The accuracy of magnetic resonance imaging scanning and its influence on management decisions in knee surgery.

PURPOSE

The purpose of our study was to evaluate the impact of preoperative magnetic resonance imaging (MRI) assessment of articular knee pathology on the clinical management of patients presenting with joint line pain.

METHODS

A preliminary study on 100 patients was performed to assess the accuracy of specific MRI sequences, using arthroscopy as a gold standard. Six hundred and eighteen consecutive patients with knee symptoms presenting to 2 specialist knee surgeons were then recruited. A clinical diagnosis of an arthroscopically treatable lesion was made in all cases. Clinical assessment data were correlated to subsequent MRI findings, recording any discrepancy and in particular whether or not MRI findings influenced management decisions.

RESULTS

In the preliminary study, MRI sequences had an overall sensitivity of 83.2% and a specificity of 94.3% for the detection of chondral lesions. However, when considering arthroscopic grade III and IV lesions, MRI sensitivity and specificity were improved to 84.5% and 97.1%. In the second phase of the study, 141 (22.8%) of the 618 patients presenting with knee symptoms had an altered clinical management subsequent to MRI. The presence of unexpected chondral lesions was found in 77 of these patients. Conversely, 22 patients with clinical symptoms suspicious for simple chondral degeneration had unstable meniscal tears.

CONCLUSIONS

We suggest that preoperative MRI scanning identifies a group of patients who have more advanced degenerative joint disease than the clinical assessment and the plain radiographs suggest. This would expedite definitive surgery in patients with advanced osteoarthritis on MRI scans.

LEVEL OF EVIDENCE

Level II, development of diagnostic criteria on basis of consecutive patients with universally applied gold standard.