Lucas F. Fonseca

Flexible Adult Acquired Flatfoot Deformity: Comparison Between Weight-Bearing and Non-Weight-Bearing Measurements Using Cone-Beam Computed Tomography

Background: The 3-dimensional nature of adult acquired flatfoot deformity can be challenging to characterize using radiographs. We tested the hypothesis that measurements on weight-bearing (WB) cone-beam computed tomography (CT) images were more useful for demonstrating the severity of the deformity than non-weight-bearing (NWB) measurements. Methods: We prospectively enrolled 12 men and 8 women (mean age, 52 years; range, 20 to 88 years) with flexible adult acquired flatfoot deformity. The subjects underwent cone-beam CT while standing (WB) and seated (NWB), and images were assessed in the sagittal, coronal, and axial planes by 3 independent observers who performed multiple measurements. Intraobserver and interobserver reliabilities were assessed with the Pearson or Spearman correlation and the intraclass correlation coefficient (ICC), respectively. Measurements were compared using paired Student t tests or Wilcoxon rank-sum tests. P < 0.05 was considered significant. Results: We found that overall the measurements had substantial intraobserver and interobserver reliability on both the NWB images (mean ICC, 0.80; range, 0.49 to 0.99) and the WB images (mean ICC, 0.81; range, 0.39 to 0.99). Eighteen of 19 measurements differed between WB and NWB cone-beam CT images, with more pronounced deformities on the WB images. The most reliable measurements, based on intraobserver and interobserver reliabilities and the difference between WB and NWB images, were the medial cuneiform-to-floor distance, which averaged 29 mm (95% confidence interval [CI] = 28 to 31 mm) on the NWB images and 18 mm (95% CI = 17 to 19 mm) on the WB images, and the forefoot arch angle (mean, 13° [95% CI = 12° to 15°] and 3.0° [95% CI = 1.4° to 4.6°], respectively) in the coronal view and the cuboid-to-floor distance (mean, 22 mm [95% CI = 21 to 23 mm] and 17 mm [95% CI = 16 to 18 mm], respectively) and the navicular-to-floor distance (mean, 38 mm [95% CI = 36 to 40 mm] and 23 mm [95% CI = 22 to 25 mm], respectively) in the sagittal view. Conclusions: Measurements analogous to traditional radiographic parameters of adult acquired flatfoot deformity are obtainable using high-resolution cone-beam CT. Compared with NWB images, WB images better demonstrated the severity of osseous derangement in patients with flexible adult acquired flatfoot deformity. Level of Evidence: Diagnostic Level II. See Instructions for Authors for a complete description of levels of evidence.

Metal artifact reduction MRI of total ankle arthroplasty implants

AbstractObjectivesTo assess high-bandwidth and compressed sensing-(CS)-SEMAC turbo spin echo (TSE) techniques for metal artifact reduction MRI of total ankle arthroplasty (TAA) implants.MethodsFollowing institutional approval and consent, 40 subjects with TAA implants underwent 1.5-T MRI prospectively. Evaluations included bone-implant interfaces, anatomical structures, abnormal findings and differential diagnoses before and after MRI. AUCs of P-P plots were used to determine superiority. Statistical differences were evaluated with McNemar and chi-square tests. P-values ≤ 0.05 were considered significant.ResultsCS-SEMAC TSE was superior to high-bandwidth TSE in showing the bone-implant interfaces (AUC=0.917), periprosthetic bone, tendons and joint capsule (AUC=0.337–0.766), bone marrow oedema (43 % difference, p=0.041), interface osteolysis (63 %, p=0.015), tendinopathy (62 %, p=0.062), periprosthetic fractures (60 %, p=0.250), synovitis (43 %, p=0.250), as well as reader confidence for bone marrow oedema (p=<0.001), fracture (p=0.001), interface osteolysis (p=0.003), synovitis (p=0.027) and tendinopathy (p=0.034). The number of differential diagnoses in symptomatic subjects after the MRI with CS-SEMAC decreased from 3 (1–4) to 1 (1–2) (p<0.001).ConclusionsMRI of TAA implants with CS-SEMAC improves the diagnosis of interface osteolysis, periprosthetic bone marrow oedema, fractures and tendinopathy when compared to high-BW TSE, and has a positive effect on patient management.Key Points• High-bandwidth TSE and compressed sensing SEMAC improve MRI of ankle arthroplasty implants. • Compressed sensing SEMAC improves bone-implant interfaces, periprosthetic bone, tendons and joint capsule visibility. • Compressed sensing SEMAC improves the diagnosis of osteolysis, tendinopathy, fractures and synovitis. • MRI decreases the number of clinical differential diagnoses of painful ankle arthroplasty implants.

