Midterm Outcomes of 3-Component Mobile-Bearing Total Ankle Arthroplasty in 124 Ankles: Influence of Preoperative Varus, Valgus, and Neutral Alignment Authors: Jung-Won Lim, Woo Sub Kim, Je-Min Im, Hong-Geun Jung Published: 2026 ABSTRACT BackgroundThe influence of preoperative coronal alignment on outcomes after total ankle arthroplasty (TAA) remains controversial, particularly in the presence of varus or valgus deformity. This study aimed to evaluate midterm clinical and radiographic outcomes of Salto mobile-bearing TAA according to preoperative varus, neutral, and valgus alignment.MethodsThis retrospective single-center study reviewed 124 ankles that underwent primary TAA between 2014 and 2021 and were followed up for a minimum of 2 years. Ankles were classified into varus, neutral, and valgus groups based on predefined radiographic criteria. Clinical and radiographic outcomes, as well as implant survivorship, were evaluated. Adjunctive corrective procedures were performed as indicated to address associated deformities.ResultsFinal clinical outcomes, including American Orthopaedic Foot and Ankle Society scores, visual analog scale pain scores, and ankle range of motion, did not differ significantly among alignment groups. Most radiographic parameters improved postoperatively; however, ankles with preoperative varus alignment demonstrated greater residual hindfoot varus. Kaplan-Meier analysis demonstrated favorable midterm revision-free and reoperation-free survivorship.ConclusionsTAA can achieve favorable midterm clinical outcomes across a spectrum of preoperative coronal alignment patterns when combined with appropriate adjunctive corrective procedures. Varus deformity, including more severe forms, often requires additional correction; however, when appropriately addressed, clinical outcomes are likely to be maintained. Given the limited number of valgus ankles and the midterm follow-up duration, cautious interpretation is warranted, and further studies are needed to clarify long-term outcomes. Total ankle arthroplasty (TAA) has become an increasingly accepted surgical option for the management of end-stage ankle arthritis, offering pain relief and restoration of motion while preserving more physiological gait patterns than ankle arthrodesis. Over the past 2 decades, advances in implant design, surgical technique, and perioperative management have resulted in improved outcomes and implant survivorship.1) Despite these advances, preoperative coronal deformity remains an important consideration in TAA. Historically, varus or valgus malalignment was regarded as a relative contraindication because of the risk of component malposition, asymmetric load distribution, and early loosening. However, several recent studies have reported comparable outcomes in ankles with varus or valgus deformities when neutral alignment was successfully restored during surgery.2345) The Salto mobile-bearing prosthesis is a 3-component system designed to replicate natural ankle kinematics and reduce bone–implant interface stress through its mobile polyethylene insert. Previous reports have shown favorable mid- to long-term results with this system,678) but evidence specifically analyzing the influence of preoperative coronal alignment remains limited.9) Therefore, the present study aimed to evaluate the midterm clinical and radiographic outcomes of Salto mobile-bearing TAA according to preoperative coronal alignment (varus, neutral, and valgus). We hypothesized that when postoperative neutral alignment is achieved, clinical and radiographic results would not differ significantly among the 3 groups. METHODS This retrospective study was approved by the Institutional Review Board of Konkuk University Medical Center where the surgeries and research were conducted (IRB No. KUMC IRB 2022-03-044). The requirement for informed consent was waived due to the retrospective nature of the study. A total of 213 consecutive total TAAs using the Salto 3-component mobile-bearing prosthesis (Salto Mobile bearing, Tornier SA) were performed between June 24, 2014, and July 27, 2021. Of these, 24 ankles were excluded due to loss to follow-up, defined as less than 6 months of postoperative clinical observation. Ultimately, 124 ankles in patients with a minimum of 2 years of follow-up were included in the final analysis (Fig. 1). Among these, 5 patients underwent bilateral procedures. The inclusion criteria were as follows: (1) a minimum follow-up period of 24 months and (2) diagnosis of end-stage ankle arthritis due to posttraumatic osteoarthritis (PTOA), primary osteoarthritis, inflammatory arthritis such as rheumatoid arthritis, or sequelae of previous infective arthritis. The exclusion criteria were as follows: (1) active ankle infection, (2) insufficient bone stock in the tibia or talus, (3) neurological disorders, (4) poor peripheral circulation, (5) uncontrolled diabetes mellitus, and (6) high physical activity demands in relatively young patients. Patients were subsequently categorized into varus, neutral, or valgus groups according to preoperative coronal alignment, which was