Musculoskeletal disorders represent a leading cause of global disability, creating an urgent demand for precise interpretation of medical imaging. Current artificial intelligence (AI) approaches in orthopedics predominantly rely on task-specific, supervised learning paradigms. These methods are inherently fragmented, require extensive annotated datasets, and often lack generalizability across different modalities and clinical scenarios. The development of foundation models in this field has been constrained by the scarcity of large-scale, curated, and open-source musculoskeletal datasets. To address these challenges, we introduce OrthoFoundation, a multimodal vision foundation model optimized for musculoskeletal pathology. We constructed a pre-training dataset of 1.2 million unlabeled knee X-ray and MRI images from internal and public databases. Utilizing a Dinov3 backbone, the model was trained via self-supervised contrastive learning to capture robust radiological representations. OrthoFoundation achieves state-of-the-art (SOTA) performance across 14 downstream tasks. It attained superior accuracy in X-ray osteoarthritis diagnosis and ranked first in MRI structural injury detection. The model demonstrated remarkable label efficiency, matching supervised baselines using only 50% of labeled data. Furthermore, despite being pre-trained on knee images, OrthoFoundation exhibited exceptional cross-anatomy generalization to the hip, shoulder, and ankle. OrthoFoundation represents a significant advancement toward general-purpose AI for musculoskeletal imaging. By learning fundamental, joint-agnostic radiological semantics from large-scale multimodal data, it overcomes the limitations of conventional models, which provides a robust framework for reducing annotation burdens and enhancing diagnostic accuracy in clinical practice.

Abstract Purpose To investigate whether bone tunnel position is associated with postoperative recurrent sprain in chronic lateral ankle instability (CLAI) patients following anatomic lateral ligament reconstruction with autograft and to propose potential safe zones for bone tunnel placement. Methods Four hundred fifty‐seven CLAI patients following anatomic reconstruction at our institution from June 2015 to September 2023 were retrospectively reviewed. Among them, 30 patients sustaining recurrent sprain within 48 months after surgery were included in the recurrent sprain group. Then, 206 consecutive patients who showed no recurrent sprain with minimum 48‐month follow‐up were selected as potential controls. Patients were further divided into two groups based on whether the calcaneofibular ligament was reconstructed, and 1:1 propensity score matching was performed separately in the two groups. The positions of the fibular, talar and calcaneal tunnel were measured on three‐dimensional computed tomography (3D‐CT) scans acquired at postoperative Day 1 and were compared. Potential safe zones for tunnel placement were identified through receiver operating characteristic analysis. Results There was no difference in fibular and calcaneal tunnel position along the reference line between groups. Patients with postoperative recurrent sprain demonstrated a significantly lower talar tunnel position on the reference line from the apex of the lateral talar process (ALTP) to the anterolateral corner of the trochlea (ACT) on the talus compared to those without postoperative recurrent sprain (58.1 ± 6.3% vs. 63.9 ± 6.0%, p = 0.001). The area under the curve value of the talar tunnel position was 0.744 (p = 0.001), and the determined cutoff value was 58.9% with a sensitivity of 90.0% and a specificity of 53.3%. Conclusion A lower talar tunnel placement is associated with postoperative recurrent sprain in CLAI patients following anatomic reconstruction, with 58.9% above the ALTP‐ACT reference line can be a potential safe zone. The talar tunnel placement needs to be carefully considered to reduce postoperative recurrent sprain. Level of Evidence Level III, retrospective study.

Background The knee meniscus has limited intrinsic repair and regenerative capacity. Safe and effective biotherapeutic products remain lacking, and enhancing cell migration and angiogenesis remains a major challenge. This study aimed to develop a novel self-assembled meniscus-specific extracellular matrix membrane (saECM-M) and evaluate its performance in scaffold-free meniscus repair and scaffold-guided whole-meniscus reconstruction, in comparison with chemically processed decellularized ECM (cpd-ECM). Methods Meniscal fibrochondrocytes (MFCs) were cultured under membrane-inducing conditions to generate saECM-M in vitro. Its bioactivity was evaluated by cell migration and endothelial network formation assays in vitro and by a scaffold-free meniscus defect repair model in vivo. For whole-meniscus reconstruction, saECM-M was integrated with a biomimetic 3D-printed poly(ε-caprolactone) (PCL) scaffold to guide ECM alignment and provide mechanical support. PCL scaffolds seeded with MFCs alone or combined with cpd-ECM served as controls. Meniscus regeneration was assessed in a rabbit meniscectomy model using imaging, histological, and biomechanical analyses, including chondroprotective evaluation. Transcriptomic profiling and proteomics were performed to investigate molecular pathways associated with saECM-M-mediated cell migration and vascular ingrowth. Results saECM-M was characterized as a cell-laden membrane enriched with native meniscus-like ECM components. In vitro, saECM-M significantly enhanced cell migration and endothelial network formation compared with cpd-ECM (approximately 1.2-fold and 2-fold, respectively; P < 0.05), accompanied by increased collagen deposition. In vivo, isolated saECM-M promoted superior meniscus defect repair. When integrated with the biomimetic PCL scaffold, saECM-M further promoted aligned collagen organization, vascular ingrowth, and robust meniscus regeneration. Compared with the PCL-cpd-ECM group, the PCL-saECM-M group showed improved host tissue integration on MRI (WORMS score: 10 ± 1 vs 32.3 ± 2.1), enhanced tensile mechanical properties (79 ± 2 MPa vs 64 ± 7.8 MPa), and superior chondroprotective effects (Mankin score: 1.67 ± 0.58 vs 6.67 ± 0.58) (P < 0.05). Transcriptomic and proteomic analyses identified enrichment of ECM-receptor interaction- and PI3K-AKT signaling-related pathways associated with saECM-M-enhanced cell migration and vascular ingrowth. Conclusion saECM-M functions as an intrinsically bioactive meniscus-derived matrix capable of promoting both scaffold-free repair and scaffold-guided reconstruction. Compared with cpd-ECM, saECM-M more effectively supports cell migration, angiogenesis, and organized collagen remodeling, leading to superior regenerative outcomes. The translational potential of this article saECM-M addresses an unmet clinical need as a cell-secreted, additive-free bioactive matrix. It may serve as a simple biologic augmentation for arthroscopic repair, while integration with an FDA-approved PCL scaffold offers a promising strategy for meniscus replacement with improved bioactivity and biocompatibility.

