Medial femoral condyle OCD (osteochondritis dissecans): correlation between imaging and arthroscopy.

Osteochondritis dissecans (OCD) describes a pathologic condition centered at the osteochondral junction that may result in an unstable subchondral fragment (progeny), disruption of the overlying cartilage, which may separate from the underlying parent bone. It is one of the causes of chronic knee pain in children and young adults. The current literature on OCD lesions focuses primarily on the medial femoral condyle (MFC), but inconsistent use of terminology, particularly in the distinction of OCD lesions between skeletally immature and mature patients has created uncertainty regarding imaging workup, treatment, and long-term prognosis. This article reviews the pathophysiology of MFC OCD lesions, highlighting the role of endochondral ossification at the secondary growth plate of the immature femoral condyles, the rationale behind the imaging work-up, and key imaging findings that can distinguish between stable lesions, unstable lesions, and physiologic variants. This overview also provides a case-based review to introduce imaging correlates with the ROCK (Research in Osteochondritis of the Knee) arthroscopic classification.

Imaging of Disease and Normal Variant Patterns in Pediatric Hips.

The pediatric hip undergoes significant changes from infancy through adolescence. Proper maturation is crucial for the development of a stable and functional hip joint. Imaging interpretation of the pediatric hip requires distinguishing normal variants and maturation patterns from pathology. We review femoral ossification centers, variants, and conditions that affect the proximal femur, such as Legg-Calvé-Perthes disease; the acetabulum, such as developmental hip dysplasia; the acetabular labrum, such as femoroacetabular impingement; and synovial pathology in children through adolescence. Understanding the spectrum of hip conditions and using advanced imaging techniques are essential for the accurate diagnosis and effective management of pediatric hip disorders.

ACR Appropriateness Criteria® Chronic Hand and Wrist Pain: 2023 Update.

Chronic hand and wrist pain is a common presenting complaint. The intricate anatomy results in a variety of pain generators-multiple bones, articular cartilage, intrinsic ligaments, triangular fibrocartilage complex, joint capsules and synovium, tendons and tendon sheaths, muscles, and nerves-in a compact space. The need for imaging and the choice of the appropriate imaging modality are best determined by the patient's presentation, physical examination, and the clinician's working differential diagnosis. Radiography is usually appropriate as the initial imaging study in the evaluation of chronic hand or wrist pain. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

Anatomical Proximity Between Sciatic Nerve and Ischial Spine and its Relationship to the Development of Deep Gluteal Pain Syndrome.

OBJECTIVE: Deep gluteal syndrome (DGS) is a medical diagnosis in which the pathoanatomy of the subgluteal space contributes to pain. The growing recognition that gluteal neuropathies can be associated with the presence of a bone-neural conflict with irritation or compression may allow us to shed some light on this pathology. This study aims to determine whether the location of the sciatic nerve (SN) in relation to the ischial spine (IS) contributes to the development of DGS. METHODS: The SN - IS relationship was analyzed based on magnetic resonance imaging (MRI) in 15 surgical patients (SPs), who underwent piriformis release, and in 30 control patients who underwent MRI of the pelvis for reasons unrelated to sciatica. The SN exit from the greater sciatic foramen was classified as either zone A (medial to the IS); zone B (on the IS); or zone C (lateral to the IS). RESULTS: The SN was significantly closer to the IS in SPs than in MRI controls (P = 0.014). When analyzing patients of similar age, SNs in SPs were significantly closer (P = 0.0061) to the IS, and located in zone B significantly more (P = 0.0216) as compared to MRI controls. Patients who underwent surgery for piriformis release showed a significant decrease in pain postoperatively (P < 0.0001). CONCLUSIONS: The results from this study suggest that the relationship between the IS and SN may play a role in the development of DGS. This may also help establish which patients would benefit more from surgical intervention.

Chronic Nonspinal Osteomyelitis in Adults: Consensus Recommendations on Percutaneous Bone Biopsies from the Society of Academic Bone Radiologists.

