• Amir Abdul-Jabbar, Steven Takemoto, Michael H Weber, Serena S Hu, Praveen V Mummaneni, Vedat Deviren, Christopher P Ames, Dean Chou, Philip R Weinstein, Shane Burch, and Sigurd H Berven.
• Department of Orthopaedic Surgery, University of California, San Francisco, CA 94143, USA. amir.abdul-jabbar@ucsf.edu
• Spine. 2012 Jul 1;37(15):1340-5.

Study DesignRetrospective analysis.ObjectiveThe objective of this study was to investigate the accuracy of using an automated approach to administrative claims data to assess the rate and risk factors for surgical site infection (SSI) in spinal procedures.Summary Of Background DataSSI is a major indicator of health care quality. A wide range of SSI rates have been proposed in the literature depending on clinical setting and procedure type.MethodsAll spinal surgeries performed at a university-affiliated tertiary-care center from July 2005 to December 2010 were identified using diagnosis-related group, current procedural terminology, and International Classification of Diseases, Ninth Revision (ICD-9) codes and were validated through chart review. Rates of SSI and associated risk factors were calculated using univariate regression analysis. Odds ratios were calculated through multivariate logistic regression.ResultsA total of 6628 hospital visits were identified. The cumulative incidence of SSI was 2.9%. Procedural risk factors associated with a statistically significant increase in rates of infection were the following: sacral involvement (9.6%), fusions greater than 7 levels (7.8%), fusions greater than 12 levels (10.4%), cases with an osteotomy (6.5%), operative time longer than 5 hours (5.1%), transfusions of red blood cells (5.0%), serum (7.4%), and autologous blood (4.1%). Patient-based risk factors included anemia (4.3%), diabetes mellitus (4.2%), coronary artery disease (4.7%), diagnosis of coagulopathy (7.8%), and bone or connective tissue neoplasm (5.0%).ConclusionUsed individually, diagnosis-related group, current procedural terminology, and ICD-9 codes cannot completely capture a patient population. Using an algorithm combining all 3 coding systems to generate both inclusion and exclusion criteria, we were able to analyze a specific population of spinal surgery patients within a high-volume medical center. Within that group, risk factors found to increase infection rates were isolated and can serve to focus hospital-wide efforts to decrease surgery-related morbidity and improve patient outcomes.

36 keywords...

hide…