 |
|
Chorioamnionitis as a Risk Factor for Cerebral Palsy
A Meta-analysis
Yvonne W. Wu, MD, MPH;
John M. Colford, Jr, MD, PhD
JAMA. 2000;284:1417-1424.
ABSTRACT
| |
Context Chorioamnionitis has been implicated in the pathogenesis of cerebral palsy, but most studies have not reported a significant association. Cystic periventricular leukomalacia (cPVL) is believed to be a precursor of cerebral palsy in preterm infants.
Objectives To determine whether chorioamnionitis is associated with cerebral palsy or cPVL and to examine factors that may explain differences in study results.
Data Sources Searches of MEDLINE (1966-1999), Index Medicus (1960-1965), Doctoral Dissertation Abstracts On-Line (1861-1999), bibliographies, and online conference proceedings (1999) were performed for English-language studies with titles or abstracts that discussed prenatal risk factors for cerebral palsy or cPVL.
Study Selection Of 229 initially identified publications, meta-analyses were performed on studies that addressed the association between clinical (n = 19) or histologic (n = 7) chorioamnionitis and cerebral palsy or cPVL in both preterm and full-term infants. Inclusion criteria were: presence of appropriate exposure and outcome measures, case-control or cohort study design, and provision of sufficient data to calculate relative risks (RRs) or odds ratios with 95% confidence intervals (CIs). Studies evaluating risk of cerebral palsy following maternal fever, urinary tract infection, or other maternal infection were collected, but not included in the meta-analysis.
Data Extraction Information from individual studies was abstracted using standardized forms by 2 independent observers blinded to authors' names, journal titles, and funding sources.
Data Synthesis Using a random effects model, clinical chorioamnionitis was significantly associated with both cerebral palsy (RR, 1.9; 95% CI, 1.4-2.5) and cPVL (RR, 3.0; 95% CI, 2.2-4.0) in preterm infants. The RR of histologic chorioamnionitis and cerebral palsy was 1.6 (95% CI, 0.9-2.7) in preterm infants, and histologic chorioamnionitis was significantly associated with cPVL (RR, 2.1; 95% CI, 1.5-2.9). Among full-term infants, a positive association was found between clinical chorioamnionitis and cerebral palsy (RR, 4.7; 95% CI, 1.3-16.2). Factors explaining differences in study results included varying definitions of clinical chorioamnionitis, extent of blinding in determining exposure status, and whether individual studies adjusted for potential confounders.
Conclusion Our meta-analysis indicates that chorioamnionitis is a risk factor for both cerebral palsy and cPVL.
INTRODUCTION
Despite improvements in perinatal medicine, the prevalence of cerebral palsy has increased over the last 2 decades,1 and the etiology of cerebral palsy remains poorly understood. Evidence suggests that 70% to 80% of cases of cerebral palsy are due to prenatal factors,2 and that birth asphyxia plays a relatively minor role.3
Chorioamnionitis, or intrauterine infection, has been implicated as a potential cause of cerebral palsy.4-6 Chorioamnionitis is common and often subclinical. Histologic signs of chorioamnionitis can be detected in more than 50% of women who deliver prematurely,7 and most patients have no clinical signs of infection.8 It is postulated that both subclinical and clinical chorioamnionitis can lead to a fetal inflammatory response, and that this inflammation contributes to neonatal brain injury and subsequent cerebral palsy.5, 9-10 Recent reports of elevated neonatal blood11 and amniotic fluid12 cytokine levels in children with cerebral palsy support the notion that cerebral palsy is preceded by a perinatal inflammatory condition.
A number of studies have assessed the relationship between chorioamnionitis and cerebral palsy in premature infants,12-20 but most have not reported a significant association. In full-term infants, the existing literature supports a relationship between chorioamnionitis and cerebral palsy,6, 21 but few studies are available.
We conducted a meta-analysis to evaluate the potential association between chorioamnionitis and cerebral palsy in both full-term and preterm infants. In addition to cerebral palsy, we were interested in the outcome of cystic periventricular leukomalacia (cPVL), a powerful predictor of cerebral palsy in preterm infants. An estimated 60% to 100% of patients with cPVL go on to develop cerebral palsy.22 We performed a systematic review of the existing literature addressing all forms of chorioamnionitis as a potential risk factor for cerebral palsy or cPVL.
