The clinical utility and diagnostic performance of magnetic resonance imaging for identification of early and advanced knee osteoarthritis: a systematic review

Integrating Radiomics and Neural Networks for Knee Osteoarthritis Incidence Prediction

OBJECTIVE

Accurately predicting knee osteoarthritis (KOA) is essential for early detection and personalized treatment. We aimed to develop and test a magnetic resonance imaging (MRI)-based joint space (JS) radiomic model (RM) to predict radiographic KOA incidence through neural networks by integrating meniscus and femorotibial cartilage radiomic features.

METHODS

In the Osteoarthritis Initiative cohort, participants with knees without radiographic KOA at baseline but at high risk for radiographic KOA were included. Patients' knees developed radiographic KOA, whereas control knees did not over four years. We randomly split the participants into development and test cohorts (8:2) and extracted features from baseline three-dimensional double-echo steady-state sequence MRI. Model performance was evaluated using an area under the receiver operating characteristic curve (AUC), sensitivity, and specificity in both cohorts. Nine resident surgeons performed the reader experiment without/with the JS-RM aid.

RESULTS

Our study included 549 knees in the development cohort (275 knees of patients with KOA vs 274 knees of controls) and 137 knees in the test cohort (68 knees of patients with KOA vs 69 knees of controls). In the test cohort, JS-RM had a favorable accuracy for predicting the radiographic KOA incidence with an AUC of 0.931 (95% confidence interval [CI] 0.876-0.963), a sensitivity of 84.4% (95% CI 83.9%-84.9%), and a specificity of 85.6% (95% CI 85.2%-86.0%). The mean specificity and sensitivity of resident surgeons through MRI reading in predicting radiographic KOA incidence were increased from 0.474 (95% CI 0.333-0.614) and 0.586 (95% CI 0.429-0.743) without the assistance of JS-RM to 0.874 (95% CI 0.847-0.901) and 0.812 (95% CI 0.742-0.881) with JS-RM assistance, respectively (P < 0.001).

CONCLUSION

JS-RM integrating the features of the meniscus and cartilage showed improved predictive values in radiographic KOA incidence.

Quantitative Magnetic Resonance Imaging Had Greater Sensitivity in Diagnosing Chondral Lesions of the Knee: A Systematic Review and Meta-Analysis

PURPOSE

To investigate the accuracy and reliability of magnetic resonance imaging (MRI) in identifying and grading chondral lesions and explore the optimal imaging technique to image cartilage.

METHOD

A comprehensive search was conducted on Medline, Embase, and Cochrane Library. Eligible cohort studies published before August 2022 were included. The study reports used MRI to diagnose and grade cartilage lesions, with intraoperative findings as the reference standard. Summary estimates of diagnostic performance were obtained. The reliability of MRI interpretation was summarized. Subgroup analyses were performed based on assessed imaging techniques, field strength, and joint surface.

RESULTS

Forty-three trials and 3,706 patients were included in the systematic review. The overall area under curve for hierarchical summarized receiver operating characteristics was 0.91 (95% confidence interval [CI] 0.88-0.93). The pooled sensitivity for quantitative MRI, 3-dimensional MRI, and 2-dimensional MRI was 0.82 (95% CI 0.64-0.92), 0.79 (95% CI 0.74-0.83), and 0.63 (95% CI 0.51-0.73), respectively. The pooled sensitivity of 3 Tesla (3T), 1.5 Tesla (1.5T), and <1.5 Tesla MRI was 0.79 (95% CI 0.72-0.85), 0.67 (95% CI 0.60-0.74), and 0.55 (95% CI 0.39-0.71), respectively. There were differences in interobserver consistency across different studies.

CONCLUSIONS

In general, MRI had high specificity in discriminating normal cartilage, but its sensitivity for identifying chondral lesions is less optimal. Further analysis showed that quantitative MRI, 3D MRI, and 3T MRI demonstrate greater sensitivity compared with 2D MRI, 1.5T MRI, and <1.5 Tesla MRI.

LEVEL OF EVIDENCE

Level III, systematic review of Level II-III studies.

In vivo MRI of knee using a metasurface-inspired wireless coil

PURPOSE

To investigate the application of a metasurface-inspired wireless coil and evaluate its performance in clinical knee MRI.

METHODS

A metasurface surface coil is designed for knee MRI at 1.5T. The image SNR and uniformity are assessed using a water phantom. In vivo studies are performed on 10 healthy volunteers (age range, 24-30 y; three males) and two patients (ages 31 and 76 y; two males) with knee conditions. A commercial 4-channel flexible coil and a 12-channel knee coil are used for comparison. The SNRs of different tissues on knee MRI images are evaluated and compared. The image quality is evaluated using a five-point Likert scale.

RESULTS

The SNRs of the images acquired by the metasurface coil with spine coil as receiving coil are similar to the 12-channel knee coil, whereas the uniformity from groups where the metasurface coil was used is higher than that acquired by conventional coils in phantom studies. For in vivo knee MRI, the SNRs of the images acquired by the metasurface coil with spine coil as receiving coil are between that of the 4- and 12-channel phased-array coils. The image quality scores evaluated by radiologists are higher when metasurface is used.

CONCLUSION

The metasurface-inspired wireless coil is applicable to clinical knee MRI. When used in conjunction with the spine coil, it provides a favorable SNR between that of the 4- and 12-channel phased-array coil at 1.5T MRI system. The metasurface coil improves image uniformity regardless of which coil is used as the receiving coil.

Application of machine learning-based multi-sequence MRI radiomics in diagnosing anterior cruciate ligament tears

OBJECTIVE

To compare the diagnostic power among various machine learning algorithms utilizing multi-sequence magnetic resonance imaging (MRI) radiomics in detecting anterior cruciate ligament (ACL) tears. Additionally, this research aimed to create and validate the optimal diagnostic model.

METHODS

In this retrospective analysis, 526 patients were included, comprising 178 individuals with ACL tears and 348 with a normal ACL. Radiomics features were derived from multi-sequence MRI scans, encompassing T1-weighted imaging and proton density (PD)-weighted imaging. The process of selecting the most reliable radiomics features involved using interclass correlation coefficient (ICC) testing, t tests, and the least absolute shrinkage and selection operator (LASSO) technique. After the feature selection process, five machine learning classifiers were created. These classifiers comprised logistic regression (LR), support vector machine (SVM), K-nearest neighbors (KNN), light gradient boosting machine (LightGBM), and multilayer perceptron (MLP). A thorough performance evaluation was carried out, utilizing diverse metrics like the area under the receiver operating characteristic curve (ROC), specificity, accuracy, sensitivity positive predictive value, and negative predictive value. The classifier exhibiting the best performance was chosen. Subsequently, three models were developed: the PD model, the T1 model, and the combined model, all based on the optimal classifier. The diagnostic performance of these models was assessed by employing AUC values, calibration curves, and decision curve analysis.

RESULTS

Out of 2032 features, 48 features were selected. The SVM-based multi-sequence radiomics outperformed all others, achieving AUC values of 0.973 and 0.927, sensitivities of 0.933 and 0.857, and specificities of 0.930 and 0.829, in the training and validation cohorts, respectively.

CONCLUSION

The multi-sequence MRI radiomics model, which is based on machine learning, exhibits exceptional performance in diagnosing ACL tears. It provides valuable insights crucial for the diagnosis and treatment of knee joint injuries, serving as an accurate and objective supplementary diagnostic tool for clinical practitioners.

Automated Radiographic Measurements of Knee Osteoarthritis

OBJECTIVE

Herewith, we report the development of Orthopedic Digital Image Analysis (ODIA) software that is developed to obtain quantitative measurements of knee osteoarthritis (OA) radiographs automatically. Manual segmentation and measurement of OA parameters currently hamper large-cohort analyses, and therefore, automated and reproducible methods are a valuable addition in OA research. This study aims to test the automated ODIA measurements and compare them with available manual Knee Imaging Digital Analysis (KIDA) measurements as comparison.

DESIGN

This study included data from the CHECK (Cohort Hip and Cohort Knee) initiative, a prospective multicentre cohort study in the Netherlands with 1,002 participants. Knee radiographs obtained at baseline of the CHECK cohort were included and mean medial/lateral joint space width (JSW), minimal JSW, joint line convergence angle (JLCA), eminence heights, and subchondral bone intensities were compared between ODIA and KIDA.

