May 2013
Volume 54, Issue 5
Free
Retina  |   May 2013
The Relationship Between Cord Blood Cytokine Levels and Perinatal Factors and Retinopathy of Prematurity: A Gestational Age-Matched Case-Control Study
Author Affiliations & Notes
  • Se Joon Woo
    Department of Ophthalmology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Kyo Hoon Park
    Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Sung Youn Lee
    Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Seong Joon Ahn
    Department of Ophthalmology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Jeeyun Ahn
    Department of Ophthalmology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
    Department of Ophthalmology, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul, Korea
  • Kyu Hyung Park
    Department of Ophthalmology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Kyung Joon Oh
    Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Aeli Ryu
    Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Korea
  • Correspondence: Kyo Hoon Park, Department of Obstetrics and Gynecology, Seoul National University Bundang Hospital, 300 Gumi-dong, Bundang-gu, Seongnam-si, Gyeonggi-do, 463-707, Korea; pkh0419@snubh.org
Investigative Ophthalmology & Visual Science May 2013, Vol.54, 3434-3439. doi:10.1167/iovs.13-11837
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Se Joon Woo, Kyo Hoon Park, Sung Youn Lee, Seong Joon Ahn, Jeeyun Ahn, Kyu Hyung Park, Kyung Joon Oh, Aeli Ryu; The Relationship Between Cord Blood Cytokine Levels and Perinatal Factors and Retinopathy of Prematurity: A Gestational Age-Matched Case-Control Study. Invest. Ophthalmol. Vis. Sci. 2013;54(5):3434-3439. doi: 10.1167/iovs.13-11837.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To investigate the relationship between cytokine levels in cord blood and perinatal factors and retinopathy of prematurity (ROP) in gestational age-matched, preterm, newborn infants.

Methods.: Each of 20 premature singleton infants with ROP (gestational age < 32 weeks) was matched for gestational age, birth weight, and sex with two control infants without ROP. The concentration of 10 cytokines in cord blood extracted at birth was measured using a multiplex bead array assay. Data on maternal factors, labor and delivery characteristics, and neonatal parameters were also collected from both groups. The variables obtained were compared using the conditional logistic regression model.

Results.: No differences in the levels of inflammatory cytokines (IL-1β, IL-4, IL-6, IL-8, IL-10, IL-12, interferon-γ, and TNF-α) and growth factors (insulin-like growth factor-1 and VEGF) were detected between the two groups. Multivariate conditional logistic regression analysis indicated that elevated maternal leukocyte count on admission and low Apgar scores at 5 minutes were significantly associated with an increased risk of ROP.

Conclusions.: Cytokine levels in cord blood are not associated with the risk of ROP, whereas elevated maternal blood leukocyte count and low Apgar score are associated with ROP. These data suggest that the determination of cytokine levels in cord blood samples in premature infants may be of little value for predicting ROP.

