Pregnancy in spina bifida patients: a comparative analysis of peripartum procedures and complications

Spina bifida (SB) is caused by a failure in neural tube closure that can present with lower extremity sensory deficits, paralysis, and hydrocephalus. Medical advances have allowed increased pregnancies among SB patients, but management and pregnancy-associated complications have not been thoroughly investigated. The objective is to delineate peripartum procedures and complications in patients with SB. A national de-identified database, TriNetX, was retrospectively queried to evaluate pregnant SB patients and the general population. Procedures and complications were investigated using corresponding ICD-10 and CPT codes within 1 year of pregnancy diagnosis. 11,405 SB patients were identified and compared to 9,269,084 non-SB patients. SB patients were significantly more likely to undergo cesarean delivery (1.200; 95% CI [1.133–1.271]) and less likely to receive neuraxial analgesia (0.406; 95% CI [0.383–0.431]). Additionally, patients with SB had an increased risk of seizures (3.922; 95% CI [3.529–4.360]) and venous thromboembolism (VTE) (3.490; 95% CI [3.070–3.969]). Risks of preeclampsia and hemorrhage were comparable. SB patients with hydrocephalus and Chiari malformation type 1 (CM-1) or type 2 (CM-2) were compared to patients without these comorbid conditions. This sub-group analysis showed a significantly increased risk of having cesarean deliveries (SB with hydrocephalus: 12.55%, S.B. with CM-1 or CM-2: 12.81% vs. SB without hydrocephalus or CM, 6.16%) and VTE (3.74%, 2.43% vs. 0.81%). There were also increased risks of hemorrhage and seizures and decreased use of neuraxial analgesia, but the sample size was insufficient. SB patients were more likely to undergo cesarean section and exhibit peripartum complications compared to those without SB.


Introduction
Spina bifida (SB) is a congenital disorder in which spinal cord development is impaired by incomplete neural tube closure [1]. The life expectancy of patients with SB has dramatically increased over the last few decades, with over 80-95% of patients surviving into adulthood [2]. The wellknown health complications of surviving SB patients are the sequela of urological and orthopedic symptoms. Best practices for clinical management in these areas have been heavily researched and implemented [2]. However, as SB patients survive longer into adulthood, new health discrepancies emerge and remain poorly defined.
Management and pregnancy outcomes in SB patients are inadequately addressed in both the literature and guidelines. Over the last 10 years, the number of patients with SB giving birth has increased by 56%, which calls for increasing necessity and urgency for this area of research [3]. A few studies have demonstrated that SB patients suffer from an increased risk of peripartum complications, but these studies are limited by small cohort sizes hindering their validity and generalizability. In 2018, Shepard et al. demonstrated that SB patients were more likely to undergo cesarean delivery than vaginal but did not describe anesthetic management or discuss other peripartum procedures [4].

3
This research aims to leverage a large population database to identify peripartum procedures and complications of patients with SB compared to those of the general population (pregnant mothers without SB). Implications of this research include augmenting the limited literature by offering more evidence on the peripartum complications and risks faced by SB patients. Furthermore, these findings will inform healthcare providers of procedural and management options when providing maternal care specific to SB patients.

Materials and methods
This study uses a real-world multi-institutional healthcare database, the TriNetX Diamond Network, to collate information about pregnant patients with SB compared to pregnant patients in the general population. TriNetX Diamond Network provides access to 92 healthcare organizations (HCOs) in the USA and their de-identified electronic medical record data. The data includes patient demographics, diagnoses, medications, laboratory values, genomics, and procedural codes. The validity of this database has been previously described [5,6]. The Institutional Review Board did not require patient consent since the data was de-identified prior to this study's use.
Data extraction and analysis from the TriNetX database were performed in December 2021. Datasets were retrospectively queried for peripartum procedures and complications using the International Classification of Diseases, Tenth Revision (ICD-10), and current procedural terminology (CPT) codes within 1 year of a pregnancy diagnosis.
Procedures and complications of interest included a cesarean section (C-section), utilization of neuraxial analgesia, development of pulmonary embolism or deep venous thrombosis (PE/DVT), preeclampsia, intrapartum/peripartum hemorrhage, and seizures. Inclusion criteria included a diagnosis of spina bifida (Q05) or spina bifida occulta (Q76.0) and pregnancy, childbirth, and the puerperium (O00-O9A). In addition, the exclusion criterion included a pregnancy with an abortive outcome within 9 months of the pregnancy diagnosis (O00-O08). A sub-group analysis was completed for all pregnant SB mothers with hydrocephalus and Chiari malformation type 1 (CM-1) or type 2 (CM-2). ICD-10 and CPT codes for diagnoses, procedures, and complications are listed in Table 1.
The following demographic information was also collected: age at pregnancy, race, and ethnicity. Cohorts were further stratified into age groups by age at pregnancy: 0-17 years, 18-29 years, 30-39 years, ≥ 40 years, and at all ages. For any outcomes with an n ≤ 10, the TriNetX database reports a value of n = 10 to protect and comply with patient health privacy guidelines. Therefore, the data was considered insufficient if the number of patients with an outcome was less than 10 (n ≤ 10) for the purposes of this study. In each age cohort, patients are also excluded if they met the index event more than 20 years ago. Hazard ratios were calculated using R's survival package v3.2-3 and validated, comparing the output to SAS version 9.4. Chi-square analysis and logistic regression were performed on categorical variables. Statistical significance was determined by p ≤ 0.05.

