Systematic Review and Meta-Analysis Neonatology 2014;105:189–199 DOI: 10.1159/000357556
Received: October 8, 2013 Accepted after revision: November 26, 2013 Published online: January 24, 2014
Chorioamnionitis as a Risk Factor for Retinopathy of Prematurity: A Systematic Review and Meta-Analysis Souvik Mitra a, e Dagfinn Aune b, c Christian P. Speer d Ola Didrik Saugstad a a Department of Paediatric Research, Oslo University Hospital, University of Oslo, Oslo, and b Department of Public Health, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway; c Department of Epidemiology and Biostatistics, Imperial College London, London, UK; d University Children’s Hospital, University of Würzburg, Würzburg, Germany; e Division of Neonatology, Department of Paediatrics, McMaster University Medical Centre, Hamilton, Ont., Canada
Key Words Chorioamnionitis · Retinopathy of prematurity · Systematic review · Meta-analysis
Abstract Background: The role of chorioamnionitis (CA) in the development of retinopathy of prematurity (ROP) has not been well established. Objective: To conduct a systematic review and meta-analysis of the association between CA and ROP in preterm infants. Data Sources: The authors searched MEDLINE, Embase, CINAHL, Cochrane Central Register of Controlled Trials and PubMed, reviewed reference lists of relevant articles, abstracts and conference proceedings (Society for Pediatric Research, European Society for Paediatric Research 1990–2012), sought results of unpublished trials, and contacted the primary authors of relevant studies. Study Selection: Studies were included if they had a comparison group, examined preterm infants, and reported primary data that could be used to measure the association between exposure to CA and the development of ROP. Data Extraction: Two reviewers independently screened the search results, applied inclusion criteria and assessed methodological quality using the Newcastle-Ottawa Scale. One
© 2014 S. Karger AG, Basel 1661–7800/14/1053–0189$39.50/0 E-Mail [email protected] www.karger.com/neo
reviewer extracted data and a second reviewer checked data extraction. Summary relative risks (RRs) were calculated using a random effects model. Data Synthesis: We identified 1,249 potentially relevant studies from the electronic databases. Twenty-seven studies involving 10,590 preterm neonates with 2,562 cases of ROP were included. Taking into account all included studies without adjusting for gestational age (GA), CA was significantly associated with ROP (any stage) [summary RR 1.33 (95% CI 1.14–1.55, I2 = 77%, pheterogeneity < 0.0001)], and a borderline significant association was observed for severe ROP (stage ≥3) [summary RR 1.27 (95% CI 0.99–1.63, I2 = 74%, pheterogeneity < 0.0001)]. There was no publication bias with Begg’s test. However, subgroup analysis of studies adjusting for GA showed no significant association on CA with ROP [summary RR 0.98 (95% CI 0.77–1.26, I2 = 0%, pheterogeneity = 0.89)]. Conclusion: Unadjusted analyses showed that CA was significantly associated with ROP (any stage) as well as with severe ROP (stage ≥ 3). However, the association disappeared on analysis of studies adjusting for GA. Hence, CA cannot be definitively considered as a risk factor for ROP, and further studies should adjust for potential confounding factors and report results by stage to clarify the association with severe ROP. © 2014 S. Karger AG, Basel
Souvik Mitra, MD Division of Neonatology, Department of Pediatrics McMaster University Medical Centre HSC-4F, 1280 Main Street West, Hamilton, ON L8S 4K1 (Canada) E-mail [email protected] or souvik4sphs @ yahoo.com
Introduction
Retinopathy of prematurity (ROP) is a retinal vascular disease of preterm infants, first described in 1942 [1]. ROP was initially connected with premature birth in the 1940s and slightly later to oxygen supplementation in these premature infants [2]. It is a major cause of blindness in children in the developing and developed world despite current surgical and laser treatment in the late stage of the disease [3]. Given the high prevalence and sequelae of ROP, effective prevention and timely treatment are essential for the preservation of lifelong vision in these premature neonates. A number of studies over the years have shown that low birth weight, early gestational age (GA) and oxygen use are the major risk factors for ROP [4–8]. However, primary prevention of ROP through elimination of risk factors like prematurity is practically not feasible [7, 9]. Therefore, there is a pressing need to identify contributing factors based on the prenatal origin of ROP, because preventive factors for ROP targeting the mothers and their fetuses may offer new opportunities for successful prevention of ROP [10]. Preterm labor and delivery, the leading identifiable causes of ROP, have been shown to be associated with maternal infections like chorioamnionitis (CA) and systemic infections [11]. Histological CA is defined by maternal inflammatory response with neutrophilic infiltration of the membranes and/or chorionic plate, with or without a fetal inflammatory co-response. On the other hand, clinical CA is defined by the attending physician in presence of maternal fever along with other suggestive clinical features. The incidence of CA has been found to increase with decreasing GA [12, 13]. It has also been shown in observational studies conducted in cohorts of extremely low-birth-weight infants that early-onset sepsis and systemic candidiasis are independently associated with increased severity of ROP and the need for laser treatment [14–16]. These findings suggest that perinatal inflammation and infection may have a deleterious effect on the developing blood vessels in the retina, making them vulnerable to the development of ROP. In a number of recently published studies, histological CA has in fact been shown to be an important risk factor for the development of early-onset sepsis as well as subsequent neonatal disorders of the lung, intestines and brain including white matter injuries and intraventricular hemorrhage [17–25]. A number of studies have tried to evaluate the relationship of CA and ROP. Intrauterine inflammation may be responsible for an adverse ‘first hit’ to the immature 190
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
fetal organism which is followed by various postnatal events (‘multiple hits’) aggravating the risk of ROP in preterm infants [26, 27]. However, other investigators have found no significant association of CA with ROP [21, 28]. This disagreement may be due to the relatively small sample size, lack of control for known risk factors (GA, birth weight and hyperoxia/hypoxia), wide variation in outcome measures [e.g. any ROP, severe (stage 3 and above) ROP, prethreshold and threshold ROP], and the lack of consensus regarding the definition of CA in previous studies. In an effort to resolve the discrepancy observed across studies, we conducted a systematic review and meta-analysis to synthesize the literature that measures the association of CA with ROP in premature infants. Methods Sources We searched the Cochrane Central Register of Controlled Trials (The Cochrane Library issue 1 of 12, January 2013) using the terms RETINOPATHY OF PREMATURITY AND CHORIOAMNIONITIS. We searched MEDLINE (January 1966 to January 2013) and PubMed (2013) using exploded MeSH terms RETINOPATHY OF PREMATURITY, CHORIOAMNIONITIS; ROP, CHORIOAMNIONITIS; CHORIOAMNIONITIS, NEURODEVELOPMENTAL OUTCOME; CHORIOAMNIONITIS, LONG TERM EFFECTS; PLACENTAL INFLAMMATION; PLACENTA HISTOLOGY, RETINOPATHY OF PREMATURITY and all possible combinations thereof in all languages (or without language restrictions). We searched EMBASE (January 1974 to January 2013) using the terms RETINOPATHY OF PREMATURITY with CHORIOAMNIONITIS. We searched the Science Citation Index (September 1984 to January 2013) for trials which cited the trials identified by our searches as MEDLINE and Embase. We sought registered details of selected trials in the U.S. National Institutes of Health resource Clinicaltrials.gov. We obtained information by personal communication, reviewing the reference lists of relevant articles, abstracts and conference proceedings (Society for Pediatric Research, European Society for Paediatric Research 1990–2012) and seeking results of unpublished trials. Study Selection Studies were included if they had a comparison group, examined preterm infants, and reported primary data that could be used to measure the association between exposure to CA and the development of ROP. We defined CA cases as clinical, histological or both clinical and histological. Clinical CA was defined as documented clinical signs and symptoms suggestive of CA. Histological CA was defined based on previously reported definitions as an acute inflammatory change in any of the tissue samples (amnion, chorion deciduas, umbilical cord and chorionic plate) [29]. ROP was primarily defined according to the International Classification of Retinopathy of Prematurity and severe ROP as stages 3, 4 or 5 [30]. Studies where data on ROP were collected from the neonatal
Mitra /Aune /Speer /Saugstad
Records identified through electronic database searching after removing duplicates (n = 1,249) Excluded after title review (n = 226) Abstracts reviewed (n = 1,023) Excluded after abstract review (n = 955) Full-text articles assessed for eligibility (n = 68) Additional records obtained through personal communication with authors (n = 3) Studies included in the systematic review and meta-analysis (n = 27)
Full-text articles excluded (n = 44) Exposure not defined (n = 29) Outcome not included (n = 9) Outcome not defined (n = 3) Exposure and outcome defined but detailed statistics not included in full text (n = 3)
Fig. 1. Study flow diagram.
medical records were also included. To identify relevant studies, 2 reviewers independently screened the results of the searches and applied inclusion criteria using a structured form. Discrepancies were resolved through discussion or in consultation with a third reviewer. Quality Assessment We assessed the methodological quality using the NewcastleOttawa Scale for case-control or cohort studies as appropriate [31, 32]. The scale assesses for potential selection bias, comparability of cohorts or cases and controls and ascertainment of exposure (cohort studies) or outcome (case-control studies). Two reviewers independently assessed the methodological quality of each study and any discrepancy was resolved through consultation with a third reviewer.
Statistical Analysis We calculated summary relative risks (RRs) comparing risk of ROP among preterm infants born to women with CA with those without CA using a random effects model which accounts for within and between study heterogeneity [33]. The average of the natural logarithm of the RRs was estimated, and the RRs from each study were weighted by the inverse of the variance. A two-tailed p < 0.05 was considered statistically significant. Heterogeneity between studies was evaluated using Q and I2 statistics [34]. I2 is the amount of total variation that is due to variation between studies. I2 values of approximately 25, 50, and 75% indicate low, moderate and high heterogeneity, respectively. Publication bias was assessed using Begg-Mazumdar’s test [35]. We conducted sensitivity analyses excluding one study at a time to test whether the results were robust. Subgroup analyses were done based on the definition of CA, severity of ROP and studies adjusting for GA.
Data Extraction One reviewer extracted data from relevant studies using a predetermined data extraction form. A second reviewer checked the data extraction for accuracy and completeness. All data included in the meta-analysis were monitored by the statistician. Discrepancies were resolved by consulting the primary report. Data extracted from each study included citation information, language of publication, country where the research was conducted, objectives, study design, definitions of histological CA and ROP, inclusion/ exclusion criteria, patient characteristics and results. Raw data, where available, were extracted from the results to analyze the effect of CA on ROP.
