BPD at trial entry. The result of the Australian BOOST trial was published in 2004.84 The aim of this double-blind study was to see whether maintaining higher oxygen saturations versus standard levels among oxygen-dependent preterm infants improved their growth and development. This study recruited 358 babies at <30 weeks gestation who remained in

388

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

supplemental oxygen at 32 weeks PMA. Trial oximeters were specially modied to keep targeted functional saturation in the range of either 9194% or 9598% while displaying a value (masked) in the range 9396%. This well-designed study showed no evidence that the growth and developmental outcome of oxygen-dependent preterm infants was improved by keeping their oxygen saturation in the high range. In keeping with the observation in the STOP-ROP trial, the BOOST trial also showed that infants in the higher oxygen saturation range had greater use of postnatal steroids (58% vs 50%) and diuretics (52% vs 44%), more readmissions (54% vs 48%), and more pulmonary-related deaths (6% vs 1%). Although these pulmonary-related outcomes in both RCTs were secondary outcomes, thus lacking in reliable statistical power, the consistency of the direction of effects lends weight to the hypothesis that targeting high oxygen saturations >95% in infants with BPD may do more harm than good. 3.3. Diuretic therapy The distal tubule diuretic thiazide, the potassium-sparing diuretic spironolactone, and the loop diuretic furosemide are commonly used in preterm infants who have established BPD. These diuretics are administered parenterally, enterally, or by inhaled aerosol to reduce alveolar and interstitial lung oedema and to improve pulmonary mechanics. A thiazide diuretic, chlorthiazide, often used in combination with spironolactone for the control of pulmonary oedema in preterm infants with severe BPD, is felt to be the safest diuretic combination for long-term use, but it can cause excessive urinary calcium loss and subsequent bone demineralisation and nephrocalcinosis in preterm infants.14 Despite their widespread use, there are surprisingly few data assessing the meaningful value of diuretic therapy in BPD. All the systematic reviews to assess the risks and benets of furosemide, including the aerosolised form, as well as thiazides (with or without spironolactone) in preterm infants with BPD showed improvements in oxygenation and lung compliance, but these short-term physiological benets are not translated into any meaningful clinical benets (mortality, duration of mechanical ventilation and oxygen dependency, and length of hospital stay).8587 Of all the adjunctive therapies used in preterm infants with BPD, diuretic therapy is one of the most abused without evidence of substantive benet. 3.4. Late postnatal corticosteroid therapy 3.4.1. Delayed/late systemic corticosteroids Delayed systemic corticosteroids given after 3 weeks of postnatal age had no effect on mortality and had a borderline effect on reducing the incidence of BPD at 36 weeks PMA. There was no increase in the occurrence of infection, hyperglycaemia, or gastrointestinal complications, but hypertension was signicantly more common with delayed corticosteroid therapy. Although there was an increased rate of abnormal neurological examination among those receiving delayed corticosteroid therapy, there were no increases in the rate of CP and major neurosensory disabilities.50 Information on this long-term outcome, however, is limited. A more recent review on late postnatal systemic corticosteroids, given any time after 7 days, analysed 19 eligible RCTs with a total of 1354 participants. This treatment was associated with a reduction in neonatal mortality at 28 days but not at discharge or at the latest reported age. The meta-analysis showed benecial effects of this late treatment such as: reduction in failure to extubate by 3, 7 and 28 days; BPD, as well as death or BPD at both 28 days and 36 weeks PMA; and reduction in the need for home oxygen therapy. There was a trend toward an increased risk of infection and gastrointestinal bleeding, but not NEC. There was a trend toward increased risk of CP

or abnormal neurological examination in the steroid-treated group, but this was partly offset by the trend toward a reduction in deaths prior to follow-up. Thus, the combined rate of death or CP was not different between steroid and control groups. No differences were found in the risk of major neurosensory disability or the combined risk of death or major neurosensory disability. Reviewers concluded that postnatal corticosteroid therapy for BPD, initiated after 7 days of age, may reduce mortality without signicantly increasing the risk of adverse neurodevelopmental outcomes. However, long-term outcome information is of limited quality, and no study was powered to detect long-term adverse neurosensory outcomes. Thus, postnatal corticosteroids should be reserved for infants who cannot be weaned from mechanical ventilation and the dose and duration should be kept to a minimum.88 The long-term risk:benet ratio of systemic postnatal corticosteroids may depend upon the risk of an infant developing BPD when steroid therapy is initiated. A meta-regression analysis of 28 RCTs, with available follow-up data from 1721 randomised infants, showed that the impact of postnatal corticosteroids on the combined outcome of death or CP was modied by the risk of BPD in the control group of infants; postnatal steroid therapy signicantly increased the chance of death or CP if used in infants with a risk of BPD <35%, whereas the same therapy reduced the chance of death or CP if used in infants with risk of BPD >65%.89 Based on the available evidence that adverse effects of postnatal corticosteroids may be related to a high dose and its early use in infants with a relatively low risk of BPD, clinicians should seriously consider re-examining the long-term efcacy and safety of a low dose, short course in infants at a high risk for BPD. 3.4.2. Late inhaled corticosteroids Although inhaled steroids have been used to treat infants with BPD who remain ventilator or oxygen dependent, meta-analysis of two trials of late treatment showed that compared with systemic therapy, inhaled steroids did not offer any advantage in either reducing the risk of BPD, or the short-term side-effects of steroid therapy.90 There are no data yet regarding the long-term safety of this treatment, and it is not justiable to use inhaled steroids to treat established BPD. 3.5. Sildenal therapy Sildenal is slowly gaining attraction as a potentially useful therapy in infants who are difcult to oxygenate because of severe BPD and pulmonary hypertension. Although animal studies have shown that sildenal improves alveolar growth and reduces pulmonary hypertension,91 there are no available clinical data of this treatment on newborns; and its use (if any) should be restricted to infants with a major difculty in oxygenation and with clear echocardiographic features of pulmonary hypertension. 3.6. Nutritional intervention Preterm infants with BPD are likely to have nutritional decits that may have an impact on mortality as well as short- and longterm morbidity, and there is a perception that increasing the daily energy intake for these infants may be of benet in terms of growth, respiratory, and neurodevelopmental outcome. However, there has been no controlled trial to provide any evidence on this important issue.92 4. Conclusion Since the rst description of BPD by Northway et al. in 1967, there have been great improvements in our knowledge and

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390

389

understanding of its pathogenesis and its risk factors. In parallel, newer strategies of ventilation and drug therapies have developed. Population characteristics of infants at high risk of BPD have also changed considerably over this 40-year period. However, BPD remains a major challenge to the clinicians, and continues to impose major risks of mortality and long-term morbidity among preterm infants. The valuable evidence for some of the drug therapies for BPD, including prenatal corticosteroids, surfactant and caffeine have evolved as a result of using the RCT gold standard tool in accumulating evidence, but several therapies used in preventing or treating BPD are still based on individual or institutional beliefs and myths that remain unsubstantiated. Only our collaborative effort to mount well-designed studies to address the long-term efcacy and safety will enable us to debunk the myths and to practise evidence-based medicine.

References
1. Northway WH, Rosen RC, Porter DY. Pulmonary disease following respiratory therapy of hyaline membrane disease: bronchopulmonary dysplasia. N Engl J Med 1967;276:35768. 2. Bancalari E, Claure N, Sosenko I. Bronchopulmonary dysplasia: changes in pathogenesis, epidemiology and denition. Semin Neonatol 2003;8:6371. 3. Crowley P. Prophylactic corticosteroids for preterm birth. Cochrane Database Syst Rev; 2000. CD000065. 4. Crowther CA, Harding J. Repeat doses of prenatal corticosteroids for women at risk of preterm birth for preventing neonatal respiratory disease. Cochrane Database Syst Rev; 2003. CD003935. 5. Soll RF. Prophylactic synthetic surfactant for preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2000. CD001079. 6. Soll RF. Prophylactic natural surfactant extract for preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2000. CD000511. 7. Soll RF, Morley CJ. Prophylactic versus selective use of surfactant in preventing morbidity and mortality in preterm infants. Cochrane Database Syst Rev; 2001. CD000510. 8. Soll RF, Blanco F. Natural surfactant extract versus synthetic surfactant for neonatal respiratory distress syndrome. Cochrane Database Syst Rev; 2001. CD000144. 9. Soll RF. Synthetic surfactant for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2000. CD001149. 10. Pster RH, Soll RF, Wiswell T. Protein containing synthetic surfactant versus animal derived surfactant extract for the prevention and treatment of respiratory distress syndrome. Cochrane Database Syst Rev; 2007. CD006069. 11. Sinha SK, Lacaze-Masmontiel T, Valls I, et al. A multicentre, randomised controlled trial of lucinactant versus poractant alpha among very premature infants at high risk of respiratory distress syndrome. Pediatrics 2005;115: 10308. 12. Moya F, Gadzinowski J, Bancalari E, et al. A multicentre, randomised, masked, comparison trial of lucinactant, colfosceril palmitate and beractant for the prevention of respiratory distress syndrome. Pediatrics 2005;115:101829. 13. Henderson-Smart DJ, Steer P. Methylxanthine treatment for apnoea in preterm infants. Cochrane Database Syst Rev; 2001. CD000140. 14. Hey E, editor. Neonatal formulary. 5th ed. London: BMJ Books; 2007. 15. Aranda JV, Gorman W, Bergsteinsson H, Gunn T. Efcacy of caffeine in treatment of apnea in the low-birth-weight infant. J Pediatr 1977;90:46772. 16. Schmidt B. Methylxanthine therapy in premature infants: sound practice, disaster or fruitless by way? J Pediatr 1999;135:5268. 17. Gunn TR, Metrakos K, Riley P, et al. Sequelae of caffeine treatment in preterm infants with apnea. J Pediatr 1979;94:1069. 18. Nelson RM, Resnick MB. Long term outcome of premature infants treated with theophylline. Semin Perinatol 1981;5:3703. 19. Ment LR, Scott DT, Ehrenkranz RA, et al. Early childhood developmental follow up of infants with GMH/IVH: effect of methylxanthine therapy. Am J Perinatol 1985;2:2237. 20. Kitchen WH, Doyle LW, Ford GW, et al. Cerebral palsy in very low birth weight infants surviving to 2 years with modern neonatal intensive care. Am J Perinatol 1987;4:2935. 21. Hoecker C, Nelle M, Poeschl J, et al. Caffeine impairs cerebral and intestinal blood ow velocity in preterm infants. Pediatrics 2002;109:7847. 22. Anonymous. New drug application: CAFCIT (NDA) 02073. Washington, DC: Food and Drug Administration; 2000. 23. Schmidt B, Roberts RS, Davis P, et al. Caffeine therapy for apnea of prematurity. N Engl J Med 2006;354:211221. 24. Schmidt B, Roberts RS, Davis P, et al. Long term effects of caffeine therapy for apnea of prematurity. N Engl J Med 2007;357:1893902. 25. Davis PG, Schmidt B, Roberts R, et al. Caffeine for Apnea of Prematurity (CAP) Trial: benets may vary in subgroups. J Pediatr, in press. 26. Saugstad OD. Bronchopulmonary dysplasia oxidative stress an anti-oxidants. Semin Neonatol 2003;8:3949. 27. Tin W, Milligan DWA, Pennefather P, et al. Pulse oximetry, severe retinopathy, and outcome at one year in babies of less than 28 weeks gestation. Arch Dis Child 2001;84:F10610. 28. Chow L, Wright KW, Sola S. Can changes in clinical practice decrease the incidence of severe retinopathy in very low birth weight infants? Pediatrics 2003;111:33945. 29. Anderson CG, Benitz WE, Madan A. Retinopathy of prematurity and pulse oximetry: a national survey of recent practices. J Perinatol 2004;24:1648. 30. Sun SC. Relation of target SpO2 levels and clinical outcome in ELBW infants on supplemental oxygen. Pediatr Res 2002;51:A350. 31. Poets C, Arand J, Hummler H, et al. Retinopathy of prematurity: a comparison between two centers aiming for different pulse oximetry saturation levels. Biol Neonate 2003;84:A267. 32. Tin W, Wariyar U. Giving small babies oxygen: 50 years of uncertainty. Semin Neonatol 2002;7:3617. 33. Silverman WA. A cautionary tale about supplemental oxygen: the albatross of neonatal medicine. Pediatrics 2004;113:3946. 34. Kang JL, Park W, Pack IS, et al. Inhaled nitric oxide attenuates acute lung injury via inhibition of nuclear factor-kappa B and inammation. J Appl Physiol 2002;92:795801.

