Search

Traditionally, two-dimensional conventional radiographs have been the primary tool to measure the complex morphology of the foot and ankle. However, the subtalar, talonavicular, and calcaneocuboid joints are challenging to assess due to their bone morphology and locations within the ankle. Weightbearing computed tomography is a novel high-resolution volumetric imaging mechanism that allows detailed generation of 3D bone reconstructions. This study aimed to develop a multi-domain statistical shape model to assess morphologic and alignment variation of the subtalar, talonavicular, and calcaneocuboid joints across an asymptomatic population and calculate 3D joint measurements in a consistent weightbearing position. Specific joint measurements included joint space distance, congruence, and coverage. Noteworthy anatomical variation predominantly included the talus and calcaneus, specifically an inverse relationship regarding talar dome heightening and calcaneal shortening. While there was minimal navicular and cuboid shape variation, there were alignment variations within these joints; the most notable is the rotational aspect about the anterior-posterior axis. This study also found that multi-domain modeling may be able to predict joint space distance measurements within a population. Additionally, variation across a population of these four bones may be driven far more by morphology than by alignment variation based on all three joint measurements. These data are beneficial in furthering our understanding of joint-level morphology and alignment variants to guide advancements in ankle joint pathological care and operative treatments.

Category: Bunion; Midfoot/Forefoot Introduction/Purpose: First metatarsal (M1) axial rotation as measured by the metatarsal pronation angle (MPA) has increasingly become recognized as a clinically relevant component of hallux valgus deformity and methods to realign the M1 in 3 dimensions have been developed. Preoperative imaging with weightbearing computed tomography (WBCT) is considered the best way to image and measure the MPA, however WBCT is not available to many foot and ankle surgeons. The aim of this project is to investigate if the MPA measured on sesamoid view weight bearing radiographs is similar to that measured on WBCT. Methods: A retrospective analysis of patient imaging studies was performed. We identified 25 feet with both sesamoid WBR and WBCT images from a consecutive patient database. Included subjects had no trauma or foot and ankle operation between acquisition of the two scans. MPA of the WBCT for each patient was initially measured. After an appropriate washout period, the observer then measured MPA of the WBR for the cohort. Demographic statistics were calculated, a paired t-test was performed to evaluate for any significant difference in angular measurement between the two scans. Results: Patient characteristics: mean age 53 +- 16 years, mean BMI 32 +- 8 kg/m2, 13 right and 12 left feet, 19 females and 6 males. Average time between sesamoid view WBR and WBCT was 5.7 +- 9.5 months. Mean difference in MPA between WBR and WBCT = 0.6 +- 3.9 degrees (p= 0.444, t=0.777). Conclusion: Among the cohort assessed, we found no significant difference present in the measurement of the metatarsal pronation angle as evaluated on sesamoid view weightbearing radiography compared with weightbearing computed tomography (Figure 1). Routine acquisition of a sesamoid view weightbearing radiograph may be considered as an appropriate substitute for weightbearing computed tomography in the preoperative evaluation of M1 rotation in the coronal plane. These findings may benefit physicians who do not have access to weightbearing computed tomography in their clinics.

Category: Midfoot/Forefoot; Bunion Introduction/Purpose: Hyperpronation of the first metatarsal (M1) is present in Hallux Valgus (HV) but its impact is still unknown. A previous biomechanical study showed that an increase in hallucal pronation might lead to a medial soft tissue failure of the first metatarsophalangeal joint (MTP1). Conversely, an increase in supination and adduction of the first ray when weight-bearing is present in case of HV. The objective of our study was to sequentially answer the following questions: (1) Does an increase in M1 pronation cause an increase in hallucal pronation? (2) Can a combination of intrinsic M1 hyperpronation and MTP1 medial soft tissue failure induce a supination motion of the first ray during weight-bearing? (3) Can a first ray supination motion during weight- bearing be accompanied by an increase in IMA and HVA? Methods: A cadaveric model allowing a simulated standing position was developed and secured with a radiolucent frame (Figure 1). A midshaft osteotomy of M1 was performed allowing either 0° or 30° in pronation. MTP1 medial soft tissue release was performed to simulate failure. Twelve specimens underwent 6 Weight-Bearing CT acquisitions under different conditions listed below. The first 3 acquisitions had 0° pronation of M1:1. Simulated non-weight-bearing condition (Figure 2a) 2. Simulated weight- bearing condition (Figure 2b) 3. Simulated weight-bearing condition with medial soft tissue failure (Figure 2c). The next 3 WBCT acquisitions followed the same sequence but with 30° pronation of M1 (Figure 2d to 2f). On each WBCT acquisition, the HVA, IMA, Metatarsal Pronation Angle (MPA, M1 head pronation relative to the ground) and the hallucal pronation (HP) were measured (Figure 3). Motions were indirectly calculated from the differential values of these angular measurements produced by these 6 different conditions. Results: 1.The increase in MPA and HP induced by the 30° pronation osteotomy of M1 in simulated non weight bearing- conditions were respectively 27.6+/-4.1 and 25.5+/-5.6 degrees (p=0.202). 2.The first ray motion induced by weight-bearing without pronation osteotomy combined with a MTP1 medial soft tissue failure was 3.7+/-3.6 degrees pronation (differential value on MRA between Figure 2a and 2c) compared to 11+/-7.5 degrees supination after the 30 degrees pronation osteotomy of M1 combined with a MTP1 medial soft tissue failure (p

Abstract Historically, conventional radiographs have been the primary tool to morphometrically evaluate the talocrural joint, which is comprised of the distal tibia, distal fibula, and proximal talus. More recently, high-resolution volumetric imaging, including computed tomography (CT), has enabled the generation of three-dimensional (3D) reconstructions of the talocrural joint. Weightbearing cone-beam CT (WBCT) technology provides additional benefit to assess 3D spatial relationships and joint congruency while the patient is load bearing. In this study we applied statistical shape modeling, a computational morphometrics technique, to objectively quantify anatomical variation, joint level coverage, joint space distance, and congruency at the talocrural joint. Shape models were developed from segmented WBCT images and included the distal tibia, distal fibula, and full talus. Key anatomical variation across subjects included the fibular notch on the tibia, talar trochlea sagittal plane rate of curvature, tibial plafond curvature with medial malleolus prominence, and changes in the fibular shaft diameter. The shape analysis also revealed a highly congruent talocrural joint with minimal inter-individual morphometric differences at the articular regions. These data are helpful to improve understanding of ankle joint pathologies and to guide refinement of operative treatments.

