What’s New in Foot and Ankle Surgery

This article provides a summary of orthopaedic foot and ankle research from September 2022 to September 2023. The included studies were published in The Journal of Bone & Joint Surgery, Foot & Ankle International, Foot and Ankle Surgery, Clinical Orthopaedics and Related Research, The American Journal of Sports Medicine, JAMA Network Open, The Bone & Joint Journal, and Knee Surgery, Sports Traumatology, Arthroscopy. Forefoot Lesser Toes In a study of 31 patients (40 osteotomies) who underwent percutaneous proximal phalanx medial closing-wedge osteotomy of the second toe to correct valgus malalignment (the Akinette procedure), Ray et al.1 reported that there was satisfaction or extreme satisfaction with the procedure for 91.4% of the cases. The second toe valgus angle was significantly lower at a mean follow-up of 1.6 years, with no major complications and only 2 cases of radiographic recurrence. Choi et al.2 retrospectively compared the outcomes of lesser toe mucous cyst excision with either osteophytectomy (n = 22) or distal interphalangeal joint arthrodesis (n = 16). Seven recurrences (31.8%) were reported with osteophytectomy, and none were reported with arthrodesis (p < 0.001). Arthrodesis provided a more reliable outcome. Hallux Rigidus In a large, retrospective, multicenter, cohort trial on nonunion after great toe metatarsophalangeal arthrodesis (794 patients), Füssenich et al.3 reported 121 nonunions (15.2%), with 88 (11.1%) symptomatic at a median follow-up of 12.7 months. Interestingly, the risk of nonunion was significantly higher with convex and concave reamers (odds ratio [OR], 2.80; p < 0.001) and hand instruments (OR, 3.75; p < 0.001) for joint preparation compared with planar cuts. Significantly higher risk of nonunion was also found with the use of crossed screws compared with a dorsal plate and interfragmentary screw for fixation. Guyton4 examined the assumptions behind the definition of surgical success in the MOTION trial comparing the Cartiva implant (Stryker) with arthrodesis. Changes in the assumed definitions of success, using pain threshold modeling, considering painless pseudarthrosis as a success, and not equating implant removal with complete revision, reversed the published finding of the noninferiority of the implant. Hallux Valgus Investigations continue into the complex relationship between first metatarsal (M1) pronation, sesamoid station, and hallux valgus deformity. Lalevée et al.5 retrospectively reviewed the weight-bearing computed tomographic (CT) scans of 70 patients (88 feet) with symptomatic hallux valgus to study the M1 head pronation distribution in hallux valgus. The mean metatarsal pronation angle and the mean α angle exceeded normative values. The authors were unable to identify a threshold for pathological pronation or accurately estimate the prevalence of hyperpronation in the hallux valgus population. In a cadaveric weight-bearing CT study, Lalevée et al.6 assessed the effect of M1 hyperpronation on first-ray alignment. The authors showed that a 30° M1 pronation osteotomy increased the metatarsal pronation angle and hallucal pronation in the non-weight-bearing state and increased the compensatory M1 supination by 8° (p < 0.001) with simulated weight-bearing. The addition of a medial soft-tissue release at the first metatarsophalangeal joint further increased compensatory supination of M1 by an additional 7.5° (p = 0.003). Compensatory first-ray supination with weight-bearing was associated with an increased hallux valgus angle and intermetatarsal angle. To assess the relationship between hallux valgus recurrence and sesamoid position on anteroposterior standing radiographs, Kim et al.7 performed a retrospective study of 269 feet (185 patients) treated for hallux valgus with a distal chevron metatarsal osteotomy. At a mean follow-up of 2.8 years, there were 11 recurrences (7.5%) (hallux valgus angle, ≥15°) in the group with sesamoid reduction (147 feet, sesamoid grades I to III) and 23 recurrences (18.9%) in the group without sesamoid reduction (122 feet, sesamoid grades IV to VII) (p = 0.406). The hallux valgus angle and intermetatarsal angle in both groups worsened over time from the initial postoperative measurement to that of the final follow-up. The authors concluded that the postoperative sesamoid position on radiographs was not associated with hallux valgus recurrence. Ji et al.8 used weight-bearing CT scans to compare the first tarsometatarsal joint (TMT1) morphology between 82 feet with hallux valgus and 79 controls. Multiple radiographic parameters differed significantly between the 2 groups, including the middle facet width, inferior lateral facet height and angle, and the TMT1 angle (p < 0.05). The authors identified 4 types of TMT1 morphology. Continuous-flat morphology was associated with more severe hallux valgus and TMT1 instability. Midfoot In a large retrospective case series (n = 95), Wong-Chung et al.9 proposed a new classification system for Müller-Weiss disease considering the age of onset, existence of a previously normal navicular bone and talonavicular joint, joints affected, lateral Meary's angle, and dorsoplantar talo-first metatarsal angle. The authors identified 3 groups: early-onset, late-onset, and Müller-Weissoid, a subgroup of patients who had normal navicular bones when they were in the sixth decade of life but developed navicular changes at a mean time of 5 years later. Iturregui et al.10 retrospectively reviewed 48 patients (55 feet) who underwent deep peroneal neurectomy for treatment of symptomatic dorsal midfoot arthritis. At a mean follow-up of 2.9 years, 80.8% of patients were satisfied with the surgical procedure, but only 56% of the living eligible patients participated and 55.6% endorsed current activity limitations at the final follow-up. Six feet (10.9%) had complications requiring reoperation. The complication rate and activity restrictions gave a less favorable impression of the procedure, and the study had a high rate of patients lost to follow-up. Ankle and Hindfoot Arthroplasty Ankle arthroplasty continues to be more widely performed. Using the IBM MarketScan database from 2009 to 2019, Karzon et al.11 found that the annual primary total ankle arthroplasty (TAA) volume in the United States increased by 136.1% and the incidence per 100,000 population increased by 120.8%. The annual growth rate of TAA was 2 times greater than that for total knee arthroplasty and 3.6 times greater than that for total hip arthroplasty over the study period. Yoon et al.12 retrospectively analyzed long-term outcomes of 151 TAAs (136 patients) using the mobile-bearing HINTEGRA prosthesis (Newdeal/Integra). At a minimum follow-up of 10 years, significant improvement was seen in clinical outcome measures and ankle range of motion (p < 0.001). However, clinical scores tended to decline from 2 years and approximately one-half of patients had a visual analog scale (VAS) pain score of >2 at the final follow-up. There were 9 cases (6.0%) of metal component revision, and implant survivorship was 93.5%. Overall, 43 ankles (28.5%) underwent a second surgical procedure, for which periprosthetic osteolysis was the most common indication. These results are broadly in line with those of other modern prostheses. The first non-inventor studies of the Cadence Total Ankle System prosthesis have raised some concerns and included a retrospective series of 48 ankles (48 patients) in which Kim et al.13 reported 3 early metal component revisions, for an implant survivorship of 93.7% at a mean follow-up of 2.8 years. Periprosthetic lucency was identified in 15 ankles, all on the tibial side, of 42 ankles available for radiographic analysis. Patients with periprosthetic lucency had significantly worse postoperative Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function scores than those without lucency. In a study of 60 TAA cases (60 patients) with this prosthesis at a mean follow-up of 2.9 years, Malherbe et al.14 reported an all-component survivorship of 98.3%. A high rate of periprosthetic lucency included 9 ankles (15%) with osteolytic cysts and 33 ankles (55%) with radiolucent lines at bone-implant interfaces. Mechanical failure in the form of peg fracture of the talar implant was identified in 5 ankles (8.3%). A viable but imperfect solution for a patient who experiences pain after ankle arthrodesis is conversion to arthroplasty. Lundeen et al.15 retrospectively reviewed 51 patients (52 ankles) who experienced pain after ankle arthrodesis and underwent conversion to TAA using a fixed-bearing prosthesis (Salto Talaris, Smith & Nephew). At a mean follow-up of 4.2 years, the Veterans RAND Health Survey (VR-12) physical score, the Ankle Osteoarthritis Scale (AOS) pain and disability subscales, and the VAS pain scores all improved significantly. The radiographic ankle range of motion was dorsiflexion of 15.5° to plantar flexion of 14.4° at the final follow-up. The final mean VAS score