Introduction
Retinopathy of prematurity (ROP) is the leading cause of blindness in preterm infants, with an incidence rate of approximately 30% in preterm infants born at or earlier than 32 weeks gestation. 1,2 Blindness from ROP, which is usually bilateral, develops in the early period of life and affects the quality of life of patients. However, a growing body of evidence indicates that early detection and treatment of ROP by screening premature infants can dramatically decrease the incidence of severe vision loss and reduce unfavorable outcomes. 3,4 Therefore, identifying biomarkers that predict ROP in high-risk infants and performing timely therapeutic interventions are of the utmost importance for salvaging vision in preterm infants with ROP. 
Although several perinatal and postnatal factors have been associated with ROP, little is known about prenatal factors. A recent study 2 by our group showed that elevated maternal leukocyte count and clinical chorioamnionitis were associated with the occurrence of ROP, suggesting that prenatal factors associated with systemic inflammation may be important for determining susceptibility to ROP. Among these prenatal factors, cord blood cytokines are important because the prenatal cytokine environment may be affected by prenatal events such as inflammation, hypoxia, and stress. Indeed, several studies 5,6 have reported increases in the levels of several cytokines and growth factors in the vitreous and blood samples obtained 1 to 4 days after birth in infants who developed ROP. However, the collection of vitreous or blood samples from premature infants is difficult and has potential risks; therefore, other biologic samples that can be easily obtained without associated risks, such as cord blood, are of interest to researchers. Currently, only one study 7 has investigated the relationship between cytokines in cord blood at birth and the development of neonatal complications, including ROP, without controlling for gestational age. To date, there are no well-designed studies investigating the association between cytokines in cord blood and the development of ROP by adjusting for known risk factors of ROP, such as gestational age and birth weight. 
The purpose of this study was to explore potential biomarkers among cytokines and growth factors in cord blood at birth and to identify clinical parameters predictive of the development of ROP in preterm newborns through a case-control study with a high degree of matching for gestational age and birth weight. 
Methods
The study followed the tenets of the Declaration of Helsinki and was approved by the institutional review board of Seoul National University Bundang Hospital. We used a retrospective matched case-control design for our study, and the collection and analysis of clinical data were of a post hoc nature. The inclusion criteria for case and control subjects were as follows: (1) live, singleton infants born earlier than 32.0 weeks gestation with survival until 38 weeks postmenstrual age, (2) infants undergoing ROP screening examinations, and (3) the absence of a major congenital anomaly. A detailed database of all obstetric patients and their neonates admitted to the neonatal intensive care unit (NICU) has been maintained at our institution since 2003. Between June 2004 and July 2010, a total of 344 newborn infants who were born earlier than 32 weeks were admitted to the NICU, and 322 survived to 36 weeks postmenstrual age. Of these 322 infants, 215 were singletons, and 107 were multiple gestation. Sixty-five singleton infants developed ROP, an incidence of 30.2% (65/215). Of these 65 singleton infants, 20 who met the inclusion criteria and had cord blood samples available comprised the present study group. A control group composed of two infants individually matched by gestational age (difference, <1 week), birth weight (difference, <100 g), and sex to each case, was selected. Written informed consent was obtained from the parents of all infants (participants) whose samples and data were used for the study. 
The screening examination for ROP followed the guidelines proposed by the American Academy of Ophthalmology and Pediatrics and the Association for Pediatric Ophthalmology and Strabismus. The first examination was performed 4 to 6 weeks after birth or 31 to 33 weeks post conception age, whichever was later. The subsequent follow up examination and laser treatment for type 1 ROP followed the clinical algorithm suggested by the Early Treatment for Retinopathy of Prematurity (ETROP) study results. 4 The stage of ROP was defined as the highest stage during follow up fundus examinations. 
Data on maternal factors, labor and delivery characteristics, and newborn parameters were retrieved from a computerized database that prospectively collected clinical data before inclusion in the study using an electronically prepared Excel-based (Microsoft Corporation, Redmond, WA) data collection tool. Maternal factors included maternal age, parity, gestational age at delivery, causes of preterm birth (i.e., preterm labor, preterm premature rupture of membranes, and pre-eclampsia), maternal blood leukocyte count on admission, clinical diagnosis of chorioamnionitis, and antenatal use of medications (tocolytics, antenatal steroids, and antibiotics before delivery). Gestational age was estimated on the basis of the last menstrual period and ultrasound information obtained in the first or second trimester. Labor, delivery, and newborn characteristics noted were as follows: sex, mode of delivery, birth weight, Apgar scores at 1 and 5 minutes, placental pathologic diagnoses, the type of respiratory support (nasal continuous positive airway pressure [CPAP], mechanical ventilation, or room air/no support), postnatal use of steroids, indomethacin, and surfactant, and the presence or absence of any of the following neonatal morbidity: respiratory distress syndrome (RDS), bronchopulmonary dysplasia (BPD), intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), necrotizing enterocolitis (NEC), and patent ductus arteriosus (PDA). 
Histologic chorioamnionitis was diagnosed on the basis of the presence of acute inflammatory changes in tissue samples (amnion, chorion-decidua, umbilical cord, or chorionic plate), using criteria 8 published previously. Funisitis was diagnosed on the basis of the presence of neutrophil infiltration into the umbilical vessel walls or Wharton's jelly. Clinical chorioamnionitis was diagnosed according to the criteria of Gibbs et al. 9 RDS, 8 BPD, 10 IVH, 11 PVL, 12 and NEC, 13 were diagnosed according to the definitions previously described in detail. 
Umbilical cord blood samples were collected in plastic tubes containing EDTA by venipuncture from the umbilical vein at birth. Samples were then centrifuged, and the supernatants were aliquoted and stored in polypropylene tube at −70°C until assay. The levels of inflammatory cytokines (IL-1β, IL-4, IL-6, IL-8, IL-10, IL-12, interferon [IFN]-γ, and TNF-α) and VEGF in cord plasma were simultaneously measured using multiplex detection kits (Bio-Rad Laboratories, Hercules, CA). Multiplex cytokine array analysis was performed with the Bio-Plex protein array system (Bio-Rad Laboratories) using Luminex-based technology. 14 The assay was performed according to the protocol, except that 50 μL of each cord plasma sample was diluted at a ratio of 1:4 in sample diluents. The lower limits of detection for cytokines were different for each cytokine and ranged from 0.03 to 2.04 pg/mL. However, because the samples were run in a 1:4 dilution, the lower limits of detection for each cytokine in umbilical cord plasma were as follows: IL-1β (0.68 pg/mL), IL-4 (0.12 pg/mL), IL-6 (1.2 pg/mL), IL-8 (4.2 pg/mL), IL-10 (0.84 pg/mL), IL-12 (1.36 pg/mL), IFN-γ (8.16 pg/mL), and TNF-α (1.6 pg/mL). For samples containing cord plasma cytokine levels below the lowest point on the standard curve, the lowest value was used. Data analyses were performed using Bio-Plex Manager software version 4.1.1 (Bio-Rad Laboratories). The mean coefficient of variance (CV) was less than 10% for all analyzed markers, except for IL-6, IL-12, and TNF-α, for which the CVs were 14% to 19%. The level of insulin-like growth factor (IGF)-1 in cord plasma was determined using a standardized ELISA (R&D Systems, Minneapolis, MN) in duplicate, according to the protocol recommended by the manufacturer. The assay for IGF-1 measured reliably in the range of 0.094 to 6 ng/mL with intra-assay and interassay CVs of 3.9% and 8.1%, respectively. For IGF-1, plasma samples were diluted 100-fold before the assay. 
Two control preterm infants were matched to each case because this ratio has been shown to provide better information than a 1:1 ratio for small relative risks in case-control studies. 15 The sample size was determined before the start of the study. Based on a previously published study, 16 the SD for cord plasma IL-6 was assumed to be 7.0 pg/mL. Power analysis with an α of 0.05 and a β of 0.2 (power = 80%) revealed that 20 experimental subjects and 40 control subjects were required to detect a 5.5 pg/mL difference in cord plasma IL-6 between control and study groups. In a previous study, 16 the SDs of IL-1β, IL-4, IL-8, IL-10, IL-12, and TNF-α were shown to range from 3.27 pg/mL to 6.01 pg/mL. Therefore, we expected that the sample size of 60 would be generally sufficient for the other cytokines to yield significant differences. 
Conditional logistic regression analysis for matched case-control samples was used to examine the association between clinical variables and cytokines in cord blood and ROP. Maternal and postnatal factors were also considered as covariates. Biomarkers in cord blood that did not show a normal distribution underwent logarithmic transformation before statistical analysis. Spearman rank correlation test was used to examine the relationship between the different cytokines and growth factors measured in the cord plasma. Continuous data were expressed as mean ± SD; categorical data were expressed as numbers with percentages. When a variable in either case or control group was 0, Firth's method of bias correction 17 was used. Variables showing a significant correlation or a tendency towards an association with the occurrence of ROP in univariate analysis (P < 0.1) were included in a multivariate conditional logistic regression model. SPSS 18.0 for windows (SPSS, Inc., Chicago, IL) was used for statistical analyses, and results were considered statistically significant at P values < 0.05. 
Results
Of the 20 infants included in the case group, 11 (55%) had stage 1 ROP, 6 (30%) had stage 2 ROP, and 3 (15%) had stage 3 ROP. None of the infants developed stage 4 or 5 ROP. Three infants with stage 3 ROP (15%) underwent laser treatment. 
Table 1 shows the demographic and delivery characteristics of the controls and infants with ROP. According to the matching, gestational age at birth, birth weight, and sex were nearly identical between the two groups. There were no differences in the mode of delivery or Apgar scores at 1 minute between the 2 groups. Apgar scores at 5 minutes tended to be lower in the ROP group than controls, although the results did not show statistical significance (6.5 vs. 7.2, P = 0.054). 
Table 1. 
 