Description of the Studies We identified 1,249 potentially relevant studies from the electronic databases. Twenty-seven studies including 10,590 preterm infants met our study requirements (fig. 1). The clinical profile of the infants and the methodological characteristics of the studies have been summarized in online supplementary appendix A (see www.karger.com/doi/10.1159/000357556 for all online suppl. material) [36–69]. Studies were published between 1997 and 2013 (median year 2011). All studies except one were in English [58]. While all the studies provided data to measure the association between ROP and CA, the objec-
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
191
Study
RR (95% CI)
Soraisham, 2013 Nasef, 2012 Perrone, 2012 Seliga, 2012 Van Vliet, 2012 Woo, 2012 Bordigato, 2011 Chen, 2011 Henderson, 2011 Lee, 2011 Moscuzza, 2011 Sato, 2011 Wirbelauer, 2011 Schlapbach, 2010 Dammann, 2009 Soraisham, 2009 Suppiej, 2009 Mehta, 2006 Rocha, 2006 Felice, 2005 Lau, 2005 Polam, 2005 Fung, 2003 Fung, 2003 Ogunyemi, 2003 Gray, 1997 Morales, 1987 Overall
1.03 (0.91 – 1.17) 0.96 (0.61 – 1.50) 1.88 (1.16 – 3.05) 1.72 (1.24 – 2.38) 4.05 (1.06 – 15.43) 0.98 (0.71 – 1.37) 1.88 (0.70 – 5.04) 1.00 (0.93 – 1.07) 1.69 (1.14 – 2.52) 1.12 (0.68 – 1.85) 6.15 (0.66 – 57.27) 1.55 (0.97 – 2.46) 2.12 (0.39 – 11.64) 0.50 (0.10 – 2.55) 1.62 (1.18 – 2.24) 1.58 (1.22 – 2.05) 1.42 (0.72 – 2.80) 2.04 (1.07 – 3.92) 1.22 (0.51 – 2.92) 3.86 (1.17 – 12.75) 0.26 (0.15 – 0.48) 1.80 (1.37 – 2.37) 0.67 (0.19 – 2.30) 1.26 (0.40 – 3.99) 1.49 (1.20 – 1.87) 2.03 (1.15 – 3.58) 0.84 (0.50 – 1.41) 1.33 (1.14 – 1.55) 0.1
0.5
1
2
3
5
10
20
Risk ratio
Fig. 2. Forest plot for 27 studies examining the effect of CA (any type) on ROP (any stage). RR >1 indicates increased risk of ROP in
infants born to mothers with CA.
SE of logRR
0
0.5
1.0
1.5 –2
–1
0 logRR
1
2
Fig. 3. Funnel plot with pseudo-95% confidence limits for publica-
tion bias assessment for studies analyzing the effect of CA on ROP (any stage). SE = Standard error.
192
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
tives of the studies varied. The risk factors for developing ROP in preterm infants were studied in 4 studies [26, 39, 51, 68]. The rest of the studies evaluated the occurrence of ROP as one of the outcomes of clinical or histological CA. Additional information was collected by personal communication with the authors from 3 studies [39, 51, 52]. Eight studies defined both clinical as well as histological CA [42– 44, 50, 52, 56, 61, 68]. Two studies dealt with only clinical CA [26, 28], while the rest of the studies only defined histological CA. In the studies that defined clinical CA, the definition was not specified in one of the studies [26]. In the rest of the studies defining clinical CA, the definitions were by and large similar in accordance with the criteria suggested by Gibbs et al. [69]. Histological CA was not defined by 2 studies [49, 50]. The criteria used for histological CA definition varied in the studies where it was mentioned [29, 37, 38, 48, 55, 59, 67]. Redline’s criteria were the most commonly used criteria to define histological CA [38]. In all except 2 studies, the mean or median GA was less than 30 weeks [21, 58]. In 19 of 27 studies, the patient characteristics were clearly differentiated between the CA and non-CA groups. In 15 out of these 19 studies, there was a significantly higher GA at birth among the non-CA group compared to the CA group
Mitra /Aune /Speer /Saugstad
Study
RR (95% CI)
Weight, %
Soraisham, 2013
1.40 (1.00 – 1.96)
9.2
Nasef, 2012
1.34 (0.50 – 3.59)
4.1
Seliga, 2012
1.72 (1.24 – 2.38)
9.3
Woo, 2012
1.00 (0.63 – 1.57)
8.1
Bordigato, 2011
3.20 (0.14 – 72.62)
0.6
Chen, 2011
0.96 (0.80 – 1.15)
10.5
Henderson, 2011
1.69 (1.14 – 2.52)
8.7
Lee, 2011
1.12 (0.68 – 1.85)
7.6
Sato, 2011
1.55 (0.97 – 2.46)
8.0
Wirbelauer, 2011
2.12 (0.39 – 11.64)
1.8
Schlapbach, 2010
0.67 (0.14 – 3.12)
2.1
Dammann, 2009
4.55 (1.69 – 12.21)
4.0
Soraisham, 2009
1.58 (1.22 – 2.05)
9.9
Lau, 2005
0.26 (0.15 – 0.48)
6.9
Fung, 2003
0.67 (0.19 – 2.30)
3.0
Ogunyemi, 2003
2.05 (1.04 – 4.03)
6.1
Overall
1.27 (0.99 – 1.63)
100.0
0.1
0.5
1
2
3
5
10
20
Risk ratio
Fig. 4. Forest plot for 16 studies examining the effect of CA (any type) on severe ROP (stage ≥ 3). RR >1 indicates increased risk of ROP
in infants born to mothers with CA.
0
SE of logRR
[21, 27, 28, 36, 41, 43, 45, 56–58, 60, 62–65]. Only 8 studies showed no difference among the groups in terms of GA [42, 43, 47, 50, 52, 58, 61, 66]. On assessment of the methodological quality using the Newcastle-Ottawa scale, there were no studies in the low-quality group (score 0–3). Five studies were included in the group with scores 4–6, and the remaining 22 studies were included in the highquality group (scores 7–9; online suppl. appendix B). ROP was mostly defined as per the International Classification for Retinopathy of Prematurity [30]. Sixteen studies met our criteria for severe ROP (stage ≥3). In one study, the requirement of laser photocoagulation therapy was considered as the diagnostic criteria for ROP, and hence these cases were also included in the severe ROP analysis [60].
0.5
1.0
1.5 –4
–2
0 logRR
2
4
Results
There was a significant positive association between CA and ROP (any stage) [summary RR 1.33 (95% CI 1.14–1.55, I2 = 77%, pheterogeneity < 0.0001); fig. 2]. In sensitivity analyses excluding the most influential studies, the summary RR ranged from 1.30 (95% CI 1.12–1.51) when excluding the Chorioamnionitis as a Risk for ROP: A Meta-Analysis
Fig. 5. Funnel plot with pseudo-95% confidence limits for publica-
tion bias assessment for studies analyzing the effect of CA on severe ROP (stage ≥ 3).
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
193
Study
RR (95% CI)
Clinical Dammann, 2009 Soraisham, 2009 Subtotal
1.62 (1.18 – 2.24) 1.58 (1.22 – 2.05) 1.60 (1.31 – 196)
Clinical and histological Schlapbach, 2010 Fung, 2003 Morales, 1987 Suppiej, 2009 Gray, 1997 Nasef, 2012 Woo, 2012 Fung, 2003 Ogunyemi, 2003 Subtotal
0.50 (0.10 – 2.55) 1.26 (0.40 – 3.99) 0.84 (0.50 – 1.41) 1.42 (0.72 – 2.80) 2.03 (1.15 – 3.58) 0.96 (0.61 – 1.50) 0.98 (0.71 – 1.37) 0.67 (0.19 – 2.30) 1.49 (1.20 – 1.87) 1.17 (0.93 – 1.48)
Histological Perrone, 2012 Van Vliet, 2012 Bordigato, 2011 Wirbelauer, 2011 Lee, 2011 Moscuzza, 2011 Mehta, 2006 Felice, 2005 Polam, 2005 Soraisham, 2013 Seliga, 2012 Chen, 2011 Henderson, 2011 Sato, 2011 Rocha, 2006 Lau, 2005 Subtotal
1.88 (1.16 – 3.05) 4.05 (1.06 – 15.43) 1.88 (0.70 – 5.04) 2.12 (0.39 – 11.64) 1.12 (0.68 – 1.85) 6.15 (0.66 – 57.27) 2.04 (1.07 – 3.92) 3.86 (1.17 – 12.75) 1.80 (1.37 – 2.37) 1.03 (0.91 – 1.17) 1.72 (1.24 – 2.38) 1.00 (0.93 – 1.07) 1.69 (1.14 – 2.52) 1.55 (0.97 – 2.46) 1.22 (0.51 – 2.92) 0.26 (0.15 – 0.48) 1.38 (1.12 – 1.71)
Overall
1.33 (1.14 – 1.55) 0.1
0.5
1
2
3
5
10
20
Risk ratio
Fig. 6. Forest plot for stratified analysis of 27 studies examining the effect of CA (any type) on ROP (any stage) by type of CA. RR >1 indicates increased risk of ROP in infants born to mothers with CA.
study by Polam et al. [27] to 1.39 (95% CI 1.22–1.60) when excluding the study by Lau et al. [21]. There was no indication of publication bias with Begg’s test (p = 0.26; fig. 3). Subgroup Analyses Analysis Based on Severity of ROP There was a marginally significant association between CA and severe ROP [summary RR 1.27 (95% CI 194
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
0.99–1.63, I2 = 74%, pheterogeneity < 0.0001); fig. 4]. In sensitivity analyses excluding the most influential studies, the summary RR ranged from 1.20 (95% CI 0.95–1.53) when excluding the study by Dammann et al. [26] to 1.40 (95% CI 1.16–1.70) when excluding the study by Lau et al. [21]. There was no evidence of publication bias for severe ROP with Begg’s test (p = 0.50; fig. 5). Mitra /Aune /Speer /Saugstad
Study
RR (95% CI)
Weight, %
Nasef, 2012 Lee, 2011 Wirbelauer, 2011 Schlapbach, 2010 Rocha, 2006 Fung, 2003a Fung, 2003b Morales, 1987 Subtotal
0.96 (0.61 – 1.50) 1.12 (0.68 – 1.85) 2.12 (0.39 – 11.64) 0.50 (0.10 – 2.55) 1.22 (0.51 – 2.92) 0.67 (0.19 – 2.30) 1.26 (0.40 – 3.99) 0.84 (0.50 – 1.41) 0.98 (0.77 – 1.26)
30.5 24.6 2.1 2.3 8.2 4.1 4.7 23.6 100.0
Fig. 7. Forest plot for stratified analysis of 8 studies (adjusted for GA) examining the effect of CA on ROP. RR >1 indicates increased
risk of ROP in infants born to mothers with CA. a [42]. b [43].
Analysis Based on Definition of CA The studies were stratified according to the definition of CA into: clinical CA, histological CA as well as clinical and histological CA. We did a stratified analysis on each subgroup (fig. 6). Clinical CA was significantly associated with ROP (any stage) [RR 1.60 (95% CI 1.31–1.96, I2 = 0%)]. Histological CA was also found to be significantly associated with ROP (any stage) [RR 1.38 (95% CI 1.12– 1.71, I2 = 81.7%)]. However, in the subgroup dealing with both clinical and histological CA, no association with ROP (any stage) [RR 1.17 (95% CI 0.93–1.48, I2 = 39.8%)] was found, although there was a nonsignificant increase in ROP with CA in this subgroup. On meta-regression analysis, there was no significant heterogeneity between the subgroups (p = 0.90).