was 29.4, but had a high standard error, indicating that many patients still experienced substantial pain. Fletcher et al.16 performed a comparative cohort study of 25 patients (50 ankles) who underwent bilateral simultaneous TAA and 25 patients (50 ankles) who underwent bilateral sequential TAA at least 3 months apart. At a mean follow-up of 4.4 years, there was no difference between cohorts in clinical outcome measures, complications, reoperations, or failure-free survival (100.0%). Positive outcomes were achieved in the bilateral simultaneous TAA group despite immediate weight-bearing on the implants. The study may provide support for early weight-bearing after TAA in general. The optimal treatment strategy for failed TAA continues to stoke debate. In a nonrandomized observational study, Pfahl et al.17 compared revision TAA (n = 40), revision total ankle arthrodesis (n = 46), and revision tibiotalocalcaneal arthrodesis (n = 25) for failed TAA. At a mean follow-up of 5.7 years, there were 5 failures (10.9%) in the revision total ankle arthrodesis group, 1 (2.5%) in the revision TAA group, and none in the revision tibiotalocalcaneal arthrodesis group (p = 0.15). Survival rates were not significantly different between groups. Jennison et al.18 performed a data linkage study combining the U.K. National Joint Registry and National Health Service (NHS) Digital data to assess the survival rate of arthrodesis after failed TAA. The authors identified 131 patients with a mean follow-up of 4 years who underwent either ankle arthrodesis or tibiotalocalcaneal arthrodesis for failed TAA. Survival was 96% at 1 year, 77.5% at 3 years, and 72% at 5 years. Overall, 29 patients (22.1%) had arthrodeses that failed, and these patients underwent revision arthrodesis (24 patients) or amputation (5 patients). Using a similar methodology, the same group19 studied the survival rates of revision TAA after failed primary TAA in 228 patients at a mean follow-up of 2.6 years. Survival was 95.4% at 1 year, 87.7% at 3 years, and 77.5% at 5 years. Twenty-nine patients (12.7%) had initial revision TAA that failed, and these patients underwent further revision arthroplasty (9 patients [3.9%]), conversion to arthrodesis (19 patients [8.3%]), or amputation (1 patient [0.4%]). Wallace et al.20 performed a prospective cohort study of 41 patients to assess the effect of low-cost, noninvasive, soft-tissue expansion strips on wound complications after anterior-approach TAA. Significantly fewer patients (p = 0.001) in the experimental group (6 [15%]) had additional clinic visits for wound-related concerns compared with a historical control group (20 [49%]). The rate of superficial infection was lower (p = 0.04) in the experimental group (2%) than in the control group (12%). There were no deep infections in either group. Arthrodesis Woods et al.21 retrospectively analyzed 271 patients (284 arthrodeses) to assess nonunion rates after arthroscopic ankle arthrodesis fixed with 2 or 3 cannulated screws. At a median follow-up of 12 months, nonunion developed after 22 arthrodeses (7.7%). Previous triple hindfoot arthrodesis was an independent risk factor for nonunion (OR, 18.3; p < 0.001). Ankle arthrodesis after triple arthrodesis likely requires more robust fixation. Lause et al.22 performed a retrospective study of 28 patients with delayed union (<75% osseous bridging on CT scan) between 2 and 6 months after ankle arthrodesis. At a mean follow-up of 5.6 years, 20 patients (71%) required a revision surgical procedure or were dissatisfied. Patients with minimal or moderate osseous bridging were significantly more likely to have a successful outcome than those with absent osseous bridging despite clinical stability. Kang et al.23 retrospectively reviewed 165 cases of subtalar arthrodesis performed for posttraumatic subtalar arthritis secondary to calcaneal fracture. There were 22 nonunions (13.3%) diagnosed at 6 months via clinical examination and radiographs, with CT scans performed only in equivocal cases. Smoking, parallel screw configuration, and use of freeze-dried iliac crest graft for distraction subtalar arthrodesis were identified as risk factors for nonunion. Sports Achilles Tendon Compelling evidence for the benefit of platelet-rich plasma in treating Achilles tendon pathology remains elusive. Keene et al.24 reported the 2-year results of a randomized multicenter trial comparing platelet-rich plasma with a placebo injection in the nonoperative treatment of 230 acute