Acta Orthopaedica最新文献

{"title":"High revision rate of metal-backed glenoid component and impact on the overall revision rate of stemless total shoulder arthroplasty: a cohort study from the Danish Shoulder Arthroplasty Registry.","authors":"Marc R K Nyring, Bo S Olsen, Steen L Jensen, Jeppe V Rasmussen","doi":"10.2340/17453674.2024.41014","DOIUrl":"10.2340/17453674.2024.41014","url":null,"abstract":"<p><strong>Background and purpose: </strong>There is controversy regarding the results of stemmed and stemless total shoulder arthroplasty (TSA) used for osteoarthritis. Therefore, we aimed to compare revision rates of stemmed and stemless TSA and to examine the impact of metal-backed glenoid components.</p><p><strong>Methods: </strong>We included all patients reported to the Danish Shoulder Arthroplasty Register from January 1, 2012 to December 31, 2022 with an anatomical TSA used for osteoarthritis. Primary outcome was revision (removal or exchange of components) for any reason.</p><p><strong>Results: </strong>3,338 arthroplasties were included. The hazard ratio for revision of stemless TSA adjusted for age and sex was 1.83 (95% confidence interval [CI] 1.21-2.78) with stemmed TSA as reference. When excluding all arthroplasties with a metal-backed glenoid component, the adjusted hazard ratio for revision of stemless TSA was 1.37 (CI 0.85-2.20). For the Eclipse stemless TSA system, the adjusted hazard ratio for revision of a metal-backed glenoid component was 8.75 (CI 2.40-31.9) with stemless Eclipse with an all-polyethylene glenoid component as reference.</p><p><strong>Conclusion: </strong>We showed that the risk of revision of stemless TSAs was increased and that it was related to their combination with metal-backed glenoid components.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"386-391"},"PeriodicalIF":2.5,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11253195/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141625627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simultaneous correction of leg length discrepancy and angular deformity of the distal femur with retrograde Precice nails: a retrospective analysis of 45 patients.","authors":"Bjoern Vogt, Caja Biermann, Georg Gosheger, Andrea Laufer, Anna Rachbauer, Carina Antfang, Milena Lueckingsmeier, Gregor Toporowski, Henning Tretow, Robert Roedl, Adrien Frommer","doi":"10.2340/17453674.2024.40947","DOIUrl":"10.2340/17453674.2024.40947","url":null,"abstract":"<p><strong>Background and purpose: </strong>Magnetically controlled motorized intramedullary lengthening nails (ILNs) can be employed for simultaneous correction of angular deformities of the distal femur and leg length discrepancy. This spares typical complications of external fixators but requires precise preoperative planning and exact intraoperative execution. To date, its results are insufficiently reported. We aimed to elucidate the following questions: (i) Is acute angular deformity correction and gradual femoral lengthening via a retrograde ILN a reliable and precise treatment option? (ii) What are the most common complications of treatment?</p><p><strong>Methods: </strong>Acute angular deformity correction and subsequent gradual lengthening of the distal femur with retrograde ILN was retrospectively analyzed in 45 patients (median patient age: 15 years, interquartile range [IQR] 13-19 and median follow-up: 40 months, IQR 31-50). Outcome parameters were accuracy, precision, reliability, bone healing, and complications of treatment.</p><p><strong>Results: </strong>The median distraction was 46 mm (IQR 29-49), median distraction and consolidation index 0.9 mm/day (IQR 0.7-1.0) and 29 days/cm (IQR 24-43), respectively. The median preoperative mechanical axis deviation (MAD) was 30 mm (IQR 23-39) in the varus cohort and -25 mm (IQR -29 to -15) in the valgus cohort and reduced to a mean of 8 mm (standard deviation [SD] 8) and -3 (SD 10), respectively. Accuracy, precision, and reliability of lengthening were 94%, 95% and 96%, respectively. Accuracy and precision of deformity correction were 92% and 89%, respectively. In total, 40/45 of patients achieved distraction with a difference of less than 1 cm from the initial plan and a postoperative MAD ranging from -10 mm to +15 mm. In 13/45 patients unplanned additional surgeries were conducted to achieve treatment goal with nonunion being the most frequent (4/45) and knee subluxation (3/45) the most severe complication.</p><p><strong>Conclusion: </strong>Acute deformity correction and subsequent lengthening of the distal femur with retrograde ILN is a reliable and accurate treatment achieving treatment goal in 89% but unplanned additional surgeries in 29% of patients should be anticipated.