Abstract
OBJECTIVE:
To evaluate severe maternal morbidity (SMM) among patients with and without epilepsy.
METHODS:
We retrospectively examined SMM using linked birth certificate and maternal hospital discharge records in California between 2007 and 2012. Epilepsy present on delivery admission was the exposure, and was subtyped into generalized, focal and other less specified, or unspecified. The outcomes were SMM and non-transfusion SMM from delivery up to 42 days postpartum, identified using Centers for Disease Control and Prevention indicators. Multivariable logistic regression models were used to adjust for confounders which were selected a priori. We also estimated the association between epilepsy and SMM independent of comorbidities, using a validated obstetric comorbidity score. SMM indicators were then compared using the same multivariable logistic regression models.
RESULTS:
Of 2,668,442 births, 8,145 (0.3%) were to patients with epilepsy; 637 (7.8%) had generalized, 6,250 (76.7%) had focal or other less specified, and 1,258 (15.4%) had unspecified subtypes. Compared to patients without epilepsy, patients with epilepsy had greater odds of SMM (4.3% versus 1.4%, adjusted odds ratio [aOR] 2.91, 95% confidence interval [CI] 2.61–3.24) and non-transfusion SMM (2.9% versus 0.7%, aOR 4.16, 95% CI 3.65–4.75). Epilepsy remained significantly associated with increased SMM and non-transfusion SMM after additional adjustment for the obstetric comorbidity score, though the effects were attenuated. When grouped by organ system, all SMM indicators were significantly more common among patients with epilepsy – most notably those related to hemorrhage and transfusion.
CONCLUSION:
Severe maternal morbidity was significantly increased in patients with epilepsy, and SMM indicators across all organ systems contributed to this.
INTRODUCTION
The number of people with epilepsy in the U.S. is rising, increasing from 2.3 million in 2010 to 3 million in 2015.1,2 Approximately 24,000 patients with epilepsy deliver annually, accounting for 0.3–0.5% of all births.3,4 In these patients, the risk of maternal mortality may be as much as 11 times higher than the general population.4–7 79% of maternal deaths in a recent cohort of patients with epilepsy were considered to be sudden and unexpected, suggesting that an improved understanding of maternal mortality with epilepsy is warranted.8
Although patients with epilepsy have increased maternal mortality, evidence has been conflicting regarding their risk of pregnancy complications.3,4,9 This may be due to variability in seizure type and control, or to the rarity of specific outcomes, such as eclampsia, which have been studied. In an attempt to understand rare complications that might contribute to maternal mortality, the severe maternal morbidity composite was developed by the Centers for Disease Control and Prevention. This composite includes 21 severe maternal morbidity indicator events, such as eclampsia and cardiac arrest.10–12 Like maternal mortality, severe maternal morbidity has also been shown to be increased among patients with epilepsy.4 However, it remains unclear which severe maternal morbidity indicator events are occurring and how severe maternal morbidity varies by type of epilepsy. Thus we sought to examine severe maternal morbidity among patients with epilepsy.
METHODS
This was a cohort study of pregnancies in California between 2007 and 2012 to assess health risks associated with maternal epilepsy in pregnancy. Data were analyzed retrospectively. Births were identified using California birth cohort files, which link data from birth certificates to maternal and infant discharge records through the Office of Statewide Health Planning and Development. The California birth and fetal death certificate provides information on maternal demographics, medical diagnoses, delivery complications, and infant outcomes which are filled out by both patients and medical staff. Inpatient discharge records were obtained from all licensed acute care facilities in California, and include International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. The linkage in these birth cohort files is complete through 2012, and as such 2012 was selected as the study endpoint.
All pregnant patients who gave birth at ≥20 weeks to a liveborn or stillborn infant were included in this study. Patients were excluded if birth certificate and maternal delivery hospitalization discharge records were not linked or if the gestational age was implausible (<20 weeks or >43 weeks). The Stanford University Institutional Review Board and the California State Committee for the Protection of Human Subjects reviewed and approved this study.