Objective This study aims to compare the long-term radiographic outcomes, ankle function, and complications after microfracture surgery in patients with cystic osteochondral lesions of the talus (COLTs) with or without the preoperative MRI "Jumping Dot Sign" (JD sign), and to propose a novel imaging predictor for longterm prognosis. Methods Patients with COLTs who underwent arthroscopic microfracture surgery between January 2010 and June 2015 were retrospectively enrolled. At the final follow-up, symptoms and ankle function were assessed using the Visual Analogue Scale (VAS), American Orthopaedic Foot & Ankle Society AnkleHindfoot Score (AOFAS), Foot and Ankle Outcome Score (FAOS), and Tegner score, and postoperative complications were recorded. Postoperative ankle MRI was performed to evaluate bone marrow edema (BME)and bone cyst dimensions. Patients were divided into the JD sign group and the non-JD sign group based on preoperative MRI, and outcomes were compared using these metrics. Results Sixty patients with COLTs were included (JD sign group: n=27, non-JD sign group: n=33). Baseline characteristics including gender (22M/5F vs.29M/4F, P=0.744), age[(36.1±9.5) years vs. (33.3±7.5) years, P=0.211)], BMI[(26.0±2.6) kg/m²vs. (26.4±4.3) kg/m²,P=0.681)], follow-up duration[(11.2±1.8) years vs. (11.4±2.1) years, P=0.712), lesion location, and lesion location and size[(81.2 ± 38.2) mm² vs. (76.0 ± 38.4) mm², P=0.605)] showed no significant differences. Regarding ankle function, the non-JD sign group showed significantly higher AOFAS, FAOS, and Tegner scores (P<0.05) and lower VAS pain levels (P<0.05). The incidence of postoperative limitation of plantarflexion and dorsiflexion was lower in the non-JD sign group (30.30% vs. 62.96%, P=0.023). Regarding postoperative radiographic outcomes,the non-JD sign group had significantly smaller postoperative BME volume, lower bone cyst recurrence rate, and smaller recurrent cyst areacompared to the JD sign group. Conclusion The preoperative MRI "Jumping Dot Sign"may serve as a radiographic predictor of long-term prognosis in patients with COLTs after microfracture surgery.Microfracture surgery should be used with caution in patients exhibiting the JD sign.

Aims Cystic osteochondral lesions of the talus (OLTs) demonstrate metabolic dysregulation that predisposes to osteonecrosis, for which timely invasive intervention is crucial. This study aimed to investigate the predictive value of bone tracer uptake (BTU) of single-photon emission CT/CT (SPECT/CT) for conservative treatment outcomes, and promote a novel indicator for decision-making of the treatment of cystic OLTs. Methods Patients with cystic OLTs who underwent SPECT/CT scanning from January 2020 to August 2024 received three-month conservative treatments. Maximum standardized uptake value (SUVmax) was quantified from SPECT/CT acquisitions. Visual analogue scales (VAS) for pain and American Orthopaedic Foot and Ankle Society (AOFAS) scores were assessed before and after the treatment. A total of 95 patients were divided into the Success (23) and Failure (72) groups based on post-treatment AOFAS scores (80). VAS, AOFAS, and SUVmax were compared, and receiver operating characteristic (ROC) curve analysis was performed. Results The Failure group had significantly higher mean SUVmax (12.35 (SD 4.94) vs 6.45 (SD 4.63), p < 0.001) and inferior mean AOFAS scores (75.03 (SD 8.97) vs 80.35 (SD 10.29), p = 0.033). Maximum BTU located predominantly on sclerotic bone rims of OLTs (90/95, 94.7%). Medial OLT had significantly greater mean SUVmax (11.77 (SD 5.45) vs 7.06 (SD 3.68), p < 0.001). When the SUVmax exceeded 8.62 (sensitivity 81.9%, specificity 82.6%), OLT patients were more likely to fail the conservative treatment. Conclusion This study demonstrates that SPECT/CT-derived BTU is a potential indicator for treatment decision-making in cystic OLTs. Elevated BTU showed a statistically significant association with poor response to nonoperative care in this cohort. The critical threshold of SUVmax > 8.62 (sensitivity 81.9%, specificity 82.6%) identifies high-risk patients who might need early surgical intervention. Additionally, the SUVmax showed a strong correlation with ankle pain, and maximum BTU located predominantly on the sclerotic rims of OLTs.

