Key Points
Question
What is the association between body mass index during early pregnancy and the risk of childhood epilepsy?
Findings
In this population-based cohort study that included 1.4 million live single births, we found that maternal overweight and obesity are associated with increased risks of childhood epilepsy. Greater risks were noted with increasing body mass index.
Meaning
Given that overweight and obesity are modifiable, prevention of obesity during early pregnancy may be an important public health strategy to reduce the incidence of childhood epilepsy.
Abstract
Importance
There is growing concern about the long-term neurologic effects of prenatal exposure to maternal overweight and obesity. The causes of epilepsy are poorly understood and, in more than 60% of the patients, no definitive cause can be determined.
Objectives
To investigate the association between early pregnancy body mass index (BMI) and the risk of childhood epilepsy and examine associations between obesity-related pregnancy and neonatal complications and risks of childhood epilepsy.
Design, Setting, and Participants
A population-based cohort study of 1 441 623 live single births at 22 or more completed gestational weeks in Sweden from January 1, 1997, to December 31, 2011, was conducted. The diagnosis of epilepsy as well as obesity-related pregnancy and neonatal complications were based on information from the Sweden Medical Birth Register and National Patient Register. Multivariate Cox proportional hazards regression models were used to estimate adjusted hazard ratios (HRs) and 95% CIs after adjusting for maternal age, country of origin, educational level, cohabitation with partner, height, smoking, maternal epilepsy, and year of delivery. Data analysis was conducted from June 1 to December 15, 2016.
Main Outcomes and Measures
Risk of childhood epilepsy.
Results
Of the 1 421 551 children born between January 1, 1997, and December 31, 2011, with covariate information available, 7592 (0.5%) were diagnosed with epilepsy through December 31, 2012. Of these 3530 (46.5%) were female. The overall incidence of epilepsy in children aged 28 days to 16 years was 6.79 per 10 000 child-years. Compared with offspring of normal-weight mothers (BMI 18.5 to <25.0), adjusted HRs of epilepsy by maternal BMI categories were as follows: overweight (BMI 25.0 to <30.0), 1.11 (95% CI, 1.04-1.17); obesity grade I (BMI 30.0 to <35.0), 1.20 (95% CI, 1.10-1.31); obesity grade II (BMI 35.0 to <40.0), 1.30 (95% CI, 1.12-1.50); and obesity grade III (BMI≥40.0), 1.82 (95% CI, 1.46-2.26). The rates of epilepsy were considerably increased for children with malformations of the nervous system (adjusted HR, 46.4; 95% CI, 42.2-51.0), hypoxic ischemic encephalopathy (adjusted HR, 23.6; 95% CI, 20.6-27.1), and neonatal convulsions (adjusted HR, 33.5; 95% CI, 30.1-37.4). The rates of epilepsy were doubled among children with neonatal hypoglycemia (adjusted HR, 2.10; 95% CI, 1.90-2.33) and respiratory distress syndrome (adjusted HR, 2.43; 2.21-2.66), and neonatal jaundice was associated with more than a 50% increased risk of epilepsy (adjusted HR, 1.47; 95% CI, 1.33-1.63). The elevated risk of epilepsy in children of overweight or obese mothers was not explained by obesity-related pregnancy or neonatal complications.
Conclusions and Relevance
The rates of childhood epilepsy increased with maternal overweight or obesity in a dose-response manner. Given that overweight and obesity are modifiable, prevention of obesity may be an important public health strategy to reduce the incidence of childhood epilepsy.
This population-based study evaluates the incidence of childhood-onset epilepsy in mothers who were overweight or obese during early pregnancy.
Introduction
Epilepsy is one of the most common neurologic disorders of childhood. Currently, 50 million persons are affected by epilepsy worldwide. The cause of epilepsy is poorly understood and no definitive cause can be determined in more than 60% of the cases. Preterm birth, low birth weight for gestational age, chromosomal and congenital anomalies, neonatal convulsions, and low Apgar scores are generally associated with increased risks of childhood epilepsy.
Maternal overweight and obesity have increased globally during the past 3 decades, and the prevalence of severe obesity, commonly defined as obesity classes II to III (body mass index [BMI]≥35.0 [calculated as weight in kilograms divided by height in meters squared]), has rapidly increased. There is growing concern about the long-term neurologic effects of children exposed to maternal obesity. Obesity in pregnancy has been linked to systemic inflammation, altered endocrine responses, folic acid deficiency, and insulin resistance. Risks of severe birth asphyxia, birth injuries, and congenital anomalies increase with maternal obesity, which in turn may influence long-term neurologic and intellectual development. A few studies have found increased risks of intellectual disability, attention-deficit/hyperactivity disorder, autism, and cerebral palsy in children whose mothers were overweight or obese during pregnancy. Only 1 previous study examined the association between maternal BMI and risk of epilepsy in the offspring, and no association was found.
We investigated the association between early pregnancy BMI and risk of childhood epilepsy in a nationwide cohort study including more than 1.4 million live singleton births in Sweden. We also examined associations between obesity-related pregnancy and neonatal complications and the risks of childhood epilepsy.
Methods
Study Design and Population
This retrospective, nationwide cohort study included all live singleton births at 22 or more completed gestational weeks in Sweden from January 1, 1997, through December 31, 2011. With the use of the person-unique national registration numbers of mothers and their offspring, information on births included in the Swedish Medical Birth Register was cross-linked with the National Patient Register, Cause of Death Register, Total Population Register, and Education Register. The Medical Birth Register includes data on more than 98% of all births in Sweden. This population-based, clinically focused database contains information on antenatal, obstetrical, and neonatal care that is prospectively recorded on standardized forms. The National Patient Register contains diagnostic codes on hospital inpatient care since 1987 and hospital outpatient care from 2001. The Cause of Death Register includes information on all deaths in Sweden. Diagnoses and causes of death since 1997 are coded using the Swedish version of the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10). Maternal educational level and country of origin were obtained from the Education Register and the Total Population Register. The study was approved by the Research Ethics Committee at Karolinska Institutet, Stockholm, Sweden. This study was based on encrypted data, for which the ethics committees do not require informed consent.
