Ankle Valgus after Hindfoot Arthrodesis: A Radiographic and Chart Comparison of the Medial Double and Triple Arthrodeses

The medial double arthrodesis, comprised of subtalar and talonavicular joint fusions, has become a popular way to correct hindfoot deformity. There is potential concern for an increase in ankle valgus postoperatively owing to extended medial dissection and possible disruption of the deltoid ligament. Although this approach is often used to correct a valgus hindfoot, little attention has been paid to date on this procedure's effect on the tibiotalar joint. Although the medial double arthrodesis has been shown to produce reproducible outcomes without violating the lateral hindfoot structures, our hypothesis was that this approach would increase the ankle valgus deformity compared with its triple counterpart. The primary goal of the present retrospective study was to identify the frequency and severity of ankle valgus after the medial double arthrodesis compared with the triple arthrodesis. A total of 77 patients (78 feet) met our inclusion criteria. Their mean age was 61.3 ± 10.7 (range 27 to 79) years, and the follow-up period was 15.7 ± 9.9 (range 6 to 46) months. There were 16 and 61 patients (62 feet) in the medial double and triple arthrodesis groups, respectively. Overall, the preoperative ankle valgus was 1.24° ± 2.02° (range 0° to 6°), and there was no statistical difference of preoperative ankle valgus noted between groups (p = .060). Collectively, postoperative ankle valgus was 3.01° ± 3.54° (0° to 17°) with an increase in ankle valgus in 4 of 16 medial double and 34 of 62 triple arthrodesis patients. With a mean follow-up of 8.75 ± 4.02 (6 to 21) months, the medial double arthrodesis cohort's ankle valgus increased from 0.5° ± 1.55° (0° to 6°) to 1.5° ± 3.14° (1° to 10°) postoperatively. The triple arthrodesis group had a mean follow-up 17.53 ± 10.17 (6 to 46) months and ankle valgus increased from 1.44° ± 2.09° (0 to 6°) to 3.40° ± 3.56° (0° to 17°). Postoperative ankle valgus was statistically significant between groups (U = 303.50, p = .013). The odds of having an increase in the valgus ankle angle for patients in the triple group was 3.64 times that for patients in the double group, while holding all other variables in the model constant.