Joint cartilage regeneration presents a significant challenge due to its limited self-repair capacity, hindered by local inflammation and metabolic dysregulation. Mitochondrial quality control (MQC) is crucial for maintaining chondrocyte differentiation and metabolism; however, strategies targeting MQC remain underexplored. Here, we generated an acellular delivery vehicle, cytoplastCXCR4, which was designed to facilitate mitochondria transport and restore MQC, from enucleated CXCR4 transfected mesenchymal stem cells (MSCs) through ultracentrifugation. CXCR4 was overexpressed to enhance migration toward inflamed cartilage via its interaction with SDF-1α, which is a chemokine elevated at the site of cartilage damage. The CXCR4 gene-transfected, nuclear-depleted cytoplastCXCR4 exhibited enhanced homing ability, via CXCR4-SDF-1α axis. In vitro results showed that cytoplastCXCR4 treatment improved mitochondrial quality-structure, oxygen consumption, and ATP synthesis- in osteoarthritic chondrocytes, thus re-establishing metabolic balance. Compared to exosomes, cytoplasts retain cellular bioactivity, including functional organelles and paracrine signaling. Notably, we observed mitochondrial transfer via cytoplast-derived vesicles, which contributed to the restoration of morphology and function in damaged mitochondria within injured chondrocytes. Furthermore, in vivo results demonstrated that cytoplastsCXCR4 had a longer retention time, and could precisely target the damaged cartilage and significantly facilitate cartilage repair compared with exosomes in a rat knee cartilage defect model. This study presents an MQC-targeted and mitochondria-preserving approach, offering an improved strategy for joint cartilage regeneration.

Background Arthroscopic microfracture (MF) is currently the most widely used surgical treatment for osteochondral lesions of the talus (OLTs), but poor cartilage regeneration and prolonged rehabilitation are the focus of improvement. This study aimed to conduct a novel procedure, subtle and precise osteo-drilling for the recovery of talus (SPORT surgery) with self-designed instruments and compare its clinical outcomes to the MF for OLTs. The novel procedure has the potential to address certain limitations of MF, including the significant subchondral bone destruction and restricted drilling position. Study design Randomized controlled trial; Level of evidence, 1. Methods A total of 56 patients with OLTs who underwent surgical treatment from September 2021 to August 2022 were randomly assigned to the MF group or the SPORT group. For the SPORT group, a novel high elastic K-wire with a 1.2 mm diameter was used to drill subchondral bone holes with 10 mm depth through a curved guide. Clinical assessments were conducted by using VAS (Visual Analog Scale), AOFAS (American Orthopaedic Foot & Ankle Society) score, FAAM-ADL (Foot and Ankle Ability Measure-Activities of Daily Living) score, and SP (Foot and Ankle Ability Measure-Sports) score. MRI evaluation included relative signal intensity of subchondral bone marrow edema (RSI-SBME), volume of the subchondral bone marrow edema (V-SBME), and the Magnetic Resonance Observation of Cartilage Repair Tissue (MOCART) score. The differences in clinical outcomes and imaging results between the two groups were compared at pre-surgery, postoperative 3, 6, 12, and 24 months. Results Both the MF and SPORT groups showed significant improvement in all clinical outcomes. At 3- and 6-month post-op, the clinical outcomes of the SPORT group were significantly better than those of the MF group (improved 44.6 % and 32.2 % at 3- and 6-month postoperative for FAAM-SP, P ≤ 0.001). The percentage of patients in the SPORT group who met the MCID of FAAM-SP is significantly higher than in the MF group at 3-month (P = 0.007) and 6-month (P = 0.005) postoperative. Besides, the SPORT group achieved better MOCART scores than the MF group at all follow-ups (P ≤ 0.002). After 12 months, there was no significant difference in clinical outcomes between the two groups. Conclusion The novel device-guided SPORT surgery has significant advantages over conventional MF technology in terms of early postoperative return to sport and long-term cartilage regeneration in OLT patients. The translational potential of this article The SPORT surgery and the novel guided subchondral bone drilling instruments have excellent translational potential in sports medicine as an alternative to replace traditional MF techniques for OLTs and can be extended to cartilage repair in other joints, such as the knee and shoulder.

