Introduction - Surgical site infections (SSIs) are among the healthcare-associated infections targeted by the national control program. Since 1999, a database has been established based on interregional surveillance networks as part of the National Network for Alert, Investigation, and Surveillance of Healthcare-Associated Infections (Raisin). Methods - Each year, every participating surgical department in the surveillance network was required to include 200 surgical patients and collect individual data, including in particular the components of the NNIS risk index (operative time, ASA class, contamination class) and other perioperative factors. All included patients were to be followed up, if possible, through the 30th postoperative day. Surgical site infections (SSIs) were defined according to standard criteria. Results - Among 620,176 surgical procedures monitored between 1999 and 2004, 10,349 SSIs were identified (crude incidence rate: 1.7%), of which 42% were deep. The incidence of SIS varied with the NNIS risk index, ranging from 0.9% for low-risk patients to 14% for the highest-risk patients. For the most frequently monitored surgical procedures, the incidence of SIS ranged from 0.2% to 9.2% depending on the procedure. The urgency of the procedure and video-assisted surgery were risk factors in addition to the NNIS index for certain specific procedures. Over the study period, the incidence of SIS in NNIS-0 decreased for certain procedures such as abdominal wall hernias (-50%). Conclusion - The SSI surveillance system based on networks of voluntary surgical departments is a useful tool for estimating the risk of infection in surgical patients and provides relevant data for evaluating the impact of the national policy to combat nosocomial infections. (R.A.)