Patient characteristics
The mean age of pregnancy among SB patients was 29.1 years of age ± 8.81 years compared to 29.2 years ± 8.39 years for the general population. A majority of the patients were classified as unknown race and ethnicity in both cohorts. Following this, the second most common races and ethnicities were White/ non-Hispanic and Latino, respectively. Table 2 provides a summary of the demographic information for all patients.

Clinical outcomes
A total of 11,405 SB pregnant patients who went to delivery were identified and compared to 9,269,084 pregnant patients without SB who went to delivery (general population). The breakdown of the number of patients in each age group at the time of pregnancy between the SB and general population groups is shown in Table 2. In addition, the breakdown of each procedure and complication outcome of interest is displayed in Table 3 by each age cohort.

C-section
At delivery, it was observed that mothers with SB were significantly more likely to undergo c-section. (

Administration of neuraxial analgesia
Pregnant mothers with SB across all age groups were significantly less likely to receive neuraxial analgesia

Intrapartum and peripartum hemorrhagic complications
Overall, and across all age groups, the risks of intrapartum and peripartum hemorrhagic complications were observed to be either non-significant or had insufficient data available for analysis in the database.

Spina bifida with hydrocephalus, type 1 Chiari malformation, or type 2 Chiari malformation
A sub-group analysis of pregnant that compared SB patients that also had hydrocephalus or Chiari malformation (Type I and II) to pregnant SB patients without these comorbid conditions (SB-WOC) was performed. This analysis identified 3506 SB patients with comorbid hydrocephalus (SB-H) and 8839 SB patients with comorbid type 1 or type 2 Chiari malformation (SB-CM). These patients exhibited higher risks of undergoing C-sections (SB-H, 12.55%; SB-CM, 12.81%, vs. SB-WOC, 6.16%).

