Variability of Surgical Site Infection with VEPTR at Eight Centers: A Retrospective Cohort Analysis

Study Design Retrospective review. Objectives To describe clinical characteristics and infection rates in modern vertical expandable prosthetic titanium rib (VEPTR) surgery. Summary of Background Data Prior studies have demonstrated infection rates from 10% to 30% with VEPTR surgery. Methods A retrospective query was done on an institutional review board-approved, multicenter prospectively collected database for patients implanted with VEPTR from 2007 to 2013 at eight sites. This identified 213 patients with appropriate data for analysis. Average follow-up was 4.1 years (range 1.7-6.3). Data collected included a Classification of Early-Onset Scoliosis (C-EOS) diagnosis, American Society of Anesthesiologists Physical Status (ASA-PS), major Cobb angle, construct type, clinical symptoms, and microbiology. The distribution of infection rates across all the study sites was compared. The exact p value was estimated by Monte Carlo simulation. Results Overall, 18% (38/213) of patients implanted with VEPTR developed infection requiring operative debridement. There were significantly different infection rates among the sites, ranging from 2.9% to 42.9% (p =.029). The average time to infection was 70 days (range 8-236) after the infecting procedure. The majority of infections were due to gram-positive bacteria (80%, 44/55), the most prevalent being methicillin-sensitive Staphylococcus aureus (45%, 25/55). There were 20 patients (53%, 20/38) with either partial or complete implant removal to resolve infection; however, only 3 of 38 (8%) of these resulted in abandonment of VEPTR treatment. There was no difference in infection rate across the primary C-EOS diagnosis categories (p =.21) or based on ASA score (p =.53). After controlling for study site, the odds ratio of an infection following an implant procedure versus an expansion was 2.8 (p =.002). There was no difference in the odds ratio of an infection between the other procedure types (implant, expansion, exchange/revision). Conclusions There were significant differences in infection rates between sites. The variability in infection rate indicates a need for guided efforts to standardize best practices for infection control in VEPTR surgery. Level of Evidence III, therapeutic study.