BACKGROUND: The most important long-term complication following total hip arthroplasty is periprosthetic femoral osteolysis. A sizeable proportion of patients who undergo arthroplasty are obese. While patient activity, implant type, and quality of fixation are known risk factors for osteolysis, the literature concerning obesity is sparse and controversial. Our primary objective was to evaluate the influence of obesity on the risk of osteolysis five and ten years after primary total hip arthroplasty with a cemented stem. Secondary objectives were to evaluate clinical outcome and patient satisfaction. METHODS: We conducted a prospective cohort study of patients undergoing hip arthroplasty with a third-generation stem-cementing technique from 1996 to 2003. All patients were seen at five or ten years postoperatively. Radiographs and information regarding body-mass index (<25 kg/m(2) = normal weight, 25 to 29.9 kg/m(2) = overweight, and ≥ 30 kg/m(2) = obese) and activity were obtained. Activity was assessed with use of the University of California at Los Angeles (UCLA) activity scale. Osteolysis was assessed radiographically. Clinical outcome measurements included the Harris hip and Merle d'Aubigne and Postel scores. RESULTS: Our study included 503 arthroplasties in 433 patients; the results of 241 (47.9%) of the arthroplasties were evaluated at five years and the results of 262 (52.1%), at ten years. Osteolysis was identified around forty-four stems, with twenty-four (13.3%) in 181 hips of normal-weight patients, eleven (5.4%) in 205 hips of overweight patients, and nine (7.7%) in 117 hips of obese patients. Normal-weight patients had the highest activity level (mean UCLA activity scale score [and standard deviation], 5.5 + or - 2.0 points), and obese patients had the lowest (mean UCLA activity scale score, 5.0 + or - 1.7 points). When adjusted for activity, cementing quality, and patient age and sex, the risk of osteolysis in obese patients was not increased as compared with that for overweight patients (adjusted odds ratio, 1.4; 95% confidence interval, 0.6 to 3.7), whereas the risk of femoral osteolysis in normal-weight patients was found to be significantly higher than that in overweight patients (adjusted odds ratio, 2.6; 95% confidence interval, 1.2 to 5.7). Clinical outcomes were similar among the groups. CONCLUSIONS: We found no increased risk of osteolysis around a cemented femoral stem in obese patients five and ten years after primary total hip arthroplasty. The highest prevalence of osteolysis was observed in normal-weight patients. LEVEL OF EVIDENCE: Prognostic Level I. See Instructions to Authors for a complete description of levels of evidence.