OBJECTIVE The purpose of the study was to test the hypothesis that stem revision in Vancouver Type B2 and B3 periprosthetic fractures using a transfemoral surgical approach and a modular, noncemented, tapered and fluted revision stem led to reproducibly good results with respect to fracture healing, stability of the prosthesis stem, dislocation, intraoperative fracture, and clinical outcome. METHODS Prospective study. METHODS Orthopaedic specialized clinic and center for joint replacement. METHODS Twenty-two patients with periprosthetic fractures of Vancouver Type B2 and 10 of Type B3 were followed of at least 24 months. METHODS Fracture healing, stability of the prosthesis stem, complications, and clinical outcome. RESULTS All fractures healed with a mean time of 14.5 ± 5.2 weeks. No cases of subsidence of the stem were observed and, according to the classification of Engh et al concerning the biologic fixation of the stem, there was bony ingrowth fixation in 28 cases and stable fibrous fixation in four cases. One dislocation occurred and there were no cases of intraoperative fracture. The Harris hip score rose continually after the operations; from a score 3-months postoperatively of 59.2 ± 14.6 points, it rose to 81.6 ± 16.5 points after 24 months. According to the classification of Beals and Tower, all results were rated as excellent. CONCLUSIONS The methods described here for stem revision lead to reproducibly good results in the treatment of periprosthetic hip fractures of Vancouver Types B2 and B3. METHODS Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.