Background
Cerebral palsy (CP) is a late sequel of pregnancy, and the role of preeclampsia is debatable.
Objective
The aims of this study were to determine the association between preeclampsia and cerebral palsy and to determine the risk factors for the development of cerebral palsy in these patients.
Study Design
A retrospective population-based cohort study was designed that included 229,192 singleton pregnancies. The study population was divided into 2 groups: (1) patients with preeclampsia (n = 9749) and (2) normotensive gestations (n = 219,443). Generalized Estimating Equation multiple logistic regression models were performed to study the associations among preeclampsia, small for gestational age, gestational age at delivery, and the risk factors for the development of cerebral palsy in neonates of women with preeclampsia.
Results
The rate of cerebral palsy was double in patients with preeclampsia than in the normotensive group (0.2% vs 0.1%; P = .015); early onset preeclampsia and small for gestational age were independent risk factors for the subsequent development of cerebral palsy (odds ratio, 8.639 [95% confidence interval, 4.269–17.480]; odds ratio, 2.737 [95% confidence interval, 1.937–3.868], respectively). A second model was conducted to determine the risk factors for the development of cerebral palsy in women with preeclampsia. Birth asphyxia, complications of prematurity, and neonatal infectious morbidity, but not small for gestational age or gestational age at delivery, were independent risk factors for the development of cerebral palsy.
Conclusion
In a comparison with normal pregnant women, the rate of cerebral palsy is double among patients with preeclampsia, especially those with early-onset disease. Early-onset preeclampsia is an independent risk factor for cerebral palsy. Among women with preeclampsia, the presence of neonatal infectious morbidity, birth asphyxia, and complications of prematurity are independent risk factors for the development of cerebral palsy, which further supports the role of a multi-hit model in the pathogenesis of this syndrome.
Premature birth, especially at <28 weeks of gestation, is the leading risk factor for the development of cerebral palsy (CP) at 2-3 years of age. This late sequel of pregnancy is a children’s disease that is a diagnostic term used to describe a group of permanent disorders of movement and posture that cause activity limitation. These disorders are attributed to nonprogressive disturbances in the developing fetal brain, alteration in fetal development, or pathologic intrauterine processes or are considered as prematurity complications. The prevalence of CP rises in a positive correlation with the severity of premature delivery and can reach up to 15% in preterm neonates who are delivered at 24–27 weeks of gestation. CP is the most common form of chronic motor disability that begins in childhood; its incidence varies from 1–3.6 per 1000 live births. The overall proportion of CP did not change in recent years; yet, it decreased in neonates who were born at term and increased in those who were delivered prematurely. Nevertheless, most of the children with CP are born at term. Indeed, in the Collaborative Prenatal Project research in which 45,000 7-year-old children were examined; most of those who had CP were born at term and had no complications during delivery.
Recent studies suggested that preeclampsia may be an additional risk factor for the subsequent development of CP. A Norwegian study reported that children born to mothers with preeclampsia were more likely to experience CP than those delivered of normotensive women. Preeclampsia, a major obstetric syndrome, is 1 of the leading causes for indicated preterm birth and perturbation of fetal growth, Indeed, mild preeclampsia is associated with a 5% reduction in fetal weight; in the severe form of this syndrome, it can reach up to 12%. Moreover, neonates of mothers who experience preeclampsia are 4 times more likely to be small for gestational age (SGA). Although the cause of CP is unknown, it is correlated strongly with preterm delivery and SGA ; it has been proposed that the delivery of an SGA neonate mediates the correlation between early onset preeclampsia and CP. Because both preeclampsia and SGA result, in many cases, from abnormal placental implantation, it might be that the processes that affect fetal growth in women with preeclampsia also predispose their fetuses to the development of CP. Therefore, the aims of this study were to determine (1) the association between preeclampsia and CP and (2) the risk factors for the development of CP in these patients.
Material and Methods
This is a retrospective population-based cohort study that includes all the deliveries that occurred at Soroka University Medical Center from 1990-2013 that met the inclusion criteria. Data were collected from an electronic database that included demographic, obstetric, and general information about the mother and fetus of all the deliveries at our medical center. The use of the database was possible because the Soroka University Medical Center is a tertiary medical center that serves the population of the Negev, and all the deliveries of the region take place in its labor and delivery suites.The Department of Obstetrics and Gynecology has a computerized database of all the deliveries. The information was captured from patient medical records and coded by trained secretaries according to the International Classification of Diseases, 9th edition (ICD-9) diagnosis. Our database is tested constantly and validated by the Department of Epidemiology at the Ben-Gurion University of the Negev (Beer Sheva, Israel). The study was approved by the Institutional Review Board Committee of the Soroka University Medical Center.
