Multivariable predictors of postoperative surgical site infection after general and vascular surgery: results from the patient safety in surgery study

Abstract

Background: Surgical site infection (SSI) is a potentially preventable complication. We developed and tested a model to predict patients at high risk for surgical site infection.

Study design: Data from the Patient Safety in Surgery Study/National Surgical Quality Improvement Program from a 3-year period were used to develop and test a predictive model of SSI using logistic regression analyses.

Results: From October 2001 through September 2004, 7,035 of 163,624 (4.30%) patients undergoing vascular and general surgical procedures at 14 academic and 128 Department of Veterans Affairs (VA) medical centers experienced SSI. Fourteen variables independently associated with increased risk of SSI included patient factors (age greater than 40 years, diabetes, dyspnea, use of steroids, alcoholism, smoking, recent radiotherapy, and American Society of Anesthesiologists class 2 or higher), preoperative laboratory values (albumin<3.5 mg/dL, total bilirubin>1.0 mg/dL), and operative characteristics (emergency, complexity [work relative value units>/=10], type of procedure, and wound classification). The SSI risk score is more accurate than the National Nosocomial Infection Surveillance score in predicting SSI (c-indices 0.70, 0.62, respectively).

Conclusions: We developed and tested an accurate prediction score for SSI. Clinicians can use this score to predict their patient's risk of an SSI and implement appropriate prevention strategies.