BMC Musculoskeletal Disorders最新文献

Objective: Osteonecrosis of the femoral head (ONFH) is a common disabling hip disorder with an annually increasing incidence due to factors such as corticosteroid administration. Core decompression is an effective surgical intervention for early-stage ONFH, and robot-assisted technology can enhance the accuracy and minimally invasiveness of this procedure. However, its clinical efficacy requires comprehensive evaluation with high-quality evidence. This study aimed to assess the clinical efficacy and safety of robot-assisted core decompression (RACD) in the treatment of ONFH.

Methods: We systematically searched Chinese and English databases up to December 1, 2025, for case series, cohort studies, and randomized controlled trials (RCTs) on RACD for ONFH. The primary outcome was the femoral head collapse rate; secondary outcomes included the Harris Hip Score (HHS), Visual Analogue Scale (VAS) score, and postoperative complications. Meta-analysis was performed using fixed-effects or random-effects models, with subgroup and sensitivity analyses conducted to verify results.

Results: A total of 12 studies were included (1 RCT, 9 cohort studies, 2 case series), involving 325 patients. Meta-analysis showed that the overall femoral head collapse rate after robot-assisted core decompression was 9% (95% CI: 7%-13%). In the subgroup of ARCO Stage II patients with follow-up > 1 year, the femoral head collapse rate was 12% (95% CI: 8%-17%). The mean postoperative improvement in HHS was 17.38 points (95% CI: 14.40-20.37), and the mean postoperative improvement in VAS score was 2.75 points (95% CI: 2.01-3.50). No complications such as infection or neurovascular injury were reported in any of the studies that reported complications (n = 7, 222 patients in total). Sensitivity analysis indicated that the results were robust. Egger's test suggested potential publication bias for HHS, VAS, and femoral head collapse rate; however, the adjusted effect sizes still supported clinical benefit. Given the high heterogeneity, the adjusted effect sizes should be interpreted with caution.

Conclusion: Available evidence supports the effectiveness of robot-assisted core decompression in the treatment of early-stage ONFH. RACD yields functional improvement and pain relief in patients with ONFH, with acceptable short-to-medium-term safety. Current evidence, however, are insufficient to confirm that RACD yields superior hip preservation efficacy compared with MCD. Further direct, high-quality comparative studies are required to determine and quantify the benefits of RACD compared to MCD.

Trial registration: PROSPERO (CRD420251266114).

Objective: To assess the effect of postoperative anti-osteoporosis therapy on titanium mesh subsidence and bone fusion time after anterior cervical surgery for spinal tuberculosis.

Methods: In this retrospective study, 66 patients undergoing anterior cervical debridement with bone grafting and fixation (2020-2024) were divided into two groups: Group A (full-course anti-osteoporosis therapy, n = 33) and Group B (no/incomplete therapy, n = 33). The primary endpoints were titanium mesh subsidence and bone fusion time. Secondary endpoints included VAS score, NDI score, Frankel grade, bone mineral density, and radiographic alignment. Clinical and radiographic outcomes were compared.

Results: The mean follow-up was 12.5 months. Neurological function (Frankel score) improved in all patients. The titanium mesh subsidence rate was significantly lower in Group A (21.2%) than in Group B (45.3%). Bone fusion time was significantly shorter in Group A (3.5 ± 0.3 months) versus Group B (4.8 ± 0.5 months). Both groups showed significant improvements in VAS and NDI scores. Radiographic alignment was satisfactory. Bone mineral density improved significantly in Group A compared to both its baseline and Group B.

Conclusion: Anterior cervical surgery is effective for spinal tuberculosis. Postoperative standardized anti-osteoporosis therapy further improves outcomes by reducing implant subsidence and accelerating bone fusion, offering significant clinical benefit.

Purpose: To determine whether adjunctive blood flow restriction (BFR) training during postoperative weeks 13-20 is associated with differences in functional, strength, balance, muscle morphology, and neuromuscular outcomes compared with sham BFR following anterior cruciate ligament reconstruction (ACLR).

Methods: In this single-centre randomized controlled trial, 48 patients aged 18-35 years who underwent primary unilateral ACLR were randomly assigned to a BFR group or a Sham-BFR group. From postoperative weeks 13 to 20, both groups performed identical low-load resistance training (30% one-repetition maximum) twice weekly. The BFR group received individualized blood flow restriction at 40% arterial occlusion pressure, whereas the Sham-BFR group underwent the same protocol with minimal cuff pressure. Outcomes assessed at postoperative week 24 included the International Knee Documentation Committee (IKDC) score, Tegner Activity Scale, knee range of motion (ROM), isometric knee extensor and flexor strength, quadriceps muscle thickness, Y-Balance Test performance, and quadriceps surface electromyography (sEMG).

