[Sensitivity of magnetic resonance tomography in detecting meniscus tears]

Verification bias: an under-recognized source of error in assessing the efficacy of MRI of the meniscii

RATIONALE AND OBJECTIVES

The sensitivity and specificity of magnetic resonance imaging (MRI) for diagnosis of meniscal tears has been studied extensively, with tears usually verified by surgery. However, surgically unverified cases are often not considered in these studies, leading to verification bias, which can falsely increase the sensitivity and decrease the specificity estimates. Our study suggests that such bias may be very common in the meniscal MRI literature, and illustrates techniques to detect and correct for such bias.

MATERIALS AND METHODS

PubMed was searched for articles estimating sensitivity and specificity of MRI for meniscal tears. These were assessed for verification bias, deemed potentially present if a study included any patients whose MRI findings were not surgically verified. Retrospective global sensitivity analysis (GSA) was performed when possible.

RESULTS

Thirty-nine of the 314 studies retrieved from PubMed specifically dealt with meniscal tears. All 39 included unverified patients, and hence, potential verification bias. Only seven articles included sufficient information to perform GSA. Of these, one showed definite verification bias, two showed no bias, and four others showed bias within certain ranges of disease prevalence. Only 9 of 39 acknowledged the possibility of verification bias.

CONCLUSION

Verification bias is underrecognized and potentially common in published estimates of the sensitivity and specificity of MRI for the diagnosis of meniscal tears. When possible, it should be avoided by proper study design. If unavoidable, it should be acknowledged. Investigators should tabulate unverified as well as verified data. Finally, verification bias should be estimated; if present, corrected estimates of sensitivity and specificity should be used. Our online web-based calculator makes this process relatively easy.

Validation of the Thessaly test for detecting meniscal tears in anterior cruciate deficient knees

Meniscal injuries are frequently associated with anterior cruciate ligament (ACL) tears. Clinical tests that are useful for detecting meniscal tears may not be valid in this setting. The Thessaly test, a newly described dynamic clinical examination, has been shown to have a very high diagnostic accuracy for detecting meniscal tears. This study evaluates the accuracy of the Thessaly test in patients with combined ACL and meniscus injuries. We examined eighty patients with ACL deficiency for meniscal injuries using the Thessaly test (at 20 degrees of knee flexion), the lateral and medial joint line tenderness tests, and the McMurray test. Examiners were blinded to the MRI report on the conditions of the menisci. All patients underwent arthroscopic ACL reconstruction during which the menisci were evaluated by direct vision and probing. During the Thessaly test, six patients developed severe pain and could not complete the test. Sensitivity, specificity, likelihood ratios and predictive values were calculated for all tests. The Thessaly test had a sensitivity of 79%, specificity of 40%, positive predictive value of 56%, negative predictive value of 66%, positive likelihood ratio of 1.33, negative likelihood ratio of 0.51%, and overall accuracy of 60%. We concluded that the Thesally test has a low specificity in patients with combined ACL and meniscal injuries and can not be recommended as a diagnostic test in this setting.

[Diagnostic approaches to acute knee injury in childhood and adolescence. Yesterday and today]

INTRODUCTION

Clinical examination of acute knee injury in childhood is often difficult and therefore magnetic resonance imaging (MRI) serves as an additional diagnostic tool. The aim of the present study was to evaluate on the one hand the indications for diagnostic arthroscopy and on the other hand the indications for MRI.

METHODS

Of the children treated between 1990 and 1999, 87 (group 1) underwent arthroscopy after clinical examination. Between 2000 and 2006 (group 2) 83 patients were examined using MRI after clinical examination and 53 were subsequently submitted to arthroscopy.

RESULTS

In group 1 the clinical diagnosis was verified by arthroscopy in 79%. In group 2 the clinical and arthroscopic diagnoses were consistent in 60% of the patients. The MRI diagnosis was correctly recognized for patella dislocation in all cases, for ligament injuries in 83% and for meniscus injuries in 56%. Due to the application of MRI before arthroscopy the fraction of diagnostic arthroscopies could be reduced from 22% to 13%.

CONCLUSION

The number of diagnostic arthroscopies in childhood can be reduced by application of MRI.

The role of preoperative MRI in knee arthroscopy: a retrospective analysis of 2,000 patients

The aim of this study was to investigate, to what extent routine preoperative MRI scans could set the indications for knee arthroscopies and reduce the number of diagnostic arthroscopies. For this retrospective cohort study, 1,000 patients who had knee arthroscopies documented in 1994/1995 were compared with 1,000 patients that were treated in 2004/2005. The preoperative diagnoses that gave indications for knee arthroscopy were compared with the intraoperative findings. The congruence of preoperative diagnosis with the intraoperative findings was evaluated comparing both study populations. The number of patients who were referred to orthopaedic trauma surgeons with MRI increased from 24% to 56%. A high congruence of preoperative diagnosis and intraoperative findings was found in 49% in 1994/1995 and 55% in 2004/2005. However, regarding the most important outcome parameter, the number of diagnostic arthroscopies, no improvement was found (3% in both periods). The presented data suggests that MRI scans are not routinely necessary as an indication for knee arthroscopy, as clinical examination and plain radiograph are sufficient. However, MRI scans do allow a more detailed characterization of the expected findings and can therefore be helpful in therapy planning.

