Key points are not available for this paper at this time.
Background: Total ankle replacements (TARs) have rapidly advanced in terms of volume, technique, design, and indications. However, TARs are still at risk for early mechanical failure and revision. Prior studies have investigated potential risk factors for failure, but have been limited to smaller series or older implants. This study sought to identify risk factors for early mechanical failure in modern TAR. Methods: This is a retrospective study of a single-institution registry. Five surgeons contributed cases involving patients who underwent a primary TAR with any implant. Implants were grouped on the basis of the type of fixation. The primary outcome was early mechanical failure (revision with component removal for a non-infectious etiology, that is, subsidence, aseptic loosening, and/or malalignment). Logistic regression determined the effects of age, weight, hindfoot arthrodesis, implant type, and radiographic deformity on failure. Results: The 731 included patients had a mean follow-up of 2.7 years. Ten percent (71 patients) had hindfoot arthrodesis. There were 33 mechanical failures (4.5%) at a mean of 1.7 years after the index surgical procedure. Our model demonstrated that hindfoot arthrodesis was associated with 2.7 times greater odds of failure (p = 0.045), every 10 kg of body weight increased the odds of tibial-sided failure by 1.29 times (p = 0.039), and implants with more extensive tibial fixation (stems or keels) lowered the odds of tibial failure by 95% (p = 0.031). Conclusions: In patients with uncontrollable risk factors (hindfoot arthrodesis) or risk factors that may or may not be modifiable by the patient (weight), implants with more robust tibial fixation may be able to reduce the risk of early mechanical failure. Further research is warranted to support efforts to decrease early failure in TAR. Level of Evidence: Therapeutic Level III . See Instructions for Authors for a complete description of levels of evidence.
Henry et al. (Tue,) studied this question.