Exposures
Maternal BMI in early pregnancy was calculated from self-reported height and weight measured in light indoor clothing without shoes at the first antenatal visit, which occurs within the first 14 weeks of gestation (ie, first trimester) for 90% of pregnant women in Sweden. We computed the median height across pregnancies for multiparous women to reduce measurement error. Maternal BMI was categorized according to the World Health Organization as underweight (BMI≤18.4), normal weight (18.5 to <25.0), overweight (25.0 to <30.0), obesity grade I (30.0 to <35.0), obesity grade II (35.0 to <40.0), and obesity grade III (≥40.0).
Maternal characteristics included age at delivery, country of origin, educational level, cohabitation with a partner, parity, height, smoking during pregnancy, maternal epilepsy, and year of delivery. Maternal age at delivery was calculated as the date of delivery minus the mother’s birth date. Parity was defined as the number of births of each mother. Information on cohabitation with the mother’s partner was obtained at the first antenatal visit. Women who reported daily smoking at the first antenatal visit and/or at 30 to 32 gestational weeks were classified as smokers, whereas women who stated that they were nonsmokers were classified as nonsmokers. Maternal epilepsy before the child’s birth was defined using the same algorithm that was used for children (eTable 1 in the Supplement). Pregnancy complications included hypertension (pregestational and preeclampsia), diabetes (gestational and mellitus), and infection-related disorders.
Information about the newborn included birth weight, gestational age, sex, congenital malformations, birth trauma, neonatal asphyxia-related complications, and more common neonatal complications. Categorizations of the covariates are provided in Table 1 and eTable 2 in the Supplement; specific ICD-10 codes for diseases in the mothers and children are provided in eTable 1 in the Supplement.
Table 1. Incidence Rates and HRs of Epilepsy According to Maternal Diseases, Pregnancy Complications, and Neonatal Characteristics.
| Characteristic | No. of Children | Epilepsy | HR (95% CI) | ||
|---|---|---|---|---|---|
| No. | Rate per 10 000 Child-years | Unadjusteda | Adjustedb | ||
| Maternal Diseases and Pregnancy Complications | |||||
| Maternal epilepsy | |||||
| No | 1 416 537 | 7494 | 6.72 | 1 [Reference] | 1 [Reference] |
| Yes | 5014 | 98 | 30.9 | 4.44 (3.62-5.45) | 4.40 (3.56-5.43) |
| P value for trendc | <.001 | <.001 | |||
| Diabetic disease | |||||
| No | 1 401 240 | 7463 | 6.77 | 1 [Reference] | 1 [Reference] |
| Gestational diabetes | 13 999 | 85 | 8.28 | 1.21 (0.98-1.50) | 1.25 (1.0-1.56) |
| Pregestational diabetes | 6312 | 44 | 9.14 | 1.35 (1.0-1.81) | 1.30 (0.94-1.78) |
| P value | .033 | .04 | |||
| Hypertensive disease | |||||
| No | 1 372 938 | 7272 | 6.74 | 1 [Reference] | 1 [Reference] |
| Pregestational hypertension | 8780 | 41 | 6.76 | 0.98 (0.72-1.33) | 0.89 (0.64-1.23) |
| Preeclampsia | 39 833 | 279 | 8.75 | 1.30 (1.16-1.47) | 1.24 (1.09-1.41) |
| P value | <.001 | .003 | |||
| Chorioamnionitis | |||||
| No | 1 418 919 | 7559 | 6.78 | 1 [Reference] | 1 [Reference] |
| Yes | 2632 | 33 | 16.4 | 2.41 (1.71-3.40) | 2.22 (1.53-3.21) |
| P value | <.001 | <.001 | |||
| Maternal infection | |||||
| No | 1 412 398 | 7526 | 6.78 | 1 [Reference] | 1 [Reference] |
| Yes | 9153 | 66 | 9.49 | 1.39 (1.10-1.78) | 1.40 (1.09-1.80) |
| P value | .007 | .009 | |||
| Birth Characteristics and Neonatal Complications | |||||
| Mode of delivery | |||||
| Vaginal noninstrumental | 1 090 372 | 5363 | 6.19 | 1 [Reference] | 1 [Reference] |
| Vaginal instrumental | 108 268 | 593 | 7.06 | 1.14 (1.04-1.24) | 1.14 (1.04-1.25) |
| Elective cesarean section | 107 501 | 776 | 9.79 | 1.56 (1.45-1.68) | 1.51 (1.40-1.64) |
| Emergency cesarean section | 108 699 | 785 | 9.65 | 1.54 (1.43-1.66) | 1.53 (1.41-1.66) |
| Data missing | 6711 | 75 | 11.06 | ||
| P value for trendd | <.001 | <.001 | |||
| Newborn’s sex | |||||
| Male | 731 102 | 4062 | 7.07 | 1 [Reference] | 1 [Reference] |
| Female | 690 449 | 3530 | 6.50 | 0.92 (0.88-0.96) | 0.91 (0.87-0.96) |
| P value | <.001 | <.001 | |||
| Gestational age at delivery, wk | |||||
| Term (≥37) | 1 348 059 | 6802 | 6.42 | 1 [Reference] | 1 [Reference] |
