Outcome after total ankle replacement or ankle arthrodesis in end-stage ankle osteoarthritis on the basis of german-wide data: a retrospective comparative study over 10 years

Abstract

Background In symptomatic end-stage osteoarthritis of the ankle joint, total ankle replacement and ankle arthrodesis are the two primary surgical options for patients for whom conservative treatment fails. Published revision rates are often biased and difficult to compare. In this study, unplanned reoperation rates and revision rates were determined for both surgical interventions based on a large dataset, and risk factors for unplanned reoperations were identified. Methods German-wide health data of the largest German health-care insurance carrier between 2001 and 2012 were retrospectively analyzed, and unplanned reoperation rates within 10 years were determined for index surgeries conducted in 2001 and 2002. Unplanned reoperation rates within 5 years for index surgeries conducted in 2001/2002 were compared to index surgeries conducted in 2006/2007. Multivariate logistic regression was used to identify risk factors for unplanned reoperations. Results After ankle arthrodesis, 19% (95% confidence interval [CI], 16–22%) of 741 patients needed to undergo an unplanned reoperation within ten years. After total ankle replacement, the unplanned reoperation rate was 38% [95% CI, 29–48%] among 172 patients. For initial surgeries conducted at a later date, unplanned reoperation rates within five years were 21% [95% CI, 19–24%] for 1,168 ankle arthrodesis patients and 23% [95% CI, 19–28%] for 561 total ankle replacement patients. Significant risk factors for unplanned reoperations after ankle arthrodesis in the initial cohort were age < 50 years (odds ratio [OR] = 4.65 [95% CI 1.10;19.56]) and osteoporosis (OR = 3.72 [95% CI, 1.06;13.11]); after total ankle replacement, they were osteoporosis (OR = 2.96 [95% CI, 1.65;5.31]), Patient Clinical Complexity Level (PCCL) grade 3 (OR = 2.19 [95% CI, 1.19;4.03]), PCCL grade 4 (OR = 2.51 [95% CI, 1.22;5.17]) and diabetes mellitus (OR = 2.48 [95% CI, 1.33;4.66]). Kaplan-Meier analyses including 1,525 ankle arthrodesis patients and 644 total ankle replacement patients revealed an average unplanned reoperation-free time of approximately 17 years for both procedures. Conclusions Similar revision rates and unplanned reoperation rates for both procedures in the later-date cohort can likely be attributed to a learning curve for surgeons as well as advances in implant design. This analysis of billing health insurance data supports an increase in total ankle replacement surgeries.