RESULTS

Of the potential 2,004 radiographs, 1,743 were included for analyses. Poor intraclass correlation coefficients (ICCs) were reported for the JLCA (0.422) and minimal JSW (0.299). The mean medial and lateral JSW, eminence height, and subchondral bone intensities reported a moderate to good ICC (0.7 or higher). Discrepancies in JLCA and minimal JSW between the 2 methods were mostly a problem in the lateral tibia plateau.

CONCLUSIONS

The current ODIA tool provides important measurements of OA parameters in an automated manner from standard radiographs of the knee. Given the automated and computerized methodology that has very high reproducibility, ODIA is suitable for large epidemiological cohorts with various follow-up time points to investigate structural progression, such as CHECK or the Osteoarthritis Initiative (OAI).

The ligamentum mucosum's potential as a preventative structure in the development of knee osteoarthritis

PURPOSE

This paper aimed to identify whether the presence, type, and/or morphology of the ligamentum mucosum could play a role in the development of knee osteoarthritis. Since its microscopic structure is alike that of other knee ligaments, it was hypothesized that its presence could facilitate knee motion and stability, thus preventing or reducing the extent of knee osteoarthritis.

METHODS

Thirty three cadavers (a total of 51 knees) were dissected. The ligamentum mucosum, if present, was measured with a digital caliber and a measuring tape in terms of length, width, and thickness. Knee osteoarthritis was assessed in six regions of the knee. The OuterBridge Classification System (Grades 0-4) was used to visually assess the extent, in addition to probing the area. Osteoarthritis was deemed present if the grade was 2 or greater.

RESULTS

The presence of the ligament was associated with a lower mean osteoarthritis level in the trochlear groove and lateral tibial plateau regions (p < 0.001 and p = 0.013, respectively). Overall osteoarthritis of the knee was also present at varying levels for each type of the ligamentum mucosum (p < 0.001). The patella and the medial condyle had the greatest levels of osteoarthritis, while the medial and lateral tibial plateaus had the lowest levels.

CONCLUSION

The presence of the ligamentum mucosum is linked with decreased osteoarthritis in the trochlear groove region. In addition, both the absent ligament and its classification as a vertical septum are associated with increased knee osteoarthritis.

LEVEL OF EVIDENCE

Five.

Deep learning applications in osteoarthritis imaging

Deep learning (DL) is one of the most exciting new areas in medical imaging. This article will provide a review of current applications of DL in osteoarthritis (OA) imaging, including methods used for cartilage lesion detection, OA diagnosis, cartilage segmentation, and OA risk assessment. DL techniques have been shown to have similar diagnostic performance as human readers for detecting and grading cartilage lesions within the knee on MRI. A variety of DL methods have been developed for detecting and grading the severity of knee OA and various features of knee OA on X-rays using standardized classification systems with diagnostic performance similar to human readers. Multiple DL approaches have been described for fully automated segmentation of cartilage and other knee tissues and have achieved higher segmentation accuracy than currently used methods with substantial reductions in segmentation times. Various DL models analyzing baseline X-rays and MRI have been developed for OA risk assessment. These models have shown high diagnostic performance for predicting a wide variety of OA outcomes, including the incidence and progression of radiographic knee OA, the presence and progression of knee pain, and future total knee replacement. The preliminary results of DL applications in OA imaging have been encouraging. However, many DL techniques require further technical refinement to maximize diagnostic performance. Furthermore, the generalizability of DL approaches needs to be further investigated in prospective studies using large image datasets acquired at different institutions with different imaging hardware before they can be implemented in clinical practice and research studies.

Detection and Quantitative Assessment of Arthroscopically Proven Long Biceps Tendon Pathologies Using T2 Mapping

This study evaluates how far T2 mapping can identify arthroscopically confirmed pathologies in the long biceps tendon (LBT) and quantify the T2 values in healthy and pathological tendon substance. This study comprised eighteen patients experiencing serious shoulder discomfort, all of whom underwent magnetic resonance imaging, including T2 mapping sequences, followed by shoulder joint arthroscopy. Regions of interest were meticulously positioned on their respective T2 maps, capturing the sulcal portion of the LBT and allowing for the quantification of the average T2 values. Subsequent analyses included the calculation of diagnostic cut-off values, sensitivities, and specificities for the detection of tendon pathologies, and the calculation of inter-reader correlation coefficients (ICCs) involving two independent radiologists. The average T2 value for healthy subjects was measured at 23.3 ± 4.6 ms, while patients with tendinopathy displayed a markedly higher value, at 47.9 ± 7.8 ms. Of note, the maximum T2 value identified in healthy tendons (29.6 ms) proved to be lower than the minimal value measured in pathological tendons (33.8 ms), resulting in a sensitivity and specificity of 100% (95% confidence interval 63.1-100) across all cut-off values ranging from 29.6 to 33.8 ms. The ICCs were found to range from 0.93 to 0.99. In conclusion, T2 mapping is able to assess and quantify healthy LBTs and can distinguish them from tendon pathology. T2 mapping may provide information on the (ultra-)structural integrity of tendinous tissue, facilitating early diagnosis, prompt therapeutic intervention, and quantitative monitoring after conservative or surgical treatments of LBT.

Application of quantitative T1, T2 and T2* mapping magnetic resonance imaging in cartilage degeneration of the shoulder joint

To investigate and compare the values of 3.0 T MRI T1, T2 and T2* mapping quantification techniques in evaluating cartilage degeneration of the shoulder joint. This study included 123 shoulder joints of 119 patients, which were scanned in 3.0 T MRI with axial Fat Suppression Proton Density Weighted Image (FS-PDWI), sagittal fat suppression T2 Weighted Image (FS-T2WI), coronal T1Weighted Image (T1WI), FS-PDWI, cartilage-specific T1, T2 and T2* mapping sequences. Basing on MRI images, the shoulder cartilage was classified into grades 0 1, 2, 3 and 4 according to the International Cartilage Regeneration & Joint Preservation Society (ICRS). The grading of shoulder cartilage was based on MRI images with ICRS as reference, and did not involve arthroscopy or histology.The T1, T2 and T2* relaxation values in the superior, middle and inferior bands of shoulder articular cartilage were measured at all grades, and the differences in various indicators between groups were analyzed and compared using a single-factor ANOVA test. The correlation between T1, T2 and T2* relaxation values and MRI-based grading was analyzed by SPSS software. There were 46 shoulder joints with MRI-based grade 0 in healthy control group (n = 46), while 49 and 28 shoulder joints with grade 1-2 (mild degeneration subgroup) and grade 3-4 (severe degeneration subgroup) in patient group (n = 73), accounting for 63.6% and 36.4%, respectively. The T1, T2 and T2* relaxation values of the superior, middle and inferior bands of shoulder articular cartilage were significantly and positively correlated with the MRI-based grading (P < 0.01). MRI-basedgrading of shoulder cartilage was markedly associated with age (r = 0.766, P < 0.01). With the aggravation of cartilage degeneration, T1, T2 and T2* relaxation values showed an upward trend (all P < 0.01), and T1, T2 and T2* mapping could distinguish cartilage degeneration at all levels (all P < 0.01). The T1, T2 and T2* relaxation values were significantly different between normal group and mild degeneration subgroup, normal group and severe degeneration subgroup, mild degeneration subgroup and severe degeneration subgroup (all P < 0.05). Quantitative T1, T2 and T2* mapping can quantify the degree of shoulder cartilage degeneration. All these MRI mapping quantification techniques can be used as critical supplementary sequences to assess shoulder cartilage degeneration, among which T2 mapping has the highest value.

fastMRI+, Clinical pathology annotations for knee and brain fully sampled magnetic resonance imaging data

Improving speed and image quality of Magnetic Resonance Imaging (MRI) using deep learning reconstruction is an active area of research. The fastMRI dataset contains large volumes of raw MRI data, which has enabled significant advances in this field. While the impact of the fastMRI dataset is unquestioned, the dataset currently lacks clinical expert pathology annotations, critical to addressing clinically relevant reconstruction frameworks and exploring important questions regarding rendering of specific pathology using such novel approaches. This work introduces fastMRI+, which consists of 16154 subspecialist expert bounding box annotations and 13 study-level labels for 22 different pathology categories on the fastMRI knee dataset, and 7570 subspecialist expert bounding box annotations and 643 study-level labels for 30 different pathology categories for the fastMRI brain dataset. The fastMRI+ dataset is open access and aims to support further research and advancement of medical imaging in MRI reconstruction and beyond.