| Moderately preterm (32-36) | 59 434 | 530 | 11.31 | 1.76 (1.61-1.93) | 1.69 (1.53-1.86) |
| Very preterm (28-31) | 6025 | 107 | 22.94 | 3.56 (2.94-4.32) | 3.50 (2.84-4.31) |
| Extremely preterm (22-27) | 2484 | 65 | 37.16 | 5.72 (4.48-7.30) | 5.51 (4.17-7.28) |
| Data missing | 5549 | 88 | 16.52 | ||
| P value for trend | <.001 | <.001 | |||
| Birth weight for gestational age (percentiles) | |||||
| <3 | 22 482 | 322 | 18.3 | 2.86 (2.56-3.20) | 2.75 (2.44-3.11) |
| 3 to <10 | 66 773 | 460 | 8.92 | 1.39 (1.26-1.53) | 1.30 (1.17-1.45) |
| 10 to 90 | 1 150 500 | 5769 | 6.40 | 1 [Reference] | 1 [Reference] |
| >90 to 97 | 123 167 | 621 | 6.25 | 0.98 (0.90-1.07) | 0.99 (0.91-1.08) |
| >97 | 53 080 | 332 | 7.72 | 1.21 (1.08-1.35) | 1.16 (1.03-1.30) |
| Data missing | 5549 | 88 | 16.5 | ||
| P value for trend | <.001 | <.001 | |||
| Birth trauma | |||||
| No | 1 395 887 | 7399 | 6.75 | 1 [Reference] | 1 [Reference] |
| Yes | 25 664 | 193 | 8.70 | 1.31 (1.13-1.51) | 1.26 (1.09-1.47) |
| P value | <.001 | .003 | |||
| Any congenital malformation | |||||
| No | 1 337 941 | 5960 | 5.67 | 1 [Reference] | 1 [Reference] |
| Yes | 83 610 | 1623 | 24.8 | 4.38 (4.14-4.63) | 4.28 (4.04-4.54) |
| P value | <.001 | <.001 | |||
| Chromosomal abnormalities | |||||
| No | 1 417 925 | 7269 | 6.52 | 1 [Reference] | 1 [Reference] |
| Yes | 3626 | 323 | 120.4 | 18.4 (16.4-20.6) | 18.6 (16.5-20.9) |
| P value | <.001 | <.001 | |||
| Circulatory malformations | |||||
| No | 1 393 495 | 7133 | 6.50 | 1 [Reference] | 1 [Reference] |
| Yes | 28 056 | 459 | 22.2 | 3.38 (3.08-3.72) | 3.29 (2.98-3.64) |
| P value | <.001 | <.001 | |||
| Nervous system malformations | |||||
| No | 1 418 642 | 6988 | 6.26 | 1 [Reference] | 1 [Reference] |
| Yes | 2909 | 604 | 300.4 | 47.7 (43.7-52.1) | 46.4 (42.2-51.0) |
| P value | <.001 | <.001 | |||
| Meconium aspiration | |||||
| No | 1 419 773 | 7545 | 6.76 | 1 [Reference] | 1 [Reference] |
| Yes | 1778 | 47 | 34.10 | 5.04 (3.78-6.73) | 4.51 (3.28-6.18) |
| P value | <.001 | <.001 | |||
| Hypoxic ischemic encephalopathy and related conditions | |||||
| No | 1 419 254 | 7329 | 6.57 | 1 [Reference] | 1 [Reference] |
| Yes | 2297 | 263 | 159.8 | 24.2 (21.3-27.5) | 23.6 (20.6-27.1) |
| P value | <.001 | <.001 | |||
| Neonatal convulsions | |||||
| No | 1 418 805 | 7158 | 6.42 | 1 [Reference] | 1 [Reference] |
| Yes | 2746 | 434 | 227.4 | 35.2 (31.9-39.0) | 33.5 (30.1-37.4) |
| P value | <.001 | <.001 | |||
| Birth asphyxia | |||||
| No | 1 407 317 | 7212 | 6.52 | 1 [Reference] | 1 [Reference] |
| Yes | 14 234 | 380 | 32.8 | 5.05 (4.56-5.61) | 4.79 (4.28-5.36) |
| P value | <.001 | <.001 | |||
| Apgar score at 5 min | |||||
| 7-10 | 1 398 680 | 7062 | 6.42 | 1 [Reference] | 1 [Reference] |
| 4-6 | 10 521 | 257 | 32.7 | 5.06 (4.47-5.73) | 4.91 (4.30-5.61) |
| 0-3 | 2189 | 132 | 78.4 | 12.3 (10.3-14.6) | 11.8 (9.79-14.2) |
| Data missing | 10 161 | 141 | 16.1 | ||
| P value for trend | <.001 | <.001 | |||
| Other Neonatal Complications | |||||
| Neonatal infections | |||||
| No | 1 400 373 | 7276 | 6.62 | 1 [Reference] | 1 [Reference] |
| Yes | 21 178 | 316 | 17.8 | 2.72 (2.43-3.05) | 2.60 (2.30-2.93) |
| P value | <.001 | <.001 | |||
| Neonatal hypoglycemia | |||||
| No | 1 383 582 | 7163 | 6.59 | 1 [Reference] | 1 [Reference] |
| Yes | 37 969 | 429 | 14.1 | 2.14 (1.94-2.36) | 2.10 (1.90-2.33) |
| P value | <.001 | <.001 | |||
| Neonatal jaundice | |||||
| No | 1 362 345 | 7131 | 6.65 | 1 [Reference] | 1 [Reference] |
| Yes | 59 206 | 461 | 10.0 | 1.50 (1.37-1.65) | 1.47 (1.33-1.63) |
| P value | <.001 | <.001 | |||
| Respiratory distress | |||||
| No | 13 77 310 | 7034 | 6.49 | 1 [Reference] | 1 [Reference] |
| Yes | 44 241 | 558 | 16.5 | 2.53 (2.32-2.76) | 2.43 (2.21-2.66) |
| P value | <.001 | <.001 | |||
Abbreviation: HR, hazard ratio.
From unadjusted Cox proportional hazards models. Robust estimates of the variance were calculated to account for women who delivered more than once during the study period.
From Cox proportional hazards models with robust estimates of variance. Covariates adjusted for maternal age, country of origin, educational level, cohabitation with a partner, parity, height, smoking during pregnancy, maternal epilepsy, and year of delivery.
Wald χ2 test.
Wald test when a variable representing ordinal categories of BMI was introduced into the model as a continuous predictor.