Abstract. The American Board of Orthopaedic Surgery (ABOS) is the national organization charged with defining education standards for graduate medical education in orthopaedic surgery. The purpose of this article is to describe initiatives taken by the ABOS to develop assessments of competency of residents to document their progress toward the independent practice of orthopaedic surgery and provide feedback for improved performance during training. These initiatives are called the ABOS Knowledge, Skills, and Behavior Program. Web-based assessment tools have been developed and validated to measure competence. These assessments guide resident progress through residency education and better define the competency level by the end of training. The background and rationale for these initiatives and how they serve as steps toward competency-based education in orthopaedic residency education in the United States will be reviewed with a vision of a hybrid of time and competency-based orthopaedic residency education that will remain 5 years in length, with residents assessed using standardized tools.

Category: Other; Hindfoot; Midfoot/Forefoot Introduction/Purpose: The interplay between forefoot and hindfoot alignment has been well established in the literature for pathologies such as the flexible cavovarus foot. However, no study has evaluated the relationship between hindfoot alignment and first metatarsal axial rotation. Therefore, the purpose of this study is to determine the relationship, if any, between hindfoot alignment and first metatarsal axial rotation in patients with mild, moderate, and severe hindfoot varus/valgus deformity using weight bearing computed tomography (WBCT). We hypothesize that patients with hindfoot valgus alignment will be associated with a first metatarsal pronation/eversion deformity. Conversely, we hypothesize that patients with hindfoot varus alignment will be associated with a first metatarsal supination/inverted deformity. Methods: Patients who underwent weight bearing radiographs (WBR) and WBCT between 2015 and 2018 were evaluated with inclusion/exclusion criteria. Patients with prior foot surgery/trauma were excluded. Hindfoot alignment was evaluated using the calcaneal moment arm (Figure 1; horizontal distance between the most inferior aspect of the calcaneus and an extended line approximating the longitudinal axis of the tibia). Six subgroups were created based on the severity of hindfoot malalignment 1) Moderate valgus, 2) moderate varus, 3) substantial valgus, 4) substantial varus, 5) severe valgus, and 6) severe varus. Moderate was 1/2-1 standard deviations (SD) from the mean. Substantial was 1-2 SDs from the mean. Severe was >2 SDs from the mean. Patients with CMA values within ½ SD of the mean were excluded to focus on patients with significant hindfoot malalignment. First metatarsal axial rotation was measured using the Kim and Saltzman angles (Figure 1). Meary's angle was obtained using WBR. Results: 196 patients were included (average age - 52.6 years). The average CMA was +6.0+-16.2mm (valgus). The average Kim and Saltzman angles were 7.7+-12.9 and 2.8+-13.1 degrees (pronated). The average Meary's angle was 182.0+-11.9 degrees (>180 degrees-apex dorsal measurement). There was a significant association between hindfoot alignment and the Saltzman (r=0.641, p

Category: Trauma; Ankle; Other Introduction/Purpose: Posterior talar body fractures (AO/OTA 81.1.B/C) are rare injuries that present unique challenges in their access to the treating surgeon. Accessibility to this structure has been investigated extensively in the context of osteochondral lesion interventions, normally requiring perpendicular access to perform operative procedures. However, techniques in gaining this access regarding fracture repair, requiring only adequate visualization, has not been described in literature. Generally, a pre-operative decision is made between a posterior, soft-tissue based approach or a peri-articular osteotomy, which is associated with comparatively higher morbidity and complication rates. The aim of this study is to evaluate the accessible area of the talar dome via two standard posterior approaches (posteromedial; PM, and posterolateral; PL) with and without external fixator distraction. Methods: Eight male through-knee matched-paired cadaveric legs (mean age: 49.0 +- 14.6; mean BMI: 24.5+- 3.9 kg/m2) were included in this study. A standard PM or PL approach was performed using a randomized crossover design for surgical sequences. The accessible area without distraction was initially outlined by drilling a 1.6-mm Kirschner wire around the periphery of the visualized talus. Five millimeters of distraction, confirmed with fluoroscopy, was then applied to the specimens using an external fixator. The accessible area was again marked using the same method. The tali specimens were then explanted and imaged using a Micro-CT scanner to acquire 3 dimensional reconstructions. The accessible area was calculated as a percentage of the total talar dome surface area. The Mann-Whitney U test was used to compare the reported areas among the two surgical approaches, where the Wilcoxon signed rank test was utilized to compare values among distracted and non-distracted conditions. Results: In reference, the average total surface area of the talus is 16.94 +- 2.47 cm2. No statistically significant differences were found among match-paired specimens (p=0.63). The PM approach allowed access to 17.1% (11.1 to 23.6%, SD 5.4) of the talar dome surface without distraction and 29.3% (20.0 to 38.6%, SD 8.6) of the talar dome surface with distraction. The PL approach provided access to 7.4% (4.7 to 11.8%, SD 3.1) and 17.0% (11.0 to 26.1%, SD 6.5) of the talar dome surface with and without distraction, respectively. A statistically significant difference was observed in talar dome accessibility among distracted and non- distracted conditions in both surgical approaches (p=0.008). Additionally, the PM approach provided significantly more access to the talar dome relative to the PL approach (p=0.043). Conclusion: This matched-paired cadaveric study provides roadmap that can assist in the pre-operative planning of talar dome access in the treatment of talar body and posterior tubercle fractures. We found no advantage to a PL approach over a PM approach to access these challenging fractures. Additionally, added distraction using an external fixator consistently increased visualization of the talar dome by a magnitude of at least 40% greater than the non-distracted conditions. These methods can be applied clinically to gain appropriate access to the talar dome, allowing fracture repair.