Measurement(s)	Pathotology annotations in knee and brain MRI images
Technology Type(s)	Expert delineation

The Importance of Staging Arthroscopy for Chondral Defects of the Knee

This study aims to evaluate the role of staging arthroscopy in the diagnosis of knee chondral defects and subsequent surgical planning prior to autologous chondrocyte implantation (ACI), osteochondral allograft transplantation (OCA), and meniscus allograft transplantation (MAT). All patients who underwent staging arthroscopy prior to ACI, OCA, or MAT at our institution from 2005 to 2015 were identified. Medical records were reviewed to document the diagnosis and treatment plan based on symptoms, magnetic resonance imaging (MRI) findings and previous operative records. Operative records of the subsequent staging arthroscopy procedure were reviewed to document the proposed treatment plan after arthroscopy. All changes in treatment plan following staging arthroscopy were recorded. Univariate analyses were performed to identify any significant predictors for likelihood to change. A total of 98 patients were included in our analysis. A change in surgical plan was made following arthroscopy in 36 patients (36.7%). Fourteen patients (14.3%) were found to have additional defects that warranted cartilage restoration surgery. In 15 patients (15.3%), at least one defect that was originally thought to warrant cartilage restoration surgery was found to be amenable to debridement alone. The surgical plan was changed from ACI to OCA in four cases (4.1%) and OCA to ACI in one case (1%). A previously proposed MAT was deemed unwarranted in one case (1%), and a planned meniscal repair was changed to MAT in another (1%). Patient age, sex, and the affected knee compartment were not predictors for a change in surgical plan. Body mass index (BMI) was significantly higher in patients who had a change in surgical plan (29.5 kg/m²) compared with those who did not (26.5 kg/m²). A change in surgical plan was more likely to occur for trochlear lesions (46.4%) compared with other articular surface lesions (p = 0.008). The results of our study indicate that staging arthroscopy is an important step in determining the most appropriate treatment plan for chondral defects and meniscal deficiency, particularly those with trochlear cartilage lesions.

Feasibility of an accelerated 2D-multi-contrast knee MRI protocol using deep-learning image reconstruction: a prospective intraindividual comparison with a standard MRI protocol

OBJECTIVES

The aim of this study was to evaluate the image quality and diagnostic performance of a deep-learning (DL)-accelerated two-dimensional (2D) turbo spin echo (TSE) MRI of the knee at 1.5 and 3 T in clinical routine in comparison to standard MRI.

MATERIAL AND METHODS

Sixty participants, who underwent knee MRI at 1.5 and 3 T between October/2020 and March/2021 with a protocol using standard 2D-TSE (TSE_S) and DL-accelerated 2D-TSE sequences (TSE_DL), were enrolled in this prospective institutional review board-approved study. Three radiologists assessed the sequences regarding structural abnormalities and evaluated the images concerning overall image quality, artifacts, noise, sharpness, subjective signal-to-noise ratio, and diagnostic confidence using a Likert scale (1-5, 5 = best).

RESULTS

Overall image quality for TSE_DL was rated to be excellent (median 5, IQR 4-5), significantly higher compared to TSE_S (median 5, IQR 4 - 5, p < 0.05), showing significantly lower extents of noise and improved sharpness (p < 0.001). Inter- and intra-reader agreement was almost perfect (κ = 0.92-1.00) for the detection of internal derangement and substantial to almost perfect (κ = 0.58-0.98) for the assessment of cartilage defects. No difference was found concerning the detection of bone marrow edema and fractures. The diagnostic confidence of TSE_DL was rated to be comparable to that of TSE_S (median 5, IQR 5-5, p > 0.05). Time of acquisition could be reduced to 6:11 min using TSE_DL compared to 11:56 min for a protocol using TSE_S.

CONCLUSION

TSE_DL of the knee is clinically feasible, showing excellent image quality and equivalent diagnostic performance compared to TSE_S, reducing the acquisition time about 50%.

KEY POINTS

• Deep-learning reconstructed TSE imaging is able to almost halve the acquisition time of a three-plane knee MRI with proton density and T1-weighted images, from 11:56 min to 6:11 min at 3 T. • Deep-learning reconstructed TSE imaging of the knee provided significant improvement of noise levels (p < 0.001), providing higher image quality (p < 0.05) compared to conventional TSE imaging. • Deep-learning reconstructed TSE imaging of the knee had similar diagnostic performance for internal derangement of the knee compared to standard TSE.

Relationship between Classification of Fabellae and the Severity of Knee Osteoarthritis: A Relevant Study in the Chinese Population

Objective

To classify the fabellae and discuss the relationship between the classification of fabellae and the severity of knee osteoarthritis (KOA) in Chinese.

Methods

From February 2019 to February 2020, 136 patients were measured and classified using three‐dimensional computed tomography (CT) reconstruction. According to the CT imaging characteristics, the fabellae were divided into five types: type I, a fabella on the lateral femoral condyle; type II, a fabella on the medial femoral condyle; type III, a fabella on the lateral femoral condyle and a fabella on the medial femoral condyle; type IV, two fabellae on the medial femoral condyle; and type V, two fabellae on the lateral femoral condyle. The severity of KOA was assessed on the Recht grade by magnetic resonance imaging (MRI). The data were analyzed with SPSS 24.0.

Results

The classification of fabellae were correlated with KOA grades (χ² = 35.026, P < 0.05). In terms of KOA grades, grade I and grade II were occupied most by fabellar type II (32, 72.8%); type II and other types showed significant statistical difference (P < 0.05). Grade I and grade II were also mainly fabellar type IV (four, 100%). Fabellar type V's biggest component was grade III and grade IV (six, 75%). Type IV and type V showed significant statistical difference (P < 0.05).

Conclusion

The classification of fabellae were correlated with KOA grades. The type II may mean the lower KOA grades while type V may mean the higher KOA grades.

Diagnostic Accuracy of Magnetic Resonance Images and Weight-Bearing Radiographs in Patients With Arthroscopic-Proven Medial Osteoarthritis of the Knee

Aims:

The aim of this study is to analyze the diagnostic value of weight-bearing radiographs, magnetic resonance images (MRI), and the combination of both in osteoarthritic knees when using arthroscopic findings as the “gold standard” to compare with.

Methods:

A total of 59 patients were studied because of chronic pain in 1 of their knees. Radiographs were classified according to Kellgren-Lawrence scale. Magnetic resonance images were classified according to Vallotton, and arthroscopic findings according to Outerbridge criteria.

Results:

Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were, respectively, 75.0%, 60.0%, 56.2%, 77.8%, and 66.1% for weight-bearing radiographs, and 70.8%, 88.6%, 81.0%, 81.6%, and 81.4% for MRI. Logistic regression analysis showed that a weight-bearing radiograph added to MRI offered no additional diagnostic value compared with MRI alone (P < .001).

Conclusions:

Magnetic resonance images presented higher specificity, positive and negative predictive values, and accuracy than weight-bearing radiographs for knee osteoarthritis. The combination of radiographs and MRI did not improve the diagnostic accuracy, compared with MRI alone.

Deep Learning Approach for Anterior Cruciate Ligament Lesion Detection: Evaluation of Diagnostic Performance Using Arthroscopy as the Reference Standard

BACKGROUND

MRI is the most commonly used imaging method for diagnosing anterior cruciate ligament (ACL) injuries. However, the interpretation of knee MRI is time-intensive and depends on the clinical experience of the reader. An automated detection system based on a deep-learning algorithm may improve interpretation time and reliability.

PURPOSE

To determine the feasibility of using a deep learning approach to detect ACL injuries within the knee joint on MRI.

STUDY TYPE

Retrospective.

POPULATION

In all, 163 subjects with an ACL tear and 245 subjects with an intact ACL. There were 285, 81, and 42 volumes for training, validation, and test sets, respectively.

FIELD STRENGTH/SEQUENCE

2D sagittal proton density-weighted spectral attenuated inversion recovery sequences at 1.5T and 3.0T.

