Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis

Abstract

This systematic review and meta-analysis aimed to assess whether compression stockings or other interventions reduce the incidence of venous ulceration after acute deep venous thrombosis.We searched PubMed and Embase for randomized controlled trials (RCTs), restricted to English, Spanish, and Hebrew, related to post-thrombotic syndrome and venous ulceration in participants with confirmed deep venous thrombosis. Our primary statistical assessment was the Peto odds ratio (OR).Our search generated 23 RCTs meeting inclusion and exclusion criteria, summing 6162 patients and 146 ulcerative events. Trials were categorized into compression, low-molecular-weight heparin (LMWH), procedural thrombolysis, medical thrombolysis, or miscellaneous. Six compression trials were identified, of which five were included in meta-analysis. Compression compared with placebo did not reduce venous ulceration (OR, 0.915; 95% confidence interval [CI], 0.475-1.765), and long-term compression was not superior to short-term compression (OR, 1.36; 95% CI, 0.014-1.31). Four LMWH trials were identified but were not subjected to meta-analysis because of intertrial heterogeneity. One trial, comparing extended tinzaparin with warfarin, demonstrated eight ulcers in the warfarin group and one ulcer in the LMWH group (relative risk, 0.125; P < .05). Three procedural thrombolysis trials were pooled into meta-analysis; fewer ulcerative events occurred in procedural thrombolysis patients, but the effect was not significant (OR, 0.677; 95% CI, 0.338-1.358). Eight medical thrombolysis trials were identified. Pooled analysis of five trials demonstrated a protective effect on ulceration in streptokinase patients vs standard heparinization (OR, 0.125; 95% CI, 0.021-0.739). However, these trials were of poor-quality study design, had small sample size, and had poor overall outcomes. Miscellaneous studies included a trial of hidrosmina, a vasoactive flavonoid, and a trial comparing 6-month warfarin treatment with 6 weeks; neither trial had significant outcomes. Intertrial heterogeneity was not adequately assessed with the I2 value as venous ulceration is a rare event; the Grading of Recommendations Assessment, Development, and Evaluation evidence for most trials was very low, with the exception of procedural thrombolysis trials, for which it was low.We found insufficient evidence to assess whether compression or other interventions protect against venous ulceration. To develop guidelines for treatment decisions related to prevention of venous ulceration, high-powered RCTs investigating venous leg ulcers as a primary outcome are required.