defined based on radiographic measurements described in the Radiographic Evaluation section. Operative Technique and Postoperative Management All procedures were performed by the senior author (HGJ) using a standardized operative technique.6) The primary goal of surgery was to restore the physiologic tibiocalcaneal axis and the anatomic level of the tibiotalar joint line. After aligning the extramedullary guide with the anterior tibial crest, the tibial cut was made perpendicular to the mechanical axis in the coronal plane, with a 7° posterior slope. The depth of the distal tibial resection was determined to accommodate the combined thickness of the metal tibial baseplate and the polyethylene insert. Particular attention was paid to proper rotational alignment of the components, with the talar component centered on the midpoint between the medial and lateral talar facets. In cases with lateral ligamentous laxity, the medial deltoid ligament was released from the tip of the medial malleolus to achieve soft-tissue balance. Ligamentous tension was then adjusted intraoperatively by modifying the polyethylene insert thickness. The tibial and talar components, both coated with calcium hydroxyapatite and titanium, were implanted without cement. A mobile-bearing ultra-high-molecular-weight polyethylene insert was positioned between the components to provide a congruent articulation. In cases where neutral coronal alignment could not be achieved through the standard technique, adjunctive corrective procedures such as medial release, lateral ligament repair, or osteotomy were performed as needed to restore a balanced alignment. Details of these adjunctive procedures are described in a later section. A short leg splint was applied for 1 week after surgery, followed by the application of a short leg walking boot with partial weight-bearing for 6 weeks. With the use of an ankle brace, full weight-bearing was permitted at 6 to 12 weeks postoperatively. Active and passive ankle range of motion (ROM) exercises were recommended after stabilization of the operative wound. For patients who underwent adjunctive osteotomy, toe-touch gait with a short leg walking boot was allowed for 6 weeks postoperatively, followed by partial weight-bearing at 6 to 12 weeks and full weight-bearing thereafter with an ankle brace. Indications for Adjunctive Procedures The indications for adjunctive procedures were determined through a combination of preoperative planning and intraoperative assessment. Preoperative weight-bearing radiographs, including hindfoot alignment views, were used to evaluate the primary source and severity of deformity. A predefined correction strategy was planned prior to surgery, and the final decision to perform adjunctive procedures was confirmed intraoperatively based on residual malalignment and soft-tissue balance after trial component implantation. Achilles tendon lengthening by percutaneous triple hemisection (TAH) was performed when ankle dorsiflexion remained less than 10° after component implantation,10) indicating persistent equinus that could compromise implant positioning or postoperative function. Lateral sliding calcaneal osteotomy (LSCO) was considered in cases of varus hindfoot alignment when residual varus deformity persisted after tibiotalar realignment, particularly when preoperative hindfoot alignment views demonstrated a hindfoot-driven varus component. The goal of LSCO was to restore a more neutral hindfoot axis and improve load distribution across the ankle joint. First tarsometatarsal dorsiflexion arthrodesis (1TMTDFA) was selectively performed in ankles with a cavus or forefoot-driven varus component, as determined by preoperative clinical examination and radiographic assessment. This procedure was intended to correct the plantarflexed first ray and reduce compensatory varus stress on the ankle. Soft-tissue procedures, including deltoid ligament release and lateral ligament repair, were performed as needed to achieve balanced ankle stability after bony realignment and component positioning, based on intraoperative assessment of residual medial or lateral instability. Although intraoperative judgement played a role in confirming the need for additional correction, the selection of adjunctive procedures followed a consistent correction algorithm based on the anatomic source of deformity rather than surgeon preference alone (Fig. 2). Clinical Evaluation Clinical outcomes were assessed preoperatively and at the final follow-up using a visual analog scale (VAS) for pain (with 0 indicating no pain and 10 indicating the worst pain) and the American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot score. Patient satisfaction was recorded at the final follow-up using a 4-grade scale (very satisfied, satisfied, fair, and dissatisfied).11) ROM was measured by a single examiner (JMI) using a goniometer as the sum of active dorsiflexion and plantarflexion, with the knee flexed to minimize gastrocnemius tightness. Postoperative complications were classified as either major or minor according to established criteria.12) Major complications