OBJECTIVE Although superficial venous reflux is an important determinant of post-thrombotic skin changes, the origin of this reflux is unknown. The purpose of this study was to evaluate the frequency and etiologic mechanisms of superficial venous reflux after acute deep venous thrombosis (DVT). METHODS Patients with a documented acute lower extremity DVT were asked to return for serial venous duplex ultrasound examinations at 1 day, 1 week, 1 month, every 3 months for the first year, and every year thereafter. Reflux in the greater saphenous vein (GSV) and lesser saphenous vein (LSV) was assessed by standing distal pneumatic cuff deflation. RESULTS Sixty-six patients with a DVT in 69 lower extremities were followed up for a mean of 48 (SD +/- 32) months. Initial thrombosis of the GSV was noted in 15 limbs (21.7%). At 8 years, the cumulative incidence of GSV reflux was 77.1% (SE +/- 0.11) in DVT limbs with GSV involvement, 28.9% (+/- 0.09%) in DVT limbs without GSV thrombosis, and 14.8% (+/- 0.05) in uninvolved contralateral limbs (P <.0001). For LSV reflux, the cumulative incidence in DVT limbs was 23.1% (+/- 0.06%) in comparison with 10% (+/- 0.06%) in uninvolved limbs (P =.06). In comparison with uninvolved contralateral limbs, the relative risk of GSV reflux for DVT limbs with and without GSV thrombosis was 8.7 (P <.001) and 1.4 (P =.5), respectively. The relative risk of LSV reflux in thrombosed extremities compared with uninvolved extremities was 3.2 (P =.07). Despite these observations, the fraction of observed GSV reflux that could be attributable to superficial thrombosis was only 49%. CONCLUSIONS Superficial venous thrombosis frequently accompanies DVT and is associated with development of superficial reflux in most limbs. However, a substantial proportion of observed reflux is not directly associated with thrombosis and develops at a rate equivalent to that in uninvolved limbs.