Results: Forty-three patients completed the study (BFR, n = 21; Sham-BFR, n = 22). At 24 weeks postoperatively, the BFR group demonstrated statistically significantly higher IKDC score and Tegner Activity Scale than the Sham-BFR group (p < 0.01). Knee extensor and flexor strength, Y-Balance Test composite scores, and quadriceps sEMG amplitudes were also significantly greater in the BFR group (p < 0.05). No significant between-group differences were observed in knee ROM or quadriceps muscle thickness.

Conclusion: Adjunctive BFR training during mid-stage ACLR rehabilitation was associated with more favourable functional, strength, balance, and neuromuscular outcomes than sham BFR at 24 weeks postoperatively. These findings should be interpreted cautiously because baseline outcome measurements and immediate post-intervention assessments were not available.

Trial registration: (Chinese Clinical Trial Registry (https://www.chictr.org.cn), No. ChiCTR2400087631, 31/07/2024).

Accurate, objective assessment of hip joint range of motion (ROM) is essential for orthopedic diagnosis and rehabilitation. Conventional tools, such as goniometers, are limited by subjectivity, inter-observer variability, and poor compatibility with telemedicine applications. This study aimed to develop and validate a markerless, video-based system for estimating hip joint ROM by integrating human pose estimation (MediaPipe) with machine-learning models using inertial sensor data as the reference standard. Twenty healthy adult males performed hip flexion/extension, abduction/adduction, and internal/external rotation movements. Skeletal coordinates extracted from the videos were converted into geometric features and used to train five regression models (linear regression, ElasticNet, support vector regression, random forest, and LightGBM). Model performance was evaluated using the coefficient of determination (R²) and mean absolute error (MAE), and Shapley additive explanations (SHAP) were used to interpret feature contributions. Among the evaluated models, tree-based methods showed high predictive accuracy, and LightGBM was used for subsequent interpretability analyses across all directions of hip ROM (R² up to 0.94; MAE 4-6°), while inertial sensor validation confirmed high measurement accuracy of the reference data. SHAP analysis revealed that distinct geometric descriptors dominated flexion/extension, abduction/adduction, and internal/external rotation, indicating direction-specific biomechanical determinants of hip motion. These findings demonstrate that markerless, video-based estimation of hip ROM is feasible with promising accuracy and highlight the potential of this approach for telemedicine, remote rehabilitation monitoring, and future integration into gait and musculoskeletal modeling workflows.

Background: This meta-analysis aims to determine the prevalence of postoperative shoulder stiffness (PSS) after rotator cuff repair surgery and to explore its potential risk factors.

Methods: Search the PubMed, Embase, Web of Science, and Cochrane Library databases from their inception to July 1, 2025, for all studies reporting the prevalence of PSS after rotator cuff repair surgery and its associated risk factors. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of the included studies. Meta-analysis was performed using Stata 15.0 software.

Results: Eight cohort studies involving 21,033 patients were included. The pooled prevalence of postoperative shoulder stiffness (PSS) after rotator cuff repair was 17% (95% CI: 9%-24%), with substantial heterogeneity (I² = 98.2%). Subgroup analysis by continent showed variation in reported prevalence estimates across regions, although these findings should be interpreted cautiously because some subgroups were based on a limited number of studies. Additional subgroup analyses according to sample size, publication year, and mean age were conducted to explore heterogeneity. In the sample size subgroup analysis, the pooled prevalence was 14% (95% CI: 2%-25%) in studies with a sample size ≥ 300 and 18% (95% CI: 15%-21%) in studies with a sample size < 300, with lower heterogeneity in the latter subgroup. By contrast, subgroup analyses by publication year and mean age did not materially reduce heterogeneity. In pooled analyses of adjusted effect estimates, age < 50 years (OR = 1.09, 95% CI: 1.03-1.15), female sex (OR = 1.66, 95% CI: 1.03-2.68), and diabetes mellitus (OR = 2.73, 95% CI: 1.75-4.26) were associated with PSS; however, the effect size for age < 50 years was small.

Conclusion: PSS after rotator cuff repair is not uncommon, but the pooled prevalence should be interpreted cautiously because of substantial heterogeneity and inconsistent diagnostic definitions across studies. Female sex and diabetes mellitus may be associated with PSS, whereas the clinical relevance of the association with age < 50 years appears limited. Further high-quality prospective studies are required.

Background: Differentiating acute from chronic wedge-shaped thoracolumbar vertebral deformities on conventional lateral lumbar radiographs remains clinically challenging, especially when osteoporosis status also needs to be considered. This study aimed to develop and evaluate a You Only Look Once (YOLO)v8n framework for vertebral-level detection and classification of thoracolumbar fractures with osteoporosis-related stratification on lateral lumbar radiographs.

