Indian J Pediatr (December 2011) 78(12):14981502 DOI 10.

1007/s12098-011-0460-7

ORIGINAL ARTICLE

The Role of Hemoglobin Variant Replacement in Retinopathy of Prematurity

Wojciech Podraza & Hanna Podraza & Karolina Jezierska & Joanna Szwed & Monika Modrzejewska & Jacek Rudnicki & Agnieszka Kordek & Hanna Domek

Received: 10 November 2010 / Accepted: 3 May 2011 / Published online: 1 June 2011 # Dr. K C Chaudhuri Foundation 2011

Abstract Objective To conduct tests of relationships between different factors that could influence the course of retinopathy of prematurity (ROP) and ROP, particularly the role of hemoglobin variant replacement in adult blood transfusions. Methods A retrospective, observational study of 83 infants born between 23 and 34 wks gestation was conducted. Results The infants without ROP, with 1 and 2 stage of ROP and with 3 stage of ROP received Q 28 (12134); 51 (14 149); 156 (38244) ml/kg of transfused blood, respectively, and the factor Qt was 1,545 (56010,045); 3,093 (614 13,419); 11,907 (1,28820,638) (ml/kg)day, respectively. For the same groups MCV35 (mean cell volume at the arbitrary time of the 35th wk post-conception) was 92.3 (82.9110.5); 91.0 (79.3101.4); 87.1 (80.294.8) fl, respectively, and factor PMCV/t was 99.5 (89.2108.8); 96.3 (84.6106.3); 90.7 (85.396.5) fl, respectively. There is high influence on the stage of ROP of the amount of transfused blood and MCV, both with or without the time factor. The statistical
W. Podraza (*) : H. Podraza : K. Jezierska : J. Szwed : H. Domek Department of Medical Physics, Pomeranian Medical University, ul. Ku Socu 12, 71-073, Szczecin, Poland e-mail: podrazaw@ams.edu.pl M. Modrzejewska Clinic of Ophthalmology with Department of Pathophysiology of Organ of Vision, Pomeranian Medical University, Szczecin, Poland J. Rudnicki : A. Kordek Department of Neonatology, Pomeranian Medical University, Szczecin, Poland

differences between PMCV/t were more significant than the differences between MCV35, for different stages of ROP. Conclusions The influence of the time factor on the statistical differences of MCV but not on the amount of transfused adult blood suggests that HbF HbA replacement may play a role in ROP development. Keywords Retinopathy of prematurity . Risk factors . Transfusions . Hemoglobin F

Introduction Despite great improvements in neonatal care, retinopathy of prematurity (ROP) is still the main ophthalmologic problem in that period of life and obviously in later life. The most severe stages of ROP may cause blindness. There are over 50,000 children blind from ROP worldwide. From a global perspective, three groups of countries can be selected with different levels of ROP problems. The prevalence of ROP blindness correlates with infant mortality rates (IMRs). There are low rates of ROP blindness in countries with high (>60/1,000) and low (< 9/1,000) IMRs, for different reasons: in the former, no infants with severe ROP survive, and in the latter, there is good detection and management of ROP. The group of countries with IMRs between 9 and 60/ 1,000 live births have the highest rate of ROP blindness. Latin America with the Caribbean Islands and the countries in the Former Socialist Economies (the authors region) belong to that group [1]. The problem of ROP is not exclusive to the last group of countries. The number of children who are blind from ROP is 15,000 and 4,500 in the Former Socialist Economies and Established Market Economies, respectively [1]. Children