Although extracellular matrix (ECM) maturation during articular cartilage development is well recognized, quantitative, temporally resolved characterizations in large-animal models remain scarce. Here, we systematically characterized porcine femoral condyle cartilage from embryonic day 45 (E45) to postnatal month 10 (P10M) using integrated morphological, biochemical, and ultrastructural approaches. Knee joint width increased from 2.5 mm at E45 and 11 mm at E100 and 46 mm at P10M, whereas chondrocyte density declined markedly after E75 (p < 0.05). The change was not statistically significant in collagen content until E75 (3.2 μg/mg), but rose significantly at E100 (6.7 μg/mg, p < 0.05). Glycosaminoglycan (GAG) content remained stable prenatally but increased markedly postnatally at P1M (163 μg/mg, p < 0.05), while water content decreased significantly at P10M (p < 0.05). Type II collagen gradually increased from E60 to P1M, whereas types III, XI, and IX accumulated from E60 onward. Collagen fibril diameter increased progressively from approximately 25 nm at E45 to approximately 80 nm at P10M (p < 0.05). Matrix metalloproteinase 1 (MMP-1) and cathepsin K exhibited elevated expression at both E45 and P1M, suggesting two distinct phases of ECM remodeling. Collectively, these findings demonstrate that cartilage ECM maturation follows a temporally ordered sequence, beginning with the establishment of a collagen framework, followed by GAG accumulation and fibril thickening. This high-resolution temporal map established in a large-animal model provides a critical reference for cartilage regeneration strategies.

Articular cartilage has limited capacity for self-repair due to its avascular nature. Successful cartilage repair requires the harmonious integration of sufficient stem cell recruitment, an optimal local microenvironment and a sustained repair timeframe. Here, we present a biocompatible, physically crosslinked silk fibroin platform with tunable β-sheet content (5-50 %) via freeze-assembly. This platform enables flexible and precise tuning of drug release kinetics without chemical cross-linkers. This system allows controlled drug release durations ranging from 1 to 35 days, suitable for both hydrophilic (MSC affinity peptide, MAP, serving a pro-recruiting role) and hydrophobic (kartogenin, KGN, pro-differentiating role) drugs. In a rat cartilage defect model, a sustained 21-day MAP release profile was identified as optimal, achieving an unprecedented high density of MSC recruitment (∼2.34 × 104 cells/mm3) within a differentiating-friendly timeframe. Synchronized with KGN delivery, the co-delivery system further promoted robust hyaline cartilage regeneration. This outcome may be attributed to the effect of Cdh2 genes involved in cell adhesion and p38 MAPK pathways. This work provides a structurally programmable, scalable strategy to achieve coordinated, high-density MSC recruitment and timed differentiation, advancing the paradigm of precise biomaterial design for tissue repair.

Background: Generalized joint laxity (GJL) has been reported as a risk factor for poor outcomes following multiligament surgery. However, no previous studies have investigated whether GJL is a risk factor of postoperative resprain following anatomic reconstruction using gracilis tendon autograft. Purpose: To investigate whether GJL is a risk factor of postoperative resprain in chronic lateral ankle instability (CLAI) patients following anatomic reconstruction using gracilis tendon autograft. Study Design: Cohort study; Level of evidence, 3. Methods: A total of 290 consecutive CLAI patients who underwent anatomic reconstruction using gracilis tendon autograft from January 2018 to September 2023 with a minimum follow-up of 24 months were retrospectively evaluated and were divided into 2 groups according to the presence of GJL (based on age and Beighton score). Clinical outcomes were measured through the presence of postoperative noncontact resprain and recurrent instability, return-to-sport status, and patient-reported outcomes including the visual analog scale (VAS) for pain, Tegner Activity Level scale, and Karlsson-Peterson ankle score (Karlsson score). Risk factors associated with postoperative noncontact resprain were determined using univariate and multivariate logistic regression analysis. Survival outcomes were compared using Kaplan-Meier analysis. Results: A total of 72 patients were included in the GJL group (mean age at surgery, 27.5 ± 9.0 years; male, n = 27; female, n = 45), and 218 patients were included in the non-GJL group (mean age at surgery, 32.8 ± 10.2 years; male, n = 129; female, n = 89). The overall noncontact resprain rate was 6.6%, and the overall recurrent instability rate was 1.7%. There was no significant difference in the postoperative VAS, Tegner, return to recreational sports rate, return to preinjury sports rate, and other surgical complications between the 2 groups. The patients with GJL demonstrated a significantly higher postoperative noncontact resprain rate (15.3% vs 3.7%; P = .001), a significantly higher postoperative recurrent instability rate (6.9% vs 0.0%; P = .001), and a significantly lower postoperative Karlsson score (90.0 ± 6.2 vs 95.6 ± 4.4; P < .001) compared with those without GJL. Multivariate logistic regression analysis revealed that GJL was an independent predictor of postoperative noncontact resprain following anatomic reconstruction using gracilis tendon autograft ( P = .004; odds ratio = 4.293; 95% CI, 1.612-11.431). The cumulative success rates of the patients with GJL were significantly lower than those of the patients without GJL ( P = .001). Conclusion: GJL is an independent predictor of postoperative noncontact resprain of CLAI patients following anatomic reconstruction using gracilis tendon autograft.