Demographic and Delivery Characteristics of the Matched Controls and Infants With ROP
Table 1. 
 
Demographic and Delivery Characteristics of the Matched Controls and Infants With ROP
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
Gestational age, wk 29.0 ± 1.3 28.9 ± 1.9 0.59 (0.18–1.91) 0.38
Birth weight, kg 1.26 ± 0.34 1.22 ± 0.29 2.46 (0.16–37.3) 0.52
Sex, female 10 (50.0) 17 (42.5) 1.42 (0.44–4.64) 0.56
Mode of delivery, cesarean 8 (40) 26 (65) 0.88 (1.49–1.14) 0.24
Apgar score at 1 min 5 (1–8) 6 (1–8) 0.85 (0.62–1.17) 0.32
Apgar score at 5 min 6.5 (5–8) 7 (4–9) 0.62 (0.38–1.01) 0.054
Prenatal characteristics and parameters were compared between the case and control subject groups using univariate analysis. (Table 2) The mothers of infants who developed ROP had a significantly higher mean leukocyte count on admission than those of control infants (15.85 × 103/mm3 vs. 12.70 × 103/mm3, odds ratio [OR] = 1.15, P = 0.031). However, there were no differences between the case and control groups in terms of maternal age, parity, pre-eclampsia and premature rupture of membrane rates, prevalence of clinical and histologic chorioamnionitis, or rates of antenatal steroid, tocolytics, and antibiotics treatment. 
Table 2. 
 
Univariate Analysis of Prenatal Characteristics of Infants With ROP and Matched Controls
Table 2. 
 
Univariate Analysis of Prenatal Characteristics of Infants With ROP and Matched Controls
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
Maternal age, y 31.30 ± 4.37 32.23 ± 3.50 0.94 (0.81–1.08) 0.383
Nulliparity 11 (55.0) 17 (42.5) 1.63 (0.57–4.69) 0.365
Pre-eclampsia 4 (20.0) 12 (30.0) 0.57 (0.16–2.09) 0.398
PROM 7 (35.0) 18 (45.0) 0.61 (0.18–2.10) 0.44
Maternal blood leukocyte count on admission, ×103 cells/mm3 15.85 ± 5.08 12.70 ± 4.46 1.15 (1.01–1.30) 0.031
Clinical chorioamnionitis 3 (15.0) 0 (0) 16.20 (0.51–519.24) 0.116*
Antenatal corticosteroids 17 (85.0) 38 (95.0) 0.33 (0.06–1.99) 0.229
Antenatal antibiotics 11 (55.0) 22 (55.0) 1 (0.32–3.17) 1
Antenatal tocolytics 14 (70.0) 27 (67.5) 1.16 (0.31–4.4) 0.823
Histologic chorioamnionitis 12 (60.0) 21 (52.5) 1.46 (0.43–4.92) 0.542
No significant differences were observed between the two groups with regard to neonatal characteristics and morbidities of infants with ROP and controls in all clinical variables, including major treatments such as continuous positive airway pressure, mechanical ventilation, blood transfusions, and indomethacin and surfactant use and major neonatal morbidities such as RDS, BPD, IVH, PVL, NEC, and PDA (Table 3). 
Table 3. 
 
Univariate Analysis of Neonatal Characteristics and Morbidities of Infants With ROP and Matched Controls
Table 3. 
 