Indian J Pediatr (December 2011) 78(12):14981502

1499

with the progressive ROP necessitate cryotherapy or laser photocoagulation. There are papers which report either increase or decrease in the number of infants treated for ROP in the last years [2, 3]. There are undeveloped control mechanisms, including those responsible for retina development in a preterm infant. The control system is replaced by the doctors treatment, but even if the treatment is good and the clinician is experienced, this system is only a rough imitation of the natural system. ROP is a two-phase disease. Phase I is characterised by delayed retinal growth because of a diminished level of insulin-like growth factor (IGF-1) in the prematurely born infant. As a result of phase I, there is insufficient vascularisation of the developing retina. The lack of vessels creates hypoxia, which stimulates vascular-endothelial growth factor (VEGF) production in excess and a slow increase in IGF-1 levels. Cytokines are responsible for the destructive neovascularisation of the retina (phase II ROP) [47]. As mentioned earlier, oxygen-regulated factors, such as VEGF, can be influenced by a doctors treatment (e.g., adequate or inadequate oxygen therapy). There are four factors clearly associated with ROP: prematurity, oxygen use, male sex, and white race; there are other factors that have been postulated, but are not as clear: hypoxemia, indomethacin therapy, vitamin E and A deficiency, inositol, patent ductus arteriosus (PDA), light exposure, Candida sepsis, numerous blood transfusions, early use of erythropoietin (EPO), hypercapnia [8], postnatal systemic steroids [9], multiple births [10], small-forgestational age [11]. The aim of the present study was to find statistical relevance for different features of preterm infants between three groups (I, II, III) created according to the severity of the disease: I - ROP 0, II - ROP 1 or 2 and III - ROP3 stage and to evaluate the potential role of not only blood transfusions, but also that of hemoglobin variant replacement. It has been shown that after five transfusions in infants, the hemoglobin F (HbF) level declines to below 15% of the total hemoglobin, compared with 87.15.1% before transfusion [12]. This change affects the hemoglobin oxygen dissociation curve and results in increased oxygen toxicity during this period of life [13, 14]. The authors have tested the statistical dependences between gestational age, birth weight, total amount of transfused blood per kg of body weight (with and without the time factor) and mean cell volume (MCV) of erythrocytes (with and without the time factor) for all three groups. MCV depends on and indirectly indicates the HbF content [12]. The authors have tested relationship between severity of ROP and gender too.

Material and Methods Patients Eighty three out of 100 preterm infants born between January 2005 and December 2008 in the Department of Neonatology, Pomeranian Medical University, Szczecin were included. All infants were born and hospitalised at a level III referral neonatal intensive care unit. The exclusion of 17 patients was due to the lack of complete blood count (CBC) at 351 post-conceptional wks i.e. lack of MCV35. The treatment was performed according to routine procedures including respiratory support, oxygen therapy and red blood cells (RBC) transfusions. The ophthalmologic examinations were conducted by an experienced ophthalmologist with indirect ophthalmoscopy several times until the retinas were fully vascularised. The findings were classified according to international criteria [15]. Methods The quantity of transfused blood Q was calculated as the sum of the quotients of the volume of the transfused blood P and body weight q. Q qk (k is the number of following transfusions). In almost all instances, it was 15 ml/kg body weight multiplied by the number of transfusions. The new attribute Qt was created to take into consideration the amount of transfused blood and the time factor. Qt is defined as the sum of qk multiplied by time tm remaining until the arbitrary time established as the 42nd wk post-conception (294 days). Time t was defined as follows: t=294 GA . 7 age (in days) in the day of transfusion. MCV35 is the mean cell volume at the arbitrary time of the 35th wk post-conception (1 wk) and can be used to indirectly determine the HbF content. This value depends on the HbF and HbA exchange, because of adult blood transfusions. The new attribute PMCV/t has been created to take into consideration the time factor. The relationship between MCV and time has been plotted for all patients. PMCV/t is the area under the curve divided by a newborns age (in days) in the day of the last MCV measurement (Fig. 1). The formula is below:
PMCV MCV1 t1 X1 2 MCVm MCVm1 tm1 tm

where m+1 is the number of MCV measurements. Figure 1. Shows the relationship between MCV and time. PMCV is the sum of the rectangle P1 MCV1 t1 and the P trapeziums P2 P3 P4 MCVm MCVm1 =2 tm1 tm , where m+1 is the number of MCV measurements. The

1500

Indian J Pediatr (December 2011) 78(12):14981502

whereas, PMCV/t is lower for infant in whom transfusions were applied earlier. The higher power of statistical difference between groups for P MCV /t compared to MCV35 shows not only the importance of the amount of transfused blood but also the importance of the action period of transfused blood. It may be the proof that changed blood hemoglobin plays a role in ROP development. The research study has been approved by local ethics committee. Statistical Analysis Relevant numerical data were presented as median and (range). Statistical analyses were performed using the Kruskal-Wallis test (Statistica, StatSoft, Inc.) A p values less than 0.05 were considered significant. Chi - square test was used to test the relationship between severity of ROP and gender. The differences were considered significant when p<0.05.