A total of 10,590 preterm infants across 27 studies were evaluated to assess the relation between CA and ROP. The meta-analysis of unadjusted data extracted from all studies showed that the RR of developing ROP was signifi-
cantly (33%) higher in infants exposed to any form of CA, either clinical, histological or both. There was also a borderline significant (27%) increase in the RR of developing severe ROP (stage ≥ 3). However, when adjusted for GA, the meta-analysis of data extracted from 8 such studies showed no significant association between CA and ROP. The included studies were of good quality according to the Newcastle-Ottawa Scale with 22 out of the 27 included studies having a high score of 7–9 on methodological quality assessment. To our knowledge, this meta-analysis is the first of its kind that analyzes the effect of prenatal inflammation in the form of CA on the development of ROP in preterm infants. Perinatal inflammation has been associated with a number of adverse neurological outcomes in preterm infants such as white matter injury, intraventricular hemorrhage, cystic periventricular leukomalacia and cerebral palsy [23, 70–72]. Maternal intrauterine infection and inflammation can induce a production of proinflammatory cytokines in the fetal brain such as tumor necrosis factor-α (TNF-α) and interleukins (IL)-1, IL-6, and IL-8. These cytokines may disrupt the blood-brain barrier, thereby leading to increased plasma protein influx and oligodendroglial damage [73, 74]. On the other hand, increased cytokine levels in the first 72 h of life (IL-6 >357 pg/ml, IL-8 >216 pg/ml, and TNF-α >245 pg/ml) have been linked with development of severe ROP [75]. It has also been observed that maternal systemic inflammation leads to decreased levels of insulin-like growth factor-1, which in turn is associated with an increased risk of developing ROP [68, 76]. Cytokine levels in the vitreous of 19 patients with stage 4 ROP were recently analyzed by Sato et al. [77], who demonstrated that increased levels of vascu-
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
Analysis Based on Studies Adjusted for GA To eliminate the effect of prematurity as a confounder, we did a subgroup analysis to include the studies that showed no significant difference in GA between the CA group and the non-CA group. Eight studies were included in the analysis. A meta-analysis of these 8 studies showed no significant association of CA with ROP [summary RR 0.98 (95% CI 0.77–1.26); fig. 7]. No significant heterogeneity was observed (I2 = 0%, pheterogeneity = 0.89). Discussion
195
lar endothelial growth factor had the highest correlation with vascular activity in ROP eyes compared to other cytokines. Sood et al. [78] studied cytokine levels in extremely low-birth-weight neonates on days 1, 3, 7, 14, and 21 and showed that some cytokines were significantly different across ROP groups, suggesting that perinatal inflammation may be involved in the pathogenesis of ROP. Thus perinatal inflammation in the form of CA does seem to play an important role in the development of ROP in preterm infants. However, establishing this relation on statistical terms becomes a major challenge due to the simultaneous interplay of a number of confounding factors. The major confounding factor in such an analysis is prematurity. Since CA is often associated with preterm birth, the inflammatory response commonly develops in a population already at high risk for brain damage and poor neurologic outcome [79]. It has been shown in a number of studies that extreme prematurity plus multiple hits of perinatal inflammation appear to be involved in ROP etiology and progression [26, 39]. Again, lower GA and birth weight are associated with increased oxygen therapy, which is an independent risk factor for development of ROP [80]. In a recent meta-analysis by Saugstad and Aune [81], it was found that a lower oxygen saturation targeting by pulse oximetry reduced severe ROP by 50%, i.e. from 20.9 to 9.5%. In our analysis, infants in the CA group had lower GAs compared to the non-CA group in 15 out of the 27 studies. This underlies the importance of a subgroup analysis adjusting for the major confounding factor GA. In our study, we found only 8 studies that have adjusted for GA. The meta-analysis of these studies incidentally showed no association between CA and ROP. Hence, prematurity itself along with other factors associated with prematurity, like oxygen therapy, could possibly confound the unadjusted results. This also brings to light the dearth of studies on this issue that adjust for potential confounders, which prevents us from drawing a definitive conclusion. Another factor that was not accounted for in the studies is the influence of genetic variation. It has been mentioned in the study by Dammann et al. [26] that single nucleotide polymorphisms in the genes encoding for increased production of IL-1β, TNF-α and reduced Tolllike receptor 4 signal transduction were associated with a tendency towards reduced likelihood for ROP occurrence. In a recent study comparing whole-genome expression in the first month of life in groups of infants with and without ROP, it was found that more genes were underexpressed than overexpressed in the ROP group [82]. 196
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
This may in part account for the complex pathophysiology of ROP and may explain why ROP progresses to a severe stage in some infants, while in others, despite similar clinical conditions, regresses spontaneously, thus leading to variable results on comparative analysis with a single factor like CA. Substantial heterogeneity was observed across the studies in the magnitude and direction of associations. We explored heterogeneity by grouping the studies by type of CA, but heterogeneity remained substantial among the subgroups, except in the clinical CA group. This may be explained by the very small number of studies in the clinical CA group compared to a much larger number of studies in each of the other subgroups [26, 28]. One important source of heterogeneity may be the substantial variation in the definition of histological CA. While few of the studies mentioned diagnostic criteria for defining histological CA such as Redline’s Criteria, College of American Pathologists Guidelines, Salafia staging system, Blanc’s criteria, criteria from Amniotic Fluid Infection Nosology Committee, Roger’s protocol and Yoon’s criteria, most of them used nonspecified diagnostic methods [36, 39, 47, 58, 62, 67, 69]. Placental microbiology was not taken into account in our analysis due to the insufficient number of studies. Interestingly, it had been shown in one of the studies by Chen et al. [39] that although presence of placental organism or placental inflammation individually did not appear to be associated with an increased risk of ROP, the two risk factors, when co-occurring, were associated with a significantly increased risk of ROP in zone I. This may suggest that underappreciation of the etiologic complexity of ROP pathogenesis in many of the other studies may have led to a lack of association between mere placental inflammation and ROP. There was less discrepancy over the defining criteria for ROP, with almost all the included studies following the International Classification of Retinopathy of Prematurity [30]. ROP stage 3 and above being a distinctive marker of proliferative retinopathy was unanimously classified as severe ROP in most of the studies. Apart from gross variation in the definition of CA, lack of adjustment for probable confounders like GA, birth weight and duration of oxygen therapy most likely contributed to the high degree of heterogeneity observed across studies which again prevented us from drawing any definitive conclusion. Interestingly, we noted no heterogeneity in the subgroup analysis across the 8 studies which adjusted for GA, and these studies showed no association between CA and ROP. This underlies the importance of Mitra /Aune /Speer /Saugstad
taking into account all possible sources of heterogeneity across studies including major confounding factors while interpreting the results of a meta-analysis. This study is a perfect example where exclusion of a major confounder in the form of GA and consequent elimination of heterogeneity vastly changed the results and interpretation of the statistical analysis. An interesting observation to note was that our study did not show any publication bias. In our review, 23 out of the 27 included studies (85.2%) were not primarily designed to assess the effect of CA on ROP. They were designed to assess the effect of placental inflammation on neonatal morbidity and neurodevelopmental outcome, and ROP was included as one of the effects. This might have negated the tendency to avoid publication of studies not showing any relation of CA with ROP, or in other words, the ‘file-drawer effect’ was not present in our analysis.
Conclusion
We conclude that though unadjusted analyses showed significant association of CA with ROP (any stage) as well as with severe ROP (stage ≥ 3), no such association was found on subanalysis of the studies adjusting for GA. Hence, CA cannot be definitively considered as a risk factor for ROP based on current evidence. However, in view of dearth of studies specifically focusing on the relation between CA and ROP, there is a need to conduct further studies on this topic adjusting for potential confounding factors and report results by stage to clarify the association with severe ROP.
Disclosure Statement The authors have no conflict of interest to disclose.
References 1 Terry TL: Extreme prematurity and fibroblastic overgrowth of persistent vascular sheath behind each crystalline lens. Am J Opthalmol 1942;25:203–206. 2 Terry TL: Retrolental fibroplasia in the premature infant. V. Further studies on fibroplastic overgrowth of the persistent tunica vasculosa lentis. Trans Am Ophthalmol Soc 1944;42:383–396. 3 Silverman W: Retrolental Fibroplasia: A Modern Parable. New York, Grune and Stratton, 1980. 4 Chen M, Çitil A, McCabe F, et al: Infection, oxygen, and immaturity: interacting risk factors for retinopathy of prematurity. Neonatology 2011;99:125–132. 5 Raghuveer TS, Bloom BT: A paradigm shift in the prevention of retinopathy of prematurity. Neonatology 2011;100:116–129. 6 Darlow BA, Hutchinson JL, HendersonSmart DJ, Donoghue DA, Simpson JM, Evans NJ: Prenatal risk factors for severe retinopathy of prematurity among very preterm infants of the Australian and New Zealand Neonatal Network. Pediatrics 2005;115:990–996. 7 Chen J, Smith LE: Retinopathy of prematurity. Angiogenesis 2007;10:133–140. 8 Holmstrom G, Broberger U, Thomassen P: Neonatal risk factors for retinopathy of prematurity – a population-based study. Acta Ophthalmol Scand 1998;76:204–220. 9 Simhan HN, Caritis SN: Prevention of preterm delivery. N Engl J Med 2007; 357: 477– 487.
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
10 Rivera JC, Sapieha P, Joyal JS, Duhamel F, et al: Understanding retinopathy of prematurity: update on pathogenesis. Neonatology 2011;100:343–353. 11 Romero R, Gomez R, Chaiworapongsa T, Conoscenti G, Kim JC, Kim YM: The role of infection in preterm labour and delivery. Paediatr Perinat Epidemiol 2001;15(suppl 2):41–56. 12 Lahra MM, Jeffry HE: A fetal response to chorioamnionitis is associated with early survival after preterm birth. Am J Obstet Gynecol 2004;190:147–151. 13 Goldenberg RL, Hauth JC, Andrews WW: Intrauterine infection and preterm delivery. N Engl J Med 2000;342:1500–1507. 14 Mittal M, Dhanireddy R, Higgins RD: Candida sepsis and association with retinopathy of prematurity. Pediatrics 1998;101:654–657. 15 Noyola DE, Bohra L, Paysse EA, Fernandez M, Coats DK: Association of candidemia and retinopathy of prematurity in very low birthweight infants. Ophthalmology 2002;109:80– 84. 16 Klinger G, Levy I, Sirota L, Boyko V, LernerGeva L, Reichman B: Outcome of early-onset sepsis in a national cohort of very low birth weight infants. Pediatrics 2010;125:736–740. 17 Ahn HM, Park EA, Cho SJ, Kim YJ, Park HS: The association of histological chorioamnionitis and antenatal steroids on neonatal outcome in preterm infants born at less than thirty-four weeks’ gestation. Neonatology 2012; 102:259–264. 18 Kramer BW: Chorioamnionitis – new ideas from experimental models. Neonatology 2011;99:320–325.