The exposure for this study was a maternal diagnosis of epilepsy. Only patients with an epilepsy diagnosis during an antepartum hospitalization or marked present on admission for delivery hospitalization were defined as having epilepsy. Epilepsy was further categorized into generalized epilepsy, focal or other less specified epilepsy, and unspecified subtypes based on ICD-9 CM codes. Generalized epilepsy was identified using ICD-9 CM codes 345.0 and 345.1. Focal or other less specified epilepsy included a composite of focal (345.4, 345.5), localization-related (345.7), and other types (345.2 “petit mal status”, 345.3 “grand mal status”, 345.8 “other forms of epilepsy and recurrent seizures”, 345.9 “epilepsy unspecified”). These types were grouped due to anticipated small numbers limiting our ability to report rare outcomes between them. Two distinct unspecified epilepsy groups were created based on frequently encountered ICD-9 CM codes; code 649.4 was used for unspecified “Epilepsy complicating pregnancy, childbirth, or the puerperium” and code 780.39 was used for “Convulsions”. If patients had 780.39 and a code for eclampsia, they were excluded from the analysis as it was unclear whether this code represented the presence of epilepsy or eclampsia on admission and could result in misclassification of the exposure. If patients had codes for multiple subtypes of seizures, only the most severe specified subtype was counted. This was done in a mutually exclusive fashion in the following hierarchy: generalized, focal or other less specified, epilepsy complicating pregnancy, childbirth, or the puerperium, then lastly convulsions. For example, a patient with ICD-9 CM codes for both generalized epilepsy (e.g. 345.0) and convulsions (780.39) was categorized under the generalized subtype only.
The primary outcome was the severe maternal morbidity composite during the delivery admission or during a subsequent hospital admission up to 42 days postpartum. The severe maternal morbidity composite was defined using the Centers for Disease Control and Prevention indicators and their corresponding ICD-9-CM codes.12 To further understand this outcome, we reported “non-transfusion severe maternal morbidity” which is a composite of the same severe maternal morbidity indicators excluding blood transfusion. Blood transfusion is the only indicator for severe maternal morbidity in up to half of cases, and the amount of blood transfused is not recorded in administrative data.10,13 As such, non-transfusion severe maternal morbidity was included as a co-primary outcome.
Next, all 21 severe maternal morbidity indicators were individually evaluated and compared between people with and without epilepsy. Due to small sample size for some specific indicator events, the following organ system groups were made: cardiac (acute myocardial infarction, aneurysm, cardiac arrest or ventricular fibrillation, conversion of cardiac rhythm, heart failure or arrest during surgery, or acute heart failure); pulmonary (adult respiratory distress syndrome, pulmonary edema, temporary tracheostomy, or mechanical ventilation); renal (acute renal failure); hemorrhage (disseminated intravascular coagulation, shock, or hysterectomy); sepsis; obstetric (amniotic fluid embolism, eclampsia, severe anesthesia complications, or air and thrombotic embolism); other medical (puerperal cerebrovascular disorders or sickle cell disease with crisis); and transfusion (blood transfusion). Transfusion severe maternal morbidity was considered as a separate group from hemorrhage because of the aforementioned ambiguity in administrative data of number of units transfused and the emerging importance of non-transfusion severe maternal morbidity. Of note, the cerebrovascular disorder severe maternal morbidity indicator includes events such as subarachnoid hemorrhage, stroke, and venous sinus thrombosis.
Secondary outcomes included obstetric complications such as preeclampsia (with and without severe features), gestational diabetes, stillbirth, preterm birth, induction of labor, and cesarean birth.
Potential confounders were selected a priori based on prior literature on severe maternal morbidity and epilepsy and causal diagrams.4,13,14 These included maternal age, race or ethnicity as a social determinant, method of payment, education level, trimester of prenatal care initiation, and parity (which were included as covariates in multivariable logistic regression Model 2, see below). Race or ethnicity were obtained from the birth certificate, where it is self-reported by the patient. This approach has been previously validated using California birth certificate data.15 Race or ethnicity were included as social constructs given prior literature showing associations with both epilepsy and severe maternal morbidity.4,13 We additionally identified comorbidities, such as chronic cardiovascular disease, as potential confounders or mediators of the association between epilepsy and severe maternal morbidity. Given this, we planned a separate multivariable logistic regression model to also account for the role comorbidities might be playing in severe maternal morbidity for patients with epilepsy. To do so, comorbidities were added as a covariate to the aforementioned Model 2 to create Model 3. Comorbidities were defined using a previously developed expanded obstetric comorbidity scoring system to create a comorbidity composite.13 This obstetric comorbidity scoring system includes 27 patient-level risk factors for severe maternal morbidity such as pre-existing diabetes, chronic hypertension, pulmonary hypertension, bleeding disorders, major mental health disorders, and autoimmune disease. Since the purpose of this comorbidity composite was to identify patient-level mediators or confounders preceding delivery or the outcome, preterm birth was excluded from the composite which is in line with other validated comorbidity scores that have been published.16,17 In addition, neuromuscular disorders were excluded from the composite given overlap with epilepsy.
