OC 51.1 - Incidence of thromboembolic events among women with severe postpartum hemorrhage treated with recombinant activated factor VII: a systematic literature review and meta-analysis

Background: Recombinant activated factor VII (rFVIIa) can be used to treat severe postpartum hemorrhage (sPPH) when uterotonics fail to stop bleeding. However, data on the safety of rFVIIa treatment of sPPH from adequately powered trials are lacking.

Aims: We performed a systematic review to analyze previously published data on the incidence of arterial and venous thromboembolic events (TEs) in women with PPH treated or not with rFVIIa.

Methods: Peer-reviewed publications (published 01.1996–08.2022) and conference abstracts (published 01.2017–08.2022) were extracted from databases (EMBASE, MEDLINE, BIOSIS, Current Contents, Cochrane Library) using search terms related to “thromboembolism” or “infarction” and “postpartum hemorrhage”. Data were extracted from eligible publications following the inclusion/exclusion criteria (Table 1). Study quality was assessed based on a checklist proposed by Downs and Black and rated as "higher" if the score was ≥15 (Downs & Black. J Epidemiol Community Health 1998;52(6):377–84). The primary analysis included descriptive summary statistics and an estimation of the proportion of TEs using a generalized linear mixed model based on the binomial distribution with study as a random effect.

Results: Among 1,637 potentially eligible studies, 55 publications were included that reported data for 611 women treated and 32,488 women not treated with rFVIIa. In publications reporting results on women with sPPH, the global estimated proportion of all TEs in women treated or not treated with rFVIIa was 1.82% (prediction interval: 0.30–10.23) and 0.72% (prediction interval: 0.03–16.47), respectively (Figure 1). Twenty-eight studies were rated as "higher-quality" and included in a separate analysis.

Conclusion(s): The estimated proportions of all TEs for women treated and not treated with rFVIIa were similarly small and prediction intervals were wide and largely overlapping. Larger, well-designed studies are required to reveal more accurate incidences of TEs in women with sPPH treated or not treated with rFVIIa.