Patients who had perinatal death (ante-, intra- and postpartum death), multiple gestation, gestational hypertension, chronic hypertension without preeclampsia, or missing data were excluded from the study.
There were 229,192 pregnancies that met the inclusion criteria and comprised the 2 study groups: (1) pregnancies that were affected by preeclampsia (n = 9749) and (2) normotensive pregnancies (n = 219,443). Preeclampsia, CP, and all other diagnoses were coded according to ICD-9 codes (642.42 for mild preeclampsia, 642.52 for severe preeclampsia, 343.9 for CP).
Clinical definitions
Ethnicity was divided in Jews and Bedouins (an Arab ethnic group, descended from nomadic tribes who historically have inhabited the Arabian and Syrian Deserts). Parity groups were defined in the following manner: multiparous (2-5 deliveries) and grand-multiparous (≥6 deliveries). Preeclampsia was diagnosed in the presence of elevated blood pressure and proteinuria of at least +1 in dipstick. Its severity was defined according to the severity of hypertension and/or 1 of the following events: +3 proteinuria by dipstick; thrombocytopenia ≤ 100,000; elevated liver enzymes; persistent headache, and/or blurred vision. Gestational diabetes mellitus was diagnosed according to oral glucose tolerance test and classified according to White’s classification. Preterm delivery was defined as delivery before complete 37 weeks of gestation; late preterm birth was any delivery between 34 and 36 6/7 weeks.
Newborn infants were classified by their birthweight with the use of Leiberman sex-specific birth curves in the following manner: SGA, birthweight <10th percentile; appropriate for gestational age, birthweight from 10-90th percentile; and large for gestational age, birthweight >90th percentile, according to regional growth curves. Pathologic Apgar score was defined as <5 at 1 minute and <7 at 5 minutes. Neonatal acidosis was defined as cord blood pH< 7.0 and/or evidence of birth asphyxia.
Short-term neonatal complications were also diagnosed with ICD-9 codes and included pneumonia, jaundice, tachypnea, hypoglycemia, dyspnea, convulsions, polycythemia, bacteremia, sepsis, hemolysis, acidosis, hypothermia, infections, ventilation, intraventricular hemorrhage (IVH), retinopathy of prematurity (ROP), necrotizing enterocolitis (NEC), bronchopulmonary dysplasia (BPD), respiratory distress syndrome (RDS), packed cell transfusion, and urinary tract infection (UTI). CP was diagnosed by the ICD-9 code, and the diagnosis was made at the age of 3 years old.
Statistical analysis
Categoric variables were processed with Pearson’s Chi-square test. Normally distributed quantitative variables were analyzed with either T -Test or 1-way analysis of variance test. Not normally distributed quantitative variables were processed with the Mann-Whitney test or Kruskal-Wallis test.
Because the data in our study were collected during a period of 23 years, we have tried to clarify whether the rates of preeclampsia and CP changed over time in 2 manners. The first was to run a trend analysis; however, because there are a small number of cases in our study, such a trend analysis is not statically significant ( P = .711). We therefore included the year of delivery as a variable in our Generalized Estimating Equation (GEE) model ( Table 1 ), which was found to have a small but significantly protective effect against CP and thus suggests that the data collection and recording were uniform during this period and can be analyzed as 1 cohort.
| Variable | Odds ratio | 95% Confidence interval | P value |
|---|---|---|---|
| Late onset preeclampsia | 1.005 | 0.549–1.84 | .957 |
| Early onset preeclampsia (<34 weeks of gestation) | 8.707 | 4.301–17.628 | < .001 |
| Small for gestational age | 2.856 | 2.025–4.028 | < .001 |
| Maternal age | 0.972 | 0.946–0.998 | .034 |
| Multiparity | 1.329 | 0.981–1.799 | .066 |
| Year of delivery a | 0.906 | 0.887–0.924 | < .001 |
a Intended as the year the patient delivered (between 1990-2013).
Multivariate analysis was performed to test the association between preeclampsia and CP by using the GEE to account for multiple births by the same mother. Models were adjusted for confounders that were found statistically significant in the univariate analysis. In the first model, we defined CP as the dependent variable; late preeclampsia (>34 weeks of gestation), early onset preeclampsia (<34 weeks of gestation), maternal age, multiparity, SGA, and gestational age at delivery were included as independent variables.
