Clinical decision making in the acutely injured knee based on repeat clinical examination and MRI

A deep learning approach for anterior cruciate ligament rupture localization on knee MR images

Purpose: To develop and evaluate a deep learning-based method to localize and classify anterior cruciate ligament (ACL) ruptures on knee MR images by using arthroscopy as the reference standard.

Methods: We proposed a fully automated ACL rupture localization system to localize and classify ACL ruptures. The classification of ACL ruptures was based on the projection coordinates of the ACL rupture point on the line connecting the center coordinates of the femoral and tibial footprints. The line was divided into three equal parts and the position of the projection coordinates indicated the classification of the ACL ruptures (femoral side, middle and tibial side). In total, 85 patients (mean age: 27; male: 56) who underwent ACL reconstruction surgery under arthroscopy were included. Three clinical readers evaluated the datasets separately and their diagnostic performances were compared with those of the model. The performance metrics included the accuracy, error rate, sensitivity, specificity, precision, and F1-score. A one-way ANOVA was used to evaluate the performance of the convolutional neural networks (CNNs) and clinical readers. Intraclass correlation coefficients (ICC) were used to assess interobserver agreement between the clinical readers.

Results: The accuracy of ACL localization was 3.77 ± 2.74 and 4.68 ± 3.92 (mm) for three-dimensional (3D) and two-dimensional (2D) CNNs, respectively. There was no significant difference in the ACL rupture location performance between the 3D and 2D CNNs or among the clinical readers (Accuracy, p < 0.01). The 3D CNNs performed best among the five evaluators in classifying the femoral side (sensitivity of 0.86 and specificity of 0.79), middle side (sensitivity of 0.71 and specificity of 0.84) and tibial side ACL rupture (sensitivity of 0.71 and specificity of 0.99), and the overall accuracy for sides classifying of ACL rupture achieved 0.79.

Conclusion: The proposed deep learning-based model achieved high diagnostic performances in locating and classifying ACL fractures on knee MR images.

Agreement Between Clinical Examination and Magnetic Resonance Imaging in Acute Knee Trauma With Hemarthrosis

OBJECTIVE

Hemarthrosis after knee trauma often indicates serious joint injury. Few studies have evaluated agreement between clinical examination and findings from magnetic resonance imaging (MRI). We aimed to describe the agreement between acute clinical examination and subacute MRI findings after acute knee trauma with hemarthrosis and the importance of the subspecialty of the examiner.

DESIGN

Longitudinal cohort study. Agreement with MRI findings was evaluated by logistic regression.

SETTING

Helsingborg hospital.

PATIENTS

Thousand one hundred forty-five consecutive patients with hemarthrosis after knee trauma.

INTERVENTIONS

Clinical examination and MRI.

MAIN OUTCOME MEASURES

agreement between clinical examination and findings from MRI. We considered the radiologist's report as the gold standard.

RESULTS

Median time (25th, 75th percentile) from injury to clinical examination was 2 (1, 7) days, and from injury to imaging was 8 (5, 15) days. The overall sensitivity and specificity of clinical examination versus MRI for major ligament injury or lateral patella dislocation (LPD) were 70% [95% confidence interval 67-73) and 66% (61-72), respectively. Orthopedic subspecialist knee had the highest agreement with anterior cruciate ligament rupture (adjusted odds ratios were 1.7 (95% confidence interval 1.2-2.3), 1.9 (1.2-3.0) and 5.9 (3.7-9.5) for orthopedic trainees, orthopedic subspecialists other, and orthopedic subspecialist knee, respectively]. For other ligament injuries and LPD, we did not find statistically significant differences.

CONCLUSIONS

Clinical diagnosis after acute knee injury is relatively unreliable versus MRI findings even when performed by orthopedic specialists. However, the agreement is improved when the examination is performed by an orthopedic knee subspecialist.

Prevalence of knee osteoarthritis features on magnetic resonance imaging in asymptomatic uninjured adults: a systematic review and meta-analysis

Background

Knee MRI is increasingly used to inform clinical management. Features associated with osteoarthritis are often present in asymptomatic uninjured knees; however, the estimated prevalence varies substantially between studies. We performed a systematic review with meta-analysis to provide summary estimates of the prevalence of MRI features of osteoarthritis in asymptomatic uninjured knees.