ASSESSMENT

Based on the architecture of 3D DenseNet, we constructed a classification convolutional neural network. We tested this deep learning approach with different inputs and two other algorithms, including VGG16 and ResNet. Then we had both inexperienced radiologists and senior radiologists read the MR images.

STATISTICAL TESTS

Using arthroscopic results as the reference standard, the performance of three different inputs and three different algorithms, the residents and senior radiologists assessed the classification accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristic curve (AUC).

RESULTS

The accuracy, sensitivity, specificity, PPV, and NPV of our customized 3D deep learning architecture was 0.957, 0.976, 0.944, 0.940, and 0.976, respectively. The average AUCs were 0.946, 0.859, 0.960 for ResNet, VGG16, and our proposed network, respectively. The diagnostic accuracy of our model, residents, and senior radiologists was 0.957, 0.814, and 0.899, respectively.

DATA CONCLUSION

Our study demonstrated the feasibility of using an automated deep-learning-based detection system to evaluate ACL injury.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 1 J. MAGN. RESON. IMAGING 2020;52:1745-1752.

Long-term results following intra-articular fractures of the medial malleolus in children and adolescents with special emphasis on MRI

PURPOSE

Presently, there is no publication combining clinical follow-up with magnetic resonance imaging (MRI) to determine possible post-traumatic alterations following paediatric intra-articular fractures of the medial malleolus. Therefore, the aims of this study were to retrospectively analyze a cohort of patients with Salter-Harris (SH) III and IV fractures of the medial malleolus and to evaluate their long-term outcome. MRI was used to assess possible changes of the articular surface that cannot be diagnosed on native radiographs.

PATIENTS AND METHODS

Fifty-four patients with SH III ( n = 38) or IV fractures ( n = 17) of the medial malleolus treated between 2001 and 2011 were invited for a follow-up examination. Clinical outcome was assessed with the Weber score and osteoarthritis with the Kellgren and Lawrence classification. MRI was rated by the Outerbridge classification. Correlations between the clinical and radiological outcomes were calculated.

RESULTS

Seventeen patients were recruited for long-term follow-up at a mean of 112 (range, 65-184) months. The Weber score was very good for 5 patients, good for 10 patients and poor for 2 patients; the Kellgren and Lawrence score revealed a favourable grade 0 in 15 patients and grade 1 in 2 patients. The MRI-based Outerbridge classification yielded grade 0 for 12 patients, grade 1 for 1 patient, grade 2 for 2 patients and grade 3 for 1 patient. The Outerbridge score significantly ( p < 0.05) correlated with the Weber score.

CONCLUSION

This study shows excellent and good outcome of SH III and IV fractures of the medial malleolus. Worse clinical outcome correlated with post-traumatic changes of the articular cartilage seen on MRI.

Beyond Cartilage Repair: The Role of the Osteochondral Unit in Joint Health and Disease

Once believed to be limited to articular cartilage, osteoarthritis is now considered to be an organ disease of the “whole joint.” Damage to the articular surface can lead to, be caused by, or occur in parallel with, damage to other tissues in the joint. The relationship between cartilage and the underlying subchondral bone has particular importance when assessing joint health and determining treatment strategies. The articular cartilage is anchored to the subchondral bone through an interface of calcified cartilage, which as a whole makes up the osteochondral unit. This unit functions primarily by transferring load-bearing weight over the joint to allow for normal joint articulation and movement. Unfortunately, irreversible damage and degeneration of the osteochondral unit can severely limit joint function. Our understanding of joint pain, the primary complaint of patients, is poorly understood and past efforts toward structural cartilage restoration have often not been associated with a reduction in pain. Continued research focusing on the contribution of subchondral bone and restoration of the entire osteochondral unit are therefore needed, with the hope that this will lead to curative, and not merely palliative, treatment options. The purpose of this narrative review is to investigate the role of the osteochondral unit in joint health and disease. Topics of discussion include the crosstalk between cartilage and bone, the efficacy of diagnostic procedures, the origins of joint pain, current and emerging treatment paradigms, and suitable preclinical animal models for safety and efficacy assessment of novel osteochondral therapies. The goal of the review is to facilitate an appreciation of the important role played by the subchondral bone in joint pain and why the osteochondral unit as a whole should be considered in many cases of joint restoration strategies.

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 .

A Prospective, Blinded, Multicenter Clinical Trial to Compare the Efficacy, Accuracy, and Safety of In-Office Diagnostic Arthroscopy With Magnetic Resonance Imaging and Surgical Diagnostic Arthroscopy

PURPOSE

The purpose of this study was to compare the efficacy, accuracy, and safety of in-office diagnostic arthroscopy with magnetic resonance imaging (MRI) and surgical diagnostic arthroscopy.

METHODS

A prospective, blinded, multicenter, clinical trial was performed on 110 patients, ages 18 to 75 years, who presented with knee pain. The study period was April 2012 to April 2013. Each patient underwent a physical examination, an MRI, in-office diagnostic imaging, and a diagnostic arthroscopic examination in the operating room. The attending physician completed clinical report forms comparing the in-office arthroscopic examination and surgical diagnostic arthroscopy findings on each patient. Two blinded experts, unaffiliated with the clinical care of the study's subjects, reviewed the in-office arthroscopic images and MRI images using the surgical diagnostic arthroscopy images as the "control" group comparison. Patients were consecutive, and no patients were excluded from the study.

RESULTS

In this study, the accuracy, sensitivity, and specificity of in-office arthroscopy was equivalent to surgical diagnostic arthroscopy and more accurate than MRI. When comparing in-office arthroscopy with surgical diagnostic arthroscopy, all kappa statistics were between 0.766 and 0.902. For MRI compared with surgical diagnostic arthroscopy, kappa values ranged from a low of 0.130 (considered "slight" agreement) to a high of 0.535 (considered "moderate" agreement). The comparison of MRI to in-office arthroscopy showed very similar results as the comparison of MRI with surgical diagnostic arthroscopy, ranging from a low kappa of 0.112 (slight agreement) to a high of 0.546 (moderate agreement). There were no patient-related or device-related complications related to the use of in-office arthroscopy.

CONCLUSIONS

Needle-based diagnostic imaging that can be used in the office setting is statistically equivalent to surgical diagnostic arthroscopy with regard to the diagnosis of intra-articular, nonligamentous knee joint pathology. In-office diagnostic imaging can provide a more detailed and accurate diagnostic assessment of intra-articular knee pathology than MRI. Based on the study results, in-office diagnostic imaging provides a safe, accurate, real-time, minimally invasive diagnostic modality to evaluate intra-articular pathology without the need for surgical diagnostic arthroscopy or high-cost imaging.

LEVEL OF EVIDENCE

Level II, comparative prospective trial.

Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging

PURPOSE

To describe and evaluate a new fully automated musculoskeletal tissue segmentation method using deep convolutional neural network (CNN) and three-dimensional (3D) simplex deformable modeling to improve the accuracy and efficiency of cartilage and bone segmentation within the knee joint.

METHODS

A fully automated segmentation pipeline was built by combining a semantic segmentation CNN and 3D simplex deformable modeling. A CNN technique called SegNet was applied as the core of the segmentation method to perform high resolution pixel-wise multi-class tissue classification. The 3D simplex deformable modeling refined the output from SegNet to preserve the overall shape and maintain a desirable smooth surface for musculoskeletal structure. The fully automated segmentation method was tested using a publicly available knee image data set to compare with currently used state-of-the-art segmentation methods. The fully automated method was also evaluated on two different data sets, which include morphological and quantitative MR images with different tissue contrasts.

RESULTS

The proposed fully automated segmentation method provided good segmentation performance with segmentation accuracy superior to most of state-of-the-art methods in the publicly available knee image data set. The method also demonstrated versatile segmentation performance on both morphological and quantitative musculoskeletal MR images with different tissue contrasts and spatial resolutions.

