The Safety of Low-Dose Aspirin on the Mode of Delivery: Secondary Analysis of the Effect of Aspirin in Gestation and Reproduction Randomized Controlled Trial.

Daily LDA taken from preconception through 36 weeks of gestation did not affect the mode of delivery. Findings were similar about those with and without prior cesarean delivery, as well as by parity, and findings were robust to missing outcome. Finding suggest that LDA does not increase risk of cesarean delivery among women with low-to-moderate risk pregnancies, which is particularly important given the increasing number of indications for prescription of LDA during the prepregnancy and antepartum period. Given the lack of association between LDA and mode of delivery, our findings further support that LDA is likely a safe medication in the prepregnancy and antepartum period. It is important to evaluate the risks and benefits associated with this now widely used medication, to better make clinical decisions for patients and counseling patients on risks.

While there are some reports that the use of aspirin in pregnancy can increase the risk of placental abruptions and bleeding, there are other reports that do not support this outcome. 16 which would theoretically increase the risk of cesarean deliveries; however, the differences in rates were not found to be statistically significant (RR = 1.17, 95% CI: 0.93–1.48, I 2 = 36.4%), therefore this likely did not contribute to the lack of differences in cesarean deliveries 12 (RR = 1.17, 95% CI: 0.93–1.48, I 2 = 36.4%; RR = 1.17, 95% CI: 0.93–1.48, I 2 = 36.4%). However, what is known, is that women who develop preeclampsia with severe features are more likely to undergo a cesarean delivery overall. 17 For example, one study reported cesarean rates of 30% with chronic hypertension alone, 40% with superimposed preeclampsia without severe features, and 56% with superimposed preeclampsia with severe features. 17 If LDA is utilized to help keep high-risk women from developing preeclampsia, it is important to ensure that there are no additional risks associated with delivery method. 18 This study suggests that the risk of cesarean delivery is not affected by the addition of LDA among a healthy population of women with one to two prior pregnancy losses. While our study population had a history of one to two prior losses, we excluded participants who self-reported current treatment for a chronic health condition. Given that women with chronic health conditions have a higher risk of cesarean delivery, the association of aspirin with mode of delivery among women with preexisting conditions is an important point for further research. 3 Our study was unique in randomizing participants to LDA during preconception and continuing throughout pregnancy, allowing for an examination of the effect of LDA on mode of delivery among a healthy cohort of women with exposure to aspirin throughout pregnancy. While we relied on medical record abstraction for defining mode of delivery, there are multiple studies that demonstrate the strong validity of medical record abstraction for assessing birth outcomes, especially mode of delivery. 19 , 20 Further, baseline risk factors for cesarean delivery include advanced maternal age, elevated prepregnancy BMI, nulliparity, gestational diabetes, induction of labor from low Bishop’s score, tobacco use, and intrauterine growth restriction. 1 , 2 All of these were accounted for by randomization to the treatment and placebo groups in EAGeR. 13 Limitations of this study include that the majority of patients were Caucasian with higher incomes and education, limiting generalizability to other populations with different patterns of risk factors. Further, risk factors for cesarean delivery are more common in non-White populations. 6 Finally, to examine the concern for increased risk of abruption was difficult to do in this study, as it is a rare outcome and this study was not powered to examine a rare outcome.