included deep infection, aseptic loosening, component subsidence, periprosthetic osteolysis, intraoperative malleolar fracture, periprosthetic fracture, and polyethylene component subluxation. Minor complications consisted of wound problems, heterotopic ossification, and postoperative neuralgia.1314) Revision was defined as any surgical procedure involving the prosthetic components, including insert exchange.15) Major revision, indicating prosthesis failure, was defined as the exchange or removal of any metallic component. Minor revision was defined as a surgical procedure limited to exchange of the polyethylene insert. Reoperation (subsequent procedures) was defined as any additional surgery performed after the index procedure that did not involve the prosthetic components, such as soft tissue procedures or hardware removal.16) Radiographic Evaluation Standardized radiographic evaluations, including weight-bearing anteroposterior and lateral views of the ankle and hindfoot alignment views, were performed preoperatively and postoperatively. Follow-up radiographs were obtained at 3, 6, and 12 months after surgery, and annually thereafter. Coronal plane ankle and hindfoot alignments were assessed using 5 radiographic parameters: the distal tibial articular surface (TAS) angle, talar tilt angle (TTA), tibial axis–talar dome angle (T-TDA), talar dome–ground surface angle (TD-GSA), and tibiocalcaneal angle (TCA). The TAS, TTA, T-TDA, and TD-GSA were measured on weight-bearing anteroposterior radiographs of the ankle. The TAS angle was defined as the angle between the tibial shaft and the tibial plafond. The TTA was measured as the angle between the tibial plafond and the superior surface of the talar dome. The T-TDA represented the angle formed by the anatomical axis of the tibia and the line perpendicular to the tangent of the talar dome or prosthetic talar component. The TD-GSA was defined as the angle between the talar dome and the horizontal ground surface. The TCA was measured on the hindfoot alignment view as the angle between the longitudinal axis of the tibia and the calcaneal axis (Fig. 3).411171819) All radiographic measurements were independently performed by 2 blinded observers (WSK and JMI) who were not involved in the surgical procedures. Measurements were performed twice for each case using standardized digital tools within the Picture Archiving and Communication System (PACS). To minimize measurement variability, the final value for each parameter was calculated as the mean of the 2 middle values after excluding the highest and lowest measurements. To assess measurement reliability, a subset of radiographs was re-evaluated by each observer after an interval of at least 2 weeks. Interobserver and intraobserver reliability were assessed using intraclass correlation coefficients (ICCs) based on a 2-way random-effects model with absolute agreement [ICC (2,1)]. Patients were categorized into 3 groups—varus, neutral, and valgus—based on a multidimensional assessment of coronal plane alignment using predefined radiographic parameters. The varus group included ankles that met at least one of the following criteria: a TAS angle less than 84°, a TTA greater than 10°, a T-TDA greater than 10°, a TD-GSA greater than 10°, or a TCA greater than 10°. The valgus group was defined as ankles with a TAS angle greater than 96°, or values of TTA, T-TDA, or TD-GSA exceeding 10° in the valgus direction (negative values). Ankles that did not meet the criteria for either varus or valgus deformity were classified as the neutral group. This composite classification approach1011) was adopted to account for the fact that coronal deformity may manifest at different anatomic levels of the ankle and hindfoot rather than being adequately represented by a single radiographic parameter. The distribution of ankles meeting each individual criterion is summarized in the Supplementary Table 1. Statistical Analyses All statistical analyses were performed using SPSS software version 23.0 (IBM Corp.). Continuous variables were presented as medians with interquartile ranges, as appropriate, and categorical variables as frequencies and percentages. The normality of continuous data was assessed using the Shapiro-Wilk test. For comparisons among the varus, neutral, and valgus groups, 1-way analysis of variance or the Kruskal-Wallis test was applied according to the normality of data distribution. