Surgical site infections: reanalysis of risk factors

Abstract

Background: Surgical site infections (SSI) are the most common nosocomial infection in surgical patients, accounting for 38% of all such infections, and are a significant source of postoperative morbidity resulting in increased hospital length of stay and increased cost. During 1986-1996 the Center for Disease Control and Prevention's National Nosocomial Infections Surveillance system reported 15,523 SSI following 593,344 operations (2.6%). Previous studies have documented patient characteristics associated with an increased risk of SSI, including diabetes, tobacco or steroid use, obesity, malnutrition, and perioperative blood transfusion. In this study we sought to reevaluate risk factors for SSI in a large cohort of noncardiac surgical patients.

Methods: Prospective data (NSQIP) were collected on 5031 noncardiac surgical patients at the Veteran's Administration Maryland Healthcare System from 1995 to 2000. All preoperative risk factors were evaluated as independent predictors of surgical site infection.

Results: The mean age of the study cohort was 61 plus minus 13. SSI occurred in 162 patients, comprising 3.2% of the study cohort. Gram-positive organisms were the most common bacterial etiology. Multiple logistic regression analysis documented that diabetes (insulin- and non-insulin-dependent), low postoperative hematocrit, weight loss (within 6 months), and ascites were significantly associated with increased SSI. Tobacco use, steroid use, and chronic obstructive pulmonary disease (COPD) were not predictors for SSI.

Conclusion: This study confirms that diabetes and malnutrition (defined as significant weight loss 6 months prior to surgery) are significant preoperative risk factors for SSI. Postoperative anemia is a significant risk factor for SSI. In contrast to prior analyses, this study has documented that tobacco use, steroid use, and COPD are not independent predictors of SSI. Future SSI studies should target early preoperative intervention and optimization of patients with diabetes and malnutrition.