Nilesh Vishwakarma1, Sarabjeet Kohli2, Abhishek Kini3, Aditya Daftary4, Anant Joshi5
OBJECTIVE
The purpose of this study was to evaluate the reliability of clinical diagnosis compared to MRI findings in ligamentous and meniscal injuries with respect to arthroscopic confirmation as a gold standard.
METHODS
485 patients with knee injuries were prospectively assessed by clinical evaluation and magnetic resonance imaging and correlated after therapeutic arthroscopy. The overall accuracy, clinically productive values of sensitivity and specificity was derived. The actual value of the test with respect to positive predictive and negative predictive value was also derived, taking arthroscopic findings as confirmatory. The overall partial and total agreement among the clinical, MRI and arthroscopy was documented.
RESULTS
The overall accuracy for clinical examination was 85, 92, 100 and 100 and accuracy for MRI was 90, 97, 97 and 97 for detecting medial meniscus, lateral meniscus, ACL and PCL tears respectively. Clinically lateral meniscus tears are difficult to diagnose clinically with negative predictive value (90) whereas ACL injuries do not need MRI for diagnosis as evident by a high negative predictive value (100) of clinical examination. Total agreement with the clinical findings confirmed by arthroscopy was 64.40% which was relatively high as compared to total agreement of MRI findings which was only 31.50%. We found similar total agreement versus total disagreement of both clinical and MRI to be only 2.74% indicating very high accuracy in clinical diagnosis of meniscal and ligamentous injuries combined.
CONCLUSION
The clinical evaluation alone is sufficient to diagnose meniscal and ACL/PCL pathologies and MRI should be considered only as a powerful negative diagnostic tool. The arthroscopy decision should not be heavily dependent on MRI for ligamentous injuries but reverse is true for meniscal lesions. MR evaluation functions as a powerful negative diagnostic tool to rule out doubtful and complex knee injuries.