BACKGROUND:Chronic ankle instability (CAI) is a common consequence of lateral ankle sprain and is characterized by recurrent episodes of giving way and impaired neuromuscular control. Neuromuscular control impairments (such as decreased postural stability and delayed muscle responses) are dominant during reduced vision. However, the neural mechanisms underlying impaired sensorimotor integration under reduced vision remain unclear. As a key structure involved in sensorimotor processing, the role of the cerebellum during vision-related neuromuscular control impairments has not been well defined. QUESTIONS/PURPOSES:(1) Are there any differences in cerebellar activation across different visual conditions in patients with CAI compared with a control group? (2) Are there any associations between vision-related cortical activation changes and static and dynamic neuromuscular control deficits in patients with CAI? METHODS:This cross-sectional study enrolled physically active adults with unilateral CAI and control participants matched for age, sex, and activity level. Between May and September 2025, a total of 55 individuals were screened. Of these, 64% (35 of 55) were considered eligible based on the inclusion criteria, consisting of 57% (20 of 35) in the CAI group (nine females, mean ± SD age 28 ± 7 years) and 43% (15 of 35) in the control group (eight females, mean age 25 ± 4 years). To assess cerebellar activation and its relation to neuromuscular control, all participants underwent task-based functional MRI (fMRI) and laboratory-based neuromuscular assessments. During fMRI, participants performed repetitive ankle dorsiflexion and plantarflexion movements using an MRI-compatible device under alternating eyes-open and eyes-closed conditions in a block design. Neuromuscular control was assessed during single-leg stance and a sudden ankle inversion (trapdoor) task. Postural stability was quantified using the Romberg ratio derived from center-of-pressure measures. Dynamic neuromuscular response was evaluated by peroneal reaction time using surface EMG synchronized with motion capture. Ankle function and perceived instability were assessed using questionnaires (Cumberland Ankle Instability Tool [CAIT], Foot and Ankle Ability Measure [FAAM], and Ankle Ligament Reconstruction-Return to Sport After Injury [ALR-RSI] tool). Whole-brain analyses were conducted to identify between-group differences in vision-related activation (eyes closed minus eyes open) using a general linear model with Gaussian random field correction. Clusters greater than 50 voxels were considered to be the minimum clinically important difference. Correlation analyses were performed to examine associations between neural activation patterns, neuromuscular outcomes, and clinical measures, with age and sex included as covariates. RESULTS:Compared with controls, individuals with CAI demonstrated reduced activation when eyes changed from open to closed in bilateral cerebellar regions including lobule VI (left hemisphere: cluster size 156 voxels, group mean difference -25.1 [95% confidence interval (CI) -40.1 to -10.1]; p < 0.001; right hemisphere: cluster size 153 voxels, group mean difference -21.7 [95% CI -36.3 to -7.6]), vermis VI (cluster size 51, group mean difference -26.6 voxels [95% CI -44.8 to -8.4]), and crus I (cluster size 112 voxels, group mean difference -21.9 [95% CI -35.8 to -8.1]), as well as in the left fusiform gyrus (cluster size 54 voxels, group mean difference -18.1 [95% CI -29.3 to -6.8]). Correlation analyses revealed that reduced activation in cerebellar and fusiform gyrus in both eyes-open and eyes-closed states was moderately associated with delayed peroneal reaction time in both visual conditions (eyes open: r = -0.38; p = 0.03; eyes closed: r = -0.40; p = 0.02) and worse self-reported ankle stability and function, as measured by the CAIT (r = 0.34; p = 0.046) and ALR-RSI (r = 0.38; p = 0.02). No correlations were observed between cerebellar activation and Romberg ratio. CONCLUSION:Patients with CAI demonstrated reduced cerebellar activation across visual conditions, which was associated with delayed dynamic muscle responses and worse perceived ankle stability. CLINICAL RELEVANCE:These findings suggest that patients with CAI had vision-related central activation strategies, such as reduced cerebellar activation when eyes were closed compared with controls. Such neural alterations were associated with decreased neuromuscular control and diminished ankle function. These vision-specific central alterations may be important in designing future interventions aimed at enhancing central sensorimotor integration.

BACKGROUND:Autologous osteoperiosteal transplantation (AOPT) is a promising treatment for large cystic osteochondral lesions of the talus (OLTs), but the influence of lesion laterality remains unclear. The purpose of this study was to compare the clinical and radiologic outcomes of lateral vs medial large cystic OLTs treated with AOPT. METHODS:Patients with lateral or medial large cystic OLTs who underwent AOPT between 2010 and 2023 were retrospectively reviewed. Patients were propensity matched in 1:1 ratio on sex, age, body mass index, affected side, and lesion profiles (area, depth, volume). Clinical outcomes were assessed using the visual analog scale for pain (VAS), the Foot Ankle Outcome Score (FAOS), and the ankle activity score (AAS). The final FAOS score was designated as the primary outcome measure. Radiologic outcomes were evaluated using the MOCART (magnetic resonance observation of cartilage repair tissue) 2.0 ankle score. RESULTS:A total of 22 matched patients per group were included, with a mean follow-up of 84.6 ± 45.9 months. The lateral group comprised 20 males and 2 females (mean age, 40.4 ± 11.2 years), whereas the medial group included 18 males and 4 females (mean age, 38.0 ± 8.4 years). Both groups demonstrated significant improvements in total FAOS scores (lateral: 53.4 ± 13.8 to 87.2 ± 10.1, P < .001; medial: 51.1 ± 12.6 to 89.7 ± 8.4, P < .001) and in all secondary clinical outcomes at final follow-up (all P < .001), with no intergroup differences (all P > .05). Similarly, there were no significant differences in the total MOCART 2.0 ankle score (73.8 ± 10.7 vs 73.0 ± 9.9; nonsignificant) or in any of its individual subcomponents. CONCLUSION:This cohort study demonstrated that AOPT yields comparable mid‑ to long‑term clinical and radiologic outcomes for medial and lateral large cystic OLTs, and that radiologic results were not significantly associated with clinical outcomes.

