Abstract
BACKGROUND: To identify, analyze, and report the patient- and procedure-related factors associated with surgical site infection (SSI) after spinal fusion (SF) surgery. METHODS: We included any SSI-SF from January 2013 to September 2015. A total of 989 spine surgeries that required instrumentation were performed. RESULTS: Twenty-four out of 989 (2.43%) patients presented with SSI. More than half of the SSI cases (54%) got infected with either exclusively gram-negative bacteria or a combination of gram-negative and gram-positive bacteria; 9.1% of the surgeries involved the sacral spine (90 out of 989 patients). SSI in long constructs (more than 3 levels) was performed in 66.7% compared with 33.3% with short constructs; 87.5 % of the reported SSI (21 patients) were done through a posterior approach. Of patients who had SSI, 87.5% received prophylactic antibiotics, 92% were operated on during the daytime shift, 50% required blood transfusion, and 79% required surgical debridement. Four patients out of 24 patients died (17%) due to unrelated SSI complications. CONCLUSIONS: The overall incidence of gram-negative infections after long SFs remains low in our study population. Despite this low overall incidence, our results demonstrate a relative higher incidence of gram-negative SSIs in surgeries involving more than 3 spinal levels and for all those involving the sacral spine. We propose that there may be a potential benefit of gram-negative prophylactic antibiotic coverage in patients falling in either 1 of these categories. Further multivariate analysis and/or randomized studies may be necessary to confirm our results. LEVEL OF EVIDENCE: 3.