Citations

Abstract Computer vision now enables deep learning models to link histopathologic features to molecular features in a quantitative manner not previously achievable by human pathologists. However, human interpretability is often limited. We trained a deep neural network to predict gene expression from digital whole slide images (WSIs) using 848 central nervous system (CNS) tumors with paired WSI and RNA-sequencing data. We used inferred RNA expression levels as a surrogate for protein expression of the CNS tumor lineage markers GFAP, OLIG2, and SSTR2, as well as the proliferation marker MKI67 (Ki67). We established thresholds for categorizing the inferred expression levels as positive or negative based on levels observed in cross-validation testing in select tumor types to approximate routine clinical interpretation of immunohistochemical (IHC) staining. We tested the sensitivity and specificity of our ‘virtual’ IHC on an independent, multi-institutional cohort of over 2,000 CNS tumors with objective diagnostic labels derived from DNA methylation-based tumor classification. As a dichotomous variable, inferred GFAP expression showed a sensitivity of 74% and a specificity of 99% as a glial marker in a cohort of gliomas and meningiomas. In the same cohort, OLIG2 showed a sensitivity of 75% and a specificity of 97% as a glial marker, while SSTR2 showed a sensitivity of 83% and a specificity of 97% as a marker of meningioma. In a cohort of ependymomas and gliomas, virtual IHC for OLIG2 showed a sensitivity of 75% and a specificity of 82% in distinguishing ependymomas from gliomas. Finally, in a cohort of gliomas, inferred expression of MKI67 demonstrated a trend consistent with what would be expected by actual IHC, with increasing MKI67 expression with increasing tumor grade. Our model provides the basis for a human interpretable and clinically applicable deep neural network to aid human pathologists in the diagnosis and grading of CNS tumors.

Abstract Molecular testing has become critical components in the diagnostic classification of the central nervous system (CNS) tumors. However, these methods require substantial resources, limiting accessibility for many patients. Recent advances in artificial intelligence (AI) and computer vision empower deep learning models to infer molecular features from histopathology images, which may often be more readily available than molecular testing, to classify CNS tumors. We trained pan-CNS tumor models to predict DNA methylation and gene expression from whole-slide images (WSIs). These molecular predictions were then used in a hierarchical machine learning framework, termed Neuropath-AI, to predict nine broad tumor families and 52 tumor types using a large diverse cohort of 5,715 histopathology samples, of which 2,988 had paired DNA methylation profiling and 848 had paired RNA-sequencing. We evaluated Neuropath-AI on another large independent multi-institutional cohort of 5,516 CNS tumors, for which Neuropath-AI predicted tumor class with an associated confidence score. Neuropath-AI designated 46% of test samples as predictable with high-confidence, which it then accuracy classified in 97% of samples. It made moderate-confidence (and above) predictions for 87% of test samples, for which it achieved a diagnostic accuracy of 80% for the top-1 prediction and 86% accuracy for a top-2 predictions. A comparison with human neuropathologists showed that Neuropath-AI is comparable at classifying tumors. A second test showed that human neuropathologists were more accurate when the Neuropath-AI classifier results were provided to them, testifying to the potential translational benefit of integrating AI classification tools into the pathologist’s workflow. Our model provides the basis for a clinically applicable deep learning assistant to improve human efficiency and accuracy of CNS tumor diagnosis. The model will be made publicly available and can be readily implemented to assist human pathologists in clinical workflows, in further expanded prospective studies.

