During the period between March 1998. and April 2000, 125 knees for 123 patients suspected to have m~niscus or cruciate ligaments injury were studied by MRI followed by arthroscopy. all included in this study. All the patients had no history of previous surgery in the same knee; there were 94 males (76.4 per cent) and 29 females (23.6 per cent), and the mean age of the patients was 34 + 10.6 years, range (19-56). The MRI findings
were studied and compared with the arthroscopic fmdings, using 5 parameters accuracy, sensitivity, specificity, positive and negative predictive values. The over-all results of MRI in our series were accuracy rate 90 per cent, sensitivity 83.3 per cent, specificity 92.7 per cent, positive predictive value 82 per cent and negative predictive value 94 per cent. We think that MRI is a considerable advance and may well preserve patients from unnecessary operations; Arthroscopy for diagnostic purposes should be used only with a specific purpose. Modern MRI can and should replace "having a look".