METHODS
Study Selection
MEDLINE (1966-1999) was searched using keywords cerebral palsy and etiology with all possible suffixes included (eg, etiologies, etiologic), as well as medical subject headings cerebral palsy, etiology; periventricular leukomalacia, etiology; periventricular leukomalacia, epidemiology; and chorioamnionitis, complications. Index Medicus (1960-1965) was searched by hand for the term cerebral palsy. Doctoral Dissertation Abstracts On-Line (North American Universities, 1861-1999) was searched for keywords cerebral palsy, chorioamnionitis, maternal infection, and leukomalacia. Conference proceedings available online were searched (Society for Pediatric Research, 1999, and Child Neurology Society, 1999), as were abstracts from the January 2000 meeting of the Society for Maternal-Fetal Medicine. Bibliographies of all relevant articles were reviewed for further potential references. Only studies published in English were included.
Studies with titles or abstracts discussing prenatal risk factors for cerebral palsy or cPVL were retrieved. Any study with primary data investigating a relevant exposure and outcome was submitted for further review. A relevant exposure included any maternal infection or sign of infection during pregnancy (eg, chorioamnionitis, urinary tract infection, fever), with the exception of known congenital infections such as toxoplasmosis and cytomegaloviral infection. Relevant outcomes included cerebral palsy by any definition, cPVL, or echolucencies on neonatal ultrasound. Studies that included the ultrasonographic findings of echodensity, echogenicity, or ventriculomegaly in the definition of cPVL were excluded from the analysis. Abnormal echodensity can be a transient finding that does not invariably lead to cPVL.22 When multiple publications reported relative risks (RRs) for the same study subjects, the most recent publication was chosen.
Data Abstraction
The studies identified by the criteria described above were then reviewed independently by both authors, with all references to author names, journal title, and funding sources removed. The following inclusion criteria were applied: (1) appropriate exposure and outcome measures, as defined above; (2) case-control or cohort study design; and (3) RR or odds ratio with 95% confidence interval (CI) provided, or able to be calculated from the data presented in the article. The following data were abstracted onto standardized forms: publication year, years of study, gestational age, method of obtaining exposure information, definition of chorioamnionitis/maternal infection, definition of cerebral palsy, minimum age at diagnosis of cerebral palsy, definition of cPVL, source of cases and controls, extent of blinding, extent of controlling for potential confounding, and relevant risk ratio estimates.
Whenever possible, chorioamnionitis was categorized as clinical, histologic, and/or microbiologic. A clinical diagnosis of chorioamnionitis is usually based on criteria such as maternal fever, uterine tenderness, malodorous amniotic fluid, maternal or fetal tachycardia, and maternal leukocytosis.23 If based on such clinical criteria alone, the diagnosis was categorized as clinical chorioamnionitis. Histologic chorioamnionitis represented pathologic findings on placental histology, including inflammation of the placental membranes, fetal vasculitis, or funisitis (inflammation of the umbilical cord).24 Microbiologic chorioamnionitis was defined as retrieval of microbial organisms from the amniotic fluid or placental cultures.
If a potential confounder was matched in the study design, or included in a multivariate analysis, the resulting risk estimate was considered to be adjusted for that confounder. Whenever possible, an odds ratio was calculated and used in the summary calculations. This allowed the individual risk estimates to be as consistent as possible since most studies reported odds ratios and not RRs. The odds ratio can be interpreted as an approximation of the RR since the prevalence of cerebral palsy is low (1.5-2.5/1000 live births).25 The Woolf26 method was used in calculating a 95% CI from the raw data if available; otherwise, we calculated a 95% CI using the z score obtained from the corresponding P value.27 If a published RR or 95% CI was inconsistent with the raw data provided by the study,28-29 the effect measure was recalculated using the available data. When a zero cell was encountered,28, 30 a value of 0.5 was added to each cell in the corresponding 2 x 2 table. When there was insufficient information to calculate an RR, an attempt was made to contact the original authors for further information. Raw data and RR estimates for each study are available from the corresponding author (Y.W.W.).