Univariate Analysis of Neonatal Characteristics and Morbidities of Infants With ROP and Matched Controls
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
Continuous positive airway pressure 18 (90.0) 35 (89.7) 1.16 (0.18–7.43) 0.88
Mechanical ventilation 11 (55.0) 22 (56.4) 0.95 (0.34–2.68) 0.93
Blood transfusions 14 (70.0) 26 (65.0) 1.31 (0.37–4.61) 0.67
Use of surfactant 9 (45) 17 (42.5) 1.11 (0.37–3.37) 0.85
Use of indomethacin 7 (35.0) 15 (37.5) 0.88 (0.26–2.96) 0.84
Postnatal use of steroids 17 (85) 38 (95) 0.33 (0.06–1.99) 0.23
Positive blood culture 1 (5.0) 3 (7.5) 0.67 (0.07–6.41) 0.73
RDS 12 (60.0) 23 (57.5) 1.11 (0.37–3.27) 0.86
BPD 9 (45.0) 16 (40.0) 1.44 (0.34–6.11) 0.62
IVH, grades 2 or more 0 (0.0) 1 (2.5) 1.61 (0.02–160.53) 0.838*
PVL 2 (10.0) 3 (7.5) 1.44 (0.19–11.12) 0.72
NEC 1 (5.0) 3 (7.5) 0.67 (0.07–6.41) 0.73
PDA 7 (35.0) 15 (37.5) 0.88 (0.26–2.96) 0.84
The proportions of cord blood samples with detectable cytokine and growth factor levels were 98.4% for IL-6, IL-10, and VEGF, and 100% for IL-1β, IL-4, IL-8, IL-12, IFN-γ, TNF-α, and IGF-1. The levels of IL-1β, IL-4, IL-10, IL-12, IFN-γ, and TNF-α in cord blood were significantly correlated with each other (r = 0.337–0.870, P < 0.01). IL-6 levels were correlated with the levels of the six other cytokines (all variables, r = 0.261–0.850, P < 0.05), except for IL-4 and TNF-α, whereas IL-8 levels were correlated only with those of IL-1β (r = 0.478, P < 0.001) and IL-6 (r = 0.703, P < 0.001). On the other hand, IGF-1 and VEGF expressions were not correlated with the levels of any cytokine. No correlation was observed between the two growth factors, IGF-1 and VEGF. Similarly, maternal blood leukocyte count showed no correlation with the levels of any cytokines. 
Cytokine and growth factor concentration profiles in cord blood and their logarithmic values were compared between infants with ROP and controls (Table 4). There were no significant differences in the levels of inflammatory cytokines (IL-1β, IL-4, IL-6, IL-8, IL-10, IL-12, IFN-γ, and TNF-α) or growth factors (IGF-1 and VEGF) between the two groups. 
Table 4. 
 
Univariate Analysis of the Levels of Cord Plasma Cytokines and Growth Factors in Infants With ROP and Matched Controls
Table 4. 
 
Univariate Analysis of the Levels of Cord Plasma Cytokines and Growth Factors in Infants With ROP and Matched Controls
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
IL-1β, pg/mL 4.69 ± 2.7 5.17 ± 4.34 0.97 (0.83–1.13) 0.68
IL-4, pg/mL 9.03 ± 4.69 8.95 ± 3.53 1 (0.88–1.14) 0.94
IL-6, pg/mL 25.54 ± 31.44 22.5 ± 34.85 1 (0.99–1.02) 0.75
IL-8, pg/mL 84.71 ± 63.13 87.67 ± 138.1 1 (1–1) 0.93
IL-10, pg/mL 9.29 ± 8.47 7.64 ± 3.51 1.05 (0.96–1.16) 0.29
IL-12, pg/mL 39.83 ± 75.49 15.67 ± 7.7 1.02 (0.99–1.05) 0.26
IFN-γ, pg/mL 384.85 ± 166.04 382.23 ± 140.4 1 (1–1) 0.95
TNF-α, pg/mL 54.89 ± 33.35 53.3 ± 25.67 1 (0.98–1.02) 0.84
VEGF, pg/mL 121.77 ± 168.41 153.06 ± 251.61 1 (1–1) 0.63
IGF-1, ng/mL 20.05 ± 7.29 21.5 ± 7.91 0.98 (0.91–1.05) 0.5
The multivariate logistic regression analysis included all variables with a P value less than 0.1 in the univariate analysis (a 5 minute Apgar score and a maternal blood leukocyte count on admission), and the results are shown in Table 5. Multivariate analysis showed that elevated maternal blood white blood cell (WBC) count on admission were significantly associated with the occurrence of ROP (OR = 1.171 per 1000 cells/mm3 incremental increase; 95% confidence interval [CI] = 1.014–1.352; P = 0.032) and a low Apgar score at 5 minutes (OR = 0.553, 95% CI = 0.317–0.965, P = 0.037). 
Table 5. 
 
Multivariate Analysis Showing the Relationship of Independent Variables With the Risk of ROP in Infants With ROP and Matched Controls
Table 5. 
 