Demographic and obstetric characteristics, such as mode of delivery, preterm birth, preeclampsia, and gestational diabetes, were compared with chi-square tests for categorical variables and t-tests for continuous variables. Variables that were not normally distributed were categorized based on clinically relevant cutoffs (e.g. body-mass index and maternal age), or frequency (e.g. obstetric comorbidity score). Next, crude logistic regression models (Model 1) were used to estimate odds ratios with 95% confidence intervals (CI) for the associations between epilepsy and severe maternal morbidity and non-transfusion severe maternal morbidity using births to patients without epilepsy as the referent group. Due to the rarity of severe maternal morbidity and non-transfusion severe maternal morbidity, odds ratios approximated risk ratios.
A series of additional analyses were then conducted to minimize potential bias. First, multivariable logistic regression models (Model 2) were adjusted for the confounding variables listed above (maternal age, race or ethnicity as a social determinant, method of payment, education level, trimester of prenatal care initiation, and parity). Secondly, the obstetric comorbidity score described above was added as a covariate to Model 2 to generate Model 3. These models were run separately in the event that comorbidities served as mediators rather than confounders of the association between epilepsy and severe maternal morbidity. Next, multivariable logistic regression models were used to compare the odds of each of the 21 CDC severe maternal morbidity indicators between pregnancies with and without maternal epilepsy, adjusted for the same potential confounders listed above in Model 2. Lastly, as an assessment of the robustness of our results and analytical decisions, propensity-score matching was done to analyze the association between epilepsy and severe maternal morbidity and non-transfusion severe maternal morbidity. We used greedy nearest neighbor matching with “method nearest” in the MatchIt package in R, which matches patients with and without epilepsy based on the closest propensity score in order to optimally balance covariates between groups. In this analysis, a distance is computed between each treated unit and each control unit, and, one by one, each treated unit is assigned a control unit as a match. The matching is “greedy” in the sense that there is no action taken to optimize an overall criterion; each match is selected without considering the other matches that may occur subsequently.18 We re-ran Models 2 and 3 using propensity score matching for both severe maternal morbidity and non-transfusion severe maternal morbidity, using the same covariates noted above. Significance was set to a two-tailed alpha=0.05. All statistical analysis was performed using SAS 9.4 (SAS Institute Inc., Cary, NC) and R 3.6.1 (Vienna, Austria).
RESULTS
Among 2,668,442 eligible births, 8,145 (0.3%) were to patients with epilepsy (Figure 1). Of these, 637 (7.8%) had generalized epilepsy, 6,250 (76.7%) had focal or other less specified epilepsy, 284 (3.5%) had epilepsy complicating pregnancy, childbirth, or the puerperium, and 974 (12.0%) had convulsions.
Figure 1.
Study cohort inclusion flowsheet. *Unable to distinguish exposure.
Patients with epilepsy were more likely to be younger, Hispanic, have high BMI at delivery, have not completed college, and have commercial insurance than patients without epilepsy. Pre-pregnancy diabetes, chronic hypertension, and major depressive disorder were also more prevalent among patients with epilepsy, as was a higher obstetric comorbidity index score (Table 1).
Table 1.
Demographic characteristics of patients with epilepsy compared to patients without epilepsy in California, 2007–2012 (N=2,668,442).
| Maternal Characteristic | Births to patients without epilepsy (N=2,660,297) |
Births to patients with epilepsy (N=8,145) |
p-value* |
|---|---|---|---|
|
| |||
| Age at delivery (years) | <0.001 | ||
| <20 | 161,953 (6.1) | 542 (6.7) | |
| 20–24 | 584,716 (22.0) | 1,987 (24.4) | |
| 25–29 | 733,199 (27.6) | 2,374 (29.2) | |
| 30–34 | 691,045 (26.0) | 1,956 (24.0) | |
| 35–39 | 388,960 (14.6) | 1,023 (12.6) | |