Category: Midfoot/Forefoot; Other Introduction/Purpose: Osteoarthritis (OA) of the midfoot can elicit significant pain, disability, and decreased quality of life in affected patients. Therefore, correct diagnosis and appropriate, timely interventions towards this degeneration is essential. Currently, weightbearing radiography (WBR) has been widely accepted as the standard method in assessing midfoot OA. However, the complex structural anatomy of this region poses unique challenges in its visualization due to the obscurance caused by the overlapping osseous structures present when observed in a two-dimensional perspective. Weightbearing computed tomography (WBCT), providing a relatively clearer visualization of the midfoot could also be used for evaluation. This study aims to perform a single center, retrospective, intra-patient analysis identifying the discrepancy in midfoot OA diagnostic sensitivity among the two imaging modalities (WBCT vs. WBR). Methods: After gaining approval from an internal review board, a retrospective analysis of patient electronic health records was performed to assess the discrepancy of midfoot OA diagnostic sensitivity among WBCT and WBR. The radiological interpretation (RI) of 761 consecutive patient WBCT images were systematically assessed for OA diagnosed in 3 midfoot joint groups (Chopart group; 2 joints, midfoot group; 6 joints, Lisfranc group; 5 joints). A case was considered positive for OA if the WBCT RI contained any explicit mention of midfoot OA or any reported signs of the pathology including mild, moderate, or severe degeneration, joint space narrowing, osteophytic changes, spurring, cystic changes, or sclerosis in at least one of the aforementioned joint groups. After an appropriate washout period, the observer then synonymously reviewed the WBR RI of each positive OA case. A sensitivity metric was calculated for the OA incidence and severity discrepancy between the two imaging modalities. Results: A total of 302 (mean age, 56.0 +- 16.0 years; 140 left and 162 right) feet were assessed in this study. 244 cases of Chopart OA were detected on WBCT RI, where only 184 cases were detected on the WBR RI of the same feet, resulting in a diagnostic discrepancy of 24.6%. In the midfoot joint group, WBCT RI and WBR RI detected 96 and 67 cases of OA respectively, yielding a 30.2% discrepancy. Finally, in the Lisfranc joint group, the WBCT and WBR RI detected 174 and 129 cases of OA, respectively, showing a 25.9% discrepancy in diagnosis. Additionally, the OA severity was downgraded between modalities in 7.0%, 4.2%, and 8.0% for the Chopart, midfoot, and Lisfranc joint groups, respectively. Conclusion: After assessing 244 Chopart, 96 midfoot, and 174 Lisfranc cases of OA detected by WBCT, a significant discrepancy has been observed in the diagnostic abilities of midfoot OA among the two imaging modalities studied. With this difference realized, in addition to the recent higher availability, decreasing cost, and low radiation exposure offered by WBCT, this imaging modality may have the ability to aid physicians considerably in their clinical practice. Detecting and correctly diagnosing the severity of this pathology may lead to more appropriate treatment options, lower complication rates, and better patient outcomes.

Category: Midfoot/Forefoot; Trauma Introduction/Purpose: Chronic Lisfranc (tarsometatarsal, TMT) joint injuries often present with various clinical symptoms and radiographic findings. If severe post-traumatic arthritis is present, a salvage arthrodesis is an appropriate intervention that provides the favorable outcomes. However, this procedure severely decreases TMT range of motion, a necessary attribute in the complete resolution of functionality, which is desired by many high-demand patients. Conversely, patients with chronic Lisfranc injuries occurred greater than six weeks prior to their diagnosis can present with no radiographic signs of degeneration. For this patient cohort, there is no consensus on a specific operative intervention. Therefore, this article will provide a systematic review of the current surgical options and associated outcomes to guide future clinical decisions in caring for these patients with no degenerative sequelae. Methods: PubMed, ScienceDirect, Scopus, and Embase medical databases were searched from inception through March 5, 2019. Original studies that assessed the outcome of patients treated surgically for chronic Lisfranc injuries without secondary osteoarthritic changes were considered for inclusion. Only studies written in English and German were included. The following data were recorded from each study: number of patients, number of feet, patient age, mechanism of injury, type of injury (purely ligamentous or ligamentous with concomitant bony fractures), time between initial injury and operation, indications for surgery, pre-operative assessment, operative techniques, postoperative follow-up time, time to return to activity or sport, clinical outcome, and study design (prospective vs. retrospective, single vs. multicenter, level of evidence). The modified Coleman Score was used to assess the methodologic quality of included studies. Results: Of the 6,845 screened, ten studies met the above criteria and were analyzed. All inclusions, with the exception of one study, were performed at a single center and were retrospective or prospective case series. The surgical techniques used to treat chronic Lisfranc injuries without arthritis include 1) open reduction and internal fixation with various types of hardware, 2) Lisfranc ligament reconstruction (utilizing the gracilis tendon or half of the extensor hallucis longus tendon), and 3) arthrodesis using screws. Overall, the studies generally reported low complication rates and acceptable functional outcomes. These postoperative outcomes were most frequently measured with the American Orthopaedic Foot & Ankle Society (AOFAS) score. Of the applicable studies, the mean preoperative AOFAS score of 55.7 significantly improved to 88.1 at final follow-up. Conclusion: No definitive consensus remains on how Lisfranc instability without concurrent degenerative arthritis should be surgically managed. Despite the delay in diagnosis, this patient cohort still experiences improved patient outcomes and few post- surgical complications with the operative techniques currently available. While the quality of these studies was satisfactory, a larger patient cohort and prospective analysis could further strengthen arguments for or against certain surgeries.

Category: Trauma; Midfoot/Forefoot Introduction/Purpose: Subtle Lisfranc instability, a spectrum of conditions resulting from trauma of individual structures within the Lisfranc ligamentous complex (LLC) including the dorsal- (DLL), interosseous- (ILL), and plantar Lisfranc ligaments (PLL) remain challenging to effectively diagnose. The current standard to asses these injuries is through bilateral weightbearing radiography. However, weight bearing computed tomography (WBCT), which provides a clearer visualization of osseous structures, joint spaces, and instability could also be utilized for evaluation. This study aims to define and report the occurrence rate of an ‘asymmetric lambda sign’ observed on axial WBCT imaging secondary to simulated purely ligamentous Lisfranc injuries of various severities subjected to increasing magnitudes of weightbearing conditions. It is hypothesized that this sign will be a reliable aid in detecting subtle Lisfranc instability, clinically. Methods: The asymmetric lambda sign was assessed on 24 match-paired cadaveric legs (mean age, 46.0 +- 14.8 years; mean weight, 80.2 +- 17.4 kg; mean body mass index, 25.3 +- 4.2 kg/m2). Dissection groups were described as follows: control intact LLC (group 1), dissection of the DLL (group 2), dissection of the DLL and ILL (group 3), dissection of the DLL, ILL, and PLL (group 4). After each dissection, CT scans were acquired in non- (NWB, 0 kg), partial- (PWB, 40 kgs), and full-weightbearing (FWB, 80 kgs) conditions. In a standardized axial view, the lambda sign was appreciated by visualizing the joint spaces between the medial cuneiform and the second metatarsal base (C1-M2), the medial and middle cuneiform (C1-C2), and the second metatarsal base and middle cuneiform (M2-C2). The asymmetric lambda sign was defined by a widening of the C1-M2 joint space relative to the C1-C2 or M2-C2 joint space (Fig). Results: The asymmetric lambda sign was observed in 25.6% (221/864) of all studies. This sign was not found in any intact specimens regardless of the weightbearing status. In group 2, this sign was observed in 4.2% of NWB, 15.3% of PWB, and 16.7% of FWB conditions. With the DLL and ILL dissected, simulating a more developed subtle LLC injury (group 3), this sign was observed in 15.3%, 27.8%, and 38.9% in NWB, PWB and FWB conditions, respectively. Additionally, the fully dissected specimens (group 4), demonstrated this specific sign in 33.3%, 72.2%, and 83.3% in NWB, PWB, and FWB conditions, respectively. The inter- and intra- observer reliability was calculated to a kappa value of .843 and .912. Conclusion: An asymmetric lambda sign viewed through axial WBCT imaging is a reliable indicator in the diagnosis of a complete Lisfranc ligamentous injury with secondary joint instability in partial and full weightbearing conditions as demonstrated using a cadaveric model. Clinically, this simple sign can be utilized as an aid to accurately evaluate and diagnose patients presenting with subtle Lisfranc instability.

