Comparison Between Polyvinyl Alcohol Implant and Cheilectomy With Moberg Osteotomy for Hallux Rigidus

Background: In 2016, the US Food and Drug Administration (FDA) approved the use of a polyvinyl alcohol (PVA) hydrogel implant for the surgical management of hallux rigidus. Though recent studies have evaluated the safety and efficacy of the implant, no study has compared outcomes following PVA implantation with those following traditional joint-preserving procedures for hallux rigidus, such as cheilectomy with Moberg osteotomy. The purpose of this study was to compare clinical and patient-reported outcomes for patients undergoing cheilectomy and Moberg osteotomy, with or without PVA implant, at a single multisurgeon academic center. Our hypothesis was that the addition of the PVA implant would result in superior clinical and patient-reported outcomes. Methods: In total, 166 patients were identified who underwent cheilectomy and Moberg osteotomy with (PVACM; n = 72) or without (CM; n = 94) a PVA implant between January 2016 and December 2018 by I of 8 foot and ankle fellowship-trained orthopedic surgeons at our institution. Of these patients, 60 PVACM and 73 CM patients had both baseline and minimum I -year postoperative Patient-Reported Outcomes Measurement Information System (PROMIS) scores. The average time to survey follow-up was 14.5 months for PVACM patients and 15.6 months for CM patients. Retrospective chart review was performed to assess the incidence of postoperative complications and reoperations, with an average clinical follow-up of 27.7 (range, 16.0-46.4) months for PVACM patients and 36.6 (range, 18.6-47.8) months for CM patients. Results: Both PVACM and CM cohorts demonstrated significant improvement in the PROMIS Physical Function, Pain Interference, Pain Intensity, and Global Physical Health domains when comparing preoperative and postoperative scores within each group (P < .01). When comparing scores between the PVACM and CM cohorts, preoperative scores were similar, while CM patients demonstrated significantly higher postoperative Physical Function (51.8 +/- 8.7 vs 48.8 +/- 8.0; P = .04) and significantly lower Pain Intensity (39.9 +/- 8.3 vs 43.4 +/- 8.7; P = .02) scores. The pre- to postoperative change in Physical Function was also significantly greater for CM patients (7.1 +/- 8.5 vs 3.6 +/- 6.2; P = .011). In the PVACM group, there were 3 revisions (5%), I reimplantation, I conversion to arthrodesis, and I revision to correct hyperdorsiflexion. In the CM group, there was I revision (1.4%), a conversion to arthrodesis (P = .21). Other postoperative complications included persistent pain (7 out of 60 PVACM patients [11.7%] and 8 out of 73 CM patients [11.0%]; P = .90) and infection in 3 PVACM patients (5%) and no CM patients (P = .05). Conclusion: Though our results generally support the safety and utility of the PVA implant as previously established by the clinical trial, at I to 2 years of follow-up, CM without a PVA implant may provide equivalent or better relief compared with a PVACM procedure, while avoiding potential risks associated with the implant.