Objective
The objective of the study was to describe neurodevelopmental outcome at the age of 4.5 years in 216 children, born after expectant management of severe early-onset hypertensive complications of pregnancy.
Study Design
This was a prospective follow-up study until age 4.5 years from maternal admission onward. Developmental outcome measurements included child intelligence quotient and behavioral, motor, and neurological outcome. Abnormal composite outcome (perinatal mortality or abnormal developmental outcome) was studied in relation to gestational age (GA), birthweight (BW), and perinatal variables.
Results
Fetal and neonatal mortality was 9% and 8%, respectively. Of the 178 survivors, 149 (84%) were seen for follow-up. Mean GA was 31.4 weeks and 90% were born growth restricted. Abnormal developmental outcome occurred in 20% and abnormal composite outcome in 37%.
Conclusion
Perinatal mortality or abnormal child development occurs in one third of pregnancies with early-onset and severe hypertensive complications and is highest in the lowest GA and BW ranges.
In early and severe hypertensive complications of pregnancy, treatment strategies have to balance between maternal and infant risks. A large number of cohort studies describe maternal and infant outcome after expectant management, aiming to carefully prolong pregnancy and thereby improve perinatal outcome without jeopardizing maternal safety, and a smaller number describe outcome after interventionist management, in which delivery is planned within 48 hours, after maternal stabilization and corticosteroid administration for fetal lung maturation. The results of these studies have recently been reviewed by Magee et al.
From the studies included in this review, it appears that maternal complications were infrequent and similar in both expectant and interventionist strategies, whereas maternal death was absent in these studies. Perinatal mortality and neonatal morbidity tended to be higher after interventionist care. A major disadvantage of these studies was the restriction to short-term infant outcome, whereas it is well known that short-term outcome measures are a poor predictor of a long-term outcome. More recently, it was suggested that expectant management was associated with increased maternal complications at a gestation of 32 weeks or longer.
Most reports on child outcome in relation to maternal hypertensive complications of pregnancy collect pregnancy data retrospectively. As a result, precise data regarding underlying maternal morbidity, classification of disease, maternal treatment strategies, prematurity and fetal growth restriction, and fetal monitoring are lacking. Often literature on outcome after small-for-gestational-age (SGA) status at birth is referred to, but not all infants born SGA are also growth restricted and vice versa.
This study presents the long-term developmental outcome, including cognition, neurological, and motor and behavioral outcome, in a, from maternal admission onward, prospective cohort of children born after expectant management of early-onset hypertensive disorders of pregnancy. The aim was to describe child outcomes including perinatal death and explore the associations of long-term outcomes with maternal and neonatal morbidity.
Materials and Methods
Participants and the PETRA project
All mothers participated in the Pre-eclampsia Eclampsia TRial Amsterdam (PETRA) study. The PETRA study was a randomized clinical trial that compared expectant management with or without plasma volume expansion in case of severe and early-onset hypertensive disorders of pregnancy. Inclusion criteria of the PETRA study were a viable singleton pregnancy at a gestational age between 24 and 34 completed weeks, complicated by fetal growth restriction in combination with pregnancy-induced hypertension, severe preeclampsia, HELLP (hemolysis, elevated liver enzymes, and low platelet count) syndrome, or eclampsia (definitions in Table 1 ).
| Inclusion diagnosis | Definition |
|---|---|
| Severe preeclampsia | Diastolic blood pressure ≥110 mm Hg and proteinuria (≥0.3 g per 24 h) |
| HELLP syndrome | Platelet count <100/10 9 per liter and aspartate aminotransferase ≥70 U/L and lactate dehydrogenase ≥600 U/L |
| Fetal growth restriction and pregnancy-induced hypertension | Estimated fetal weight <p10 and diastolic blood pressure ≥90 mm Hg |
The study took place in 2 tertiary care centers (ie, the Academic Medical Center and the Vrije Universiteit Medical Center, both located in Amsterdam, The Netherlands), in the period between April 1, 2000, and May 31, 2003. Of 216 patients included in the PETRA trial, 111 women were randomized to the plasma volume expansion group and 105 women to the control group. Details of the randomization process and original study design have been described elsewhere. The primary endpoint of the trial was a neurological assessment at corrected term age.
In all cases, there was expectant management, which consisted of intensive monitoring of fetal and maternal condition. Corticosteroids were administered to patients if delivery was thought to be imminent before 33 weeks, based on maternal disease deterioration or estimated fetal condition. All but 3 mothers of infants, who were delivered alive before 33 weeks, had been given corticosteroids. Fetal indications for delivery were repeated decelerations or prolonged low variability on fetal heart rate tracings. Maternal indications were therapy-resistant hypertension, pulmonary edema, and recurrent HELLP syndrome. Decisions to abstain from intervention were made, after extensive discussions between obstetricians, neonatologists, and parents if fetuses were deemed nonviable according to individualized assessments and predefined criteria.
From the 216 mother/child pairs enrolled in our trial, 174 children remained for follow-up at age 4.5 years ( Figure 1 ). They were all invited for follow-up assessments.
There were no statistically significant differences in primary and secondary maternal or neonatal short-term outcomes or 1 year child development between randomization groups; therefore, we intended to study 4.5 year outcome irrespective of trial allocation, after it had been checked that also at 4.5 year outcomes were not different between study groups.