In Sweden, all women are offered ultrasonography scanning at 18 gestational weeks or earlier, and gestational age (in completed weeks) was estimated using the following hierarchy: the date of early second trimester ultrasonography (87.7%), the date of the last menstrual period (7.4%), or a postnatal assessment (4.9%).
In the study population of 1 441 623 live singleton births, we excluded 2334 (0.2%) children who died before age 28 days and 17 738 (1.2%) children with missing maternal and infant personal registration numbers. We also excluded 168 645 children (11.7%) for whom information on maternal BMI was missing. There were 1 421 551 births with complete information on covariates.
Outcome
The definition of epilepsy in children was based on the 2011 recommendations of the International League Against Epilepsy’s Commission on Epidemiology, using a combination of hospital and outpatient diagnoses from the National Patient Registry. A case of epilepsy was identified if it met any of the following conditions between January 1, 1997, and December 31, 2012:
an occurrence of 2 or more ICD-10 codes G40 and/or G41 (epilepsy and status epilepticus, respectively) on separate dates or
an occurrence of 1 or more ICD-10 code R56 (convulsions not elsewhere classified) and 1 or more ICD-10 codes G40 and/or G41 on separate dates; diagnosis of R56 had to precede that of G40 and/or G41.
The date associated with the first record of epilepsy was considered the date of disease onset. Diagnosis of epilepsy was restricted to the period after the child’s 27th day from birth.
Statistical Analysis
Incidence rates of epilepsy in children were quantified by dividing the number of children with epilepsy by the total follow-up time in child-years. We followed up each child from the 28th day of life until the date of the first diagnosis of epilepsy, emigration, death, or end of follow-up (December 31, 2012), whichever occurred first.
We compared crude rates between categories of maternal BMI and other maternal characteristics as well as pregnancy and neonatal complications for the whole cohort and for children born at term (≥37 weeks). To account for the varying lengths of follow-up, we calculated hazard ratios (HRs) and 95% CIs with the use of Cox proportional hazards models. Multivariate Cox proportional hazards regression was used to compare the rates of epilepsy between BMI categories and by levels of maternal and neonatal risk factors. We specified the robust sandwich estimate of the covariance matrix to account for the correlations of sequential births to the same mother in the study. The adjusted Cox proportional hazards model included maternal age, country of origin, educational level, cohabitation with partner, height, smoking, maternal epilepsy, and year of delivery. A linear trend in the association of BMI with the offspring’s epilepsy in the child was assessed using the Wald test by introducing a variable representing ordinal categories of BMI as a continuous predictor into the models. The percentage hazard difference per BMI unit was estimated as 100% × (HR − 1) from models with the BMI as a continuous variable. Interaction analyses were performed to investigate whether the association between maternal BMI and rates of epilepsy was different in children born preterm (≤36) or term (≥37 weeks). The interaction term was tested with the use of the χ2 test.
We conducted additional analyses to ascertain potentially mediating effects of obesity-related pregnancy and/or neonatal complications by introducing them as covariates into the multivariate models. These analyses assumed a lack of interaction between maternal BMI and maternal and neonatal complications and no unmeasured confounders of the potential mediators and rates of epilepsy. Regression model fit was assessed using the likelihood ratio test, and a 2-sided P value <.05 was used to determine statistical significance. All analyses were carried out with the use of SAS, version 9.4 (SAS Institute). Data analysis was conducted from June 1 to December 15, 2016.
Results
Of 1 421 551 children born between 1997 and 2011, 7592 children (0.5%) were diagnosed with epilepsy through December 31, 2012. A total of 3530 (46.5%) of the children were female. The overall incidence of epilepsy was 6.79 per 10 000 child-years. The rates and HRs of epilepsy were slightly higher among offspring of mothers who were born outside the Nordic countries, had lower educational levels, lived without a partner, were multiparous (≥4 births), or were smokers (eTable 2 in the Supplement).
A maternal diagnosis of epilepsy was associated with a more than 4-fold increase in epilepsy rates in the offspring (Table 1). Pregnancy complications (gestational diabetes, preeclampsia, infections, and chorioamnionitis), cesarean delivery, and birth trauma were also associated with increased rates of epilepsy. Most children who developed epilepsy were born at term, but the rates of epilepsy increased by decreasing gestational age, ranging from a 1.7-fold increased rate in moderately preterm infants (32-36 weeks) to an almost 5-fold increased rate in extremely preterm (22-27 weeks) infants.
The overrepresentation of congenital malformations among children with epilepsy was mainly attributed to malformations of the central nervous system (Table 1). Neonatal convulsions, hypoxic ischemic encephalopathy, birth asphyxia, and low Apgar scores (0-3 or 4-6) at 5 minutes were associated with considerably increased epilepsy rates. Rates of epilepsy were also increased among children with neonatal infections, hypoglycemia, respiratory distress syndrome, and jaundice. Restricting the analyses to children born at term did not change our results (eTable 3 in the Supplement).
The adjusted rate of epilepsy increased linearly by maternal BMI (Table 2) from 6.30 per 10 000 child-years among normal-weight women to 12.4 per 10 000 child-years among women with obesity grade III. The rates of epilepsy increased by 11% in children of overweight mothers compared with children of normal-weight mothers; obesity grades I and II were associated with 20% and 30% increased rates, respectively; and obesity grade III was associated with an 82% increase in epilepsy (Table 2). In term births, the rates of epilepsy increased consistently with maternal overweight and obesity severity; only obesity grade III was associated with epilepsy in preterm births (test for interaction P = .89) (Table 3).
