Infection After Surgery: Incidence, Bacteriology and Trends from 2008-2015
*William O'Brien, *Kalpana Gupta, Kamal M.F. Itani
VA Boston Healthcare System, Boston, MA
Objective: Infection after surgery is a serious complication associated with excess morbidity and cost. Staphylococcus aureus accounts for a large proportion of these infections. Our objectives are to assess S. aureus infection incidence one year postoperatively, identify risk factors for infection, and calculate the association between multiple site infections.
Design: Retrospective database study of non-cardiac surgeries in the Veterans Health Administration. Logistic regressions estimated factors associated with 1-year infection.
Setting: VHA medical centers.
Patients: Veterans undergoing major surgery (n=740,419) during 2008-2015>.
Main Outcome Measures: Incidence of S. aureus infection (MRSA or MSSA), as well as other organisms 1-year postoperatively. 30-day infection was defined using the VASQIP chart review and a positive culture. Infections beyond 30 days were defined as a positive culture combined with antibiotic treatment.
Results: S. aureus infection incidence was 2.3%. MRSA incidence was 0.9%, MSSA 1.5%, Staph. non-aureus 1.2%, and other bacteria 6.0%. Rates decreased from 2008 to 2015 (S. aureus 3.0% to 2.0%; MRSA 1.0% to 0.8%; MSSA 2.1% to 1.3%; other Staph. 1.6% to 1.0%; other organism 6.8% to 5.6%). Factors associated with increased infection risk were ASA class >2, diabetes, COPD, smoking, male sex, and surgery RVU > 10. Incidence of 30-day urinary tract infection with any bacteria was 3.4% among those who also had a surgical site infection (SSI), versus 0.8% among those who had no SSI; bloodstream infections were 7.6% v. 0.7%; and pneumonia 3.1% v. 0.3% (all p<0.001).
Conclusions: Postoperative S. aureus infection has decreased over time. MRSA incidence decreased relatively less than MSSA. S. aureus infection is associated with patient characteristics and surgery complexity. SSIs are associated with higher risk of other-site infections.