The diagnosis and management of chronic nonspinal osteomyelitis can be challenging, and guidelines regarding the appropriateness of performing percutaneous image-guided biopsies to acquire bone samples for microbiological analysis remain limited. An expert panel convened by the Society of Academic Bone Radiologists developed and endorsed consensus statements on the various indications for percutaneous image-guided biopsies to standardize care and eliminate inconsistencies across institutions. The issued statements pertain to several commonly encountered clinical presentations of chronic osteomyelitis and were supported by a literature review. For most patients, MRI can help guide management and effectively rule out osteomyelitis when performed soon after presentation. Additionally, in the appropriate clinical setting, open wounds such as sinus tracts and ulcers, as well as joint fluid aspirates, can be used for microbiological culture to determine the causative microorganism. If MRI findings are positive, surgery is not needed, and alternative sites for microbiological culture are not available, then percutaneous image-guided biopsies can be performed. The expert panel recommends that antibiotics be avoided or discontinued for an optimal period of 2 weeks prior to a biopsy whenever possible. Patients with extensive necrotic decubitus ulcers or other surgical emergencies should not undergo percutaneous image-guided biopsies but rather should be admitted for surgical debridement and intraoperative cultures. Multidisciplinary discussion and approach are crucial to ensure optimal diagnosis and care of patients diagnosed with chronic osteomyelitis.

Comparison of MRI and arthroscopy findings for transitional zone cartilage damage in the acetabulum of the hip joint.

OBJECTIVE: To assess the performance of morphologic and hypointense signal changes on MRI to predict grades and types of acetabular cartilage damage in the chondrolabral transitional zone (TZ) of the hip identified at arthroscopy. MATERIALS AND METHODS: This retrospective single-center study reviewed conventional 3T MRI hip studies from individuals with symptomatic femoroacetabular impingement (FAI) and subsequent hip arthroscopy surgery within 6 months. Independent review was made by three radiologists for the presence of morphologic damage or a hypointense signal lesion in the TZ on MRI. Fleiss' kappa statistic was used to assess inter-reader agreement. The degree of TZ surfacing damage (modified Outerbridge grades 1-4) and presence of non-surfacing wave sign at arthroscopic surgery were collected. Relationship between sensitivity and lesion grade was examined. RESULTS: One hundred thirty-six MRI hip studies from 40 males and 74 females were included (mean age 28.5 years, age range 13-54 years). MRI morphologic lesions had a sensitivity of 64.9-71.6% and specificity of 48.4-67.7% for arthroscopic surfacing lesions, with greater sensitivity seen for higher grade lesions. Low sensitivity was seen for wave sign lesions (34.5-51.7%). MRI hypointense signal lesions had a sensitivity of 26.3-62% and specificity of 43.8-78.0% for any lesion. Inter-reader agreement was moderate for morphologic lesions (k = 0.601) and poor for hypointense signal lesions (k = 0.097). CONCLUSION: Morphologic cartilage damage in the TZ on MRI had moderate sensitivity for any cartilage lesion, better sensitivity for higher grade lesions, and poor sensitivity for wave sign lesions. The diagnostic value of hypointense signal lesions was uncertain.

ACR Appropriateness Criteria® Imaging After Total Hip Arthroplasty.

This article reviews evidence for performing various imaging studies in patients with total hip prostheses. Routine follow-up is generally performed with radiography. Radiographs are also usually the initial imaging modality for patients with symptoms related to the prosthesis. Following acute injury with pain, noncontrast CT may add information to radiographic examination regarding the presence and location of a fracture, component stability, and bone stock. Image-guided joint aspiration, noncontrast MRI, and white blood cell scan and sulfur colloid scan of the hip, are usually appropriate studies for patients suspected of having periprosthetic infection. For evaluation of component loosening, wear, and/or osteolysis, noncontrast CT or MRI are usually appropriate studies. Noncontrast MRI is usually appropriate for identifying adverse reaction to metal debris related to metal-on-metal articulations. For assessing patients after hip arthroplasty, who have trochanteric pain and nondiagnostic radiographs, ultrasound, or MRI are usually appropriate studies. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

ACR Appropriateness Criteria® Imaging After Total Knee Arthroplasty: 2023 Update.

