Compression stockings to prevent post-thrombotic syndrome in adults, a Bayesian meta-analysis

Abstract

Background: Prescription of compression stockings to prevent post-thrombotic syndrome (PTS) in adults is controversial. We sought to estimate the efficacy of compression stockings vs. placebo/no intervention (control) in preventing PTS, and to estimate the probability of observing a benefit when prescribing compression stockings to prevent PTS.

Methods: We conducted a systematic review of the literature in MEDLINE, EMBASE, and the Cochrane Central Register of Randomized Trials, searching for randomized controlled trials that compared compression stockings, applied in the acute setting of deep vein thrombosis, vs. control to prevent PTS. We used a Bayesian approach for data analysis.

Results: Four studies met our inclusion criteria. When comparing intervention vs. control, the estimated odds ratio (OR) was 0.57 (95% Credible Interval (CrI): 0.21 to 1.20) for PTS vs. no PTS and 0.79 (95% CrI 0.31 to 1.67) for severe vs. no/mild/moderate PTS. The probabilities of observing treatment benefits in the population if prescribing compression stockings ranged between 47% (large benefit, OR < 0.50) and 95% (small benefit, OR < 1.00) for any PTS and between 16% and 82% (from large to small benefit) for severe PTS. The probabilities of observing benefit of compression stockings in a future study ranged 44%-76% and 25%-72% (from large to small benefit) for any PTS and severe PTS, respectively.

Conclusion: Despite heterogeneity, data show that it is still probable to observe some degree of treatment benefit when prescribing compression stockings and to observe some degree of treatment benefit in a future study.

Keywords: Adult; Bayes theorem; Odds ratio; Postphlebitic syndrome; Postthrombotic syndrome; Probability; Venous thrombosis.