This is a secondary analysis of the EAGeR trial, a multicenter, double-blind, block-randomized, placebo-controlled trial evaluating the effects of LDA on live birth. Details of this trial have been outlined previously. 14 , 15 Briefly, 1,228 women aged 18 to 40 years with a history of one to two previous pregnancy losses were included from 2006 to 2012 at four U.S. clinical centers. Exclusion criteria included a history of infertility, pelvic inflammatory disease, tubal occlusion, endometriosis, anovulation, uterine abnormalities, polycystic ovarian syndrome, and contraindications to aspirin. Institutional review board authorization was obtained for the data coordinating center and at all clinical centers. At the University of Utah Medical Center, the institutional review board (IRB) of Inter-mountain Healthcare Office of Research approved IRB no.: 1002521-EAGeR on November 4, 2010. At the University of Buffalo, the Health science IRB approved HSIRB project no.: SPM0900107A on September 18, 2007. For the Denver review, the COMIRB number was 08–0982 and the Wright Center for Graduate Medical Education Institutional Review Board approved Protocol No: HHSN275200403394 on March 6, 2013. Patient safety was optimized by a Data Safety and Monitoring Board (DSMB) and the trial was registered with ClinicalTrials.gov , number NCT00467363 . Women were randomized 1:1 to take daily aspirin of 81 mg with folic acid of 400 μg or placebo with folic acid of 400 μg. Participants began taking LDA or placebo on days 2 to 4 of the first menstrual cycle of follow-up, and continued for up to six menstrual cycles of attempting pregnancy and, if they became pregnant, through 36 weeks’ gestation. The primary outcome for this analysis was mode of delivery. Of 1,228 women enrolled, 597 delivered a live-born infant and were evaluated for mode of delivery. Mode of delivery was collected as part of the study design using medical record abstraction. Patients were categorized as having a cesarean versus vaginal delivery. Differences in cesarean versus vaginal delivery were further evaluated by parity and history of prior cesarean delivery ( Fig. 1 ). Participant characteristics were collected at enrollment into the study, including age, prepregnancy body mass index (BMI), race/ethnicity (white versus non-White), education (high school), household income (<$40,000, 40,000 to <100,000, and ≥100,000), smoking in the past year (yes vs. no), number of previous pregnancies (1–4), number of previous losses (1–2), number of previous live births (0–2), and number of previous vaginal deliveries and cesarean deliveries. Complications in the current pregnancy were assessed through medical record abstraction and include preterm delivery, premature rupture of membranes, gestational diabetes, and hypertensive disorders of pregnancy. Descriptive statistics were summarized as counts and percentages or means and standard deviations. Differences in descriptive statistics across LDA and placebo groups were calculated using Fisher’s exact test for categorical variables and t -tests for continuous variables. Participants missing information on mode of delivery ( n = 19; 8 in LDA group and 11 in placebo group) were excluded from analysis. Log-binomial regression was used to estimate the relative risk (RR) and 95% confidence intervals (CIs) for delivery outcomes among those assigned to LDA versus placebo. We first used an intent-to-treat approach, evaluating the association of LDA with mode of delivery among all 1,088 participants with complete follow-up data, therefore excluding the 19 patients with missing information. We then evaluated the relationship of LDA with mode of delivery among the 597 participants who had a live birth, addressing the potential for selection bias by excluding participants who became pregnant or who experienced a loss by weighting models for the inverse probability of having a live birth given the following predictors: treatment assignment, number of prior live births, number of prior losses, number of prior cesarean sections, age, BMI, race/ethnicity and cigarette smoking prior to pregnancy. In both approaches, we evaluated the effect of LDA treatment with cesarean versus vaginal delivery, and by history of cesarean delivery and parity through inclusion of interaction terms. We additionally conducted a sensitivity analysis to determine whether missing information on mode of delivery may have biased our findings. Participants with missing data were randomly assigned to (1) spontaneous or induced, meaning requiring an induction of labor, vaginal delivery versus cesarean; or (2) SVD versus induced vaginal delivery or cesarean, with the proportion assigned to each group ranging from 0 to 1.0 in intervals of 0.2. To evaluate the potential for differential bias by assignment to LDA versus placebo, all potential combinations of proportions across groups were evaluated. All analyses were conducted in SAS 9.4 (Cary, NC).

Overall, we found that daily LDA taken from preconception through 36 weeks was not associated with mode of delivery; this did not change in preterm versus term deliveries. There was no effect on success of TOLAC or differences by parity. However, future studies should evaluate whether maternal morbidity and mortality are decreased with the use of LDA in patients with elevated risk for abnormal placentation and placental insufficiency including preeclampsia, 21 – 23 other hypertensive disorders of pregnancy, and assisted reproductive technology. This study is also not able to generalize to high-risk patients including those with risk of preeclampsia or gestational diabetes. In all, our findings suggest a lack of association between LDA and mode of delivery in preconception and pregnancy and thus may provide reassurance to women who may benefit from the medication and the physicians who prescribe them.