Comparative effectiveness of lower-dose vs higher-dose aspirin for preeclampsia prevention: a systematic review and...

To determine whether higher-dose aspirin (150-162 mg/d) reduces preeclampsia risk compared with lower-dose aspirin (75-81 mg/d) in pregnant women at elevated risk, using pooled evidence from all available head-to-head randomized controlled trials.

PubMed/MEDLINE, Cochrane CENTRAL, Embase, Scopus, and ClinicalTrials.gov were searched from inception through April 27, 2026.

Parallel-group randomized controlled trials comparing at least two aspirin dose arms (head-to-head dose comparison) in pregnant women at elevated risk for preeclampsia, with a binary preeclampsia outcome reported. Aspirin-vs-placebo trials were excluded.

Log odds ratios were pooled using restricted maximum likelihood (REML) estimation with the Hartung-Knapp-Sidik-Jonkman (HKSJ) correction. Risk of bias was assessed using the Cochrane RoB 2 tool; evidence certainty was graded using GRADE. Egger's test, Duval-Tweedie trim-and-fill, and univariate meta-regression (geography, dose ratio, gestational age at initiation) were performed.

Six randomized controlled trials enrolling 1099 participants across five countries were included. REML+HKSJ pooled OR 1.93 (95% CI 0.84-4.42, P=.096; I²=64.9%), representing approximately 48% lower odds of preeclampsia with higher-dose aspirin, a clinically substantial effect size that did not reach conventional statistical significance, primarily due to limited sample size (N=1099) and substantial between-study heterogeneity. Egger's test was significant (P=.010); trim-and-fill estimated three missing studies (adjusted OR 1.92, 95% CI 1.47-2.50). No significant dose advantage was seen in North American trials (OR 1.25, 95% CI 0.67-2.34). A safety signal for placental abruption was identified with higher-dose aspirin in one large trial (8 vs 0 events). GRADE certainty: Very Low.

Applying conservative statistical methods (REML+HKSJ), higher-dose aspirin (150-162 mg/d) was associated with approximately 48% lower odds of preeclampsia vs lower-dose aspirin (75-81 mg/d; OR 1.93), a clinically meaningful effect that did not reach conventional statistical significance (95% CI 0.84-4.42, P=.096) due to limited sample size and heterogeneity. Clinical significance and statistical significance must be considered independently; the magnitude of this effect warrants serious attention in guideline discussions. A placental abruption safety signal warrants further investigation. Larger, harmonized head-to-head trials with preterm preeclampsia as the primary endpoint are needed.