Residual postural instability following the modified Broström procedure for chronic ankle instability (CAI) remains a clinical challenge. Current preoperative assessments emphasize structural and demographic factors, neglecting dynamic biomechanical aspects which may better predict functional recovery. This study investigated whether preoperative plantar pressure distribution and gait parameters are associated with early postoperative postural control. A total of fifty-five CAI patients (age, 19 to 40 years) scheduled for a modified Broström procedure underwent smartphone-based gait evaluation before surgery. At three months postoperatively, participants were stratified into stable (n = 28) or unstable (n = 27) groups based on the ability to maintain a single-leg stance for 10 s with eyes open. Preoperative temporal-spatial gait parameters and regional plantar pressure distributions were compared between group. Linear regression model and ROC curve analysis were used to identify predictors of post-operative instability. The unstable group exhibited significantly lower preoperative peak pressure under the ipsilateral fifth metatarsal head (p = 0.042), toes (p = 0.026), and forefoot (p = 0.019), alongside prolonged bilateral step duration and contralateral stance duration. A model containing contralateral stance duration and ipsilateral forefoot peak pressure significantly predicted group allocation. A forefoot pressure below 40 kPa predicted postoperative instability with 84.3

Musculoskeletal disorders represent a significant global health burden and are a leading cause of disability worldwide. While MRI is essential for accurate diagnosis, its interpretation remains exceptionally challenging. Radiologists must identify multiple potential abnormalities within complex anatomical structures across different imaging planes, a process that requires significant expertise and is prone to variability. We developed OrthoDiffusion, a unified diffusion-based foundation model designed for multi-task musculoskeletal MRI interpretation. The framework utilizes three orientation-specific 3D diffusion models, pre-trained in a self-supervised manner on 15,948 unlabeled knee MRI scans, to learn robust anatomical features from sagittal, coronal, and axial views. These view-specific representations are integrated to support diverse clinical tasks, including anatomical segmentation and multi-label diagnosis. Our evaluation demonstrates that OrthoDiffusion achieves excellent performance in the segmentation of 11 knee structures and the detection of 8 knee abnormalities. The model exhibited remarkable robustness across different clinical centers and MRI field strengths, consistently outperforming traditional supervised models. Notably, in settings where labeled data was scarce, OrthoDiffusion maintained high diagnostic precision using only 10% of training labels. Furthermore, the anatomical representations learned from knee imaging proved highly transferable to other joints, achieving strong diagnostic performance across 11 diseases of the ankle and shoulder. These findings suggest that diffusion-based foundation models can serve as a unified platform for multi-disease diagnosis and anatomical segmentation, potentially improving the efficiency and accuracy of musculoskeletal MRI interpretation in real-world clinical workflows.

Background: Bone grafting is a common and effective treatment for anterior shoulder instability. Graft healing is critical for the success of this procedure; however, few studies have investigated methods to enhance the healing process. Furthermore, suitable animal models are scarce for this type of surgery.Purpose: To (1) establish an animal model of anterior shoulder instability, and (2) evaluate whether a surgical modification based on an inlay structure (creating a groove on the glenoid and shaping the graft to match it) along with postoperative administration of the bone anabolic agent parathyroid hormone 1-34 (PTH1-34) could accelerate graft healing.Study Design: Controlled laboratory study.Methods: A rabbit model of anterior shoulder instability was established, and autologous iliac bone grafting was performed. Gross morphological observation, micro-computed tomography imaging and analysis, and histological staining and evaluation were employed to assess whether the inlay-based surgical modification and postoperative intermittent subcutaneous injection of PTH1-34 could enhance graft healing.Results: The modified inlay technique increased the expression of Runx2 and type I collagen within the graft, accelerated graft integration with the glenoid, promoted more rapid callus remodeling and maturation, and reduced graft resorption. Additionally, for both the modified inlay and classic onlay bone grafting procedures, postoperative intermittent subcutaneous injection of PTH1-34 enhanced osteogenic capacity of the autograft and glenoid, increased new bone volume, and shortened the graft healing time.Conclusion: We successfully developed an animal model of autologous bone grafting for anterior shoulder instability. Using this model, we demonstrated that the modified inlay bone grafting procedure improves osteogenic ability, shortens healing time, and promotes callus maturation. Intermittent subcutaneous administration of PTH1-34 after surgery further enhanced graft-glenoid healing.Clinical Relevance: The modified inlay technique and postoperative intermittent PTH1-34 administration may improve graft healing rates after bone grafting procedures for anterior shoulder instability.