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"364-372"},"PeriodicalIF":2.5,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11248716/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141615627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Responders to first-line osteoarthritis treatment had reduced frequency of hip and knee joint replacements within 5 years: an observational register-based study of 44,311 patients.","authors":"Kristin Gustafsson, Anna Cronström, Ola Rolfson, Eva Ageberg, Therese Jönsson","doi":"10.2340/17453674.2024.41011","DOIUrl":"10.2340/17453674.2024.41011","url":null,"abstract":"<p><strong>Background and purpose: </strong>First-line treatment (education, exercise) for patients with hip and knee osteoarthritis (OA) aims to reduce pain and improve function. We aimed to compare progression to joint replacement within 5 years between responders and non-responders to first-line treatment for hip and knee OA, respectively.</p><p><strong>Methods: </strong>This observational study included data for 30,524 knee OA and 13,787 hip OA patients from the Swedish Osteoarthritis Register, linked with the Swedish Arthroplasty Register, Statistics Sweden, and the Swedish Prescribed Drug Register. The primary prognostic factor was change in pain between baseline and 3-month follow-up, measured on a numeric rating scale (0-10, best to worst) where an improvement of ≥ 2 was classified as responder and ≤ 1 as non-responder. The main outcome was progression to joint replacement surgery within 5 years, assessed using baseline adjusted multivariable Cox regression analyses.</p><p><strong>Results: </strong>At 5 years, in hip OA, 35% (95% confidence interval [CI] 32.2-37.2) of the responders and 48% (CI 45.9-49.5) of the non-responders and in knee OA 14% (CI 13.0-15.3) of the responders and 20% (CI 18.8-20.8) of the non-responders had progressed to joint replacement. Being a responder to the treatment was associated with having a lower probability of progression to surgery for both hip OA (hazard ratio [HR] 0.4, CI 0.4-0.5) and knee OA (HR 0.6, CI 0.5-0.6).</p><p><strong>Conclusion: </strong>Patients with hip or knee OA who experienced pain relief after a first-line OA treatment program were less likely to progress to joint replacement surgery.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"373-379"},"PeriodicalIF":2.5,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11249020/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141615610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mortality after major lower extremity amputation and association with index level: a cohort study based on 11,205 first-time amputations from nationwide Danish databases.","authors":"Anna Trier Heiberg Brix, Katrine Hass Rubin, Tine Nymark, Hagen Schmal, Martin Lindberg-Larsen","doi":"10.2340/17453674.2024.40996","DOIUrl":"10.2340/17453674.2024.40996","url":null,"abstract":"<p><strong>Background and purpose: </strong>Mortality after major lower extremity amputations is high and may depend on amputation level. We aimed to examine the mortality risk in the first year after major lower extremity amputation divided into transtibial and transfemoral amputations.</p><p><strong>Methods: </strong>This observational cohort study used data from the Danish Nationwide Health registers. 11,205 first-time major lower extremity amputations were included from January 1, 2010, to December 31, 2021, comprising 3,921 transtibial amputations and 7,284 transfemoral amputations.</p><p><strong>Results: </strong>The 30-day mortality after transtibial amputation was overall 11%, 95% confidence interval (CI) 10-12 (440/3,921) during the study period, but declined from 10%, CI 7-13 (37/381) in 2010 to 7%, CI 4-11 (15/220) in 2021. The 1-year mortality was 29% overall, CI 28-30 (1,140 /3,921), with a decline from 31%, CI 21-36 (117/381) to 20%, CI 15-26 (45/220) during the study period. For initial transfemoral amputation, the 30-day mortality was overall 23%, CI 22-23 (1,673/7,284) and declined from 27%, CI 23-31 (138/509) to 22%, CI 19-25 (148/683) during the study period. The 1-year mortality was 48% overall, CI 46-49 (3,466/7,284) and declined from 55%, CI 50-59 (279/509) to 46%, CI 42-50 (315/638).</p><p><strong>Conclusion: </strong>The mortality after major lower extremity amputation declined in the 12-year study period; however, the 1-year mortality remained high after both transtibial and transfemoral amputations (20% and 46% in 2021). Hence, major lower extremity amputation patients constitute one of the most fragile orthopedic patient groups, emphasizing an increased need for attention in the pre-, peri-, and postoperative setting.