Predictors of postoperative infection in spinal deformity surgery - which curves are at greatest risk?

Numerous studies have reported on post- operative infection following spinal arthrodesis for spinal deformity. The incidence of infection associated with spinal surgery overall ranges from less than 1% to 15%. Certain factors may increase the risk of surgical site infection (SSI) in patients who undergo surgical treatment for spinal deformities. These include a diagnosis of neuromuscular disease, use of surgical drain or lack thereof, patient comorbidities, increased blood transfusions, and an allergic inflammatory response to the use of instrumentation. Many of these studies, however, have been limited by small numbers and confounded by multiple surgeons and hospitals.

The purpose of this study was to determine the rate of spinal wound infection following spinal deformity surgery in a single surgeon's practice and to identify risk factors for infection, particularly those related to type of deformity and surgical procedures performed.

This is a retrospective review of prospectively collected data from a single surgeon. Only spinal deformity patients' records were reviewed.

We reviewed 941 spinal deformity operative procedures (806 patients, 584 females, 222 males) done by a single surgeon from 1999 to 2009. Outcome measures were superficial wound infection, deep wound infection, possible infection, and no infection. The following variables were assessed: gender, age, height, weight, body mass index (BMI), primary diagnosis, associated comorbidities, prior spinal surgeries, surgical approach, type and number of procedures performed, radiographic measurements, operative data (including operative time), and complications. A one-way ANOVA and Turkey post-hoc analysis was used.

The average age was 22.3 years (range 1.8 to 82 years). There were 13 deep wound infections (1.4%) and 17 incidences of superficial wound infection (1.8%), for an overall infection rate of 3.2%. The following variables were found to be independent predictors of deep infection: neuromuscular scoliosis (odd ratio (O.R. 9.2) and Lenke 3 or 4 curve types in adolescent idiopathic scoliosis patients (O.R. 7.4). Kyphosis (O.R. 4.38), combined approach (O.R. 0.81), total number of levels fused (O.R. 1.16), and Lenke 3 or 4 curve types in AIS (O.R. 4.54) were independent predictors of superficial infection. Lastly, logistic regression identified the following as predictors of any infection: BMI (O.R. 1.11), total number of comorbidities (O.R. 1.24), osteotomy procedure (O.R. 3.14), number of levels fused (O.R. 1.13), and Lenke 3 or 4 curve types (O.R. 1.67).

Age, BMI, number of levels, Lenke 3-4 curves, osteotomy, and number of comorbidities were found to be predictors of any infection. Of these, only Lenke 3-4 and neuromuscular comorbidity was correlated with deep infection.