CONCLUSION

The study demonstrates that the combined CNN and 3D deformable modeling approach is useful for performing rapid and accurate cartilage and bone segmentation within the knee joint. The CNN has promising potential applications in musculoskeletal imaging. Magn Reson Med 79:2379-2391, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

The Benefits of an In-Office Arthroscopy in the Diagnosis of Unresolved Knee Pain

We report a patient who developed persistent knee pain with mechanical symptoms after an uncomplicated patellofemoral arthroplasty. The etiology of his knee pain remained inconclusive following magnetic resonance imaging due to metallic artifact image distortion. With the use of an in-office needle arthroscopy, an immediate and definitive diagnosis was obtained, preventing an unnecessary surgery for a diagnostic arthroscopy. We discovered a lateral meniscus tear, an anterior cruciate ligament tear, and a medial femoral condyle chondral defect for which the patient underwent arthroscopic partial meniscectomy, ligament reconstruction, and osteochondral allograft transplantation, with resolution of his knee pain.

MRI T2 Mapping of Knee Articular Cartilage Using Different Acquisition Sequences and Calculation Methods at 1.5 Tesla

OBJECTIVE

This study aims to determine how magnetic resonance imaging (MRI) acquisition techniques and calculation methods affect T2 values of knee cartilage at 1.5 tesla and to identify sequences that can be used for high-resolution T2 mapping in short scanning times.

MATERIALS AND METHODS

This study was performed on phantom and 29 patients who underwent MRI of the knee joint at 1.5 tesla. The protocol includes T2 mapping sequences based on Single-Echo Spin Echo (SESE), Multi-Echo Spin Echo (MESE), Fast Spin Echo (FSE) and Turbo Gradient Spin Echo (TGSE). The T2 relaxation times were quantified and evaluated using three calculation methods (MapIt, Syngo Offline and mono-exponential fit). signal-to-noise ratios (SNR) were measured in all sequences. All statistical analyses were performed using the t-test.

RESULTS

The average T2 values in phantom were 41.7 ± 13.8 ms for SESE, 43.2 ± 14.4 ms for MESE, 42.4 ± 14.1 ms for FSE and 44 ± 14.5 ms for TGSE. In the patient study, the mean differences were 6.5 ± 8.2 ms, 7.8 ± 7.6 ms and 8.4 ± 14.2 ms for MESE, FSE and TGSE compared to SESE, respectively; these statistical results were not significantly different (p > 0.05). The comparison between the three calculation methods showed no significant difference (p > 0.05). The t-test showed no significant difference between SNR values for all sequences.

CONCLUSION

T2 values depend not only on the sequence type but also on the calculation method. None of the sequences revealed significant differences compared to the SESE reference sequence. TGSE with its short scanning time can be used for high-resolution T2 mapping.

Comparison of a fast 5-min knee MRI protocol with a standard knee MRI protocol: a multi-institutional multi-reader study

PURPOSE

To compare diagnostic performance of a 5-min knee MRI protocol to that of a standard knee MRI.

MATERIALS AND METHODS

One hundred 3 T (100 patients, mean 38.8 years) and 50 1.5 T (46 patients, mean 46.4 years) MRIs, consisting of 5 fast, 2D multi-planar fast-spin-echo (FSE) sequences and five standard multiplanar FSE sequences, from two academic centers (1/2015-1/2016), were retrospectively reviewed by four musculoskeletal radiologists. Agreement between fast and standard (interprotocol agreement) and between standard (intraprotocol agreement) readings for meniscal, ligamentous, chondral, and bone pathology was compared for interchangeability. Frequency of major findings, sensitivity, and specificity was also tested for each protocol.

RESULTS

Interprotocol agreement using fast MRI was similar to intraprotocol agreement with standard MRI (83.0-99.5%), with no excess disagreement (≤ 1.2; 95% CI, -4.2 to 3.8%), across all structures. Frequency of major findings (1.1-22.4% across structures) on fast and standard MRI was not significantly different (p ≥ 0.215), except more ACL tears on fast MRI (p = 0.021) and more cartilage defects on standard MRI (p < 0.001). Sensitivities (59-100%) and specificities (73-99%) of fast and standard MRI were not significantly different for meniscal and ligament tears (95% CI for difference, -0.08-0.08). For cartilage defects, fast MRI was slightly less sensitive (95% CI for difference, -0.125 to -0.01) but slightly more specific (95% CI for difference, 0.01-0.5) than standard MRI.

CONCLUSION

A fast 5-min MRI protocol is interchangeable with and has similar accuracy to a standard knee MRI for evaluating internal derangement of the knee.

Is arthroscopic videotape a reliable tool for describing early joint tissue pathology of the knee?

BACKGROUND

The aim of this study was to assess the reliability of the videotape scoring system collected during knee arthroscopy for meniscal tears, and to test the consistency with intra-operative findings.

METHODS

Fifty-seven patients undergoing arthroscopic treatment of meniscal tears were included in the analysis. The cartilage damage was assessed intra-operatively, at six sites, using the Outerbridge scoring system. The meniscal tears were graded according to the International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine (ISAKOS) classification. The presence of synovial inflammation was scored, at four different sites, according to the macro-score method. The total cartilage, meniscal and synovial scores were calculated for each knee by the sum of the maximum grade of the lesions at each anatomic site. In order to assess the reliability of the arthroscopic scoring system, the videotapes recorded during the arthroscopy were reviewed by two independent observers one year after the surgery and compared with the scores completed by the surgeon during the operation.

RESULTS

The total cartilage score showed substantial inter-rater and intra-rater reliability, and moderate consistency with the intra-operative score provided by the surgeon. The total meniscal score showed substantial inter-rater and intra-rater reliability, and good consistency with the intra-operative findings. The total synovial score showed substantial inter-rater agreement, substantial intra-rater reliability, and moderate consistency with intra-operative findings.

CONCLUSIONS

The videotape scoring system provided a reliable and reproducible system for recording the intra-operative state of the whole joint pathology in patients undergoing arthroscopic treatment of meniscal tears.

In-office arthroscopy for the evaluation of chronic knee pain: A case report

This is a case report detailing the use of in-office needle arthroscopy (mi-eye 2™) in a patient with chronic knee pain and inconclusive magnetic resonance imaging findings. The patient is a 40-year-old male who presented to our clinic after an extended history of right knee pain along the medial aspect with previous failed treatments. Magnetic resonance imaging without contrast had demonstrated full-thickness chondral fissuring of the lateral patellar facet, mild abnormal signals of the proximal patellar tendon and Hoffa’s fat pad, and intact anterior cruciate ligament and posterior cruciate ligament. The patient was previously treated with an ultrasound-guided injection of 2 cm³ of 1% lidocaine without epinephrine and 1 cm³ of Kenalog-40 and scheduled for follow-up. At follow-up, clinical examination showed antalgic gait, minimal tenderness along medial joint line, medial pain in deep flexion, and no pain when in varus or valgus. Due to continued discomfort with a negative magnetic resonance imaging, in-office diagnostic arthroscopy was performed using mi-eye 2 revealing a tear of the mid-body of the medial meniscus. The patient subsequently underwent arthroscopic repair and is recovering well with complete resolution of medial joint pain. This report highlights the clinical utility of in-office diagnostic arthroscopy in the management of patients with persistent knee pain and negative or equivocal findings on magnetic resonance imaging.

Routine clinical knee MR reports: comparison of diagnostic performance at 1.5 T and 3.0 T for assessment of the articular cartilage

OBJECTIVE

Accurate assessment of knee articular cartilage is clinically important. Although 3.0 Tesla (T) MRI is reported to offer improved diagnostic performance, literature regarding the clinical impact of MRI field strength is lacking. The purpose of this study is to compare the diagnostic performance of clinical MRI reports for assessment of cartilage at 1.5 and 3.0 T in comparison to arthroscopy.

MATERIALS AND METHODS

This IRB-approved retrospective study consisted of 300 consecutive knees in 297 patients who had routine clinical MRI and arthroscopy. Descriptions of cartilage from MRI reports of 165 knees at 1.5 T and 135 at 3.0 T were compared with arthroscopy. The sensitivity, specificity, percent of articular surfaces graded concordantly, and percent of articular surfaces graded within one grade of the arthroscopic grading were calculated for each articular surface at 1.5 and 3.0 T. Agreement between MRI and arthroscopy was calculated with the weighted-kappa statistic. Significance testing was performed utilizing the z-test after bootstrapping to obtain the standard error.

RESULTS AND CONCLUSIONS

The sensitivity, specificity, percent of articular surfaces graded concordantly, and percent of articular surfaces graded within one grade were 61.4%, 82.7%, 62.2%, and 77.5% at 1.5 T and 61.8%, 80.6%, 59.5%, and 75.6% at 3.0 T, respectively. The weighted kappa statistic was 0.56 at 1.5 T and 0.55 at 3.0 T. There was no statistically significant difference in any of these parameters between 1.5 and 3.0 T. Factors potentially contributing to the lack of diagnostic advantage of 3.0 T MRI are discussed.