Aims Heterotopic ossification (HO) is a recognized complication following periacetabular osteotomy (PAO), but its patterns, incidence, and clinical relevance remain poorly defined. The commonly used Brooker classification, developed for total hip arthroplasty, is not fully applicable to HO after PAO. This study aimed to propose a new classification system specific to HO after PAO, evaluate its clinical relevance, and report the cumulative probabilities of radiological and symptomatic HO. Methods This retrospective study included 643 patients who underwent PAO between January 2006 and July 2024. A total of 308 patients (47.9%) had concomitant arthroscopic procedures. Radiographs were analyzed to identify HO using both the Brooker classification and a newly developed system. The cumulative probabilities of radiological and symptomatic HO were calculated using Kaplan-Meier analysis. Difference of variables between patients with and without HO was analyzed. Receiver operating characteristic (ROC) curve analysis was used to compare the predictive performance of the two classification systems. Results The cumulative probabilities of radiological and symptomatic HO at five years post-PAO were 30.4% and 3.8%, respectively. The probabilities were 41.9% in males and 29.2% in females. Four patients (2.8%) required surgical excision of HO, all classified as type 3 or 4 under the new system. The ROC analysis demonstrated superior predictive performance for the new classification system (AUC = 0.928) compared to the Brooker classification (AUC = 0.697; p = 0.007). No significant differences were found between the patients with and without HO in terms of age, sex, concomitant arthroscopic procedures, or receiving prophylaxis in addition to low-dose aspirin. Conclusion This study highlighted the cumulative probabilities of HO following PAO and introduced a new classification system that improved clinical relevance and predictive accuracy. These findings have provided insights into the incidence and management of HO after PAO, and laid the groundwork for standardized reporting and future research. Cite this article: Bone Jt Open 2025;6(11):1436–1445.

Background: Hip decompression effectively treats early-stage osteonecrosis of the femoral head (ONFH) by slowing disease progression and potentially delaying joint replacement. Biological adjuvants like bone marrow aspirate concentrate (BMAC) and platelet-rich plasma (PRP) support bone regeneration and improve outcomes 1–7 . The present video article demonstrates a simple, coreless hip decompression technique with BMAC and PRP injection for early-stage ONFH. Description: The procedure is performed in the same operating room setting as traditional core decompression, with the patient supine on a radiolucent table for fluoroscopic guidance. One or both legs are draped free for access to the iliac crests. Bone marrow is harvested percutaneously from the anterior superior iliac crest with a trocar needle kit, centrifuged, and prepared for injection. We recommend precoating needles and syringes with 1:1,000 heparin to prevent clotting. The BioCUE System (Zimmer Biomet) is typically utilized for centrifugation. Hip decompression is performed with use of a trocar and cannula (PerFuse System; Zimmer Biomet), with subsequent injection through the cannula into the femoral head. A 0.5-cm skin incision is made. The trocar is placed lateral to the femur and advanced percutaneously through the lateral femoral cortex, with a starting point proximal to the lesser trochanter. The trocar is then advanced along the femoral neck into the necrotic region by performing mallet strikes on the instrument’s strike cap. Anteroposterior and frog-leg lateral views assist in positioning the trocar within the necrotic area. Internal leg rotation, which aligns the patella upward, helps position the trocar horizontally parallel to the floor. Positioning is adjusted using repeated imaging as needed. Once the patient is positioned, the trocar is removed, leaving the cannula in place. With the cannula retracted 1 cm, a 30-mL syringe is utilized to inject BMAC and PRP into the necrotic lesion. Because of sclerotic resistance, substantial pressure is needed, but retraction of the cannula helps. Following injection, the cannula is withdrawn another 1 cm, and demineralized bone matrix is injected to prevent escape of the BMAC. Alternatives: Alternative treatments for ONFH include traditional core decompression with a sliding hip screw drill or an X-REAM device (Stryker), both of which carry a higher risk of fracture because of the larger diameter of the tract and require limited weightbearing postoperatively. Bone-cement injection can stabilize the femoral head but lacks regenerative properties. Core decompression with either BMAC or PRP alone, rather than in combination, also serves as an alternative treatment strategy. Open approaches, like osteotomy, are more invasive, have longer recovery times, and may complicate future hip arthroplasty if unsuccessful. Rationale: This technique enables minimally invasive hip decompression and delivery of adjuvant cell therapy or grafting, typically without the use of power instruments. This approach avoids the risk of injuring the bone due to the heat from power tools, protecting the BMAC injection site. Patients are generally discharged the same day and permitted full weight-bearing immediately, even in bilateral surgeries. Expected Outcomes: Hip decompression for ONFH has shown variable rates of success 8,9 , but adding BMAC or PRP may improve outcomes 1–3 . Houdek et al. reported that among 35 hips treated with decompression plus BMAC and PRP for corticosteroid-induced ONFH, 88% avoided THA at 3 years 2 and 70%, at 7 years 3 . Patients with grade-1 or 2 Kerboul angles had a 90% survivorship rate, underscoring the benefits of BMAC and PRP. Important Tips: Insert the trocar into the lateral cortex, positioned distal to the vastus ridge and proximal to the lesser trochanter, to reduce iatrogenic subtrochanteric fracture risk. Avoid advancing closer than 5 mm to the subchondral cortex to prevent joint-surface disruption or collapse, especially with eccentric lesions. If resistance occurs during injection, retract the cannula a few millimeters laterally to increase delivery space and reduce pressure. Acronyms and Abbreviations: BMAC = bone marrow aspirate concentrate ONFH = osteonecrosis of the femoral head PRP = platelet-rich plasma AP = anteroposterior THA = total hip arthroplasty ARCO = Association Research Circulation Osseous classification MRI = magnetic resonance imaging