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Post hoc pairwise analyses were conducted with Bonferroni correction. A p-value < 0.05 was considered statistically significant. Implant survivorship was analyzed using the Kaplan–Meier method. Two endpoints were defined: (1) revision-free survival, with revision defined as any procedure involving exchange or removal of 1 or more prosthetic components and (2) reoperation-free survival, defined as freedom from any subsequent surgical procedure performed after the index TAA that did not involve exchange or removal of the prosthetic components (e.g., soft-tissue procedures or hardware removal). Time to event was calculated from the date of index surgery to the date of the first event or last follow-up. Survivorship curves were generated for the overall cohort. RESULTS A total of 124 ankles (including 5 bilateral cases) were analyzed. The mean age of patients was 67.0 ± 8.8 years (range, 34–91 years), and the mean follow-up duration was 41.4 ± 18.1 months (range, 24–106 months). Based on preoperative coronal alignment, ankles were classified into varus (n = 71), neutral (n = 42), and valgus (n = 11) groups. The baseline demographic and clinical characteristics of the patients according to preoperative coronal alignment are summarized in Table 1. The valgus group comprised a smaller number of ankles compared with the varus and neutral groups. There were no statistically significant differences in any demographic parameters among the 3 alignment groups (all p > 0.05). PTOA was more prevalent in the varus and valgus groups, whereas the neutral group included a higher proportion of primary and inflammatory arthritis cases. Clinical Outcomes In the varus group, significant postoperative improvements were observed in the VAS pain score, AOFAS score, total ROM, and plantar flexion (all p < 0.05), whereas dorsiflexion showed no significant change (p > 0.05). In the neutral group, all clinical parameters—VAS, AOFAS, and all ROM components—demonstrated significant improvement after surgery (all p < 0.05). In the valgus group, significant improvement was noted in the VAS pain score, AOFAS score, and dorsiflexion angle (all p < 0.05), while total and plantar flexion ranges did not show significant change (p > 0.05). When comparing the 3 alignment groups (Table 2), there were no significant differences in pain reduction or total ROM improvement (all p > 0.05). However, the valgus group demonstrated higher final AOFAS scores (100.0 [88.0–100.0] vs. 87.0 [80.5–90.5] in varus and 90.0 [84.0–95.5] in neutral; p = 0.023) and greater final dorsiflexion angles (15.0 [12.5–20.0] vs. 10.0 [5.0–15.0] and 10.0 [6.3–15.0]; p = 0.036) compared with the other groups. Patient satisfaction assessed using a 4-grade scale is summarized in Table 3. The proportions of patient satisfaction did not differ significantly among the varus, neutral, and valgus groups (p = 0.751). When the responses were simplified into a 2-grade scale by combining very satisfied and satisfied as “satisfied,” and fair and dissatisfied as “dissatisfied,” the overall satisfaction rate remained high in all groups (87.3%, 92.9%, and 90.9%, respectively), with no significant intergroup difference (p = 0.642). Radiographic Outcomes Radiographic parameters according to preoperative coronal alignment are summarized in Table 4. When the changes in radiographic parameters before and after surgery were analyzed according to the preoperative alignment, all parameters (TAS, TTA, T-TDA, TD-GSA, and TCA) in the varus group showed significant improvement after surgery (all p < 0.05). In the neutral group, no significant changes were observed in any radiographic variable (all p > 0.05). In the valgus group, significant improvements were noted in all parameters except for the TAS angle, which changed from 90.0 (6.1) preoperatively to 88.7 (1.6) at the final follow-up (p = 0.203). Preoperatively, the varus, neutral, and valgus groups showed distinct differences in coronal alignment parameters, including TAS, TTA, T-TDA, TD-GSA, and TCA (all p < 0.001). At the final follow-up, there were no significant intergroup differences in TAS, TTA, T-TDA, or TD-GSA (all p > 0.05), whereas the final TCA angle remained significantly different among the groups (p < 0.001). The magnitude of change (Δ), defined as the difference between preoperative and final follow-up values for each radiographic parameter, varied significantly among the 3 groups for several radiographic parameters. Post hoc analysis revealed that the varus group showed significantly greater correction than the other groups in TAS, T-TDA, and TD-GSA (all p < 0.05), whereas no difference was observed between the neutral and valgus groups. For TTA, significant differences were observed among all 3 groups (p < 0.05). In contrast, for TCA, the neutral group showed significantly different changes compared with the varus and valgus groups (p < 0.05), while no difference was found between the latter 2. Interobserver and intraobserver reliability for all radiographic parameters was good to excellent. The ICC values ranged from 0.800 to 0.951 for interobserver reliability and from 0.927 to 0.990 for intraobserver reliability, indicating high measurement consistency across observers. Detailed interobserver and intraobserver reliability results are provided in Supplementary Table 2. Adjunctive Procedures Adjunctive procedures performed at the time of TAA according to preoperative coronal alignment are summarized in Table 5. Overall, concomitant adjunctive procedures were infrequently performed, with no significant intergroup differences in the average proportion of adjunctive procedures per patient (p = 0.165). The most common concomitant procedures were deltoid ligament release (38.0% in the varus