Total knee arthroplasty is the most commonly performed joint replacement procedure in the United States. This manuscript will discuss the recommended imaging modalities for six clinical variants; 1. follow-up of symptomatic or asymptomatic patients with a total knee arthroplasty. Initial imaging, 2. Suspected infection after total knee arthroplasty. Additional imaging following radiographs, 3. Pain after total knee arthroplasty. Infection excluded. Suspect aseptic loosening or osteolysis or instability. Additional imaging following radiographs, 4. Pain after total knee arthroplasty. Suspect periprosthetic or hardware fracture. Additional imaging following radiographs, 5. Pain after total knee arthroplasty. Measuring component rotation. Additional imaging following radiographs, and 6. Pain after total knee arthroplasty. Suspect periprosthetic soft-tissue abnormality unrelated to infection, including quadriceps or patellar tendinopathy. Additional imaging following radiographs. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

ACR Appropriateness Criteria® Chronic Extremity Joint Pain-Suspected Inflammatory Arthritis, Crystalline Arthritis, or Erosive Osteoarthritis: 2022 Update.

Evaluation for suspected inflammatory arthritis as a cause for chronic extremity joint pain often relies on imaging. It is essential that imaging results are interpreted in the context of clinical and serologic results to add specificity because there is significant overlap of imaging findings among the various types of arthritis. This document provides recommendations for imaging evaluation of specific types of inflammatory arthritis, including rheumatoid arthritis, seronegative spondyloarthropathy, gout, calcium pyrophosphate dihydrate disease (or pseudogout), and erosive osteoarthritis. The ACR Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances in which peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

ACR Appropriateness Criteria® Chronic Hip Pain: 2022 Update.

Chronic hip pain is a frequent chief complaint for adult patients who present for evaluation in a variety of clinical practice settings. Following a targeted history and physical examination, imaging plays a vital role in elucidating the etiologies of a patient's symptoms, as a wide spectrum of pathological entities may cause chronic hip pain. Radiography is usually the appropriate initial imaging test following a clinical examination. Depending on the clinical picture, advanced cross-sectional imaging may be subsequently performed for further evaluation. This documents provides best practice for the imaging workup of chronic hip pain in patients presenting with a variety of clinical scenarios. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

ACR Appropriateness Criteria® Chronic Shoulder Pain: 2022 Update.

Chronic shoulder pain is an extremely common presenting complaint. Potential pain generators include the rotator cuff tendons, biceps tendon, labrum, glenohumeral articular cartilage, acromioclavicular joint, bones, suprascapular and axillary nerves, and the joint capsule/synovium. Radiographs are typically the initial imaging study obtained in patients with chronic shoulder pain. Further imaging may often be required, with modality chosen based on patient symptoms and physical examination findings, which may lead the clinician to suspect a specific pain generator. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

Ultrasound-guided nerve block prior to biopsy of suspected neurogenic tumors: safety and feasibility in a pilot study.

OBJECTIVE: The aim of this study is to investigate the safety and feasibility of ultrasound-guided nerve block prior to biopsy of potentially neurogenic tumors. MATERIALS AND METHODS: A retrospective review of the medical record from June 2017 to June 2022 identified ultrasound-guided biopsies of potentially neurogenic tumors that were performed with a pre-procedural nerve block. Patient demographics, biopsy site, number of passes, needle gauge, use of sedation, pathology results, and procedural complications were recorded and summarized. RESULTS: The structured search found 16 patients that underwent biopsies of 18 potentially neurogenic tumors with the use of a pre-procedural nerve block at a variety of upper and lower extremity locations. Average patient age was 52 (range 18-78) and 9 patients (56%) were female. Of the 16 patients, 10 were performed without intravenous sedation. Three patients were unable to tolerate biopsy until a nerve block was used. All biopsies yielded a diagnostic sample with 13 of the tumors neurogenic in origin. One patient reported mild postprocedural pain which resolved with conservative treatment; no other complications were reported. CONCLUSION: Nerve block prior to ultrasound-guided biopsy of potentially neurogenic tumors is a safe and feasible technique. Further study is needed to determine the extent to which nerve block can decrease intra-procedural pain and reduce or eliminate the need for sedation during biopsy.

