PURPOSE: To assess the MR findings of chondromalacia patella and correlate the grade and associated lesionswith the arthroscopic findings. MATERIALS AND METHODS: Twenty-five patients with pain in the anterior part of theknee underwent fat-suppressed axial and coronal T2-weighted, dual echo sagittal proton density-weighted andT2-weighted, and in some cases, T1-weighted and T2*-weighted imaging, using a 10-cm field of view, and a 5-inchgeneral purpose coil. We retrospectively assessed these find dings, and the locations, grades and associatedlesions, and correlated these with the arthroscopic findings. RESULTS: Fourteen cases(56%) of chondromalaciapatellae were arthroscopic grade I / II, nine(36%) were grade III, and two(8%) were grade IV; associated lesionswere medial synovial plicae (16 cases, 64%), meniscus tear (10 ; 40%), cruciate ligament injury (two ; 8%),complete or incomplete discoid meniscus (four ; 16%), bipartite patella (one ; 4%) and Osgood-Schlatter disease(one ; 4%). In 24 cases, MR imaging indicated chondromalacia patella ; 17 cases were grade I / II, five were gradeIII, and two were grade IV. the location of chondromalacia patella was the medial facet (five cases ; 20%), lateralfacet (three ; 12%), junction of the medial and odd facet (13 ; 52%), and diffuse involvement (four ; 16%). Thesensivity and specificity of MR imaging were 72% and 96% respectively. CONCLUSION: We evaluated the exact locationand grade of chondromalacia patella and associated lesions, as seen on MR images. These and the arthroscopicfindings showed close correlation, and in cases involving this condition, MRI is thus a useful indicator of anappropriate surgical method and plan.