Statistical Analysis
In choosing from multiple effect estimates in a single study, the least adjusted estimate (ie, the estimate adjusted for the fewest number of confounding variables) was used as the primary outcome measure. Choosing the least adjusted estimate provided the maximum opportunity for comparison of consistent estimates across studies since the majority of studies only reported RRs that were uncontrolled for potential confounders. Studies with adjusted estimates were analyzed separately. In the summary estimate for either clinical or histologic chorioamnionitis, the RR for clinical chorioamnionitis was chosen over histologic chorioamnionitis, when studies provided both values, since the majority of studies involved chorioamnionitis that was defined by clinical criteria. Some studies presented multiple effect estimates for the relationship between histologic chorioamnionitis and cerebral palsy or cPVL, based on separate pathologic findings.9, 18, 31 For each of these studies, a single RR was calculated by combining the separate effect estimates using the Mantel-Haenszel method.32
Summary RR estimates were calculated by taking a weighted average of individual study results using a random effects model. The weight for each study is the inverse of the sum of 2 terms: the study variance and a term accounting for between-study variability.33 Homogeneity was formally tested using the general variance-based method,33 in which a conservative value of P<.10 was used to classify study results as heterogeneous. Heterogeneous studies may not measure the same underlying association between exposure and disease, so that differences in study results are unlikely to be due to random variation alone.
Finally, separate meta-analyses were performed after stratification by factors pertaining to study quality, including study design, definition of chorioamnionitis, diagnostic criteria for cerebral palsy, and extent of control for potential confounders. Differences in the stratified summary estimates were evaluated using a z score.34 Potential publication bias was evaluated with a Kendall  test.35
RESULTS
A total of 229 articles were retrieved. One hundred six articles were from MEDLINE, 11 from Index Medicus, 2 from Doctoral Dissertation Abstracts On-line, 3 from conference proceedings, and 107 from bibliography review. Forty-four publications were submitted for blinded review. The remaining articles were discarded because of redundancy in data reporting,4, 36-40 because they did not contain original data, or because they lacked information regarding maternal infection as a risk factor for cerebral palsy or cPVL. The blinded review further excluded 14 studies because of insufficient information to calculate an RR.41-54 The remaining 30 studies are summarized in Table 1.
|
|
|
|
Table 1. Summary of Study Characteristics
|
|
|
Premature Infants
Seventeen studies evaluated the association between clinical chorioamnionitis and cerebral palsy12-20,55-56 or cPVL29, 31, 57-60 in preterm infants, with RRs ranging from 0.95 to 7.3 (Figure 1). Ten of these 17 studies found no significant association. When the study results were analyzed using the random effects model, clinical chorioamnionitis was significantly associated with both cerebral palsy (RR, 1.9; 95% CI, 1.4-2.5) and cPVL (RR, 3.0; 95% CI, 2.2-4.0).
|
|
|
|
Figure. Chorioamnionitis and Cerebral Palsy (CP) or Cystic Periventricular Leukomalacia (cPVL) in Premature Infants
Asterisks indicate that P values refer to the test of homogeneity. This test evaluates the likelihood that differences between studies are due to random variation alone. If P>.10, the studies are considered homogeneous, and variability in study results is more likely to be due to chance alone.
|
|
|
Five studies evaluated the association between histologic chorioamnionitis and cerebral palsy in premature infants,12, 17-19,24 with only 1 reporting a significant association (Figure 1).19 The meta-analysis of these 5 studies produced an RR of 1.6 (95% CI, 0.9-2.7). Two studies evaluated the association between histologic chorioamnionitis and cPVL,9, 31 and the summary estimate was significant (RR, 2.1; 95% CI, 1.5-2.9). When clinical and histologic chorioamnionitis were combined into a single category as a potential risk factor for cerebral palsy, the test of homogeneity suggested that the combined studies were heterogeneous (P = .07). The single study that evaluated microbiologic evidence of chorioamnionitis found no significant association with cerebral palsy.12
Stratified Analysis in Preterm Infants
A stratified analysis was performed to investigate possible sources of heterogeneity in studies of clinical chorioamnionitis in premature infants (Table 2). Both cohort studies and case-control studies produced significant summary RRs, but a stronger effect was seen in cohort studies (RR, 3.0 vs 1.8; P = .001). Studies that included only very low-birth-weight infants (<1500 g) or infants born before 32 weeks' gestation were statistically heterogeneous (P = .02).