Abstract The Guardian table is widely utilized for combined hip arthroscopy and periacetabular osteotomy (PAO) due to its postless traction design and the ability to perform both procedures without the need for redraping. This retrospective cohort study reports one surgeon’s use of the Guardian table extension during combined hip arthroscopy and PAO and the occurrence of intraoperative posterior column fractures. Seventy-two consecutive PAO procedures performed at a single institution between February 2023 and January 2025 were analysed. Procedures involving concomitant hip arthroscopy (n = 26) utilized the Guardian table, whereas those without arthroscopy (n = 46) used a conventional radiolucent table. Patient demographics, clinical data, and complications were reviewed. Additionally, mechanical testing quantified table stability by measuring peak absolute acceleration (‘table bounce’) using accelerometry during simulated malleting. The incidence of intraoperative posterior column fractures was significantly higher with the Guardian table (4/26, 15.4%) compared to the conventional table (0/46, 0%; P = .0062). Mechanical testing demonstrated significantly greater table bounce for the Guardian table (69.1 ± 3.4 m/s2 versus 28.2 ± 4.4 m/s2, P < .001). Patient characteristics and surgical complexity did not differ significantly between fracture and non-fracture cases. These findings suggested that the Guardian table extension system was associated with increased risk of posterior column fractures, likely due to lack of distal support and stability. Surgeons should consider alternative table setups or intraoperative support modifications to mitigate this risk during PAO procedures.

AimsPeriacetabular osteotomy (PAO) is a complex procedure for treating hip dysplasia by reorienting the acetabulum to slow osteoarthritis progression. This study aimed to assess the types and rates of complications after PAO and provide recommendations for prevention.MethodsA total of 715 consecutive patients underwent unilateral PAO by a single surgeon between December 2006 and January 2024, with at least one year of follow-up. Indications included hip dysplasia, acetabular retroversion, and protrusio. Hip arthroscopy was performed concurrently in 325 cases. Demographic factors, perioperative data, and complications – graded using the modified Clavien-Dindo system – were analyzed. Risk factors and learning curves were also assessed.ResultsOf the 715 patients, 121 (16.6%) experienced 144 complications. There were 68 grade I, 52 grade II, 23 grade III, one grade IV, and 0 grade V complications. The 23 grade III complications requiring reoperations included four revision PAOs for overcorrection, three fixations for posterior column fracture, four neurolysis procedures for symptomatic lateral femoral cutaneous nerve (LFCN) dysesthesias, four heterotopic bone excisions, and eight wound debridements. All type I and II complications were treated without sequelae. The addition of hip arthroscopy did not increase complication rates. Age < 20 years was associated with a lower risk of complications (odds ratio (OR) = 0.53, 95% CI 0.33 to 0.84, p = 0.008), whereas smoking history (OR = 1.72, 95% CI 1.01 to 2.87, p = 0.040) and correction of both acetabular dysplasia and retroversion (OR = 2.46, 95% CI 1.08 to 5.24, p = 0.024) were linked to an increased risk of complications. The operating time and incidence of complications decreased with increasing experience.ConclusionPAO is an effective procedure with an acceptable complication rate. Risk factors should be discussed during preoperative counselling. While patient optimization may help to reduce complications, precise intraoperative technique remains critical for minimizing risk.Cite this article: Bone Jt Open 2025;6(10):2022–2031.

