PII: S0022-3468(18)30490-1
DOI: doi:10.1016/j.jpedsurg.2018.08.003
Reference: YJPSU 58768
To appear in: Journal of Pediatric Surgery
Received date: 20 January 2018
Revised date: 13 June 2018
Accepted date: 2 August 2018
Please cite this article as: Lukas Wisgrill, Anja Weinhandl, Lukas Unterasinger, Gabriele
Amann, Rudolf Oehler, Martin L. Metzelder, Angelika Berger, Thomas M. Benkoe ,
Interleukin-6 serum levels predict surgical intervention in infants with necrotizing
enterocolitis. Yjpsu (2018), doi:10.1016/j.jpedsurg.2018.08.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
ACCEPTED MANUSCRIPT
necrotizing enterocolitis
Lukas Wisgrill1, Anja Weinhandl2, Lukas Unterasinger1, Gabriele Amann3, Rudolf Oehler4,
PT
1
Department of Pediatrics and Adolescent Medicine, Division of Neonatology, Pediatric
RI
Intensive Care & Neuropediatrics, Medical University of Vienna, Vienna, Austria
2
SC
Department of Pediatric Surgery, Medical University of Vienna, Vienna, Austria
3
Institute of Clinical Pathology, Medical University of Vienna, Vienna, Austria
4
NU
Surgical Research Laboratories, Department of Surgery and Comprehensive Cancer Center,
Medical University of Vienna, Vienna, Austria
MA
Correspondence:
PT
eMail: thomas.benkoe@meduniwien.ac.at
Funding: This research did not receive any specific grant from funding agencies in the
1
ACCEPTED MANUSCRIPT
Abstract
specific and several biomarkers have been evaluated for their discriminative power to both
diagnose and predict the course from NEC suspicion to complicated disease requiring surgical
intervention. Thus, we aimed to assess the utility of interleukin-6 (IL-6) to predict surgical
PT
intervention in infants suffering from NEC and, furthermore, to discriminate infants with
RI
starting NEC or late-onset sepsis (LOS).
SC
Methods: IL-6 serum levels at disease onset were retrospectively analyzed in 24 infants
NEC necessitating surgical intervention in the disease course compared to infants with
MA
medical NEC (5000 [785-5000] vs. 370 [78-4716] pg/ml, p=0.0008) as well as gram-positive
LOS (5000 [785-5000] vs. 84 [12-269] pg/ml, p=0.0001). Infants suffering from gram-
ED
negative LOS exhibited elevated IL-6 serum levels at disease onset comparable to infants
with surgical NEC (5000 [1919-5000] vs. 5000 [785-5000] pg/ml, p =1.00).
PT
distinguish surgical NEC from medical NEC at the onset of disease but cannot discriminate
Level of Evidence: II
GN - gram-negative; GP - gram-positive
2
ACCEPTED MANUSCRIPT
1 Introduction
premature infants and is associated with high morbidity and mortality, leading to long-term
sequelae in surviving neonates [1, 2]. Despite different preventive approaches, the number of
PT
excessive inflammatory response of the immature intestinal epithelium of neonates to luminal
RI
bacteria, leading to exaggerated inflammatory damage of the epithelial gut barrier.
SC
Additionally, imbalance of intestinal microperfusion as well as intestinal ischemia might
provide further damage to the gut epithelium, leading to a “leaky gut” with increased bacterial
NU
translocation and, subsequently, a systemic inflammatory immune response [4, 6].
Plasma cytokine levels are very attractive as diagnostic tools in neonates because they are
MA
easily accessible and can be processed in a standardized fashion requiring only small blood
volumes. Recent data suggest that only few cytokines are significantly influenced by the
ED
presence of NEC, two of the most relevant pro-inflammatory cytokines being interleukin (IL)-
PT
8 and IL-6 [7-9]. Previous studies from our institution on IL-8 levels and NEC indicated a
significant correlation with disease extent, need for surgery, and mortality [10-12]. In contrast
CE
to IL-8, clinical data on IL-6 and NEC is sparse. Although there is direct evidence that the
onset of NEC results in significantly elevated IL-6 levels, no studies thus-far have
AC
investigated the potential value of IL-6 to discriminate NEC from LOS as well as to predict
We hypothesized that IL-6 levels are significantly elevated during the initial phase of NEC
and that the levels of IL-6 correlate with the subsequent need for surgical intervention.