midsubstance Achilles tendon ruptures. Injections were performed within 12 days of the injury. At the 2-year follow-up, there was no difference in the Achilles Tendon Rupture Score (ATRS) (p = 0.76) or quality-of-life measures between the platelet-rich plasma group (n = 85) and the placebo group (n = 92). McCormick et al.25 retrospectively analyzed 9 patients who underwent Achilles tendon reconstruction with dual semitendinosus allografts for extensive pathology. The grafts were secured proximally with a Pulvertaft weave and distally with bone tunnels in the calcaneus without interference screws. At a median follow-up of 66 months, the median ATRS was 93 of 100, with 1 graft that failed at 9.5 years. All patients were able to walk without assistive devices at the final follow-up. Aufwerber et al.26 prospectively randomized 47 patients to functional mobilization (29 patients) or standard cast treatment (18 patients) during the first 2 weeks after surgical repair of acute Achilles tendon ruptures. At 6 months postoperatively, there were no significant differences between the groups in tendon length, muscle volume, heel-rise tests, or ATRS. This finding supports the clinical decision to immobilize and allow for wound-healing before starting range of motion after Achilles repair. Cramer et al.27 used radiotracers to assess collagen turnover in the Achilles tendon in 18 patients treated surgically for Achilles ruptures. Turnover in the years prior to rupture was assessed on the basis of correlating carbon-14 levels with the known atmospheric variations of that isotope, and turnover immediately after rupture was assessed using an orally administered tracer given shortly after rupture. Turnover after repair was assessed through a third tracer administered during the surgical procedure. Higher-than-normal collagen turnover was found to precede Achilles rupture, a finding that supports the theory that most tendons that rupture are already pathologic. New tendon collagen formation was not seen immediately after rupture or repair. Ankle Instability The potential benefits of suture tape utilization in ankle ligament reconstruction remain difficult to quantify. Comfort et al.28 retrospectively compared patient-reported outcomes after lateral ligament repair with suture tape augmentation (n = 37) and without it (n = 48). At the median follow-up of 5 years, there was no difference in the Foot and Ankle Ability Measure (FAAM), 12-Item Short-Form Health Survey (SF-12), Tegner Activity Scale, or patient satisfaction between groups. Only 1 patient (in the suture tape group) required revision for recurrent lateral ankle instability. Peroneal Tendons Nishimura et al.29 compared magnetic resonance imaging (MRI) and CT scans of 30 patients with recurrent peroneal tendon dislocation and the scans of 30 control patients. No difference was observed between groups in the shape of the retromalleolar groove, posterior tilting angle of the fibula, or presence of accessory peroneal tendon muscles. Peroneal muscle volume was significantly larger (p < 0.001), and the height of peroneus brevis muscle tissue was significantly lower (p = 0.001), in the group with tendon dislocation. The study suggested that soft-tissue volume, not retromalleolar morphology, is a primary factor in recurrent peroneal tendon dislocation. Osteochondritis Dissecans of the Talus The treatment of large talar osteochondral lesions remains a clinical challenge. Dahmen et al.30 reported 2-year results of 43 patients who underwent novel press-fit talar osteoperiostic grafting from the iliac crest (TOPIC) for large (>1-cm) osteochondral lesions of the talus. The authors reported a 100% union rate of the medial malleolar osteotomy and 100% consolidation of the talar grafts on CT scans. Although all clinical outcome measures improved, the final median pain score of 2 indicates that one-half of the patients still had unacceptable pain levels at the final follow-up. The morbidity associated with a medial malleolar osteotomy and iliac crest graft harvest must also be considered. Vannini et al.31 published the minimum 10-year results of 101 patients with a mean age of 32.9 ± 10.9 years who underwent a 1-step hyaluronic scaffold transplantation with bone marrow concentrate for large osteochondral lesions of the talus (mean lesion size, 2.2 cm2). Significant improvement was observed in clinical outcome scores, but only modest improvement was observed in return to sport, with the activity level being below that of the