| ≥40 | 100,424 (3.8) | 263 (3.2) | |
| Race or ethnicity | <0.001 | ||
| Hispanic | 759,019 (28.5) | 3,030 (37.2) | |
| Non-Hispanic White | 1,370,901 (51.5) | 3,554 (43.6) | |
| Asian or Pacific Islander | 150,401 (5.7) | 944 (11.6) | |
| Black | 366,187 (13.8) | 501 (6.2) | |
| Other† | 13,789 (0.5) | 116 (1.4) | |
| BMI at delivery (kg/m2) | <0.001 | ||
| BMI <24.9 | 307,176 (11.5) | 918 (11.3) | |
| BMI 25–29.9 | 989,878 (37.2) | 2,799 (34.4) | |
| BMI 30–34.9 | 784,219 (29.5) | 2,357 (28.9) | |
| BMI 35–39.9 | 365,089 (13.7) | 1,233 (15.1) | |
| BMI ≥40 | 213,935 (8.0) | 838 (10.3) | |
| Education level | <0.001 | ||
| High school or less | 601,631 (22.6) | 1,946 (23.9) | |
| Completed high school | 711,152 (26.7) | 2,541 (31.2) | |
| Some college | 655,946 (24.7) | 2,298 (28.2) | |
| Completed college or higher | 691,568 (26.0) | 1,360 (16.7) | |
| Delivery Payment Method | <0.001 | ||
| Medi-Cal | 1,283,529 (48.3) | 3,317 (40.7) | |
| Commercial insurance | 1,291,654 (48.6) | 4,664 (57.3) | |
| Other | 85,114 (3.2) | 164 (2.0) | |
| Trimester of prenatal care initiation | <0.001 | ||
| First | 2,187,846 (82.2) | 6,484 (79.6) | |
| Second | 347,739 (13.1) | 1,146 (14.1) | |
| Third or None | 77,781 (2.9) | 306 (3.8) | |
| Unknown | 46,931 (1.8) | 209 (2.6) | |
| Pre-pregnancy diabetes (type 1 or 2) | 23,771 (0.9) | 175 (2.2) | <0.001 |
| Chronic hypertension | 39,355 (1.5) | 262 (3.2) | <0.001 |
| Major depressive disorder | 3,239 (0.1) | 43 (0.5) | <0.001 |
| Obstetric comorbidity score‡ | |||
| 0 | 1,346,263 (50.6) | 3,245 (39.8) | <0.001 |
| 1–4 | 562,598 (21.2) | 1,312 (16.1) | |
| 5–9 | 261,895 (9.8) | 849 (10.4) | |
| ≥10 | 489,541 (18.4) | 2,739 (33.6) | |
Shown as number (column percent), and analyzed with chi square test.
Race or ethnicity groups defined based on state reporting. Other defined as American Indian, Alaska native, or other.
Defined using a validated comorbidity scoring system which has been shown to aid in prediction of severe maternal morbidity (SMM).13 This scoring system includes 27 patient-level risk factors for SMM such as pulmonary hypertension, preexisting cardiac disease or bleeding disorders, and autoimmune disease. Due to overlap with our exposure, we excluded neuromuscular diseases from this comorbidity score. In addition, only comorbidities present prior to delivery were included; therefore, preterm birth was also excluded from this comorbidity score. See text for details.
Compared to patients without epilepsy, patients with maternal epilepsy were more likely to be nulliparous and develop preeclampsia without severe features or gestational hypertension and severe preeclampsia and to deliver preterm. The rate of cesarean birth with maternal epilepsy was significantly higher than without epilepsy (Table 2).
Table 2.
Pregnancy characteristics and outcomes among patients with epilepsy compared to those without epilepsy in California, 2007–2012 (N=2,668,442).
| Characteristic | Births to patients without epilepsy (N=2,660,297) |
Births to patients with epilepsy (N=8,145) |
p-value* |
|---|---|---|---|
|
| |||
| Nulliparous | 996,833 (37.5) | 3,168 (38.9) | 0.01 |
| Multiple gestation | 42,049 (1.6) | 136 (1.7) | 0.52 |
| Preeclampsia without severe features or gestational hypertension | 117,674 (4.4) | 487 (6.0) | <0.001 |
| Severe preeclampsia | 286,662 (1.1) | 177 (2.2) | <0.001 |
| Gestational diabetes | 203,336 (7.6) | 617 (7.6) | 0.82 |
| Stillbirth | 9,385 (0.4) | 35 (0.4) | 0.24 |
| Preterm birth | 210,877 (7.9) | 1,102 (13.5) | <0.001 |
| <34w0d | 154,920 (5.8) | 759 (9.3) | |
| 34w0d-36w6d | 55,957 (2.1) | 343 (4.2) | |
| Average gestational age at delivery (weeks) | 38.6 (2.0) | 38.2 (2.4) | <0.001 |
| Induction of labor | 927,054 (34.9) | 2,918 (35.8) | 0.06 |
| Cesarean birth | 889,026 (33.4) | 3,415 (41.9) | <0.001 |
| Repeat cesarean birth | 404,963 (15.2) | 1,474 (18.1) | <0.001 |
| Severe maternal morbidity | 37,960 (1.4) | 350 (4.3) | <0.001 |
| Non-transfusion severe maternal morbidity | 17,309 (0.7) | 234 (2.9) | <0.001 |
Number (column percent) shown for categorical variables and analyzed with chi square test, mean (standard deviation) shown for continuous variables and analyzed with t-test.
