Background: A prospective observational study was undertaken in 2,481 patients undergoing elective colon resection in 114 German centers to identify optimal drug and dosing modalities and risk factors for postoperative infection. Methods: Patients were pair matched using six risk factors and divided into 672 pairs (ceftriaxone vs, other cephalosporins, group A) and 400 pairs (ceftriaxone vs. penicillins, group B). End points were local and systemic postoperative infection and cost effectiveness. Results: Local infection rates were 6.0 versus 6.5% (group A) and 4.0 versus 10.5% (group B); systemic infection rates in groups A and B were 4.9 versus 6.3% and 3.3 versus 10.5%, respectively. Ceftriaxone was more effective than penicillins overall (6.8 vs. 17.8%, p < 0.001). Length of postoperative hospital stay was 16.2 versus 16.9 days (group A) and 15.8 versus 17.6 days (group B). Of the six risk factors, age and concomitant disease were significant for systemic infection, and blood loss, rectum resection and immunosuppressive therapy were significant for local infection. Penicillin was a risk factor compared to ceftriaxone (p < 0.0001). Ceftriaxone saved Q160.7 versus other cephalosporins and O416.2 versus penicillins. Conclusion: Clinical and microbiological efficacy are responsible for the cost effectiveness of ceftriaxone for perioperative prophylaxis in colorectal surgery. Copyright (C) 2000 S. Karger AG, Basel.