The second GEE model was aimed to study the risk factors for CP among patients with preeclampsia. CP was set as the dependent variable; SGA, birth asphyxia, neonatal infectious morbidity (pneumonia, bacteremia, sepsis, UTI, other infections, chorioamnionitis, and sepsis), sequel of prematurity (BPD, NEC, IVH, ROP, RDS), and gestational age at delivery served as covariates.
The statistical analysis was performed with IBM Statistics SPSS software (version 20; SPSS Inc, Chicago, IL), and statistical significance was determined as a probability value of <.05.
Results
Table 2 presents the demographic characteristics of the patients. Women in the preeclampsia group were older (29.18 ± 6.525 vs 28.51 ± 5.77 years; P < .0001), had a lower median gravidity (median, 2 [range, 1–5] vs 3 [range, 2–5]; P < .0001) and parity (median, 2 [range, 1–4] vs 3 [range, 2–5]; P < .0001) and had a higher proportion of previous obstetric complications that included preterm birth (5.7% vs 4.4%; P < .0001), placental abruption (0.5% vs 0.4%; P = .016), and preeclampsia (12.4% vs 1.9%; P < .0001) than did normotensive pregnancies. In addition, the rate of assisted reproduction technologies (3.7% vs 1.7%; P < .0001) and the median length of hospitalization were higher in the preeclampsia group than in the normotensive group ( P < .0001; Table 2 ).
| Variable | Group | P value | |
|---|---|---|---|
| Preeclampsia (n = 9749) | Normotensive (n = 219,443) | ||
| Maternal age at delivery a | 29.18 ± 6.525 | 28.51 ± 5.77 | < .0001 |
| Bedouin origin, % (n) | 46.9 (4574) | 53.8 (118,158) | < .0001 |
| Gravidity b | 2 (1–5) | 3 (2–5) | < .0001 |
| Parity b | 2 (1–4) | 3 (2–5) | < .0001 |
| History of preterm birth, % (n) | 5.7 (552) | 4.4 (9,725) | < .0001 |
| Infertility treatment, % (n) | 3.7 (356) | 1.7 (3696) | < .0001 |
| Hospitalization, d b | 4 (1–35) | 2 (1–34) | < .0001 |
| History of placental abruption, % (n) | 0.5 (52) | 0.4 (823) | .016 |
| History of small for gestational age, % (n) | 4.1 (399) | 4.1 (8,952) | .990 |
| History of preeclampsia, % (n) | 12.4 (1206) | 1.9 (4,079) | < .0001 |
| Placental abruption history, % (n) | 0.5 (52) | 0.4 (823) | .016 |
a Data are given as mean ± standard deviation
Clinical characteristics of the study groups demonstrated that women with preeclampsia had a higher rate of placental abruption, pregestational and gestational diabetes mellitus, oligohydramnios, polyhydramnios, nonreassuring fetal heart monitoring during labor, cesarean delivery, and a lower mean gestational age at delivery than normotensive patients ( P < .0001 for all comparisons). In addition, women with preeclampsia had a significantly higher rate of both preterm and early preterm delivery than normotensive patients ( P < .0001; Table 3 ).
