The influence of surveillance methods on surgical wound infection rates in a tertiary care spinal surgery service.
STUDY DESIGN: A 1-year prospective study of the influence of surveillance methods on the surgical wound infection rates in a tertiary care spinal surgery unit. OBJECTIVES: To assess the effect of postdischarge surveillance, the diagnostic indications for surgery, and the type of procedure on the surgical wound infection rates. SUMMARY OF BACKGROUND DATA: Use of the National Nosocomial Infection Surveillance system for surgical wound infection resulted in infection rates above the published values for procedures performed by the Spinal Surgical Service. A preliminary review failed to find causes for these higher rates, and a study was undertaken to assess the influence of the surveillance methods used. METHODS: Patient information collected by the Spinal Surgical Service and surveillance data obtained by infection control were combined in a relational database. Surveillance after discharge was performed by regularly sending questionnaires to surgeons' offices. The diagnostic indications were assessed by dividing all patients into three groups: Class D (disc disease or spinal stenosis). Class T (spinal trauma within 60 days), and Class M (mostly complex spinal procedures for deformity and instability). Infection rates for the three diagnostic indication classes and for procedures with and without instrumentation were calculated. RESULTS: Postdischarge surveillance significantly increased the infection rates, mostly by detecting superficial infections that did not require readmission. Significant differences were noted between Class T and Class M patients undergoing lumbar posterior segmental instrumentation, despite the fact that they had a similar incidence of risk factors according to the National Nosocomial Infection Surveillance system. The surgical wound infection rates of the National Nosocomial Infection Surveillance system may not be appropriate standards for specialized units with a high incidence of complex clinical problems and complicated surgical procedures. CONCLUSIONS: Postdischarge surveillance, surgical procedure classification methods, and the indications for surgery (e.g., trauma, congenital deformity) influence the surgical wound infection rate. Current adjustments for some of these factors in the National Nosocomial Infection Surveillance system appear to be inadequate when used in a tertiary care facility.