|
|
|
|
Table 2. Summary Estimates of Clinical Chorioamnionitis and Cerebral Palsy or Cystic Periventricular Leukomalacia in Premature Infants Stratified by Study Characteristics*
|
|
|
Six articles provided no specific criteria for the diagnosis of clinical chorioamnionitis. The summary RR from these studies was lower than that calculated from the 7 studies with the most stringent diagnostic criteria for clinical chorioamnionitis (ie, requiring 2 clinical signs or symptoms; RR, 1.8 vs 2.9; P = .005; Table 2). Evidence for heterogeneity was found among studies without specific diagnostic criteria (P = .08). Five studies examining the risk of maternal fever alone9, 18, 57, 61-62 were also statistically heterogeneous (P = .004).
The baseline prevalence of clinical chorioamnionitis in preterm infants estimated from control populations varied from 3%29, 57, 63 to 56%.14 The 7 studies reporting a relatively high (>15%) baseline prevalence were statistically heterogeneous (P = .006; Table 2). Studies that did not explicitly state that ascertainment of chorioamnionitis was performed blinded to outcome status produced a larger summary estimate than studies that blinded the exposure ascertainment (RR, 3.0 vs 1.8; P = .002).
The baseline prevalence of histologic chorioamnionitis in healthy premature control infants ranged from 33%31 to 72%.18 Of the 7 studies addressing histologic chorioamnionitis in preterm infants, only 2 studies used the same pathologic findings in defining chorioamnionitis.
The definition of cerebral palsy differed greatly among studies of premature infants (Table 2). Of the 8 studies that provided adequate information, 4 did not require a minimum age of 2 years for the diagnosis of cerebral palsy, and most did not confirm the diagnosis by an examination performed by study personnel. Some studies excluded children with congenital anomalies, while others did not specify any exclusion criteria. Cystic PVL was defined as echolucencies seen in the periventricular white matter on head ultrasound. Of the 2 studies9, 64 that noted age at which ultrasounds were obtained, only 1 routinely performed follow-up scans after 2 weeks of life.9
Six studies of clinical chorioamnionitis in premature infants reported RRs controlled for gestational age.13, 17-19,29, 58 The summary RR calculated from adjusted estimates was smaller than that derived from unadjusted estimates (RR, 1.8 vs 2.5; P = .03; Table 2). Two positive associations became insignificant after controlling for gestational age,15, 58 and only 1 study provided an RR that remained significant after adjustment,57 but this study controlled only for prolonged rupture of membranes.
Full-Term Infants
For full-term infants, 2 studies of clinical chorioamnionitis and cerebral palsy yielded a summary RR of 4.7 (95% CI, 1.3-16.2).6, 21 The single study examining the association between histologic chorioamnionitis and cerebral palsy in full-term infants found an RR of 8.9 (95% CI, 1.9-40).6 No studies evaluated microbiologic chorioamnionitis.
Sensitivity Analysis
The association between clinical chorioamnionitis and cerebral palsy was determined for studies including all gestational ages. The resulting summary RR differed minimally from that calculated for premature infants only, although evidence of heterogeneity was present (RR, 2.1; 95% CI, 1.5-3.0; P = .08). The addition of the single study addressing histologic chorioamnionitis in full-term infants6 did not appreciably change the association between histologic chorioamnionitis and cerebral palsy.
Publication Bias
Any meta-analysis is subject to potential publication bias (ie, nonpublication of studies that show no association between exposure and outcome). A Kendall test (P = .27) did not support the presence of significant publication bias. Of note, only 2 negative studies were identified among the excluded studies that lacked sufficient information to calculate an RR. One reported no association between chorioamnionitis and cPVL,43 and the other found no association between chorioamnionitis and cerebral palsy or death.45
Five studies evaluating the risk of cerebral palsy following maternal urinary tract infection during pregnancy6, 15, 30, 62, 65 found widely varying results, with RRs ranging from 0.16 to 3.9. It is likely that these studies are heterogeneous (P = .05), and therefore not evaluating the same underlying association.
