PURPOSE: The purpose of this study was to evaluate the extent of necrosis of the femoral head inLegg-Clave-Perthes (LCP) disease. This involved the use of MRI, followed by volume measurement and the use of theCatterall classification system; the difference between the grade obtained using each of these approaches was thendetermined. MATERIALS AND METHODS: We retrospectively reviewed 28 hip-joint MR images on which a diagnosis of LCPdisease had been based. According to the necrotic portion of femoral head, and on the basis of the catterallclassification, LCP was graded 1 to 4, as follows: Grade I=0~ <25%; II=2 5~ <50%; III=50~ <75; IV=75-100%. Using atransparent paper on which 1mm squares had been drawn, the necrotic area of each MR image was measured; thefollowing equation was then used to calculate the volume of the necrotic portion: volume=necrotic area x slicethickness. On the basis of this measurement, each femoral head was graded and the results were compared with thoseobtained using the Catterall classification. RESULTS: In 24 joints(85.7%), grades according to the Catterallclassification and MRI volume measurement were not the same. As compared with the volume measurement method, useof the Catterall classification led to grade overestimation in 18 joints (64.3%) and underestimation in six(21.4%). CONCLUSION: The grade according to the Catterall classification was different from that obtained usingthe volume measurement method. This study thus indicates the need for a new system of classifying LCP diseasebased on the volume measurement method and using MR imaging.