19 Wu YW, Colford JM Jr: Chorioamnionitis as a risk factor for cerebral palsy: a meta-analysis. JAMA 2000;284:1417–1424. 20 Ragouilliaux CJ, Keeney SE, Hawkins HK, Rowen JL: Maternal factors in extremely low birth weight infants who develop spontaneous intestinal perforation. Pediatrics 2007; 120:1458–1464. 21 Lau J, Magee F, Qiu Z, Hoube J, Von Dadelszen P, Lee SK: Chorioamnionitis with a fetal inflammatory response is associated with higher neonatal mortality, morbidity, and resource use than chorioamnionitis displaying a maternal inflammatory response only. Am J Obstet Gynecol 2005;193:708–713. 22 Watterberg KL, Demers LM, Scott SM, Murphy S: Chorioamnionitis and early lung inflammation in infants in whom bronchopulmonary dysplasia develops. Pediatrics 1996; 97:210–215. 23 Dammann O, Leviton A, Bartels DB, Dammann CEL: Lung and brain damage in preterm newborns. Are they related? How? Why? Biol Neonate 2004;85:305–313. 24 Thomas W, Speer CP: Chorioamnionitis: important risk factor or innocent bystander for neonatal outcome? Neonatology 2011; 99: 177–187. 25 Speer CP: Neonatal respiratory distress syndrome: an inflammatory disease? Neonatology 2011;99:316–319. 26 Dammann O, Brinkhaus MJ, Bartels DB, et al: Immaturity, perinatal inflammation, and retinopathy of prematurity: a multi-hit hypothesis. Early Hum Dev 2009;85:325–329.
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
197
27 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. 28 Soraisham AS, Singhal N, McMillan DD, Sauve RS, Lee SK: A multicenter study on the clinical outcome of chorioamnionitis in preterm infants. Am J Obstet Gynecol 2009;200: 371–376. 29 Yoon BH, Romero R, Kim CJ, et al: 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. 30 An international classification of retinopathy of prematurity. Pediatrics 1984;74:127–133. 31 Wells G, Shea B, O’Connell B, et al: The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Non-Randomized Studies in Metaanalysis. Ottawa, University of Ottawa, 2009. 32 Hartling L, Liang Y, Lacaze-Masmonteil T: Chorioamnionitis as a risk factor for bronchopulmonary dysplasia: a systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed 2012;97:8–17. 33 DerSimonian R, Laird N: Meta-analysis in clinical trials. Control Clin Trials 1986;7:177– 188. 34 Higgins JP, Thompson SG: Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21:1539–1558. 35 Begg CB, Mazumdar M: Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088–1101. 36 Bordigato MA, Piva D, Gangi IMD, Giordano G, Chiandetti L, Filippone M: Asymmetric dimethylarginine in ELBW newborns exposed to chorioamnionitis. Early Hum Dev 2011;87: 143–145. 37 Langston C, Kaplan C, Macpherson T, et al: Practice guideline for examination of the placenta: developed by the Placental Pathology Practice Guideline Development Task Force of the College of American Pathologists. Arch Pathol Lab Med 1997;121:449–476. 38 Redline RW: Inflammatory responses in the placenta and umbilical cord. Semin Fetal Neonatal Med 2006;11:296–301. 39 Chen ML, Allred EN, Hecht JL, et al: Placenta microbiology and histology and the risk for severe retinopathy of prematurity. Invest Ophthalmol Vis Sci 2011;52:7052–7058. 40 Driscoll SG, Langston C: College of American Pathologists Conference XIX on the Examination of the Placenta: report of the Working Group on Methods for Placental Examination. Arch Pathol Lab Med 1991;115:704–708. 41 Felice CD, Toti P, Parrini S, et al: Histologic chorioamnionitis and severity of illness in very low birth weight newborns. Pediatr Crit Care Med 2005;6:298–302.
198
42 Fung G, Bawden K, Chow P, Yu V: Long-term outcome of extremely preterm infants following chorioamnionitis. HK J Pediatr (new series) 2003;8:87–92. 43 Fung G, Bawden K, Chow P, Yu V: Chorioamnionitis and outcome in extremely preterm infants. Ann Acad Med Singapore 2003; 32:305–310. 44 Gray PH, Hurley TM, Rogers YM, et al: Survival and neonatal and neurodevelopmental outcome of 24–29 week gestation infants according to primary cause of preterm delivery. Aust NZ J Obstet Gynaecol 1997;31:161. 45 Hendson L, Russell L, Robertson CMT, et al: Neonatal and neurodevelopmental outcomes of very low birth weight infants with histologic chorioamnionitis. J Pediatr 2011; 158: 397–402. 46 Reese AB, Ling MJ, Owens WC: A classification of retrolental fibroplasias. Am J Opthalmol 1953;36:1333. 47 Lee HJ, Kim EK, Kim HS, Choi CW, Kim BII, Choi JH: Chorioamnionitis, respiratory distress syndrome and bronchopulmonary dysplasia in extremely low birth weight infants. J Perinatol 2011;31:166–170. 48 Salafia CM, Weigl C, Silberman L: The prevalence and distribution of acute placental inflammation in uncomplicated term pregnancies. Obstet Gynecol 1989;73:383–389. 49 Mehta R, Nanjundaswamy S, Shen-Schwarz S, Petrova A: Neonatal morbidity and placental pathology. Indian J Pediatr 2006;73:25–28. 50 Morales WJ: The effect of chorioamnionitis on the developmental outcome of preterm infants at one year. Obstet Gynaecol 1987; 70: 183–186. 51 Moscuzza F, Belcari F, Nardini V, et al: Correlation between placental histopathology and fetal/neonatal outcome: chorioamnionitis and funisitis are associated to intraventricular haemorrage and retinopathy of prematurity in preterm newborns. Gynecol Endocrinol 2011;27:319–323. 52 Nasef N, Shabaan AE, Schurr P, et al: Effect of clinical and histological chorioamnionitis on the outcome of preterm infants. Am J Perinatol 2013;30:59–68. 53 International Committee for the Classification of Retinopathy of Prematurity: The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005;123:991–999. 54 Newton ER: Chorioamnionitis and intraamniotic infection. Clin Obstet Gynecol 1993;36: 795–808. 55 Redline RW, Faye-Petersen O, Heller D, Qureshi F, Savell V, Vogler C, Society for Pediatric Pathology, Perinatal Section, Amniotic Fluid Infection Nosology Committee: Amniotic infection syndrome: nosology and reproducibility of placental reaction patterns. Pediatr Dev Pathol 2003;6:435–448.
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
56 Ogunyemi D, Murillo M, Jackson U, Hunter N, Alperson B: The relationship between placental histopathology findings and perinatal outcome in preterm infants. J Mat Fetal Neonatal Med 2003;13:102–109. 57 Perrone S, Toti P, Toti MS, et al: Perinatal outcome and placental histological characteristics: a single-center study. J Matern Fetal Neonatal Med 2012;25(suppl 1):110–113. 58 Rocha G, Proença E, Quintas C, Rodrigues T, Guimarães H: Corioamnionite e morbidade neonatal. Acta Med Port 2006;19:207–212. 59 Blanc WA: Pathology of the placenta, membranes, and umbilical cord in bacterial, fungal, and viral infections in man. Monogr Pathol 1981;22:67–132. 60 Sato M, Nishimaki S, Yokota S, et al: Severity of chorioamnionitis and neonatal outcome. J Obstet Gynaecol Res 2011;37:1313–1319. 61 Schlapbach LJ, Ersch J, Adams M, Bernet V, Bucher HU, Latal B: Impact of chorioamnionitis and preeclampsia on neurodevelopmental outcome in preterm infants below 32 weeks gestational age. Acta Paediatr 2010;99: 1504–1509. 62 Seliga-Siwecka JP, Kornacka MK: Neonatal outcome of preterm infants born to mothers with abnormal genital tract colonisation and chorioamnionitis: a cohort study. Early Hum Dev 2013;89:271–275. 63 Soraisham AS, Trevenen C, Wood S, Singhal N, Sauve R: Histological chorioamnionitis and neurodevelopmental outcome in preterm infants. J Perinatol 2013;33:70–75. 64 Suppiej A, Franzoi M, Vedovato S, Marucco A, Chiarelli S, Zanardo V: Neurodevelopmental outcome in preterm histological chorioamnionitis. Early Hum Dev 2009; 85: 187– 189. 65 van Vliet EOG, de Kieviet JF, van der Voorn JP, et al: Placental pathology and long-term neurodevelopment of very preterm infants. Am J Obstet Gynecol 2012;206:489.e1–7. 66 Wirbelauer J, Thomas W, Speer CP: Response of leukocytes and nucleated red blood cells in very low-birth weight preterm infants after exposure to intrauterine inflammation. J Mat Fetal Neonatal Med 2011;24:348–353. 67 Rogers BB, Alexander JM, Head J, McIntire D, Leveno KJ: Umbilical vein interleukin-6 levels correlate with the severity of placental inflammation and gestational age. Hum Pathol 2002;33:335–340. 68 Woo SJ, Park KH, Jung HJ, et al: Effects of maternal and placental inflammation on retinopathy of prematurity. Graefes Arch Clin Exp Ophthalmol 2012;250:915–923. 69 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. 70 Murata Y, Itakura A, Matsuzawa K, Okumura A, Wakai K, Mizutani S: Possible antenatal and perinatal related factors in development of cystic periventricular leukomalacia. Brain Dev 2005;27:17–21.
Mitra /Aune /Speer /Saugstad
71 Wu YW, Colford JM Jr: Chorioamnionitis as a risk factor for cerebral palsy: a meta-analysis. JAMA 2000;284:1417–1424. 72 Shatrov JG, Birch SC, Lam LT, Quinlivan JA, McIntyre S, Mendz GL: Chorioamnionitis and cerebral palsy: a meta-analysis. Obstet Gynecol 2010;116:387–392. 73 Dammann O, Leviton A: Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 1997; 42: 1–8. 74 Stolp HB, Dziegielewska KM: Review: role of developmental inflammation and bloodbrain barrier dysfunction in neurodevelopmental and neurodegenerative diseases. Neuropathol Appl Neurobiol 2009;35:132–146. 75 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.
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
76 Hellström A, Peruzzi C, Ju M, et al: Low IGF1 suppresses VEGF-survival signaling in retinal endothelial cells: direct correlation with clinical retinopathy of prematurity. Proc Natl Acad Sci 2001;98:5804–5808. 77 Sato T, Kusaka S, Shimojo H, Fujikado T: Simultaneous analyses of vitreous levels of 27 cytokines in eyes with retinopathy of prematurity. Ophthalmology 2009;116:2165–2169. 78 Sood BG, Madan A, Saha S, et al: On behalf of the NICHD Neonatal Research Network. Perinatal systemic inflammatory response syndrome and retinopathy of prematurity. Pediatr Res 2010;67:394–400.
79 Reiman M, Kujari H, Maunu J, et al, PIPARI Study Group: Does placental inflammation relate to brain lesions and volume in preterm infants? J Pediatr 2008; 152: 642–647, 647.e1– 2. 80 Fortes Filho JB, Eckert GU, Procianoy L, Barros CK, Procianoy RS: Incidence and risk factors for retinopathy of prematurity in very low and in extremely low birth weight infants in a unit-based approach in southern Brazil. Eye 2009;23:25–30. 81 Saugstad OD, Aune D: In search of the optimal oxygen saturation for extremely low birth weight infants: a systematic review and metaanalysis. Neonatology 2011;100:1–8. 82 Pietrzyk JJ, Kwinta P, Bik-Multanowski M, et al: New insight into the pathogenesis of retinopathy of prematurity: assessment of wholegenome expression. Pediatr Res 2013;73:476– 483.
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
199
Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.