general population. Although the pain score improved from baseline to the final follow-up, pain increased significantly after the 5-year time point, indicating deterioration of results over time. Trauma Lisfranc Injuries Koroneos et al.32 performed a cadaveric biomechanical study comparing suture tape fixation including a dorsal intercuneiform limb (n = 8) and traditional cannulated screws (n = 8) for the fixation of ligamentous Lisfranc injuries. No significant differences were found in gapping at the Lisfranc articulation or the intercuneiform articulation. There were 2 failures (gapping of ≥2 mm at the Lisfranc articulation) in the screw group and none in the suture tape group. Clinical data have been limited. Ankle Fractures Chronic kidney disease is emerging as an independent risk factor for poor outcomes after a surgical procedure for lower-extremity trauma. Sinkler et al.33 retrospectively reviewed 136 patients (68 with and 68 without chronic kidney disease) who underwent operative and nonoperative treatment for torsional ankle fractures. Patients with kidney disease were more likely to undergo unplanned arthrodesis and almost 4 times more likely to experience any complication. In a prospective, noninferiority, randomized controlled trial, Gregersen et al.34 assessed the influence of stable and unstable gravity stress radiographs on the outcomes of nonoperatively treated Weber B ankle fractures (149 patients) that had stable weight-bearing ankle radiographs. At the 2-year follow-up, the difference between groups in the Manchester-Oxford Foot and Ankle Questionnaire (MOXFQ) fell within the predefined noninferiority margin of 7.5 points. The results suggested that, if a weight-bearing radiograph shows no talar shift, ankles may be successfully treated nonoperatively without further stress views. There were limitations, as noted in a commentary on the article35. The study did not address whether long-term arthritic changes would develop, what should be done with patients who could not tolerate weight-bearing radiographs, or whether the findings were generalizable given the high prevalence of neuropathy in the general population. O'Keefe et al.36 reported the minimum 5-year results from the CROSSBAT (Combined Randomised and Observational Study of Surgery vs No Surgery for Type B Ankle Fracture Treatment) study, which randomized patients to operative treatment (80 patients) or nonoperative treatment (80 patients) of Weber B ankle fractures with a minimal talar shift, where a greater than minimal talar shift was defined as medial clear space >2 mm wider than the superior clear space on initial non-weight-bearing radiographs. With 40 operatively treated patients and 37 nonoperatively treated patients available at a mean follow-up of 7.3 years, no difference was found between the groups in the American Academy of Orthopaedic Surgeons (AAOS) Foot and Ankle Outcomes Questionnaire or the Physical Component Summary of the SF-12 version 2. As expected, the surgical group had a higher rate of adverse events (OR, 3.7; p = 0.04). Talar Fractures Mechas et al.37 retrospectively reviewed 137 talar fractures at a level-I trauma center to determine whether fracture extension into the talar body (proximal to a line subtended by the anterior edge of the dorsal cartilage and the lateral process) increased rates of osteonecrosis. The authors identified 80 fractures (58%) isolated to the talar neck and 57 fractures (42%) with proximal extension. At a median follow-up of 10 months, after adjusting for diabetes, smoking status, and Hawkins classification type, the proximal extension group was >3 times more likely to develop osteonecrosis. The results suggest a need to reconsider the classification of talar neck fractures to account for fracture extension proximally into the talar body. Diabetes Contrary to intuition, neuroarthropathy outcomes in patients without diabetes may be similar to those in patients with diabetes. Waibel et al.38 retrospectively compared the clinical course of 159 patients with diabetes with a mean follow-up of 6 years and 60 patients without diabetes with a mean follow-up of 5 years treated nonoperatively for Charcot neuro-osteoarthropathy. There was no difference at 10 years in the risk of major amputation, any surgical procedure on the foot, or Charcot reactivation between the groups. Racial differences and social settings continue to predict complications. Using 2013 to 2014 data from the U.S. National Medicare Claims Data Database, Brennan