Graduated compression stockings did not prevent deep venous thrombosis after stroke and increased skin complications

Question In patients with stroke, do thigh-length graduated compression stockings (GCSs) prevent deep venous thrombosis (DVT)? Methods Design Randomized controlled trial (Clots in Legs Or sTockings after Stroke [CLOTS] trial 1). ISRCTN28163533. Allocation Concealed.* Blinding Blinded (outcome assessors for DVT).* Follow-up period 30 days. Setting 64 hospitals in the UK, Italy, and Australia. Patients 2518 patients (mean age 76 y, 51% women) who were admitted to hospital within 1 week of an acute stroke and were immobile. Exclusion criteria included peripheral vascular disease, diabetic or sensory neuropathy, suspicion of potential for skin damage from the stockings, and subarachnoid hemorrhage. Intervention Continuous use of GCSs on both legs until the patient was independently mobile, was discharged, or refused to wear them or the staff had concerns about the patients skin (n =1256); or avoidance of GCSs unless an indication for their use developed (n =1262). Outcomes Primary outcome was definite or probable symptomatic or asymptomatic DVT in the popliteal or femoral veins, detected on screening compression Doppler ultrasonography or in response to symptoms. Secondary outcomes included symptomatic proximal DVT, any DVT, pulmonary embolism, and skin complications. The trial had 90% power to detect a 4% absolute reduction in the primary outcome. Patient follow-up 89% (intention-to-treat analysis). Main results GCSs did not reduce risk for proximal DVT or other measures of venous thromboembolism (Table). Skin complications (breaks, ulcers, blisters, or necrosis) were more frequent in the GCS group (Table). Conclusion In patients with recent stroke, thigh-length graduated compression stockings did not prevent deep venous thrombosis and increased skin complications. Thigh-length graduated compression stockings vs avoidance of stockings to prevent deep venous thrombosis (DVT) in patients with recent stroke Outcomes Stockings Avoidance of stockings At 30 d RRR (95% CI) NNT Proximal DVT 10.0% 10.5% 1.8% (23 to 22) Not significant Symptomatic proximal DVT 2.9% 3.4% 16% (30 to 46) Not significant Any DVT 16% 18% 8.4% (9 to 23) Not significant Pulmonary embolism 1.0% 1.6% 35% (30 to 68) Not significant RRI (CI) NNH (CI) Skin complications 5.1% 1.3% 301% (136 to 572) 26 (14 to 57) Abbreviations defined in Glossary. RRR, RRI, NNT, NNH, and CI calculated from data in article. Excluding patients who were lost to follow-up or died before any primary outcome, and adjusted for delay from onset to randomization, stoke severity, and leg strength. Commentary DVT and pulmonary embolism are important causes of morbidity and mortality after ischemic stroke. CLOTS 1 is the first large assessment of the safety and efficacy of thigh-length GCSs for DVT prophylaxis after recent stroke. Contrary to the reduced incidence of DVT seen in postsurgical patients, this trial failed to show a reduction in venous thromboembolism (VTE) with GCSs. This apparently discordant result may be explained by the level of functional impairment and advanced age of poststroke patients, both of which have been identified as risk factors for VTE after stroke. Several issues may affect the validity of results from CLOTS 1. The lower-than-expected rate of proximal DVT in the control group (11% vs 15%) reduced the power of the trial. Instead of the more rigorous venography, Doppler ultrasonography was used to screen for asymptomatic DVT; however, a recent review strongly suggests that asymptomatic proximal DVT assessed with ultrasonography can serve as a valid surrogate parameter (1). A relatively small percentage of patients in CLOTS 1 received unfractionated or low-molecular-weight heparin (LMWH). This finding is surprising given guideline recommendations supporting LMWH use in patients at high risk for VTE in the poststroke period (2). However, these agents have caused concern because their reduction in VTE is matched by a similar increase in hemorrhagic conversion of ischemic stroke (3). The apparent lack of efficacy with GCSs may ultimately make LMWH the best choice for preventing VTE after stroke (4). Results from CLOTS 3, which assesses the efficacy of intermittent pneumatic compression boots in reducing DVT risk after stroke, are still pending. For now, guidelines recommending GCSs in virtually all patients with stroke require revision.