Novel animal model for Achilles tendinopathy: Controlled experimental study of serial injections of collagenase in rabbits

Our goal was to develop a novel technique for inducing Achilles tendinopathy in animal models which more accurately represents the progressive histological and biomechanical characteristic of chronic Achilles tendinopathy in humans. In this animal research study, forty-five rabbits were randomly assigned to three groups and given bilateral Achilles injections. Low dose (LD group) (n = 18) underwent a novel technique with three low-dose (0.1mg) injections of collagenase that were separated by two weeks, the high dose group (HD) (n = 18) underwent traditional single high-dose (0.3mg) injections, and the third group were controls (n = 9). Six rabbits were sacrificed from each experimental group (LD and HD) at 10, 12 and 16 weeks. Control animals were sacrificed after 16 weeks. Histological and biomechanical properties were then compared in all three groups. At 10 weeks, Bonar score and tendon cross sectional area was highest in HD group, with impaired biomechanical properties compared to LD group. At 12 weeks, Bonar score was higher in LD group, with similar biomechanical findings when compared to HD group. After 16 weeks, Bonar score was significantly increased for both LD group (11,8±2,28) and HD group (5,6±2,51), when compared to controls (2±0,76). LD group showed more pronounced histological and biomechanical findings, including cross sectional area of the tendon, Young’s modulus, yield stress and ultimate tensile strength. In conclusion, Achilles tendinopathy in animal models that were induced by serial injections of low-dose collagenase showed more pronounced histological and biomechanical findings after 16 weeks than traditional techniques, mimicking better the progressive and chronic characteristic of the tendinopathy in humans.

Metal Artifact Reduction MRI for Total Ankle Replacement Sagittal Balance Evaluation

BACKGROUND Restoration of anatomical relationship between talus and tibia is crucial for longevity of total ankle replacement (TAR). Weight-bearing (WB) radiographs are the standard for evaluating the sagittal balance alignment, but are prone to rotational misalignment and altered measurements. Metal artifact reduction sequence (MARS) MRI allows visualization of periprosthetic landmarks and alignment of the image plane to the true sagittal axis of the implant. The purpose of this study was to compare TAR sagittal balance measurements on MARS MRI and WB radiographs. METHODS Twenty-three subjects with TAR [10 men/13 women, age 60 (41-73) years; 13 (3-24) months post-op] underwent MARS MRI and standard lateral WB radiographs. Standardized MARS MR images were aligned to the sagittal talar component axis. Three observers performed sagittal balance alignment measurements twice in an independent, random and blinded fashion. Lateral Talar Station (LTS), tibial axis-to-talus (T-T) ratio and normalized tibial axis-to-lateral-process (T-L) distance were measured. Concordance correlation coefficients (CCC) and intraclass correlation coefficients (ICC) were used for statistical analysis. In addition, mixed effects linear models were employed to assess overall concordance of the two image types. RESULTS The intraobserver agreement was excellent for radiographic (CCC=0.96) and MRI (CCC=0.90-0.97) measurements. Interobserver agreements were good-to-excellent with overall slightly higher agreements for MRI (ICC=0.78-0.94) than radiography (ICC=0.78-0.90) measurements. The T-T ratios of radiographs and MRI showed a high degree of concordance, whereas LTS was significantly lower on MRI when compared with radiographs, and T-L distance showed notable disagreement between the two imaging types. CONCLUSION Sagittal balance measurements performed on standardized weight-bearing radiographs and standardized MARS MRI demonstrate substantial correlation and similarity. Given its high intra and interobserver agreement, MARS MRI may be helpful for the evaluation of TAR sagittal balance. LEVEL OF EVIDENCE Level II - Prospective Comparative Study.

