Deep vein thrombosis in rehabilitating stroke patients. Incidence, risk factors and prophylaxis

A total of 503 consecutive cases of suspected stroke were examined for potential eligibility based on recent development of a paralyzed limb. Of 123 otherwise eligible subjects, 22 were found by ultrasound to have deep vein thrombosis (DVT) on admission. Therefore, 101 patients were assigned randomly to one of the treatment groups or to the control group. The three treatments were adjusted-dose heparin, external pneumatic compression and functional electrical muscle stimulation. An ultrasound examination of the lower extremities was conducted twice a week on each patient until completion of the study (28 days or discharge, whichever came first). Electrical muscle stimulation was discontinued after 4 mo of the study because of discomfort, blister formation and high drop-out rate. Ten patients developed DVT during the study period. In 17 of the 32 cases of DVT, venography was performed, which confirmed the ultrasound findings in every case. The 32 cases of DVT differed from those without DVT by having a higher prevalence of hypertension (P = 0.02), cholesterol (P = 0.08) and a longer time interval between stroke and admission (P less than 0.05). We conclude that ultrasound is effective for DVT detection in the rehabilitation setting, and two-thirds of such cases are detectable on admission.