Fig. 1 The graph plots the relationship between MCV and time. PMCV is the sum of the rectangle P1 MCV1 t1 and the trapeziums P P2 P3 P4 1MCVm MCVm1 tm1 tm 2

graph relationship between MCV and time has been plotted for all patients. If the first complete blood count (CBC) was tested in time t0 (i.e., up to 12 h of life), the equation is: PMCV P1 2 MCVm MCVm1 tm1 tm MCV of adults erythrocyte is lower compared to newborns one. PMCV/t depends on time and number of adult RBC transfusions and corresponds with total HbA impact on newborns body because of HbF HbA replacement. Diminishing of MCV35 reflects also the amount of transfused blood, but doesnt indicate the time (age of the infant) of transfusion. Early in the life transfusions have longer impact on oxygen exchange and possible ROP development. PMCV/t reflects either the amount or the time of transfusions. For the similar two infants and the same amount of transfused blood, MCV35 should be similar

Results Eighty three preterm infants born at 29 (range 2334) wks of gestational age (GA) with a birth weight of 1,080 g (range 5702,145) were included. All calculations were also performed for the subgroup of patients born up to 30 wks of GA. This subgroup included 57 patients with an birth weight of 980 g (range 5701,550). The diagnosis of no ROP and ROP stages 1, 2, 3, 4 and 5 were made for 26, 23, 18, 12, 3 and 1 patients, respectively. All infants with progressive disease underwent laser photocoagulation.

Table 1 Selected clinical and laboratory data [median (range)where appropriate] Q* Quantity of transfused blood/body weight; Qt** Attribute, where t is the time remaining from the day of transfusion to the 294th day post conception; MCV35*** Mean cell volume in 35 (1) wk of GA; PMCV/t****- Area under the curve PMCV =f(t) divided by the newborns age in the day of the last MCV measurement; ROP Retinopathy of prematurity; cs/nd - Cesarean section/ natural delivery Gestational age (wks) GA Birth weight (g) BW Sex (female/male) Delivery method (cs/nd) Apgar score at 1st min. APG1 ROP (0/1/2/3/4/5) Number of transfusions Q* (ml/kg) Qt** ((ml/kg) day) MCV35*** (fl) PMCV/t **** (fl)

All patients (n=83) 29 (2334) 1,080 (5702,145) 42/41 66/17 6.0 (110) 26/23/18/12/3/1 4.8 (116) 51.3 (12244) 2,812 (56020,638) 90.4 (79.3110.5) 96.0 (84.6108.8)

30 wk GA patients (n=57) 28 (2330) 980 (5701,550) 29/28 44/13 5.0 (110) 12/14/16/11/3/1 6.0 (116) 80 (15244) 6,331 (73220,638) 88.6 (79.397.9) 94.3 (84.6106.0)

Indian J Pediatr (December 2011) 78(12):14981502 Table 2 p values for different features p values, all patients I (n=26) vs II (n=41) 0.173 0.129 0.051 0.256 1.00 0.274 0.415 p values, 30 wk GA patients I (n=12) vs II (n=30) 0.467 0.335 0.361 0.319 1.00 0.824 0.362

1501

Q* Qt** GA BW APG1 MCV35*** PMCV/t****

I (n=26) vs III (n=16) <0.0001 <0.0001 <0.0001 <0.0001 0.103 0.00114 <0.0001 I (n=12) vs III (n=15) <0.0001 <0.0001 <0.0001 0.00071 0.698 0.110 0.00074

II (n=41) vs III (n=16) <0.0001 0.00014 0.00142 0.00229 0.0190 0.0497 0.00097 II (n=30) vs III (n=15) 0.00017 0.00014 0.00087 0.017 0.182 0.505 0.015