Real time spatial compound ultrasound in the evaluation of meniscal injuries: a comparison study with conventional ultrasound and MRI

The aim of this study was to compare real time spatial compound ultrasound (RTCS) and conventional ultrasound with MRI in the evaluation of meniscal injuries. Thirty five patients with clinical suspicion of meniscal injury underwent ultrasound of knee alternating between conventional and compound sonography followed by MRI of the knee on a 1.5 T MR system. The images were compared for their quality and ability to detect meniscal tear. Compound ultrasound significantly improved definition of soft tissue planes, reduced speckle and other noise, improved contrast resolution (p value<0.05) with improved confidence in detection of tears in doubtful cases. Interobserver variability was seen only for conventional ultrasound and the sensitivity, specificity, accuracy were 83.3%, 87.5% and 85.7% for the first observer and 80%, 87.5% and 84.3% for second observer. For compound ultrasound it was same for both the observers with sensitivity, specificity and accuracy being 90% each. Although compound ultrasound improves image quality in the evaluation of meniscal injuries the benefit with respect to detection of tears is only marginal. It is a valuable tool for screening the patients before MRI and we recommend that instead of conventional ultrasound, compound ultrasound should be routinely used for sonographic evaluation of meniscal injuries.

Diagnostics of degenerative meniscal tears at 3-Tesla MRI compared to arthroscopy as reference standard

INTRODUCTION

The goal of this study was to evaluate the diagnostic performance of 3-Tesla MRI for the assessment of degenerative meniscal tears in clinical practice.

MATERIALS AND METHODS

In patients with chronic knee pain, a negative history of acute trauma and a mean age of 52 years, 3-Tesla MRI were performed a few days prior to arthroscopy. In 86 menisci, diagnostic values of 3-Tesla MRI for the detection of degenerative tears were evaluated using arthroscopy as reference standard. The MRI classification, for meniscus diagnostics, described by Crues was used.

RESULTS

At arthroscopy, all tears identified (19 horizontal, 7 complex, 3 radial) were degenerative as confirmed by histological examination. MRI grade II lesions had a prevalence of 24% and a rate of 24% of missed tears, whereas grade I lesions were not associated with a torn meniscus at arthroscopy. For meniscal tears, (grade III) sensitivity and specificity of 3-Telsa MRI was 79 and 95% for both menisci, 86 and 100% for the medial meniscus, and 57 and 92% for the lateral meniscus. The best diagnostic performance was found for complex tears, horizontal tears showed relatively good results, poor results were documented for radial tears.

CONCLUSION

For the medial meniscus, where horizontal and complex tears were more prevalent, 3-Tesla MRI shows a higher accuracy than for the lateral meniscus. Particularly with regard to the medial meniscus, 3-Tesla MRI could be effectively used when a degenerative tear is suspected. Nevertheless, in regard to a remarkable number of false positive and false negative findings the diagnostic value of a 3-Tesla MRI investigation should not be overestimated.

Prediction of Soft-Tissue Injuries in Schatzker II Tibial Plateau Fractures based on Measurements of Plain Radiographs

BACKGROUND

Split-depression fractures of the lateral tibial plateau (Schatzker II) are associated with a significant risk of capsuloligamentous and meniscal injury. We hypothesized that the amount of fracture depression and widening on anteroposterior (AP) plain radiographs would correlate with the incidence of injury to these structures on magnetic resonance imaging (MRI).

METHODS

Sixty-two consecutive patients with Schatzker II tibial plateau fractures had a knee x-ray series and MRI preoperatively. AP plain radiographs were measured for lateral joint line depression and condylar widening, and MRIs were evaluated for injury to soft-tissue structures around the knee. For each structure, the threshold of depression and widening that led to the greatest disparity in soft-tissue injury was determined. Multiple logistic regressions were applied to calculate whether depression and/or widening above the thresholds were predictive for injury to individual soft-tissue structures.

RESULTS

When depression was greater than 6 mm and widening was greater than 5 mm, lateral meniscal injury occurred in 83% of fractures, compared with 50% of fractures with less displacement (p < 0.05). When either depression or widening was at least 8 mm, medial meniscal injury occurred more frequently (depression 53%, p < 0.05; widening 78%, p < 0.05; versus neither 15%). Lateral collateral ligament and posterior cruciate ligament tears were not seen with minimally displaced fractures (< 4 mm), but the incidence of injury approached 30% with increasing displacement.

CONCLUSIONS

Due to the limited availability of MRI in some centers, correlation of lateral condylar depression and widening, as measured on plain radiographs, to injury of various soft-tissue structures may be extremely helpful in planning open or arthroscopic treatment methods. Using these guidelines, Schatzker II fractures with depression or widening approaching 5 mm deserve heightened vigilance in diagnosing and treating these concomitant soft-tissue injuries.