BACKGROUND:Insertional reattachment of the Achilles tendon is the recommended surgical treatment for acute sleeve avulsion fractures at the tendon insertion in patients fit for surgery and who are comfortable taking the risks associated with the procedure. However, there is a paucity of evidence regarding the longitudinal clinical outcomes and sustained sports participation at minimum 2- and 5-year follow-up after this procedure. QUESTIONS/PURPOSES:(1) What are the 2- and 5-year patient-reported outcomes after insertional reattachment for acute sleeve avulsion fractures of the Achilles tendon? (2) What proportion of patients return to sports after insertional reattachment for acute sleeve avulsion fractures of the Achilles tendon? (3) What patient factors are associated with inferior patient-reported outcomes, including the VAS, the American Orthopaedic Foot and Ankle Society (AOFAS) Ankle-Hindfoot score, the Foot Function Index (FFI), and Tegner scores, at minimum 5-year follow-up? METHODS:A retrospective study was performed that evaluated patients who underwent insertional reattachment for acute sleeve avulsion fractures of the Achilles tendon between December 2011 and December 2019 at our institution. During this period, we treated 55 patients for this injury. Surgery was generally offered to patients who presented with acute posterior heel pain and functional loss consistent with insertional rupture, radiographic or MRI evidence of a proximally displaced avulsion fragment, and a Pomeranz classification ≥ Type IB; these criteria were applied consistently throughout the study period. Patients with anesthesia contraindications, immunosuppressive disease, or those who refused surgery were treated nonoperatively and were not included. Five patients met the exclusion criteria, leaving 50 patients, of whom 78% (39) completed both minimum 2- and 5-year follow-up; their results were analyzed in this study. Baseline data were recorded: mean ± SD age was 45 ± 11 years, 92% (36 of 39) of patients were men, mean ± SD BMI was 27.2 ± 2.9 kg/m 2 , 69% (27 of 39) of patients sustained high-energy injuries, 41% (16 of 39) of patients had preinjury insertional pain, and 10% (4 of 39) of patients had prior steroid injection. Patient-reported outcomes (PROs)-including the VAS for pain, the AOFAS Ankle-Hindfoot score, the FFI, and the Tegner activity scale-were collected preoperatively, at minimum 2- and 5-year follow-up. Sports participation, time to return to sport, single-leg heel-rise performance, delayed wound healing, and rerupture were also assessed. A Spearman correlation analysis was used to explore factors associated with minimum 5-year PROs. RESULTS:Overall, PROs consistently improved and remained improved out to 5 years postoperatively. VAS pain scores improved at 2 years and remained improved at 5 years (4.7 ± 1.3 preoperatively versus 1.6 ± 1.4 at 2 years and 0.7 ± 1.0 at 5 years, mean differences -3.1 and -0.9, respectively; p < 0.001 for both). AOFAS scores improved at 2 years and remained improved at 5 years (50 ± 17 preoperatively versus 88 ± 7 at 2 years and 96 ± 5 at 5 years, mean differences 38 and 8, respectively; p < 0.001 for both). FFI scores improved at 2 years and remained improved at 5 years (110 ± 20 preoperatively versus 11 ± 10 at 2 years and 5 ± 4 at 5 years, mean differences -99 and -6, respectively; p < 0.001 for both). Tegner scores improved at 2 years and remained stable at 5 years (1 ± 1 preoperatively versus 4 ± 1 at 2 years and 4 ± 2 at 5 years, mean differences 3 and 0, respectively; p < 0.001 and p = 0.18). At minimum 5-year follow-up, 64% (25 of 39) of patients were engaged in sports activities, with a mean ± SD time to return to sport of 8 ± 3 months. For those with preinjury sporting demands, 78% (25 of 32) were still participating in sports at minimum 5-year follow-up. In all, 90% (35 of 39) of patients completed a single-leg heel-rise test. Delayed wound healing occurred in 13% (5 of 39), and rerupture occurred in 8% (3 of 39). Increasing age was modestly associated with poorer 5-year Tegner activity score (r = -0.47 [95% CI -0.71 to -0.16]; p = 0.002). A preinjury history of insertional pain was modestly to substantially associated with poorer postoperative VAS (r = 0.32 [95% CI 0.00 to 0.62]; p = 0.047), AOFAS (r = -0.35 [95% CI -0.63 to -0.04]; p = 0.03), and FFI scores (r = 0.52 [95% CI 0.24 to 0.73]; p = 0.001). Similarly, a history of corticosteroid injection was negatively correlated with FFI scores (r = 0.41 [95% CI 0.16 to 0.60]; p = 0.01). Calcaneoplasty performed during surgery was positively correlated with better postoperative AOFAS scores (r = 0.36 [95% CI 0.17 to 0.52]; p = 0.03). No correlations were found between clinical outcomes and other variables. CONCLUSION:Insertional reattachment for acute Achilles sleeve avulsion fractures yields meaningful and durable improvements in pain, function, and activity level at minimum 2- and 5-year follow-up. Most patients experience continued improvement over time, with stable sports participation at the minimum 5-year follow-up. Importantly, among patients engaged in sports preinjury, nearly 80% maintained sports participation at 5 years, highlighting the durability of this operation. However, a history of preinjury insertional pain and prior corticosteroid injection, features suggestive of underlying tendinopathy, may be associated with poorer minimum 5-year outcomes. LEVEL OF EVIDENCE:Level III, therapeutic study.