Q* Qt** GA BW APG1 MCV35*** PMCV/t****

I - ROP 0, II - ROP 1 or 2 and III - ROP3 stage Q* - Quantity of transfused blood/body weight; Qt** - Attribute, where t is the time remaining from the day of transfusion to the 294th day post conception; MCV35*** - Mean cell volume in 35 (1) wk of GA; PMCV/t**** - Area under the curve PMCV =f(t) divided by the newborns age in the day of the last MCV measurement; GA - Gestational age; BW - Body weight; APG1 - Apgar score at 1st min of life

The severity of ROP did not depend on gender. The Chi- square test was not significant (pI,II =0.37, pI,III = 0.63, pII,III =0.80). The selected clinical and laboratory data and the tested relations are presented in Tables 1, 2 and 3.
Table 3 The amount of transfused blood and MCV with and without the time factor for three groups (I, II, III) of patients created according to the severity of disease

Discussion ROP is a multifactorial disease. The detection and evaluation of the risk factors of ROP are important from both a scientific and a practical point of view. The ROP risk

Q* (ml/kg) Qt** [(ml/kg)day] MCV35*** (fl) PMCV/t**** (fl)

median range median range median range median range

All patients (n=83) I (n=26) 28 12134 1,545 56010,045 92.3

II (n=41) 51 14149 3,093 61413,419 91.0

III (n=16) 156 38244 11,907 1,28820,638 87.1 80.294.8 90.7 85.396.5 III (n=15) 158 38244 12,137 4,38720,638 87.1 80.294.8 90.4 85.396.5

I - ROP 0, II - ROP 1 or 2 and III - ROP3 stage Q* - Quantity of transfused blood/ body weight; Qt** - Attribute, where t is the time remaining from the day of transfusion to the 294th day post conception; MCV35*** - Mean cell volume in 35 (1) wk of GA; PMCV/t****- Area under the curve PMCV = f(t) divided by the newborns age in the day of the last MCV measurement; ROP Retinopathy of prematurity Q* (ml/kg) Qt** [(ml/kg)day] MCV35*** (fl) PMCV/t**** (fl)

median range median range median range median range

82.9110.5 79.3101.4 99.5 96.3 89.2108.8 84.6106.3 30 wk GA patients (n=57) I (n=12) II (n=30) 31 72 15134 16149 2,093 4,845 73210,045 87013,419 89.9 88.6 82.996.8 79.397.9 99.5 94.6 89.2105.3 84.6106.0

1502

Indian J Pediatr (December 2011) 78(12):14981502 Conflict of Interest None. Role of Funding Source None.

factors have been previously documented. Sola et al. [8] emphasised such factors as prematurity, oxygen use, male sex, and white race; however, in the largest cohort of newborns analysed for ROP out of 4.67 million live births, the multivariate regression model revealed that only respiratory distress and intraventricular haemorrhage were predictive of the development of ROP [16]. Technically, one cannot count prematurity, in itself, as a risk factor, because ROP is exclusively restricted to that period of life. However, the degree of prematurity, expressed as gestational age and/or birth weight, can be and is treated as a ROP risk factor. The multiple transfusions of red blood cells (RBC), so called small volume transfusions are routine life saving procedures in clinical practice [17, 18]. Small volume transfusion is a confusing term. A routine dosage of RBC 15 ml/kg body mass (i.e., 15 ml of single volume), which seems small, when transfused to a preterm infant weighting 1,000 g, is exactly the same concentration as a 1.050 l of blood transfused to an adult patient weighting 70 kg. The infants are transfused with adult blood, which contains almost 100% hemoglobin A (HbA). The physiological content of hemoglobin F (HbF) of over 90% in that period of life is replaced by HbA, resulting in a dissociation curve shift, which can be responsible for oxygen toxicity [7, 1214]. The number of transfusions received by an infant has been suspected as a risk factor of ROP [8, 19]. There is a general recommendation to introduce more restrictions on the indications for blood transfusion [18]. On the other hand, there are no unique neonatal transfusion guidelines, and every centre works with its own guidelines [20]. In the authors department, among comparable groups of preterm infants, the average number of transfusions was 5.35 in the years 20002004 [21], but this has increased to 6.00 in the years 20052008. The authors examined the role of HbF in the development of ROP. There are similar high differences between groups I and III or II and III for amount of transfused blood with and without the time factor. A meaningful increase in the differences for MCV between I and III or II and III groups (as an indirect indicator of HbF content in the blood) was observed when the time factor was taken into account (Table 2). The influence of the time factor on the differences between MCV but not on the amount of transfused adult blood, suggests that HbF HbA replacement may play a role in ROP development. The present study has certain limitations: the relatively small sample size of the study group and the fact that relationships do not always indicate cause-and-effect phenomenon. Further studies based on direct evaluation of HbF content may clarify the above thesis.

