Factors Associated with Surgical Site Infection after Lower Extremity Bypass in the SVS Vascular Quality Initiative (VQI) – a National Perspective
Jeffrey Kalish1, Alik Farber1, Karen Homa2, Magdiel Trinidad-Hernandez3, Adam Beck4, Mark Wyers5, Mark Davies6, Jason Chiriano7, Margaret Tracci8, Larry Kraiss9, Jack Cronenwett10 on behalf of the SVS PSO Quality Committee
1Boston Medical Center, Boston, MA, United States; 2SVS PSO, Chicago, IL, United States; 3University of Arizona Medical Center, Tucson, AZ, United States; 4University of Florida College of Medicine, Gainesville, FL, United States; 5Beth Israel Deaconess Medical Center, Boston, MA, United States; 6Methodist Debakey Heart and Vascular Center, Houston, TX, United States; 7Loma Linda University Medical Center, Loma Linda, CA, United States; 8University of Virginia Medical Center, Charlottesville, VA, United States; 9University of Utah Hospital, Salt Lake City, UT, United States; 10Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States.
OBJECTIVES: Surgical site infection (SSI) is a major source of morbidity after infrainguinal lower extremity bypass (LEB). This study identified process of care variables associated with SSI after LEB that could be modified to improve outcomes.
METHODS: The VQI database (2003-2012) was queried to identify in-hospital SSI after 7908 consecutive LEB procedures performed by 365 surgeons at 91 academic and community hospitals in 45 states. Variables associated with SSI were identified using multivariable logistic regression and hierarchical clustering. Expected and observed SSI rates were calculated for each hospital.
RESULTS: The overall in-hospital SSI rate after LEB was 4.8%. By univariate analysis, obesity, dialysis, tissue loss, ABI<0.35, distal target, vein graft, continuous vein harvest incision, transfusion>2units PRBC, procedure time>220 min, and EBL>100mL had higher SSI rates, while chlorhexidine (compared to iodine) skin prep was protective. By multivariate analysis, independent predictors of SSI included ABI<0.35 (OR 1.53, 95%CI 1.03-2.30, p<0.04), transfusion>2 units (OR 3.30, 95%CI 2.17-5.02, p<0.001), and procedure time>220 minutes (OR 2.11, 95%CI 1.05- 4.23, p<0.04). Chlorhexidine was protective against SSI (OR 0.53, 95% CI 0.35-0.79, p=0.002). Stratified analyses based on the presence of tissue loss yielded similar results. Across VQI hospitals, observed SSI rates ranged from 0-30%, while expected SSI rates adjusted by the four predictors ranged from 0-8%.
CONCLUSIONS: In-hospital SSI after LEB varies substantially across VQI hospitals. Three modifiable processes of care (transfusion rate, procedure time, and type of skin preparation) were identified and may be used by hospitals to reduce SSI rates. This study demonstrates the value of the SVS VQI detailed shared clinical registry to identify improvement opportunities directly pertinent to providers that are not available in typical administrative datasets.
AUTHOR DISCLOSURES: A. Beck, No financial disclosure; J. Chiriano, No financial disclosure; J. Cronenwett, No financial disclosure; M. Davies, No financial disclosure; A. Farber, No financial disclosure; K. Homa, No financial disclosure; J. Kalish, No financial disclosure; L. Kraiss, No financial disclosure; M. Tracci, No financial disclosure; M. Trinidad-Hernandez, No financial disclosure; M. Wyers, No financial disclosure
Updated May 2013