Secondary aims of the study were to analyse IL-6 levels in LOS according to the presence of
gram-positive and gram-negative bacteria and to evaluate the ability of IL-6 levels to
3
ACCEPTED MANUSCRIPT
2 Methods
The study was conducted at the Division of Neonatology and the Department of Pediatric
Surgery at the Medical University of Vienna and was approved by the local ethics committee
(EC no. 1374/2017). Neonates with diagnosed NEC and blood-culture positive late-onset
sepsis were recruited during October 2015 and January 2017. Infants exhibiting clinical signs
PT
and radiographic findings of NEC were staged according to the modified Bell’s staging
RI
criteria [13]. Depending on the subsequent clinical course, infants with NEC were divided
SC
into two different treatment groups, medical and surgical NEC. Infants assigned to the
medical group had no surgical intervention during the disease course. Indications for surgery
NU
included evidence of intestinal perforation and/or clinical deterioration despite maximum
conservative treatment [14]. Decision for surgical intervention was dictated by the clinical
MA
judgement of the attending pediatric surgeon in coordination with the attending neonatologist
and was in no way influenced by this study. The diagnosis of NEC was confirmed intra-
ED
operatively in all patients. Macroscopic necrotic tissue was resected saving as much as
PT
possible viable tissue. The diagnosis of NEC was confirmed by histopathology in all patients
who received bowel resections. According to macroscopically evident affected bowel, disease
CE
extent was defined as focal NEC when it was limited to a well defined single intestinal
intestinal segments and panintestinal NEC when the majority of the small and/or large
intestines were involved. Infants with isolated intestinal perforation were excluded from the
study. Since 2015, IL-6 is part of the routine sepsis surveillance protocol at our department.
Blood samples were obtained from deteriorating infants at the onset of clinical signs of sepsis
or NEC. To compare IL-6 values at the time of NEC diagnosis with preoperative IL-6 levels,
we additionally evaluated preoperative IL-6 levels in the group of surgical NEC, which were
4
ACCEPTED MANUSCRIPT
was based on NeoKISS definitions: Culture-proven blood stream infections required the
recognition of a pathogen from one or more blood cultures, not related to another infection
site, and at least 2 of the following clinical signs: temperature >38°C or <36.5°C or
PT
staphylococci (CoNS) additionally required at least one of the following positive laboratory
RI
parameters of infection: C-reactive protein > 2 mg/dl or IL-6 > 50 pg/ml, immature/total
SC
neutrophil ratio > 0.2, white blood cell (WBC) count < 5000/µL and platelets <100/nl.
using the patient data management system ICCA of Phillips Healthcare. Blood culture results
and antibiotic resistance profiles were analysed using the monitoring of microorganism
ED
IL-6 serum concentrations were routinely tested using the IL-6 electrochemiluminescent
CE
assay, IL-6 values less than 50 pg/ml are considered normal for infants [16]. The assay was
AC
calibrated up to 5000 pg/ml. At this level the normal value was exceeded by more than 100-
fold and therefore not further diluted to determine the exact amount of IL-6.
Patient characteristics are presented as mean ± standard deviation or as frequency within the
study group if not stated otherwise. The Kruskal-Wallis-Test with post-hoc Dunn-Bonferroni
correction was used to test statistical significance. ROC analysis was performed with SPSS. A
5
ACCEPTED MANUSCRIPT
p-value <0.05 was considered statistically significant. Statistical analysis was performed
3 Results
PT
During the 16-month study period, a total of 24 neonates with NEC were included as well as
RI
16 control infants with diagnosis of culture-proven LOS. Clinical characteristics of the
SC
enrolled infants are summarized in table 1. Both groups exhibited similar gestational age and
age at diagnosis of infection. Infants in the sepsis group display significantly lower birth
NU
weights compared to infants with NEC (p = 0.044). In the sepsis group, gram-negative late
MA
onset sepsis (GN-LOS) was observed in 8 infants, and gram-positive LOS (GP-LOS) was
detected in 8 cases.