Category: Trauma; Midfoot/Forefoot Introduction/Purpose: Introduction: The incidence of Lisfranc injury is higher than previously estimated, and more than two- thirds of these injuries are caused by low-energy trauma. This low-energy mechanism, common to the athletic population, tends to be distinctly characterized by a primarily ligamentous Lisfranc injury. Bilateral weightbearing radiographs are widely accepted as the gold standard and are able to diagnose instability by contralateral comparison. However, contrary to radiography, CT imaging is not obscured by metatarsals overlapping each other and their adjacent tarsal bone and thus provides a better visualization of bony structure. This study’s purpose was to model a purely ligamentous Lisfranc joint injury to determine widening between the medial cuneiform (C1) and second metatarsal (M2) after each ligament dissection under non-weightbearing (NWB, 0 kg), partial weightbearing (PWB, 40 kgs.), and full weightbearing (FWB, 80 kgs.) conditions evaluated by weightbearing computed tomography (WBCT). Methods: Methods: Twelve paired through-knee cadavers were incised dorsally to visualize the Lisfranc joint complex. An intact ligament condition was the control (group 1). The dorsal ligament (group 2), interosseous ligament (group 3), and plantar ligament (group 4) were sequentially dissected. The twelve pairs were then equally randomized (groups 5a, 5b, and 5c). The first tarsometatarsal joint capsule (group 5a), the second tarsometatarsal joint capsule (group 5b), or the intercuneiform ligament (group 5c) was transected. The remaining intact ligaments of each subgroup were similarly randomized for transection (e.g., group 6ac: transection of the first tarsometatarsal joint, followed by transection of the intercuneiform ligament). Finally, the last ligament was transected (group 7). After each dissection, CT scans were taken for NWB, PWB, and FWB conditions. Distance between the lateral border of C1 and the medial border of M2 was used to evaluate diastasis for both coronal and axial views. Analysis of variance (ANOVA) tested for significant differences between all groups. Dependent t-test evaluated for significance of the weightbearing variable. Results: In group 1, C1-M2 distance was 3.9 mm (3.5-4.3, SD 0.3) for coronal imaging and 4.0 mm (3.6-4.5, SD 0.4) for axial imaging, and no statistically significant differences were found between paired feet (P = 0.61) or for weightbearing conditions (P = 0.34). In group 4, an average widening of 1.4-mm (0.8-2.1 mm, SD 0.4) and 1.8-mm (1.4-2.8 mm, SD 0.5) were observed for PWB and FWB in the coronal view, and an average widening of 1.5-mm (0.9-2.3 mm, SD 0.5) and 2.0-mm (1.4-3.0 mm, SD 0.5) were observed for PWB and FWB in the axial view, respectively. For the subsets of group 5, only 5c was found to have a majority of specimen (7/8, 87.5%) that had significant widening - at least 2 mm - between C1-M2 for the PWB condition in the axial view; for the FWB condition, significant widening was found in both the axial and coronal view. In groups 6ca, 6cb, and 7, a significant widening - at least 2 mm - was measured in both planes for the NWB condition. Conclusion: WBCT scans improve the ability to detect incomplete ligamentous Lisfranc injuries by comparing C1-M2 distance to the uninjured side. Widening exceeding 1 mm more than the normal contralateral side for the PWB condition could clinically indicate complete Lisfranc ligament injury (sensitivity 83.3%, specificity 88.9%, accuracy 87.5%).

Abstract Background This study evaluated the responsiveness of several PROMIS patient-reported outcome measures in patients with hand and upper extremity disorders and provided comparisons with the qDASH instrument. Methods The PROMIS Upper Extremity computer adaptive test (UE CAT) v1.2, the PROMIS Physical Function (PF) CAT v1.2, the PROMIS Pain Interference (PI) CAT v1.1 and the qDASH were administered to patients presenting to an orthopaedic hand clinic during the years 2014–2016, along with anchor questions. The responsiveness of these instruments was assessed using anchor based methods. Changes in functional outcomes were evaluated by paired-sample t-test, effect size, and standardized response mean. Results There were a total of 255 patients (131 females and 124 males) with an average age of 50.75 years (SD = 15.84) included in our study. Based on the change and no change scores, there were three instances (PI at 3 months, PI >3 months, and qDASH >3 months follow-ups) where scores differed between those experiencing clinically meaningful change versus no clinically meaningful change. Effect sizes for the responsiveness of all instruments were large and ranged from 0.80–1.48. All four instruments demonstrated high responsiveness, with a standardized response mean ranging from 1.05 to 1.63. Conclusion The PROMIS UE CAT, PF CAT, PI CAT, and qDASH are responsive to patient-reported functional change in the hand and upper extremity patient population.