| Variable | Group | P value | |
|---|---|---|---|
| Preeclampsia (n = 9749) | Normotensive (n = 219,443) | ||
| Mode of delivery, % (n) | < .0001 | ||
| Forceps | 0.1 (7) | 0.004 (81) | |
| Cesarean section | 29.3 (2859) | 13.3 (29,164) | |
| Vacuum | 3.5 (342) | 3.2 (7,088) | |
| Vaginal | 67.1 (6541) | 83.4 (183,110) | |
| Placental abruption, % (n) | 1.7 (163) | 0.6 (1,291) | < .0001 |
| Placenta previa, % (n) | 0.3 (32) | 0.4 (915) | .2 |
| Abnormal presentation, % (n) | 5.2 (508) | 4 (8,678) | < .0001 |
| Gestational diabetes mellitus, % (n) | 7 (685) | 3.9 (8,634) | < .0001 |
| Diabetes mellitus, % (n) | 1.8 (180) | 0.6 (1,397) | < .0001 |
| Oligohydramnios, % (n) | 3.5 (339) | 2.4 (5,264) | < .0001 |
| Polyhydramnion, % (n) | 4.4 (432) | 3.6 (7,880) | < .0001 |
| Non reassuring fetal heart rate, % (n) | 3.1 (298) | 1.4 (3,164) | < .0001 |
| Gestational age at delivery, wk a | 38.03 ± 2.66 | 39.1 ± 1.914 | < .0001 |
| Preterm delivery <37 weeks, % (n) | 19.7 (1923) | 7 (15,401) | < .0001 |
| Early preterm delivery <34 weeks, % (n) | 5.8 (568) | 1.4 (3,236) | < .0001 |
In terms of early and late neonatal outcomes, neonates of women in the preeclampsia group had a lower mean birthweight and higher rate of SGA, Apgar score <5 at 1 and 5 minutes, and birth asphyxia compared with normotensive pregnancies ( P < .001 for all comparisons; Table 4 ). In addition, the rate of jaundice, tachypnea of the newborn infant, hypoglycemia, polycythemia, hemolysis, hypothermia, sepsis, ROP, NEC, BPD, and RDS was higher in patients from the preeclampsia than the normotensive group group ( Supplemental Table ). Among patients with preeclampsia, the rate of all neonatal morbidities that included convulsions ( P = .006), sepsis, ROP, NEC, BPD, and RDS ( P < .0001 for all comparisons) was higher in neonates of women with early-onset preeclampsia than in those with late-onset disease ( Table 5 ).
| Variable | Group | P value | |
|---|---|---|---|
| Preeclampsia (n = 9749) | Normotensive (n = 219,443) | ||
| Birthweight, g | 2958 ± 692.8 | 3207 ± 505 | < .0001 |
| Male, % (n) | 51.4 (5012) | 51.3 (112,615) | .99 |
| Intrauterine growth, % (n) | < .0001 | ||
| Small for gestational age | 11.1 (1078) | 5.3 (11,654) | |
| Appropriate for gestational age | 78.3 (7635) | 84.4 (185,165) | |
| Large for gestational age | 10 (975) | 9.6 (21,013) | |
| Cerebral palsy, % (n) | 0.2 (20) | 0.1 (253) | .015 |
| Apgar score, % (n) | |||
| 1 min <5 | 11.6 (1128) | 5.5 (11,993) | < .0001 |
| 5 min <5 | 0.6 (62) | 0.4 (780) | < .0001 |
| pH a | 7.35 ± 0.41 | 7.37 ± 1.07 | .974 |
| Asphyxia, % (n) | 1.5 (146) | 1.1 (2,487) | .001 |
| Variable | Preeclampsia group, % (n) | P value | |
|---|---|---|---|
| Early-onset (n = 789) a | Late-onset (n = 8960) b | ||
| Pneumonia | 8.9 (70) | 4.2 (380) | < .0001 |
| Jaundice | 51.8 (409) | 4.1 (368) | < .0001 |
| Tachypnea | 11.3 (89) | 1.6 (145) | < .0001 |
| Hypoglycemia | 9.9 (39) | 2.6 (237) | < .0001 |
| Dyspnea | 4.9 (39) | 1.9 (170) | < .0001 |
| Convulsions | 2.8 (22) | 1.5 (134) | .006 |
| Polycythemia | 2.5 (20) | 1.1 (101) | < .0001 |
| Bacteremia | 1.8 (14) | 0.8 (70) | .004 |
| Sepsis | 1.5 (12) | 0.4 (38) | < .0001 |
| Hemolysis | 0.4 (3) | 0.3 (28) | .519 |
| Acidosis | 4.3 (34) | 1 (87) | < .0001 |
| Hypothermia | 0.5 (4) | 0.1 (12) | .013 |
| Infections | 1 (8) | 0.2 (22) | < .0001 |
| Ventilation | 0.5 (4) | 0.2 (15) | .038 |
| Intraventricular hemorrhage | 3.4 (27) | 0 | < .0001 |
| Retinopathy of prematurity | 6.1 (48) | 0 | < .0001 |
| Necrotizing enterocolitis | 4.2 (33) | 0.1 (6) | < .0001 |
| Bronchopulmonary dysplasia | 5.6 (44) | 0 | < .0001 |
| Respiratory distress syndrome | 33.8 (267) | 0.3 (28) | < .0001 |
| Packed cells transfusion | 22.6 (178) | 0.7 (63) | < .0001 |
| Urinary tract infection | 0.6 (5) | 0.2 (16) | .008 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