Table 2. Incidence Rates and HRs of Epilepsy According to Maternal Early Pregnancy BMI .
| Characteristic | No. of Children | Child-years of Follow-up | No. of Casesa | Rate per 10 000 Child-years | HR (95% CI) | |
|---|---|---|---|---|---|---|
| Unadjustedb | Adjustedc | |||||
| Maternal BMI | ||||||
| <18.5 | 30 733 | 240 213 | 152 | 6.33 | 1.01 (0.85-1.19) | 1.00 (0.85-1.18) |
| 18.5-<25.0 | 786 811 | 6 163 602 | 3880 | 6.3 | 1 [Reference] | 1 [Reference] |
| 25.0-<30.0 | 315 486 | 2 430 474 | 1739 | 7.15 | 1.13 (1.07-1.20) | 1.11 (1.04-1.17) |
| 30.0-<35.0 | 99 737 | 740 447 | 590 | 7.97 | 1.26 (1.15-1.37) | 1.20 (1.10-1.31) |
| 35.0-<40.0 | 29 837 | 214 899 | 187 | 8.7 | 1.36 (1.18-1.58) | 1.30 (1.12-1.50) |
| ≥40.0 | 10 374 | 71 613 | 89 | 12.4 | 1.94 (1.56-2.40) | 1.82 (1.46-2.26) |
| Missing | 148 573 | 13 15 139 | 955 | 7.26 | ||
| P value for trendd | <.001 | <.001 | ||||
| Hazard difference per unit BMI, % | 2.4 (1.8-2.9) | 2.0 (1.5-2.6) | ||||
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HR, hazard ratio.
Number of epilepsy diagnoses by age: younger than 2 years, 2097 (27.6%); 2 to younger than 4 years, 1434 (18.9%); 4 to younger than 6 years, 1346 (17.7%); 6 to younger than 8 years, 993 (13.1%); 8 to younger than 10 years, 829 (10.9%); and 10 years or older, 893 (11.8%). Median age at diagnosis was 4.0 years (interquartile range, 1-7 years). The last case was diagnosed when the patient was 16 years.
From a Cox proportional hazards model with age at diagnosis of epilepsy as the outcome. Analyses included 6637 cases among 1 272 978 children.
From a Cox proportional hazards model adjusted for maternal age, country of origin, educational level, cohabitation with a partner, parity, height, smoking during pregnancy, maternal epilepsy, and year of delivery. Complete case analyses including 6422 cases among 1 228 641 children.
Wald test when a variable representing ordinal categories of BMI was introduced into the model as a continuous predictor.
Table 3. Incidence Rates and HRs of Epilepsy In Childhood According to Early Pregnancy BMI Stratified by Gestational Age at Delivery.
| Characteristic | Deliverya | |||||
|---|---|---|---|---|---|---|
| Preterm (n = 1 348 059) | Term (n = 67 943) |
|||||
| No. | Rate per 10 000 Child-years | Adjusted HR (95% CI)b (n = 1 168 495) | No. | Rate per 10 000 Child-years | Adjusted HR (95% CI)b (n = 56 236) | |
| Maternal BMI, total | ||||||
| <18.5 | 14 | 9.63 | 0.77 (0.45-1.32) | 135 | 6.00 | 0.99 (0.83-1.19) |
| 18.5-<25.0 | 337 | 12.6 | 1 [Reference] | 3516 | 5.99 | 1 [Reference] |
| 25.0-<30.0 | 134 | 11.9 | 0.92 (0.75-1.13) | 1577 | 6.83 | 1.12 (1.05-1.19) |
| 30.0-<35.0 | 55 | 14.1 | 1.03 (0.77-1.38) | 533 | 7.64 | 1.21 (1.11-1.33) |
| 35.0-<40.0 | 22 | 16.8 | 1.21 (0.79-1.88) | 163 | 8.13 | 1.28 (1.09-1.50) |
| ≥40.0 | 13 | 26.7 | 1.91 (1.10-3.33) | 75 | 11.4 | 1.76 (1.39-2.23) |
| Missing | 127 | 15.8 | 803 | 6.58 | ||
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HR, hazard ratio.
Analyses are stratified by gestational age at delivery. Preterm was considered 22 to 36 weeks’ gestation; term, 37 or more weeks’ gestation. The number of children denotes the sample size for complete case analyses including only women with nonmissing values in all covariates.
From Cox proportional hazards models with robust estimates of variance. Covariates adjusted for maternal age, country of origin, educational level, cohabitation with a partner, parity, height, smoking during pregnancy, maternal epilepsy, and year of delivery.
In the mediation analyses of BMI and epilepsy, neither obesity-related pregnancy nor neonatal complications substantially changed the association between maternal BMI and rates of epilepsy (Table 4, adjusted models 2 and 3). When both neonatal and obesity-related pregnancy complications were added to the multivariate model, the association between maternal BMI and rates of epilepsy was attenuated, but not eliminated (Table 4, adjusted model 4). The associations also persisted after excluding children with cerebral palsy (eTable 4 in the Supplement).
Table 4. Early Pregnancy BMI and Adjusted HRs for Epilepsy in Children Born at Term.
| Characteristic | Adjusted HR (95% CI) | |||
|---|---|---|---|---|
| Model 1a | Model 2b | Model 3c | Model 4d | |
| Maternal BMI | ||||
| <18.5 | 1.00 (0.83-1.19) | 1.00 (0.84-1.19) | 1.00 (0.83-1.19) | 1.00 (0.84-1.19) |
| 18.5 to <25.0 | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| 25.0 to <30.0 | 1.12 (1.05-1.19) | 1.10 (1.03-1.17) | 1.07 (1.01-1.14) | 1.06 (1.00-1.13) |
| 30.0 to <35.0 | 1.22 (1.11-1.34) | 1.18 (1.08-1.30) | 1.13 (1.03-1.24) | 1.11 (1.09-1.22) |
| 35.0 to <40.0 | 1.28 (1.09-1.50) | 1.22 (1.04-1.44) | 1.15 (0.98-1.35) | 1.12 (0.96-1.32) |
| ≥40.0 | 1.76 (1.39-2.24) | 1.66 (1.31-2.11) | 1.54 (1.21-1.95) | 1.48 (1.17-1.88) |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HR, hazard ratio.
Covariates adjusted for maternal age, country of origin, educational level, cohabitation with a partner, parity, height, smoking during pregnancy, and year of delivery.