Outcomes of Flexor Digitorum Longus (FDL) Tendon Transfer for the Treatment of Chronic Achilles Tendinopathy

Category: Hindfoot, Sports Introduction/Purpose: Flexor hallucis longus (FHL) tendon transfer is a common surgical technique used for augmentation during the surgical treatment of chronic Achilles tendinopathy and reconstruction. Flexor digitorum longus (FDL) tendon transfer represents a viable surgical alternative for patients with failed FHL transfers or athletes where compromise of the hallux push off strength could negatively impact their level of activity. There is no reported clinical outcome data about this technique in the current literature. Our study describes the clinical and functional results after FDL tendon transfer for the treatment of patients with chronic Achilles tendinopathy. Methods: We retrospectively assessed prospectively collected data on patients that underwent FDL tendon transfer in the treatment of chronic Achilles tendinopathy (March 2012 - March 2015). Charts were reviewed for clinical data, associated treatments and complications. Preoperative assessment included the Visual Analogue Score (VAS), SF-36 health status survey and the lower extremity functional scale (LEFS). At final follow up we evaluated pain level, range of motion of the ankle and the toes, ability to perform single leg raise and toe walking, calf atrophy and complications. Postoperative outcomes were assessed by Visual Analogue Score (VAS), SF-36 health survey, Lower Extremity Functional Scale (LEFS), Foot Function Index (FFI), VISA-A score and the Foot and Ankle Ability Measure (FAAM). Fifteen patients (seventeen feet), 6 males and 9 females, mean age of 53.6 years (27- 76 years) and an average body mass index of 31.4 kg/m2 (20.5 to 45.4 kg/m2) were included in the study. Results: Mean follow-up was 27.5 months (15-49). Four patients (6 feet) had prior surgeries, including two patients with failed FHL transfer. We found significant clinical improvement when comparing pre-operative and postoperative VAS scores (6.0±3.3 versus 1±1.36; p<0.001), SF-36 physical component summary (28.2±10.7 versus 45.0±11.1; p<0.002) and LEFS (36.4±22 versus 57.9±20.5; p<0.011). At final follow up, 6/7 patients (86%) returned to prior levels of recreational sport activities. No differences were found on single leg raise test when compared to uninvolved side. One patient reported weakness for plantar flexion of the toes, without gait complaints. Mean VISA-A was 52.6 points (15-85), Foot Function Index (FFI) 21.2% (0-65%) and FAAM 86.2% (55.3-100%) for the FAAM. Three patients had superficial infection and two patients had deep infection, requiring surgical debridement. Conclusion: FDL tendon transfer represents a safe surgical alternative as a method of augmentation during the treatment of chronic Achilles tendinopathy. Our study showed comparable clinical and functional outcomes to FHL tendon transfer and minimal complications or donor site morbidity.

Hindfoot Alignment of Adult Acquired Flatfoot Deformity: A Comparison of Clinical Assessment and Weightbearing ConeBeam CT Examinations