COMMENT
Based on the available literature, chorioamnionitis appears to be associated with both cerebral palsy and cPVL in premature infants. Full-term infants exposed to chorioamnionitis also exhibit an increased risk of cerebral palsy in the 2 studies that are available. A growing body of evidence supports the notion that fetal inflammation caused by maternal infection contributes to neonatal brain injury. Experimental maternal intrauterine infection induces cytokine production in fetal rat brains,66 and maternal intrauterine infection leads to a robust fetal inflammatory response in humans.67 The cytokine hypothesis10 states that elevated blood and brain cytokine levels resulting from maternal infection lead to central nervous system damage in the fetus, and subsequent cerebral palsy. Indeed, experimental evidence indicates that inflammatory cytokines can be neurotoxic both in vitro68 and in vivo,69 and that cytokines inhibit oligodendrocytes in the developing white matter.70 In a rabbit model, experimentally induced intrauterine infection has been shown to cause fetal white matter brain lesions.71
Neonatal blood inflammatory cytokine levels are significantly higher in full-term infants who develop cerebral palsy than in controls.11 In the preterm population, a positive association has been reported between intra-amniotic fluid inflammatory cytokines and cerebral palsy72; and intra-amnionitic,12 brain,73-74 and umbilical cord75 cytokine levels are all associated with cPVL. Although these findings do not establish a causal link between cytokines and perinatal brain injury, they do provide evidence that cerebral palsy is associated with an inflammatory environment at birth.
To better understand the relationships between chorioamnionitis, inflammation, and cerebral palsy, it is crucial to develop consensus definitions of clinical and histologic chorioamnionitis. Most published reports do not apply specific diagnostic criteria for clinical chorioamnionitis, and as a group, these studies produced heterogeneous results. The results of our stratified analysis suggest that a stronger association is seen when studies use more stringent diagnostic criteria. Not surprisingly, studies evaluating the effect of fever alone were statistically heterogeneous since fever during labor may be due to epidural analgesia76 or infections other than chorioamnionitis. Studies that do not include strict clinical criteria beyond the presence of fever may lead to an overdiagnosis of chorioamnionitis, which may explain why studies reporting a relatively high prevalence (>15%) of clinical chorioamnionitis also demonstrated statistical evidence of heterogeneity.
When chorioamnionitis was diagnosed without knowledge of outcome status, a significantly lower summary RR was found. This demonstrates the importance of using blinded observers when evaluating the presence of chorioamnionitis.
The definition of cerebral palsy varied between studies; some explicitly excluded children with postnatal causes or congenital anomalies, while others did not. Although cerebral palsy can be difficult to diagnose in children younger than 2 years, 1 study made the diagnosis in infants as young as 6 months. A consensus definition of cerebral palsy is needed.
Few studies report significant associations between chorioamnionitis and cerebral palsy after adjusting for potential confounders. Although gestational age appears to be a possible confounder, it may also lie directly in the causal pathway between maternal infection and cerebral palsy. Chorioamnionitis is associated with premature delivery,7 and low gestational age is in turn associated with a host of intrinsic vulnerabilities within the brain that have been implicated in the pathogenesis of cPVL and cerebral palsy.77 Therefore, if low gestational age resulting from maternal infection in itself plays a direct role in the pathogenesis of cerebral palsy, then adjusting for its effect will falsely diminish the observed association between chorioamnionitis and cerebral palsy. Two studies of clinical chorioamnionitis in premature infants reported positive univariate associations, as well as adjusted estimates of RR15, 58; both became insignificant after controlling for the effect of gestational age. It is unclear if this is because chorioamnionitis does not contribute independently to cerebral palsy, or perhaps because gestational age lies on the causal pathway.