Severe maternal morbidity occurred in 1.4% of all births, while non-transfusion severe maternal morbidity occurred in 0.7% of all births. The risk of severe maternal morbidity was significantly increased in births with maternal epilepsy compared to births without epilepsy (4.3% versus 1.4%, crude odds ratio [OR] 3.10 [95% confidence interval (CI) 2.79–3.45], adjusted OR [aOR] 2.91 [95% CI 2.61–3.24], Table 3) as was non-transfusion severe maternal morbidity (2.9% versus 0.7%, crude OR 4.52 [95% CI 3.97–5.15], aOR 4.16 [95% CI 3.65–4.75], Table 4). Generalized epilepsy was associated with the highest risk of both severe maternal morbidity and non-transfusion severe maternal morbidity. Focal or other less specified epilepsy subtypes were also associated with significantly increased risk of both severe maternal morbidity and non-transfusion severe maternal morbidity. When examining individual codes for unspecified epilepsy, severe maternal morbidity remained similarly increased for unspecified epilepsy complicating pregnancy, childbirth, or the puerperium as well as for convulsions (Tables 3 and 4).
Table 3.
Adjusted risk of severe maternal morbidity among patients with epilepsy compared to patients without epilepsy in California, 2007–2012 (N=2,668,442).
| Exposure group | N (row %) | Model 1 Crude OR (95% CI) |
Model 2* Adjusted OR (95% CI) |
Model 3† Adjusted OR (95% CI) |
|---|---|---|---|---|
|
| ||||
| Patients without epilepsy (N=2,660,297) | 37,960 (1.4) | Reference | Reference | Reference |
| All patients with epilepsy‡ (N=8,145) | 350 (4.3) | 3.10 (2.79–3.45) | 2.91 (2.61–3.24) | 2.13 (1.90–2.39) |
| Generalized epilepsy (N=637) | 49 (7.7) | 5.76 (4.30–7.71) | 5.32 (3.97–7.14) | 3.81 (2.79–5.21) |
| Focal epilepsy and other less specified epilepsies (N=6,250) | 241 (3.9) | 2.77 (2.44–3.15) | 2.61 (2.29–2.97) | 2.01 (1.75–2.30) |
| Unspecified epilepsy complicating pregnancy, childbirth, or the puerperium (N=284) | 14 (4.9) | 3.58 (2.09–6.13) | 3.23 (1.88–5.54) | 2.31 (1.30–4.11) |
| Convulsions only (N=974) | 46 (4.7) | 3.43 (2.55–4.61) | 3.20 (2.38–4.31) | 1.80 (1.30–2.49) |
Model 2 adjusted for maternal age, race or ethnicity as a social determinant, method of payment, education level, trimester of prenatal care initiation, and parity.
Model 3 additionally adjusted for validated obstetric comorbidity score. See text for details.
Refer to text for details regarding categorization of epilepsy subtypes.
Table 4.
Adjusted risk of non-transfusion severe maternal morbidity among patients with epilepsy compared to patients without epilepsy in California, 2007–2012 (N=2,668,442).
| Exposure group | N (row %) | Model 1 Crude OR (95% CI) |
Model 2* Adjusted OR (95% CI) |
Model 3† Adjusted OR (95% CI) |
|---|---|---|---|---|
|
| ||||
| Patients without epilepsy (N=2,660,297) | 17,309 (0.7) | Reference | Reference | Reference |
| All patients with epilepsy‡ (N=8,145) | 234 (2.9) | 4.52 (3.97–5.15) | 4.16 (3.65–4.75) | 2.99 (2.61–3.43) |
| Generalized epilepsy (N=637) | 38 (6.0) | 9.69 (6.98–13.45) | 8.83 (6.34–12.89) | 6.29 (4.44–8.92) |
| Focal epilepsy and other less specified epilepsies (N=6,250) | 162 (2.6) | 4.07 (3.48–4.76) | 3.78 (3.21–4.40) | 2.85 (2.42–3.36) |
| Unspecified epilepsy complicating pregnancy, childbirth, or the puerperium (N=284) | <15§ | 3.30 (1.47–7.40) | 2.87 (1.28–6.47) | 1.87 (0.80–4.39) |
| Convulsions only (N=974) | 28 (2.9) | 4.52 (3.10–6.58) | 4.17 (2.86–6.08) | 2.21 (1.47–3.33) |
Model 2 adjusted for maternal age, race or ethnicity as a social determinant, method of payment, education level, trimester of prenatal care initiation, and parity.
Model 3 additionally adjusted for validated obstetric comorbidity score. See text for details.
Refer to text for details regarding categorization of epilepsy subtypes.
Cell sizes <15 not shown per data use agreement.
After accounting for comorbidities in Model 3, the associations between any type of epilepsy and both severe maternal morbidity and non-transfusion severe maternal morbidity were attenuated, but remained significant (Tables 3 and 4).
