Transfer of abductor hallucis tendon combined with scarf osteotomy versus single scarf osteotomy in moderate to severe hallux valgus deformity: A comparative retrospective cohort study

Background: Scarf osteotomy (SO) was broadly applied in moderate to severe hallux valgus (MSHV), and the results were satisfactory. However, due to the complicated pathologic changes in hallux valgus, the ideal surgical treatment is still controversial. Transfer of the abductor hallucis tendon combined with Scarf osteotomy (TAHTCSO) was an innovative alternative technique. This retrospective cohort study aimed to define if TAHTCSO mode resulted in improved outcomes as compared with the single SO in MSHV. Methods: Of 73 patients (92 ft) with MSHV, 36 (45 ft) were treated through TAHTCSO and 37 ones (47 ft) through SO. The patients were assessed clinically and radiographically with a 24-month follow-up. They were assessed pre-operatively and post-operatively with intermetatarsal angle (IMA), hallux valgus angle (HVA), distal metatarsal articular angle (DMAA), first metatarsophalangeal joint range of motion (1#MTP ROM), as well as American Orthopaedic Foot and Ankle Society (AOFAS) forefoot scores and postoperative complications of surgery. Results: Both cohorts had the same baseline feathures. All patients were followed up from 24 to 40 months, with a mean of 28.3 months. Patients in the TAHTCSO cohort had significantly decreased HVA at 6 months (p < 0.0001), 12 months (p < 0.0001), and 24 months (p < 0.0001) after surgery. 1#MTP had been increased slightly with non-statistic sense (p >0.05). IMA, DMAA and AOFAS also had not significantly difference at all followed time after surgery as compared with the SO cohort. The healing of osteotomies was observed within 8 weeks in the two cohorts. Two cases of hallus varus had been found in SO cohort and there were no cases of delayed healing and bone non-union in both cohorts. Conclusion: In this retrospective cohort study, TAHTCSO had sufficient maintenance of the correction and improved functional performance thereby was a good alternative for MSHV, though it did not display a better result for MSHV compared to SO.