Methods

We searched six electronic databases for studies reporting MRI osteoarthritis feature prevalence (ie, cartilage defects, meniscal tears, bone marrow lesions and osteophytes) in asymptomatic uninjured knees. Summary estimates were calculated using random-effects meta-analysis (and stratified by mean age: <40 vs ≥40 years). Meta-regression explored heterogeneity.

Results

We included 63 studies (5397 knees of 4751 adults). The overall pooled prevalence of cartilage defects was 24% (95% CI 15% to 34%) and meniscal tears was 10% (7% to 13%), with significantly higher prevalence with age: cartilage defect <40 years 11% (6%to 17%) and ≥40 years 43% (29% to 57%); meniscal tear <40 years 4% (2% to 7%) and ≥40 years 19% (13% to 26%). The overall pooled estimate of bone marrow lesions and osteophytes was 18% (12% to 24%) and 25% (14% to 38%), respectively, with prevalence of osteophytes (but not bone marrow lesions) increasing with age. Significant associations were found between prevalence estimates and MRI sequences used, physical activity, radiographic osteoarthritis and risk of bias.

Conclusions

Summary estimates of MRI osteoarthritis feature prevalence among asymptomatic uninjured knees were 4%–14% in adults aged <40 years to 19%–43% in adults ≥40 years. These imaging findings should be interpreted in the context of clinical presentations and considered in clinical decision-making.

How sensitive and specific is 1.5 Tesla MRI for diagnosing injuries in patients with knee dislocation?

PURPOSE

The purpose of this study was to assess the sensitivity and specificity of 1.5 T magnetic resonance imaging (MRI) in diagnosing and identifying the specific injury pattern in patients with knee dislocation. The hypothesis was that the sensitivity and specificity are low in patients with posterolateral corner injury and/or PCL tear.

METHODS

A retrospective study was performed on 38 patients (m:f = 29:9, mean age ± SD 34.3 ± 14.0) with traumatic knee dislocation, who underwent 1.5 T MRI prior to surgery. MRI scans were analysed by a musculoskeletal radiologist, and the presence and type of tears to ligaments, tendons and meniscus or bone were recorded. Comparison was made with the intraoperative findings from the surgical records using the same reporting scheme. The agreement between MRI and surgical findings was assessed using kappa statistics, and the sensitivity and specificity were calculated.

RESULTS

In patients with knee dislocation, MRI was found to have low sensitivity (25-38 %) but high specificity (94-97 %) for diagnosing injury to the posterolateral corner. There was high sensitivity in the diagnosis of tears in the cruciate and collateral ligaments (97-100 %); the specificity, however, was lower (50-67 %). The diagnosis of meniscal injury showed low sensitivity (36-56 %) and moderate specificity (69-83 %).

CONCLUSIONS

MRI is a sensitive measure of cruciate and collateral ligament injury in acute knee dislocation; however, it does not reliably diagnose injury to the posterolateral corner or meniscus, and therefore, a higher index of suspicion is required during arthroscopy to prevent misdiagnosis which could affect long-term clinical outcome.

LEVEL OF EVIDENCE

Diagnostic study, Level II.

Epidemiology of intra- and peri-articular structural injuries in traumatic knee joint hemarthrosis - data from 1145 consecutive knees with subacute MRI

OBJECTIVE

In patients with traumatic knee distorsion and hemarthrosis, to investigate the panorama of structural injury, as visualized on magnetic resonance imaging (MRI).

DESIGN

Cohort study of 1145 consecutive patients with traumatic knee distorsion who underwent MRI within median 8 days after injury. We present structural injury as visualized on MRI in relation to age, sex and activity at injury. Population based gender specific annual incidences of common structural injuries were calculated.

RESULTS

The majority of injuries (72%) occurred during sports. Overall, anterior cruciate ligament (ACL) rupture was the most common structural injury (52%), followed by meniscal tear (41%) and lateral patella dislocation (LPD, 17%). Only 12% of ACL tears were isolated with meniscal tear being the most common associated injury (55%). The annual incidence of ACL injury was 77 (70-85, 95% CI) per 100,000 inhabitants with significant differences between men (91, 80-103) and women (63, 53-73). In those aged 16 years and younger, LPD was the most frequent structural injury, both in boys (39%) and girls (43%). In this age group, the annual incidence of LPD was 88 (68-113) and higher in boys (113, 81-154) than in girls (62, 39-95).