Covariates adjusted for all variables included in model 1 and for any obesity-related pregnancy complications (gestational diabetes or diabetes mellitus, chronic hypertension, preeclampsia, maternal epilepsy, infection during pregnancy, and mode of delivery).
Covariates adjusted for all variables included in model 1 and for any neonatal complications: small birth weight for gestational age, Apgar score of between 0 and 6, neonatal infection, congenital malformation, asphyxia-related neonatal complications (including meconium aspiration, hypoxic ischemic encephalopathy and related conditions, and neonatal convulsions), birth asphyxia, neonatal hypoglycemia, respiratory distress, and birth trauma.
Adjusted for all variables included in model 2 and model 3.
Discussion
In this large, population-based study, the rate of childhood epilepsy increased with maternal overweight and obesity severity in a dose-response manner. We also found strong evidence that several asphyxia-related neonatal complications are associated with substantially increased rates of epilepsy and that the less severe but more prevalent neonatal complications, including neonatal jaundice, hypoglycemia, and respiratory distress, could potentially increase the rates of childhood epilepsy. However, the elevated rates of epilepsy in offspring of overweight or obese mothers could not be explained by obesity-related pregnancy or neonatal complications.
Overall, we found a steady increase in the rate of childhood epilepsy across the entire range of maternal BMI. We are aware of only 1 previous study investigating this issue, and no association was found between maternal obesity and the risk of childhood epilepsy. Reliance on a non–population-based data source (ie, Medicaid records) and use of a strict case ascertainment for diagnosis of epilepsy may have induced an underestimation of the risk of childhood epilepsy and contributed to the lack of association. In our population-based study, the definition of epilepsy was based on international recommendations.
Few studies have reported associations between prepregnancy overweight and obesity and other long-term neurodevelopmental outcomes of the children, such as intellectual disability, lower cognitive performance, attention-deficit/hyperactivity disorder, cerebral palsy, and autism. Given the importance and sensitivity of fetal brain development and its susceptibility to hormonal and inflammatory reactions, even at the early embryonic stage, it is possible that maternal overweight and obesity increase the risk of brain injury, leading to a range of neurodevelopmental disorders. A meta-analysis reported a 2-fold increased risk of major malformations in the central nervous system, specifically neural tube defects, in the children of obese mothers. This finding lends support to the hypothesis that epilepsy may result from the gradual accumulation of environmental insults to the central nervous system. Another potential mechanism by which maternal obesity might affect neurodevelopment could be through obesity-induced inflammation. There is increasing evidence that maternal inflammatory conditions are associated with fetal systemic inflammation, which increases the risk of neonatal brain injury leading to outcomes such as cerebral palsy and epilepsy. In our study, fetal exposure to maternal infections and chorioamnionitis were associated with overall increased risks of epilepsy, but not when the analysis was restricted to term births. Maternal obesity during pregnancy, regardless of infections, is also associated with endothelial dysfunction and inflammatory upregulation, which may have a lasting effect on fetal neurodevelopment. Evidence also indicates that high levels of leptin that are observed in obesity are associated with placental dysfunction potentially disrupting normal neurodevelopment in the children.
In addition, maternal overweight and obesity are related to increased risks of pregnancy-associated diseases and neonatal complications that may be contributing to the increased risk of epilepsy in children. A Swedish population-based study has shown that maternal obesity is associated with increased risks of severe asphyxia-related complications in term infants. This increased risk may be attributed partly to shoulder dystocia and otherwise traumatic labor due to fetal macrosomia. However, our findings did not change substantially after adjusting for obesity-related pregnancy and neonatal complications, suggesting that maternal overweight or obesity per se may play a critical role in fetal neurodevelopment. Other underlying issues, including unmeasured genetic, lifestyle, and environmental factors that interact with obesity, could also contribute to this association.
We found strong evidence that asphyxia-related neonatal complications other than neonatal convulsions or low Apgar scores, including meconium aspiration and hypoxic ischemic encephalopathy, are associated with markedly increased rates of epilepsy. Previous studies have shown that children with a history of hypoxic ischemic encephalopathy had 5 times the risk of developing epilepsy. Furthermore, our study showed that neonatal jaundice, hypoglycemia, and respiratory distress may increase the risk of epilepsy. These findings persisted when analyses were restricted to children born at term.
Few previous studies have explored the potential role of more-prevalent neonatal complications on the risk of epilepsy in children. Moderate hypoglycemia for 3 or more days is associated with impaired neurologic development at 18 months. Respiratory distress has been linked to changes in cerebral blood flow, which in turn can increase the risks of periventricular or intraventricular hemorrhage in the developing brain. Even moderate degrees of hyperbilirubinemia have been associated with an increase in the risk of minor neurologic dysfunction throughout the first year of life.
Moreover, we found higher rates of childhood epilepsy among children whose mothers had epilepsy, gestational diabetes, and preeclampsia. Other factors associated with higher rates of childhood epilepsy were preterm birth, low birth weight for gestational age, birth trauma, birth asphyxia, and congenital malformations. Chromosomal abnormalities, malformation of the nervous system, and neonatal convulsions were all associated with substantially increased risks of childhood epilepsy. These findings are in line with those of previous studies demonstrating robust associations between congenital anomalies and neonatal convulsions and the development of epilepsy. Convulsions in newborns with perinatal asphyxia are independently associated with early life brain injury and later neurodevelopmental impairment.
The strengths of our study include the use of a large, comprehensive, nationwide population-based data source, which allowed us to study the rates of childhood epilepsy in relatively small groups, such as offspring of mothers with BMI of 40 or more and in children with specific neonatal complications. We used a previously validated case definition for epilepsy. In addition, prospective data collection minimized the risks of selection and recall bias.
