Surgical-site infection in spinal injury: incidence and risk factors in a prospective cohort of 518 patients.

Abstract

To investigate the incidence of surgical-site infection (SSI) and determinate the risk factors of SSI in the context of spinal injury. From February 1, 2011 to July 31, 2011, for a multicentre cohort of patients with acute spinal injury, we prospectively censored those with SSI for at least 12 months. We recorded epidemiologic characteristics and details of surgical procedure and postoperative care for each patient. We calculated the incidence of SSI at 1, 3 and 12 months after surgery. Univariate and multivariate analysis were used to establish the association of risk factors and SSI. We studied clinical outcomes by a visual analog scale for pain and physical and mental component summaries (PCS and MCS) of the Medical Outcomes Survey 36-Item Short Form (SF-36). At 1 year, among 518 patients, we recorded 25 SSI events, with median occurrence at 16 days (25-75 % quartile: 13-44 days). Incidence of SSI was 3.2 % (95 % confidence interval [1.9-5.3 %]) at 1 month, 3.7 % (95 % [2.2-5.8 %]) at 3 months and 4.6 % (95 % CI [3-6.9 %]) at 12 months. On multivariate analysis, age, presence of diabetes and surgical duration were predictors of SSI (p = 0.009, p = 0.047, and p = 0.015 respectively). At 12 months, infected and non-infected patients did not differ in pain (p = 0.58) or SF-36 PCS (p = 0.8) or MCS (p = 0.68). In this large prospective multicentre study in the context of spinal injury, we obtained an equivalent incidence rate and risk factors of SSI as found in the literature for elective spinal surgery.