Objective To evaluate the effectiveness and safety of surgical thrombectomy for acute deep venous thrombosis of lower extremities. Methods Randomized controlled trials of surgery versus conservative treatment were sought from MEDLINE (1966-Jun.2006), EMbase (1974-Jun.2006), The Cochrane Library (Issue 2, 2006), CBM (1989-Jun. 2006) and CMCC (1994-Jun. 2006). Collections of Chinese Congress on Vascular Surgery (1991-Jun.2006) and the journal of Vascular Surgery (2000-Jun. 2006) were handsearched. Two reviewers independently extracted data into a designed extraction form. The guidance in The Cochrane Collaboration’s Handbook was consulted for quality evaluation and data analysis. Results Six potentially eligible studies were identified. Six were included according to the inclusion criteria. The 6-month total patency was significantly higher in the surgical treatment group than in the conservative treatment group with OR 7.26 and 95%CI 2.40 to 21.94, while the 5-year total patency was not different between the two groups with OR 2.59 and 95%CI 0.88 to 7.67. At month 6 and year 5, the incidence of post-thrombosis syndrome (PTS) was significantly higher in the conservative treatment group than in the surgical treatment group with OR 0.11, 95%CI 0.59 to 1.59, OR0.18, 95%CI 0.06 to 0.60 respectively. The incidence of 10-year PTS and the results of valvular function measurements were similar between the two groups. The incidence of pulmonary thrombosis was also comparable between the two groups with OR 1.40 and 95%CI 0.39 to 4.97. Conclusion Surgical thrombectomy may improve the extent of patency and venous valvular sufficiency in the short term, but without increasing the patency rate. There is no enough evidence to assess whether surgical throbectomy improves long-term outcomes. It is safe to preform surgical thrombectomy. The small number of patients randomised and the low quality of the trials decreases the reliability of the current evidence. Therefore, more high quality randomised controlled studies should be done, to determine the long-term outcomes of surgical thrombectomy.