Practice points  A more objective and physiological denition of BPD is available and clinicians should carefully scrutinise the diagnosis of BPD before initiating any treatment.  Oxygen: despite being the most commonly used therapy, there is still no good evidence to guide us as to how much oxygen is appropriate to give to preterm infants.  Evidence for the long-term efcacy and safety of caffeine therapy in preterm infants is now available through a well-designed RCT, but it took more than 30 years to establish this evidence since its introduction.  The most recent data from the European Collaborative Study concluded that low dose inhaled nitric oxide is not effective in preventing BPD.  The most recent review suggested that late use of postnatal steroids may have benecial effects in reducing the combined outcome of death and BPD, but its use should still be restricted in accordance with current guidelines until further evidence is available.

Research directions  It is expected that the ve oxygen saturation targeting trials will recruit a total of more than 5000 babies, and the prospective meta-analysis of these trials may clarify some of the myths and uncertainties of oxygen therapy.  Clinicians should consider re-examining the long-term efcacy and safety of low dose, short course postnatal steroids for infants with high risk of BPD.  Available data from systematic reviews that nutritional supplements such as vitamin A and inositol reduce the risk of BPD highlight the need for further RCTs of these relatively low cost preventative interventions.  Clinicians have a moral obligation to assess all therapies for meaningful efcacy and long-term safety by collaborative RCTs.

Conict of interest statement None declared. Funding sources None.

390

W. Tin, T.E. Wiswell / Seminars in Fetal & Neonatal Medicine 14 (2009) 383390 63. Shah SS, Ohlsson A, Halliday HL, et al. Inhaled versus systemic corticosteroids for preventing chronic lung disease in ventilated very low birth weight preterm neonates. Cochrane Database Syst Rev; 2003. CD002058. 64. Tyson JE, Wright LL, Oh W, et al. Vitamin A supplementation for extremely-lowbirth infants. N Engl J Med 1999;340:19628. 65. Wardle SP, Hughes A, Chen S, et al. Randomised controlled trial of vitamin A supplementation in preterm infants to prevent chronic lung disease. Arch Dis Child 2001;84:F913. 66. Darlow BA, Graham PJ. Vitamin A supplementation to prevent mortality and short- and long-term morbidity in very low birth weight infants. Cochrane Database Syst Rev 2007;(4). CD000501. 67. Brion LP, Bell EF, Rughuveer TS. Vitamin E supplementation for prevention of morbidity and mortality in preterm infants. Cochrane Database Syst Rev 2003;(4). CD003665. 68. Suresh G, Davis JM, Soll R. Superoxide dismutase for preventing chronic lung disease in mechanically ventilated infants. Cochrane Database Syst Rev; 2001. CD001968. 69. Ng GY, da Silva O, Ohlsson A. Bronchodilators for the prevention and treatment of chronic lung disease in preterm infants. Cochrane Database Syst Rev 2001;(3). CD003214. 70. Holt WJ, Greenspan JS, Wiswell TE, et al. Pulmonary response to an inhaled bronchodilator in chronically ventilated preterm infants with bronchopulmonary dysplasia. Respir Care 1995;40:14551. 71. Wilkie RA, Bryan MH. Effect of bronchodilators on airway resistance in ventilator-dependent neonates with chronic lung disease. J Pediatr 1987;111:27882. 72. Brundage KL, Mohsini KG, Froese AB, et al. Bronchodilator response to ipratropium bromide in infants with bronchpulmonary dysplasia. Am Rev Respir Dis 1990;142:113742. 73. Yuksel B, Greenough A. Ipratropium bromide for symptomatic preterm infants. Eur J Pediatr 1991;150:8547. 74. Brion LP, Soll RF. Diuretics for respiratory distress syndrome in premature infants. Cochrane Database Syst Rev; 2001. CD001454. 75. Schmidt B, Davis P, Moddemann D, et al. Long term effects of indomethacin prophylaxis in extremely-low-birth weight infants. N Engl J Med 2001;344:196672. 76. Fowlie PW, Davis PG. Prophylactic intravenous indomethacin for preventing mortality and morbidity in preterm infants. Cochrane Database Syst Rev; 2002. CD000174. 77. Shah SS, Ohlsson A. Ibuprofen for the prevention of patent ductus arteriosus in preterm and/or low birth weight infants. Cochrane Database Syst Rev; 2006. CD004213. 78. Howlett A, Ohlsson A. Inositol for respiratory distress syndrome in preterm infants. Cochrane Database Syst Rev; 2003. CD000366. 79. Mabanta CG, Pryhuber GS, Weinburg GA, et al. Erythromycin for the prevention of chronic lung disease in intubated preterm infants at high risk for, or colonised or infected with Ureaplasma urealyticum. Cochrane Database Syst Rev; 2003. CD003744. 80. Bancalari E, Abdenour GE, Feller R, et al. Bronchopulmonary dysplasia: clinical presentation. J Pediatr 1979;85:81923. 81. Jobe A, Bancalari E. NICHD/NHLBI/ORD workshop summary bronchpulmonary dysplasia. Am J Respir Crit Care Med 2001;163:17239. 82. Walsh M, Yao Q, Gettner P, et al. Impact of a physiologic denition on bronchpulmonary dysplasia rates. Pediatrics 2004;114:130511. 83. The STOP-ROP Multicenter Study Group. Supplemental Therapeutic Oxygen for Prethreshold Retinopathy Of Prematurity (STOP-ROP), a randomized, controlled trial. I: primary outcomes. Pediatrics 2000;105:295310. 84. Askie LM, Henderson-Smart DJ, Irwig L, et al. Oxygen-saturation targets and outcomes in extremely preterm infants. N Engl J Med 2003;349:95967. 85. Brion LP, Primhak RA. Intravenous or enteral loop diuretics for preterm infants with (or developing) chronic lung disease. Cochrane Database Syst Rev; 2002. CD001453. 86. Brion LP, Primhak RA, Yong W. Aerosolized diuretics for preterm infants with (or developing) chronic lung disease. Cochrane Database Syst Rev; 2006. CD001694. 87. Brion LP, Primhak RA, Ambrosio-Perez I. Diuretics acting on the distal renal tubule for preterm infants with (or developing) chronic lung disease. Cochrane Database Syst Rev; 2002. CD001817. 88. Halliday HL, Ehrenkranz RA, Doyle L. Late (>7days) postnatal steroids for chronic lung disease in preterm infants. Cochrane Database Syst Rev; 2009. CD001145. 89. Doyle LW, Halliday HL, Ehrenkranz RA, et al. Impact of postnatal systemic corticosteroids on mortality and cerebral palsy in preterm infants: effect modication by risk for chronic lung disease. Pediatrics 2005;115:65561. 90. Shah SS, Ohlsson A, Halliday HL, et al. Inhaled versus systemic corticosteroid for the treatment of chronic lung disease in ventilated very low birth weight preterm infants. Cochrane Database Syst Rev; 2007. CD002057. 91. Ladha F, Bonnet S, Eaton F. Sildenal improves alveolar growth and pulmonary hypertension in hyperoxia-induced lung injury. Am J Respir Crit Care Med 2005;172:7506. 92. Lai NM, Rajadurai SV, Tan K. Increased energy intake for preterm infants with (or developing) bronchopulmonary dysplasia/chronic lung disease. Cochrane Database Syst Rev; 2006. CD005093.