ED
In the NEC group, surgical treatment was performed in 12 infants (50%), and medical
treatment in 12 cases (50%). Of 24 infants with NEC, six exhibited Bell Stage I, six Bell stage
PT
II and 12 patients displayed Bell stage III. The median time from NEC diagnosis to surgery
was 24 [8 - 96] hours. Indications for surgery were radiographic signs of intestinal perforation
CE
Of the operated infants, none had focal NEC (0%), 9 infants had multifocal NEC (75%) and 3
infants had panintestinal NEC (25%). Intestinal perforations were present in 11 infants with
surgical NEC. Intestinal resections were done in 9 patients, 4 infants were treated with small
bowel and 5 infants with small and large bowel resections. Two infants presented with
characteristic morphologic changes of NEC (e.g. mosaic like pattern of affected bowel and
6
ACCEPTED MANUSCRIPT
segment. These infants were treated with a diverting enterostomy and the intestinal
perforation was sutured but not resected. One infant was treated with a diverting enterostomy
3.2 IL-6 is significantly higher in surgical NEC patients compared to medical NEC patients
PT
RI
Of all evaluated laboratory parameters, only serum concentrations of IL-6 were significantly
SC
higher in surgical NEC compared to cases with medical NEC (5000 [785-5000] vs. 370 [78-
4716], p = 0.008). In 8 of 12 cases with surgical NEC, IL-6 serum levels exceeded the assay
NU
range of 5000 pg/ml (Table 3/ Figure 1). The comparison of IL-6 levels at disease onset with
preoperative IL-6 values did not reach a statistical significance (p=0.11). IL-6 values tended
MA
to remain constant or slightly decrease between the time of NEC diagnosis and the time of
operation (Table 2). An explorative ROC analysis was performed to discriminate surgical
ED
NEC from medical NEC based on the IL-6 level at the time of NEC diagnosis. The AUC was
0.931 (CI 0.83 – 1.0), and an exploratory cut-off point could be established at 1440 pg/ml. At
PT
this value IL-6 reached a sensitivity of 91.7% and a specificity of 83.3% (Figure 2). IL-6
CE
levels were elevated according to the infants Bell´s stage. IL-6 levels showed a stage-
depending increase of IL-6 levels with higher levels in Bell stage III compared to lower levels
AC
in Bell stage II and I (5000 [785-5000] vs. 596 [152-4716] vs. 158 [78-2578] pg/ml,
p=0.001). In addition, we excluded Bell stage I cases in the medical NEC group and found no
difference in the AUC (surgical NEC vs. medical NEC without Bell stage I AUC 0.931; CI
Furthermore, we investigated if NEC infants with proven blood stream infection showed
different IL-6 levels when compared to NEC cases with negative blood culture results. In the
7
ACCEPTED MANUSCRIPT
surgical NEC group, 6 of 12 infants showed positive blood cultures (4 E. coli and 2 K.
E. faecalis). In the surgical group, positive blood cultures were not associated with altered IL-
6 serum levels (5000 [785-5000] vs. 3402 [1443-5000] pg/ml, p = 0.36). In contrast, medical
NEC infants with positive blood cultures showed a trend towards decreased IL-6 serum levels
PT
RI
3.3 IL-6 levels distinguish GN-LOS from GP-LOS and medical NEC, but not from surgical
SC
NEC
NU
IL-6 is an established early biomarker for neonatal sepsis. Therefore, we included infants with
aureus). Infants suffering GN-LOS showed significantly higher serum concentrations of IL-6
compared to neonates with medical NEC (5000 [1919-5000] vs. 370 [78-4716] pg/ml, p =
ED
0.014) and GP-LOS (5000 [1919-5000] vs. 84 [12-269] pg/ml, p = 0.0001). IL-6 serum levels
did not differ in patients with surgical NEC and GN-LOS (5000 [785-5000] vs. 5000 [1919-
PT
5000] pg/ml, p = 1.00). Interestingly, NEC infants treated medically displayed higher
CE
thrombocyte counts compared to the other study groups, reaching borderline statistical
significance compared to the GP-LOS group (260 [42-385] vs. 98 [60-211] x 103/µl, p =
AC
0.048). WBC and CRP levels did not differ among the study groups. IL-6 serum levels
showed no correlation with other laboratory values (data not shown). However, IL-6 serum
levels could not discriminate between surgical NEC and GN-LOS (AUC 0.458; 0.19 – 0.71).