Multivariate Analysis Showing the Relationship of Independent Variables With the Risk of ROP in Infants With ROP and Matched Controls
Parameter OR 95% CI P Value
Maternal blood WBC count on admission, × 103 cells/mm3 1.179 1.019–1.364 0.027
Apgar score at 5 min 0.553 0.317–0.965 0.037
Discussion
The present gestational age-matched case-control study on high-risk preterm infants revealed that levels of cytokines and growth factors in cord blood samples of infants with ROP were not significantly different from those of healthy controls. High maternal leukocyte counts were associated with ROP, which was consistent with our previous report. 2 In addition, a low Apgar score at 5 minutes was significantly associated with ROP. However, from a clinical standpoint, its clinical significance may be low because the difference in the median Apgar score at 5 minutes between the ROP and control groups was only 0.5 and the upper boundary of the 95% CI for the OR of the Apgar score was close to 1.0 (Table 5). 
To date, only two studies have used cord blood plasma to identify biomarkers for ROP. Takahashi et al. 7 investigated the association between cytokine profiles of cord blood and perinatal morbidities in 158 preterm and 66 term infants. They showed that IL-6, IL-8, and monocyte chemotactic protein-1 (MCP-1) were closely related to certain neonatal diseases such as RDS, chronic lung disease, and PDA in preterm neonates. The development of ROP was not correlated with any of the pro-inflammatory markers (IL-1β, IL-6, and TNF-α), Th1 cytokines (IFN-γ, IL-2, and IL-12), Th2 cytokines (IL-4, 5, 10, and 13), growth factors (IL-7, granulocyte-macrophage colony-stimulating factor, granulocyte colony-stimulating factor), and chemokines (IL-8, MCP-1, and macrophage inflammatory protein-1β) in cord blood. However, the study did not control for gestational age or birth weight in the comparative analysis, despite the fact that gestational age and birth weight are the strongest risk factors of ROP. Madan et al. 18 performed a proteomic analysis of cord blood from preterm infants and found that deamidation of globin chains was significantly higher in infants who later developed ROP stage 2 or higher. However, their study was limited by the inclusion of a small number of infants with ROP without adjustment for gestational age, and the significance of the proteomic result was not validated. In addition, the current proteomic research methodology cannot detect small molecular weight cytokines. 
Compared with the previous studies, our gestational age-matched case control study, including high-risk infants was more likely to prove the true association between cord blood cytokines and the development of ROP by controlling for the strongest risk factor of ROP. Our results showed that the candidate inflammatory cytokines in the cord blood were not associated with the development of ROP. Our results were in agreement with those of Takahashi et al., 7 who showed that the prevalence of ROP was not correlated with any cytokines in the cord blood of preterm infants. Cytokine levels in cord blood at birth reflect the perinatal events, whereas the subsequent levels in infant blood may be more reflective of the postnatal state. Therefore, the discrepancy between the negative result in our study using cord blood and the positive result using postnatal blood 5 suggests that postnatal blood cytokine profiles may be more suitable for investigating the pathogenic condition of the retina in infants with ROP. However, the infantile blood cytokine assay has limited value for the prediction of ROP, as the screening examination for ROP is usually performed in the postnatal period and infantile blood is difficult to obtain compared with cord blood. 
In our study, cord blood levels of VEGF and IGF-1, which are well known factors that play critical roles in the development of ROP, 19 were not associated with ROP. Systemic and ocular levels of VEGF and IGF-1 in preterm infants with ROP have not been studied in detail. In one study, 6 no statistically significant differences were observed in postnatal plasma VEGF levels of infants with ROP and those of controls, whereas a statistically significant difference was observed between the two groups in the aqueous and vitreous samples. This implies that the VEGF levels in the local environment, the retina, rather than the systemic levels, are indicative of the pathologic state of ROP. Therefore, predicting ROP development using the VEGF level from systemic blood samples seems to be impractical. On the other hand, IGF-1 from postnatal blood may be of value as a biomarker for predicting the occurrence of ROP, as the mean serum levels of IGF-1 in age-matched premature infants were correlated with the severity of clinical ROP. 1922 However, the postnatal sampling of peripheral blood also limit the use of blood IGF-1 as a biomarker of ROP 