et al.39 performed a retrospective cohort study to examine risk factors associated with major limb amputation and death in 124,487 patients with diabetes who were hospitalized. Overall, 17.6% of the cohort underwent major limb amputation or died within 30 days of admission. The rate increased to 28.0% for patients who identified as Black and were living in a rural area, an absolute increase of >10.0%. Bullock et al.40 retrospectively examined the risk of subsequent ulceration in transmetatarsal amputations with Achilles tendon lengthening (n = 30) and without it (n = 55). At a mean follow-up of almost 2 years, 19 feet (35%) developed late forefoot ulceration in the group without Achilles lengthening compared with 1 foot (3%) in the group with Achilles lengthening (p < 0.001). Of the 19 feet with ulceration, almost one-half (47%) developed osteomyelitis and 16% underwent proximal amputation. The study reinforced the teaching that Achilles tendon lengthening should always be performed concomitantly with transmetatarsal amputation, especially in younger patients. Ron et al.41 compared 185 patients who underwent transmetatarsal amputation that progressed to below-the-knee or above-the-knee amputation and 156 patients who underwent transmetatarsal amputation but did not undergo more proximal amputation. Renal impairment stood out as an independent risk factor and was more common in patients with failed transmetatarsal amputation (p = 0.002). A Charlson Comorbidity Index of >7.5 was also a predictive threshold for failure. Hong et al.42 retrospectively reviewed 89 partial first-ray amputations in patients with diabetes to determine the risk factors for a poor outcome, focusing on the effect of the residual length of the first metatarsal. The preservation of more than one-third of the length of the first metatarsal was associated with a lower chance of the transfer amputation of lesser rays. Patient-Reported Outcome Measures Bernstein et al.43 randomized new patients presenting to an academic foot and ankle clinic into 2 groups: a group who reviewed and discussed their PROMIS scores with the surgeon (n = 130) and a control group who did not discuss their PROMIS scores with the surgeon (n = 145). Of those patients who completed the study, there was no difference between groups on the Clinician & Group Survey-Consumer Assessment of Healthcare Providers and Systems (CG-CAHPS) questions or the Patient Activation Measure (PAM) (p > 0.05). When clustered by surgeon, patients in the group who discussed scores were less likely to indicate a top score for the understanding domain of the CG-CAHPS survey and had decreased odds of high patient activation. Surprisingly, the study indicated that collecting patient-reported outcome measures and discussing them with patients did not improve the patient experience or activation and might have a negative effect. Evidence-Based Orthopaedics The editorial staff of JBJS reviewed a large number of recently published studies related to the musculoskeletal system that received a higher Level of Evidence grade. In addition to articles cited already in this update, 3 other articles relevant to foot and ankle surgery are appended to this review after the standard bibliography, with a brief commentary about each article to help guide your further reading, in an evidence-based fashion, in this subspecialty area. Evidence-Based Orthopaedics Goldberg AJ, Chowdhury K, Bordea E, Hauptmannova I, Blackstone J, Brooking D, Deane EL, Bendall S, Bing A, Blundell C, Dhar S, Molloy A, Milner S, Karski M, Hepple S, Siddique M, Loveday DT, Mishra V, Cooke P, Halliwell P, Townshend D, Skene SS, Doré CJ, Brown R, Butler M, Chadwick C, Clough T, Cullen N, Davies M, Davies H, Harries B, Khoo M, Makwana N, Murty A, Najefi A, O'Donnell P, Raglan M, Thomas R, Torres P, Welck M, Winson I, Zaidi R; TARVA Study Group. Total ankle replacement versus arthrodesis for end-stage ankle osteoarthritis: a randomized controlled trial. Ann Intern Med. 2022 Dec;175(12):1648-57. [Erratum in: Ann Intern Med. 2023 Feb;176]. The TARVA (Total Ankle Replacement Versus Ankle Arthrodesis) trial was a multicenter, randomized controlled trial comparing TAA (n = 152) with ankle arthrodesis (n = 151) for end-stage ankle arthritis. Fixed-bearing and mobile-bearing total ankle implants and arthroscopic and open approaches for ankle arthrodesis were used by 33 experienced surgeons across 17 centers in the NHS of the United Kingdom, with the primary outcome as change