4 Discussion
8
ACCEPTED MANUSCRIPT
The present study demonstrates that IL-6 plasma levels strongly correlate with disease
severity and the need for surgical intervention in neonates with NEC. At disease onset IL-6
plasma levels are significantly higher in neonates with surgical NEC compared to patients
with medically treated NEC. The observed IL-6 levels further correlate with advanced Bell
stages in infants suffering from NEC. The present data supports our hypothesis that elevated
IL-6 levels correspond with the disease severity of NEC. It further highlights the importance
PT
of IL-6 as a promising diagnostic marker in neonates with NEC.
RI
The cytokine response at the onset of NEC has received much scientific attention [17, 18].
SC
This has revealed an extensive dysregulation of gene expression in human NEC tissues with
recently, we were able to elucidate the role of IL-8 in NEC patients and demonstrate that IL-8
MA
levels predicted the extent of involved intestine [10], the need for surgical intervention, and
60-day mortality [12]. Routine IL-8 measurements were exchanged for IL-6 due to a
ED
therefore unable to provide information on both IL-6 and IL-8 levels in the present patient
PT
cohort. Based on our previous experience with IL-8 and the valuable information it provided
CE
in the risk assessment in infants with NEC, we found it promising to evaluate IL-6 as a
The diagnosis of NEC currently relies on clinical symptoms and radiological features such as
pneumatosis intestinalis and pneumoperitoneum, underlying the Bell´s stages of disease [13].
Initial clinical signs are often unspecific and mimic feeding intolerance or other forms of
benign gastroenteritis, which hampers early diagnosis [4]. However, the typical clinical
scenario of a premature infant who is at first developing subtle signs of feeding intolerance
and then progressing to abdominal distention and bloody stool, often with sepsis and
9
ACCEPTED MANUSCRIPT
few hours, presenting with pan-intestinal disease extent. It is important to note that early
clinical signs in the initial phase of both NEC and sepsis are overlapping and thus difficult to
discriminate. This early phase of the disease with unspecific clinical signs is often referred as
Despite all limitations of the Bell Staging criteria, infants with NEC Stage I were included in
PT
the study to determine the utility of IL-6 as an early diagnostic marker. The early recognition
RI
or exclusion of NEC, even in the stated Bell Stage I, might have important impacts on the
SC
prospective clinical management of those neonates (e.g. stop enteral feeding, radiographic
monitoring, change of antibiosis, early transfer into a tertiary care center). This circumstance
NU
mandates future research focused on implementing targeted therapy at an early stage.
Data on the predictive value of IL-6 in neonates with NEC are promising but sparse. The
MA
presented results are in line with a prior study on plasma concentrations of 6 different
cytokines that reported a significant increase of IL-6, IL-8, TNF and IL-1ß levels in surgical
ED
NEC compared to spontaneous intestinal perforation and controls [8]. Despite concomitant
up-regulation of IL-6 and IL-8, only the latter was able to discriminate between limited NEC
PT
and NEC totalis in that study. In contrast to our study, infants with medical NEC and culture-
CE
As a consequence of this diagnostic dilemma we further aimed to investigate the utility of IL-
AC
6 as a reliable marker to distinguish patients with NEC or LOS at disease onset. Harris and
colleagues [7] reported that IL-6 levels were 6- to 10-fold elevated in infants with NEC
compared to infants suffering from sepsis and healthy controls. These elevated IL-6 levels
were only observed in NEC infants with positive blood culture results. In their study IL-6
levels were not able to discriminate between sepsis and NEC. However, a further
Although we observed significantly higher IL-6 levels in gram-negative LOS compared with
10
ACCEPTED MANUSCRIPT
gram-positive LOS, IL-6 levels at disease onset were unable to discriminate between infants
bacteria, was identified as the top upstream transcriptional regulator in NEC. LPS induces IL-
1, IL-6, TNF, and IFN, which in turn activate inflammatory transcription factors [21]. The
PT
identification of LPS as the top upstream regulator is consistent with increasing evidence that
RI
infants who develop NEC display a gut microbial dysbiosis with overrepresentation of gram-
SC
negative bacteria prior to disease onset [22, 23]. This hypothesis is supported by clinical
bacteria in infants with sepsis and sepsis syndrome would be highly desirable. In adults,
higher values in gram-negative bacteremia [20]. In the context of adapted empirical antibiotic
schemes IL-6 values could provide valuable additional information at sepsis suspicion.
