Investigation of Surgical Site Infection Outbreak Among Neurosurgical Patients

Abstract

Background: The infection control service of a private hospital in Belo Horizonte, Brazil, performs continuous surveillance of surgical patients according to the CDC NHSN protocols. In a routine analysis of the neurosurgical service, we identified a subtle increase in the incidence of surgical site infection (SSI): in 5 months (June–October 2018), 6 patients developed an SSI. From January 2017 until May 2018, there were no cases of infection in neurosurgery, which led us to suspect an outbreak. Methods: A cohort study was used to investigate the factors associated with risk of SSI. We investigated the following variables: ASA score, number of hospital admissions, age, preoperative hospital length of stay, duration of surgery, wound class, general anesthesia, emergency, trauma, prosthesis, surgical procedures, surgeon. Furthermore, 9 key steps were followed to investigate the outbreak: case definition (step 1), search for new SSI cases (step 2); confirmation of the outbreak (step 3); analysis of SSI cases by London Protocol (step 4); analysis of the cohort data (step 5); inspections in the surgical ward (step 6); qualitative and quantitative reports sent to the neurosurgical departments (step 7); continuing with active surveillance (stage 8); announcement of research findings (step 9). Results: The outbreak was confirmed: SSI incidence in the pre-epidemic period (January–May 2018) was 0 of 218 (0%); in the epidemic period (June–October 2018), SSI incidence was 6 of 94 (6.4%) (P < .001). We identified 3 SSI etiologic agents: 2 Klebsiella pneumoniae, 2 S. aureus, and 1 Serratia marcescens. It was unlikely that there was a common source for the outbreak. We identified the following risk factors: second or third hospital admissions (RR, 3.7; P = .041), and preoperative hospital length of stay: SSI patients (4.3±5.7 days) versus control patients (0.7 ± 2.1 days) (P = .048). None of the surgeons presented an SSI rate significantly different from each other. We used the London protocol to identify antibiotic prophylaxis failures in most cases. Conclusions: New cases of infections can be prevented if the length of preoperative hospital stay becomes as short as possible and, most importantly, if antibiotic prophylaxis does not fail.

Funding: None

Disclosures: None