35. Ballard PL, Gonzales LW, Godinez RI, et al. Surfactant composition and function in a primate model of infant chronic lung disease: effect of inhaled nitric oxide. Pediatr Res 2006;59:15762. 36. McCurnin DC, Pierce RA, Chang LY, et al. Inhaled NO improves early pulmonary function and modies lung growth and elastin deposition in a baboon model of neonatal chronic lung disease. Am J Physiol Lung Cell Mol Physiol 2005;288:L4509. 37. Finer N, Barrington KJ. Nitric oxide for respiratory failure in infants born at or near term. Cochrane Database Syst Rev; 2006. CD000339. 38. Barrington KJ, Finer NN. Inhaled nitric oxide for respiratory failure in preterm infants. Cochrane Database Syst Rev; 2007. CD000509. 39. Kinsella JP, Walsh WF, Bose CL, et al. Inhaled nitric oxide in premature neonates with severe hypoxemic respiratory failure: a randomised controlled trial. Lancet 1999;354:10615. 40. Anonymous. Early compared with delayed inhaled nitric oxide in moderately hypoxaemic neonates with respiratory failure: a randomised controlled trial. The Franco-Belgium Collaborative NO Trial Group. Lancet 1999;354:106671. 41. Subhedar NV, Ryan SW, Shaw NJ. Open randomised controlled trial of inhaled nitric oxide and early dexamethasone in high risk preterm infants. Arch Dis Child 1997;77:F18590. 42. Field D, Elbourne D, Truesdale A, et al. Neonatal ventilation with inhaled nitric oxide versus ventilatory support without inhaled nitric oxide for preterm infants with severe respiratory failure: the INNOVO Multicentre Randomised Controlled Trial. Pediatrics 2005;115:92636. 43. Van Meurs KP, Wright LL, Ehrenkranz RA, et al. Inhaled nitric oxide for premature infants with severe respiratory failure. N Engl J Med 2005;353: 1322. 44. Schreiber MD, Gin-Mestan K, Marks JD, et al. Inhaled nitric oxide in premature infants with the respiratory distress syndrome. N Engl J Med 2003;349: 2099107. 45. Kinsella JP, Cutter GR, Walsh WF, et al. Early inhaled nitric oxide therapy in premature newborns with respiratory failure. N Engl J Med 2006;355: 35464. 46. Ballard RA, Truog WE, Cnann A, et al. Inhaled nitric oxide in preterm infants undergoing mechanical ventilation. N Engl J Med 2006;355:34353. 47. Mercier J, Hummler H, Dummeyer X, et al. Inhaled nitric oxide (iNO) for prevention of bronchopulmonary dysplasia (BPD) in preterm infants. The EUNO Trial. EPAS 2009; 3212.5. 48. Halliday HL, Ehrenkranz RA, Doyle LW. Early (<96 hours) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev 2003;(1). CD001146. 49. Halliday HL, Ehrenkranz RA, Doyle LW. Moderately early (714 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev 2003;(1). CD001144. 50. Halliday HL, Ehrenkranz RA, Doyle LW. Delayed (> 3 weeks) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev 2003;(1). CD001145. 51. Cummings JJ, DEugenio DB, Gross SJ. A controlled trial of dexamethasone in preterm infants at high risk for bronchopulmonary dysplasia. N Engl J Med 1989;320:150510. 52. OShea TM, Kothadia JM, Klinepeter KL. Randomised placebo controlled trial of a 42-day tapering course of dexamethasone to reduce the duration of ventilator dependency in very low birth weight infants: outcome of study participants at 1-year adjusted age. Pediatrics 1995;104:1521. 53. Yeh TF, Lin YJ, Huang CC. Early dexamethasone therapy in preterm infants: a follow up study. Pediatrics 1998;101:E7. 54. Shinwell ES, Karplus M, Reich D. Early postnatal dexamethasone treatment and increased incidence of cerebral palsy. Arch Dis Child 2000;83:F17781. 55. American Academy of Pediatrics, Committee on Fetus and Newborn, Canadian Pediatric Society, Fetus and Newborn Committee. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Pediatrics 2002;109:3308. 56. Halliday HL. Guidelines on neonatal steroids. Prenat Neonat Med 2001;6:3713. 57. Doyle LW, Davis PG, Morley CJ, et al. Low-dose dexamethasone facilitates extubation among chronically ventilator-dependent infants: a multicenter, international, randomised, controlled trial. Pediatrics 2006;117:7583. 58. Doyle LW, Davis PG, Morley CJ, et al. Outcome at 2 years of age of infants from the DART Study: a multicenter, international, randomised controlled trial of low-dose dexamethasone. Pediatrics 2007;119:71621. 59. Groneck P, Reuss D, Gotze-Speer B. Effects of dexamethasone on chemotactic activity and inammatory mediators in tracheobronchial aspirates of preterm infants at risk for chronic lung disease. J Pediatr 1993;122:93844. 60. Zimmerman JJ, Gabbert D, Shivpuri C. Meter-dosed, inhaled beclomethasone attenuates bronchoalveolar oxyradical inammation in premature infants at risk for bronchopulmonary dysplasia. Am J Perinatol 1998;15:56776. 61. Merz U, Kusenbach G, Hausler M. Inhaled budesonide in ventilator-dependent preterm infants: a randomised double-blind pilot study. Biol Neonate 1999;75:4653. 62. Shah V, Ohlsson A, Halliday HL, Dunn M. Early administration of inhaled corticosteroids for preventing chronic lung disease in ventilated very low birth weight preterm neonates. Cochrane Database Syst Rev; 2007. CD001969.

Seminars in Fetal & Neonatal Medicine 14 (2009) 391395

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Long-term outcomes of bronchopulmonary dysplasia


Lex W. Doyle a, b, c, e, *, Peter J. Anderson d, e
a

Department of Obstetrics and Gynaecology, University of Melbourne, Melbourne, Australia Department of Paediatrics, University of Melbourne, Melbourne, Australia c Research Ofce, The Royal Womens Hospital, Melbourne, Australia d Department of Psychology, University of Melbourne, Melbourne, Australia e Murdoch Childrens Research Institute, Melbourne, Australia
b

s u m m a r y
Keywords: Behaviour Bronchopulmonary dysplasia Cognition Infant Lung function Preterm

As more very preterm infants survive, more survivors will have bronchopulmonary dysplasia (BPD). Children with BPD have higher rates of cognitive, educational and behavioural impairments, and also reduced lung function, through childhood and into early life than would normally be expected. The importance of these neurological and respiratory problems later into adult life needs to be determined. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Mechanical ventilation was introduced into neonatal nurseries in the 1960s, and shortly thereafter the rst reports of bronchopulmonary dysplasia (BPD) appeared, mostly in babies who were not very preterm (>31 weeks of gestation) and who had birth weights >1499 g.1 As more babies survived mechanical ventilation, more survivors with the classical scarring and cystic type of BPD were discharged from neonatal nurseries. These survivors born in the 1970s and early 1980s had poor neurodevelopmental outcomes,2 and abnormal lung function.3 The new BPD is characterised more by alveolar arrest and less by brosis,4 and since survivors with BPD are now much smaller and more immature at birth, the outcomes for survivors of the new BPD need to be continually reviewed. Outcomes of most interest are neurological and respiratory. 2. Neurological outcomes

2.1. Neurosensory problems Cerebral palsy occurs more frequently in children with BPD. In one study of infants of very low birth weight (VLBW, <1500 g), 15% of survivors with BPD, dened as oxygen dependence >27 days, had cerebral palsy compared with 34% in those treated with oxygen for <28 days.6 BPD was a signicant risk factor for cerebral palsy at 1822 months corrected age in a study of 827 infants <25 weeks born between 1993 and 1999 [adjusted odds ratio: 1.66, 95% condence interval (CI): 1.012.74].7 Apart from cerebral palsy, some infants with BPD have a specic movement disorder affecting the limbs, neck, trunk, and oralbuccallingual movements,8 and they also exhibit poorer ne and gross motor skills than VLBW children without BPD.9,10 BPD is associated with visual11 and auditory12 problems less severe than legal blindness or deafness requiring amplication.

2.2. General cognitive functioning Neurosensory problems occur more frequently in preterm survivors with the new BPD compared with preterm survivors without BPD.5 A summary of some of the more common neurological problems follows, updated where necessary, since an earlier review in 2006.5 Children with BPD score lower on tests of early development.1316 In addition, rates of cognitive and motor delay in a large cohort of VLBW preschoolers with BPD were more than double that of VLBW peers without BPD.15 In general, school-aged children with BPD have lower IQs than VLBW children without BPD,10,1721 although some studies have failed to nd signicant differences between groups.9,22 In those studies that have reported a difference in IQ between VLBW children with and without BPD, the difference ranges from one-quarter to two-thirds of a standard deviation.10,1821

* Corresponding author. Address: Department of Obstetrics and Gynaecology, The Royal Womens Hospital, 20 Flemington Rd, Parkville, NSW 3052, Australia. Tel.: 61 3 8345 3716; fax: 61 3 8345 3702. E-mail address: lwd@unimelb.edu.au (L.W. Doyle). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.004

392

L.W. Doyle, P.J. Anderson / Seminars in Fetal & Neonatal Medicine 14 (2009) 391395

2.3. Attention Attentional impairments are more frequent and severe in children with BPD; in one study 59% of VLBW children with BPD were classied as having an attentional impairment (i.e. scores >1 SD below normal expectations), signicantly greater than VLBW children without BPD (32%).23 Similarly, on a Continuous Performance Task, BPD children made more errors than controls, suggestive of impaired sustained attention.10 Consistent with these ndings, the rate of attention decithyperactivity disorder (ADHD) in 8-year-old VLBW children with BPD has been reported to be 15%, double that of non-BPD VLBW children.10 2.4. Language Speech and language disorders are more common in preterm children with BPD.24 Singer et al.25 reported that children with BPD at preschool age displayed less developed receptive and expressive language skills than VLBW children without BPD; 49% of the BPD preschoolers were signicantly delayed in receptive language development, and 43% were signicantly delayed in expressive language development. School-aged VLBW children with BPD also have more frequent receptive language impairments than do VLBW children without BPD.17,23,26 In one study, 15% of children with BPD had a signicant receptive language impairment and 9% had a signicant expressive language impairment, more than double that of non-BPD VLBW peers.26 Approximately 50% of school-aged VLBW children with BPD were enrolled in speechlanguage therapy in one study, in contrast with about 20% of VLBW children without BPD.10,26 2.5. Memory and learning Although studies are few, children with BPD exhibit more memory difculties than expected from prematurity per se. In one study of 17 VLBW children with BPD and 28 non-BPD peers, 65% of the BPD group exhibited a memory impairment (>1 SD below expectations) compared with 29% of the non-BPD group, and they had particular difculty with immediate auditory memory and working memory.23 2.6. Visualspatial perception A number of studies have found that children with BPD perform signicantly less well than controls on the Developmental Test of VisualMotor Integration (VMI),17,20,21 suggesting visualspatial perceptual decits. Nearly 30% of children with BPD are reported to perform below age expectations on the VMI.21 Consistent with these ndings, a longitudinal study by Taylor et al.19,27 reported an association between perceptual difculties and BPD. They found that elementary school-aged children with BPD had a specic impairment on perceptual motor tasks,19 and a later follow-up (at a mean age of 16 years) revealed that weeks of oxygen requirement was strongly associated with a number of measures that require visualspatial perceptual skills,27 including the VMI, among other tests. 2.7. Executive skills Executive function refers to a collection of interrelated processes responsible for purposeful, goal-directed behaviour, and is important in a childs cognitive functioning, behaviour, emotional control, and social interaction.28 The principal components of executive function include anticipation, goal selection, planning and organisation, initiation of activity, self-regulation, mental exibility,

deployment of attention, working memory, and utilisation of feedback.29 Executive decits have been reported in very preterm children,27,30,31 but again research examining the long-term effects of BPD on executive functioning is lacking. Taylor et al.27 found measures of planning ability, mental exibility, working memory, and self-monitoring to be negatively correlated with duration of oxygen requirement in VLBW children, suggesting that children with BPD may be at increased risk for executive dysfunction. 2.8. Academic performance Children with a history of BPD have more school-based problems than those without BPD.9,10,17,20,21 In one study, children with BPD performed signicantly below VLBW children without BPD and full-term controls on tests of reading and mathematics,20 and in another study, the rate of problems in reading and mathematics was reported to be as high as 47%.23 Poorer spelling skills have also been reported in school-aged children with BPD.17 Children with BPD are more likely to require educational assistance,9,23 with one study reporting that more than 50% of children with BPD received special education services.10 2.9. Behavioural problems Preterm children with BPD exhibit more behavioural problems than peers without BPD.17,19 In a recent study, BPD children scored worse than classroom controls on several aspects of behaviour, including internalising behaviours and inattention.32 3. Respiratory outcomes of BPD 3.1. Pulmonary outcomes for the oldest survivors of BPD The oldest subjects with BPD to have lung function data reported have been in the late teens or early twenties. Northway et al.3 reported the lung function in late adolescence (mean age 18 years) of 26 subjects who had the old BPD (cystic changes with scarring) and who were born between 1964 and 1973 compared with 26 age-matched controls of similar birthweight and gestational age who had not been ventilated as infants, and 53 age-matched normal subjects who were not born prematurely, who had no history of chronic lung disease and who were non-smokers. In their study, 68% of BPD subjects had airway obstruction; this was reversible in most, but xed in 24%. In addition, those with BPD had reductions in variables reecting airow [lower forced expired volume in 1 s (FEV1), forced vital capacity (FVC), maximum forced expiratory ow between 25% and 75% of vital capacity (FEF2575%), instantaneous ow at 50% of vital capacity (VEMAx50%), and peak expiratory ow rate (PEFR)], and increased gas trapping [higher residual volume (RV)/total lung capacity (TLC) ratio] compared with both the preterm controls and the normal controls. A quarter had one or more severe abnormalities in lung function. BPD subjects had more wheezing, episodes of pneumonia, limitation of exercise capacity, and long-term medication use than either control group, but precise rates of these outcomes were not reported. They also had more chronic changes on chest radiography. Halvorsen et al.33 reported the pulmonary outcomes for 46 subjects of birth weight <1001 g or gestational ages <29 weeks at a mean age of 17.7 years from a geographically based cohort of subjects born between 1982 and 1985 in Western Norway. Twelve (26%) of the subjects had moderate or severe BPD based on oxygen requirement at 36 weeks postmenstrual age, 24 (52%) had mild BPD based on oxygen requirement at 28 days but not 36 weeks, and 10 (22%) had no BPD. They compared results with 46 temporally related controls who were born at term and of the same gender.