Category: Ankle Introduction/Purpose: Syndesmotic ankle injuries are challenging to diagnose, since current 2D imaging techniques try to quantify a 3D displacement. Therefore, our aim was two-fold: to determine displacement of sequential syndesmotic ankle injuries under various amounts of load using a 3D weightbearing CT (WBCT) and to assess the relation with current 2D imaging. Methods: Seven paired male cadaver specimens were included (tibia plateau to toe-tip) and mounted into a custom-built frame. WBCT scans were obtained after different patterns of load (0 kg or 85 kg) were combined with torque (0 Nm or 10 Nm external rotation). These conditions were repeated after each ligament condition: intact ligaments, sequential sectioning of the anterior inferior tibiofibular ligament (AITFL), deltoid ligament (DL), and interosseous membrane (IOM). CT images were segmented to obtain 3D models. These allowed quantification of displacement based on the position of computed anatomical landmarks in reference to the intact position of the fibula. A correlation analysis was performed between the 2D and 3D measurements. Results: The effect of torque caused significant displacements in all directions (P0.05). Weight caused a significant lateral (mean=-1.4 mm, SD=1.5) and posterior translation (mean=-0.6 mm, SD=1.8). The highest displacement consisted of external rotation (mean=-9.4°, SD=6.5) and posterior translation (mean=6.1 mm, SD=2.3) after IOL sectioning combined with torque (Fig. 1). Pearson correlation coefficients were moderate (range 0.31-0.51, P

Category: Hindfoot, Trauma Introduction/Purpose: Accurate assessment of subtalar joint stability is difficult. Weightbearing CT (WBCT) scans have increased in popularity among foot and ankle surgeons as it offers an accurate representation of hindfoot joint alignment under weightbearing conditions. However, the clinical utility of this imaging modality to diagnose subtalar joint instability has yet to be investigated. The purpose of this study was to assess the impact of load and torque application on the assessment of subtalar joint instability when using WBCT scans. We hypothesized that load and torque application would improve the ability to identify subtalar joint instability. Methods: Fourteen paired male cadavers (tibial plateau to toe-tip) were included. A radiolucent frame held specimens in a plantigrade position while non-weightbearing and weightbearing CT scans (with and without torque application) were taken. First, intact ankles (Native) were scanned. Second, one specimen from each pair underwent interosseous talo-calcaneal ligament (ITCL) transection (Condition 1A), while the contralateral underwent calcaneo-fibular ligament (CFL) transection (Condition 1B). Third, the lesions were reversed on the same specimens and the remaining intact ITCL or CFL was transected (Condition 2). Finally, the deltoid ligament was transected in all ankles (Condition 3). Two radiographic measurements were performed on the level of the ankle joint (talar tilt [TT], anterior talar translation [ATT]), while the subtalar tilt (STT) was measured at the anterior, middle, and posterior plane of the subtalar joint. Inter- and intra-observer agreement calculations were additionally performed. Results: Weight did not impact most measurements, while the majority of measurements differed significantly within each condition after torque application. Measurements performed at the level of the subtalar joint were more reliable and better predictors for subtalar joint instability compared to measurements performed at the level of the ankle joint. Discrete ITFL, combined ITFL/ CFL, and combined ITFL/ CFL/ deltoid ligament injuries had significantly different STT values than native ankles (Figure 1). Conclusion: Measurements to assess the stability of the subtalar joint should primarily be performed at the level of the subtalar joint rather than at the level of the ankle joint when using WBCT scans. Torque application is crucial for identifying subtalar joint instability, while the application of load should be avoided. Future clinical studies will therefore face substantial technical challenges should they attempt to determine the clinical utility of WBCT scans for assessment of subtalar joint instability.

Category: Ankle, Basic Sciences/Biologics, Trauma Introduction/Purpose: Injury to the distal tibio-fibular syndesmosis can occur as an isolated ligamentous injury or can be associated with bony ankle trauma. The missed diagnosis of such injuries is believed to lead to chronic pain and early degeneration. Weightbearing CT (WBCT) has increased in popularity among foot and ankle surgeons as it offers an accurate representation of joint alignment under weightbearing conditions. However, the use of this imaging modality to diagnose syndesmotic injuries has not been investigated. The purpose of this study was to assess the influence of torque on the assessment of syndesmotic injuries using axial CT images under weightbearing conditions. We hypothesized that torque application would improve the ability to detect syndesmotic injuries. Methods: Seven pairs of male cadavers (tibia plateau to toe-tip) were included. CT scans with axial load application (85 kilograms, kg) and with (10 newton meters, Nm) or without torque to the tibia (corresponding to external rotation of the foot and ankle) were taken during four test conditions. First, intact ankles (Native) were scanned. Second, one specimen from each pair underwent anterior inferior tibio-fibular ligament (AITFL) transection (Condition 1A) while the contralateral underwent deltoid transection (Condition 1B). Third, the lesions were reversed on the same specimens and the remaining intact deltoid or AITFL was transected (Condition 2). Finally, the distal tibiofibular interosseous membrane (IOM) was transected in all ankles (Condition 3). Eight different measurements were performed to assess the integrity of the distal tibio-fibular syndesmosis on axial CT scans. Results: Torque impacted axial CT scan measurements in almost all conditions (Table I). The ability to diagnose syndesmotic injuries using axial CT measurements improved when torque was applied. No significant syndesmotic morphological change was observed with or without torque for either isolated AITFL or deltoid ligament transection. Conclusion: Low magnitude torque rotational stress application had a notable impact on 2D measurements used to diagnose syndesmotic injuries with axial CT scans (cadaveric model). Two dimensional measurements are not effective to distinguish between single AITFL and single deltoid ligament injuries. Because weightbearing conditions allow for a standardized positioning of the foot while radiographs or CT scans are taken, a combination of axial load and torque application is desirable. Further studies including patients with syndesmotic injuries will be necessary to clarify the clinical potential of WBCT scans with torque applied to identify syndesmotic widening.

Category: Hindfoot Introduction/Purpose: Patients with a hindfoot deformity pose a particular challenge when performing a total knee arthroplasty (TKA). This could be attributed to the lack of insights concerning hindfoot alignment outcome. Our objective was to perform a systematic review of the literature to investigate the influence of TKA on hindfoot alignment and vice-versa. Methods: This systematic review was performed in accordance to PRISMA guidelines; the original protocol is registered on PROSPERO (ID: 106980). The following electronic databases were searched to identify capable studies: Pubmed, EMBASE, Web of Science, Google Scholar, and Cochrane Library. To ensure the quality of the review, a quality assessment was performed according to the MINORS criteria. The mean MINORS score was 9.5 (range 7-12) and concurred with a good quality of the obtained studies. Results: Sixteen identified articles met the review criteria: six prospective cohort studies, nine retrospective studies, and one case-control study. Five articles reporting on clinical outcomes were identified; a pre-operative hindfoot deformity negatively influenced TKP outcome (N=4) or demonstrated no influence (N=1). Thirteen articles reporting on radiographic outcomes were identified. Five studies reported a significant improvement (P < 0.05) of hindfoot alignment after TKA, two studies demonstrated improvement limited to valgus hindfeet, seven studies observed a significant (P < 0.05) correlation between knee and hindfoot alignment, and two studies reported the association between hindfoot OA in patients undergoing TKA. Two studies reporting biomechanical outcomes were identified; these reported a shift of the weighbearing axis after TKA as a result of the remaining hindfoot deformity. Conclusion: TKA can improve the alignment of valgus hindfeet, but improvement was less pronounced in varus hindfeet. Furthermore, a worse clinical outcome is expected after TKA in patients with concomitant ankle OA. A remaining hindfoot deformity shifts the weightbearing axis, which could shorten the survival of the total knee prosthesis.