group, 31.0% in the neutral group, and 27.3% in the valgus group) and TAH (26.8% in the varus group and 40.5% in the neutral group). Significant intergroup differences were found for TAH and LSCO, both of which were more frequently performed in the varus group (p = 0.025 and p = 0.015, respectively). Other procedures—including 1TMTDFA, medial malleolar fixation, subtalar arthrodesis, and lateral ligament repair—showed no significant differences among the groups (all p > 0.05). Although not statistically significant, 1TMTDFA was more frequently performed in the varus group (5 ankles), whereas no such procedure was performed in the neutral or valgus groups. Complications and Revisions/Reoperations The incidence of postoperative complications is summarized in Table 6. The overall rates of major, minor, and total complications were comparable among the 3 alignment groups, with no statistically significant differences (p = 0.776, p = 0.949, and p = 0.975, respectively). Asymptomatic osteolysis was observed in 20% of the varus group and 26% of the neutral group, whereas no cases were identified in the valgus group; however, this difference did not reach statistical significance (p = 0.155). The total complication rate per patient was 53.5% (38 / 71) in the varus group, 40.5% (17 / 42) in the neutral group, and 54.5% (6 / 11) in the valgus group. A total of 6 minor revisions (4.8%) and 11 reoperations (8.8%) were performed during the follow-up period. Kaplan-Meier survivorship analyses demonstrated favorable midterm revision-free and reoperation-free survival of TAA (Fig. 4). Revision-free survival remained high throughout the follow-up period, while reoperation-free survival showed a gradual decline over time. No major revision involving a metallic component change occurred. The detailed causes and distribution of these procedures are summarized in Table 7. Minor revisions were performed in 6 cases. In the varus group, 2 cases involved lateral ligament reconstruction with polyethylene insert exchange for residual varus deformity, and 1 case involved polyethylene insert exchange for insert subluxation. In the neutral group, 2 cases underwent irrigation and debridement with polyethylene insert exchange for deep infection, and 1 case underwent bony spur excision with polyethylene insert exchange for bony impingement. Reoperations included procedures for periprosthetic fracture, arthroscopic debridement for impingement, lateral ligament reconstruction, and superficial infection management. Among these, periprosthetic fractures, occurring both intraoperatively and postoperatively, involved the medial malleolus. There were no significant differences in the incidence of revision or reoperation among the 3 alignment groups (all p > 0.05). Supplementary Analyses Sensitivity and subgroup analyses were performed to assess the robustness of the primary findings and are summarized in the Supplementary Tables 3, 4, 5, 6 and Supplementary Fig. 1. DISCUSSION The principal finding of this study is that comparable midterm clinical outcomes were achieved across different preoperative coronal alignment groups when TAA was performed with active adjunctive corrective strategies. Rather than suggesting that coronal alignment itself is unrelated to postoperative outcomes, the present results indicate that unfavorable alignment can be effectively managed through appropriate adjunctive procedures, resulting in maintained clinical outcomes. In this context, the absence of significant differences in clinical outcomes among alignment groups should be interpreted as the effect of targeted correction strategies rather than as evidence that coronal deformity has no clinical relevance. This interpretation is particularly relevant for varus ankles, in which adjunctive corrective procedures were more frequently required and performed. Given the heterogeneity of varus deformity, we further explored the impact of deformity severity on outcomes by stratifying varus ankles according to the preoperative TCA. Increasing varus severity was associated with a higher utilization of adjunctive corrective procedures, particularly hindfoot bony correction, reflecting the greater structural complexity of more severe deformities.20) Despite these differences in deformity severity and surgical complexity, final clinical outcomes remained comparable across varus severity subgroups (Supplementary Table 4). Similarly, within the varus group, ankles requiring adjunctive correction demonstrated clinical outcomes comparable to those without adjunctive correction (Supplementary Table 5), despite exhibiting greater residual hindfoot varus postoperatively. These findings suggest that, although severe varus deformity necessitates more extensive corrective measures, appropriately applied adjunctive procedures may mitigate the potential negative impact of deformity severity on midterm clinical outcomes.420212223) The frequent use of adjunctive corrective procedures in the varus group reflects a deliberate, alignment-driven correction strategy rather than inconsistent surgical decision-making, and this structured