Limitations
Even though we relied on both hospital inpatient and outpatient registries to capture maternal and neonatal complications, misclassification and underreporting of some diagnoses, such as maternal epilepsy, maternal infections, and chorioamnionitis, should be considered. Any misclassification would have been nondifferential, which generally results in an underestimation of rates. In addition, the cause of epilepsy may be multidimensional, with interaction between genetic and environmental factors. Environmental factors, including maternal smoking, alcohol use, socioeconomic factors, receipt of antiepileptic drugs, and deficiency of vitamins, could potentially increase the risk of epilepsy in offspring. Although we controlled for a broad range of potential confounders, unmeasured and residual confounding may have occurred due to factors not available in our data sources.
Conclusions
This population-based study demonstrates that the risk of epilepsy in offspring increased with maternal BMI category in a dose-response pattern. Furthermore, our study provides strong evidence that asphyxia-related neonatal complications, as well as less severe neonatal complications, including neonatal hypoglycemia, jaundice, and respiratory distress, independently increase the risk of childhood epilepsy. Given that overweight and obesity are potentially modifiable risk factors, prevention of obesity in women of reproductive age may be an important public health strategy to reduce the incidence of epilepsy.
eTable 1. ICD-10 Codes for Maternal, Neonatal, and Childhood Diseases
eTable 2. Incidence Rates and Hazard Ratios of Epilepsy in Children According to Maternal Characteristics
eTable 3. Incidence Rates and Hazard Ratios of Epilepsy According to Maternal and Neonatal Complications
eTable 4. Early Pregnancy Body-Mass Index and Rates of Epilepsy in Children Without Cerebral Palsy
References
- 1.Guerrini R. Epilepsy in children. Lancet. 2006;367(9509):499-524. [DOI] [PubMed] [Google Scholar]
- 2.Hirtz D, Thurman DJ, Gwinn-Hardy K, Mohamed M, Chaudhuri AR, Zalutsky R. How common are the “common” neurologic disorders? Neurology. 2007;68(5):326-337. [DOI] [PubMed] [Google Scholar]
- 3.World Health Organization (WHO) Epilepsy fact sheet. http://www.who.int/mediacentre/factsheets/fs999/en/. Published 2016. Updated February 2017. Accessed July 18, 2016.
- 4.Chang BS, Lowenstein DH. Epilepsy. N Engl J Med. 2003;349(13):1257-1266. [DOI] [PubMed] [Google Scholar]
- 5.Walsh S, Donnan J, Fortin Y, et al. A systematic review of the risks factors associated with the onset and natural progression of epilepsy. Neurotoxicology. 2016;S0161-813X(16)30030-4. [DOI] [PubMed] [Google Scholar]
- 6.Whitehead E, Dodds L, Joseph KS, et al. Relation of pregnancy and neonatal factors to subsequent development of childhood epilepsy: a population-based cohort study. Pediatrics. 2006;117(4):1298-1306. [DOI] [PubMed] [Google Scholar]
- 7.Seikku L, Gissler M, Andersson S, et al. Asphyxia, neurologic morbidity, and perinatal mortality in early-term and postterm birth. Pediatrics. 2016;137(6):e20153334. [DOI] [PubMed] [Google Scholar]
- 8.Ehrenstein V, Sørensen HT, Pedersen L, Larsen H, Holsteen V, Rothman KJ. Apgar score and hospitalization for epilepsy in childhood: a registry-based cohort study. BMC Public Health. 2006;6:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dalenius K, Brindley P, Smith B, Reinold C, Grummer-Strawn L. Pregnancy nutrition surveillance 2010 report. Atlanta, GA: US Dept. of Health and Human Services, Centers for Disease Control and Prevention; 2012. [Google Scholar]
- 10.Mendez MA, Monteiro CA, Popkin BM. Overweight exceeds underweight among women in most developing countries. Am J Clin Nutr. 2005;81(3):714-721. [DOI] [PubMed] [Google Scholar]
- 11.Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the United States, 2005 to 2014. JAMA. 2016;315(21):2284-2291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Schmatz M, Madan J, Marino T, Davis J. Maternal obesity: the interplay between inflammation, mother and fetus. J Perinatol. 2010;30(7):441-446. [DOI] [PubMed] [Google Scholar]
- 13.Johansson S, Villamor E, Altman M, Bonamy A-KE, Granath F, Cnattingius S. Maternal overweight and obesity in early pregnancy and risk of infant mortality: a population based cohort study in Sweden. BMJ. 2014;349:g6572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hemond J, Robbins RB, Young PC. The effects of maternal obesity on neonates, infants, children, adolescents, and adults. Clin Obstet Gynecol. 2016;59(1):216-227. [DOI] [PubMed] [Google Scholar]
- 15.Persson M, Johansson S, Villamor E, Cnattingius S. Maternal overweight and obesity and risks of severe birth-asphyxia–related complications in term infants: a population-based cohort study in Sweden. PLoS Med. 2014;11(5):e1001648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Stothard KJ, Tennant PW, Bell R, Rankin J. Maternal overweight and obesity and the risk of congenital anomalies: a systematic review and meta-analysis. JAMA. 2009;301(6):636-650. [DOI] [PubMed] [Google Scholar]
- 17.Mehta SH, Kerver JM, Sokol RJ, Keating DP, Paneth N. The association between maternal obesity and neurodevelopmental outcomes of offspring. J Pediatr. 2014;165(5):891-896. [DOI] [PubMed] [Google Scholar]
- 18.Forthun I, Wilcox AJ, Strandberg-Larsen K, et al. Maternal prepregnancy BMI and risk of cerebral palsy in offspring. Pediatrics. 2016;138(4):e20160874. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Pan C, Deroche CB, Mann JR, McDermott S, Hardin JW. Is prepregnancy obesity associated with risk of cerebral palsy and epilepsy in children? J Child Neurol. 2014;29(12):NP196-NP201. [DOI] [PubMed] [Google Scholar]
- 20.Ludvigsson JF, Almqvist C, Bonamy AK, et al. Registers of the Swedish total population and their use in medical research. Eur J Epidemiol. 2016;31(2):125-136. [DOI] [PubMed] [Google Scholar]
- 21.Swedish National Board of Health and Welfare The Swedish Medical Birth Register: a summary of content and quality. http://www.socialstyrelsen.se/Lists/Artikelkatalog/Attachments/8306/2009-125-15_200912515_rev2.pdf. Accessed February 15, 2016.