Preeclampsia has been shown to be strongly protective against cerebral palsy in preterm infants.63, 78 Failure to adjust for this potential confounder could bias the RR away from the null, giving falsely high estimates of association. The single study that adjusted for the effect of preeclampsia,58 as well as the 1 study that excluded women with preeclampsia,18 found no significant association between chorioamnionitis and cerebral palsy in premature infants. This only reinforces the need to control for the potential positive confounding effect of preeclampsia in future studies.
Other factors may interact with chorioamnionitis in the pathogenetic pathway leading to cerebral palsy. For instance, 1 study suggests that there may be a statistical interaction between birth asphyxia, or hypoxic-ischemic brain injury, and maternal infection.79 Genetic predisposition to cytokine functional polymorphisms may also play a role in determining which infants exposed to intrauterine infection go on to develop brain damage.80 Individuals with a tumor necrosis factor  promoter region polymorphism demonstrate an increased production of this cytokine, and an increased risk of preterm rupture of membranes.81 Such genetic factors may in part determine the extent of inflammation that occurs in the face of maternal infection.
All meta-analyses are subject to limitations. The biases inherent to observational studies are not eliminated in a quantitative synthesis. There may be publication bias, incomplete ascertainment of published studies, and errors in data abstraction. We did not find statistical evidence for publication bias. Blinded independent abstractors were used in an attempt to maximize accuracy in data abstraction.
The attributable risk of chorioamnionitis and spastic cerebral palsy has been reported to be as high as 12% (95% CI, 4%-20%) in normal-birth-weight children.6 For premature infants, a conservative calculation using the lower end of our 95% CI (RR, 1.4) yields an attributable risk of 28%. The total annual cost of cerebral palsy in the United States has been estimated at $5 billion.25 Even a partial reduction in the incidence of cerebral palsy could substantially reduce both financial costs and personal suffering. Would efforts to prevent chorioamnionitis and fetal inflammation lead to a decrease in the prevalence of cerebral palsy? Do the tocolytic, antibiotic, and steroid agents currently used in perinatal medicine alleviate or perhaps aggravate the fetal brain damage postulated to occur in response to maternal infection and fetal inflammation? Multicenter collaborations between obstetricians, neonatologists, and neurodevelopmental specialists are needed to address these important and complex questions.
AUTHOR INFORMATION
Funding/Support: This study was funded in part by the Neurological Sciences Academic Development Award grant 5 K12 NS01692-05.
Acknowledgment: We are indebted to S. Claiborne Johnston, MD, MPH, for his invaluable assistance with statistical methods and his review of the manuscript. We also wish to thank Donna Ferriero, MD, and Judith Grether, PhD, for their help with the manuscript, and Farhad Sahebkar, MD, for his research assistance.
Corresponding Author and Reprints: Yvonne W. Wu, MD, MPH, Department of Neurology, University of California, PO Box 0114, San Francisco, CA 94143-0114 (e-mail: yvonne{at}itsa.ucsf.edu).
Author Affiliations: Departments of Neurology and Pediatrics, University of California, San Francisco (Dr Wu); and the Division of Epidemiology, School of Public Health, University of California, Berkeley (Dr Colford).
REFERENCES
| |
1. Bhushan V, Paneth N, Kiely JL. Impact of improved survival of very low birth weight infants on recent secular trends in the prevalence of cerebral palsy. Pediatrics. 1993;91:1094-1100.
FREE FULL TEXT
2. Blair E, Stanley FJ. Intrapartum asphyxia: a rare cause of cerebral palsy. J Pediatr. 1988;112:515-519. [published erratum appears in J Pediatr. 1988;113:420].
FULL TEXT
3. Perlman JM. Intrapartum hypoxic-ischemic cerebral injury and subsequent cerebral palsy: medicolegal issues. Pediatrics. 1997;99:851-859.
FREE FULL TEXT
4. Dammann O, Leviton A. Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res. 1997;42:1-8.
ISI
| PUBMED
5. Adinolfi M. Infectious diseases in pregnancy, cytokines and neurological impairment: an hypothesis. Dev Med Child Neurol. 1993;35:549-553.