PT
However, larger studies are needed to assess the potential role of IL-6 to predict gram-
CE
It is of utmost importance to note that the dominant factor that defines the further course,
subsequent gastrointestinal morbidity, and outcome of infants with NEC is the extent of
affected intestine. The surgical option employed is strongly influenced by clinical judgement
and estimation of the involved intestinal tissue [25]. The evaluation of different surgical
procedures in neonates with NEC in the past three decades has clearly shown that the
modality of surgical treatment is not able to significantly influence the outcome [4]. Surgical
intervention is associated with a variety of inherent stresses that may aggravate the
11
ACCEPTED MANUSCRIPT
physiologic derangement observed in NEC [26]. One of the key questions that remains to be
answered is the timing of surgery. While the traditional concept to proceed to surgery in the
the clinical status of the patient despite optimal treatment, the additional information provided
by biomarkers such as IL6 might justify re-assessment of current surgical algorithms. It might
be prudent to undergo surgery earlier in patients with excessively high cytokine levels, at a
PT
time point when these fragile but otherwise stable infants might be able to tolerate surgery
RI
better than at a later more advanced disease stage.
SC
A limitation of our study is the upper detection limit of IL-6, which represents a 100-fold
increase of its normal value. The upper limit was reached in 8 out of 12 cases with surgical
NU
NEC and in 5 out of 8 cases with gram-negative LOS, which limits further evaluation. Since
the patients sera were not further diluted to measure the exact quantity, we are not able to
MA
assess potential differences in elevated IL-6 serum concentrations in those two groups. As a
further consequence of the excessively high IL-6 levels observed in surgical NEC we are not
ED
able to correlate IL-6 levels with intraoperative findings. As a consequence of those results,
samples for IL-6 measurement are now being diluted to measure the exact serum
PT
In conclusion, the proinflammatory cytokine IL-6 was able to predict the subsequent need for
AC
surgical intervention at the disease onset in our patient cohort. It is important to note that the
observed IL-6 levels remained highly elevated between NEC onset and operation. This further
underlines the diagnostic value of IL-6 to predict the need for surgical intervention at disease
onset. Whether an early surgical intervention in the presence of extremely high cytokine
levels, even in the absence of evidence for intestinal perforation, can improve the outcome of
this devastating disease needs to be tested in prospective treatment trials. Further studies are
12
ACCEPTED MANUSCRIPT
also needed to assess the association of IL-6 serum levels with the extent of affected intestine
Funding: This research did not receive any specific grant from funding agencies in the
PT
public, commercial, or not-for-profit sectors.
RI
SC
NU
MA
ED
PT
CE
AC
13
ACCEPTED MANUSCRIPT
References
[1] Hintz SR, Kendrick DE, Stoll BJ, Vohr BR, Fanaroff AA, Donovan EF, et al.
PT
[3] Patel RM, Myers LS, Kurundkar AR, Maheshwari A, Nusrat A, Lin PW. Probiotic
RI
bacteria induce maturation of intestinal claudin 3 expression and barrier function.
SC
Am J Pathol 2012;180(2):626-35.
[4] Neu J, Walker WA. Necrotizing enterocolitis. The New England journal of medicine
NU
2011;364(3):255-64.