Category: Hindfoot Introduction/Purpose: Assessment of hindfoot alignment in adult acquired flatfoot deformity (AAFD) can be challenging. Clinical judgment and radiograph studies while important may not represent the accurate valgus alignment of the affected patients. Weightbearing (WB) ConeBeam CT (CBCT) is an emerging imaging modality that may potentially better demonstrate the three-dimensional (3D) deformity, facilitating visualization of important soft-tissue and bony landmarks and helping in surgical planning. Based on the relative position of bone and soft-tissue axes, different measurements of hindfoot alignment can be obtained with CT images. Therefore, we compared clinical assessment of hindfoot valgus alignment in AAFD patients with different possible measurements performed on WB CBCT images. Methods: In this prospective, IRB-approved study, 20 patients (20 feet, 15 right and 5 left) with clinical diagnosis of flexible AAFD were included. There were 12 males and 8 females, with a mean age of 52.2 years (range, 20 – 88 years of age), and average BMI of 30.35 kg/m2 (range, 19.00 – 46.09 kg/m2). Patients underwent clinical assessment of hindfoot alignment as well as WB CBCT. Two independent and blinded foot and ankle board-certified surgeons performed different hindfoot alignment measurements on the WB CBCT images that included: 3D “clinical” alignment; Achilles tendon axis/calcaneal tuberosity angle; angles formed between the tibial axis and the calcaneal tuberosity, calcaneal axis and line connecting midpoint of subtalar joint and most inferior part of calcaneal tuberosity. Positive values were considered valgus alignment. Mean differences between the measurements modalities were compared by paired T-test. Intra- and Inter-observer reliability for the WB CBCT measurements were calculated using Pearson correlation. Results: The mean clinical hindfoot valgus measured was 15.15o (SD 7.7o). It was found to be significantly different from the mean values of all WB CBCT angles modalities: 3D “clinical” alignment (10.42o, p < 0.015); Achilles tendon/calcaneal tuberosity angle (2.96o, p < 0.0001); tibial axis/calcaneal tuberosity angle (5.42o, p < 0.0001); tibial axis/subtalar joint angle (7.52o, p < 0.0001) and tibial axis/calcaneal axis angle (20.39o, p < 0.017). We found an excellent intra-observer agreement for all WB CBCT 3D measurements (range, 0.8863 – 0.9713, p < 0.0001). There was also good to excellent inter-observer reliability, with the exception of the 3D “clinical” alignment (r=0.450, p < 0.04), that showed moderate correlation. Conclusion: The use of 3D WB CBCT imaging can help characterize the valgus hindfoot alignment in patients with adult acquired flatfoot deformity. We found the different CBCT measurements modalities to be reliable and repeatable, and to significantly differ from the clinical evaluation of hindfoot valgus alignment.

Comparison between Weight Bearing Radiographs and Weight Bearing ConeBeam CT Examinations in the Assessment of Adult Acquired Flatfoot Deformity

Category: Hindfoot Introduction/Purpose: Adult acquired flatfoot deformity (AAFD) represents a biomechanical derangement involving the three- dimensional (3D) midfoot and hindfoot osseous complex, and can be challenging to optimally characterize using conventional two- dimensional (2D) plain radiographs. Weightbearing (WB) ConeBeam CT (CBCT) can better demonstrate the deformity of the 3D structures during WB. Therefore, we compared validated AAFD measurements between WB conventional radiographs and WB CBCT images. Methods: In this prospective, IRB approved and HIPAA compliant study, 20 patients (20 feet, 15 right and 5 left) with clinical diagnosis of flexible AAFD were included, 12 males and 8 females, with a mean age of 52.2 years (range, 20 to 88 years of age), and average BMI of 30.35 kg/m2 (range, 19.00 to 46.09 kg/m2). Involved feet underwent standing (WB) anteroposterior (AP) and lateral radiographs, and were also scanned by WB CBCTs. Both imaging modalities were assessed with traditional AAFD measurements obtained at sagittal (lateral view on radiograph) and axial (anteroposterior view on radiograph) planes using predefined anatomical landmarks, by two independent and blinded foot and ankle fellowship-trained observers. Intra- and Inter- observer reliabilities for both imaging modalities were calculated using Pearson correlation. WB radiograph and WB CBCT measurements were compared by T-Test of the means. P- values < 0.05 were considered significant. Results: There was good to excellent intra and inter-observer agreements for most of the measurements on both radiographs and WB CBCT images, with slightly better results favoring WBCT measurements. When comparing WB radiographs and WB CBCT images, we found significant differences in the mean values for some of the measurements, including: talus-first metatarsal angle in the sagittal plane (11.34° x 21.73°, p<0.0001), navicular-medial cuneiform angle (13.19° x 7.63°, p<0.0004), medial cuneiform to floor distance (6.70 mm x 5.50 mm, p<0.0003) and navicular to floor distance (31.34 mm x 23.22 mm, p<0.0001). No significant differences were found when measuring: talus-first metatarsal angle in the axial plane, talar uncoverage angle, cuboid to floor distance and calcaneal inclination angle. Conclusion: Traditional adult acquired flatfoot deformity radiographic measurements are obtainable using high resolution 3D WB CBCT imaging. Measurements performed on WB CBCT have similar intra-observer and overall higher inter-observer reliability when compared to WB radiographs. The statistically significant differences found in some of the measurements, when comparing both imaging techniques, might be related to a better characterization of the three-dimensional deformity on WB CBCT images.