L.W. Doyle, P.J. Anderson / Seminars in Fetal & Neonatal Medicine 14 (2009) 391395

393

Within the preterm group, the FEV1 (% predicted) was 9.9% (95% CI: 1.018.8%) lower in the moderate/severe BPD subjects compared with the remainder. Doyle et al.34 studied 147 survivors of birth weight <1500 g from the Royal Womens Hospital, Melbourne, who were born during 19771982 and who had lung function tests at a mean age of 18.9 (SD: 1.1) years. Of the 147 subjects, 33 (22%) had BPD in the newborn period, as dened in the original Northway study.1 There were also 37 normal birth weight (NBW) controls with lung function tests. All lung function variables reecting airow were substantially diminished in the BPD group, but lung volumes were not signicantly different. Within the preterm group, the FEV1 (% predicted) was 11.3% (95% CI: 5.716.9%) lower in the BPD subjects than in preterm controls, and both groups were lower than NBW controls. More subjects in the BPD group had reductions in airow in the clinically signicant range (e.g. FEV1 < 75% predicted, BPD 30.3%, no BPD 7.9%, c2 11.4, P 0.001; FEV1/FVC <75%, BPD 42.4%, no BPD 15.8%, c2 10.7, P 0.001). The rates of asthma were similar in those with BPD (24.2%) and no BPD (21.9%). In a recent study of 74 survivors with either birth weight <1001 g or gestational age <29 weeks born in Norway in either 19821985 or 19911992 and studied at 18 years or 10 years of age, respectively, lung parenchymal abnormalities detected by pulmonary high-resolution computed tomography (HRCT) were strongly related to increasing durations of oxygen therapy and worse lung function, as reected in lower FEV1 with higher HRCT abnormality scores.35 These changes relatively early in life are worrisome, especially as lung function normally deteriorates in all subjects after the mid-twenties, and it may decline even more quickly in those who had BPD in the newborn period. 3.2. Changes in lung function with increasing age in BPD survivors There are few studies with longitudinal data in the same subjects, and only one that has reported changes from early school age up to late adolescence/early adulthood.34 In this study, data at 8 years and 18 years of age in 129 subjects of birth weight <1500 g were described; 29 of the 129 subjects had BPD. Compared with lung function variables measured at 8 years, the only variable with a statistically signicant difference over time in BPD subjects was a larger fall in the FEV1/FVC ratio between 8 and 18 years of age compared with non-BPD preterm subjects (mean difference: 3.4%; 95% CI: 0.2% to 6.7%). This greater than expected fall in FEV1/FVC ratio is worrisome, as lung function may decline more rapidly in BPD subjects in later adulthood. Others have suggested that lung function in survivors with BPD might improve as they grow older. Blayney et al.36 studied 32 children with BPD at both 7 and 10 years of age. The 19 with an FEV1 < 80% predicted at age 7 years improved by 10 years of age, but this may just reect regression towards the mean. In another study where lung function tests were repeated in 17 subjects with BPD between 8 and 15 years, Koumbourlis et al.37 reported that reductions in airow persisted over time, although there was improvement in air trapping. 3.3. Effect of exogenous surfactant on respiratory outcome for new BPD survivors The effect of surfactant administered soon after birth on lung function on small numbers of children enrolled in clinical trials has been reported to be minimal,38 or possibly benecial.39 However, surfactant has ushered in the era of the new BPD, where lung scaring and cystic change are minimal, and alveolar arrest is the dominant pathology.4 In one study of the effect of new BPD in the surfactant era on lung function, 34 VLBW children at 78 years of

age who had BPD diagnosed by oxygen dependency at 28 days of age, 14 of whom still had oxygen dependency at 36 weeks postmenstrual age, were compared with age- and sex-matched controls comprising 34 VLBW cases without BPD and 34 term children. Compared with term controls, the new BPD cases had lower FEV1, and those with oxygen dependency had lower FEV1 than both control groups, results similar to the pre-surfactant era.40 However, the results were somewhat more encouraging in another study, which reported the results of a geographical cohort study of 298 consecutive ELBW/very preterm survivors born in 19911992, after surfactant was introduced.41 Lung function was measured in 81% (240/298) of ELBW/very preterm children at a mean age of 8.7 (SD: 0.3) years, and in 79% (208/262) of NBW controls at a mean age of 8.9 (SD: 0.4) years. Of the 240 ELBW/very preterm children cohort, 89 (37%) had BPD in the newborn period, dened as oxygen dependency at 36 weeks postmenstrual age in children who had received assisted ventilation via an endotracheal tube. Most children with BPD had lung function within the expected range. However, some variables reecting airow were reduced in children with BPD, compared with both ELBW/very preterm children without BPD, as well as with NBW controls, but the differences were not as marked as in the pre-surfactant era. For example, within the ELBW/very preterm group, the mean (SD) for FEV1 (% predicted) was 81.1% (13.7%) in the BPD group, and 87.1% (11.5%) in the non-BPD group, a mean difference of 6.0% (95% CI: 2.79.3%). This is approximately half the difference of 11.3% (95% CI: 5.716.9%) described above for the pre-surfactant era, although the denition of BPD was different, and the subjects were older and less premature in the early study.34 3.4. Other respiratory health issues related to BPD In most studies, high rehospitalisation rates are reported in children with BPD, with rates varying from 40% to 60% in the rst 2 years of life.4248 Rates of rehospitalisation are generally higher in early life in children with BPD; in one study Gross et al.47 reported that 53% of children <32 weeks gestational age with BPD required rehospitalisation during the rst 2 years of life compared with 26% in those without BPD (P < 0.01). Children with BPD are also more likely to require multiple rehospitalisations and longer hospital stays. Cunningham et al.42 reported that 26% of the BPD group versus 5% of the non-BPD group had multiple rehospitalisations, as well as a cumulative hospital stay of >31 days in 12% of the BPD group compared with 5% in the non-BPD group. Chien et al.48 found a signicant difference in the length of hospital stay in infants with BPD compared with those without BPD (median 10 days vs 3 days). As children with a history of BPD enter school age, it appears that the risk of rehospitalisation may be similar to their non-BPD peers. In a study of VLBW children at ages 810 years, McCormick et al.49 reported that the risk of rehospitalisation in the preceding year was similar for BPD (6%) and non-BPD (7%) groups. In another study of VLBW and NBW cohorts at 14 years of age, rehospitalisation for respiratory illness was infrequent in all groups, including those with BPD.50 Rates of wheezing or asthma, or of bronchial hyper-responsiveness, are generally reported to be higher in BPD survivors.33,47,5153 However, in another study, the rates of asthma (dened as requiring bronchodilator therapy for recurrent wheezing in the previous 12 months) were similar in the late adolescents/young adults with BPD (24.2%) and without BPD (21.9%).34 4. Conclusions Compared with children without BPD, those with BPD have higher rates of adverse neurological outcomes, including motor,