ISI
| PUBMED
6. Grether JK, Nelson KB. Maternal infection and cerebral palsy in infants of normal birth weight. JAMA. 1997;278:207-211. [published erratum appears in JAMA. 1998;279:118].
FREE FULL TEXT
7. Hillier SL, Martius J, Krohn M, Kiviat N, Holmes KK, Eschenbach DA. A case-control study of chorioamnionic infection and histologic chorioamnionitis in prematurity. N Engl J Med. 1988;319:972-978.
ABSTRACT
8. Russell P. Inflammatory lesions of the human placenta, I: clinical significance of acute chorioamnionitis. Am J Diagn Gynecol Obstet. 1979;1:127-137.
9. Leviton A, Paneth N, Reuss L, et al. Maternal infection, fetal inflammatory response, and brain damage in very low birth weight infants. Pediatr Res. 1999;46:566-575.
ISI
| PUBMED
10. Dammann O, Leviton A. Infection remote from the brain, neonatal white matter damage, and cerebral palsy in the preterm infant. Semin Pediatr Neurol. 1998;5:190-201.
FULL TEXT
| PUBMED
11. Nelson KB, Dambrosia JM, Grether JK, Phillips TM. Neonatal cytokines and coagulation factors in children with cerebral palsy. Ann Neurol. 1998;44:665-675.
FULL TEXT
|
ISI
| PUBMED
12. Yoon BH, Jun JK, Romero R, et al. Amniotic fluid inflammatory cytokines (interleukin-6, interleukin-1 beta, and tumor necrosis factor-alpha), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol. 1997;177:19-26.
FULL TEXT
|
ISI
| PUBMED
13. Wilson-Costello D, Borawski E, Friedman H, Redline R, Fanaroff AA, Hack M. Perinatal correlates of cerebral palsy and other neurologic impairment among very low birth weight children. Pediatrics. 1998;102:315-322.
FREE FULL TEXT
14. Grether JK, Nelson KB, Emery III ES, Cummins SK. Prenatal and perinatal factors and cerebral palsy in very low birth weight infants. J Pediatr. 1996;128:407-414.
FULL TEXT
|
ISI
| PUBMED
15. Allan WC, Vohr B, Makuch RW, Katz KH, Ment LR. Antecedents of cerebral palsy in a multicenter trial of indomethacin for intraventricular hemorrhage. Arch Pediatr Adolesc Med. 1997;151:580-585.
ABSTRACT
16. Cooke RW. Cerebral palsy in very low birthweight infants. Arch Dis Child. 1990;65:201-206.
ABSTRACT
17. O'Shea TM, Klinepeter KL, Meis PJ, Dillard RG. Intrauterine infection and the risk of cerebral palsy in very low-birthweight infants. Paediatr Perinat Epidemiol. 1998;12:72-83.
FULL TEXT
|
ISI
| PUBMED
18. Grether JK, Nelson KB. Intrauterine infection and cerebral palsy in preterm children. Am J Obstet Gynecol. 2000;182(suppl):S95.
19. Jacobsson B, Hagberg G, Hagberg B, Ladfors L, Niklasson A. Cerebral palsy in preterm infants: a population based analysis of antenatal risk factors. Am J Obstet Gynecol. 2000;182(suppl):S29.
20. Nelson KB, Ellenberg JH. Predictors of low and very low birth weight and the relation of these to cerebral palsy. JAMA. 1985;254:1473-1479.
ABSTRACT
21. Nelson KB, Ellenberg JH. Antecedents of cerebral palsy, I: univariate analysis of risks. Am J Dis Child. 1985;139:1031-1038.
ABSTRACT
22. Leviton A, Paneth N. White matter damage in preterm newborns—an epidemiologic perspective. Early Hum Dev. 1990;24:1-22.
PUBMED
23. Sweet RL, Gibbs RS. Infectious Diseases of the Female Genital Tract. 3rd ed. Baltimore, Md: Williams & Wilkins; 1995:554.
24. Redline RW, Wilson-Costello D, Borawski E, Fanaroff AA, Hack M. Placental lesions associated with neurologic impairment and cerebral palsy in very low-birth-weight infants. Arch Pathol Lab Med. 1998;122:1091-1098.