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"358-363"},"PeriodicalIF":2.5,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11186348/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and validation of an artificial intelligence model for the classification of hip fractures using the AO-OTA framework.","authors":"Ehsan Akbarian, Mehrgan Mohammadi, Emilia Tiala, Oscar Ljungberg, Ali Sharif Razavian, Martin Magnéli, Max Gordon","doi":"10.2340/17453674.2024.40949","DOIUrl":"10.2340/17453674.2024.40949","url":null,"abstract":"<p><strong>Background and purpose: </strong>Artificial intelligence (AI) has the potential to aid in the accurate diagnosis of hip fractures and reduce the workload of clinicians. We primarily aimed to develop and validate a convolutional neural network (CNN) for the automated classification of hip fractures based on the 2018 AO-OTA classification system. The secondary aim was to incorporate the model's assessment of additional radiographic findings that often accompany such injuries.</p><p><strong>Methods: </strong>6,361 plain radiographs of the hip taken between 2002 and 2016 at Danderyd University Hospital were used to train the CNN. A separate set of 343 radiographs representing 324 unique patients was used to test the performance of the network. Performance was evaluated using area under the curve (AUC), sensitivity, specificity, and Youden's index.</p><p><strong>Results: </strong>The CNN demonstrated high performance in identifying and classifying hip fracture, with AUCs ranging from 0.76 to 0.99 for different fracture categories. The AUC for hip fractures ranged from 0.86 to 0.99, for distal femur fractures from 0.76 to 0.99, and for pelvic fractures from 0.91 to 0.94. For 29 of 39 fracture categories, the AUC was ≥ 0.95.</p><p><strong>Conclusion: </strong>We found that AI has the potential for accurate and automated classification of hip fractures based on the AO-OTA classification system. Further training and modification of the CNN may enable its use in clinical settings.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"340-347"},"PeriodicalIF":2.5,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11184710/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Demographics and risk for containment surgery in patients with unilateral Legg-Calvé-Perthes disease: a national population-based cohort study of 309 patients from the Swedish Pediatric Orthopedic Quality Register.","authors":"Miriam G Wadström, Nils P Hailer, Yasmin D Hailer","doi":"10.2340/17453674.2024.40907","DOIUrl":"10.2340/17453674.2024.40907","url":null,"abstract":"<p><strong>Background and purpose: </strong>It is controversial as to which patients affected by Legg-Calvé-Perthes disease (LCPD) benefit from containment surgery. This population-based study based on data from a national quality registry aims to assess the incidence of LCPD and to explore which factors affect the decision for surgical intervention.</p><p><strong>Methods: </strong>This observational study involved 309 patients with unilateral LCPD reported between 2015 and 2023 to the Swedish Pediatric Orthopedic Quality Register (SPOQ). Descriptive statistics and logistic regression models were used for analysis.</p><p><strong>Results: </strong>In 2019, the assessed incidence of LCPD in the Swedish population of 2-12-year-olds was 4.2 per 105. 238 (77%) were boys with a mean age of 6 years. At diagnosis, 55 (30%) were overweight or obese, rising to 17 patients (39%) and 16 patients (40%) at 2-year follow-up for surgically and non-surgically treated groups, respectively. At diagnosis, affected hips had reduced abduction compared with healthy hips, and their abduction remained restricted at the 2-year follow-up. Surgically treated patients had inferior abduction compared with non-surgically treated ones at diagnosis. The adjusted risk for containment surgery increased with age and in the presence of a positive Trendelenburg sign but decreased with greater hip abduction.</p><p><strong>Conclusion: </strong>We found a lower national yearly incidence (4.2 per 105) than previously reported in Swedish studies. A higher proportion of overweight or obese patients compared with the general Swedish population of 4-9-year-olds was identified. Increasing age, positive Trendelenburg sign, and limited hip abduction at diagnosis correlated with increased surgical intervention likelihood.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"333-339"},"PeriodicalIF":2.5,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11184411/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Total joint arthroplasty for thumb carpometacarpal joint osteoarthritis: a systematic review and meta-analysis of randomized controlled trials.","authors":"Rasmus Liukkonen, Venla-Linnea Karjalainen, Reetta Kvist, Matias Vaajala, Ville Ponkilainen, Teemu Karjalainen","doi":"10.2340/17453674.2024.40816","DOIUrl":"10.2340/17453674.2024.40816","url":null,"abstract":"<p><strong>Background and purpose: </strong>Thumb carpometacarpal (CMC) joint osteoarthritis (OA) is increasingly treated with total joint arthroplasty (TJA). We aimed to perform a systematic review and meta-analysis of the benefits and harms of the TJA for thumb CMC OA compared with other treatment strategies.