394

L.W. Doyle, P.J. Anderson / Seminars in Fetal & Neonatal Medicine 14 (2009) 391395 21. Gray PH, OCallaghan MJ, Rogers YM. Psychoeducational outcome at school age of preterm infants with bronchopulmonary dysplasia. J Paediatr Child Health 2004;40:11420. 22. Bohm B, Katz-Salamon M. Cognitive development at 5.5 years of children with chronic lung disease of prematurity. Arch Dis Child Fetal Neonatal Ed 2003;88:F1015. 23. Farel AM, Hooper SR, Teplin SW, Henry MM, Kraybill EN. Very-low-birthweight infants at seven years: an assessment of the health and neurodevelopmental risk conveyed by chronic lung disease. J Learn Disabil 1998;31:11826. 24. Casiro OG, Moddemann DM, Stanwick RS, Panikkar-Thiessen VK, Cowan H, Cheang MS. Language development of very low birth weight infants and fullterm controls at 12 months of age. Early Hum Dev 1990;24:6577. 25. Singer LT, Siegel AC, Lewis B, Hawkins S, Yamashita T, Baley J. Preschool language outcomes of children with history of bronchopulmonary dysplasia and very low birth weight. J Dev Behav Pediatr 2001;22:1926. 26. Lewis BA, Singer LT, Fulton S, et al. Speech and language outcomes of children with bronchopulmonary dysplasia. J Commun Disord 2002;35:393406. 27. Taylor HG, Minich N, Bangert B, Filipek PA, Hack M. Long-term neuropsychological outcomes of very low birth weight: associations with early risks for periventricular brain insults. J Int Neuropsychol Soc 2004;10: 9871004. 28. Gioia G, Isquith P, Guy S. Assessment of executive functions in children with neurological impairment. In: Simeonsson R, Rosenthal S, editors. Psychological and developmental assessment: children with disabilities and chronic conditions. New York: Guilford Press; 2001. p. 31756. 29. Anderson P. Assessment and development of executive function (EF) during childhood. Child Neuropsychol 2002;8:7182. 30. Espy KA, Stalets MM, McDiarmid MM, Senn TE, Cwik MF, Hamby A. Executive functions in preschool children born preterm: application of cognitive neuroscience paradigms. Neuropsychol Dev Cogn Sect C Child Neuropsychol 2002;8:8392. 31. Anderson PJ, Doyle LW, for the Victorian Infant Collaborative Study Group. Executive functioning in school-aged children who were born very preterm or with extremely low birth weight in the 1990s. Pediatrics 2004;114:507. 32. Gray PH, OCallaghan MJ, Poulsen L. Behaviour and quality of life at school age of children who had bronchopulmonary dysplasia. Early Hum Dev 2008;84:18. 33. Halvorsen T, Skadberg BT, Eide GE, Roksund OD, Carlsen KH, Bakke P. Pulmonary outcome in adolescents of extreme preterm birth: a regional cohort study. Acta Paediatr 2004;93:1294300. 34. Doyle LW, Faber B, Callanan C, Freezer N, Ford GW, Davis NM. Bronchopulmonary dysplasia in very low birth weight subjects and lung function in late adolescence. Pediatrics 2006;118:10813. 35. Aukland SM, Rosendahl K, Owens CM, Fosse K, Eide GE, Halvorsen T. Neonatal bronchopulmonary dysplasia predicts abnormal pulmonary HRCT in long term survivors of extreme preterm birth. Thorax 2009;64:40510. 36. Blayney M, Kerem E, Whyte H, OBrodovich H. Bronchopulmonary dysplasia: improvement in lung function between 7 and 10 years of age. J Pediatr 1991;118:2016. 37. Koumbourlis AC, Motoyama EK, Mutich RL, Mallory GB, Walczak SA, Fertal K. Longitudinal follow-up of lung function from childhood to adolescence in prematurely born patients with neonatal chronic lung disease. Pediatr Pulmonol 1996;21:2834. 38. Gappa M, Berner MM, Hohenschild S, Dammann CE, Bartmann P. Pulmonary function at school-age in surfactant-treated preterm infants. Pediatr Pulmonol 1999;27:1918. 39. Pelkonen AS, Hakulinen AL, Turpeinen M, Hallman M. Effect of neonatal surfactant therapy on lung function at school age in children born very preterm. Pediatr Pulmonol 1998;25:18290. 40. Korhonen P, Laitinen J, Hyodynmaa E, Tammela O. Respiratory outcome in school-aged, very-low-birth-weight children in the surfactant era. Acta Paediatr 2004;93:31621. 41. Doyle LW, the Victorian Infant Collaborative Study Group. Respiratory function at age 89 years in extremely low birthweight/very preterm children born in Victoria in 199192. Pediatr Pulmonol 2006;41:5706. 42. Cunningham CK, McMillan JA, Gross SJ. Rehospitalization for respiratory illness in infants of less than 32 weeks gestation. Pediatrics 1991;88:52732. 43. Chye JK, Gray PH. Rehospitalization and growth of infants with bronchopulmonary dysplasia: a matched control study. J Paediatr Child Health 1995;31:10511. 44. Furman L, Hack M, Watts C, et al. Twenty-month outcome in ventilatordependent, very low birth weight infants born during the early years of dexamethasone therapy. J Pediatr 1995;126:43440. 45. Furman L, Baley J, Borawski-Clark E, Aucott S, Hack M. Hospitalization as a measure of morbidity among very low birth weight infants with chronic lung disease. J Pediatr 1996;128:44752. 46. Gregoire MC, Lefebvre F, Glorieux J. Health and developmental outcomes at 18 months in very preterm infants with bronchopulmonary dysplasia. Pediatrics 1998;101:85660. 47. Gross SJ, Iannuzzi DM, Kveselis DA, Anbar RD. Effect of preterm birth on pulmonary function at school age: a prospective controlled study. J Pediatr 1998;133:18892. 48. Chien YH, Tsao PN, Chou HC, Tang JR, Tsou KI. Rehospitalization of extremelylow- birth-weight infants in rst 2 years of life. Early Hum Dev 2002;66:3340.

visual and auditory problems. They exhibit low average IQ, more academic difculties, delayed speech and language development, more visualmotor integration impairments and behaviour problems, and they have more attention problems, memory and learning decits, and executive dysfunction. Subjects with BPD have worse respiratory function and more respiratory ill-health than those of similar size and gestational age who did not have BPD. As survival rates of ELBW and very preterm babies are increasing rapidly, with no decrease in the rate of BPD in survivors, the increasing number of children with BPD is going to add to the burden of neurological and respiratory disease into later life. Reducing the rate of BPD remains one of the biggest challenges in neonatal care.4 Conict of interest statement None declared. Funding sources None. References
1. Northway Jr WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med 1967;276:35768. 2. Saigal S, OBrodovich H. Long-term outcome of preterm infants with respiratory disease. Clin Perinatol 1987;14:63550. 3. Northway Jr WH, Moss RB, Carlisle KB, et al. Late pulmonary sequelae of bronchopulmonary dysplasia. N Engl J Med 1990;323:17939. 4. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J Respir Crit Care Med 2001;163:17239. 5. Anderson PJ, Doyle LW. Neurodevelopmental outcome of bronchopulmonary dysplasia. Semin Perinatol 2006;30:22732. 6. Skidmore MD, Rivers A, Hack M. Increased risk of cerebral palsy among very low- birthweight infants with chronic lung disease. Dev Med Child Neurol 1990;32:32532. 7. Hintz SR, Kendrick DE, Stoll BJ, et al. Neurodevelopmental and growth outcomes of extremely low birth weight infants after necrotizing enterocolitis. Pediatrics 2005;115:696703. 8. Perlman JM, Volpe JJ. Movement disorder of premature infants with severe bronchopulmonary dysplasia: a new syndrome. Pediatrics 1989;84:2158. 9. Vohr BR, Coll CG, Lobato D, Yunis KA, ODea C, Oh W. Neurodevelopmental and medical status of low-birthweight survivors of bronchopulmonary dysplasia at 10 to 12 years of age. Dev Med Child Neurol 1991;33:6907. 10. Short EJ, Klein NK, Lewis BA, et al. Cognitive and academic consequences of bronchopulmonary dysplasia and very low birth weight: 8-year-old outcomes. Pediatrics 2003;112:e359. 11. McGinnity FG, Halliday HL. Perinatal predictors of ocular morbidity in school children who were very low birthweight. Paediatr Perinat Epidemiol 1993;7:41725. 12. Gray PH, Sarkar S, Young J, Rogers YM. Conductive hearing loss in preterm infants with bronchopulmonary dysplasia. J Paediatr Child Health 2001;37: 27882. 13. Goldson E. Severe bronchopulmonary dysplasia in the very low birth weight infant: its relationship to developmental outcome. J Dev Behav Pediatr 1984;5:1658. 14. Landry SH, Fletcher JM, Denson SE, Chapieski ML. Longitudinal outcome for low birth weight infants: effects of intraventricular hemorrhage and bronchopulmonary dysplasia. J Clin Exp Neuropsychol 1993;15:20518. 15. Singer L, Yamashita T, Lilien L, Collin M, Baley J. A longitudinal study of developmental outcome of infants with bronchopulmonary dysplasia and very low birth weight. Pediatrics 1997;100:98793. 16. Schmidt B, Asztalos EV, Roberts RS, Robertson CM, Sauve RS, Whiteld MF. Impact of bronchopulmonary dysplasia, brain injury, and severe retinopathy on the outcome of extremely low-birth-weight infants at 18 months: results from the trial of indomethacin prophylaxis in preterms. J Am Med Assoc 2003;289:11249. 17. Robertson CM, Etches PC, Goldson E, Kyle JM. Eight-year school performance, neurodevelopmental, and growth outcome of neonates with bronchopulmonary dysplasia: a comparative study. Pediatrics 1992;89:36572. 18. OShea TM, Goldstein DJ, deRegnier RA, Sheaffer CI, Roberts DD, Dillard RG. Outcome at 4 to 5 years of age in children recovered from neonatal chronic lung disease. Dev Med Child Neurol 1996;38:8309. 19. Taylor HG, Klein N, Schatschneider C, Hack M. Predictors of early school age outcomes in very low birth weight children. J Dev Behav Pediatr 1998;19: 23543. 20. Hughes CA, OGorman LA, Shyr Y, Schork MA, Bozynski ME, McCormick MC. Cognitive performance at school age of very low birth weight infants with bronchopulmonary dysplasia. J Dev Behav Pediatr 1999;20:18.

L.W. Doyle, P.J. Anderson / Seminars in Fetal & Neonatal Medicine 14 (2009) 391395 49. McCormick MC, Workman-Daniels K, Brooks-Gunn J, Peckham GJ. Hospitalization of very low birth weight children at school age. J Pediatr 1993;122:3605. 50. Doyle LW, Cheung MM, Ford GW, Olinsky A, Davis NM, Callanan C. Birth weight <1501 g and respiratory health at age 14. Arch Dis Child 2001;84:404. 51. Ng DK, Lau WY, Lee SL. Pulmonary sequelae in long-term survivors of bronchopulmonary dysplasia. Pediatr Int 2000;42:6037.

395

52. Palta M, Sadek-Badawi M, Sheehy M, et al. Respiratory symptoms at age 8 years in a cohort of very low birth weight children. Am J Epidemiol 2001;154:5219. 53. Halvorsen T, Skadberg BT, Eide GE, Roksund O, Aksnes L, Oymar K. Characteristics of asthma and airway hyper-responsiveness after premature birth. Pediatr Allergy Immunol 2005;16:48794.

Seminars in Fetal & Neonatal Medicine 14 (2009) 396400

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

End of life decisions in chronic lung disease


Malcolm Chiswick*
University of Manchester, and Saint Marys Hospital, Manchester, UK

s u m m a r y
Keywords: Bronchopulmonary dysplasia Ethics Neonatal intensive care Palliative care

Staff may be reluctant to discuss end of life decisions in chronic lung disease (CLD) as it is usual for the disease to take a prolonged course and most infants recover to be discharged home without supplemental oxygen. A minority suffer a protracted and very severe illness in spite of treatments. Further intensive care may prolong a distressing death rather than offer any hope of survival. An end of life decision may be made after discussions with parents. Assisted ventilation may be withdrawn, or care redirected to withhold further episodes of assisted ventilation. A lingering death is a risk in infants who have not yet reached the point of dying but whose care has been redirected. Tachypnoea, rib retractions and agitation are distressing for the infant and parents. Palliative care must meet the needs of parents as well as their baby. It includes the legal use of drugs to relieve the infants distress. 2009 Elsevier Ltd. All rights reserved.

1. Introduction Most infants with chronic lung disease (CLD) pursue a relatively benign course and, in spite of a prolonged period of oxygen dependency in the neonatal unit, they are discharged home without the need for supplemental oxygen. However, in a minority the disease progressively worsens culminating in severe respiratory failure, pulmonary hypertension, and in some cases right heart failure with cardiomegaly, hepatomegaly and oedema features that were more common in earlier descriptions of bronchopulmonary dysplasia.1 Deterioration may be episodic and precipitated by pulmonary infection with the infant never quite regaining the ground that has been lost, and in due course further deterioration occurs. In some infants airway damage is manifest as bronchomalacia with signs of airway obstruction. Periods of extreme agitation can make their ventilator care difcult with endotracheal tube stability being a problem. There may be a gradual realisation that treatment is prolonging death rather than offering a reasonable hope of survival, or there may be concern that the price of survival will be signicant longterm neurodisability. The withdrawal of life support treatment for critically ill newborn, or redirection of their care towards facilitating a peaceful death, is widely practised and has featured in guidance recommended by professional paediatric organisations.2,3 When considering end of life decisions there is a need to ensure

that parents have been well informed about their infants medical state including the prognosis and treatment options; that they have actively participated in decision-making; that consideration has been given to the comfort of the infant at all times; and that decisions are made in the best interests of the infant.