Methods: We retrospectively collected 1352 lateral lumbar radiographs from 1352 patients, with one radiograph per patient. A total of 1774 vertebral fracture segments were manually annotated. Lumbar magnetic resonance imaging (MRI) and dual-energy X-ray absorptiometry (DXA) were used as reference standards to stratify vertebral targets into three categories: acute fracture with osteoporosis, acute fracture without osteoporosis and chronic fracture with osteoporosis. The dataset was divided into training and validation subsets at the patient level. A YOLOv8n detector was trained as the primary model. To strengthen methodological rigor, additional baseline comparison experiments were conducted under the same patient-level training/validation split using YOLOv5n and Faster R-CNN. Detection performance was assessed using precision, recall, F1-score, mean average precision (mAP) 50 and mAP50-95.

Results: On the validation set, the YOLOv8n model achieved a precision of 0.495, recall of 0.482, F1-score of 0.490, mAP50 of 0.506, and mAP50-95 of 0.397. In comparative experiments, YOLOv5n achieved a precision of 0.451, recall of 0.549, F1-score of 0.495, mAP50 of 0.494, and mAP50-95 of 0.367, whereas Faster R-CNN achieved a precision of 0.273, recall of 0.814, F1-score of 0.409, mAP50 of 0.300, and mAP50-95 of 0.217. These findings indicate that YOLOv8n provided the most balanced overall detection performance in the present dataset.

Conclusion: The proposed YOLOv8n framework demonstrated preliminary feasibility for automated vertebral-level detection and classification of thoracolumbar fractures with osteoporosis-related stratification on lateral lumbar radiographs. However, given the moderate overall performance and lack of external validation, the current model should be regarded as an assistive screening tool rather than a standalone diagnostic system.

Background: Subtrochanteric femur fractures are challenging to treat due to complex anatomy and high mechanical stresses. Cephalomedullary nailing is the standard treatment, often performed with the patient supine on a traction table, but this can prolong setup and cause complications (pressure or nerve injuries). Lateral decubitus positioning without a traction table may facilitate reduction and reduce traction-related risks, yet comparative data in subtrochanteric fractures are limited. This study compared outcomes of lateral decubitus versus traction table positioning for subtrochanteric fracture nailing.

Methods: A retrospective cohort study included 68 patients with subtrochanteric femur fractures treated with cephalomedullary nails. Patients underwent surgery supine on a traction table (n = 37) or in lateral decubitus without a traction table (n = 31). Operative metrics (entry point determination time, operative time, fluoroscopy duration, and blood loss) and postoperative outcomes (transfusion requirements, time to mobilization, complications, and final Harris Hip Score) were compared between groups.

Results: Lateral decubitus positioning was associated with shorter entry-point determination time (median 2 vs. 7 min; p = 0.05), shorter operating-room time (80.3 ± 23.6 vs. 108.2 ± 23.2 min; p = 0.03), reduced fluoroscopy time (151.2 ± 19.2 vs. 178.2 ± 16.1 s; p = 0.03), and lower estimated blood loss (752 ± 346 vs. 1459 ± 611 mL; p = 0.01) compared with traction-table positioning. No statistically significant differences were observed in transfusion requirements, time to mobilization, final Harris Hip Score, or complication rates (all p > 0.05).

Conclusion: Compared with traction-table positioning, lateral decubitus cephalomedullary nailing was associated with better functional outcome, shorter operative and fluoroscopy times, and lower estimated blood loss, while union and complication rates were similar between groups. These findings suggest that lateral decubitus positioning is a viable alternative for subtrochanteric fracture fixation, particularly when traction-table setup may compromise operating-room efficiency. Given the observational design and potential confounding related to reduction techniques and surgeon factors, these findings should be interpreted as associations; prospective studies are warranted.

Background: Epidemiological data on injuries and fractures in children under two years of age is limited, thus, we aimed to explore the incidence, patterns and mechanisms.

Methods: Retrospective, population-based cross-sectional study including children under two, seen at the A&E department due to a trauma, or having a radiograph due to birth trauma. All radiographs were re-reviewed.

Results: Four hundred thirty children (49.8% female), mean age 14.7 months (SD 6.8), were included, of whom four had a high-energy trauma and 42 were birth-related. Of 388 children (50.3% female) (mean age 16.3 months, SD 5.1 months) with non-birth related injury, 163 (42.0%) had a fracture (annual incidence 4.3 per 1,000 children; 2.4 per 1,000 infants vs. 6.1 per 1,000 children aged 12 to 24 months of age. 47/163 (28.8%) fractures involved the forearm and 41 (25.2%) involved the leg. Fracture mechanisms 39.9% fall from more than own height/furniture; 17.2% fall from own height, 6.7% crush injury and 6.1% dropped by parent. In 12.9% of the traumas, no mechanism was provided. The number of fractures increased significantly by age group (p=0.027), and the distribution differed, with skull fractures predominating in 0-6-months-old (33.0%), clavicle fractures (33.0%) in 6-12-months-old and forearm fractures in the two older age groups (35.0% and 34.4%, respectively) (p < 0.001). No classic metaphyseal lesions were found in the entire cohort. Four children, all of whom had radiological "red flags", had inflicted injury. Forty-two children had birth-related injury, of whom 50.0% had a fracture (1.1 per 1,000 live births).