The posterior crescent sign on MRI and MR arthrography: is it a marker of hip dysplasia and instability?

OBJECTIVE: To evaluate the prevalence of the 'posterior crescent sign' in symptomatic patients referred for MRI/MR arthrogram of the hip and identify any correlation with imaging features of joint pathology. MATERIALS AND METHODS: Retrospective imaging assessment of a cohort of 1462 hips, from 1380 included MR examinations (82 bilateral) retrieved from a search of all examinations in patients 16-50 years old from June 2018 to June 2021, with median age 45.8 years (range 17.8-50.0) and 936 hips (64%) in women. Radiographic and MR findings related to hip dysplasia, femoroacetabular impingement and osteoarthritis were assessed. RESULTS: Fifty-one hips (3.5%) were positive for the posterior crescent sign, median age of 45.8 years (range 17.8-50.0) and 29 (58%) in women. Radiographic findings included the following: mean lateral centre edge angle (LCEA) 22.2° (± 7.8°) with LCEA < 20° in 15 (31%) and LCEA 20-25° in 17 (35%) and mean acetabular index (AI) of 13.1° (± 5.8°) with AI > 13° in 22 (45%). MR findings included the following: mean anterior acetabular sector angle (AASA) 54.3° (± 9.8°), mean posterior acetabular sector angle (PASA) 92.7° (± 7.0°), labral tear at 3-4 o'clock in 20 (39%), high-grade acetabular chondral loss in 42 (83%) and ligamentum teres abnormality in 20 (39%). CONCLUSION: The posterior crescent sign occurs in 3.5% of symptomatic young and middle-aged adults on MR. It is associated with overt and borderline hip dysplasia and other findings of hip instability. It is also associated with osteoarthritis in some cases and should be interpreted with caution in these patients.

ACR Appropriateness Criteria® Osteoporosis and Bone Mineral Density: 2022 Update.

Osteoporosis constitutes a significant public health risk. An estimated 10.2 million adults in the United States >50 years of age have osteoporosis, a systemic condition that weakens the bones increasing the susceptibility for fractures. Approximately one-half of women and nearly one-third of men >50 years of age will sustain an osteoporotic fracture. These fractures are associated with a decrease in quality of life, diminished physical function, and reduced independence. Dual-energy X-ray absorptiometry (DXA) is the primary imaging modality used to screen for osteoporosis in women >65 years of age and men >70 years of age. DXA may be used in patients <65 years of age to evaluate bone mass density if there are additional risk factors. In certain situations, vertebral fracture assessment and trabecular bone score may further predict fracture risk, particularly in patients who are not yet osteoporotic but are in the range of osteopenia. Quantitative CT is useful in patients with advanced degenerative changes in the spine. Given the proven efficacy of pharmacologic therapy, the role of imaging to appropriately identify and monitor high-risk individuals is critical in substantially reducing osteoporosis-associated morbidity and mortality, and reducing the considerable cost to the health care system. The ACR Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.

Contrast-enhanced ultrasound-guided musculoskeletal biopsies: our experience and technique.

OBJECTIVE: To present our experience with contrast-enhanced ultrasound (CEUS)-guided musculoskeletal soft tissue biopsies in a busy interventional clinic. MATERIALS AND METHODS: After IRB approval was obtained and informed consent was waived, we retrospectively reviewed all CEUS-guided musculoskeletal biopsies performed from December 1, 2018 to March 2, 2020. Relevant pre-procedure imaging was reviewed. Number of samples, suspected necrosis on pre-procedure imaging, specimen adequacy for pathologic analysis, correlation with pathologic diagnosis of surgical resection specimens, and procedural complications were recorded. RESULTS: Thirty-six CEUS-guided musculoskeletal biopsies were performed in 32 patients (mean age 57, range 26-88; 22 males, 10 females). All procedures were performed using 16-gauge biopsy needles, and all procedures provided adequate samples for pathologic analysis as per the final pathology report. Between two and seven core specimens were obtained (mean 3.7). In 30/36 cases (83%), a contrast-enhanced MRI was obtained prior to biopsy, and 10/30 (33%) of these cases showed imaging features suspicious for necrosis. In 15/36 cases, surgical resection was performed, and the core biopsy and surgical resection specimens were concordant in 14/15 cases (93%). One patient noted transient leg discomfort at the time of microbubble bursting. Otherwise, no adverse reactions or procedural complications were observed. CONCLUSION: CEUS is an accurate way to safely target representative areas of soft tissue lesions for biopsy and can be implemented in a busy interventional clinic. Our early experience has shown this to be a promising technique, especially in targeting representative areas of heterogeneous lesions and lesions with areas of suspected necrosis on prior imaging.