Our study showed that clinical factors of prenatal inflammation such as maternal leukocyte counts are associated with ROP. Previous studies, including ours, 2,5,2325 suggested the relationship between the development of ROP and perinatal inflammation. Sood et al. investigated 20 cytokines in dried blood spots collected from preterm infants over the first 3 weeks after birth, and from among these infants, those with ROP showed a coordinated pattern of increased and decreased levels of six inflammatory markers and two growth factors. 23 In their study, a greater number of inflammatory cytokines was associated with ROP in the late postnatal period than in the early postnatal period, suggesting that the prediction of ROP using plasma inflammatory markers might be more accurate using the blood obtained at the later postnatal life. 23 Another study revealed that plasma levels of IL-6 and TNF-α cytokines in the first 72 hours of life in preterm infants with sepsis were associated with the later development of severe ROP. 5 Polam et al. showed that infants with chorioamnionitis have a higher incidence of ROP. 25 These studies, including ours, suggest the possibility that perinatal inflammation is involved in the pathogenesis of ROP. However, the mechanisms responsible for the development of ROP in the context of maternal systemic inflammation remain unclear. Given the results of the current study and those of a previous study 26 showing that cytokines do not cross the placenta, maternal systemic inflammation may not directly cause fetal systemic inflammation and ROP development. Rather, maternal exposure to inflammatory stimuli during pregnancy may impair the newborn infant's innate responses and increase the susceptibility to infection and inflammation. This is supported by the observations of Beloosesky et al., 27 who demonstrated that in rats, maternal lipopolysaccharide exposure suppresses the innate immune responses of the offspring to inflammatory stimuli. Infectious organisms and their products can stimulate the production of proinflammatory cytokines, which may exert a direct effect on retinal neovascularization via inflammation-regulated VEGF availability. 28  
Despite the reported value of the detection of markers in postnatal blood, 23 it may be difficult to obtain blood samples for predicting the development of ROP in preterm infants. Cord blood sampling, however, is not associated with any significant risk or burden to preterm infants. In addition, cord blood sampling is reported to have potential as a biomarker for predicting the occurrence of several pediatric diseases, including BPD, 29 asthma and allergic diseases, 30 and hyperbilirubinemia. 31 Because the complications of prematurity are associated with gestational age and birth weight, future studies focusing on various morbidities of premature infants should be designed by controlling for known risk factors similar to our gestational age-matched case-control study. 
There are several limitations in our study that require consideration. First, although this study was designed as a gestational age-matched case control study, which was done for the first time, the possibility of selection bias during the inclusion and matching process cannot be excluded. Second, the number of cases was relatively small to draw a definitive conclusion. In particular, the number of cases with severe ROP requiring treatment was small (15% of all ROP cases in our study); therefore, we cannot exclude the possibility that there may be some cord blood cytokines that show significant differences in their levels between infants with severe ROP and healthy controls. Future research including a larger number of cases with severe ROP is required. Third, because the study was of a retrospective design, certain parameters including Apgar scores could not be defined and objectively assessed at the time of data collection. Fourth, our study used stored frozen samples to measure cytokine levels. This may not truly reflect clinical conditions because in an actual clinical setting, fresh samples are used to assess the risk of ROP development, although previous reports 32,33 showed a significant correlation between cytokine levels in fresh and stored plasma samples. The strength of our study is that the individuals who measured cytokine levels, collected blood samples, recorded maternal and neonatal data, and confirmed the presence of ROP were blinded to each other's findings, which reduced the potential for bias. 
In conclusion, this gestational age-matched case-control study revealed that the development of ROP, especially the mild form, was not associated with cord blood levels of inflammatory cytokines and angiogenic growth factors, whereas it was associated with a high maternal blood leukocyte count and a low Apgar score at 5 minutes. Future research is warranted to reveal the association of cord blood cytokines with severe ROP and the utility of postnatal blood cytokines for the prediction of ROP development. 
Acknowledgments
The authors thank Soyeon Ahn, PhD, Jaebong Lee, MS, and the Medical Research Collaboration Center (MRCC) of Seoul National University Bundang Hospital for assistance in statistical analysis. 
Disclosure: S.J. Woo, None; K.H. Park, None; S.Y. Lee, None; S.J. Ahn, None; J. Ahn, None; K.H. Park, None; K.J. Oh, None; A. Ryu, None 
Supported by grants from the Seoul National University Bundang Hospital, Republic of Korea (02-2010-011) and in part by Grant A111161 from the Korea Health Technology R&D Project, Ministry of Health and Welfare, Republic of Korea, and National Research Foundation of Korea (NRF) Grant 2012R1A2A2A02012821, funded by the Ministry of Education, Science, and Technology. 
References
Mathew MR Fern AI Hill R. Retinopathy of prematurity: are we screening too many babies? Eye (Lond) . 2002; 16: 538–542. [CrossRef] [PubMed]
Woo SJ Park KH Jung HJ Kim S Choe G Ahn J. Effects of maternal and placental inflammation on retinopathy of prematurity. Graefes Arch Clin Exp Ophthalmol . 2012; 250: 915–923. [CrossRef] [PubMed]
Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter Trial of Cryotherapy for Retinopathy of Prematurity: ophthalmological outcomes at 10 years. Arch Ophthalmol . 2001; 119: 1110–1118. [CrossRef] [PubMed]
Early Treatment For Retinopathy Of Prematurity Cooperative Group. Revised indications for the treatment of retinopathy of prematurity: results of the early treatment for retinopathy of prematurity randomized trial. Arch Ophthalmol . 2003; 121: 1684–1694. [CrossRef] [PubMed]