Category: Ankle, Trauma, Imaging Introduction/Purpose: Injury to the distal tibio-fibular syndesmosis is common and appears in up to 20% of patients with an ankle sprain or ankle fracture. While pronounced injuries can be reliably diagnosed using conventional radiographs, assessment of subtle syndesmotic injuries is challenging. With the introduction of weightbearing CT (WBCT) scans, detailed assessment of foot and ankle disorders under load bearing conditions became possible. The purpose of this cadaver study was to assess the influence of weight on assessment of incomplete and more complete syndesmotic injuries using two-dimensional (2D) measurements on axial CT images. We hypothesized that weight would significantly impact assessment of both incomplete and more complete injuries to the distal tibio-fibular syndesmosis. Methods: Fourteen paired male cadavers (tibial plateau to toe-tip) were included. A radiolucent frame held specimens in a plantigrade position while both non-weightbearing and weightbearing CT scans were taken. Four conditions were tested: First, intact ankles (Native) were scanned. Second, one specimen from each pair underwent anterior inferior tibio-fibular ligament (AITFL) transection (Condition 1A), while the contralateral underwent deltoid transection (Condition 1B). Third, the remaining intact deltoid or AITFL was transected (Condition 2). Finally, the distal tibio-fibular interosseous membrane (IOM) was transected in all ankles (Condition 3). For each condition, non-weightbearing, half-bodyweight (42.5 kg), and full-bodyweight (85 kg) CT scans were taken. Six measurements were performed to assess the integrity of the distal tibio-fibular syndesmosis on axial CT scans 1 cm above the ankle joint (Figure 1A/ B) and two measurements at the level of the talar surface (Figure 1C). Inter- and intra- observer agreement were additionally calculated. Results: Inter- and intra-observer agreement differed between measurements. Excellent agreement was evident for the tibio- fibular clear space (TFCS) and tibio-fibular overlap (TFO) with an intra-observer agreement of 0.79 and 0.94, respectively. Poor agreement was evident for Angle 1 (inter-observer, 0.39). Agreement of the other measurements (inter- and intra-observer) was either rated as fair or good and ranged from 0.44 to 0.71. Weightbearing had no significant influence on measurements. Only more complete injuries (Condition 3) differed from native ankles when using either the anterior tibio-fibular distance (ATFD) or TFO. For these two measurements, no significant differences were observed within each condition between non-, half-, and full- weightbearing. Also, no significant differences were evident between single AITFL and deltoid ligament transection for the ATFD and TFO. Conclusion: Load application does not impact the ability of WBCT scans to diagnose incomplete and also more complete syndesmotic injuries in a cadaver model. The utility of current 2D measurements on axial WBCT scans for diagnosing incomplete syndesmotic injuries is questionable. Nevertheless, the ability to reliably position the foot during imaging is an advantage of WBCT technology over other imaging options. Further investigations using more precise measurement options (e.g. 3-dimensional [3D] measurements) are necessary to better understand the potential role of weight bearing to diagnose syndesmotic injuries with CT scan imaging technology.

Category: Ankle, Ankle Arthritis Introduction/Purpose: Treatment options for ankle arthritis in younger patients are currently limited. Since the longevity of modern total ankle replacements is not sufficient for this patient population, ankle arthrodesis is typically utilized when joint preserving treatment is not a viable option. A new procedure using a pyrocarbon ankle spacer has been developed as a potential alternative, allowing for talar articular resurfacing for pain relief with minimal bone resection. The objective of this study was to assess whether this pyrocarbon ankle spacer could provide normal ankle kinematics as the native ankle joint using cadaveric gait simulation. Methods: Five mid-tibia cadaveric specimens without deformity and no history of lower limb injury or surgery were utilized. The stance phase of gait was simulated for each specimen using a six degree-of-freedom robotic device. A force plate was moved relative to stationary specimen through an inverse tibial kinematic path calculated from in vivo data while extrinsic tendons were actuated using physiologic loads (Figure 1A). Magnitudes of load were scaled to that of 25% bodyweight. Ankle kinematics were measured from reflective markers attached to the tibia and talus via surgical pins. The pyrocarbon ankle spacer (Exactech, Gainesville, FL, USA) was implanted in a nest formed 3-4 mm in depth on the talar articular surface using a custom burring technique (Figure 1B). Ankle spacer kinematics were compared to 95% confidence intervals of native, intact ankle joint kinematics to assess agreement. Results: Outcomes revealed no significant difference in ankle joint kinematics between the native, intact condition and post- pyrocarbon spacer implantation (Figure 1C). This result was consistent for the sagittal, coronal and axial planes of motion. Conclusion: The results of this study demonstrate that a pyrocarbon spacer permits normal ankle kinematics. Further, the device was observed to be stable in the joint throughout simulations. While the testing was performed at 25% bodyweight for analyses on all specimens, load magnitudes were also increased up to 75% on a subset of specimens and the structural integrity of the device remained pristine. With these findings, we concluded that the pyrocarbon spacer device offers promising potential as a treatment option for ankle arthritis.