CONCLUSIONS

ACL injury occurs in one out of two knees with traumatic hemarthrosis but only 12% are without concomitant structural injury. The overall rate of traumatic knee hemarthrosis and ACL injury is higher in men. In those aged 10-19 years, ACL rupture is more common among girls than boys whereas in those 16 years and younger, LPD is more common among boys than girls.

Presentation, examination, investigation and early treatment of acute knee injuries

Acute knee injuries are common presentations that frequently occur during sporting activities and can result in permanent disability. However, they often result in a single visit, seldom receive surgical intervention and, in the main, are cared for entirely by General Practitioners or Emergency Department physicians. Thorough clinical assessment and appropriate early treatment at the primary presentation is vital in order to offer the best chance of maximal functional recovery and to minimize long-term disability. This requires a sound knowledge of basic knee anatomy, careful assessment of the mechanism of injury, detection and consideration of physical ndings and the results of adjunctive investigations. This article aims to give a comprehensive review of the examination, investigation and early treatment of acute knee injuries at the primary presentation. In addition, comments on important epidemiological and aetiological factors and a brief description of basic knee anatomy are also provided.

Diagnostic needle arthroscopy and the economics of improved diagnostic accuracy: a cost analysis

BACKGROUND

Hundreds of thousands of surgical arthroscopy procedures are performed annually in the United States (US) based on MRI findings. There are situations where these MRI findings are equivocal or indeterminate and because of this clinicians commonly perform the arthroscopy in order not to miss pathology. Recently, a less invasive needle arthroscopy system has been introduced that is commonly performed in the physician office setting and that may help improve the accuracy of diagnostic findings. This in turn may prevent unnecessary follow-on arthroscopy procedures from being performed.

OBJECTIVE

The purpose of this analysis is to determine whether the in-office diagnostic needle arthroscopy system can provide cost savings by reducing unnecessary follow on arthroscopy procedures.

METHODS

Data obtained from a recent trial and from a systematic review were used in comparing the accuracy of MRI and VisionScope needle arthroscopy (VSI) with standard arthroscopy (gold standard). The resultant false positive and false negative findings were then used to evaluate the costs of follow-on procedures. These differences were then modeled for the US patient population diagnosed and treated for meniscal knee pathology (most common disorder) to determine if a technology such as VSI could save the US healthcare system money. Data on surgical arthroscopy procedures in the US for meniscal knee pathology were used (calendar year [CY] 2010). The costs of performing diagnostic and surgical arthroscopy procedures (using CY 2013 Medicare reimbursement amounts), costs associated with false negative findings, and the costs for treating associated complications arising from diagnostic and therapeutic arthroscopy procedures were assessed.

RESULTS

In patients presenting with medial meniscal pathology (International Classification of Diseases, 9th edition, Clinical Modification [ICD9CM] diagnosis 836.0), VSI in place of MRI (standard of care) resulted in a net cost savings to the US system of US$115-US$177 million (CY 2013) (use of systematic review and study data, respectively). In patients presenting with lateral meniscus pathology (ICD9CM 836.1), VSI in place of MRI cost the healthcare system an additional US$14-US$97 million (CY 2013). Overall aggregate savings for meniscal (lateral plus medial) pathology were identified in representative care models along with more appropriate care as fewer patients were exposed to higher risk surgical procedures.

CONCLUSIONS

Since in-office arthroscopy is significantly more accurate, patients can be treated more appropriately and the US healthcare system can save money, most especially in medial meniscal pathology.

Efficacy of repeat magnetic resonance imaging of the knee

A patient often initially presents to an orthopedic surgeon with the magnetic resonance image (MRI) ordered by his or her primary care physician in hand. Often, a significant period of time elapses after injury before the patient is assessed by the orthopedic surgeon; therefore, the initial MRI may be considered outdated because of a new injury or a change in symptoms or because the orthopedist may prefer a new study, a stronger magnet, or a special imaging protocol. However, the decision to repeat a knee MRI is presently an arbitrary one because no clinical guidelines exist to justify this practice.All repeat knee MRIs performed at our academic institution in the past 9 years were retrospectively examined. Inclusion criterion was repeat MRI of the same knee with no surgical intervention. The formal radiology reports were grouped into 3 categories: change, no change, and unclear. Knee pathology was further grouped into 6 categories indicating what specific structures were pathological or injured. Logistic regression analysis was used to test the association of time vs category or condition change between MRIs. Of 3501 knee MRI studies, 88 patients had a total of 101 repeat MRIs. The average number of days between repeats for those with category or condition change was 612 vs 504 for those with no change. Age, sex, and time between MRIs were not significantly associated with a category or condition change. Repeat knee MRI prior to surgical intervention is becoming more prevalent and may have clinical merit. A further prospective study is warranted.