</p><p><strong>Patients and methods: </strong>We performed a systematic search on MEDLINE and CENTRAL databases on August 2, 2023. We included randomized controlled trials investigating the effect of TJA in people with thumb CMC joint OA regardless of the stage or etiology of the disease or comparator. The outcomes were pooled with a random effect meta-analysis.</p><p><strong>Results: </strong>We identified 4 studies randomizing 420 participants to TJA or trapeziectomy. At 3 months, TJA's benefits for pain may exceed the clinically important difference. However, after 1-year follow-up TJA does not improve pain compared with trapeziectomy (mean difference 0.53 points on a 0 to 10 scale; 95% confidence interval [CI] 0.26-0.81). Furthermore, it provides a transient benefit in hand function at 3 months (measured with Disabilities of Arm, Shoulder, and Hand questionnaire, scale 0-100, lower is better) compared with trapeziectomy with or without ligament reconstruction tendon interposition. The benefit in function diminished to a clinically unimportant level at 1-year follow-up (4.4 points better; CI 0.42-8.4).</p><p><strong>Conclusion: </strong>Transient benefit in hand function for TJA implies that it could be a preferable option over trapeziectomy for people who consider fast postoperative recovery important. However, current evidence fails to inform us if TJA carries long-term higher risks of revisions compared with trapeziectomy.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"325-332"},"PeriodicalIF":2.5,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11184324/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"International trends in shoulder replacement: a meta-analysis from 11 public joint registers.","authors":"Neal Rupani, Christophe Combescure, Alan Silman, Anne Lübbeke, Jonathan Rees","doi":"10.2340/17453674.2024.40948","DOIUrl":"10.2340/17453674.2024.40948","url":null,"abstract":"<p><strong>Background and purpose: </strong>International variation exists in the types of shoulder replacement used for treatment of specific diseases. Implant choice continues to evolve without high-quality evidence. Our aim was to evaluate trends in incidence rates of shoulder replacement and assess any recent changes in practice between countries by using registry data.</p><p><strong>Methods: </strong>Patient characteristics, indication and year of surgery, type of replacement, and collection methods of patient-reported outcomes (PROMs) was extracted from 11 public joint registries. Meta-analyses examined use of reverse total shoulder replacement (RTSR) for osteoarthritis, cuff tear arthropathy, and acute fracture; use of anatomical total shoulder replacement (TSR) for osteoarthritis; and use of humeral hemiarthroplasty for fracture.</p><p><strong>Results: </strong>The annual growth rate of shoulder replacements performed is 6-15% (2011-2019). The use of RTSR has almost doubled (93%). RTSR is now universally performed for cuff tear arthropathy (97.3%, 95% confidence interval [CI] 96.0-98.1). Its use for avascular necrosis, trauma, and inflammatory arthropathy is increasing. The use of RTSR was similar (43.1%, CI 30.0-57.2) versus TSR (44.7%, CI 31.1-59.1) for osteoarthritis. The types of PROMs used, collection time points, and response rates lack standardization. COVID-19 had a varying inter-registry impact on incidence rates.</p><p><strong>Conclusion: </strong>The incidence of shoulder replacements has grown. Use of RTSR has increased for all disease indications despite limited high-quality evidence driving this change in indications outside of cuff arthropathy. Consequently, less variation is observed in international practice. Existing differences now relate to use of newer implant types and methodology of PROMs collection, which prevents international comparison and outcome analysis.