Silveira RC Fortes Filho JB, Procianoy RS. Assessment of the contribution of cytokine plasma levels to detect retinopathy of prematurity in very low birth weight infants. Invest Ophthalmol Vis Sci . 2011; 52: 1297–1301. [CrossRef] [PubMed]
Velez-Montoya R Clapp C Rivera JC Intraocular and systemic levels of vascular endothelial growth factor in advanced cases of retinopathy of prematurity. Clin Ophthalmol . 2010; 4: 947–953. [CrossRef] [PubMed]
Takahashi N Uehara R Kobayashi M Cytokine profiles of seventeen cytokines, growth factors and chemokines in cord blood and its relation to perinatal clinical findings. Cytokine . 2010; 49: 331–337. [CrossRef] [PubMed]
Yoon BH Romero R Kim CJ Amniotic fluid interleukin-6: a sensitive test for antenatal diagnosis of acute inflammatory lesions of preterm placenta and prediction of perinatal morbidity. Am J Obstet Gynecol . 1995; 172: 960–970. [CrossRef] [PubMed]
Gibbs RS Blanco JD St Clair PJ Castaneda YS. Quantitative bacteriology of amniotic fluid from women with clinical intraamniotic infection at term. J Infect Dis . 1982; 145: 1–8. [CrossRef] [PubMed]
Jobe AH Bancalari E. Bronchopulmonary dysplasia. Am J Respir Crit Care Med . 2001; 163: 1723–1729. [CrossRef] [PubMed]
Papile LA Burstein J Burstein R Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr . 1978; 92: 529–534. [CrossRef] [PubMed]
Yoon BH Jun JK Romero R Amniotic fluid inflammatory cytokines (interleukin-6, interleukin-1beta, and tumor necrosis factor-alpha), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol . 1997; 177: 19–26. [CrossRef] [PubMed]
Richter D Bartmann P Pohlandt F. Prevention of necrotizing enterocolitis in extremely low birth weight infants by IgG feeding? Eur J Pediatr . 1998; 157: 924–925. [CrossRef] [PubMed]
de Jager W te Velthuis H Prakken BJ Kuis W Rijkers GT. Simultaneous detection of 15 human cytokines in a single sample of stimulated peripheral blood mononuclear cells. Clin Diagn Lab Immunol . 2003; 10: 133–139. [PubMed]
Holford TR Bracken MB Eskenazi B. Log-linear models for the analysis of matched cohort studies. Am J Epidemiol . 1989; 130: 1247–1253. [PubMed]
Takahashi N Uehara R Kobayashi M Cytokine profiles of seventeen cytokines, growth factors and chemokines in cord blood and its relation to perinatal clinical findings. Cytokine . 2010; 49: 331–337. [CrossRef] [PubMed]
Firth D. Bias reduction of maximum likelihood estimates. Biometrika . 1993; 80: 27–38. [CrossRef]
Madan A El-Ferzli G Carlson SM A potential biomarker in the cord blood of preterm infants who develop retinopathy of prematurity. Pediatr Res . 2007; 61: 215–221. [CrossRef] [PubMed]
Chen J Smith LE. Retinopathy of prematurity. Angiogenesis . 2007; 10: 133–140. [CrossRef] [PubMed]
Hellstrom A Perruzzi C Ju M Low IGF-I suppresses VEGF-survival signaling in retinal endothelial cells: direct correlation with clinical retinopathy of prematurity. Proc Natl Acad Sci U S A . 2001; 98: 5804–5808. [CrossRef] [PubMed]
Hellstrom A Engstrom E Hard AL Postnatal serum insulin-like growth factor I deficiency is associated with retinopathy of prematurity and other complications of premature birth. Pediatrics . 2003; 112: 1016–1020. [CrossRef] [PubMed]
Lofqvist C Engstrom E Sigurdsson J Postnatal head growth deficit among premature infants parallels retinopathy of prematurity and insulin-like growth factor-1 deficit. Pediatrics . 2006; 117: 1930–1938. [CrossRef] [PubMed]
Sood BG Madan A Saha S Perinatal systemic inflammatory response syndrome and retinopathy of prematurity. Pediatr Res . 2010; 67: 394–400. [CrossRef] [PubMed]
Dammann O Brinkhaus MJ Bartels DB Immaturity, perinatal inflammation, and retinopathy of prematurity: a multi-hit hypothesis. Early Hum Dev . 2009; 85: 325–329. [CrossRef] [PubMed]
Polam S Koons A Anwar M Shen-Schwarz S Hegyi T. Effect of chorioamnionitis on neurodevelopmental outcome in preterm infants. Arch Pediatr Adolesc Med . 2005; 159: 1032–1035. [CrossRef] [PubMed]
Aaltonen R Heikkinen T Hakala K Laine K Alanen A. Transfer of proinflammatory cytokines across term placenta. Obstet Gynecol . 2005; 106: 802–807. [CrossRef] [PubMed]
Beloosesky R Maravi N Weiner Z Maternal lipopolysaccharide-induced inflammation during pregnancy programs impaired offspring innate immune responses. Am J Obstet Gynecol . 2010; 203: 185 e181- e184.
Lee J Dammann O. Perinatal infection, inflammation, and retinopathy of prematurity. Semin Fetal Neonatal Med . 2012; 17: 26–29. [CrossRef] [PubMed]
Tsao PN Wei SC Su YN Placenta growth factor elevation in the cord blood of premature neonates predicts poor pulmonary outcome. Pediatrics . 2004; 113: 1348–1351. [CrossRef] [PubMed]
Kaan A Dimich-Ward H Manfreda J Cord blood IgE: its determinants and prediction of development of asthma and other allergic disorders at 12 months. Ann Allergy Asthma Immunol . 2000; 84: 37–42. [CrossRef] [PubMed]
Bernaldo AJ Segre CA. Bilirubin dosage in cord blood: could it predict neonatal hyperbilirubinemia? Sao Paulo Med J . 2004; 122: 99–103. [CrossRef] [PubMed]
Hosnijeh FS Krop EJ Portengen L Stability and reproducibility of simultaneously detected plasma and serum cytokine levels in asymptomatic subjects. Biomarkers . 2010; 15: 140–148. [CrossRef] [PubMed]
Ito Y Nakachi K Imai K Stability of frozen serum levels of insulin-like growth factor-I, insulin-like growth factor-II, insulin-like growth factor binding protein-3, transforming growth factor beta, soluble Fas, and superoxide dismutase activity for the JACC study. J Epidemiol . 2005; 15: S67–S73. [CrossRef] [PubMed]
Table 1. 
 