Cartilage morphology at 2-3 years following anterior cruciate ligament reconstruction with or without concomitant meniscal pathology

PURPOSE

To examine differences in cartilage morphology between young adults 2-3 years post-anterior cruciate ligament reconstruction (ACLR), with or without meniscal pathology, and control participants.

METHODS

Knee MRI was performed on 130 participants aged 18-40 years (62 with isolated ACLR, 38 with combined ACLR and meniscal pathology, and 30 healthy controls). Cartilage defects, cartilage volume and bone marrow lesions (BMLs) were assessed from MRI using validated methods.

RESULTS

Cartilage defects were more prevalent in the isolated ACLR (69 %) and combined group (84 %) than in controls (10 %, P < 0.001). Furthermore, the combined group showed higher prevalence of cartilage defects on medial femoral condyle (OR 4.7, 95 % CI 1.3-16.6) and patella (OR 7.8, 95 % CI 1.5-40.7) than the isolated ACLR group. Cartilage volume was lower in both ACLR groups compared with controls (medial tibia, lateral tibia and patella, P < 0.05), whilst prevalence of BMLs was higher on lateral tibia (P < 0.001), with no significant differences between the two ACLR groups for either measure.

CONCLUSIONS

Cartilage morphology was worse in ACLR patients compared with healthy controls. ACLR patients with associated meniscal pathology have a higher prevalence of cartilage defects than ACLR patients without meniscal pathology. The findings suggest that concomitant meniscal pathology may lead to a greater risk of future OA than isolated ACLR.

LEVEL OF EVIDENCE

III.

The prevalence of patellofemoral osteoarthritis: a systematic review and meta-analysis

OBJECTIVE

To determine the prevalence of radiographic patellofemoral osteoarthritis (OA) from population- and symptom-based cohorts and to evaluate if knee pain, physical function and quality of life (QOL) differ between people with isolated patellofemoral OA, isolated tibiofemoral OA and combined patellofemoral and tibiofemoral OA.

METHOD

Terms associated with "patellofemoral OA", "prevalence" and "clinical features" were used to search Medline, EMBASE, CINAHL, SCOPUS, AMED and Web of Science databases with no language restriction' from inception to August 2014. Two independent reviewers screened papers for eligibility. Studies were included if they reported prevalence of compartmental patterns of radiographic knee OA in population- or symptom-based cohorts. Studies were excluded if they evaluated a targeted sample (e.g., occupation-specific participants) or repeated already reported data from the same cohorts. Point prevalence estimates of patellofemoral OA were extracted from eligible studies, pooled and quantitatively analysed. A critical appraisal tool was used to evaluate methodological quality.

RESULTS

The search yielded 1891 records. The inclusion criteria were met by 32 studies. The crude prevalence of patellofemoral OA was 25% in the population-based cohorts (aged >20 years) and 39% in the symptom-based cohorts (aged >30 years). Eight studies reported knee pain, physical function and QOL in people with different compartmental disease; however no significant differences were found.

CONCLUSION

These findings confirm the substantial prevalence of patellofemoral OA, demonstrating the need to specifically consider the patellofemoral joint in knee OA research and clinical settings.

Diagnosis and classification of chondral knee injuries: comparison between magnetic resonance imaging and arthroscopy

PURPOSE

To compare the magnetic resonance imaging (MRI) findings of patients undergoing knee arthroscopy for chondral lesions. The hypothesis was that MRI displays low sensitivity in the diagnosis and classification of chondral injuries.

METHODS

A total of 83 knees were evaluated. The MRIs were performed using the same machine (GE SIGNA HDX 1.45 T). The MRI results were compared with the arthroscopy findings, and an agreement analysis was performed. Thirty-eight of the 83 MRI exams were evaluated by another radiologist for inter-observer agreement analysis. These analyses were performed using the kappa (κ) coefficient.

RESULTS

The highest incidence of chondral injury was in the patella (14.4 %). The κ coefficient was 0.31 for the patellar surface; 0.38 for the trochlea; 0.46 for the medial femoral condyle; 0.51 for the lateral femoral condyle; and 0.19 for the lateral plateau. After dividing the injuries into two groups (ICRS Grades 0-II and Grades III and IV), the following κ coefficients were obtained as follows: 0.49 (patella); 0.53 (trochlea); 0.46 (medial femoral condyle); 0.43 (medial plateau); 0.67 (lateral femoral condyle); and 0.51 (lateral plateau). The MRI sensitivity was 76.4 % (patella), 88.2 % (trochlea), 69.7 % (medial femoral condyle), 85.7 % (medial plateau), 81.8 % (lateral femoral condyle) and 75 % (lateral plateau). Comparing the radiologists' evaluations, the following κ coefficients were obtained as follows: 0.73 (patella); 0.63 (trochlea); 0.84 (medial femoral condyle); 0.72 (medial plateau); 0.77 (lateral femoral condyle); and 0.91 (lateral plateau).

CONCLUSION

Compared with arthroscopy, MRI displays moderate sensitivity for detecting and classifying chondral knee injuries. It is an important image method, but we must be careful in the assessment of patients with suspected chondral lesions.

LEVEL OF EVIDENCE

III.

Magnetic resonance tomography of the knee joint

OBJECTIVES

To compare the diagnostic performance of magnetic resonance imaging (MRI) in terms of sensitivity and specificity using a field strength of <1.0 T (T) versus ≥1.5 T for diagnosing or ruling out knee injuries or knee pathologies.

METHODS

The systematic literature research revealed more than 10,000 references, of which 1598 abstracts were reviewed and 87 full-text articles were retrieved. The further selection process resulted in the inclusion of four systematic reviews and six primary studies.

RESULTS

No differences could be identified in the diagnostic performance of low- versus high-field MRI for the detection or exclusion of meniscal or cruciate ligament tears. Regarding the detection or grading of cartilage defects and osteoarthritis of the knee, the existing evidence suggests that high-field MRI is tolerably specific but not very sensitive, while there is literally no evidence for low-field MRI because only a few studies with small sample sizes and equivocal findings have been performed.

CONCLUSIONS

We can recommend the use of low-field strength MRI systems in suspected meniscal or cruciate ligament injuries. This does, however, not apply to the diagnosis and grading of knee cartilage defects and osteoarthritis because of insufficient evidence.

Cartilage status in knees with recurrent patellar instability using magnetic resonance imaging T2 relaxation time value

PURPOSE

The purpose of the study was to analyse the patellar status in patients with recurrent patellar instability and in healthy controls using magnetic resonance imaging (MRI).

METHODS

One hundred and thirteen patients with patellar instability (patellar instability group) and 50 healthy controls (control group) were studied. All patients underwent computed tomography (CT) to measure the patellofemoral joint anatomy. Meanwhile, MRI was used to investigate cartilage status and to determine the T2 relaxation time value of the patellar cartilage plate. The mean values of these parameters for the patellar instability group and the control group were compared.

RESULTS

The CT images of the patellar instability group revealed greatly abnormal anatomy of the patellofemoral joint compared with the control group. By MRI assessment, 32 patients were determined to have a cartilage defect, which was confirmed under arthroscopy. Among the other 81 patients, the T2 relaxation time value of the patellar instability group was significantly higher than that of the control group on the middle (p = 0.032) or medial sites (p = 0.041) of the patellar cartilage.

CONCLUSIONS

The patellar instability group exhibits a higher risk of cartilage lesions with abnormal patellofemoral joint anatomy. MRI may enable early detection of these cartilage defects within the patellofemoral joint, enabling clinicians to adopt strategies to delay or prevent cartilage degeneration.

LEVEL OF EVIDENCE

III.