Interobsever Variability of Measurements for Flatfoot Deformity Using High Resolution Weightbearing Cone-Beam CT Examination According to Reader Experience

Category: Other Introduction/Purpose: To evaluate interobserver reliability among readers of different clinical experience by applying measurements for adult acquired flatfoot deformity (AAFD) using high-resolution three-dimensional (3D) weightbearing (WB) cone-beam CT (CBCT) examination. Methods: In this IRB approved study, 20 patients with flexible AAFD [12 male, 8 female; mean age 54.2 (20-88) years] were scanned with standing (weightbearing) CTs. Two blinded observers, a medical student and a foot/ankle surgeon, applied validated AAFD measurements in sagittal, coronal, and axial planes using predefined anatomical landmarks. Interobserver reliability was calculated using Pearson correlation. Results: There was significant interobserver agreement with high correlation for the following measurements(p < 0.0001): distances between medial cuneiform to-floor (r=0.981) and to-skin (r=0.986), navicular to-floor (0.992) and to-skin (r=0.900); cuboid to-floor (r=0.975) and to-skin (r=0.978), and calcaneus-to-fibula (r=0.808); calcaneal inclination angle (r=0.795); forefoot arch angle (r=0.983); and subtalar horizontal angles at 25%, 50%, and 75% of the anteroposterior joint length (r=0.784, 0.891, 0.809). Significant agreement with moderate correlation was additionally demonstrated for talar-first metatarsal angle (r=0.553, p < 0.014), medial cuneiform-first metatarsal angle (r=0.668, p < 0.001), and navicular-cuneiform angle (r=0.746, p < 0.0002). Level of training did not influence the reliability of any measurements except medial cuneiform-first metatarsal angle (specialist: 8.83°; student: 1.61°; p < 0.01). Statistically insignificant difference between readers (p>0.05) was noted in mean talar-first metatarsal and subtalar horizontal angles. Conclusion: While literature describes large variability for AAFD measurements from plain radiographs among readers of varying medical experience, 3D WB CBCT can yield similar measurements using predefined planes with high reliability, independent of reader experience.

Flexible Adult Acquired Flatfoot Deformity: Comparison Between Weightbearing and Nonweightbearing Cone-Beam CT Examinations

Category: Other Introduction/Purpose: Adult acquired flatfoot deformity (AAFD) is a biomechanical derangement involving the three- dimensional (3D) midfoot and hindfoot osseous complex, which can be challenging to optimally characterize using conventional two-dimensional (2D) plain radiographs. Weightbearing (WB) Cone-Beam CT (CBCT) can better demonstrate the deformity of the 3D structures during WB. Therefore, we compared validated AAFD measurements between non-weightbearing (NWB) and WB CBCT images. Methods: In this prospective, IRB approved study, 20 patients were included, 12 males and 8 females, mean age of 54.21 (20-88) years, with clinical diagnosis of flexible AAFD. Subjects were scanned with standing (WB) and seated (NWB) CBCTs. WB and NWB CBCT images were assessed with traditional flatfoot measurements obtained at sagittal, coronal, and axial planes using predefined anatomical landmarks, by two independent observers. Interobserver reliability was calculated using Pearson correlation. Results: The measurements in patients with AAFD differed significantly between WB and NWB CBCT images. Specifically, WB images showed, when compared to NWB, decreased forefoot arch angle (mean difference: 9.91°, p < 0.0001), increased talus-first metatarsal angle (10.59°, p < 0.0001), increased navicular-medial cuneiform angle (13.89°, p < 0.0001), decreased navicular-floor (coronal 14.05mm/sagittal 14.91mm, p < 0.0001) and navicular-skin distances (coronal 5.87mm/sagittal 8.25mm, p < 0.0001), decreased medial cuneiform-floor (coronal 10.79mm/sagittal 11.07mm, p < 0.0001) and medial cuneiform-skin distances (coronal 4.45mm/sagittal 5.78mm, p < 0.0001), and decreased cuboid-floor (5.78mm, p < 0.0001) and cuboid-skin distances in the sagittal plane (4.60mm, p < 0.0001). Interobserver reliability was good to excellent (0.610-0.991). Conclusion: Traditional adult acquired flatfoot deformity radiographic measurements are obtainable using high resolution 3D WB CBCT imaging, and can help characterize the biomechanical derangements during weightbearing in subjects with flexible AAFD.

