Impact of surgical site infections on length of stay and costs in selected colorectal procedures.

Background: Length of stay (LOS) and inpatient costs for open-abdomen colorectal procedures have not been examined recently. The aim of this study was to determine LOS and costs for several colorectal procedures in the context of factors potentially associated with surgical site infection (SSI).

Methods: We used a large U.S. hospital database to identify the variables associated with longer LOS and higher costs for colorectal procedures from January 1, 2005, through June 30, 2006. The study population consisted of all patients >18 years, identified via International Classification of Disease, Ninth Revision, procedural codes for elective colorectal surgery. Patient demographics, surgical procedure, and a modified Study of the Efficacy of Nosocomial Infection Control (SENIC) infection risk score were examined using logistic regression as predictors of LOS >or=1 week and cost >or=$15,000. Patients given cefotetan as surgical prophylaxis were compared with patients given cefazolin/metronidazole. Superficial and deep SSIs were considered; intra-abdominal infection was not.

Results: The 25,825 patients were of average age 63 years, with 53% being female and 75% being Caucasian. The overall infection rate was 3.7%. The mean LOS was 7.25 days, and the mean +/- standard deviation total cost per patient $13,746 +/- $13,330. Rates of infection, LOS, and mean hospital costs were all greater for patients with a high SENIC score and increasing disease acuity. Values for these outcome variables were highest for procedures involving stoma formation, followed by operations on the small bowel and large bowel. Variables independently predictive of longer LOS were SSI (odds ratio [OR] 11.74; 95% confidence interval [CI] 9.67, 14.26), age >or=65 years (OR 1.90; 95% CI 1.81, 2.01), and high SENIC score (OR 1.79; 95% CI 1.67, 1.92), whereas Caucasian race (OR 0.86; 95% CI 0.81, 0.91) was predictive of a shorter LOS. Cefazolin/metronidazole was not predictive of a shorter LOS compared with cefotetan (OR 1.06; 95% CI 0.96, 1.17) but was associated with significantly more hospitalizations with costs >or=$15,000 (OR 1.39; 95% CI 1.23, 1.56).

Conclusions: Length of stay and cost rise proportionally with SENIC score, disease acuity, and patient characteristics such as age. Surgical site infections are significantly and independently associated with LOS and cost and contribute to inpatient morbidity and expense. Cefotetan has limited availability, and substitutions are utilized increasingly. Although equally efficacious in elective colon procedures, cefotetan used as surgical prophylaxis was associated with lower hospitalization costs than cefazolin plus metronidazole.