Received: October 8, 2013 Accepted after revision: November 26, 2013 Published online: January 24, 2014
Chorioamnionitis as a Risk Factor for Retinopathy of Prematurity: A Systematic Review and Meta-Analysis Souvik Mitra a, e Dagfinn Aune b, c Christian P. Speer d Ola Didrik Saugstad a a Department of Paediatric Research, Oslo University Hospital, University of Oslo, Oslo, and b Department of Public Health, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway; c Department of Epidemiology and Biostatistics, Imperial College London, London, UK; d University Children’s Hospital, University of Würzburg, Würzburg, Germany; e Division of Neonatology, Department of Paediatrics, McMaster University Medical Centre, Hamilton, Ont., Canada
Key Words Chorioamnionitis · Retinopathy of prematurity · Systematic review · Meta-analysis
Abstract Background: The role of chorioamnionitis (CA) in the development of retinopathy of prematurity (ROP) has not been well established. Objective: To conduct a systematic review and meta-analysis of the association between CA and ROP in preterm infants. Data Sources: The authors searched MEDLINE, Embase, CINAHL, Cochrane Central Register of Controlled Trials and PubMed, reviewed reference lists of relevant articles, abstracts and conference proceedings (Society for Pediatric Research, European Society for Paediatric Research 1990–2012), sought results of unpublished trials, and contacted the primary authors of relevant studies. Study Selection: Studies were included if they had a comparison group, examined preterm infants, and reported primary data that could be used to measure the association between exposure to CA and the development of ROP. Data Extraction: Two reviewers independently screened the search results, applied inclusion criteria and assessed methodological quality using the Newcastle-Ottawa Scale. One
© 2014 S. Karger AG, Basel 1661–7800/14/1053–0189$39.50/0 E-Mail [email protected] www.karger.com/neo
reviewer extracted data and a second reviewer checked data extraction. Summary relative risks (RRs) were calculated using a random effects model. Data Synthesis: We identified 1,249 potentially relevant studies from the electronic databases. Twenty-seven studies involving 10,590 preterm neonates with 2,562 cases of ROP were included. Taking into account all included studies without adjusting for gestational age (GA), CA was significantly associated with ROP (any stage) [summary RR 1.33 (95% CI 1.14–1.55, I2 = 77%, pheterogeneity < 0.0001)], and a borderline significant association was observed for severe ROP (stage ≥3) [summary RR 1.27 (95% CI 0.99–1.63, I2 = 74%, pheterogeneity < 0.0001)]. There was no publication bias with Begg’s test. However, subgroup analysis of studies adjusting for GA showed no significant association on CA with ROP [summary RR 0.98 (95% CI 0.77–1.26, I2 = 0%, pheterogeneity = 0.89)]. Conclusion: Unadjusted analyses showed that CA was significantly associated with ROP (any stage) as well as with severe ROP (stage ≥ 3). However, the association disappeared on analysis of studies adjusting for GA. Hence, CA cannot be definitively considered as a risk factor for ROP, and further studies should adjust for potential confounding factors and report results by stage to clarify the association with severe ROP. © 2014 S. Karger AG, Basel
Souvik Mitra, MD Division of Neonatology, Department of Pediatrics McMaster University Medical Centre HSC-4F, 1280 Main Street West, Hamilton, ON L8S 4K1 (Canada) E-mail [email protected] or souvik4sphs @ yahoo.com
Introduction
Retinopathy of prematurity (ROP) is a retinal vascular disease of preterm infants, first described in 1942 [1]. ROP was initially connected with premature birth in the 1940s and slightly later to oxygen supplementation in these premature infants [2]. It is a major cause of blindness in children in the developing and developed world despite current surgical and laser treatment in the late stage of the disease [3]. Given the high prevalence and sequelae of ROP, effective prevention and timely treatment are essential for the preservation of lifelong vision in these premature neonates. A number of studies over the years have shown that low birth weight, early gestational age (GA) and oxygen use are the major risk factors for ROP [4–8]. However, primary prevention of ROP through elimination of risk factors like prematurity is practically not feasible [7, 9]. Therefore, there is a pressing need to identify contributing factors based on the prenatal origin of ROP, because preventive factors for ROP targeting the mothers and their fetuses may offer new opportunities for successful prevention of ROP [10]. Preterm labor and delivery, the leading identifiable causes of ROP, have been shown to be associated with maternal infections like chorioamnionitis (CA) and systemic infections [11]. Histological CA is defined by maternal inflammatory response with neutrophilic infiltration of the membranes and/or chorionic plate, with or without a fetal inflammatory co-response. On the other hand, clinical CA is defined by the attending physician in presence of maternal fever along with other suggestive clinical features. The incidence of CA has been found to increase with decreasing GA [12, 13]. It has also been shown in observational studies conducted in cohorts of extremely low-birth-weight infants that early-onset sepsis and systemic candidiasis are independently associated with increased severity of ROP and the need for laser treatment [14–16]. These findings suggest that perinatal inflammation and infection may have a deleterious effect on the developing blood vessels in the retina, making them vulnerable to the development of ROP. In a number of recently published studies, histological CA has in fact been shown to be an important risk factor for the development of early-onset sepsis as well as subsequent neonatal disorders of the lung, intestines and brain including white matter injuries and intraventricular hemorrhage [17–25]. A number of studies have tried to evaluate the relationship of CA and ROP. Intrauterine inflammation may be responsible for an adverse ‘first hit’ to the immature 190
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
fetal organism which is followed by various postnatal events (‘multiple hits’) aggravating the risk of ROP in preterm infants [26, 27]. However, other investigators have found no significant association of CA with ROP [21, 28]. This disagreement may be due to the relatively small sample size, lack of control for known risk factors (GA, birth weight and hyperoxia/hypoxia), wide variation in outcome measures [e.g. any ROP, severe (stage 3 and above) ROP, prethreshold and threshold ROP], and the lack of consensus regarding the definition of CA in previous studies. In an effort to resolve the discrepancy observed across studies, we conducted a systematic review and meta-analysis to synthesize the literature that measures the association of CA with ROP in premature infants. Methods Sources We searched the Cochrane Central Register of Controlled Trials (The Cochrane Library issue 1 of 12, January 2013) using the terms RETINOPATHY OF PREMATURITY AND CHORIOAMNIONITIS. We searched MEDLINE (January 1966 to January 2013) and PubMed (2013) using exploded MeSH terms RETINOPATHY OF PREMATURITY, CHORIOAMNIONITIS; ROP, CHORIOAMNIONITIS; CHORIOAMNIONITIS, NEURODEVELOPMENTAL OUTCOME; CHORIOAMNIONITIS, LONG TERM EFFECTS; PLACENTAL INFLAMMATION; PLACENTA HISTOLOGY, RETINOPATHY OF PREMATURITY and all possible combinations thereof in all languages (or without language restrictions). We searched EMBASE (January 1974 to January 2013) using the terms RETINOPATHY OF PREMATURITY with CHORIOAMNIONITIS. We searched the Science Citation Index (September 1984 to January 2013) for trials which cited the trials identified by our searches as MEDLINE and Embase. We sought registered details of selected trials in the U.S. National Institutes of Health resource Clinicaltrials.gov. We obtained information by personal communication, reviewing the reference lists of relevant articles, abstracts and conference proceedings (Society for Pediatric Research, European Society for Paediatric Research 1990–2012) and seeking results of unpublished trials. Study Selection Studies were included if they had a comparison group, examined preterm infants, and reported primary data that could be used to measure the association between exposure to CA and the development of ROP. We defined CA cases as clinical, histological or both clinical and histological. Clinical CA was defined as documented clinical signs and symptoms suggestive of CA. Histological CA was defined based on previously reported definitions as an acute inflammatory change in any of the tissue samples (amnion, chorion deciduas, umbilical cord and chorionic plate) [29]. ROP was primarily defined according to the International Classification of Retinopathy of Prematurity and severe ROP as stages 3, 4 or 5 [30]. Studies where data on ROP were collected from the neonatal
Mitra /Aune /Speer /Saugstad
Records identified through electronic database searching after removing duplicates (n = 1,249) Excluded after title review (n = 226) Abstracts reviewed (n = 1,023) Excluded after abstract review (n = 955) Full-text articles assessed for eligibility (n = 68) Additional records obtained through personal communication with authors (n = 3) Studies included in the systematic review and meta-analysis (n = 27)
Full-text articles excluded (n = 44) Exposure not defined (n = 29) Outcome not included (n = 9) Outcome not defined (n = 3) Exposure and outcome defined but detailed statistics not included in full text (n = 3)
Fig. 1. Study flow diagram.
medical records were also included. To identify relevant studies, 2 reviewers independently screened the results of the searches and applied inclusion criteria using a structured form. Discrepancies were resolved through discussion or in consultation with a third reviewer. Quality Assessment We assessed the methodological quality using the NewcastleOttawa Scale for case-control or cohort studies as appropriate [31, 32]. The scale assesses for potential selection bias, comparability of cohorts or cases and controls and ascertainment of exposure (cohort studies) or outcome (case-control studies). Two reviewers independently assessed the methodological quality of each study and any discrepancy was resolved through consultation with a third reviewer.
Statistical Analysis We calculated summary relative risks (RRs) comparing risk of ROP among preterm infants born to women with CA with those without CA using a random effects model which accounts for within and between study heterogeneity [33]. The average of the natural logarithm of the RRs was estimated, and the RRs from each study were weighted by the inverse of the variance. A two-tailed p < 0.05 was considered statistically significant. Heterogeneity between studies was evaluated using Q and I2 statistics [34]. I2 is the amount of total variation that is due to variation between studies. I2 values of approximately 25, 50, and 75% indicate low, moderate and high heterogeneity, respectively. Publication bias was assessed using Begg-Mazumdar’s test [35]. We conducted sensitivity analyses excluding one study at a time to test whether the results were robust. Subgroup analyses were done based on the definition of CA, severity of ROP and studies adjusting for GA.
Data Extraction One reviewer extracted data from relevant studies using a predetermined data extraction form. A second reviewer checked the data extraction for accuracy and completeness. All data included in the meta-analysis were monitored by the statistician. Discrepancies were resolved by consulting the primary report. Data extracted from each study included citation information, language of publication, country where the research was conducted, objectives, study design, definitions of histological CA and ROP, inclusion/ exclusion criteria, patient characteristics and results. Raw data, where available, were extracted from the results to analyze the effect of CA on ROP.