MA
[6] Markel TA, Crisostomo PR, Wairiuko GM, Pitcher J, Tsai BM, Meldrum DR. Cytokines
[7] Harris MC, Costarino AT, Jr., Sullivan JS, Dulkerian S, McCawley L, Corcoran L, et al.
CE
Cytokine elevations in critically ill infants with sepsis and necrotizing enterocolitis. J
Pediatr 1994;124(1):105-11.
AC
[8] Bhatia AM, Stoll BJ, Cismowski MJ, Hamrick SE. Cytokine levels in the preterm infant
14
ACCEPTED MANUSCRIPT
[10] Benkoe T, Reck C, Gleiss A, Kettner S, Repa A, Horcher E, et al. Interleukin 8 correlates
[11] Benkoe TM, Mechtler TP, Weninger M, Pones M, Rebhandl W, Kasper DC. Serum
PT
enterocolitis in preterm infants. J Pediatr Surg 2014;49(10):1446-51.
[12] Benkoe T, Reck C, Pones M, Weninger M, Gleiss A, Stift A, et al. Interleukin-8 predicts
RI
60-day mortality in premature infants with necrotizing enterocolitis. J Pediatr Surg
SC
2014;49(3):385-9.
[13] Walsh MC, Kliegman RM. Necrotizing enterocolitis: treatment based on staging
NU
criteria. Pediatr Clin North Am 1986;33(1):179-201.
MA
[14] Henry MC, Moss RL. Current issues in the management of necrotizing enterocolitis.
[15] Adlassnig KP, Berger A, Koller W, Blacky A, Mandl H, Unterasinger L, et al. Healthcare-
associated infection surveillance and bedside alerts. Stud Health Technol Inform
PT
2014;198:71-8.
CE
neonates with early and late onset infection. The Pediatric infectious disease journal
AC
1997;16(4):370-5.
[17] Maheshwari A, Schelonka RL, Dimmitt RA, Carlo WA, Munoz-Hernandez B, Das A, et
15
ACCEPTED MANUSCRIPT
[18] Chan KY, Leung FW, Lam HS, Tam YH, To KF, Cheung HM, et al. Immunoregulatory
[19] Chan KY, Leung KT, Tam YH, Lam HS, Cheung HM, Ma TP, et al. Genome-wide
PT
perforation in human intestinal tissues: dysregulation of functional pathways. Ann
Surg 2014;260(6):1128-37.
RI
[20] Guo SY, Zhou Y, Hu QF, Yao J, Wang H. Procalcitonin is a marker of gram-negative
SC
bacteremia in patients with sepsis. Am J Med Sci 2015;349(6):499-504.
[21] MohanKumar K, Namachivayam K, Cheng F, Jiang RH, Flores-Torres J, Torres BA, et al.
NU
Trinitrobenzene sulfonic acid-induced intestinal injury in neonatal mice activates
MA
[22] Warner BB, Deych E, Zhou Y, Hall-Moore C, Weinstock GM, Sodergren E, et al. Gut
[23] Wang Y, Hoenig JD, Malin KJ, Qamar S, Petrof EO, Sun J, et al. 16S rRNA gene-based
analysis of fecal microbiota from preterm infants with and without necrotizing
AC
[24] Bury RG, Tudehope D. Enteral antibiotics for preventing necrotizing enterocolitis in
[25] Hall NJ, Eaton S, Pierro A. Royal Australasia of Surgeons Guest Lecture. Necrotizing
67.