Background: An extended trochanteric osteotomy (ETO) is commonly used to improve exposure and facilitate femoral component removal in revision total hip arthroplasty (THA). An additional medial corticotomy may be used in conjunction with an ETO to correct a femoral deformity, particularly varus remodeling in association with a failed femoral component. This study evaluated the outcomes of combining an ETO with a medial corticotomy in revision THA, with emphasis on implant fixation, femoral alignment, bone union, and clinical outcomes. Methods: Of the 612 ETOs performed between 2003 and 2013, 13 patients (9 men and 4 women) underwent ETO combined with a medial corticotomy to correct varus remodeling, representing 2% of all ETOs during that period. The average follow-up was 8 ± 3.5 years. The mean age at surgery was 67 ± 13.5 years. The mean body mass index was 32 ± 7.7 kg/m 2 . Radiographs were reviewed to measure preoperative and postoperative femoral deformity, time to consolidation, and femoral fixation. Clinical outcomes were evaluated using the Harris Hip Score (HHS). Results: All patients had preoperative varus femoral deformity (mean 16.3° ± 5.7°). The mean postoperative alignment was 4.3 ± 1.6° varus achieving an average correction of 12° (95% CI −15.7 to −8.3, p < 0.0001). Both the ETO and the medial corticotomy consolidated in 10 of 11 patients (91%) with available 1-year radiographs at a mean of 11 ± 7.7 months. The mean HHS improved significantly from 42 preoperatively to 82 at 5-year follow-up (p = 0.0002). Complications related to the ETO and corticotomy occurred in 4 patients (30%), including 1 intraoperative fracture, 1 postoperative greater trochanteric fracture, 1 nonunion of the medial corticotomy, and 1 postoperative wound-hematoma. All femoral components remained well fixed at final follow-up. Conclusion: The combination of ETO and medial corticotomy in revision THA effectively corrected femoral alignment in patients with a preoperative varus deformity and was associated with significant functional improvement at the final follow-up. Level of Evidence: Level IV . See Instructions for Authors for a complete description of levels of evidence.

Background: Gluteus medius tears and atrophy cause lateral hip pain, limp, and functional impairment. We developed the “broken wing sign,” a novel physical examination test for detecting gluteus medius tendon tears and muscle degeneration. This study evaluated its diagnostic accuracy and clinical utility against magnetic resonance imaging (MRI) and intraoperative findings. Methods: We prospectively examined 59 patients (75 hips; mean age, 69.5 ± 10.8 years; 48 women) with suspected hip abductor insufficiency. The broken wing sign was tested with patients prone, the knee flexed at 90°, and the hip actively extended. A positive sign involved ≥10° of compensatory external hip rotation, visible as an inward drift of the foot, indicating gluteus medius (and resulting internal rotation) weakness. MRI served as the reference standard for classifying gluteus medius integrity as no tear, partial, full-thickness, or massive. Fatty infiltration was graded (Goutallier grades 0 to 4), quantified, and analyzed for a correlation with the examination findings. Diagnostic accuracy metrics were calculated. Results: The broken wing sign demonstrated high accuracy (sensitivity, 81.8%; specificity, 80.0%; positive predictive value [PPV], 91.8%; negative predictive value, 61.5%; diagnostic odds ratio, 17.8). An external rotation threshold of ≥30° yielded 100% specificity and 100% PPV for a tear. The sign detected acute tears with no or minimal fatty infiltration (100% sensitivity for massive tears). The degree of the external rotation angle was strongly correlated with muscle atrophy with fatty infiltration, showing 88.0% sensitivity for Goutallier grade ≥3 and 100% sensitivity for grade 4. Conclusions: The broken wing sign is clinically useful for diagnosing gluteus medius tendon tears and muscle atrophy, particularly massive tears and advanced fatty degeneration. It effectively guides MRI utilization and surgical planning, serving as a valuable initial diagnostic tool in the clinical setting. Level of Evidence: Diagnostic Level II . See Instructions for Authors for a complete description of levels of evidence.