ISI
| PUBMED
25. Kuban KC, Leviton A. Cerebral palsy. N Engl J Med. 1994;330:188-195.
FREE FULL TEXT
26. Woolf B. On estimating the relation between blood group and disease. Ann Hum Genet. 1955;19:251-253.
PUBMED
27. Rothman KJ, Greenland S. Modern Epidemiology. 2nd ed. Philadelphia, Pa: Lippincott-Raven; 1998:648.
28. al-Rajeh S, Bademosi O, Awada A, Ismail H, al-Shammasi S, Dawodu A. Cerebral palsy in Saudi Arabia: a case-control study of risk factors. Dev Med Child Neurol. 1991;33:1048-1052.
ISI
| PUBMED
29. Spinillo A, Capuzzo E, Stronati M, Ometto A, De Santolo A, Acciano S. Obstetric risk factors for periventricular leukomalacia among preterm infants. Br J Obstet Gynaecol. 1998;105:865-871.
ISI
| PUBMED
30. Evans P. Antecedents of infantile cerebral palsy. Arch Dis Child. 1948;23:213-219.
PUBMED
31. Roland EH, Magee JF, Rodriguez E, Lupton BA, Hill A. Placental abnormalities: insights into pathogenesis of cystic periventricular leukomalacia. Ann Neurol. 1996;40:3213.
32. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst. 1959;22:719-748.
ISI
| PUBMED
33. Petitti DB. Meta-analysis, decision analysis, and cost-effectiveness. In: Kelsey JL, Marmot MG, Stolley PD, Vessey MP, eds. Monographs in Epidemiology and Biostatistics. 2nd ed. Oxford, England: Oxford University Press; 2000.
34. Daniel WW. Biostatistics: A Foundation for Analysis in the Health Sciences. 6th ed. New York, NY: John Wiley & Sons Inc; 1995:224.
35. Cooper HM, Hedges LV. The Handbook of Research Synthesis. New York, NY: Russell Sage Foundation; 1994:402.
36. Murphy DJ, Hope PL, Johnson A. Ultrasound findings and clinical antecedents of cerebral palsy in very preterm infants. Arch Dis Child Fetal Neonatal Ed. 1996;74:F105-F109.
37. Murphy DJ, Hope PL, Johnson A. Neonatal risk factors for cerebral palsy in very preterm babies: case-control study. BMJ. 1997;314:404-408.
FREE FULL TEXT
38. Spinillo A, Capuzzo E, Orcesi S, Stronati M, Di Mario M, Fazzi E. Antenatal and delivery risk factors simultaneously associated with neonatal death and cerebral palsy in preterm infants. Early Hum Dev. 1997;48:81-91.
FULL TEXT
|
ISI
| PUBMED
39. Polivka B. Risk Factors for Cerebral Palsy: A Multivariate Model. Columbus: Ohio State University; 1990:132.
40. O'Shea TM, Klinepeter KL, Dillard RG. Prenatal events and the risk of cerebral palsy in very low birth weight infants. Am J Epidemiol. 1998;147:362-369.
FREE FULL TEXT
41. Bejar R, Wozniak P, Allard M, et al. Antenatal origin of neurologic damage in newborn infants, I: preterm infants. Am J Obstet Gynecol. 1988;159:357-363.
ISI
| PUBMED
42. Roboz P. Etiology of congenital cerebral palsy. Arch Pediatr. 1962;79:233-250.
PUBMED
43. Carbine DN. Neonatal CNS Injury and Developmental Outcome After Prolonged Premature Rupture of Membranes. San Francisco, Calif: Pediatric Academic Societies; 1999.
44. Verma U, Tejani N, Klein S, et al. Obstetric antecedents of intraventricular hemorrhage and periventricular leukomalacia in the low-birth-weight neonate. Am J Obstet Gynecol. 1997;176:275-281.
FULL TEXT
|
ISI
| PUBMED
45. de Vries LS, Regev R, Dubowitz LM, Whitelaw A, Aber VR. Perinatal risk factors for the development of extensive cystic leukomalacia. Am J Dis Child. 1988;142:732-735.
ABSTRACT
|