Category: Ankle Introduction/Purpose: Measurements of joint angles and translations (i.e. kinematics) are essential to understand the pathomechanics of ankle disease and functional changes following treatment. Traditional motion capture techniques, which track the positions of reflective markers adhered to the skin, cannot measure motion of the tibiotalar and subtalar joints independent of one another. To overcome this limitation, we used high-speed dual fluoroscopy (DF), an x-ray videography technique, to quantify in-vivo kinematics of healthy asymptomatic ankles during activities of daily living. Using these kinematics as baseline data, our secondary objective was to assess preliminary kinematic differences between chronic ankle instability (CAI) patients and asymptomatic control subjects. Methods: High-speed DF images of the hindfoot of ten healthy, asymptomatic adults and four adults with CAI were acquired during treadmill walking at 0.5 m/s and 1.0 m/s and during a single-leg, balanced heel-rise. Three-dimensional (3D) CT models of the calcaneus, tibia, and talus and DF images served as input to the validated model-based markerless tracking software that quantified in vivo kinematics for the tibiotalar and subtalar joints. Dynamic joint kinematics and mean range of motion (ROM) were calculated and reported as dorsi/plantarflexion (D/P), inversion/eversion (In/Ev) and internal/external rotation (IR/ER) angles or translations along the medial/lateral (ML), anterior/posterior (AP), and superior/inferior (SI) directions. Results: During gait, the tibiotalar joint had significantly greater D/P ROM than the subtalar joint (0.5 m/s: p=0.004; 1.0 m/s: p=0.003). The subtalar joint had significantly greater In/Ev (0.5 m/s: p < 0.001; 1.0 m/s: p < 0.001) and IR/ER (0.5 m/s: p=0.01; 1.0 m/s: p=0.02) ROM than the tibiotalar joint. However, during balanced heel-rise, D/P and In/Ev were significantly different between the two joints (p < 0.001; p < 0.001). For AP translation, subtalar ROM was significantly greater than tibiotalar ROM during walking at 0.5m/s (p=0.002). CAI patients often demonstrated rotational profiles with dynamic trends that fell outside the 95% confidence intervals of the asymptomatic subjects (Figure 1). CAI patients exhibited smaller ROM than asymptomatic subjects. However, only 0.5 m/s tibiotalar SI translational (p=0.049) and 1.0 m/s subtalar In/Ev (p=0.03) ROM were significant. Conclusion: To our knowledge, this is the first study to quantify in-vivo joint angles and translations in asymptomatic and CAI subjects. Our results support the belief that the tibiotalar joint is primarily responsible for D/P, while the subtalar joint facilitates In/Ev and IR/ER. Secondary rotational contributions suggest that both joints undergo complex, 3D motion. Our comparison of CAI and asymptomatic subjects is not conclusive, yet suggests that a larger sample size will detect significant differences. With a larger sample size, dual-fluoroscopy may provide insight into the clinical relevance of altered kinematics and the pathomechanics responsible for ankle instability and other pathologies.

Category: Ankle Arthritis Introduction/Purpose: The current standard treatment for failed ankle arthrodesis is a revision ankle arthrodesis. However, there is limited literature addressing postoperative outcomes in patients with revision tibiotalar arthrodesis. The objectives of the retrospective study were to compare (1) demographics, (2) surgical technique, (3) postoperative fusion rates, and (4) postoperative complication rates in patients with primary vs. revision tibiotalar arthrodesis. Methods: Between March 2002 and November 2014, 455 ankle arthrodeses were performed in our institution. There were 234 male and 221 female patients with a mean age of 55.6 ± 15.1 years (18.0-88.8). The mean weight, height, and body mass index (BMI) were 90.3 ± 21.3 kg (46-218), 172.8 ± 13.6 cm (147-208), and 30.2 ± 6.3 kg/m2 (18.9-61.7), respectively. Both patient groups were compared with regard to demographics including gender, weight, height, BMI, ASA classification, smoking, alcohol use, and comorbidities. The surgical technique has been analyzed in both groups including surgical approach, main fixation type, allograft/autograft use, and use of bone morphogenic protein. Finally, fusion rate and time to complete osseous fusion were analyzed. Complication rates including wound complications, deep vein thrombosis/pulmonary embolism, and any secondary procedures were described in both groups. The mean time to final follow-up was 38.3 ± 27.4 months (12.0-150.4). Results: There were 385 and 70 patients with primary and revision arthrodesis, respectively. Demographics and comorbidities were comparable in both groups. All revision surgeries were open procedures, while 63 of 385 primary ankle arthrodeses were performed arthroscopically. The most common main fixation type in patients with primary ankle arthrodesis was a screw construct, while plates were most common for revisions (P < 0.001). The use of autograft was comparable in both groups (P = 0.886), however allograft was used more frequently in the revision arthrodesis group (P < 0.001). The rate of osseous union was comparable in both groups with 90.4% and 91.4% in patients with primary and revision arthrodesis, respectively (P = 0.735). The complication rate was comparable in both groups. Conclusion: The osseous union rates and complication rates were comparable in both patient groups, with primary and revision ankle arthrodesis.

Category: Trauma Introduction/Purpose: Although choices physicians make profoundly impact the cost of healthcare, few surgeons know actual costs. Without valid cost information, surgeons cannot understand how their choices impact the total cost of care. We leveraged a validated value analytics framework to efficiently allocate clinical care costs to individual patient encounters in an effort to understand the sources and variation of cost of care for a putatively straightforward and common orthopaedic problem. Methods: We conducted a retrospective cost analysis on all isolated, operatively treated ankle fractures from a Level 1 trauma hospital and affiliated outpatient surgery center between 2013 and 2015. Patients were categorized based on whether they were treated on an inpatient or outpatient basis, and records were reviewed to determine the presence of confounding variables as well as readmission and emergency department (ED) visits within 90 days after surgery. Actual costs were determined using a validated episode of care costing system and analyzed using multivariate regression analysis. Results: 148 patients (61 inpatients, 87 outpatients) with isolated, operatively treated ankle fractures were included. After controlling for confounding variables, outpatient care was associated with 31.6% (95% CI: 19.8% - 41.8%) lower costs compared to inpatient care. Obese patients had 21.6% (95% CI: 5.8% - 39.8%) higher costs compared to patients who were not obese. There was no difference in reoperation, readmission or return visits to the ED for patients treated on an inpatient or outpatient basis. Conclusion: Inpatient surgical care is clearly more expensive than outpatient care primarily due to higher facility and labor costs without a clear advantage relative to lower readmission or ER visit rates. Where medically appropriate, this analysis suggests ankle fracture surgery should be provided in an outpatient surgical facility to provide the greatest value to the patient and society.