2. The best interests concept A recurring theme among medical ethicists is the need to act in the best interests of patients. We need to examine what practical help this might offer when considering end of life decisions in infants with deteriorating CLD. Leaving aside the philosophical issue of whether extremely preterm babies with or without CLD have any interest in living or dying, there seem to be two different but overlapping concepts in the best interests argument. First is the notion that the interests of the infant are of prime importance and surpass any interest that others might have in the infant. This is easy to understand if we argue that the infants interests are more important than, say, those of an inexperienced neonatologist who might see the continuing battle to save the life of an extremely preterm infant with CLD as a medical challenge. However, the babys interests are intimately linked with those of the parents. Their own interests cannot be ignored indeed that is the point of ensuring that parents are involved in decision-making, and it is probably impossible for them to offer a view without them being inuenced by their own feelings and prejudices. Fortunately, on most occasions, but not all, there is agreement with what the parents want for their baby and what the neonatologist sees as their babys best interests from a medical perspective.4

* Highclere, Parkeld Road, Altrincham, Cheshire WA14 2BT, UK. Tel.: 44 (0)161 928 8579; fax: 44 (0)161 929 5564. E-mail address: m.chiswick@manchester.ac.uk 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.08.005

M. Chiswick / Seminars in Fetal & Neonatal Medicine 14 (2009) 396400

397

The other interpretation of best interests is more difcult. It does not depend on a competing interest between the infant and others, but relies instead on the idea that there exists a true and unbiased understanding of the infants best interests. The limitations of this concept and the difculties of avoiding prejudices have recently been reviewed.5 How can we test what the infants best interests are instead of using phrases like I acted in the infants bests interests as a conclusion? Perhaps the closest we can get to a test is by assessing the balance between the likely outcome for the baby against the burden of treatment, which essentially means the degree of pain and suffering. It is a medical judgement whether an infant with severe and deteriorating CLD has reached the stage where death is likely to occur within a few days or weeks in spite of all available care. If that is the case, then further treatment is likely to prolong death rather than offering the infant any reasonable hope of survival. Treatment would not only be futile but it would entail an unnecessary burden and so would not be in the infants best interests. A more difcult situation occurs when it is suggested that the infant will survive but only at the cost of signicant neurodisability. The relationship between CLD and neurodevelopment has been widely referred to.69 However, in individual infants with deteriorating CLD it is rarely possible to predict the long-term outcome should the infant survive let alone make predictions about the degree of neurological impairment. The issues that ethicists urge us to consider are often unanswerable for example, whether the child will lead a reasonably independent existence, be capable of establishing relationships with others and experience pleasure, or whether there will be an intolerable burden of disability. The predictive value of abnormal neurological signs in infants being ventilated for CLD is limited as many are receiving muscle relaxants, analgesics, or sedatives. Perhaps the closest we can get to prediction of neurodisability is when there is brain scan evidence of severe periventricular haemorrhage or ischaemic damage. However, these infants are often on a downhill course in spite of treatments, and withdrawal of ventilatory support may be indicated because death is likely, not because of concerns about survival with an intolerable burden of disability. 2.1. The burden of intensive care We cannot know for certain the true burden of intensive care for an ill preterm infant as this relies on knowledge of the conscious perception of pain, discomfort and stress in the preterm newborn. It would be foolhardy and incoherent to assume that the infant is oblivious to the numerous interventions to which it is exposed. If that was the case then almost anything would be acceptable and the burden of care for the infant would not feature in determining best interests. The pragmatic approach is to assume that there is little difference between what we as adults would consider as pain and discomfort if we were receiving the same sort of intensive care. There is the danger of a shifting sands attitude towards the burden of treatments. For example, if it is felt that continuing full intensive care is appropriate in a critically ill infant with CLD it may be argued that the burden of care has been carefully considered and reduced by the judicious use of a muscle relaxant, analgesics and sedatives, and a policy of minimal handling. Yet in another baby managed in the same way the huge burden of intensive care may be used to justify withdrawal of assisted ventilation. In summary, the best interests tenet reminds us that the infants interest are paramount but the parents have a large say in determining what those interests are. Parents also suffer the burden of intensive care for their baby but we should not be driven by that when considering end of life decisions. We are not very good at predicting neurodevelopmental outcomes in individual

infants with CLD in whom survival is anticipated. What it boils down to is that end of life decisions are more likely to be based on the notion that the infant is so ill that death is inevitable within hours or days and that further treatments would be futile, rather than being based on quality of life considerations. 3. The realisation of futility As the majority of infants with CLD gradually improve, there is a powerful motivation to offer the deteriorating infant continuing medical support. Indeed, the longer the infant has been provided with care the more difcult it can become for staff to step back from the incubator and ponder on whether the infant has a realistic chance of surviving. The infant with episodic deterioration is a particular challenge. There is the temptation for some neonatologists to be drawn on a roller, repeatedly modifying treatments without it dawning that any improvements are short-lived and that overall the infant is losing ground. When this happens, engaging parents in discussions about the possible limits of future care may come very late, giving them insufcient time to feel that they have materially contributed to decision-making. 3.1. Staff behaviour and unspoken signals Neonatal staff are more likely to feel at ease with their own thoughts about potential end of life decisions when they are shared with other members of the team and when they feel that everyone is working in the same direction. This is facilitated in neonatal units that are characterised by strong leadership, teamwork, and a forum for discussing ethical matters. In their absence, end of life decisions in CLD become an enormous challenge and may result in unclear and untimely advice being given to parents which reect uncertainties and sometimes conict between staff members. Attention has previously been drawn to unspoken signals on neonatal units when staff are reluctant to raise openly their suspicion that intensive care is becoming a futile exercise.10 This behavior is probably less common now. However, the insidious deterioration of some infants with CLD when the experience is of so many making good progress can inhibit staff from raising concerns about the direction of further care and instead they may revert to unspoken signals. For example, previously well-motivated staff seem to switch off and show a lack of interest in discussing the further care of the infant, or they engage in window gazing or some other diversion. It is out of keeping with their usual behaviour. Nurses and junior medical staff may use expressions that dramatise clinical signs probably as a desperate attempt to indicate that the infant is making no progress. For example, an infant being ventilated may be described as struggling for breath. Skin trauma from monitoring devices may be described as the skin peeling off in layers. When different treatments are discussed on ward rounds staff may respond by highlighting the potential untoward effects of all treatments that are suggested (therapeutic nihilism). They may well be correct but there is a reluctance to consider the balance of harm against benet simply because they feel that the infant is out of reach of any benet. They are reluctant to say so. Staff may suggest calling in a specialist such as a nephrologist or cardiologist to advise on organ system failure. This is usually a cry for help and an indication that hopefully the specialist might indicate that there is nothing further that can help the infant. Small groups may form among staff who discuss the futility of further intensive care while being reluctant to express this openly on ward rounds. In contrast, when staff are comfortable about

398

M. Chiswick / Seminars in Fetal & Neonatal Medicine 14 (2009) 396400

reecting on, discussing, and questioning the care they are providing for a critically ill infant with CLD then it is more likely that information will be shared in a timely way with the parents. 4. Talking with parents When withdrawal or redirection of intensive care is a consideration, parents want to know what the options are and to be part of decision-making. Engaging them in discussion is considered to be good practice.2,3 However, infants deteriorating with CLD present special problems in this respect. When good progress had been made during the rst week of life, the subsequent slow evolution of increasing oxygen or ventilator dependency may be difcult for parents to understand. Parents of extremely preterm infants may have been advised before the birth that a good response to resuscitation in the delivery room would lead to full support being continued in the neonatal intensive care unit. The benet of antenatal corticosteroids and postnatal surfactant in reducing the risk of respiratory distress syndrome may have been stressed. Progressive and protracted deterioration may be met with disbelief or anger, tinged with a suspicion that something must have gone wrong with their babys management on the intensive care unit. Suddenly presenting parents with the option of withdrawing assisted ventilation at the time of a crisis is unhelpful. In contrast, keeping parents informed on a daily basis about their babys condition, treatments and progress facilitates their participation in end of life decisions. It allows time for them to reect, and for some it is an opportunity to discuss issues with other family members or their religious advisor. There should rarely be an urgent need for parents to express their opinion. If it is felt that the infant is rapidly approaching a moribund state where death is likely within a few hours then there is little point in seeking emergency consent to force an end of life decision to cease assisted ventilation. It is far better to let nature take its course. Parents should not be made to feel that they alone carry the burden of decision-making. It is especially important that they are not burdened with contradictions by being advised on the one hand that death within a few days is inevitable, and on the other hand that even if the baby should survive it is likely that he or she will be severely disabled. The uncertainties of predicting disability were alluded to earlier. Conict with parents in end of life situations is more likely to occur when the issues are portrayed as a risk of long-term disability.4 Although the burden of continuing intensive care is a consideration it should not be focused on as a way of encouraging parents to reach a decision to withdraw intensive care. Their response may well be to question the standard of care being provided; they may, with some justication, argue that it is the job of doctors and nurses to relieve pain and discomfort associated with the care they are giving. 5. Clinical circumstances of end of life decisions in CLD There are two rather distinct circumstances when end of life decisions in infants with CLD are a consideration and they inuence the direction that the discussion takes with parents. Most infants will be receiving assisted ventilation and will have done so for many weeks together with parenteral nutrition, and a range of supportive drugs including pulmonary vasodilators, muscle relaxants, sedatives, and antibiotics. In spite of this they have deteriorated to the point where death within hours or days appears likely. It is assisted ventilation that is prolonging their life and it is extubation and detachment from the ventilator that is the primary end of life consideration. Parents are generally uncomfortable about asking how long it will take before their baby dies. It should be made clear that there is

a variable length of time before death occurs and the aim is to ensure that their baby is comfortable and free of pain and distress. They should be invited to contribute to decisions about the most appropriate way of ensuring this, and also they should be asked how they see their own role as parents during that time. For example, if circumstances allow they may wish to bath and dress their baby. A more difcult situation is when the infant is not receiving assisted ventilation but has previously had repeated episodes of ventilator dependency and the general course has been relentlessly downhill. Parents may well express the idea that their baby has gone through enough and if previous attempts to retrieve the situation by assisted ventilation have failed, why should the next one succeed? If there are medical grounds supporting this outlook, then, following informed discussion with the parents, it may be agreed that their babys further care will be redirected. Rather than prolonging a struggle for survival, further assisted ventilation will be withheld, and care will instead be directed towards ensuring a peaceful death acting within legal limits. 6. Compassionate aftercare Once an end of life decision has been made the focus should be on ways of relieving signs of distress, avoiding a lingering death, and attending to the needs of the parents.11 Privacy is important and a single room facilitates extended visiting for family members. It also provides an opportunity for parents to bath and dress their baby when the circumstances permit. There are choices in palliative care, and parents should be encouraged to express their needs because they are not always obvious. As well as religious practices there may be a range of personal requests that they found difcult to mention in advance of the end of life decision. They may wish to place in the crib beside their baby a photograph of one or more family members, or a family memento that has special signicance. Not all parents wish for an overtly close involvement with their dying baby and their views must be respected. They are not lacking in parental feelings if they decide that they simply want to go home and be called in when their baby dies. Their grief is no less. As well as attending to the parents needs, doctors and nurses have a duty of care which demands that they act within the law and within an ethical (moral) framework. In practice, this can raise a number of dilemmas. In research-based interviews with parents who had agreed to the withdrawal of intensive care for their baby, nearly one-quarter expressed concern about the length of the dying process.12 It would appear that a swift death conrmed the wisdom of the end of life decision whereas a protracted death raised doubts. This reinforces the need for parents to be properly prepared for what might happen after assisted ventilation is withdrawn, or after it is withheld. When care has been redirected to withhold further assisted ventilation, infants with endstage CLD are vulnerable to a protracted death which can be particularly distressing unless managed effectively it. A proportion of these infants are mature with respect to their post-conceptional age. Their nutritional support may have been carefully supervised so they may be relatively well-grown. Some have tachypnoea, intercostal retractions, and features of airways obstruction due to bronchomalacia. Hypoxia may result in agitation. Issues that may be raised by parents are the use of supplemental oxygen, especially when their baby appears cyanosed, and the provision of nutrition and uids. It is helpful in discussions with parents to focus on the net benet to the baby. If a treatment is of no net benet then it cannot be in the babys bests interests. Having reached the endstage of CLD with an agreement that further assisted ventilation will be of no benet then it is unlikely that a cyanosed infant would materially benet from supplemental