Description of the Studies We identified 1,249 potentially relevant studies from the electronic databases. Twenty-seven studies including 10,590 preterm infants met our study requirements (fig. 1). The clinical profile of the infants and the methodological characteristics of the studies have been summarized in online supplementary appendix A (see www.karger.com/doi/10.1159/000357556 for all online suppl. material) [36–69]. Studies were published between 1997 and 2013 (median year 2011). All studies except one were in English [58]. While all the studies provided data to measure the association between ROP and CA, the objec-
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
191
Study
RR (95% CI)
Soraisham, 2013 Nasef, 2012 Perrone, 2012 Seliga, 2012 Van Vliet, 2012 Woo, 2012 Bordigato, 2011 Chen, 2011 Henderson, 2011 Lee, 2011 Moscuzza, 2011 Sato, 2011 Wirbelauer, 2011 Schlapbach, 2010 Dammann, 2009 Soraisham, 2009 Suppiej, 2009 Mehta, 2006 Rocha, 2006 Felice, 2005 Lau, 2005 Polam, 2005 Fung, 2003 Fung, 2003 Ogunyemi, 2003 Gray, 1997 Morales, 1987 Overall
1.03 (0.91 – 1.17) 0.96 (0.61 – 1.50) 1.88 (1.16 – 3.05) 1.72 (1.24 – 2.38) 4.05 (1.06 – 15.43) 0.98 (0.71 – 1.37) 1.88 (0.70 – 5.04) 1.00 (0.93 – 1.07) 1.69 (1.14 – 2.52) 1.12 (0.68 – 1.85) 6.15 (0.66 – 57.27) 1.55 (0.97 – 2.46) 2.12 (0.39 – 11.64) 0.50 (0.10 – 2.55) 1.62 (1.18 – 2.24) 1.58 (1.22 – 2.05) 1.42 (0.72 – 2.80) 2.04 (1.07 – 3.92) 1.22 (0.51 – 2.92) 3.86 (1.17 – 12.75) 0.26 (0.15 – 0.48) 1.80 (1.37 – 2.37) 0.67 (0.19 – 2.30) 1.26 (0.40 – 3.99) 1.49 (1.20 – 1.87) 2.03 (1.15 – 3.58) 0.84 (0.50 – 1.41) 1.33 (1.14 – 1.55) 0.1
0.5
1
2
3
5
10
20
Risk ratio
Fig. 2. Forest plot for 27 studies examining the effect of CA (any type) on ROP (any stage). RR >1 indicates increased risk of ROP in
infants born to mothers with CA.
SE of logRR
0
0.5
1.0
1.5 –2
–1
0 logRR
1
2
Fig. 3. Funnel plot with pseudo-95% confidence limits for publica-
tion bias assessment for studies analyzing the effect of CA on ROP (any stage). SE = Standard error.
192
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
tives of the studies varied. The risk factors for developing ROP in preterm infants were studied in 4 studies [26, 39, 51, 68]. The rest of the studies evaluated the occurrence of ROP as one of the outcomes of clinical or histological CA. Additional information was collected by personal communication with the authors from 3 studies [39, 51, 52]. Eight studies defined both clinical as well as histological CA [42– 44, 50, 52, 56, 61, 68]. Two studies dealt with only clinical CA [26, 28], while the rest of the studies only defined histological CA. In the studies that defined clinical CA, the definition was not specified in one of the studies [26]. In the rest of the studies defining clinical CA, the definitions were by and large similar in accordance with the criteria suggested by Gibbs et al. [69]. Histological CA was not defined by 2 studies [49, 50]. The criteria used for histological CA definition varied in the studies where it was mentioned [29, 37, 38, 48, 55, 59, 67]. Redline’s criteria were the most commonly used criteria to define histological CA [38]. In all except 2 studies, the mean or median GA was less than 30 weeks [21, 58]. In 19 of 27 studies, the patient characteristics were clearly differentiated between the CA and non-CA groups. In 15 out of these 19 studies, there was a significantly higher GA at birth among the non-CA group compared to the CA group
Mitra /Aune /Speer /Saugstad
Study
RR (95% CI)
Weight, %
Soraisham, 2013
1.40 (1.00 – 1.96)
9.2
Nasef, 2012
1.34 (0.50 – 3.59)
4.1
Seliga, 2012
1.72 (1.24 – 2.38)
9.3
Woo, 2012
1.00 (0.63 – 1.57)
8.1
Bordigato, 2011
3.20 (0.14 – 72.62)
0.6
Chen, 2011
0.96 (0.80 – 1.15)
10.5
Henderson, 2011
1.69 (1.14 – 2.52)
8.7
Lee, 2011
1.12 (0.68 – 1.85)
7.6
Sato, 2011
1.55 (0.97 – 2.46)
8.0
Wirbelauer, 2011
2.12 (0.39 – 11.64)
1.8
Schlapbach, 2010
0.67 (0.14 – 3.12)
2.1
Dammann, 2009
4.55 (1.69 – 12.21)
4.0
Soraisham, 2009
1.58 (1.22 – 2.05)
9.9
Lau, 2005
0.26 (0.15 – 0.48)
6.9
Fung, 2003
0.67 (0.19 – 2.30)
3.0
Ogunyemi, 2003
2.05 (1.04 – 4.03)
6.1
Overall
1.27 (0.99 – 1.63)
100.0
0.1
0.5
1
2
3
5
10
20
Risk ratio
Fig. 4. Forest plot for 16 studies examining the effect of CA (any type) on severe ROP (stage ≥ 3). RR >1 indicates increased risk of ROP
in infants born to mothers with CA.
0
SE of logRR
[21, 27, 28, 36, 41, 43, 45, 56–58, 60, 62–65]. Only 8 studies showed no difference among the groups in terms of GA [42, 43, 47, 50, 52, 58, 61, 66]. On assessment of the methodological quality using the Newcastle-Ottawa scale, there were no studies in the low-quality group (score 0–3). Five studies were included in the group with scores 4–6, and the remaining 22 studies were included in the highquality group (scores 7–9; online suppl. appendix B). ROP was mostly defined as per the International Classification for Retinopathy of Prematurity [30]. Sixteen studies met our criteria for severe ROP (stage ≥3). In one study, the requirement of laser photocoagulation therapy was considered as the diagnostic criteria for ROP, and hence these cases were also included in the severe ROP analysis [60].
0.5
1.0
1.5 –4
–2
0 logRR
2
4
Results
There was a significant positive association between CA and ROP (any stage) [summary RR 1.33 (95% CI 1.14–1.55, I2 = 77%, pheterogeneity < 0.0001); fig. 2]. In sensitivity analyses excluding the most influential studies, the summary RR ranged from 1.30 (95% CI 1.12–1.51) when excluding the Chorioamnionitis as a Risk for ROP: A Meta-Analysis
Fig. 5. Funnel plot with pseudo-95% confidence limits for publica-
tion bias assessment for studies analyzing the effect of CA on severe ROP (stage ≥ 3).
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
193
Study
RR (95% CI)
Clinical Dammann, 2009 Soraisham, 2009 Subtotal
1.62 (1.18 – 2.24) 1.58 (1.22 – 2.05) 1.60 (1.31 – 196)
Clinical and histological Schlapbach, 2010 Fung, 2003 Morales, 1987 Suppiej, 2009 Gray, 1997 Nasef, 2012 Woo, 2012 Fung, 2003 Ogunyemi, 2003 Subtotal
0.50 (0.10 – 2.55) 1.26 (0.40 – 3.99) 0.84 (0.50 – 1.41) 1.42 (0.72 – 2.80) 2.03 (1.15 – 3.58) 0.96 (0.61 – 1.50) 0.98 (0.71 – 1.37) 0.67 (0.19 – 2.30) 1.49 (1.20 – 1.87) 1.17 (0.93 – 1.48)
Histological Perrone, 2012 Van Vliet, 2012 Bordigato, 2011 Wirbelauer, 2011 Lee, 2011 Moscuzza, 2011 Mehta, 2006 Felice, 2005 Polam, 2005 Soraisham, 2013 Seliga, 2012 Chen, 2011 Henderson, 2011 Sato, 2011 Rocha, 2006 Lau, 2005 Subtotal
1.88 (1.16 – 3.05) 4.05 (1.06 – 15.43) 1.88 (0.70 – 5.04) 2.12 (0.39 – 11.64) 1.12 (0.68 – 1.85) 6.15 (0.66 – 57.27) 2.04 (1.07 – 3.92) 3.86 (1.17 – 12.75) 1.80 (1.37 – 2.37) 1.03 (0.91 – 1.17) 1.72 (1.24 – 2.38) 1.00 (0.93 – 1.07) 1.69 (1.14 – 2.52) 1.55 (0.97 – 2.46) 1.22 (0.51 – 2.92) 0.26 (0.15 – 0.48) 1.38 (1.12 – 1.71)
Overall
1.33 (1.14 – 1.55) 0.1
0.5
1
2
3
5
10
20
Risk ratio
Fig. 6. Forest plot for stratified analysis of 27 studies examining the effect of CA (any type) on ROP (any stage) by type of CA. RR >1 indicates increased risk of ROP in infants born to mothers with CA.
study by Polam et al. [27] to 1.39 (95% CI 1.22–1.60) when excluding the study by Lau et al. [21]. There was no indication of publication bias with Begg’s test (p = 0.26; fig. 3). Subgroup Analyses Analysis Based on Severity of ROP There was a marginally significant association between CA and severe ROP [summary RR 1.27 (95% CI 194
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
0.99–1.63, I2 = 74%, pheterogeneity < 0.0001); fig. 4]. In sensitivity analyses excluding the most influential studies, the summary RR ranged from 1.20 (95% CI 0.95–1.53) when excluding the study by Dammann et al. [26] to 1.40 (95% CI 1.16–1.70) when excluding the study by Lau et al. [21]. There was no evidence of publication bias for severe ROP with Begg’s test (p = 0.50; fig. 5). Mitra /Aune /Speer /Saugstad
Study
RR (95% CI)
Weight, %
Nasef, 2012 Lee, 2011 Wirbelauer, 2011 Schlapbach, 2010 Rocha, 2006 Fung, 2003a Fung, 2003b Morales, 1987 Subtotal
0.96 (0.61 – 1.50) 1.12 (0.68 – 1.85) 2.12 (0.39 – 11.64) 0.50 (0.10 – 2.55) 1.22 (0.51 – 2.92) 0.67 (0.19 – 2.30) 1.26 (0.40 – 3.99) 0.84 (0.50 – 1.41) 0.98 (0.77 – 1.26)
30.5 24.6 2.1 2.3 8.2 4.1 4.7 23.6 100.0
Fig. 7. Forest plot for stratified analysis of 8 studies (adjusted for GA) examining the effect of CA on ROP. RR >1 indicates increased
risk of ROP in infants born to mothers with CA. a [42]. b [43].
Analysis Based on Definition of CA The studies were stratified according to the definition of CA into: clinical CA, histological CA as well as clinical and histological CA. We did a stratified analysis on each subgroup (fig. 6). Clinical CA was significantly associated with ROP (any stage) [RR 1.60 (95% CI 1.31–1.96, I2 = 0%)]. Histological CA was also found to be significantly associated with ROP (any stage) [RR 1.38 (95% CI 1.12– 1.71, I2 = 81.7%)]. However, in the subgroup dealing with both clinical and histological CA, no association with ROP (any stage) [RR 1.17 (95% CI 0.93–1.48, I2 = 39.8%)] was found, although there was a nonsignificant increase in ROP with CA in this subgroup. On meta-regression analysis, there was no significant heterogeneity between the subgroups (p = 0.90).
