OBJECTIVE This study compared vascularized and nonvascularized bone grafts for the reconstruction of segmental defects of the mandible. METHODS The results in 39 patients having vascularized bone grafts (38 fibulas and one iliac crest) and 29 patients having nonvascularized bone grafts (26 iliac crest [22 corticocancellous block grafts, four cancellous bone grafts in a tray] and three rib grafts) for segmental mandibular reconstruction were evaluated in terms of overall success rate, total number of surgeries performed, total blood loss, total number of hospital days, and total number of hours in the operating room. RESULTS Of 39 vascularized bone grafts, two failed (95% success rate), whereas of 29 nonvascularized bone grafts, seven failed (76% success rate). Failure for the nonvascularized bone grafts was closely correlated to the length of the defect. Nonvascularized bone graft patients underwent an average of one more surgical procedure for total reconstruction than vascularized bone graft patients, including osseointegrated implants. However, vascularized bone graft patients spent a mean of over 14 additional days in the hospital for all of their reconstructive procedures and an additional 3 hours in the operating room as compared with nonvascularized bone graft patients. Blood loss was similar in both groups (1,100 mL). Only 20% to 24% of patients in each treatment group have completed reconstruction to include osseointegrated implants. CONCLUSIONS The success rate for vascularized bone grafting is high and is the treatment of choice when primary reconstruction is required, when the patient has been previously irradiated, or when simultaneous replacement of soft tissue is required. Vascularized bone grafts are also the treatment of choice for mandibular replacements over 9 cm in length. Nonvascularized bone grafts create a better contour and bone volume for facial esthetics and subsequent implant insertion, and may be the treatment of choice for secondary reconstruction of defects less than 9 cm in length.