Objective:

We conducted this review to support the United States Preventive Services Task Force (USPSTF) in updating its 2014 recommendation on daily low dose aspirin use during pregnancy for individuals at increased risk for preeclampsia. We systematically reviewed updated evidence on the effectiveness and potential harms of daily aspirin use in pregnancy to prevent morbidity and mortality associated with preeclampsia.

Data Sources:

We performed comprehensive searches of MEDLINE, PubMed (publisher-supplied only), Embase, and the Cochrane Collaboration Registry of Controlled Trials for studies published between January 2013 and July 2019. A research librarian developed and executed the search strategy. Studies included in the prior review to support the 2014 recommendation and studies referenced in recently published reviews were also considered for inclusion.

Study Selection:

We reviewed 3749 newly identified abstracts and 183 articles against prespecified inclusion criteria developed for the current review. Studies were included if they evaluated daily treatment during pregnancy with aspirin not combined with any other medication at a dosage of ≥50mg daily compared with placebo or no treatment. Included studies reported incidence of preeclampsia or serious maternal, perinatal, and longer-term health outcomes. For studies evaluating aspirin effectiveness, eligible study designs included randomized controlled trials (RCTs) and individual participant data meta-analyses that enrolled pregnant persons who were at increased risk for preeclampsia based on personal sociodemographic characteristics, medical history, diagnostic measurements or assays, or risk prediction models. For studies evaluating harms of aspirin use, the study design criteria were expanded to include observational studies and trials conducted in average-risk populations.

Data Analysis:

Eligible studies that were determined to have high risk of bias were excluded for poor quality according to standard USPSTF procedures. Descriptions of study populations, study design characteristics, and outcomes were developed for all included studies. The most consistently reported health outcomes were analyzed using random effects meta-analytic model to calculate pooled intervention effects and subgroup comparisons were tested with meta-regression tests for interaction. We examined whether the results were influenced by small-study effects using the Peters test and visual inspection of forest plots.

Results:

Twenty-three studies (33 articles) met our inclusion criteria. Four studies included in the previous review (including two observational studies) were not included and four new trials were added. We included 18 trials evaluating aspirin effectiveness in individuals at increased risk for preeclampsia; all but one used a matching placebo comparator. Dosages of aspirin ranged from 50-150 mg per day, usually starting in the second or third trimester and continuing until delivery or near term. These interventions were associated with reduced risks of perinatal mortality (pooled RR 0.79 [95% CI 0.66, 0.96], I² 0%), preterm birth (pooled RR 0.80 [95% CI 0.67, 0.95], I² 49%), and small for gestational age/intrauterine growth restriction (SGA/IUGR) (pooled RR 0.82 [95% CI 0.68, 0.99], I² 41%). There was also a statistically significant reduction in the risk of preeclampsia (pooled RR, 0.85 [95% CI, 0.75 to 0.95], I² 0%). Rare maternal health outcomes, such as eclampsia and maternal mortality, occurred too infrequently to estimate preventive effectiveness. In meta-analytic subgroup comparisons, we found no consistent evidence for effect differences related to intervention or population characteristics such as the timing of treatment initiation, the dosage of aspirin used, or participant characteristics. There was evidence of small-study effects for several of the pooled health outcomes.

Studies in average and increased-risk populations did not provide any clear evidence of harms associated with daily aspirin use (<150mg) taken during the second or third trimester of pregnancy, and there was no evidence of differences in harms by aspirin dosage or timing, or for specific populations identified in limited subgroup comparisons. Bleeding related harms were uncommon; our analyses showed null effects for differences in risk of postpartum hemorrhage (pooled RR 1.03, [95% CI, 0.94, 1.12], I² 0%, k=9), or intracranial fetal bleeding (pooled RR, 0.90 [95% CI, 0.51, 1.57]; I² 19%, k=6) between intervention and control groups. There was also no difference in rates of placental abruption (pooled RR, 1.15 [95% CI, 0.76, 1.72], I² 25%, k=10). Longer-term followup from one large trial found no differences in child developmental outcomes between aspirin and placebo exposed groups. No differences were found in a limited set of studies reporting other rare perinatal harms.

Limitations:

Our search was limited to English-language literature and trials conducted in settings other than very high Human Development Index settings were excluded. Large trials conducted in other settings could provide additional relevant information, but evidence from other reviews without this exclusion do not find substantively different results. The meta-analysis results for some outcomes were limited by low numbers of included studies and very few events. Conservative approaches to estimation of pooled effects and testing of subgroup comparisons were used but cannot fully address some limitations in the data. The two largest studies used the same low aspirin dosage (60mg) and enrolled a majority of participants after 16 weeks of gestation and small-study effects cannot be entirely disentangled from other design and population features of the included studies. Despite inequities in preeclampsia rates by race and ethnicity in the United States, most trials enrolled predominantly White populations. Evidence also was lacking on potential longer-term health consequences of low dose aspirin exposure during pregnancy.

Conclusions:

Daily aspirin use in pregnancy for individuals at increased risk for preeclampsia consistently led to beneficial effects on perinatal mortality, preterm birth, fetal growth restriction, and preeclampsia diagnosis across a clinically heterogeneous set of trials. A large body of trial evidence shows no clear signal of serious harms associated with daily low dose aspirin use in the second and third trimesters of pregnancy. Further research from pragmatic and comparative effectiveness trials is needed to identify the best way to identify patients at increased preeclampsia risk who are most likely to benefit from aspirin use, and to obtain greater clarity on the optimal regimen. Finally, the benefits of aspirin prophylaxis may not always reach those at risk, especially among populations with limited access to high quality prenatal care. Implementing strategies that promote equitable access to this intervention could help address preeclampsia health inequities, especially those observed for Black women in the United States.