BACKGROUND: Charcot neuroarthropathy is a morbid and expensive complication of diabetes that can lead to lower extremity amputation. Current treatment of unstable midfoot deformity includes lifetime limb bracing, primary transtibial amputation, or surgical reconstruction of the deformity. In the absence of a widely adopted treatment algorithm, the decision to pursue more costly attempts at reconstruction in the United States continues to be driven by surgeon preference. QUESTIONS/PURPOSES: To examine the cost effectiveness (defined by lifetime costs, quality-adjusted life-years [QALYs] and incremental cost-effectiveness ratio [ICER]) of surgical reconstruction and its alternatives (primary transtibial amputation and lifetime bracing) for adults with diabetes and unstable midfoot Charcot neuroarthropathy using previously published cost data. METHODS: A Markov model was used to compare Charcot reconstruction and its alternatives in three progressively worsening clinical scenarios: no foot ulcer, uncomplicated (or uninfected) ulcer, and infected ulcer. Our base case scenario was a 50-year-old adult with diabetes and unstable midfoot deformity. Patients were placed into health states based on their disease stage. Transitions between health states occurred annually using probabilities estimated from the evidence obtained after systematic review. The time horizon was 50 cycles. Data regarding costs were obtained from a systematic review. Costs were converted to 2019 USD using the Consumer Price Index. The primary outcomes included the long-term costs and QALYs, which were combined to form ICERs. Willingness-to-pay was set at USD 100,000/QALY. Multiple sensitivity analyses and probabilistic analyses were performed to measure model uncertainty. RESULTS: The most effective strategy for patients without foot ulcers was Charcot reconstruction, which resulted in an additional 1.63 QALYs gained and an ICER of USD 14,340 per QALY gained compared with lifetime bracing. Reconstruction was also the most effective strategy for patients with uninfected foot ulcers, resulting in an additional 1.04 QALYs gained, and an ICER of USD 26,220 per QALY gained compared with bracing. On the other hand, bracing was cost effective in all scenarios and was the only cost-effective strategy for patents with infected foot ulcers; it resulted in 6.32 QALYs gained and an ICER of USD 15,010 per QALY gained compared with transtibial amputation. As unstable midfoot Charcot neuroarthropathy progressed to deep infection, reconstruction lost its value (ICER USD 193,240 per QALY gained) compared with bracing. This was driven by the increasing costs associated with staged surgeries, combined with a higher frequency of complications and shorter patient life expectancies in the infected ulcer cohort. The findings in the no ulcer and uncomplicated ulcer cohorts were both unchanged after multiple sensitivity analyses; however, threshold effects were identified in the infected ulcer cohort during the sensitivity analysis. When the cost of surgery dropped below USD 40,000 or the frequency of postoperative complications dropped below 50%, surgical reconstruction became cost effective. CONCLUSIONS: Surgeons aiming to offer both clinically effective and cost-effective care would do well to discuss surgical reconstruction early with patients who have unstable midfoot Charcot neuroarthropathy, and they should favor lifetime bracing only after deep infection develops. Future clinical studies should focus on methods of minimizing surgical complications and/or reducing operative costs in patients with infected foot ulcers. LEVEL OF EVIDENCE: Level II, economic and decision analysis.
ASJC Scopus subject areas
- Orthopedics and Sports Medicine