M. Chiswick / Seminars in Fetal & Neonatal Medicine 14 (2009) 396400

399

oxygen. It is unlikely that a dying infant at the endstage of CLD would feel any sensation of hunger, and so satiation is unnecessary, nor would nutritional support offer any other benet for the infant. Withholding uids is unlikely to lead to the infant having a conscious feeling of thirst, although one can never be certain. However, intravenous uids may well prolong dying and the net benet is in favour of withholding this treatment. Oral hygiene, including moistening of the mouth and lips, should be an integral part of the infants care. Infants who have already entered the process of dying before assisted ventilation was withdrawn present less of a dilemma because, with such a short anticipated lifespan, these treatments cannot benet the infant. 7. Use of drugs The judicious use of drugs to alleviate pain and distress in end of life situations is common in neonatal practice, and opiates are the most commonly prescribed drugs. It is well recognised that such drugs, as well as relieving pain and distress, may hasten death. The moral context is the doctrine of double effect, which argues that there is a distinction between intending an outcome and foreseeing an outcome. This has been adopted in case law such that it is within the law to administer a drug with the intention of relieving pain and distress even if it is foreseeable that the drug will shorten the infants life. However, the administration of a drug with the intention of shortening life, albeit as a way of relieving pain and distress, is illegal. In practice, the foreseeable effect of shortening life can easily become an intention, especially in infants who are experiencing a lingering death. Indeed, the results of condential surveys indicate that in some cases the shortening of life is indeed the intention.1315 Given that the practice is revealed by condential surveys it is likely that it is more widespread than is supposed. Some neonatologists have strong personal opinions about the use of drugs in end of life decisions which are often well thought out and based on a deep understanding of medical ethics. They may argue that when an end of life decision is made it can contravene the best interests of a distressed dying infant for doctors to be driven by the legalities of intention versus foreseeable outcomes. However, as long as euthanasia remains illegal, neonatologists should avoid heroically challenging the law. It is not only they who face the consequences, but it is the parents, and it is irrelevant whether or not they have given explicit consent. In the current climate of media interest, parents may have to cope not only with the grief of their baby dying but a public analysis of the circumstances of the death. Agitation and tachypnoea in the terminally ill infant with CLD can usually be controlled with opiates such as morphine, fentanyl or meperidine. It is important to commence at the recommended dose, and to titrate subsequent increases in dose according to the babys signs of distress. In some infants their tachypnoea and distress is replaced by terminal gasping, often referred to as agonal gasping. Gasping may also be observed in infants shortly after withdrawal of assisted ventilation. It may persist for many hours and often cannot be relieved by opiates. It can be particularly distressing for parents but it is very unlikely that the infant has a conscious sensation of distress. Critically ill adult patients who exhibit the same terminal gasping pattern are unconscious, and it is was established around 50 years ago that gasping was associated with electrical activity at brain stem level with no cortical activity.16 7.1. Muscle relaxants The use of muscle relaxants in neonatal palliative care is controversial. It is difcult to imagine how their use can be seen as anything other than an intention to promote death rather than for the relief of distress. Their specic use to alleviate terminal gasping

continues to be debated with some arguing that there is no certainty that affected infants have no sensation of distress.17,18 In its guidance, the Royal College of Paediatrics and Child Health indicates that when an end of life decision entails the withdrawal of assisted ventilation it would not be in an infants best interests to withdraw the paralysing drugs before discontinuing ventilation. This action would subject the child to a period of suboptimal ventilation leading up to withdrawal of ventilatory treatment. However, it would be unlawful to prescribe a paralysing agent prior to withdrawal of treatments simply to hasten death and to avoid subsequent terminal gasping (and so by implication it would be unlawful to introduce a muscle relaxant after withdrawal of ventilation). 8. Post-mortem examination The response of parents to discussions about a post-mortem examination for their infant can be inuenced by whether the death occurred in spite of all treatments, or whether it was the result of an end of life decision. Occasionally, parents will raise the issue of a post-mortem examination early on, during the discussion about the possibility of withdrawing or redirecting care. In other cases it is best not to raise the issue at this time, as it would mean the parents having to manage two burdens at the same time. The difculties about requesting a post-mortem examination following neonatal death and how they might be addressed have been widely discussed.1921 Parents of infants with CLD will have been kept informed of the nature of the condition and their babys progress. It was this that helped them decide that intensive care should be withdrawn or withheld. A subsequent request for a postmortem examination may be interpreted by some as uncertainty about the diagnosis and this can create anxiety and guilt. The chance nding at post-mortem of an associated unsuspected abnormality such as congenital heart disease can occasionally occur. It is very rare (but not impossible) to discover that a preterm infant who displayed the typical clinical and radiological features of CLD for many weeks had a primary underlying and undiagnosed pulmonary cause, revealed at post-mortem or by subsequent lung histology. It is important to give the parents the opportunity for their baby to have a post-mortem examination while stressing that the purpose is not to establish the cause of death. Most parents who consent to a post-mortem are usually relieved to have conrmation that the disease was so severe that survival was not possible. They can be reassured that having faced a most difcult parenting challenge they truly acted in the best interests of their baby.

Practice points  A minority of infants with CLD run a protracted course and remain critically ill, in spite of treatments.  Episodic deterioration may occur, leading to many episodes of assisted ventilation.  An end of life decision may be made when further intensive care is deemed futile and is prolonging death.  Staff may be reluctant to raise this issue, as the usual course of CLD is recovery after prolonged oxygen dependency.  Parents need to be part of decision-making, and to do this effectively they have to be regularly informed about their babys progress beforehand.  When a decision is made to redirect care and withhold further assisted ventilation, palliative care must include the relief of the infants distress.  It is legally acceptable to use drugs with the intention of relieving distress but not with the aim of shortening life.

400

M. Chiswick / Seminars in Fetal & Neonatal Medicine 14 (2009) 396400 9. Anjari M, Counsell SJ, Srinivasan L, et al. The association of lung disease with cerebral white matter abnormalities in preterm infants. Pediatrics 2009;124: 26876. 10. Chiswick M. Parents and end of life decisions in neonatal practice. Arch Dis Child Fetal Neonatal Ed 2001;85:F13. 11. Laing IA, Freer Y. Reorientation of care in the NICU. Semin Fetal Neonatal Med 2008;13:3059. 12. McHafe HE, Lyon AJ, Fowlie PW. Lingering death after treatment withdrawal in the neonatal intensive care unit. Arch Dis Child Fetal Neonatal Ed 2001;85: F812. 13. Provoost V, Cools F, Bilsen J. The use of drugs with a life-shortening effect in end-of-life care in neonates and infants. Intensive Care Med 2006;32:1339. 14. Verhagen AAE, Dorscheidt J, Engels B, Hubben JH, Sauer PJ. Analgesics, sedatives and neuromuscular blockers as part of end-of-life decisions in Dutch NICUs. Arch Dis Child Fetal Neonatal Ed, in press. 15. Provoost V, Cools F, Mortier F, et al. Medical end-of-life decisions in neonates and infants in Flanders. Lancet 2005;365:131520. 16. Gurvitch AM. Rhythmic bursts in the medullary reticular formation and their connection with agonal respiration during hypoxia and the post-hypoxic period. Electroencephalogr Clin Neurophysiol 1966;21:35564. 17. Perkin RM, Resnik DB. The agony of agonal respiration: is the last gasp necessary? J Med Ethics 2002;28:1649. 18. Hawryluck L. Neuromuscular blockers a means of palliation? J Med Ethics 2002;28:1702. 19. Chiswick M. Perinatal and infant post-mortem examination. Br Med J 1995;310: 1412. 20. Khong TY. Improving perinatal autopsy rates: who is counselling bereaved parents for autopsy consent? Birth 1997;24:557. 21. Lyon A. Perinatal autopsy remains the gold standard. Arch Dis Child Fetal Neonatal Ed 2004;89:F284.

Conict of interest statement None declared. Funding sources None. References


1. Northway W, Rosan R, Porter D. Pulmonary disease following respirator therapy of hyaline membrane disease: bronchopulmonary dysplasia. N Engl J Med 1967;276:35768. 2. American Academy of Pediatrics, Committee on Fetus and Newborn. Noninitiation or withdrawal of intensive care for high-risk newborns. Pediatrics 2007;119:4013. 3. Royal College of Paediatrics and Child Health. Withholding or withdrawing life sustaining treatment in children. A framework for practice. London: RCPCH; May 2004. 4. Verhagen AAE, de Vos M, Dorscheidt J, et al. Conicts about end-of-life decisions in NICUs in the Netherlands. Pediatrics 2009;124:e1129. 5. Bellieni CV, Buonocore G. Flaws in the assessment of the best interests of the newborn. Acta Paediatr 2009;98:6137. 6. Katz-Salamon M, Gerner EM, Jonsson B, Lagercrantz H. Early motor and mental development in very preterm infants with chronic lung disease. Arch Dis Child Fetal Neonatal Ed 2000;83:F16. 7. Anderson PJ, Doyle LW. Neurodevelopmental outcome of bronchopulmonary dysplasia. Semin Perinatol 2006;30:22732. 8. Short EJ, Kirchner HL, Asaad GR, et al. Developmental sequelae in preterm infants having a diagnosis of bronchopulmonary dysplasia: analysis using a severitybased classication system. Arch Pediatr Adolesc Med 2007;161:10827.

Seminars in Fetal & Neonatal Medicine 14 (2009) 401

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine


journal homepage: www.elsevier.com/locate/siny

Corrigendum

Corrigendum to: Normal and abnormal maternal metabolism during pregnancy [Seminars in Fetal & Neonatal Medicine 14 (2009) 6671]
David R. Hadden a, *, Ciara McLaughlin b
a b

Royal Victoria Hosptial, Belfast BT12 6BA, UK Regional Centre for Endocrinology & Diabetes, Royal Victoria Hospital, Belfast BT12 6BA, UK

It has been brought to the authors attention that there was an error in gure 4 of this published paper. The correct version of gure 4 is reproduced below.