- 22.Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish National Inpatient Register. BMC Public Health. 2011;11:450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Swedish National Board of Health and Welfare Quality and content in the Swedish Patient Register http://www.socialstyrelsen.se/Lists/Artikelkatalog/Attachments/19490/2014-8-5.pdf. Accessed February 19, 2016.
- 24.Swedish National Board of Health and Welfare Cause of death. http://www.socialstyrelsen.se/statistics/statisticaldatabase/help/causeofdeath. Accessed February 19, 2016.
- 25.Statistics Sweden Evaluation of the Swedish Register of Education. http://www.scb.se/statistik/_publikationer/BE9999_2006A01_BR_BE96ST0604.pdf. Accessed February 19, 2016.
- 26.Villamor E, Bosch RJ. Optimal treatment of replicate measurements in anthropometric studies. Ann Hum Biol. 2015;42(5):507-510. [DOI] [PubMed] [Google Scholar]
- 27.World Health Organization BMI classification. http://apps.who.int/bmi/index.jsp?introPage=intro_3.html. Accessed February 19, 2016.
- 28.Helmers SL, Thurman DJ, Durgin TL, Pai AK, Faught E. Descriptive epidemiology of epilepsy in the US population: A different approach. Epilepsia. 2015;56(6):942-948. [DOI] [PubMed] [Google Scholar]
- 29.Thurman DJ, Beghi E, Begley CE, et al. ; ILAE Commission on Epidemiology . Standards for epidemiologic studies and surveillance of epilepsy. Epilepsia. 2011;52(suppl 7):2-26. [DOI] [PubMed] [Google Scholar]
- 30.VanderWeele T. Explanation in Causal Inference: Methods for Mediation and Interaction. Oxford, England: Oxford University Press; 2015. [Google Scholar]
- 31.Pugh SJ, Richardson GA, Hutcheon JA, et al. Maternal obesity and excessive gestational weight gain are associated with components of child cognition. J Nutr. 2015;145(11):2562-2569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Goeden N, Velasquez J, Arnold KA, et al. Maternal inflammation disrupts fetal neurodevelopment via increased placental output of serotonin to the fetal brain. J Neurosci. 2016;36(22):6041-6049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Rees S, Harding R, Walker D. The biological basis of injury and neuroprotection in the fetal and neonatal brain. Int J Dev Neurosci. 2011;29(6):551-563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Shatrov JG, Birch SC, Lam LT, Quinlivan JA, McIntyre S, Mendz GL. Chorioamnionitis and cerebral palsy: a meta-analysis. Obstet Gynecol. 2010;116(2, pt 1):387-392. [DOI] [PubMed] [Google Scholar]
- 35.Ramsay JE, Ferrell WR, Crawford L, Wallace AM, Greer IA, Sattar N. Maternal obesity is associated with dysregulation of metabolic, vascular, and inflammatory pathways. J Clin Endocrinol Metab. 2002;87(9):4231-4237. [DOI] [PubMed] [Google Scholar]
- 36.Hauguel-de Mouzon S, Lepercq J, Catalano P. The known and unknown of leptin in pregnancy. Am J Obstet Gynecol. 2006;194(6):1537-1545. [DOI] [PubMed] [Google Scholar]
- 37.Hogan L, Ingemarsson I, Thorngren-Jerneck K, Herbst A. How often is a low 5-min Apgar score in term newborns due to asphyxia? Eur J Obstet Gynecol Reprod Biol. 2007;130(2):169-175. [DOI] [PubMed] [Google Scholar]
- 38.Heslehurst N, Simpson H, Ells LJ, et al. The impact of maternal BMI status on pregnancy outcomes with immediate short-term obstetric resource implications: a meta-analysis. Obes Rev. 2008;9(6):635-683. [DOI] [PubMed] [Google Scholar]
- 39.Bergamasco B, Benna P, Ferrero P, Gavinelli R. Neonatal hypoxia and epileptic risk: a clinical prospective study. Epilepsia. 1984;25(2):131-136. [DOI] [PubMed] [Google Scholar]
- 40.Cansu A, Serdaroğlu A, Yüksel D, et al. Prevalence of some risk factors in children with epilepsy compared to their controls. Seizure. 2007;16(4):338-344. [DOI] [PubMed] [Google Scholar]
- 41.Lucas A, Morley R, Cole TJ. Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia. BMJ. 1988;297(6659):1304-1308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Perlman JM, McMenamin JB, Volpe JJ. Fluctuating cerebral blood-flow velocity in respiratory-distress syndrome: relation to the development of intraventricular hemorrhage. N Engl J Med. 1983;309(4):204-209. [DOI] [PubMed] [Google Scholar]
- 43.Soorani-Lunsing I, Woltil HA, Hadders-Algra M. Are moderate degrees of hyperbilirubinemia in healthy term neonates really safe for the brain? Pediatr Res. 2001;50(6):701-705. [DOI] [PubMed] [Google Scholar]
- 44.van Handel M, Swaab H, de Vries LS, Jongmans MJ. Long-term cognitive and behavioral consequences of neonatal encephalopathy following perinatal asphyxia: a review. Eur J Pediatr. 2007;166(7):645-654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Kjeldsen MJ, Kyvik KO, Christensen K, Friis ML. Genetic and environmental factors in epilepsy: a population-based study of 11900 Danish twin pairs. Epilepsy Res. 2001;44(2-3):167-178. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. ICD-10 Codes for Maternal, Neonatal, and Childhood Diseases
eTable 2. Incidence Rates and Hazard Ratios of Epilepsy in Children According to Maternal Characteristics
eTable 3. Incidence Rates and Hazard Ratios of Epilepsy According to Maternal and Neonatal Complications
eTable 4. Early Pregnancy Body-Mass Index and Rates of Epilepsy in Children Without Cerebral Palsy