Microinstability of the hip: a systematic review of the imaging findings.

OBJECTIVES: To undertake a systematic review of the morphologic features associated with hip microinstability and determine whether there are suggestive or diagnostic imaging findings. METHODS: Four electronic databases were searched up to September 2019 to identify original research reporting morphologic features in individuals with either a clinical diagnosis of hip microinstability (instability without overt subluxation/dislocation) or those with symptomatic laxity demonstrated on imaging (increased femoral head translation/distraction or capsular volume). Studies focussing on individuals with pre-existing hip conditions (including definite dysplasia (lateral centre edge angle < 20°), significant trauma, previous dislocation or surgery were excluded. Methodological quality was assessed by the Quality Assessment of Diagnostic Accuracy Studies 2 tool. RESULTS: Twenty-two studies met inclusion criteria (clinical diagnosis of microinstability n = 15 and demonstration of laxity n = 7). Imaging information gathered from the studies includes radiographs (n = 14), MRI (n = 6), MR arthrography (n = 4), CT (n = 1) and intraoperative examination. Most studies exhibited design features associated with an overall high or unclear risk of bias. Some dysplastic features are associated with microinstability or laxity reference measures; however, microinstability is frequently diagnosed in those with a lateral centre edge angle > 25°. Other associated imaging findings reported include impingement morphology, anterior labral tearing, femoral head chondral injury, ligamentum teres tears and capsular attenuation. CONCLUSIONS: The current literature does not provide strong evidence for imaging features diagnostic of microinstability. In the appropriate clinical context, dysplastic morphology, anterior labral tears and ligamentum teres tears may be suggestive of this condition although further research is needed to confirm this. PROSPERO REGISTRATION: CRD42019122406.

Overuse Injuries of the Shoulder.

This article discusses the most common and important overuse injuries of the shoulder with attention to MR imaging and ultrasound findings. Pathologic conditions occurring in athletes and nonathletes are included, with review of relevant anatomy, predisposing factors, and treatment considerations. Specific overuse injuries involving the rotator cuff, long head of the biceps tendon, and subacromial-subdeltoid bursa are reviewed. Impingement syndromes of the shoulder, Little Leaguer's shoulder, and stress-induced distal clavicular osteolysis are also discussed.

Incidence and implications of fracture in core muscle injury.

OBJECTIVE: To determine the pubic bone fracture incidence and associated injury patterns in patients with core muscle injury. MATERIALS AND METHODS: Ninety-three consecutive patients with core muscle injury protocol MRI showing rectus abdominis-adductor longus aponeurotic plate injuries from June 2007 through August 2017 were independently analyzed in blinded fashion by two musculoskeletal radiologists for the presence or absence of pubic bone fracture. A variety of other osseous and soft tissue injury characteristics were recorded. Pain duration prior to MRI and return to play time were taken from the clinical record. Statistical analysis included fracture incidence as well as the association of fracture with other injury characteristics, duration of pain, and return to play time. RESULTS: Eighty-seven men and six women with a mean age of 34.4 years (range, 16-66 years) were included in the study cohort. Overall fracture incidence was 18.3% (17/93) including 13 fatigue fractures of the pubic body and four elevated cortical fractures/fragments. After correction for multiple comparisons, no strong association was identified with osseous or soft tissue injury characteristics, pain duration, or return to play time. CONCLUSIONS: Pubic fractures-particularly fatigue fractures-are a common co-existing injury in patients with a wide range of core muscle injury patterns. The presence of fracture did not have a strong correlation with injury patterns, pain duration, or return to play time but may have implications for patient management.

