Comparison of Surveillance of Surgical Site Infections by a National Surveillance Program and by Institutional Audit.

BACKGROUND Reported incidence of surgical site infections (SSI) after colonic surgery varies widely. These variations depend not only on patient- and surgery-related parameters but are influenced by type and quality of follow-up. The aim of the study was to compare SSI assessed by two independent prospective surveillance systems, a national surveillance program based on recommendations of the National Healthcare Safety Network (Swissnoso) versus an international audit system, the ERAS Interactive Audit System (EIAS; Encare, Stockholm, Sweden). METHODS Comparative study of a consecutive cohort of colonic resections at a single institution from September 2015 to March 2017. Independent prospective SSI monitoring was available from Swissnoso and EIAS. Inter-observer reliability was calculated using Cohen k. Sensitivity, specificity, and accuracy of EIAS in assessing SSI was compared with Swissnoso, considered as gold standard. RESULTS The final sample included 143 patients. Of these, 136 (95.1%) were classified into the same category by both systems, identifying 17 patients (12.5%) with SSI and 119 patients (87.5%) without SSI, respectively. Discrepant results were found for the remaining seven patients (4.9%) with four SSI categorization according to Swissnoso but not EIAS, and three SSI categorization in EIAS but not in Swissnoso; all miscategorized patients presented superficial SSI. Sensitivity, specificity, and accuracy of EIAS for SSI recording was 81%, 97.5%, and 95.1%, respectively. Inter-observer agreement was high (Cohen k value of 0.801, p < 0.001). Case-by-case analysis of discrepant findings revealed mainly discrepant interpretation of clinical symptoms and erroneous labeling of non-procedure-related infections. CONCLUSIONS Surgical site infection recording by two independent systems showed high concordance and good inter-rater reliability.