References
1. Gilbert C. Retinopathy of prematurity: a global perspective of the epidemics, population of babies at risk and implications for control. Early Hum Dev. 2008;84:7782. 2. Todd DA, Wright A, Smith J. Severe retinopathy of prematurity in infants <30 weeks gestation in New South Wales and the Australian Capital Territory from 1992 to 2002. Arch Dis Child Fetal Neonatal Ed. 2007;92:F2514. 3. Dhaliwal CA, Fleck BW, Wright E, Graham C, McIntosh N. Incidence of retinopathy of prematurity in Lothian, Scotland, from 1990 to 2004. Arch Dis Child Fetal Neonatal Ed. 2008;93:F4226. 4. Smith LEH. IGF-1 and retinopathy of prematurity in the preterm infant. Biol Neonate. 2005;88:23744. 5. Fleck BW, McIntosh N. Pathogenesis of retinopathy of prematurity and possible preventive strategies. Early Hum Dev. 2008;84:838. 6. Kvanta A. Ocular angiogenesis: the role of growth factors. Acta Ophthalmol Scand. 2006;84:2828. 7. Saugstad OD. Oxidative stress in the newborna 30-year perspective. Biol Neonate. 2005;88:22836. 8. Sola A, Chow L, Rogido M. Retinopathy of prematurity and oxygen therapy: a changing relationship. An Pediatr (Barc). 2005;62:4861. 9. Smolkin T, Steinberg M, Sujov P, Mezer E, Tamir A, Makhoul IR. Late postnatal systemic steroids predispose to retinopathy of prematurity in very-low-birth-weight infants: a comparative study. Acta Paediatr. 2008;97:3226. 10. Nodgaard H, Andreasen H, Hansen H, Sorensen HT. Risk factors associated with retinopathy of prematurity (ROP) in Northern Jutland, Denmark 19901993. Acta Ophthalmol Scand. 1996;74:30610. 11. Dhaliwal CA, Fleck BW, Wright E, Graham C, McIntosh N. Retinopathy of prematurity in small-for-gestational age infants compared with those of appropriate size for gestational age. Arch Dis Child Fetal Neonatal Ed. 2009;94:F1935. 12. Barkemeyer BM, Hempe JM. Effects of transfusion on hemoglobin variants in preterm infants. J Perinatol. 2000;20:3558. 13. De Halleux V, Truttmann A, Gangon C, Bard H. The effects of blood transfusion on the hemoglobin oxygen dissociation curve of very early preterm infants during the first week of life. Semin Perinatol. 2002;26:4115. 14. Bain A, Blackburn S. Issues in transfusing preterm infants in the NICU. J Perinat Neonat Nurs. 2004;18:17082. 15. The Committee for the Classification of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. Arch Ophthalmol. 1984;102:11304. 16. Lad EM, Nguyen TC, Morton JM, Moshfeghi DM. Retinopathy of prematurity in the United States. Br J Ophthalmol. 2008;92:3205. 17. Luban NLC. Neonatal red blood cell transfusions. Vox Sanguinis. 2004;87:1848. 18. Fabres J, Wehrli G, Marques MB, et al. Estimating blood needs for very-low-birth-weight infants. Transfusion. 2006;46:191520. 19. Dani C, Reali MF, Bertini G, Martelli E, Pezzati M, Rubatelli FF. The role of blood transfusions and iron intake on retinopathy of prematurity. Early Hum Dev. 2001;62:5763. 20. Strauss RG. Data-driven blood banking practices for neonatal RBC transfusions. Transfusion. 2000;40:152840. 21. Podraza W, Nowak J, Domek H, et al. Neonatal RBC transfusions comparison of two patterns. Transfus Med Hemother. 2006;33:5159.