Beyond static risk scores: Precision thromboprophylaxis for pregnancy-associated venous thromboembolism – A systematic review and dynamic risk-adapted maternal–perinatal prevention framework

Pregnancy-associated venous thromboembolism (VTE) remains an important cause of maternal morbidity and mortality, yet decisions about thromboprophylaxis are commonly based on risk factors assessed at discrete time points and on guideline-specific thresholds. We systematically reviewed contemporary evidence to examine whether a more dynamic approach to risk assessment is justified and to develop an evidence-informed conceptual framework for future investigation. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 principles, studies published from January 1, 2015, to April 30, 2026, were searched across major bibliographic databases; 27 studies met the predefined eligibility criteria. Evidence from epidemiological, prediction, biomarker, and prevention studies indicates that VTE risk changes across pregnancy, delivery, and the postpartum period as hospitalization, immobility, infection, obstetric complications, mode of delivery, hemorrhage, and recovery emerge or resolve. Existing risk-assessment models vary considerably in structure, performance, and intervention thresholds, and external validation remains limited. Biomarker-enhanced and artificial intelligence approaches are promising but are not sufficiently validated for pregnancy-specific clinical use. Evidence for pharmacological prevention is less certain: randomized and systematic-review data leave important uncertainty regarding patient selection, low-molecular-weight heparin intensity, duration, and individual net benefit. These findings support reassessment at clinically meaningful transitions rather than reliance on a single early-pregnancy classification, while recognizing that higher predicted VTE risk does not itself establish benefit from anticoagulation. Our contribution is preliminary: we harmonize the available evidence into an evidence-informed conceptual framework to guide prospective data collection, external validation, and future treatment-benefit research. The framework is not a validated prescribing algorithm and should not replace contemporary guideline-based management.