Correlation between the Arthroscopic and MRI Findings in Patients Active in Sports with Clinically Suspected Acetabular Labral Tear - Prospective Study Results
Abstract
PURPOSE OF THE STUDY The aim of our prospective study was to establish correlation between the arthro-MRI and arthroscopic findings in patients active in sports clinically suspected of having a hip labral tear. We sought to identify the benefits of arthro-MRI for clinical practice prior to the considered hip arthroscopy.
MATERIAL AND METHODS The prospective study included 34 patients with hip pain and clinically suspected of having a labral tear. The group consisted of 19 men (55.9%) and 15 women (44.1%) of the mean age of 32.2 years (18 to 50 years).
All our patients were active athletes, either professional (12; 35.3%) or recreational (22; 64.7%). Included in the study were exclusively patients aged 18-50 with hip pain lasting longer than 3 months, not responding to conservative therapy, with normal X-ray findings, or FAI syndrome.
The patients underwent arthro-MRI and subsequently also hip arthroscopy, namely within no more than 3 months. The description of findings was translated into simple numerical scales, assessed using Spearman s rank correlation coefficient.
We also examined the specificity, sensitivity, positive and negative predictive values and the odds ratio at various cut-off values of the scale for arthro-MRI. The statistical significance was set at the alpha level = 5%.
The quality of tests was also evaluated using the ROC curve. RESULTS In hips assessed by the radiologist as reliably with no pathological finding (MRI type 1) the negative predictive value achieves 83.3% and the positive predictive value of 85.7%, with 96.0% sensitivity and 55.6% specificity.
When evaluating the reliably negative finding together with the likely negative finding (MRI type 2) as against the pathological and certainly pathological findings (MRI type 3 and 4), the negative predictive value is 57.1%, the positive predictive value 95.0%, the specificity 88.9% and the sensitivity 76.0%. DISCUSSION In our study, the values achieved were comparable to or better than those reported by foreign authors in their studies establishing correlations between the arthro-MRI and intraoperative findings.
Significantly higher was primarily the negative predictive value. CONCLUSIONS The results of our study indicate that the arthro-MRI performed as a part of preoperative examination prior to hip arthroscopy has adequate specificity and sensitivity in patients with suspected labral tear.
Based on this examination a decision can be made as to whether to perform arthroscopy. Our study also shows that it is essential to have a highly competent radiologist to evaluate the arthro-MRI results.