Background: Cholestatic liver diseases, including primary sclerosing cholangitis, are characterized by biliary fibroinflammation. TGFβ-activated cholangiocytes release signals that recruit immune cells and activate myofibroblasts, promoting inflammation and extracellular matrix (ECM) deposition. TGFβ also regulates stearoyl-CoA desaturase (SCD), an enzyme involved in lipid signaling. Yet, the role of SCD or its inhibitor, Aramchol, in biliary fibroinflammation had not been studied. Methods and Results: Mdr2 -/- with established biliary fibrosis and 3,5-diethoxycarboncyl-1,4-dihydrocollidine (DDC) diet-fed mice were treated with Aramchol meglumine (12.5 mg/kg/day). Hepatic fibrosis was assessed by qPCR, Picrosirius red staining, immunofluorescence, and hydroxyproline content. Human H69 or murine large cholangiocyte cell lines stimulated with TGFβ, as well as PSC-derived cholangiocytes (PSC-C), were treated with Aramchol or SCD siRNA. RNA-seq, fibroinflammatory marker expression, peroxisome proliferator-activated receptor (PPAR) activity, and targeted fatty acid profiling were performed. Aramchol treatment significantly reduced hepatic ECM gene expression, inflammatory cytokines ( Il6 , Tnfa ), collagen content, and myofibroblast activation (aSMA staining) in both mouse models. In TGFβ-stimulated H69 cells, Aramchol suppressed hepatic fibrosis pathways and enhanced PPAR signaling. Aramchol also reduced the expression of fibrotic markers, myofibroblast-activating mediators ( VEGFA and PDGFB ), and IL6, mirroring the effects of SCD knockdown. In PSC-C, Aramchol significantly downregulated SCD, VEGFA and IL6 . Conversely, PPARα and -γ activity and fatty acid agonist, linoleic acid levels were increased in cholangiocyte cell lines. Conclusions: Aramchol attenuates and prevents biliary fibrosis in mouse models of cholestatic liver disease by inhibiting TGFβ-induced fibroinflammatory mediators and activating PPARa/γ in cholangiocytes. These findings, combined with its favorable clinical safety profile, support the potential of Aramchol as a therapeutic candidate for PSC.

Importance Nontraumatic subarachnoid hemorrhage (SAH) represents the third most common stroke type with unique etiologies, risk factors, diagnostics, and treatments. Nevertheless, epidemiological studies often cluster SAH with other stroke types leaving its distinct burden estimates obscure. Objective To estimate the worldwide burden of SAH. Design, Setting, and Participants Based on the repeated cross-sectional Global Burden of Disease (GBD) 2021 study, the global burden of SAH in 1990 to 2021 was estimated. Moreover, the SAH burden was compared with other diseases, and its associations with 14 individual risk factors were investigated with available data in the GBD 2021 study. The GBD study included the burden estimates of nontraumatic SAH among all ages in 204 countries and territories between 1990 and 2021. Exposures SAH and 14 modifiable risk factors. Main Outcomes and Measures Absolute numbers and age-standardized rates with 95% uncertainty intervals (UIs) of SAH incidence, prevalence, mortality, and disability-adjusted life-years (DALYs) as well as risk factor–specific population attributable fractions (PAFs). Results In 2021, the global age-standardized SAH incidence was 8.3 (95% UI, 7.3-9.5), prevalence was 92.2 (95% UI, 84.1-100.6), mortality was 4.2 (95% UI, 3.7-4.8), and DALY rate was 125.2 (95% UI, 110.5-142.6) per 100 000 people. The highest burden estimates were found in Latin America, the Caribbean, Oceania, and high-income Asia Pacific. Although the absolute number of SAH cases increased, especially in regions with a low sociodemographic index, all age-standardized burden rates decreased between 1990 and 2021: the incidence by 28.8% (95% UI, 25.7%-31.6%), prevalence by 16.1% (95% UI, 14.8%-17.7%), mortality by 56.1% (95% UI, 40.7%-64.3%), and DALY rate by 54.6% (95% UI, 42.8%-61.9%). Of 300 diseases, SAH ranked as the 36th most common cause of death and 59th most common cause of DALY in the world. Of all worldwide SAH-related DALYs, 71.6% (95% UI, 63.8%-78.6%) were associated with the 14 modeled risk factors of which high systolic blood pressure (population attributable fraction [PAF] = 51.6%; 95% UI, 38.0%-62.6%) and smoking (PAF = 14.4%; 95% UI, 12.4%-16.5%) had the highest attribution. Conclusions and Relevance Although the global age-standardized burden rates of SAH more than halved over the last 3 decades, SAH remained one of the most common cardiovascular and neurological causes of death and disabilities in the world, with increasing absolute case numbers. These findings suggest evidence for the potential health benefits of proactive public health planning and resource allocation toward the prevention of SAH.