Conclusions: Fractures in infants are rare, in particular classic metaphyseal fractures. The occurrence of radiological "red flags" should raise suspicion of non-accidental injury and instigate further assessment.

Background: The surgical rate for distal radius fractures is steadily rising despite limited evidence of its benefits over non-surgical treatment. Using a natural experimental approach, we aimed to compare patient-reported outcomes following surgical versus non-surgical treatment of distal radius fractures.

Methods: Registered in the Swedish Fracture Register by 36 Swedish hospitals from 2013 to 2018, we included a cohort of 13,356 fractures on 13,031 patients aged 18 years or older with distal radius fractures Arbeitsgemeinschaft fur Osteosynthesefragen (AO) 23-A2.1-2, A3, and C1-C3. The observational study utilized differences in the frequency of surgical treatment across hospitals as a source of random treatment assignment and a natural experiment. We assumed that all hospitals encountered a similar range of fractures each year. Therefore, the annual frequency of surgery per hospital was used as a proxy for randomization between surgical and nonsurgical treatment, regardless of each patient's actual treatment. The outcome was the individual Patient Reported Outcome Measures (PROM) at 1 year, with the Arm and Hand Function Index from the Short Musculoskeletal Function Assessment (SMFA) as the primary measure.

Results: The surgical rate per hospital year ranged from 7 to 66%. Surgical treatment was associated with lower Arm and Hand Function Index scores in comminuted intraarticular fractures of type C2 (11.9 units, p = 0.004) and type C3 (19.4 units, p = 0.029). There was a tendency for a positive association with surgical treatment in dorsally angulated extraarticular fractures (23A2.2), but the difference of 5.1 units (p = 0.079) was below the Minimal Clinically Important Difference (MCID). In other extra-articular fractures (23-A2.1 and 23-A3) and simple intra-articular fractures (23-C1), the benefits of surgical treatment were small and also not statistically significant. Several sensitivity analyses were conducted to test the study design, and all supported the primary results.

Conclusions: In this comparison of surgical and non-surgical treatment for distal radius fractures across hospitals with varying surgical rates, comminuted intra-articular distal radius fractures (AO 23-C2/C3) treated surgically were associated with better one-year patient-reported outcomes than those treated non-surgically.

Background: Differentiating between spinal tuberculosis, pyogenic (bacterial) spondylitis and spinal metastasis remains a major diagnostic challenge because their radiological features often overlap. Delayed or incorrect diagnosis may lead to inappropriate treatment, permanent disability or death.

Objective: To develop and evaluate deep learning models for automated classification of spinal tuberculosis, pyogenic infection, and spinal metastasis using magnetic resonance imaging (MRI).

Methods: T2-weighted sagittal MRI scans from 120 patients (40 per disease group) with pathologically or microbiologically confirmed diagnoses between 2014 and 2019 were retrospectively analyzed. Lesion regions were manually annotated by radiologists, and data were split into 80% training and 20% testing sets at the patient level. Extensive data augmentation (rotation ± 5°, zoom 1.1-1.2×, shearing ± 5°, grid distortion 2 × 2) was applied to mitigate overfitting. Three models were trained and compared: (1) a single-layer perceptron baseline, (2) a custom dense neural network (2 × 1024 neurons), and (3) pre-trained convolutional neural networks (ResNet50, VGG16, InceptionV3). Model performance was evaluated using accuracy, precision, recall, and F1-score on both whole and segmented images.

Results: After augmentation, 1,000 synthetic samples were generated per class. The baseline model achieved 27-33% accuracy, whereas the dense and pre-trained models achieved 98-100% accuracy on the test set. Although pre-trained networks demonstrated marginally higher performance, the difference compared with the dense model was not statistically significant. Activation heatmaps revealed inconsistent localization of attention regions, suggesting potential overfitting and limitations in visualization interpretability.

Conclusion: Deep learning models demonstrated strong potential in distinguishing between spinal tuberculosis, bacterial spondylitis, and spinal metastasis on MRI. However, the near-perfect performance likely reflects dataset homogeneity and augmentation effects rather than full generalization. External, multi-center validation and improved interpretability methods (e.g., Grad-CAM) are warranted to confirm clinical applicability and ensure reliable decision support for radiologists.