Appendix 1, available online at http://links.lww.com/xxx, demonstrates the risk of each severe maternal morbidity indicator compared between births with and without maternal epilepsy. When grouped by organ system, all severe maternal morbidity indicators were significantly more common among patients with epilepsy (Figure 2). For simplification in Figure 2, cardiac and pulmonary categories were combined but are shown separately in Appendix 1. Compared to patients without epilepsy, the most common events among patients with epilepsy were transfusion, eclampsia, disseminated intravascular coagulation, and puerperal cerebrovascular disorders.
Figure 2.
Severe maternal morbidity (SMM) indicator event rates compared between births to patients with and without epilepsy in California, 2007‒2012.
Lastly, the estimated associations between epilepsy and both severe maternal morbidity and non-transfusion severe maternal morbidity were overall unchanged in sensitivity analyses that used propensity score matching to account for confounders (see Appendixes 2–4, available online at http://links.lww.com/xxx).
DISCUSSION
Patients with epilepsy had a 3-fold increased risk of severe maternal morbidity and a 4-fold increased risk of non-transfusion severe maternal morbidity compared to patients without epilepsy, and these risks persisted despite adjustment for maternal comorbidities. Although generalized epilepsy was rare, it was associated with the highest risk. To put our results into context, the risk of SMM with epilepsy was higher than what has been demonstrated with other conditions such as autoimmune disease (aRR 1.80, 95% CI 1.73–1.87).13 Overall, our findings emphasize the contribution of hemorrhage to severe maternal morbidity for patients with epilepsy and highlight who might be at highest risk of complications.
Our results are important in light of conflicting evidence regarding the risk of pregnancy complications for patients with epilepsy.3–6, 9,19,20 While it is reassuring that the majority of these patients (95.7%) did not experience severe maternal morbidity, the increased risk we identified is similar to what has been previously reported.4 The persistent risk even after accounting for comorbidities further underscores the importance of physician awareness when managing patients with epilepsy.
Seizure control is a critical component of prenatal management and is important when interpreting the risk of severe maternal morbidity for patients with epilepsy.9 Seizures themselves can cause hypoxia, aspiration, or trauma from falls, which can contribute to severe maternal morbidity events such as pulmonary edema.3 While these types of events were more common for patients with epilepsy, other events such as those related to hemorrhage were also increased. Furthermore, though eclampsia contributed to severe maternal morbidity, it did not occur in 84.6% of patients with epilepsy who experienced severe maternal morbidity. Taken together, it seems that a seizure event alone – either epileptic or eclamptic – does not completely explain the increased risk of severe maternal morbidity with epilepsy.
To this end, research has also explored the role of antiseizure medications (ASMs) in augmenting the risk of adverse outcomes with epilepsy.1,21–28 However, understanding the association between ASMs and obstetric outcomes is challenging. While greater ASM use might reflect improved seizure control and better outcomes, it can also be a proxy for disease severity and bias the signal in either direction.14,19 As we were unable to comment on use of ASMs in this analysis, future research interrogating whether ASMs affect the risk of severe maternal morbidity is warranted.
Strengths of our study include the large population, which enabled us to examine rare but clinically significant outcomes, and the pragmatic approach to associating epilepsy subtype with a standardized measure of maternal morbidity, which is applicable to clinical practice. These results can help risk-stratify patients and provide guidance for future research on risk mitigation strategies. Additionally, our analysis demonstrates that patients with epilepsy have higher rates of comorbidities which are also risk factors for severe maternal morbidity. Furthermore, the robustness of our results was confirmed using propensity score matching.
Our results must be interpreted within the context of the study design and limitations. Though timing of events was not available retrospectively, we attempted to address causality between epilepsy and severe maternal morbidity by restricting only to epilepsy codes present on admission. There likely remain unmeasured confounders which were not accounted for or were not included due to concerns about coding reliability (e.g. smoking status, cesarean indication, or epilepsy disease control). Due to data use agreements, we are unable to report rates of very rare outcomes such as maternal mortality. Though our results may not be generalizable to populations outside of California, California is diverse and accounts for the greatest total number of births in the United States.29
In conclusion, the risk of severe maternal morbidity and non-transfusion severe maternal morbidity were significantly increased among patients with epilepsy, though reassuringly the absolute risks remain low.
Supplementary Material
PRECIS:
Patients with epilepsy had a threefold increased risk of severe maternal morbidity
Acknowledgements:
The authors thank Dr. Anna Girsen M.D., PhD. for her assistance with development of the study idea.
Funding Source: Funding from NIH NR017020 supported this work.
Financial Disclosure
Kimford J. Meador disclosed that money was paid to his institution by Eisai Inc. and Sunovion Pharmaceuticals. The Epilepsy Study Consortium pays Kimford J. Meador’s university for his research consultant time related to Eisai, GW Pharmaceuticals, NeuroPace, Novartis, Supernus, Upsher-Smith Laboratories, UCB Pharma, and Vivus Pharmaceuticals.