16
ACCEPTED MANUSCRIPT
[26] Raval MV, Moss RL. Current concepts in the surgical approach to necrotizing
PT
RI
SC
NU
MA
ED
PT
CE
AC
17
ACCEPTED MANUSCRIPT
Figure Legends
Figure 1 Laboratory parameters in neonates among the study population. White blood cell
count (WBC; A), platelet count (B) as well as serum concentrations of C-reactive protein
(CRP; D) and interleukin 6 (IL-6; E) in surgical NEC (n=12), medical NEC (n=12), gram
negative (GN) sepsis (n=8) and gram positive (GP) sepsis (n=8) patients. IL-6 serum levels in
PT
infants suffering NEC stage I (n=6), stage II (n=6) and stage III (n=12; C). Boxes display the
25th and 75th percentiles and error bars indicate the 5th and the 95th percentiles. Median values
RI
are represented by the box middle line. * = p<0.05; n.s. = not significant
SC
Figure 2 Receiver operating characteristics curve analysis of IL-6 serum levels in infants with
NU
surgical NEC (sNEC) compared to medical NEC (mNEC- solid line), sNEC compared to late
onset sepsis (LOS - large dotted line) and sNEC compared to gram-negative sepsis (GN -
MA
small dotted line) sepsis. The calculated exploratory IL-6 cut-off value between sNEC and
mNEC was 1440 pg/ml with a sensitivity of 91.7% and a specificity of 83.3%. Data are
ED
presented as area under the curve (AUC) with 5 – 95% confidence interval.
PT
CE
AC
18
ACCEPTED MANUSCRIPT
PT
Birth weight, g (median 750 [630 - 902 [590 - 692 [585 – 665 [565 –
[min-max]) 3160] 2110] 1090] 915]
RI
Age at Onset, d (median 14.5 [4 - 28] 13.5 [3 - 34] 15 [5 – 25] 11
SC
[min-max]) [4 – 26]
Mechanical Ventilation, 12/12 (100%) 4/12 (33%) 6/8 (75%) 0/8 (0%)
NU
n (%)
Use of inotropes, n (%) 7/12 (58%) 3/12 (25%) 5/8 (63%) 0/8 (0%)
Twins, n (%) 1/12 (8.3%) 3/12 (25%) 2/8 (25%) 3/8 (37.5%)
MA
Steroids, n (%) 8/12 (66.7%) 8/12 (66.7%) 5/8 (62.5%) 5/8 (62.5%)
IVH, n (%) 3/12 (25%) 1/12 (8.3%) 1/8 (12.5%) 1/8 (12.5%)
ED
BPD, n (%) 2/12 (16.7%) 3/12 (25%) 3/8 (37.5%) 3/8 (37.5%)
ROP, n (%) 7/12 (58.3%) 3/12 (25%) 3/8 (37.5%) 4/8 (50%)
PT
PDA, n (%) 7/12 (58.3%) 6/12 (50%) 8/8 (100%) 7/8 (87.5%)
Medical closure of PDA, 7/12 (58.3%) 6/12 (50%) 8/8 (100%) 7/8 (87.5%)
CE
n (%)
Ligation of PDA, n (%) 1/12 (8.3%) 1/12 (8.3%) 1/8 (12.5%) 0/8 (0%)
AC
60d mortality, n (%) 4/12 (33.3%) 0/12 (0%) 2/8 (25%) 0/8 (0%)
NEC, necrotizing enterocolitis; GN, gram-negative; GP, gram-positive; LOS, late-onset
19
ACCEPTED MANUSCRIPT
Table 2. Demographic and laboratory features of infants with surgically and medically
treated NEC
Gestational age, wks 25.9 [23.1 - 38.4] 27.3 [24.1 – 33.4] 0.838
PT
Birth weight, g 750 [630 - 3160] 902 [590 - 2110] 0.963
RI
NEC Bell stage 12/12 (100%) III 6/12 (50%) I n.a.
SC
6/12 (50%) II
Colon: 0
Both: 5
CE
None: 3
AC
20
ACCEPTED MANUSCRIPT
PT
4711]
CRP (mg/dl) 6.3 [0.2-19.2] 1.7 [0.2-6.5] 3.4 [0.2-6.6] 3.6 [1.0-4.4]
RI
IL-6 (pg/ml) 5000 [785- 370 [78-4716] 5000 [1919- 84 [12-269]
SC
5000]) 5000]
Positive blood 6/12 (50%) 3/12 (25%) 8/8 (100%) 8/8 (100%)
NU
cultures, n (%)
Data shown as median [min-max]; NEC, necrotizing enterocolitis; GN, gram-negative; GP,
MA
gram-positive; LOS, late-onset sepsis; WBC, white blood count; CRP, C-reactive protein; IL,
interleukin
ED
PT
CE
AC
21
Figure 1
Figure 2