A prospective, randomized study was performed in 559 patients to compare two doses of oral cefadroxil with three doses of intravenous cefuorxime as antibiotic prophylaxis in intra-and subtrochanteric hip fracture surgery. Antibiotic concentrations in the wound fluid were determined at the start and at the end of the operation. The first dose of cefadroxil was given about 2 h before surgery and cefuroxime about 30 min before operation. In 226/ 242 (93%) patients randomized to oral cefadroxil, the concentration in the wound during surgery was on average 15 ~tg/ml, i.e., well above the minimum inhibitory concentration (MIC-90) for Staphylococcus aureus. In the cefuroxime group, antibiotic levels in the wound exceeded the MIC-90 for S. aureus in 204/210 (97%) of the patients at the start and/or at the end of surgery. All patients were followed up for 4 months. One deep and five superficial infections occurred in the cefuroxime group and no deep but one superficial infection in the cefadroxil group (P = 0.07). S. aureus was cultured in three of the infected cases while cultures were negative in four patients. Four of the seven infected patients had adequate levels of antibiotic in the wound during surgery, and in three patients no antibiotic assay was performed. The infected patients did not