Category: Ankle Arthritis Introduction/Purpose: In the last decades, different surgical techniques with various approaches and fixation methods have been described for ankle arthrodesis. Tibiotalar arthrodesis can be performed with or without distal tibiofibular fusion. The objectives of this retrospective study were to compare (1) demographics, (2) surgical techniques, (3) postoperative fusion rates, and (4) postoperative complication rates in patients with primary open tibiotalar arthrodesis with vs. without distal tibiofibular fusion. Methods: Between March 2002 and November 2014, 322 primary open ankle arthrodeses were performed at our institution. There were 183 male and 139 female patients with a mean age of 56.0 ± 14.0 years (18.0-88.8). The mean weight, height, and body mass index (BMI) were 90.0 ± 20.4 kg (46-168), 172.9 ± 11.4 cm (147-208), and 30.0 ± 5.7 kg/m2 (18.9-54.9), respectively. Both patient groups were compared with regard to demographics including gender, weight, height, BMI, ASA classification, smoking, alcohol use, and comorbidities. The surgical technique has been analyzed in both groups including surgical approach, main fixation type, and allograft/autograft use. Finally, fusion rate and time to complete osseous fusion were analyzed. Complication rates including wound complications, deep vein thrombosis/pulmonary embolism, and any secondary procedures were described in both groups. The mean time to final follow-up was 36.7 ± 26.7 months (12.0-150.4). Results: 214 had a combined distal tibiofibular fusion, while 108 did not. The most common surgical approach was lateral and anterior in patients with and without distal tibiofibular fusion, respectively (P < 0.001). The main fixation type was different between groups, with the most common technique being screws for patients with tibiofibular fusion, and plates in those without (P < 0.001). Autograft and allograft were used significantly less frequently in patients without distal tibiofibular fusion. The rate of osseous union was comparable in both groups with 92.2% and 93.0% in patients with and without distal tibiofibular fusion, respectively (P = 0.675). The incidence of wound and thrombembolic complications was similar in both groups. Conclusion: The osseous union rates and complication rates were comparable in both patient groups, with and without distal tibiofibular fusion.

Category: Ankle Arthritis Introduction/Purpose: Previous hindfoot surgeries present a unique challenge in performing hindfoot arthrodesis. The use of a blade plate construct is widely accepted, however there is limited data supporting the use of a posterior approach to blade plate arthrodesis. The purpose of this study was to (1) describe demographics of patients who underwent posterior hindfoot arthrodesis using a blade plate, (2) describe our surgical technique, (3) discuss outcomes, and (4) compare patients with and without complications. Methods: Between December 2001 and July 2014, 42 patients underwent hindfoot arthrodesis using a posterior blade plate and 40 patients were included in this study. Demographic data including age, gender, body mass index, smoking status, and comorbidities were analyzed. Surgical data including indication for the surgery, previous surgical treatment, and additional surgical procedures were reviewed. Weight-bearing radiographs were used to assess the fusion rate. Clinic and surgery notes were reviewed for possible intraoperative, perioperative, and postoperative complications. Univariate analysis was performed to compare patients who experienced complications with those who did not. There were 27 male and 13 female patients with a mean age of 56.4 ± 13.4 years. Twenty-eight patients had a tibiotalocalcaneal arthrodesis in a primary (n=6), primary staged (n=10), revision (n=9), or revision staged (n=3) setting. Eleven patients had ankle arthrodesis (primary n=7, revision n=4). The latest follow-up averaged 46.5 ± 27.5 months (range, 13.7-137.2 months). Results: Patients had a median of two previous hindfoot or ankle surgeries (range, 0-9 surgeries). Thirty-three of 40 (82%) procedures fused at an average of 24.4 ± 21.2 weeks. Four patients had a delayed osseous union. Seven patients had a nonunion, including ankle (n=3), subtalar (n=3), and both (n=1) joints. Patient groups with, and without primary solid osseous unions were comparable in terms of demographic data and surgical details. Eighteen major and eight minor complications were observed. Patients with or without complications were comparable in terms of demographic data and surgical characteristics. In total six patients (15%) underwent below knee amputation due to unsatisfactory results. Conclusion: Indications for hindfoot arthrodesis using posterior blade plate fixation include a diverse patient population. These surgeries may be performed as primary, revision, primary staged, or revision staged procedures. Most of the patients in our cohort had previous ankle/hindfoot surgeries. The fusion rate is lower than in primary hindfoot arthrodesis as reported in the current literature. The complications rate is high.

Category: Ankle Arthritis Introduction/Purpose: Ankle arthrodesis is the gold-standard treatment option for patients with end-stage ankle osteoarthritis resulting in measurable improvements in postoperative pain relief. Arthroscopic ankle arthrodesis has gained increasing popularity, however there remains a lack of comparative data with open ankle arthrodesis. The objectives of this retrospective study were to compare (1) demographics, (2) surgical technique, (3) postoperative fusion rates, and (4) postoperative complication rates in patients with primary open vs. arthroscopic tibiotalar arthrodesis. Methods: Between March 2002 and November 2014, 385 primary ankle arthrodeses were performed at our institution. There were 212 male and 173 female patients with a mean age of 56 ± 14.7 years (18.0-88.8). The mean weight, height, and body mass index (BMI) were 89.4 ± 19.7 kg (46-168), 172.9 ± 11.1 cm (147-208), and 29.9 ± 5.8 kg/m2 (18.9-54.9), respectively. There were 322, and 63 patients with primary open and arthroscopic arthrodesis, respectively. Both patient groups were compared with regard to demographics including gender, weight, height, BMI, ASA classification, smoking, alcohol use, and comorbidities. The surgical technique was analyzed in both groups including approach, main fixation type, and allograft/autograft use. Finally, fusion rates and time to complete union were analyzed. Complication rates including wound complications, deep vein thrombosis/pulmonary embolism, and any secondary procedures were described in both groups. The mean time to final follow-up was 37.9 ± 27.0 months (12.0-150.4). Results: Demographics and comorbidities were comparable in both groups. All but one arthroscopy were performed using anterior portals. All arthroscopic ankle arthrodeses were performed using screw fixation, while fixation types varied in open arthrodesis (P < 0.001). Autograft use was higher in open arthrodesis at 84.2% vs. 6.3% (P < 0.001), while allograft was favored for arthroscopic arthrodesis at 66.7% vs. 20.2% (P = 0.101).Osseous union was with 92.2% and 90.5% in patients with open and arthroscopic arthrodesis, respectively (P = 0.529). However, the time to complete osseous fusion was significantly shorter in patients with arthroscopic ankle fusion, at 4.3 vs. 5.1 months (P = 0.034). Wound and thrombembolic complications occurred significantly more often in patients with open ankle arthrodesis. Conclusion: Osseous union rates were comparable in both patient groups,however union time was significantly shorter in the arthroscopic group. Wound healing problems and thrombembolic complications were more common in patients who underwent open ankle arthrodesis.