The timing, extent, progression and regression of deep vein thrombosis in immobile stroke patients: observational data from the CLOTS multicenter randomized trials

Summary.  Background: Deep vein thrombosis (DVT) is an important complication of stroke, but the evidence to support commonly used prophylactic strategies is conflicting. Objectives: To describe the incidence, extent, associated clinical features and evolution of DVT after stroke. Patients/Methods: The CLOTS trials 1 and 2 together randomized 5632 immobile stroke patients in 135 hospitals in nine countries. We screened patients for asymptomatic DVT with compression duplex ultrasound (CDU) at about 7-10 days and again at about 25-30 days after enrollment. Results: Six hundred and forty-one (11.4%) of 5632 patients had DVT detected on the first CDU scan at a median of 8 days (interquartile range [IQR] 7-10 days) after enrollment, and an additional 176 (3.1%) had a DVT on the second CDU scan at a median of 28 days (IQR 26-30 days). Of the 817 with DVTs, 289 (35%) were symptomatic and 39 (5%) had pulmonary embolism (PE) confirmed by imaging. Six hundred and seventy-six (83%) were unilateral, 141 (17%) were bilateral, 322 (39%) were limited to calf veins, 172 (21%) were popliteal, and 323 (40%) were femoral. Among the 542 patients with DVT and a weak leg, the DVT affected the weaker leg in 396 (73%), the stronger leg in 59 (11%), and was bilateral in 87 (16%). Among the 318 patients with a DVT detected on the first CDU scan who had a second scan, the DVT regressed in 148 (47%), stayed the same in 140 (44%), and progressed in only 30 (9%). Conclusions: Although most DVTs develop within the first week, some develop later, and some early DVTs progress. Any prophylaxis needs to be started early but ideally continued for at least 4 weeks.