Birthweight ratio (BWR) was calculated by birthweight divided by the expected weight for gestation age, using the Gardosi customized growth chart p50 value. Major maternal morbidity was defined as eclampsia, encephalopathy, pulmonary edema, liver hematoma, placental abruption, severe infectious morbidity, and severe thrombotic morbidity. Major neonatal morbidity was defined as chronic lung disease, grades 3 and 4 cerebral hemorrhage, grades 2 or higher periventricular leucomalacia, or hydrocephalus.
Follow-up procedures
At the corrected age of 4.5 years, the participating families were asked to visit 1 of the 2 outpatient follow-up clinics for sequential assessment of child intelligence, motor development, and neurological outcome. The parents were asked to complete a questionnaire about their child’s behavior (see the following text).
Intelligence quotient (IQ) was assessed using the Revised Amsterdam Child Intelligence Test (RAKIT), short form. The short form for children aged 4 years 2 months to 5 years 2 months years consists of 5 subtests, measuring visual perception, logical reasoning, word knowledge, visuomotor integration, and word fluency. This test was standardized for the Dutch and Flemish population in the early 1980s.
The standardized IQ index has a mean of 100. The IQ index ranges from 56 to 145. The correlation of the short version with the complete test is 0.93. Although qualitatively an excellent test, RAKIT norms have become outdated. In accordance with the literature on the Flynn effect, it is estimated that the mean IQ on intelligence tests improves about 3-5 points each decade. We therefore used a mean of 108 IQ index points as an estimated contemporary population mean.
In a recent study on school outcome of moderately preterm children, a mean of 108 was indeed found for the term control group. With a population SD of 15 index points, an IQ score of 93 or greater was classified as normal development, a score between 78 and 92 was classified as moderately delayed, and a score below 78 was classified as abnormal.
Motor assessment
Motor development was examined using the first version of the Movement Assessment Battery for Children (M-ABC). The M-ABC measures motor functioning in daily life. A score above the 15th percentile is considered normal; a score between the 15th and the 5th percentile is considered at risk and a score below the 5th percentile is considered abnormal.
Neurological assessment
Neurological development was assessed by using a method adapted from Touwen’s examination of the child with minor neurological dysfunction. This examination focuses on (minor) signs of neurological dysfunction. It examines hand function, quality of walking, postural control, passive muscle tone, and coordination. The child’s outcome was expressed as a percentage of the optimal score. Performance was rated as normal (75-100%), nonoptimal (50-74%), or abnormal (<50%). Cerebral palsy was diagnosed according to the latest standardized definitions.
Behavior
The child’s behavioral and emotional functioning was assessed with the Child Behaviour Check List (CBCL) for ages 1.5-5 years (CBCL 1.5-5). The CBCL 1.5-5 is a parent report form that consists of 99 statements about the child’s behavior. It yields scores for internalizing and externalizing behavior problems and a total problem score. Higher scores represent more behavior problems. Total problem scores are defined as normal (<82nd percentile), moderately problematic (82nd and 90th percentile), or abnormal (≥90th percentile).
Definition of subnormal and abnormal developmental outcome and abnormal composite outcome
Developmental outcome was scored if at least 2 development measures were available.
Subnormal developmental outcome was defined as an IQ between 78 and 93 and/or an M-ABC score between the 15th and the 5th percentile and/or a neurological examination total score between 50% and 75% and/or a borderline CBCL total problem score without any of the scores in the abnormal range.
Abnormal developmental outcome was defined as an IQ below 78 and/or an M-ABC score up or below the 5th percentile and/or a neurological examination total score below 50% and/or a clinical CBCL total problem score.
Abnormal composite outcome was defined as fetal or neonatal mortality or abnormal developmental outcome.
The Committee of Medical Ethics in both the Academic Medical Center and the Vrije Universiteit University Medical Center in Amsterdam (The Netherlands) approved this follow-up study.
Statistical analysis
Statistical analyses were performed with SPSS 16.0.2 (SPSS Inc, Chicago, IL). Differences were considered statistically significant at P < .05.
First, the association of the 4.5 year outcome with trial allocation was analyzed by using χ 2 tests (2 sided), Student t tests, and nonparametric Mann-Whitney U tests, when appropriate. Second, it was intended to analyze the complete cohort for further exploration of data in the absence of differences in relation to trial allocation, as has been done in earlier publications. For the complete cohort, logistic regression was done to find predictors of abnormal developmental outcome (abnormal vs not abnormal).
In addition, linear regression analyses were done with IQ, motor scores, neurological scores, and behavioral scores as 4 separate dependent outcomes. The sets entered in the stepwise regression analyses were as follows: The first set comprised sociodemographic variables (ie, maternal and paternal education and ethnic background). The second set comprised variables during maternal disease and childbirth (ie, HELLP syndrome, major maternal morbidity, the highest ratio of the umbilical artery divided by the median cerebral artery pulsatility index (U/C ratio), highest systolic and diastolic blood pressure, gestational age at delivery, BWR, sex, and Apgar score at 5 minutes less than 7). The third set comprised major neonatal morbidity.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