Thomas F. McElrath disclosed receiving compensation from Hoffman – LaRoche, NxPreatal, Comanche Biopharma, and Momenta Pharmaceuticals, Inc. for serving on their scientific advisory boards in work that is unrelated to the present analysis.
Deirdre J. Lyell disclosed receiving funds from Bloomlife (stock options for consulting, unrelated to this work). She also received payment from SMFM and UCSF for lectures. The other authors did not report any potential conflicts of interest.
Each author has confirmed compliance with the journal’s requirements for authorship.
REFERENCES
- 1.French JA, Kanner AM, Bautista J, Abou-Khalil A, Browne T, Harden CL, et al. Practice Parameters, Efficacy, and Tolerability of the New Antiepileptic Drugs: Treatment of New-Onset Epilepsy. Report of the TTA and QSS Subcommittees of the American Academy of Neurology and the American Epilepsy Society. Epilepsia. 2004;45(5):401–409. [DOI] [PubMed] [Google Scholar]
- 2.Centers for Disease Control and Prevention. More Americans have epilepsy than ever before. Published 2017. Accessed February 25, 2019. https://www.cdc.gov/media/releases/2017/p0810-epilepsy-prevalence.html
- 3.Hessler A, Dolbec K. Clinical updates in women’s healthcare: Seizures. Am Coll Obstet Gynecol. 2021;XX(I). [DOI] [PubMed] [Google Scholar]
- 4.MacDonald SC, Bateman BT, McElrath TF, Hernández-Díaz S. Mortality and morbidity during delivery hospitalization among pregnant women with epilepsy in the United States. JAMA Neurol. 2015;72(9):981–988. doi: 10.1001/jamaneurol.2015.1017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Leach J, Smith P, Craig J, Bagary M, Cavanagh D, Duncan S, et al. Epilepsy and Pregnancy: For healthy pregnancies and happy outcomes. Suggestions for service improvements from the Multispecialty UK Epilepsy Mortality Group. Seizure. 2017;50(2017):67–72. doi: 10.1016/j.seizure.2017.05.004 [DOI] [PubMed] [Google Scholar]
- 6.Viale L, Allotey J, Cheong-See F, Arroyo-Manzano D, Mccorry D, Bagary M, et al. Epilepsy in pregnancy and reproductive outcomes: A systematic review and meta-analysis. Lancet. 2015;386(10006):1845–1852. doi: 10.1016/S0140-6736(15)00045-8 [DOI] [PubMed] [Google Scholar]
- 7.Stephen LJ, Harden C, Tomson T, Brodie MJ. Management of epilepsy in women. Lancet Neurol. Published online 2019:1–11. doi: 10.1016/S1474-4422(18)30495-2 [DOI] [PubMed] [Google Scholar]
- 8.Edey S, Moran N, Nashef L. SUDEP and epilepsy-related mortality in pregnancy. Epilepsia. 2014;55(7):72–74. doi: 10.1111/epi.12621 [DOI] [PubMed] [Google Scholar]
- 9.Harden CL, Hopp J, Ting TY, Pennell PB, French JA, Hauser WA, et al. Practice parameter update: Management issues for women with epilepsy - Focus on pregnancy (an evidence-based review): Obstetrical complications and change in seizure frequency: Report of the Quality Standards Subcommittee and Therapeutics and Technology A. Neurology. 2009;73(2):126–132. doi: 10.1212/WNL.0b013e3181a6b2f8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Main EK, Abreo A, McNulty J, Gilbert W, McNally C, Poeltler D, et al. Measuring severe maternal morbidity: Validation of potential measures. Am J Obstet Gynecol. 2016;214(5):643.e1–643.e10. 10.1016/j.ajog.2015.11.004 [DOI] [PubMed] [Google Scholar]
- 11.Main EK. Reducing Maternal Mortality and Severe Maternal Morbidity Through State-based Quality Improvement Initiatives. Clin Obstet Gynecol. 2018;61(2):319–331. doi: 10.1097/GRF.0000000000000361 [DOI] [PubMed] [Google Scholar]
- 12.Centers for Disease Control and Prevention. Severe Maternal Morbidity Indicators and Corresponding ICD Codes during Delivery Hospitalizations. Accessed March 11, 2019. https://www.cdc.gov/reproductivehealth/maternalinfanthealth/smm/severe-morbidity-ICD.htm