Imaging following acute knee trauma

Joint injury has been recognized as a potent risk factor for the onset of osteoarthritis. The vast majority of studies using imaging technology for longitudinal assessment of patients following joint injury have focused on the injured knee joint, specifically in patients with anterior cruciate ligament injury and meniscus tears where a high risk for rapid onset of post-traumatic osteoarthritis is well known. Although there are many imaging modalities under constant development, magnetic resonance (MR) imaging is the most important instrument for longitudinal monitoring after joint injury. MR imaging is sensitive for detecting early cartilage degeneration and can evaluate other joint structures including the menisci, bone marrow, tendons, and ligaments which can be sources of pain following acute injury. In this review, focusing on imaging following acute knee trauma, several studies were identified with promising short-term results of osseous and soft tissue changes after joint injury. However, studies connecting these promising short-term results to the development of osteoarthritis were limited which is likely due to the long follow-up periods needed to document the radiographic and clinical onset of the disease. Thus, it is recommended that additional high quality longitudinal studies with extended follow-up periods be performed to further investigate the long-term consequences of the early osseous and soft tissue changes identified on MR imaging after acute knee trauma.

In vitro toxicity in long-term cell culture of MR contrast agents targeted to cartilage evaluation

OBJECTIVE

Contrast-enhanced magnetic resonance (MR) imaging methods have been proposed for non-invasive evaluation of osteoarthritis (OA). We measured cell toxicities of cartilage-targeted low-generation dendrimer-linked nitroxide MR contrast agents and gadopentetate dimeglumine (Gd-DTPA) on cultured chondrocytes.

DESIGN

A long-term Swarm rat chondrosarcoma chondrocyte-like cell line was exposed for 48-h to different salts (citrate, maleate, tartrate) and concentrations of generation one or two diaminobutyl-linked nitroxides (DAB4-DLN or DAB8-DLN), Gd-DTPA, or staurosporine (positive control). Impact on microscopic cell appearance, MTT spectrophotometric assays of metabolic activity, and quantitative PicoGreen assays of DNA content (cell proliferation) were measured and compared to untreated cultures.

RESULTS

Chondrocyte cultures treated with up to 7.5 mM Gd-DTPA for 48-h had no statistical differences in DNA content or MTT reaction compared to untreated cultures. At all doses, DAB4-DLN citrate treated cultures had results similar to untreated and Gd-DTPA-treated cultures. At doses >1 mM, DAB4-DLN citrate treated cultures showed statistically greater DNA and MTT reaction than maleate and tartrate DAB4-DLN salts. Cultures exposed to 5 mM or 7.5 mM DAB8-DLN citrate exhibited rounded cells, poor cell proliferation, and barely detectable MTT reaction. Treatment with 0.1 μM staurosporine caused chondrocyte death.

CONCLUSION

Long-term exposure, greater than clinically expected, to either DAB4-DLN citrate or Gd-DTPA had no detectable toxicity with results equivalent to untreated cultures. DAB4-DLN citrate was more biocompatible than either the maleate or tartrate salts. Cells exposed for 48-h to 5 mM or 7.5 mM DAB8-DLN salts demonstrated significant cell toxicity. Further evaluation of DAB8-DLN with clinically appropriate exposure times is required to determine the maximum useful concentration.

Cartilage repair surgery: outcome evaluation by using noninvasive cartilage biomarkers based on quantitative MRI techniques?

BACKGROUND

New quantitative magnetic resonance imaging (MRI) techniques are increasingly applied as outcome measures after cartilage repair.

OBJECTIVE

To review the current literature on the use of quantitative MRI biomarkers for evaluation of cartilage repair at the knee and ankle.

METHODS

Using PubMed literature research, studies on biochemical, quantitative MR imaging of cartilage repair were identified and reviewed.

RESULTS

Quantitative MR biomarkers detect early degeneration of articular cartilage, mainly represented by an increasing water content, collagen disruption, and proteoglycan loss. Recently, feasibility of biochemical MR imaging of cartilage repair tissue and surrounding cartilage was demonstrated. Ultrastructural properties of the tissue after different repair procedures resulted in differences in imaging characteristics. T2 mapping, T1rho mapping, delayed gadolinium-enhanced MRI of cartilage (dGEMRIC), and diffusion weighted imaging (DWI) are applicable on most clinical 1.5 T and 3 T MR scanners. Currently, a standard of reference is difficult to define and knowledge is limited concerning correlation of clinical and MR findings. The lack of histological correlations complicates the identification of the exact tissue composition.

CONCLUSIONS

A multimodal approach combining several quantitative MRI techniques in addition to morphological and clinical evaluation might be promising. Further investigations are required to demonstrate the potential for outcome evaluation after cartilage repair.

Effect of chondral defect size, shape, and location on MRI diagnostic performance in the porcine knee

The purpose of this study was to determine the sensitivity and positive predictive value of magnetic resonance imaging (MRI) in the identification of full-thickness articular cartilage defects in the porcine knee. Seventy-two full-thickness chondral defects (small or large; circular, oval, or triangular) were created in 12 porcine knees. The authors used 3.0-T MRI with 3-dimensional gradient echo water-selective/fluid (WATSf) sequences acquired in axial, coronal, and sagittal planes. Sensitivity and positive predictive value parameters were calculated for 2 readers. Magnetic resonance imaging was highly sensitive for detection of full-thickness defects in the knee (85%). The highest sensitivity was observed at the medial femoral condyle (93%), while the lowest was observed at the medial patella (71%). The sensitivities for detecting different shapes were unique to each shape, with oval lesions identified with greatest sensitivity (93%). Small lesions (86%) were detected at a similar sensitivity as large lesions (83%). The positive predictive values for accurate true-positive reads were low for all lesion shapes (18%-57%) and moderate for small (69%) and large (59%) sizes, with significant differences observed between the 2 readers. Magnetic resonance imaging has a high sensitivity in the detection of full-thickness articular cartilage defects in the porcine knee. Variability in defect shape and intra-articular location affects MRI sensitivity, while size does not. Magnetic resonance imaging was not effective in describing lesion shape or size. Further, there was subjectivity in reading defect shape and size between 2 radiologists.

Theranostic immunoliposomes for osteoarthritis

Although there have been substantial advancements in the treatment of inflammatory arthritis, treatments for osteoarthritis (OA) have lagged and currently are primarily palliative until joints become totally dysfunctional and prosthetic replacement is needed. One obstacle for developing a preventive therapy for OA is the lack of good tools for efficiently diagnosing the disease and monitoring its progression during the early stages when the effect of therapeutic drugs or biologics is most likely to be effective. We have developed near infrared immunoliposomes conjugated with type II collagen antibody for diagnosis and treatment of early OA. These immunoliposomes bind to damaged but not normal cartilage. Utilizing these reagents, we can quantitate exposure of type II collagen during cartilage degradation in individual joints in vivo in a guinea pig. Immunoliposomes could be used to determine the effectiveness of therapeutic interventions in small animals as well as vehicles for localized drug delivery to OA chondrocytes.

FROM THE CLINICAL EDITOR

This team of authors have developed near infrared immunoliposomes conjugated with type II collagen antibody for diagnosis and treatment of early OA, with promising results demonstrated in a guinea pig model.

Cartilage and meniscal T2 relaxation time as non-invasive biomarker for knee osteoarthritis and cartilage repair procedures

OBJECTIVE

The purpose of this work was to review the current literature on cartilage and meniscal T2 relaxation time.

METHODS

Electronic searches in PubMed were performed to identify relevant studies about T2 relaxation time measurements as non-invasive biomarker for knee osteoarthritis (OA) and cartilage repair procedures.

RESULTS

Initial osteoarthritic changes include proteoglycan loss, deterioration of the collagen network, and increased water content within the articular cartilage and menisci. T2 relaxation time measurements are affected by these pathophysiological processes. It was demonstrated that cartilage and meniscal T2 relaxation time values were significantly increased in subjects with compared to those without radiographic OA and focal knee lesions, respectively. Subjects with OA risk factors such as overweight/obesity showed significantly greater cartilage T2 values than normal controls. Elevated cartilage and meniscal T2 relaxation times were found in subjects with vs without knee pain. Increased cartilage T2 at baseline predicted morphologic degeneration in the cartilage, meniscus, and bone marrow over 3 years. Furthermore, cartilage repair tissue could be non-invasively assessed by using T2 mapping. Reproducibility errors for T2 measurements were reported to be smaller than the T2 differences in healthy and diseased cartilage indicating that T2 relaxation time may be a reliable discriminatory biomarker.

CONCLUSIONS

Cartilage and meniscal T2 mapping may be suitable as non-invasive biomarker to diagnose early stages of knee OA and to monitor therapy of OA.