A total of 10,590 preterm infants across 27 studies were evaluated to assess the relation between CA and ROP. The meta-analysis of unadjusted data extracted from all studies showed that the RR of developing ROP was signifi-
cantly (33%) higher in infants exposed to any form of CA, either clinical, histological or both. There was also a borderline significant (27%) increase in the RR of developing severe ROP (stage ≥ 3). However, when adjusted for GA, the meta-analysis of data extracted from 8 such studies showed no significant association between CA and ROP. The included studies were of good quality according to the Newcastle-Ottawa Scale with 22 out of the 27 included studies having a high score of 7–9 on methodological quality assessment. To our knowledge, this meta-analysis is the first of its kind that analyzes the effect of prenatal inflammation in the form of CA on the development of ROP in preterm infants. Perinatal inflammation has been associated with a number of adverse neurological outcomes in preterm infants such as white matter injury, intraventricular hemorrhage, cystic periventricular leukomalacia and cerebral palsy [23, 70–72]. Maternal intrauterine infection and inflammation can induce a production of proinflammatory cytokines in the fetal brain such as tumor necrosis factor-α (TNF-α) and interleukins (IL)-1, IL-6, and IL-8. These cytokines may disrupt the blood-brain barrier, thereby leading to increased plasma protein influx and oligodendroglial damage [73, 74]. On the other hand, increased cytokine levels in the first 72 h of life (IL-6 >357 pg/ml, IL-8 >216 pg/ml, and TNF-α >245 pg/ml) have been linked with development of severe ROP [75]. It has also been observed that maternal systemic inflammation leads to decreased levels of insulin-like growth factor-1, which in turn is associated with an increased risk of developing ROP [68, 76]. Cytokine levels in the vitreous of 19 patients with stage 4 ROP were recently analyzed by Sato et al. [77], who demonstrated that increased levels of vascu-
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
Analysis Based on Studies Adjusted for GA To eliminate the effect of prematurity as a confounder, we did a subgroup analysis to include the studies that showed no significant difference in GA between the CA group and the non-CA group. Eight studies were included in the analysis. A meta-analysis of these 8 studies showed no significant association of CA with ROP [summary RR 0.98 (95% CI 0.77–1.26); fig. 7]. No significant heterogeneity was observed (I2 = 0%, pheterogeneity = 0.89). Discussion
195
lar endothelial growth factor had the highest correlation with vascular activity in ROP eyes compared to other cytokines. Sood et al. [78] studied cytokine levels in extremely low-birth-weight neonates on days 1, 3, 7, 14, and 21 and showed that some cytokines were significantly different across ROP groups, suggesting that perinatal inflammation may be involved in the pathogenesis of ROP. Thus perinatal inflammation in the form of CA does seem to play an important role in the development of ROP in preterm infants. However, establishing this relation on statistical terms becomes a major challenge due to the simultaneous interplay of a number of confounding factors. The major confounding factor in such an analysis is prematurity. Since CA is often associated with preterm birth, the inflammatory response commonly develops in a population already at high risk for brain damage and poor neurologic outcome [79]. It has been shown in a number of studies that extreme prematurity plus multiple hits of perinatal inflammation appear to be involved in ROP etiology and progression [26, 39]. Again, lower GA and birth weight are associated with increased oxygen therapy, which is an independent risk factor for development of ROP [80]. In a recent meta-analysis by Saugstad and Aune [81], it was found that a lower oxygen saturation targeting by pulse oximetry reduced severe ROP by 50%, i.e. from 20.9 to 9.5%. In our analysis, infants in the CA group had lower GAs compared to the non-CA group in 15 out of the 27 studies. This underlies the importance of a subgroup analysis adjusting for the major confounding factor GA. In our study, we found only 8 studies that have adjusted for GA. The meta-analysis of these studies incidentally showed no association between CA and ROP. Hence, prematurity itself along with other factors associated with prematurity, like oxygen therapy, could possibly confound the unadjusted results. This also brings to light the dearth of studies on this issue that adjust for potential confounders, which prevents us from drawing a definitive conclusion. Another factor that was not accounted for in the studies is the influence of genetic variation. It has been mentioned in the study by Dammann et al. [26] that single nucleotide polymorphisms in the genes encoding for increased production of IL-1β, TNF-α and reduced Tolllike receptor 4 signal transduction were associated with a tendency towards reduced likelihood for ROP occurrence. In a recent study comparing whole-genome expression in the first month of life in groups of infants with and without ROP, it was found that more genes were underexpressed than overexpressed in the ROP group [82]. 196
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
This may in part account for the complex pathophysiology of ROP and may explain why ROP progresses to a severe stage in some infants, while in others, despite similar clinical conditions, regresses spontaneously, thus leading to variable results on comparative analysis with a single factor like CA. Substantial heterogeneity was observed across the studies in the magnitude and direction of associations. We explored heterogeneity by grouping the studies by type of CA, but heterogeneity remained substantial among the subgroups, except in the clinical CA group. This may be explained by the very small number of studies in the clinical CA group compared to a much larger number of studies in each of the other subgroups [26, 28]. One important source of heterogeneity may be the substantial variation in the definition of histological CA. While few of the studies mentioned diagnostic criteria for defining histological CA such as Redline’s Criteria, College of American Pathologists Guidelines, Salafia staging system, Blanc’s criteria, criteria from Amniotic Fluid Infection Nosology Committee, Roger’s protocol and Yoon’s criteria, most of them used nonspecified diagnostic methods [36, 39, 47, 58, 62, 67, 69]. Placental microbiology was not taken into account in our analysis due to the insufficient number of studies. Interestingly, it had been shown in one of the studies by Chen et al. [39] that although presence of placental organism or placental inflammation individually did not appear to be associated with an increased risk of ROP, the two risk factors, when co-occurring, were associated with a significantly increased risk of ROP in zone I. This may suggest that underappreciation of the etiologic complexity of ROP pathogenesis in many of the other studies may have led to a lack of association between mere placental inflammation and ROP. There was less discrepancy over the defining criteria for ROP, with almost all the included studies following the International Classification of Retinopathy of Prematurity [30]. ROP stage 3 and above being a distinctive marker of proliferative retinopathy was unanimously classified as severe ROP in most of the studies. Apart from gross variation in the definition of CA, lack of adjustment for probable confounders like GA, birth weight and duration of oxygen therapy most likely contributed to the high degree of heterogeneity observed across studies which again prevented us from drawing any definitive conclusion. Interestingly, we noted no heterogeneity in the subgroup analysis across the 8 studies which adjusted for GA, and these studies showed no association between CA and ROP. This underlies the importance of Mitra /Aune /Speer /Saugstad
taking into account all possible sources of heterogeneity across studies including major confounding factors while interpreting the results of a meta-analysis. This study is a perfect example where exclusion of a major confounder in the form of GA and consequent elimination of heterogeneity vastly changed the results and interpretation of the statistical analysis. An interesting observation to note was that our study did not show any publication bias. In our review, 23 out of the 27 included studies (85.2%) were not primarily designed to assess the effect of CA on ROP. They were designed to assess the effect of placental inflammation on neonatal morbidity and neurodevelopmental outcome, and ROP was included as one of the effects. This might have negated the tendency to avoid publication of studies not showing any relation of CA with ROP, or in other words, the ‘file-drawer effect’ was not present in our analysis.
Conclusion
We conclude that though unadjusted analyses showed significant association of CA with ROP (any stage) as well as with severe ROP (stage ≥ 3), no such association was found on subanalysis of the studies adjusting for GA. Hence, CA cannot be definitively considered as a risk factor for ROP based on current evidence. However, in view of dearth of studies specifically focusing on the relation between CA and ROP, there is a need to conduct further studies on this topic adjusting for potential confounding factors and report results by stage to clarify the association with severe ROP.
Disclosure Statement The authors have no conflict of interest to disclose.
References 1 Terry TL: Extreme prematurity and fibroblastic overgrowth of persistent vascular sheath behind each crystalline lens. Am J Opthalmol 1942;25:203–206. 2 Terry TL: Retrolental fibroplasia in the premature infant. V. Further studies on fibroplastic overgrowth of the persistent tunica vasculosa lentis. Trans Am Ophthalmol Soc 1944;42:383–396. 3 Silverman W: Retrolental Fibroplasia: A Modern Parable. New York, Grune and Stratton, 1980. 4 Chen M, Çitil A, McCabe F, et al: Infection, oxygen, and immaturity: interacting risk factors for retinopathy of prematurity. Neonatology 2011;99:125–132. 5 Raghuveer TS, Bloom BT: A paradigm shift in the prevention of retinopathy of prematurity. Neonatology 2011;100:116–129. 6 Darlow BA, Hutchinson JL, HendersonSmart DJ, Donoghue DA, Simpson JM, Evans NJ: Prenatal risk factors for severe retinopathy of prematurity among very preterm infants of the Australian and New Zealand Neonatal Network. Pediatrics 2005;115:990–996. 7 Chen J, Smith LE: Retinopathy of prematurity. Angiogenesis 2007;10:133–140. 8 Holmstrom G, Broberger U, Thomassen P: Neonatal risk factors for retinopathy of prematurity – a population-based study. Acta Ophthalmol Scand 1998;76:204–220. 9 Simhan HN, Caritis SN: Prevention of preterm delivery. N Engl J Med 2007; 357: 477– 487.
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
10 Rivera JC, Sapieha P, Joyal JS, Duhamel F, et al: Understanding retinopathy of prematurity: update on pathogenesis. Neonatology 2011;100:343–353. 11 Romero R, Gomez R, Chaiworapongsa T, Conoscenti G, Kim JC, Kim YM: The role of infection in preterm labour and delivery. Paediatr Perinat Epidemiol 2001;15(suppl 2):41–56. 12 Lahra MM, Jeffry HE: A fetal response to chorioamnionitis is associated with early survival after preterm birth. Am J Obstet Gynecol 2004;190:147–151. 13 Goldenberg RL, Hauth JC, Andrews WW: Intrauterine infection and preterm delivery. N Engl J Med 2000;342:1500–1507. 14 Mittal M, Dhanireddy R, Higgins RD: Candida sepsis and association with retinopathy of prematurity. Pediatrics 1998;101:654–657. 15 Noyola DE, Bohra L, Paysse EA, Fernandez M, Coats DK: Association of candidemia and retinopathy of prematurity in very low birthweight infants. Ophthalmology 2002;109:80– 84. 16 Klinger G, Levy I, Sirota L, Boyko V, LernerGeva L, Reichman B: Outcome of early-onset sepsis in a national cohort of very low birth weight infants. Pediatrics 2010;125:736–740. 17 Ahn HM, Park EA, Cho SJ, Kim YJ, Park HS: The association of histological chorioamnionitis and antenatal steroids on neonatal outcome in preterm infants born at less than thirty-four weeks’ gestation. Neonatology 2012; 102:259–264. 18 Kramer BW: Chorioamnionitis – new ideas from experimental models. Neonatology 2011;99:320–325.