- 13.Leonard SA, Kennedy CJ, Carmichael SL, Lyell DJ, Main EK. An Expanded Obstetric Comorbidity Scoring System for Predicting Severe Maternal Morbidity. Obstet Gynecol. 2020;136(3):440–449. doi: 10.1097/AOG.0000000000004022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Razaz N, Tomson T, Wikström AK, Cnattingius S. Association between pregnancy and perinatal outcomes among Women with epilepsy. JAMA Neurol. 2017;74(8):983–991. doi: 10.1001/jamaneurol.2017.1310 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Baumeister L, Marchi K, Pearl M, Williams R, Braveman P. The validity of information on “race” and “Hispanic ethnicity” in California birth certificate data. Health Serv Res. 2000;35(4):869–883. http://www.ncbi.nlm.nih.gov/pubmed/11055453%0Ahttp://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC1089157 [PMC free article] [PubMed] [Google Scholar]
- 16.Bateman BT, Mhyre JM, Hernandez-Diaz S, Huybrechts KF, Fischer MA, Creanga AA, et al. Development of a comorbidity index for use in obstetric patients. Obstet Gynecol. 2013;122(5):957–965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Easter SR, Bateman BT, Sweeney VH, Manganaro K, Lassey SC, Gagne JJ, et al. A comorbidity-based screening tool to predict severe maternal morbidity at the time of delivery. Am J Obstet Gynecol. 2019;221(3):271.e1–271.e10. 10.1016/j.ajog.2019.06.025 [DOI] [PubMed] [Google Scholar]
- 18.Ho DE, King G, Stuart EA, Imai K. MatchIt : Nonparametric Preprocessing for. J Stat Softw. 2011;42(8):1–28. http://www.jstatsoft.org/v42/i08/paper [Google Scholar]
- 19.Borthen I, Eide MG, Daltveit AK, Gilhus NE. Obstetric outcome in women with epilepsy: A hospital-based, retrospective study. BJOG An Int J Obstet Gynaecol. 2011;118(8):956–965. doi: 10.1111/j.1471-0528.2011.03004.x [DOI] [PubMed] [Google Scholar]
- 20.Melikova S, Bagirova H, Magalov S. The impact of maternal epilepsy on delivery and neonatal outcomes. Child’s Nerv Syst. 2020;36(4):775–782. doi: 10.1007/s00381-019-04435-2 [DOI] [PubMed] [Google Scholar]
- 21.Marson AG, Appleton R, Baker GA, Chadwick DW, Doughty J, Eaton B, et al. A randomised controlled trial examining the longer-term outcomes of standard versus new antiepileptic drugs. The SANAD trial. Health Technol Assess. 2007;11(37):iii–iv, ix-x, 1–134. [DOI] [PubMed] [Google Scholar]
- 22.French JA, Gazzola DM. New generation antiepileptic drugs: What do they offer in terms of improved tolerability and safety? Ther Adv Drug Saf. 2011;2(4):141–158. doi: 10.1177/2042098611411127 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kim H, Faught E, Thurman DJ, Fishman J, Kalilani L. Antiepileptic Drug Treatment Patterns in Women of Childbearing Age With Epilepsy. JAMA Neurol. 2019;94304:1–8. doi: 10.1001/jamaneurol.2019.0447 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Meador KJ, Pennell PB, May RC, Gerard E, Kalayjian L, Velez-Ruiz N, et al. Changes in antiepileptic drug-prescribing patterns in pregnant women with epilepsy. Epilepsy Behav. 2018;84:10–14. doi: 10.1016/j.yebeh.2018.04.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Veroniki AA, Cogo E, Rios P, Al. E. Comparative safety of anti-epileptic drugs during pregnancy: a systematic review and network meta-analysis of congenital malformations and prenatal outcomes. BMC Med. 2017;15(1):1–20. doi: 10.1186/s12916-017-0845-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Meador KJ, Loring DW. Developmental effects of antiepileptic drugs and the need for improved regulations. Neurology. 2016;86(3):297–306. doi: 10.1212/WNL.0000000000002119 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Vajda FJE, Hollingworth S, Graham J, Hitchcock AA, O’Brien TJ, Lander CM, et al. Changing patterns of antiepileptic drug use in pregnant Australian women. Acta Neurol Scand. 2010;121(2):89–93. doi: 10.1111/j.1600-0404.2009.01260.x [DOI] [PubMed] [Google Scholar]
- 28.Vajda F, O’Brien T, Graham J, Hitchcock A, Lander C, Eadie M. Anti-epileptic drug exposure and risk of foetal death in utero. Acta Neurol Scand. 2017;137(1):20–23. doi: 10.1111/ane.12816 [DOI] [PubMed] [Google Scholar]
- 29.Martin J, Hamilton B, Osterman M, Driscoll A. Births: Final Data for 2019. Natl Vital Stat Reports. 2021;70(2):1–51. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.