Methodologic quality of knee articular cartilage studies

PURPOSE

(1) To evaluate the quality of knee articular cartilage surgery literature using established methodologic quality instruments, and (2) to assess whether study quality has improved with time.

METHODS

A systematic review was performed using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Studies of autologous chondrocyte implantation (ACI), osteochondral autograft and allograft transplant, and microfracture were analyzed. Study methodologic quality was assessed by the level of evidence and 9 different methodologic quality questionnaires. Comparisons were made between different surgical technique groups by use of Student's t tests. Assessment of study quality improvement with time was performed by comparison of the Coleman Methodology Score (CMS) from the included studies (2004 to present) and CMS from a prior study assessing quality of articular cartilage studies (1985 to 2004). Furthermore, assessment of study quality improvement with time was performed over the period of the included studies (2004 to present).

RESULTS

We included 194 studies (11,787 subjects). Most evidence was Level IV (76%) and nonrandomized (91%). ACI was the most commonly reported technique (62% of studies). Only 34% of studies denied the presence of a financial conflict of interest. The mean subject age was 33.5 ± 8.2 years, and the mean length of follow-up was 3.7 ± 2.3 years. By use of study quality questionnaires, the methodologic quality of articular cartilage studies was poor. However, study quality (after 2004) was significantly improved versus that reported from a prior study (before 2004) using the CMS (P < .01). The mean level of evidence, CMS, CONSORT (Consolidated Standards of Reporting Trials) score, and Jadad score showed no significant improvement over the period of the included studies (P > .05). The quality of randomized controlled trials (RCTs) was significantly higher than that of non-RCTs (P < .05). The most common study weaknesses included blinding, subject selection process, study type, sample size calculation, and outcome measures and assessment.

CONCLUSIONS

The methodologic quality of knee articular cartilage surgery studies was poor overall and also for individual techniques (ACI, osteochondral autograft transplant, osteochondral allograft transplant, and microfracture). However, the overall quality of the investigations in this review (after June 2004) has significantly improved in comparison to those published before 2004. The quality of RCTs was significantly higher than that of non-RCTs. Level of evidence, CMS, CONSORT score, and Jadad score did not significantly improve with later publication date within the period of the studies analyzed. Methodologic quality deficiencies identified in this investigation may be used to guide future articular cartilage studies' design, conduct, and reporting.

LEVEL OF EVIDENCE

Level IV, systematic review of studies with Levels of Evidence I-IV.

Application of advanced magnetic resonance imaging techniques in evaluation of the lower extremity

This article reviews current magnetic resonance imaging (MR imaging) techniques for imaging the lower extremity, focusing on imaging of the knee, ankle, and hip joints. Recent advancements in MR imaging include imaging at 7 T, using multiple receiver channels, T2* imaging, and metal suppression techniques, allowing more detailed visualization of complex anatomy, evaluation of morphologic changes within articular cartilage, and imaging around orthopedic hardware.

Non-terminal animal model of post-traumatic osteoarthritis induced by acute joint injury

OBJECTIVE

Develop a non-terminal animal model of acute joint injury that demonstrates clinical and morphological evidence of early post-traumatic osteoarthritis (PTOA).

METHODS

An osteochondral (OC) fragment was created arthroscopically in one metacarpophalangeal (MCP) joint of 11 horses and the contralateral joint was sham operated. Eleven additional horses served as unoperated controls. Every 2 weeks, force plate analysis, flexion response, joint circumference, and synovial effusion scores were recorded. At weeks 0 and 16, radiographs (all horses) and arthroscopic videos (OC injured and sham joints) were graded. At week 16, synovium and cartilage biopsies were taken arthroscopically from OC injured and sham joints for histologic evaluation and the OC fragment was removed.

RESULTS

OC fragments were successfully created and horses were free of clinical lameness after fragment removal. Forelimb gait asymmetry was observed at week 2 (P = 0.0012), while joint circumference (P < 0.0001) and effusion scores (P < 0.0001) were increased in injured limbs compared to baseline from weeks 2 to 16. Positive flexion response of injured limbs was noted at multiple time points. Capsular enthesophytes were seen radiographically in injured limbs. Articular cartilage damage was demonstrated arthroscopically as mild wear-lines and histologically as superficial zone chondrocyte death accompanied by mild proliferation. Synovial hyperemia and fibrosis were present at the site of OC injury.

CONCLUSION

Acute OC injury to the MCP joint resulted in clinical, imaging, and histologic changes in cartilage and synovium characteristic of early PTOA. This model will be useful for defining biomarkers of early osteoarthritis and for monitoring response to therapy and surgery.

MRI characteristics of cysts and "cyst-like" lesions in and around the knee: what the radiologist needs to know

Objectives and Methods

A variety of benign cystic or “cyst-like” lesions may be encountered during a routine magnetic resonance imaging (MRI) of the knee. These lesions comprise a diverse group of entities from benign cysts to complications of underlying diseases. In addition, normal anatomic bursae and recesses may be misdiagnosed as an intra-articular cystic lesion when they are distended. However, the majority of the aforementioned lesions have characteristic MR appearances that allow a confident diagnosis, thus obviating the need for additional imaging or interventional procedures.

Results

This article includes a comprehensive pictorial essay of the characteristic MRI features of common and uncommon benign cysts and “cyst-like” lesions in and around the knee joint.

Discussion

For accurate assessment of the “cystic structure”, a radiologist should be able to identify typical MRI patterns that contribute in establishing the correct diagnosis and thus guiding specific therapy and avoiding unwarranted interventional procedures such as biopsy or arthroscopy.

Teaching points

• Cystic lesions are common in knee MRI and the commonest, the Baker’s cyst, has an incidence of 38 %.

• Synovial cysts, meniscal cysts, normal knee bursae and recesses have characteristic MR appearances.

• Miscellaneous “cyst-like” lesions may require a more dedicated MR protocol for a correct diagnosis.

Preoperative MRI underestimates articular cartilage defect size compared with findings at arthroscopic knee surgery

BACKGROUND

Magnetic resonance imaging (MRI) is widely used as a preoperative tool to estimate the size of articular cartilage defects to optimize treatment selection. However, the reliability of MRI sizing of cartilage defects is not well understood. Hypothesis/

PURPOSE

The purpose of this investigation was to compare the size of knee articular cartilage defects on MRI to arthroscopic visualization after debridement. It was hypothesized that MRI sizing would produce measurements that were no different than those made during arthroscopic knee surgery.

STUDY DESIGN

Cohort study (diagnosis); Level of evidence, 2.

METHODS

Seventy-seven patients (age [mean ± SD], 38 ± 10.7 years) who met inclusion criteria underwent preoperative knee MRI of at least 1.5 T within 1 year of arthroscopic knee surgery for a high-grade cartilage defect. Postdebridement defect sizes were obtained from intraoperative surgery notes and compared with retrospective MRI estimates.

RESULTS

Ninety-two total cartilage defects were analyzed with an average of 1.2 high-grade defects per knee and average postdebridement defect area of 2.99 cm(2) per lesion (95% CI, 1.63-2.26 cm(2)). Preoperative MRI analysis estimated a lesion area that was an average of 1.04 cm(2) smaller (95% CI, 0.70-1.39 cm(2); P < .0001). In 74% of the lesions analyzed, defect size was larger on arthroscopic visualization than was estimated by MRI sizing. On average, MRI underestimated the defect area by 70% compared with arthroscopic visualization.

CONCLUSION

Magnetic resonance imaging underestimates the size of articular cartilage defects compared with final postdebridement size as measured during arthroscopic knee surgery. Thus, before arthroscopic surgery, orthopaedic surgeons should consider treatment strategies that are appropriate for a larger defect than predicted by preoperative MRI.

Osteoarthritis: a review of strengths and weaknesses of different imaging options

Slowing of radiographic joint space narrowing represents the only recommended imaging-based outcome measure to assess structural disease progression in osteoarthritis (OA) clinical trials. There are no effective disease-modifying OA drugs. The ability of magnetic resonance (MR) to image structures within the knee and to visualize cartilage morphology and composition gives MR imaging a critical role in understanding the natural history of the disease and in the search for therapies. In this article, the roles and limitations of conventional radiography and MR imaging, focusing on knee OA, and the use of other modalities in clinical practice and OA research are described.