Abstract
BACKGROUND: Previous studies stratified postoperative infection risk by patient comorbidities. However, it is unclear whether the incidence varies by surgical approach in a specialized orthopaedic setting. This study aims to compare infection rates and microbiologic characteristics of postoperative spine infections requiring return to the operating room for debridement by hospital setting: a dedicated orthopaedic and spine hospital versus a general hospital serving multiple surgical specialties. METHODS: The study is a retrospective review of prospectively collected data. Procedures performed between March 2006 and August 2008 at the multispecialty university hospital were compared with cases at an orthopaedic specialty hospital from September 2008 through August 2016. The surgeons, residents, and patients were similar, but the operative venue changed in 2008. RESULTS: The overall general university hospital infection rate was 2.03%, higher than the overall infection rate at the dedicated orthopaedic and spine hospital of 1.31% (P < .0104). The general university infection rate was 2.27% in the final years of practice, compared with 0.91% at the dedicated orthopaedic and spine hospital (P < .0001) during a recent 2-year time frame. Demographic variables did not significantly differ between the 2 settings. The overall proportion of Gram-negative infection rates was not statistically different (21.7% vs 18.6%), despite an increased proportion of Gram-negative infections at the general university hospital following surgery from an anterior approach. Most of the organisms isolated in both facilities were Staphylococcus species. There was no difference in the seasonality of postoperative spine infections in either setting. CONCLUSIONS: In transitioning from a multispecialty university hospital to a dedicated orthopaedic hospital, the incidence of postoperative spine infections was significantly reduced to 0.91%. Despite the change in venue, the proportion of Gram-negative infections (∼20%) following spine surgery did not significantly change. These results suggest improved infection rates during the course of the last 10 years with consistent proportions of Gram-negative infections. LEVEL OF EVIDENCE: 3.