Magnetic resonance imaging of the knee in Norway 2002-2004 (national survey): rapid increase, older patients, large geographic differences

Background

Magnetic resonance imaging (MRI) of the knee is the second most common MRI examination in Norway after head/brain MRI. Little has been published internationally on trends in the use of knee MRI after 1999. This study aimed to describe levels and trends in ambulant knee MRI utilisation in Norway 2002–2004 in relation to type of radiology service, geographic regions, number of MRI-scanners, patient age and gender, and type of referring health care provider.

Methods

We analysed administrative data on all claims for reimbursement of ambulant knee MRI performed in Norway in 2002, 2003 and 2004 and noted nominal reimbursement. We also recorded the referring health care provider from clinical requests of ambulant knee MRI done consecutively during two months in 2004 at one private institute and three hospitals. Number of MRI-scanners was given by manufacturers and radiology services.

Results

In Norway, the rate of knee MRI claims for 2004 was 15.6 per 1000 persons. This rate was 74% higher in East than in North region (18.4 vs. 10.6), slightly higher for men than women (16.4 vs. 14.7) and highest for ages 50–59 years (29.0) and 60–69 years (21.2). Most claims (76% for 2004) came from private radiology services. In 2004, the referring health care provider was a general practitioner in 63% of claims (unspecified in 24%) and in 83.5% (394/472) of clinical requests. From 2002 to 2004, the rate of knee MRI claims increased 64%. In the age group 50 years or above the increase was 86%. Rate of MRI-scanners increased 43% to 21 scanners per million persons in 2004. Reimbursement for knee MRI claims (nominal value) increased 80% to 70 million Norwegian kroner in 2004.

Conclusion

Ambulant knee MRI utilisation in Norway increases rapidly especially for patients over 50, and shows large geographic differences. Evaluation of clinical outcomes of this activity is needed together with clinical guidelines for use of knee MRI.

Clinical Course of Knees with Asymptomatic Meniscal Abnormalities: Findings at 2-year Follow-up after MR Imaging–based Diagnosis

PURPOSE

To prospectively evaluate the clinical course of asymptomatic meniscal lesions diagnosed by using magnetic resonance (MR) imaging.

MATERIALS AND METHODS

Institutional review board approval and informed consent were obtained. The clinical courses of meniscal lesions in 84 asymptomatic knees (in 48 men and 36 women; mean age, 43.6 years; age range, 18-73 years) were assessed. Thirty-one asymptomatic meniscal lesions were depicted among the 84 knees at MR imaging. The follow-up period was at least 2 years (mean, 29.8 months; range, 24-36 months). Knee pain, stiffness, and function during daily and sports activities were assessed by using a visual analogue scale (VAS), on which a score of 0 indicated no pain or complaints and a score of 100 indicated maximal pain and/or complaints. The chi2 test was used for statistical analysis.

RESULTS

At follow-up, 12 (39%) of the 31 patients with and 10 (19%) of the 53 patients without meniscal lesions reported having knee pain (P = .046). Nine (29%) patients with and five (9%) patients without meniscal lesions reported having knee stiffness (P = .02). Ten (32%) patients with and five (9%) patients without meniscal lesions reported having impaired function during daily activities (P = .008). Thirteen (42%) patients with and 15 (28%) patients without meniscal lesions reported having impaired function during sports activities (P = .20). Mean VAS scores ranged from 22 to 30; mean VAS scores for pain, stiffness, impaired daily function, and impaired sports function were 26, 22, 27, and 30, respectively. None of the initially asymptomatic knees was treated with surgery during the follow-up period.

CONCLUSION

The clinical course of knees with initially asymptomatic meniscal lesions, compared with the clinical course of knees without meniscal lesions, is characterized by an increased frequency of complaints. However, severities of pain and knee function impairment remain low.