approach likely contributed to the maintenance of clinical outcomes despite greater preoperative deformity. Despite overall postoperative improvement in most radiographic alignment parameters, a greater degree of residual hindfoot varus, as reflected by the TCA, persisted in ankles with preoperative varus alignment. This finding suggests that correction of coronal alignment at the ankle joint alone may not fully address underlying subtalar or hindfoot malalignment, particularly in cases with long-standing deformity.19) Residual hindfoot varus may influence load distribution across the ankle and hindfoot, potentially affecting functional mechanics and long-term implant behavior,192425) although its precise clinical significance remains incompletely understood. Accordingly, these findings underscore the importance of comprehensive preoperative assessment of hindfoot alignment and raise the possibility that more aggressive or tailored hindfoot correction may be warranted in selected cases. Further studies with longer follow-up are required to clarify the impact of residual hindfoot malalignment on implant survival and functional outcomes following TAA. The midterm clinical outcomes and implant survivorship observed in this study are broadly consistent with previously reported results of the Salto mobile-bearing TAA.6721) Prior Salto series have demonstrated reliable functional improvement and satisfactory implant survival when appropriate alignment and soft-tissue balance are achieved.6782126) In this context, the present findings suggest that unfavorable coronal alignment does not necessarily preclude satisfactory midterm outcomes when managed through a structured, alignment-driven correction strategy incorporating appropriate adjunctive procedures. These results extend existing Salto literature by emphasizing the role of surgical strategy, rather than alignment category alone, in determining postoperative outcomes. The overall incidence of complications, revisions, and reoperations did not differ significantly among the varus, neutral, and valgus groups during the midterm follow-up period. These findings suggest that, within the follow-up duration of the present study, preoperative coronal alignment was not associated with an increased risk of adverse events when TAA was performed with appropriate intraoperative alignment correction and balancing. This observation is consistent with previous reports indicating that satisfactory implant survival and complication profiles can be achieved across different alignment patterns when neutral postoperative alignment is attained.4822) Although varus ankles have traditionally been considered at higher risk for wound complications, edge-loading, or aseptic loosening, meticulous intraoperative balancing and the use of a mobile-bearing design may help mitigate these risks in selected cases.2021) Nevertheless, as the current analysis reflects midterm outcomes, longer-term follow-up is required to determine whether subtle alignment-related differences in implant survivorship or complication patterns may emerge over time.27) Consistent with these findings, Kaplan–Meier analysis demonstrated favorable midterm revision-free and reoperation-free survivorship in the overall cohort. This study has several limitations that should be acknowledged. First, the retrospective single-center design inherently carries the risk of selection bias, although all procedures were performed by a single senior surgeon, which minimized technical variability. Second, the relatively small number of ankles in the valgus group limits statistical power and precludes definitive conclusions regarding this subgroup. Third, coronal alignment was classified using multiple radiographic parameters rather than a single validated cutoff, which may have introduced heterogeneity within alignment groups despite the application of predefined criteria. Fourth, although bilateral ankles were included in the primary analysis, sensitivity analyses excluding bilateral cases yielded results consistent with those of the primary findings. Finally, the midterm follow-up duration may be insufficient to fully assess the long-term impact of residual hindfoot malalignment on implant survival. In this retrospective cohort study, TAA achieved favorable midterm clinical outcomes across different patterns of preoperative coronal alignment when combined with appropriate adjunctive corrective procedures. Varus deformity, including more severe forms, frequently required additional corrective interventions; however, when such strategies were applied, clinical outcomes were maintained. These findings suggest that preoperative coronal malalignment should not be viewed as an absolute contraindication to TAA, provided that deformity is actively addressed through tailored corrective techniques. Given the limited number of valgus ankles and the midterm duration of follow-up, further studies with larger cohorts and longer observation periods are warranted to better define the long-term impact of coronal alignment and residual hindfoot malalignment on implant survivorship.