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"348-357"},"PeriodicalIF":2.5,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11184711/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The effectiveness of a protocol without routine radiographs for follow-up of adolescent idiopathic scoliosis patients (CURVE): a study protocol.","authors":"Jurre T F Baetsen, Miranda L Hooff, Pepijn Bisseling, Johanna M Van Dongen, Dineke G Van de Fliert, Eric Hoebink, Diederik H R Kempen, Joost P H J Rutges, Tom P C Schlösser, Hanneke M Van West, Philip J Van der Wees, Paul C Willems, Marinus De Kleuver","doi":"10.2340/17453674.2024.40904","DOIUrl":"10.2340/17453674.2024.40904","url":null,"abstract":"<p><strong>Background and purpose: </strong>Current follow-up protocols for adolescent idiopathic scoliosis (AIS) are based on consensus and consist of regular full-spine radiographs to monitor curve progression and surgical complications. Consensus exists to avoid inappropriate use of radiographs in children. It is unknown whether a standard radiologic follow-up (S-FU) approach is necessary or if a patient-empowered follow-up (PE-FU) approach can reduce the number of radiographs without treatment consequences.</p><p><strong>Methods and analyses: </strong>A nationwide multicenter pragmatic randomized preference trial was designed for 3 follow-up subgroups (pre-treatment, post-brace, post-surgery) to compare PE-FU and S-FU. 812 patients with AIS (age 10-18 years) will be included in the randomized trial or preference cohorts. Primary outcome is the proportion of radiographs with a treatment consequence for each subgroup. Secondary outcomes consist of the proportion of patients with delayed initiation of treatment due to non-routine radiographic follow-up, radiation exposure, societal costs, positive predictive value, and interrelation of clinical assessment, quality of life, and parameters for initiation of treatment during follow-up. Outcomes will be analyzed using linear mixed-effects models, adjusted for relevant baseline covariates, and are based on intention-to-treat principle. Study summary: (i) a national, multicenter pragmatic randomized trial addressing the optimal frequency of radiographic follow-up in patients with AIS; (ii) first study that includes patient-empowered follow-up; (iii) an inclusive study with 3 follow-up subgroups and few exclusion criteria representative for clinical reality; (iv) preference cohorts alongside to amplify generalizability; (v) first study conducting an economic evaluation comparing both follow-up approaches.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"298-306"},"PeriodicalIF":2.5,"publicationDate":"2024-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11181918/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141417148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial intelligence can be used in the identification and classification of shoulder osteoarthritis and avascular necrosis on plain radiographs: a training study of 7,139 radiograph sets.","authors":"Martin Magnéli, Michael Axenhus, Johan Fagrell, Petter Ling, Jacob Gislén, Yilmaz Demir, Erica Domeij-Arverud, Kristofer Hallberg, Björn Salomonsson, Max Gordon","doi":"10.2340/17453674.2024.40905","DOIUrl":"10.2340/17453674.2024.40905","url":null,"abstract":"<p><strong>Background and purpose: </strong>Knowledge concerning the use AI models for the classification of glenohumeral osteoarthritis (GHOA) and avascular necrosis (AVN) of the humeral head is lacking. We aimed to analyze how a deep learning (DL) model trained to identify and grade GHOA on plain radiographs performs. Our secondary aim was to train a DL model to identify and grade AVN on plain radiographs.</p><p><strong>Patients and methods: </strong>A modified ResNet-type network was trained on a dataset of radiographic shoulder examinations from a large tertiary hospital. A total of 7,139 radiographs were included. The dataset included various projections of the shoulder, and the network was trained using stochastic gradient descent. Performance evaluation metrics, area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to assess the network's performance for each outcome.</p><p><strong>Results: </strong>The network demonstrated AUC values ranging from 0.73 to 0.93 for GHOA classification and > 0.90 for all AVN classification classes. The network exhibited lower AUC for mild cases compared with definitive cases of GHOA. When none and mild grades were combined, the AUC increased, suggesting difficulties in distinguishing between these 2 grades.</p><p><strong>Conclusion: </strong>We found that a DL model can be trained to identify and grade GHOA on plain radiographs. Furthermore, we show that a DL model can identify and grade AVN on plain radiographs. The network performed well, particularly for definitive cases of GHOA and any level of AVN. However, challenges remain in distinguishing between none and mild GHOA grades.</p>","PeriodicalId":6916,"journal":{"name":"Acta Orthopaedica","volume":"95 ","pages":"319-324"},"PeriodicalIF":2.5,"publicationDate":"2024-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11182033/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141330083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}