in the MOXFQ walking/standing (W/S) domain scores at 52 weeks postoperatively. Both groups demonstrated clinically meaningful improvement in MOXFQ-W/S scores. The adjusted difference in the change from baseline in MOXFQ-W/S scores was −5.6, favoring the TAA group, but the difference was not clinically important or significant. Patients with fixed-bearing implants showed nonsignificantly greater improvement in MOXFQ-W/S scores. Subgroup analysis modeled for adjacent joint arthritis favored TAA, especially when fixed-bearing implants were compared with ankle arthrodesis. There were more wound complications (13.8% compared with 5.5%) and nerve complications (4.3% compared with <1%) in the TAA group and more thromboembolic events (5% compared with 3%) in the arthrodesis group. There was a 7% rate of symptomatic nonunion in the arthrodesis group. Overall, the study suggested that both TAA and ankle arthrodesis are good options for treating end-stage ankle arthritis, with no differences in patient-reported function at 1 year and similar rates of adverse events. The authors should be commended on a well-designed and well-executed study, especially given patient preference for arthroplasty. However, the results should be interpreted with caution because multiple types of arthroplasty implants and ankle arthrodesis techniques were utilized and follow-up was short. The advantage of TAA over ankle arthrodesis in the long term likely lies in objective improvements in gait and sparing of adjacent joint degeneration, as subgroup analysis in this study suggested, not in early pain relief or self-reported function. Haque A, Parsons H, Parsons N, Costa ML, Redmond AC, Mason J, Nwankwo H, Kearney RS; the AIR Trial collaborators. Use of cast immobilization versus removable brace in adults with an ankle fracture: 2-year follow-up of a multicentre randomized controlled trial. Bone Joint J. 2023 Mar 15;105-B(4):382-8. This multicenter randomized controlled trial reported the 2-year results of the Ankle Injury Rehabilitation (AIR) Trial, which compared plaster casting with a fixed-angle, removable brace for the treatment of operatively and nonoperatively managed acute ankle fractures. The primary outcome was the Olerud-Molander Ankle Score (OMAS). Secondary outcomes included the EuroQol-5 Dimension-5 Level (EQ-5D-5L) measure, the Disability Rating Index (DRI), and complications. At the final follow-up of 2 years, 436 of the original 669 randomized patients were available for analysis. Both the cast and the brace group improved substantially compared with baseline, and there was no significant or clinically meaningful difference between groups in OMAS scores (−0.3; p = 0.866). No differences were found in the EQ-5D-5L or the DRI. The limitations included the high rate of loss to follow-up. Based on these results, surgeons may consider either a cast or a brace when treating ankle fractures, although a brace has the advantage of allowing hygiene and earlier range of motion. Tillmann F, Hoepelman RJ, Cagienard F, Link BC, Babst R, van de Wall BJM, Beeres FJP. Primary closure or secondary wound healing of pin sites after external fixator removal: a single-center blinded randomized controlled trial. J Bone Joint Surg Am. 2023 Feb 1;105(3):207-13. This single-center, randomized controlled noninferiority trial compared primary wound closure with suture (123 pin sites) and secondary healing (118 pin sites) for pin sites after the removal of a temporary upper-extremity or lower-extremity external fixator, with the primary outcome of local pin-site infection requiring antibiotics or wound debridement. The median duration of external fixation was 6 days. Patients were followed at 2, 6, 12, 24, and 52 weeks postoperatively. There were no pin-site infections in either group. The primary closure group had faster wound-healing time (median, 2 compared with 6 weeks; p = 0.013), with 82% healed in the primary closure group compared with 61% in the secondary healing group (p = 0.001) at 2 weeks. There was no difference between groups in satisfaction (p = 0.157), VAS pain (p = 0.940), the Vancouver Scar Scale (p = 0.722), or total complications (p = 0.438) at the final follow-up. The results suggest that the traditional teaching of treating external fixator pin sites with secondary wound-healing should be reconsidered. Given the faster healing time and similar low risk of infection, primary closure of pin sites is a reasonable alternative after removing temporary external fixators.