Demographic and Delivery Characteristics of the Matched Controls and Infants With ROP
Table 1. 
 
Demographic and Delivery Characteristics of the Matched Controls and Infants With ROP
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
Gestational age, wk 29.0 ± 1.3 28.9 ± 1.9 0.59 (0.18–1.91) 0.38
Birth weight, kg 1.26 ± 0.34 1.22 ± 0.29 2.46 (0.16–37.3) 0.52
Sex, female 10 (50.0) 17 (42.5) 1.42 (0.44–4.64) 0.56
Mode of delivery, cesarean 8 (40) 26 (65) 0.88 (1.49–1.14) 0.24
Apgar score at 1 min 5 (1–8) 6 (1–8) 0.85 (0.62–1.17) 0.32
Apgar score at 5 min 6.5 (5–8) 7 (4–9) 0.62 (0.38–1.01) 0.054
Table 2. 
 
Univariate Analysis of Prenatal Characteristics of Infants With ROP and Matched Controls
Table 2. 
 
Univariate Analysis of Prenatal Characteristics of Infants With ROP and Matched Controls
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
Maternal age, y 31.30 ± 4.37 32.23 ± 3.50 0.94 (0.81–1.08) 0.383
Nulliparity 11 (55.0) 17 (42.5) 1.63 (0.57–4.69) 0.365
Pre-eclampsia 4 (20.0) 12 (30.0) 0.57 (0.16–2.09) 0.398
PROM 7 (35.0) 18 (45.0) 0.61 (0.18–2.10) 0.44
Maternal blood leukocyte count on admission, ×103 cells/mm3 15.85 ± 5.08 12.70 ± 4.46 1.15 (1.01–1.30) 0.031
Clinical chorioamnionitis 3 (15.0) 0 (0) 16.20 (0.51–519.24) 0.116*
Antenatal corticosteroids 17 (85.0) 38 (95.0) 0.33 (0.06–1.99) 0.229
Antenatal antibiotics 11 (55.0) 22 (55.0) 1 (0.32–3.17) 1
Antenatal tocolytics 14 (70.0) 27 (67.5) 1.16 (0.31–4.4) 0.823
Histologic chorioamnionitis 12 (60.0) 21 (52.5) 1.46 (0.43–4.92) 0.542
Table 3. 
 
Univariate Analysis of Neonatal Characteristics and Morbidities of Infants With ROP and Matched Controls
Table 3. 
 
Univariate Analysis of Neonatal Characteristics and Morbidities of Infants With ROP and Matched Controls
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
Continuous positive airway pressure 18 (90.0) 35 (89.7) 1.16 (0.18–7.43) 0.88
Mechanical ventilation 11 (55.0) 22 (56.4) 0.95 (0.34–2.68) 0.93
Blood transfusions 14 (70.0) 26 (65.0) 1.31 (0.37–4.61) 0.67
Use of surfactant 9 (45) 17 (42.5) 1.11 (0.37–3.37) 0.85
Use of indomethacin 7 (35.0) 15 (37.5) 0.88 (0.26–2.96) 0.84
Postnatal use of steroids 17 (85) 38 (95) 0.33 (0.06–1.99) 0.23
Positive blood culture 1 (5.0) 3 (7.5) 0.67 (0.07–6.41) 0.73
RDS 12 (60.0) 23 (57.5) 1.11 (0.37–3.27) 0.86
BPD 9 (45.0) 16 (40.0) 1.44 (0.34–6.11) 0.62
IVH, grades 2 or more 0 (0.0) 1 (2.5) 1.61 (0.02–160.53) 0.838*
PVL 2 (10.0) 3 (7.5) 1.44 (0.19–11.12) 0.72
NEC 1 (5.0) 3 (7.5) 0.67 (0.07–6.41) 0.73
PDA 7 (35.0) 15 (37.5) 0.88 (0.26–2.96) 0.84
Table 4. 
 
Univariate Analysis of the Levels of Cord Plasma Cytokines and Growth Factors in Infants With ROP and Matched Controls
Table 4. 
 
Univariate Analysis of the Levels of Cord Plasma Cytokines and Growth Factors in Infants With ROP and Matched Controls
ROP Group, n = 20 Control Group, n = 40 OR (95% CI) P Value
IL-1β, pg/mL 4.69 ± 2.7 5.17 ± 4.34 0.97 (0.83–1.13) 0.68
IL-4, pg/mL 9.03 ± 4.69 8.95 ± 3.53 1 (0.88–1.14) 0.94
IL-6, pg/mL 25.54 ± 31.44 22.5 ± 34.85 1 (0.99–1.02) 0.75
IL-8, pg/mL 84.71 ± 63.13 87.67 ± 138.1 1 (1–1) 0.93
IL-10, pg/mL 9.29 ± 8.47 7.64 ± 3.51 1.05 (0.96–1.16) 0.29
IL-12, pg/mL 39.83 ± 75.49 15.67 ± 7.7 1.02 (0.99–1.05) 0.26
IFN-γ, pg/mL 384.85 ± 166.04 382.23 ± 140.4 1 (1–1) 0.95
TNF-α, pg/mL 54.89 ± 33.35 53.3 ± 25.67 1 (0.98–1.02) 0.84
VEGF, pg/mL 121.77 ± 168.41 153.06 ± 251.61 1 (1–1) 0.63
IGF-1, ng/mL 20.05 ± 7.29 21.5 ± 7.91 0.98 (0.91–1.05) 0.5
Table 5. 
 
Multivariate Analysis Showing the Relationship of Independent Variables With the Risk of ROP in Infants With ROP and Matched Controls
Table 5. 
 
Multivariate Analysis Showing the Relationship of Independent Variables With the Risk of ROP in Infants With ROP and Matched Controls
Parameter OR 95% CI P Value
Maternal blood WBC count on admission, × 103 cells/mm3 1.179 1.019–1.364 0.027
Apgar score at 5 min 0.553 0.317–0.965 0.037
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×