Metal Artifact Reduction MRI of Total Ankle Arthroplasty

Category: Ankle Arthritis Introduction/Purpose: Total ankle arthroplasty (TAA) is an often successful treatment for end-stage ankle arthritis; however, a subset of patient presents with ankle pain following TAA and concern for early failure. Although radiographs are often diagnostic, patients with normal radiographs and continued pain pose a diagnostic challenge. In those patients, magnetic resonance imaging (MRI) may be helpful to evaluate periprosthetic bone and soft tissues. Traditional high-bandwidth (high-BW) metal artifact reduction sequence (MARS) MRI lessens metal artifacts, but image distortions remain with Cobalt-Chromium implants. Slice- Encoding-Metal-Artifact-Correction (SEMAC) is a recently FDA-approved, advanced MARS technique that promises more powerful metal suppression. Therefore, we prospectively tested the hypothesis that SEMAC MARS MRI facilitates better metal reduction and visibility of periprosthetic structures than does traditional MARS MRI in patients with TAA. Methods: In this IRB-approved study, 20 volunteers [10 females/10 males; age, 59 (41-73) years, 15 (3-24) months post- operatively] with TAA were prospectively enrolled. The research protocol consisted of high-BW and SEMAC MARS MRI using a clinical 1.5 Tesla MRI scanner and a dedicated boot-shaped MRI coil. For each technique, intermediate-weighted and fat- suppressed MR images were obtained in axial, sagittal and coronal orientation. Three observes (2 foot/ankle surgeons and 1 musculoskeletal radiologist) evaluated the high-BW and SEMAC MR images in an independent, random and blinded fashion. Equidistance 5-point Likert scales (1=non-diagnostic, 5=very good) were used to grade image quality as well as the quality of metal suppression and visibility of bone-metal interfaces, tendons, ligaments, bone and joints. Differences between the two MARS techniques were assessed with a Kruskall-Wallis test, and interobserver agreement was graded with Intraclass-Correlation- Coefficient (ICC). Bonferroni-corrected p-values ≤ 0.01 were considered significant. Results: All 20 subjects completed the research protocol. There was good agreement between observers (ICC = 0.79; 95% CI, 0.78-0.80). Image quality of high-BW and SEMAC MR images was similar for tissue contrast, fat suppression, and fluid brightness, whereas image sharpness was one interval higher on high-BW images than on SEMAC images (p < 0.01). Metal reduction of TAA components was significantly better (p < 0.01) on SEMAC images (very good) than on high-BW images (poor-to-adequate). At level of the implants, long flexor and peroneal tendons, deep deltoid, syndesmotic and lateral collateral ligaments, and medial and lateral gutters were significantly better seen (p < 0.01) on SEMAC images (good-to-very good) than on high-BW images (poor-to- adequate). Similar visibility (good-to-very good) was found for the remainder of the tendons, ligaments, bones and joints. Conclusion: Based on the results of our study, we accept the hypothesis that SEMAC MARS MRI outperforms traditional high- BW MARS MRI in the degree of metal artifact reduction as well as visibility of bone implant interfaces and periprosthetic tendons and ligaments at level of the joint line. We believe advanced MARS MRI can be a valuable clinical tool to assess osseous integration and soft tissue lesions in patients following TAA.