Cost-Effectiveness Analysis of Syndesmotic Screw Versus Suture Button Fixation in Tibiofibular Syndesmotic Injuries.

Abstract

To evaluate the cost effectiveness of suture buttons compared with syndesmotic screws for repair of tibiofibular syndesmotic injuries. A decision tree model was constructed to describe outcomes after syndesmosis repair using suture buttons and syndesmotic screws from the perspective of a capitated health care system. Outcomes were uneventful healing, removal of symptomatic implants, deep infection, and persistent diastasis requiring revision. Weighted literature averages were used to estimate variables for a baseline model. Outcomes were measured in quality adjusted life years. Procedure and implant costs were derived from Medicare reimbursement rates and the University Health System Consortium. An incremental cost-effectiveness ratio threshold of $50,000 per quality-adjusted life years was used to evaluate cost effectiveness. The baseline model did not identify suture buttons to be cost effective. Sensitivity analysis demonstrates the model to be exquisitely sensitive to small changes in reoperation rates and implant price. At median University Health System Consortium implant prices, if the removal rate for symptomatic screws is below 13.7%, then screws are cost effective. If the screw removal rate is greater than 17.5%, then a suture button is cost effective. Within this interval, detailed analysis of the model suggests that screws may be the cost-effective strategy, but that determination should be taken with caution. Moving away from the practice of routinely removing all syndesmotic screws has changed the financial landscape of syndesmosis repair. At their median cost, suture buttons are likely to be cost effective over screws for symptomatic screw removal rates greater than 17.5%. Cost effectiveness is sensitive to changes in implant removal rates and the number of devices used per patient. Economic Level IV. See Instructions for Authors for a complete description of levels of evidence.