19 Wu YW, Colford JM Jr: Chorioamnionitis as a risk factor for cerebral palsy: a meta-analysis. JAMA 2000;284:1417–1424. 20 Ragouilliaux CJ, Keeney SE, Hawkins HK, Rowen JL: Maternal factors in extremely low birth weight infants who develop spontaneous intestinal perforation. Pediatrics 2007; 120:1458–1464. 21 Lau J, Magee F, Qiu Z, Hoube J, Von Dadelszen P, Lee SK: Chorioamnionitis with a fetal inflammatory response is associated with higher neonatal mortality, morbidity, and resource use than chorioamnionitis displaying a maternal inflammatory response only. Am J Obstet Gynecol 2005;193:708–713. 22 Watterberg KL, Demers LM, Scott SM, Murphy S: Chorioamnionitis and early lung inflammation in infants in whom bronchopulmonary dysplasia develops. Pediatrics 1996; 97:210–215. 23 Dammann O, Leviton A, Bartels DB, Dammann CEL: Lung and brain damage in preterm newborns. Are they related? How? Why? Biol Neonate 2004;85:305–313. 24 Thomas W, Speer CP: Chorioamnionitis: important risk factor or innocent bystander for neonatal outcome? Neonatology 2011; 99: 177–187. 25 Speer CP: Neonatal respiratory distress syndrome: an inflammatory disease? Neonatology 2011;99:316–319. 26 Dammann O, Brinkhaus MJ, Bartels DB, et al: Immaturity, perinatal inflammation, and retinopathy of prematurity: a multi-hit hypothesis. Early Hum Dev 2009;85:325–329.
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
197
27 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. 28 Soraisham AS, Singhal N, McMillan DD, Sauve RS, Lee SK: A multicenter study on the clinical outcome of chorioamnionitis in preterm infants. Am J Obstet Gynecol 2009;200: 371–376. 29 Yoon BH, Romero R, Kim CJ, et al: 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. 30 An international classification of retinopathy of prematurity. Pediatrics 1984;74:127–133. 31 Wells G, Shea B, O’Connell B, et al: The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Non-Randomized Studies in Metaanalysis. Ottawa, University of Ottawa, 2009. 32 Hartling L, Liang Y, Lacaze-Masmonteil T: Chorioamnionitis as a risk factor for bronchopulmonary dysplasia: a systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed 2012;97:8–17. 33 DerSimonian R, Laird N: Meta-analysis in clinical trials. Control Clin Trials 1986;7:177– 188. 34 Higgins JP, Thompson SG: Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21:1539–1558. 35 Begg CB, Mazumdar M: Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088–1101. 36 Bordigato MA, Piva D, Gangi IMD, Giordano G, Chiandetti L, Filippone M: Asymmetric dimethylarginine in ELBW newborns exposed to chorioamnionitis. Early Hum Dev 2011;87: 143–145. 37 Langston C, Kaplan C, Macpherson T, et al: Practice guideline for examination of the placenta: developed by the Placental Pathology Practice Guideline Development Task Force of the College of American Pathologists. Arch Pathol Lab Med 1997;121:449–476. 38 Redline RW: Inflammatory responses in the placenta and umbilical cord. Semin Fetal Neonatal Med 2006;11:296–301. 39 Chen ML, Allred EN, Hecht JL, et al: Placenta microbiology and histology and the risk for severe retinopathy of prematurity. Invest Ophthalmol Vis Sci 2011;52:7052–7058. 40 Driscoll SG, Langston C: College of American Pathologists Conference XIX on the Examination of the Placenta: report of the Working Group on Methods for Placental Examination. Arch Pathol Lab Med 1991;115:704–708. 41 Felice CD, Toti P, Parrini S, et al: Histologic chorioamnionitis and severity of illness in very low birth weight newborns. Pediatr Crit Care Med 2005;6:298–302.
198
42 Fung G, Bawden K, Chow P, Yu V: Long-term outcome of extremely preterm infants following chorioamnionitis. HK J Pediatr (new series) 2003;8:87–92. 43 Fung G, Bawden K, Chow P, Yu V: Chorioamnionitis and outcome in extremely preterm infants. Ann Acad Med Singapore 2003; 32:305–310. 44 Gray PH, Hurley TM, Rogers YM, et al: Survival and neonatal and neurodevelopmental outcome of 24–29 week gestation infants according to primary cause of preterm delivery. Aust NZ J Obstet Gynaecol 1997;31:161. 45 Hendson L, Russell L, Robertson CMT, et al: Neonatal and neurodevelopmental outcomes of very low birth weight infants with histologic chorioamnionitis. J Pediatr 2011; 158: 397–402. 46 Reese AB, Ling MJ, Owens WC: A classification of retrolental fibroplasias. Am J Opthalmol 1953;36:1333. 47 Lee HJ, Kim EK, Kim HS, Choi CW, Kim BII, Choi JH: Chorioamnionitis, respiratory distress syndrome and bronchopulmonary dysplasia in extremely low birth weight infants. J Perinatol 2011;31:166–170. 48 Salafia CM, Weigl C, Silberman L: The prevalence and distribution of acute placental inflammation in uncomplicated term pregnancies. Obstet Gynecol 1989;73:383–389. 49 Mehta R, Nanjundaswamy S, Shen-Schwarz S, Petrova A: Neonatal morbidity and placental pathology. Indian J Pediatr 2006;73:25–28. 50 Morales WJ: The effect of chorioamnionitis on the developmental outcome of preterm infants at one year. Obstet Gynaecol 1987; 70: 183–186. 51 Moscuzza F, Belcari F, Nardini V, et al: Correlation between placental histopathology and fetal/neonatal outcome: chorioamnionitis and funisitis are associated to intraventricular haemorrage and retinopathy of prematurity in preterm newborns. Gynecol Endocrinol 2011;27:319–323. 52 Nasef N, Shabaan AE, Schurr P, et al: Effect of clinical and histological chorioamnionitis on the outcome of preterm infants. Am J Perinatol 2013;30:59–68. 53 International Committee for the Classification of Retinopathy of Prematurity: The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005;123:991–999. 54 Newton ER: Chorioamnionitis and intraamniotic infection. Clin Obstet Gynecol 1993;36: 795–808. 55 Redline RW, Faye-Petersen O, Heller D, Qureshi F, Savell V, Vogler C, Society for Pediatric Pathology, Perinatal Section, Amniotic Fluid Infection Nosology Committee: Amniotic infection syndrome: nosology and reproducibility of placental reaction patterns. Pediatr Dev Pathol 2003;6:435–448.
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
56 Ogunyemi D, Murillo M, Jackson U, Hunter N, Alperson B: The relationship between placental histopathology findings and perinatal outcome in preterm infants. J Mat Fetal Neonatal Med 2003;13:102–109. 57 Perrone S, Toti P, Toti MS, et al: Perinatal outcome and placental histological characteristics: a single-center study. J Matern Fetal Neonatal Med 2012;25(suppl 1):110–113. 58 Rocha G, Proença E, Quintas C, Rodrigues T, Guimarães H: Corioamnionite e morbidade neonatal. Acta Med Port 2006;19:207–212. 59 Blanc WA: Pathology of the placenta, membranes, and umbilical cord in bacterial, fungal, and viral infections in man. Monogr Pathol 1981;22:67–132. 60 Sato M, Nishimaki S, Yokota S, et al: Severity of chorioamnionitis and neonatal outcome. J Obstet Gynaecol Res 2011;37:1313–1319. 61 Schlapbach LJ, Ersch J, Adams M, Bernet V, Bucher HU, Latal B: Impact of chorioamnionitis and preeclampsia on neurodevelopmental outcome in preterm infants below 32 weeks gestational age. Acta Paediatr 2010;99: 1504–1509. 62 Seliga-Siwecka JP, Kornacka MK: Neonatal outcome of preterm infants born to mothers with abnormal genital tract colonisation and chorioamnionitis: a cohort study. Early Hum Dev 2013;89:271–275. 63 Soraisham AS, Trevenen C, Wood S, Singhal N, Sauve R: Histological chorioamnionitis and neurodevelopmental outcome in preterm infants. J Perinatol 2013;33:70–75. 64 Suppiej A, Franzoi M, Vedovato S, Marucco A, Chiarelli S, Zanardo V: Neurodevelopmental outcome in preterm histological chorioamnionitis. Early Hum Dev 2009; 85: 187– 189. 65 van Vliet EOG, de Kieviet JF, van der Voorn JP, et al: Placental pathology and long-term neurodevelopment of very preterm infants. Am J Obstet Gynecol 2012;206:489.e1–7. 66 Wirbelauer J, Thomas W, Speer CP: Response of leukocytes and nucleated red blood cells in very low-birth weight preterm infants after exposure to intrauterine inflammation. J Mat Fetal Neonatal Med 2011;24:348–353. 67 Rogers BB, Alexander JM, Head J, McIntire D, Leveno KJ: Umbilical vein interleukin-6 levels correlate with the severity of placental inflammation and gestational age. Hum Pathol 2002;33:335–340. 68 Woo SJ, Park KH, Jung HJ, et al: Effects of maternal and placental inflammation on retinopathy of prematurity. Graefes Arch Clin Exp Ophthalmol 2012;250:915–923. 69 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. 70 Murata Y, Itakura A, Matsuzawa K, Okumura A, Wakai K, Mizutani S: Possible antenatal and perinatal related factors in development of cystic periventricular leukomalacia. Brain Dev 2005;27:17–21.
Mitra /Aune /Speer /Saugstad
71 Wu YW, Colford JM Jr: Chorioamnionitis as a risk factor for cerebral palsy: a meta-analysis. JAMA 2000;284:1417–1424. 72 Shatrov JG, Birch SC, Lam LT, Quinlivan JA, McIntyre S, Mendz GL: Chorioamnionitis and cerebral palsy: a meta-analysis. Obstet Gynecol 2010;116:387–392. 73 Dammann O, Leviton A: Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 1997; 42: 1–8. 74 Stolp HB, Dziegielewska KM: Review: role of developmental inflammation and bloodbrain barrier dysfunction in neurodevelopmental and neurodegenerative diseases. Neuropathol Appl Neurobiol 2009;35:132–146. 75 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.
Chorioamnionitis as a Risk for ROP: A Meta-Analysis
76 Hellström A, Peruzzi C, Ju M, et al: Low IGF1 suppresses VEGF-survival signaling in retinal endothelial cells: direct correlation with clinical retinopathy of prematurity. Proc Natl Acad Sci 2001;98:5804–5808. 77 Sato T, Kusaka S, Shimojo H, Fujikado T: Simultaneous analyses of vitreous levels of 27 cytokines in eyes with retinopathy of prematurity. Ophthalmology 2009;116:2165–2169. 78 Sood BG, Madan A, Saha S, et al: On behalf of the NICHD Neonatal Research Network. Perinatal systemic inflammatory response syndrome and retinopathy of prematurity. Pediatr Res 2010;67:394–400.
79 Reiman M, Kujari H, Maunu J, et al, PIPARI Study Group: Does placental inflammation relate to brain lesions and volume in preterm infants? J Pediatr 2008; 152: 642–647, 647.e1– 2. 80 Fortes Filho JB, Eckert GU, Procianoy L, Barros CK, Procianoy RS: Incidence and risk factors for retinopathy of prematurity in very low and in extremely low birth weight infants in a unit-based approach in southern Brazil. Eye 2009;23:25–30. 81 Saugstad OD, Aune D: In search of the optimal oxygen saturation for extremely low birth weight infants: a systematic review and metaanalysis. Neonatology 2011;100:1–8. 82 Pietrzyk JJ, Kwinta P, Bik-Multanowski M, et al: New insight into the pathogenesis of retinopathy of prematurity: assessment of wholegenome expression. Pediatr Res 2013;73:476– 483.
Neonatology 2014;105:189–199 DOI: 10.1159/000357556
199
Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.
![[Football as risk factor for a non-injury-related knee osteoarthritis - results from a systematic review and metaanalysis].](/assets/img/hold.png)