Aims:An os subfibulare (OS) is frequently seen in adolescents with chronic lateral instability of the ankle, potentially impairing participation in sport and the quality of life. The aim of this study was to compare the clinical outcomes of the modified Broström-Gould procedure in adolescent patients with chronic lateral instability of the ankle with and without an OS. Methods:We retrospectively reviewed adolescent patients who underwent a modified Broström-Gould procedure for chronic instability of the ankle between August 2019 and July 2023. Propensity score matching (1:1) was undertaken based on age, sex, side, BMI, the duration of symptoms, pre-injury Tegner score, and joint hypermobility, yielding 82 matched cases (41 OS and 41 non-OS). Preoperative imaging parameters and intraoperative findings were recorded. Primary outcomes included patient-reported outcome measures (PROMs): a visual analogue scale (VAS) for pain, the Karlsson score, the Tegner score, and the Foot and Ankle Outcome Score (FAOS). Secondary outcomes included the incidence of re-sprain, the time to return to sport, satisfaction, and complications. Logistic regression analysis identified risk factors for re-sprain. Results:The mean follow-up was 47.8 months (SD 11.2) in the OS group and 47.9 months (SD 9.6) in the non-OS group (p = 0.949). Both groups had significant improvements in all PROMs, with no differences between the groups for the VAS, Karlsson score, Tegner score, or FAOS. Re-sprain occurred in five in the OS group (12.2%) and 12 in the non-OS group (29.3%) (p = 0.081). Multivariate analysis identified higher pre-injury Tegner score (p = 0.001) and concomitant calcaneofibular ligament injury (p = 0.028) as significant independent risk factors for re-sprain. Conclusion:In adolescent patients with chronic lateral instabilty of the ankle, the presence of an OS does not compromise the outcomes of a modified Broström-Gould procedure. Excision of the OS combined with ligamentous repair provides improvements which are comparable to those of an isolated repair, representing a safe and effective form of surgical treatment. Nonetheless, greater attention should be given to patients with high sporting demands or concomitant calcaneofibular ligament injury, due to their increased risk of re-sprain postoperatively.

OBJECTIVES:This study aimed to develop KOA-Diff, a multimodal diffusion model predicting future knee radiographs for accurate, visually interpretable forecasting of knee osteoarthritis (KOA) progression. METHODS:KOA-Diff utilizes a latent diffusion backbone and a multimodal fusion network to integrate baseline X-rays, MRIs, and clinical characteristics. A feature-targeting attention mechanism guides image generation toward critical anatomical structures. The model was trained and internally validated on the Osteoarthritis Initiative (OAI) dataset and externally validated on a private cohort (PUTH-KOA, n=80). We evaluated the synthesized future radiographs using joint space width error (JSWE, error relative to the actual width) and binary Kellgren-Lawrence (KL) grade accuracy (early [KL ≤2] vs. advanced [KL ≥3]). Additionally, blinded experts graded structural features. Finally, clinical utility was tested by using model outputs to assist clinicians in identifying high-risk patients exhibiting a longitudinal KL grade increase ≥2 levels over 8 years. RESULTS:In the OAI dataset, KOA-Diff achieved a JSWE of 0.71%-1.21% and binary KL-grade accuracy of 84.0%-91.8% over 96 months. External validation demonstrated promising performance with 90.0% binary KL-grade accuracy and 2.06% JSWE. In the blinded evaluation, experts rated the accuracy of synthesized features on a 5-point scale. For the OAI cohort, the model scored 4.162 (joint space narrowing), 3.895 (osteophyte formation), and 3.943 (subchondral bone sclerosis). PUTH-KOA scores were consistently high (4.050, 3.925, 3.775, respectively). With model assistance, clinicians' accuracy in identifying high-risk patients improved from 58.3% to 77.5%. CONCLUSIONS:KOA-Diff synthesizes future radiographs reflecting accurate structural progression, providing an interpretable tool for KOA prediction.

Anterior cruciate ligament (ACL) injury is a prevalent sports-related trauma with significant impact on both professional athletes and young individuals. While often viewed as accidental events, high recurrence rates suggest the presence of underlying structural risk factors. This study investigated the association between 3D bone morphology and ACL injury susceptibility. Leveraging a multicenter dataset of 5000 MRIs, we developed ACL-injury Patterning (ACL-P), a deep learning model that utilizes 3D bone morphology point clouds for injury classification. Given the short-term stability of bone geometry following trauma, this classification serves as a proxy task to quantify the correlation between pre-existing morphological traits and ACL injury through discriminative performance. The model's discriminative performance was validated across an external civilian test set (n = 7797) and a professional athlete test set (n = 269), yielding an Area Under the Curve (AUC) of 0.887 (95% CI: 0.879-0.895) and 0.907 (95% CI: 0.872-0.941), respectively. Robustness checks confirmed that the model's performance remained consistent across both acute and chronic injury phases (0-365 days) and was independent of secondary joint dislocation. These findings establish 3D bone morphology as a significant anatomical factor linked to injury susceptibility, enhancing our understanding of the structural correlates of knee trauma.