Preoperative MRI Shoulder Findings Associated with Clinical Outcome 1 Year after Rotator Cuff Repair.

Background Investigation of the use of preoperative MRI for providing prognostic information regarding clinical outcome following rotator cuff repair has been limited. Purpose To determine whether patients with more severe rotator cuff tears of the shoulder at preoperative MRI have a greater degree of residual pain and disability after rotator cuff repair. Materials and Methods This retrospective study included a cohort of 141 patients who underwent surgical repair of a full-thickness rotator cuff tear at a single institution between April 16, 2012, and September 3, 2015. The mean patient age was 56.8 years, and there were 100 men (mean age, 56.1 years) and 41 women (mean age, 56.3 years). Patients completed the Disabilities of the Arm, Shoulder, and Hand (DASH) survey (lower score indicates less pain and disability) before and 1 year after surgery. One musculoskeletal radiologist blinded to the DASH scores measured the maximal anterior-posterior width and medial-lateral retraction of the rotator cuff tear on the preoperative MRI and assessed tendon degeneration and composite muscle atrophy and fatty infiltration using categorical grading scales (grade 0 indicates no tendon degeneration or muscle atrophy and fatty infiltration, and higher grades indicate incrementally more severe tendon degeneration or muscle atrophy and fatty infiltration). Generalized estimating equation models were used to determine the association between preoperative MRI findings and the postoperative DASH score. Results There was a significant positive association (P < .05) between the measured tear width (estimate, 2.05), measured tear retraction (estimate, 3.52), and tendon degeneration grade (estimate, 1.59) and the postoperative DASH score. There was no significant association (P = .49) between the composite muscle atrophy and fatty infiltration grade (estimate, 0.31) and the postoperative DASH score. Conclusion Patients with larger rotator cuff tears, more tendon retraction, and more severe tendon degeneration have worse clinical outcome scores 1 year after rotator cuff repair. © RSNA, 2019.

Deep Learning Approach for Evaluating Knee MR Images: Achieving High Diagnostic Performance for Cartilage Lesion Detection.

Purpose To determine the feasibility of using a deep learning approach to detect cartilage lesions (including cartilage softening, fibrillation, fissuring, focal defects, diffuse thinning due to cartilage degeneration, and acute cartilage injury) within the knee joint on MR images. Materials and Methods A fully automated deep learning-based cartilage lesion detection system was developed by using segmentation and classification convolutional neural networks (CNNs). Fat-suppressed T2-weighted fast spin-echo MRI data sets of the knee of 175 patients with knee pain were retrospectively analyzed by using the deep learning method. The reference standard for training the CNN classification was the interpretation provided by a fellowship-trained musculoskeletal radiologist of the presence or absence of a cartilage lesion within 17 395 small image patches placed on the articular surfaces of the femur and tibia. Receiver operating curve (ROC) analysis and the κ statistic were used to assess diagnostic performance and intraobserver agreement for detecting cartilage lesions for two individual evaluations performed by the cartilage lesion detection system. Results The sensitivity and specificity of the cartilage lesion detection system at the optimal threshold according to the Youden index were 84.1% and 85.2%, respectively, for evaluation 1 and 80.5% and 87.9%, respectively, for evaluation 2. Areas under the ROC curve were 0.917 and 0.914 for evaluations 1 and 2, respectively, indicating high overall diagnostic accuracy for detecting cartilage lesions. There was good intraobserver agreement between the two individual evaluations, with a κ of 0.76. Conclusion This study demonstrated the feasibility of using a fully automated deep learning-based cartilage lesion detection system to evaluate the articular cartilage of the knee joint with high diagnostic performance and good intraobserver agreement for detecting cartilage degeneration and acute cartilage injury. © RSNA, 2018 Online supplemental material is available for this article .