Discussion
At delivery, SB mothers were significantly more likely to undergo cesarean delivery and less likely to receive neuraxial analgesia. SB patients also had an increased risk of complications such as seizures and PE/DVT. Over the last few decades, more patients with SB have been giving birth due to multiple medical advancements that have significantly improved their morbidity and mortality. Still, very limited research addresses issues related to pregnancy in this population [3]. Furthermore, patients with SB report having inadequate information and knowledge about their sexual and reproductive health [7,8]. This study aims to help bridge the knowledge gap on pregnancy outcomes among mothers with SB.
In 2019, Shepard et al. conducted the first large-scale study that analyzed pregnancy outcomes associated with vaginal and cesarean deliveries in SB mothers [4]. The present study also found a higher risk of undergoing c-section in SB mothers, which is in line with previously reported findings [4,9]. Higher rates of wheelchair use and the use of other assistive devices in the SB population have been associated with increased rates of c-section, which may provide a partial explanation for this finding [10,11]. Additionally, the limited range of motion due to orthopedic deformities can make proper positioning required for vaginal delivery unsuitable for this patient population [12]. Furthermore, neurogenic bladder requiring urological procedures is very common among SB patients. A c-section is often the safest delivery method that avoids disrupting these reconstructions and prevents further injury [13,14]. Our study also found insignificant differences in the risk of hemorrhagic complications for SB versus non-SB. Delivering mothers is also observed in Shepard et al. study regardless of delivery method [4].
We elaborated on these findings by looking at specific outcomes of the administration of neuraxial analgesia, PE/ DVT, preeclampsia, and seizures. Neuraxial analgesia is less commonly used in patients with preexisting central nervous system disorders, such as SB, due to worsening neurologic outcomes. In addition, debate exists over the safety profile of administering neuraxial analgesia in this patient population due to reports of the increased risk of accidental puncture of the dura due to anatomic variations [15]. There also exists a potential presence of a tethered cord, which must first be excluded by magnetic resonance imaging (MRI) and may not be necessarily feasible for a patient that presents in active labor [16]. Furthermore, scar tissue and scoliosis are common in SB patients, which can also make the administration of neuraxial analgesia more challenging and cumbersome [17,18]. This is reflected in our results that showed a statistically significant decrease in the administration of neuraxial analgesia across all age cohorts. The risk of developing venous thromboembolism (PE/ DVT) is multifactorial. Our study showed a threefold increase in the risk of developing PE/DVT. Pregnancy puts mothers at a significantly increased risk of venous thromboembolism due to its propensity for a hormone-mediated hypercoagulable state [19,20]. Other factors may include flow stasis due to uterine obstruction and decreased mobility. Obesity and limited mobility are other risk factors for VTE, which are more prevalent in the spina bifida population [10,[21][22][23][24]. The interplay of pregnancy, immobility, and obesity could explain the statistically significant increase in PE/DVT seen in delivering SB patients.
The incidence of preeclampsia and eclampsia in SB patients is unknown. Many studies report an increased risk of hypertension in SB patients, but little is known about the incidence of preeclampsia compared to the general population [25,26]. In a qualitative study by Tong et al., three out of six patients reported a diagnosis of preeclampsia [26]. Our study reports an increased risk of preeclampsia in patients ≥ 40 years old. Comparable risks of preeclampsia were seen across the younger cohorts, which may be partially explained by an existing high prevalence of hypertension among SB patients prior to 20 weeks of pregnancy [25]. Furthermore, with the high prevalence of kidney disease and the interplay of renal function and blood pressure, we expected to see an increased risk of preeclampsia across all cohorts [27][28][29][30]. In our study, the decreased risk in the ≥ 40-year-old cohort was most likely due to a small sample size. The upper bound of the confidence interval is nearly 1.
Hydrocephalus and Chiari malformation are common comorbidities seen in SB patients that may influence pregnancy outcomes [31,32]. Their pregnancies can be complicated by orthopedic deformities and previous neurological surgeries for decompression and placement of ventriculoperitoneal and ventriculoatrial shunts. This analysis demonstrated that hydrocephalus and CM were significantly associated with increased risks of C-section delivery, seizures, and PE/DVT compared to SB patients without these comorbidities. Additionally, symptoms associated with Chiari malformation, such as a headache, may worsen the Valsalva maneuver, making cesarean sections preferred to vaginal deliveries [33][34][35]. Additionally, a large percent of SB patients require shunting for hydrocephalus which may explain the statistically significant nearly fourfold increase in the risk of seizures for mothers with spina bifida in each age cohort [10,32,36,37].
Overall, a complex interplay between these three different comorbidities can present multiple challenges during pregnancy and delivery. Chiari malformation can cause accumulation of cerebrospinal fluid leading to hydrocephalus [38]. Furthermore, Chiari malformation type 2 is associated with myelomeningocele, a severe form of spina bifida [1]. MRI should be obtained at the beginning of and throughout pregnancy to accurately assess the spinal dysraphism and amount of cerebrospinal fluid because shunt malfunctions that require revisions have commonly occurred and been reported in the literature during pregnancy for these patients [17,39,40]. When comparing the risk of complications in these sub-groups, our results for the sub-group analysis are largely consistent with those of our SB cohort without other comorbidities.
Although SB patients are living longer, increasing the number of pregnant patients with SB, there is a significant lack of information regarding sexual and reproductive health in those with SB. Larger cohort studies are needed to thoroughly understand the challenges of pregnancy in order to provide better preconception counseling and peripartum care for patients with SB.
It is thought that spina bifida is a sporadic condition and results from a combination of genetic and environmental factors [1,41]. Genetic counseling and emphasis on adequate folate intake during pregnancy should be discussed to prevent a neural tube defect in the fetus; however, having SB in the present day is not a contraindication in itself to vaginal or Cesarean delivery [42][43][44][45]. Further studies should investigate if gravida affects the risk of developing pregnancy complications.
No study is without limitations. The outcomes with insufficient data were seen in the 0-17 and ≥ 40 population, which was expected with the average pregnancy age, 29.1 years ± 8.81, of SB mothers. Furthermore, there is a greater percentage of abortive outcomes in patients of older maternal age [46,47]. Our study excluded pregnancies with an abortive outcome which may have removed more data on pregnant mothers who were ≥ 40 years old. These two reasons may also explain why our data show an increased risk of preeclampsia among the general population compared to those with spina bifida in mothers ≥ 40 years old.

3
The major limitation of this study is that it was retrospective. The data in the TriNetX database is collected for billing purposes but not for clinical use, which eliminates a lot of clinical data and is subject to coding errors. Patient-level information regarding specific diagnostic tests and radiographic information was not provided to confirm disease, gravida, medications, and insurance status. The risks of peripartum outcomes may vary depending on the level of defect. As a database study, our study relies on the accuracy of each healthcare organization's CPT and ICD 10 coding, which is a potential source of inaccurate or incomplete reporting. Further limitations include the data source is restricted to healthcare organizations in the USA that participate in the database retrieval system.
Mothers with SB were more likely to have a c-section and exhibit peripartum complications compared to those without SB. Further studies are needed so that multidisciplinary care teams can address the complexity of conditions in SB mothers to provide optimal comprehensive peripartum care.
Author contribution All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Bao Y. Sciscent, Debarati Bhanja, Lekhaj C. Daggubati, Casey Ryan, David R. Hallan, and Elias Rizk. The first draft of the manuscript was written by Bao Y. Sciscent, Debarati Bhanja, Lekhaj C. Daggubati, Casey Ryan, and David R. Hallan, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.