Objective To evaluate the effectiveness of reamed versus nonreamed intramedullary nailing for femoral fractures. Methods Randomized controlled trials (RCTs) and clinical controlled trials (CCTs) were identified from MEDLINE (1966-2004.5), EMBASE (1966-2004.5), Cochrane Library (Issue 2, 2004), Cochrane Musculoskeletal Injuries Group Database (2004.5), and CBM disc (1979-2004.5). We handsearched Chinese Journal of Orthopaedy (from establishment to May 2004) and Orthopaedic Journal of China (from establishment to May 2004) . RCTs and CCTs were included. Data were extracted by two reviewers with designed extraction form. RevMan 4.2.3 software was used for data analysis. Results Five RCTs and two CCTs were included. The combined results of meta-analysis showed that reamed intramedullary nailing for femoral fractures can reduce the rate of nonunion (RR=0.38, 95%CI 0.17 to 0.83, P=0.01) and the rate of implant failure (RR=0.42, 95%CI 0.20 to 0.89, P=0.02). Conclusions Compared with nonreamed intramedullary nailing for femoral fractures, reamed intramedullary nailing can reduce the rates of nonunion and implant failure. However, the relation between reaming or pulmonary complications, the time of union, infection, malunion, operative time, and blood loss needs further study.