Changes in plasma concentrations of glucose, alanine, free fatty acids and b-hydroxybutrate in non-pregnant and pregnant women between 12 hours fasting and 18 hours fasting during the third trimester. *Denotes signicant change from 12 hours. (Adapted from Metzger et al, 2003)5

DOI of original article: 10.1016/j.siny.2008.09.004. * Corresponding author. E-mail address: davidrhadden@btinternet.com (D.R. Hadden). 1744-165X/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.siny.2009.04.002

Seminars in Fetal & Neonatal Medicine 14 (2009) III

AUTHOR INDEX TO VOLUME 14


Adhikari M, 234 Almeida OFX, 130 Ambalavanan N, 21 Anderson PJ, 391 Appel IM, 311 Arvin AM, 209 Bancalari E, 35 Barker DJ, 119 Bassi L, 278 Baud O, 28 Benders MJNL, 272 Bose C, 374 Boynton BR, 345 Brian Smith P, 374 Broliden K, 218 Champagne DL, 136 Cheong JLY, 267 Chiswick M, 396 Chitayat L, 72 Claure N, 35 Cnossen MH, 311 Cornette L, 61, 126, 178, 241, 243, 329 Cowan FM, 267 Culhane JF, 200 Curtis N, 222 Davis PG, 14 de Heus R, 151 de Veber G, 243 De Vries LS, 272 de Vries WB, 171 Deming DD, 345 Denney JM, 200 Deprest JA, 8 Donn SM, 332, 367 Doyle LW, 391 Dudink J, 299, 323 Eriksson UJ, 85 Eventov-Friedman S, 164 Flemmer AW, 8 Fok T-F, 49 Fontaine R, 28 Forsgren M, 181 Fumagalli M, 278 Goldenberg RL, 182 Golomb MR, 318 Govaert P, 243, 250, 278, 284, 323 Gratacos E, 8 Greenough A, 339 Grigsby PL, 190 Groenendaal F, 272 Gupta S, 367 Hadden DR, 66, 401 Heijnen CJ, 127 Hod M, 63, 72, 94 Hummler H, 42 Hummler HD, 1 Jobe AH, 2, 157 Jols M, 136 Jovanovic L, 72 Kallapur S, 2 Kramer BW, 2 Lal M, 125 Laughon MM, 374 Leo P, 130 Lequin MH, 299 Loron G, 28 Lynch JK, 245 Malm G, 204 Maury L, 28 McClure EM, 182 McLaughlin C, 66, 401 Mercier J-C, 28 Merritt TA, 345 Mesquita AR, 130 Metzger BE, 94 Morley CJ, 14 Mosca F, 278 Mulder EJH, 151 Navr L, 181 Newnham J, 2 Newnham JP, 157 Nicolai T, 56 Nicolaides K, 8 Nizard J, 101

II

Author Index

Northway WH, 331 Novy MJ, 190 Obenaus A, 299 Oliveira M, 130 Olivier P, 28 Owen LS, 14 Patchev AV, 130 Persson B, 106 Pettersson K, 181 Philip AGS, 333 Posfay-Barbe KM, 228 Rademaker KJ, 171 Ramenghi LA, 278 Ronald de Kloet E, 136 Schelonka RL, 190 Schulze A, 1, 42 Shinwell ES, 164 Simeoni U, 63, 119 Sinha SK, 332, 367 Smith CK, 209 Smith L, 323 Sousa N, 130

Tebruegge M, 222 Thome UH, 21 Tin W, 383 Tolfvenstam T, 218 Tong KA, 299

van Bel F, 127 Van Marter LJ, 358 van Ommen CH, 311 Ville Y, 101 Visser GHA, 77, 151

Waites KB, 190 Wald ER, 228 Weaver ICG, 143 Wegerich Y, 130 Weindling AM, 111 Wiswell TE, 383

Xiao L, 190

Yogev Y, 77, 94

Seminars in Fetal & Neonatal Medicine 14 (2009) IIIV

KEYWORD INDEX TO VOLUME 14


Abnormal metabolism, 6671 Adaptation, 136142 Adaptive ventilation, 3541 Antenatal corticosteroid therapy, 151156 Antenatal corticosteroids, 157163 Anti-inflammatory agents, 4955 Antiviral therapy, 222227 Anxiety, 130135 Arterial oxygen saturation, 3541 Arterial spin labeling, 299310 Arterial, 284298 Artificial pancreas, 7276 Bacterial vaginosis, 200203 Behaviour, 391395 Betamethasone, 151156 Brain, 136142, 284298 Bronchodilators, 4955 Bronchopulmonary dysplasia, 27, 2127, 2834, 164170, 190199, 333338, 339344, 345357, 358366, 367373, 374382, 383390, 391395, 396400 Caffeine, 358366, 374382 Cardiovascular abnormalities, 8593 Cardiovascular effects, 171177 Cerebral infarction, 267271 Cerebral palsy, 164170, 318322 Chaos, 5660 Chorioamnionitis, 27, 182189, 190199 Chorioamniotitis, 345357 Chromatin, 143150 Chronic lung disease, 171177, 333338, 383390 Closed-loop oxygen control, 3541 Cognition, 391395 Cognitive impairment, 318322 Congenital abnormalities, 8593 Congenital diaphragmatic hernia, 813 Congenital infection, 190199, 218221 Congenital malformation, 101105 Congenital varicella syndrome, 209217 Continuous positive airway pressure, 1420, 358366 Corticosteroids, 164170, 374382 Coxsackievirus, 222227 Craniofacial abnormalities, 8593 Cystic adenomatoid malformation, 5660 Cystic periventricular leucomalacia, 2834 Cytokines, 339344 Depression, 130135 Development, 136142 Developmental programming, 119124 Dexamethasone, 151156, 164170, 171177 Diabetes mellitus, 7276, 101105, 111118, 119124 Diagnosis, 94100, 209217, 278283, 323328 Diffusion tensor imaging, 299310 Diffusion-weighted imaging, 299310 Disease, 136142 Diuretics, 4955 DNA methylation, 143150 Doppler velocity waveforms, 151156 Drug therapies, 383390 Early-onset sepsis, 228233 Echovirus, 222227 Encephalitis, 204208 Enterovirus, 222227 Epidemiology, 245249, 358366 Epilepsy, 318322 Erythrovirus, 218221 Ethics, 396400 EXIT (ex-utero intrapartum) procedure, 5660 Fetal heart rate, 151156 Fetal infection, 209217 Fetal inflammatory response, 345357 Fetal movement, 151156 Fetal tracheal occlusion, 813 Folic acid, 8593 Free radical scavengers, 8593 Free radicals, 8593 Genetic predisposition, 333338 Gestational diabetes, 7784, 94100 Glucocorticoid receptor, 143150 Glucocorticoid, 130135 Glucocorticoids, 339344 Glucoregulation, 106110 Glucose monitoring, 7276 Glyburide, 7276 Heart rate variability, 151156 Herpes zoster, 209217 Hippocampus, 143150 Human immunodeficiency virus, 234240 Human parvovirus B19, 218221 Hydranencephaly, 250266

IV

Keyword Index

Hydrocortisone, 171177 Hypercapnia, 2127 Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study, 94100 Hypertension, 119124 Hypocapnia, 2127 Infant of diabetic mother, 111118 Infant, 1420, 391395 Infant, newborn, 4248, 391395 Infarction (stroke), 250266 Infection, 182189, 339344 Inhaled nitric oxide, 2834, 358366 Insulin, 7276 Intermittent positive-pressure ventilation, 4248 Intracranial haemorrhage, 2834 Ischemic stroke, 245249 Laryngeal mask, 5660 Late-onset sepsis, 228233 Listeriosis, 228233 Lung and brain angiogenesis, 2834 Lung function, 391395 Lung injury, 27 Macrosomia, 101105 Magnetic resonance imaging, 272277 Magnetic resonance spectroscopy, 299310 Management, 5660 Maternal care, 143150 Maternal hyperglycemia/fetal hyperinsulinism theory, 106110 Maternal metabolism, 6671 Mechanical ventilation, 2127, 3541, 367373 Metabolic syndrome, 7784 Metformin, 7276 Morbitity, 200203 Mortality, 200203 Mucolytic agents, 4955 Multicystic encephalopathy, 250266 Multidrug resistance, 234240 Mycoplasma hominis, 190-199 Myelination, 157163 Neonatal arterial ischaemic stroke, 323328 Neonatal cerebral sinovenous thrombosis, 323328 Neonatal herpes, 204208 Neonatal hypoglycaemia, 106110 Neonatal intensive care, 396400 Neonatal stroke, 272277 Neonatal, 111118 Neonate upper airway, 5660 Neonate, 119124, 222227 Neonates, 228233 Neurodevelopment, 130135 Neurodevelopmental outcome, 171177 Neuroendocrinology, 136142 Neuroimaging, 272277, 278283 Neuroprotection, 2834

Newborn, 1420, 267271, 271276, 278283 Newborns, 4955 Non-invasive ventilation, 358366 Normal metabolism, 6671 Obesity, 7784 Offspring of diabetic mothers, 106110 Outcome, 204208, 245249 Oxygen therapy, 3541 Palliative care, 396400 Perfusion-weighted imaging, 299310 Perinatal arterial stroke, 272277 Perinatal infection, 228233 Perinatal stroke, 245249, 311317, 318322 Pharyngeal ventilation, 5660 Porencephaly, 250266 Positive-pressure respiration, 4248 Postnatal, 130135 Prediction of outcome, 813 Pregnancy in diabetics, 8593 Pregnancy, 6671, 7276, 7784, 101105, 111118, 119124, 200203, 209217, 218221, 234240 Premature infant, 2127 Prematurity, 2834, 333338, 339344, 367373 Prenatal diagnosis, 813, 218221 Prenatal, 130135, 250266 Preterm birth, 157163 Preterm infant, 3541 Preterm infants, 164170 Preterm labor, 190199 Preterm, 171177, 383390, 391395 Prevention, 374382 Programming, 136142 Prophylaxis, 374382 Prothrombotic coagulation factors, 311317 Respiration, artificial, 4248 Respiratory distress syndrome, 27, 1420, 157163, 345357, 367373, 383390 Risk factors, 245249, 267271, 311317 Screening, 94100 Seizures, 267271 Sensory impairment, 318322 Sinovenous thrombosis, 278283, 318322 Stillbirth, 182189 Stress, 130135, 136142 Stroke (or infarction), 284298 Stroke, 267271, 323328 Surfactant deficiency, 345357 Targeted minute ventilation, 3541 Templates, 284298 Treatment, 204208, 222227, 323328 Tuberculosis, 234240 Ultrasound examination, 101105 Ultrasound, 284298

Keyword Index

Ureaplasma, 27 Ureaplasma parvum, 190-199 Ureaplasma urealyticum, 190-199 Vaccination, 209217 Varicella, 209217

Vascular endothelial growth factor, 2834 Venous, 284298 Ventilator weaning, 4248 Vitamin A, 374382 WilsonMikity syndrome, 333338