Poster Presentations/Osteoarthritis and Cartilage 19S1 (2011) S53–S236 S195 Fig. 1: Mid-portion patellar cartilage of mCT and CTa. Attenuation of cartilage regions is visualized in colour and representative for cartilage quality. Results: CTa acquired using the maximum dose correlated excellent with reference mCT values(R = 0.86; R 2 =0.73; p < 0.0001). The CTa scans acquired using the lower radiation doses correlated well with the mCT values (R = 0.75–0.74; R 2 =0.57–0.55; p < 0.0001) (Table 1). The visual agreement between the CTa and the reference mCT showed that CTa acquired using the maximum radiation dose agreed most with mCT in locally determining good and bad cartilage quality. The agreement decreased with the decrease in radiation dose. CTa acquired using 10% can not distinguish between small regions of good and bad cartilage quality (Figure 1). Conclusions: Despite a lower radiation dose and hence an increase in noise in the images, CTa acquired using a low radiation dose is capable of detecting overall cartilage quality in large anatomical ROIs. The results also suggest that CTa may be capable of determining local good and bad cartilage quality when acquired using enough radiation. In conclusion, low radiation dose CTa may be a potential clinical application to measure overall cartilage quality in humans. 420 OSTEOPOROTIC CHANGES OF SUBCHONDRAL TRABECULAR BONE IN OSTEOARTHRITIS OF THE KNEE: A 3-T MRI STUDY K. Chiba, M. Uetani, M. Ito, N. Okazaki, K. Taguchi, M. Osaki. Nagasaki Univ., Nagasaki, Japan Purpose: To investigate structural features of subchondral trabecular bone of knee osteoarthritis (OA). Methods: Sixty knees with KL grade 0–4 (all female) were examined. Fast imaging employing steady-state acquisition-cycled phases (FIESTA-c) and FatSat Spoiled gradient recalled acquisition in the steady state (SPGR) images were acquired by 3-T MRI. At four sites (the medial femur, medial tibia, lateral femur, and lateral tibia), subchondral trabecular bone structure was analyzed by FIESTA-c imaging, cartilage area was measured by SPGR imaging, and their correlation was analyzed. In addition, the subjects were classified into four groups from the cartilage area measured by SPGR imaging, and subchondral trabecular bone structure in each group was compared. Results: As cartilage area decreased in the medial joint, bone volume fraction and trabecular thickness in the medial tibia increased, and bone volume fraction, trabecular thickness, number, and connectivity in the lateral femur and lateral tibia decreased (r ≥ 0.4 or ≤-0.4, p ≤ 0.001). Compared to medially, the changes laterally showed a higher correlation. When the medial-lateral ratio of trabecular thickness in the tibia was determined, it had the highest correlation coefficient (r = -0.7, p < 0.001). These changes were not significantly detected in the early stage. Conclusions: To more sensitively detect OA changes in subchondral trabecular bone structure, a focus on osteoporotic changes in the lateral joint and the medial-lateral ratio would be useful. Detectability of early OA remains unknown, but based on a strong correlation with the degree of OA progression, trabecular structural analysis of subchondral bone may be a useful parameter to evaluate OA severity and evaluate treatment. 421 SIZE AND POSITION OF THE HEALTHY MENISCUS, AND ITS CORRELATION WITH SEX, HEIGHT, WEIGHT, BONE SIZE, AND AGE K. Bloecker 1 , W. Wirth 1 , M. Hudelmaier 1 , R. Burgkart 2 , R. Frobell 3 , F. Eckstein 4 . 1 Paracelsus Med. Univ., Inst. of Anatomy and musculoskeletal Res., Salzburg, Austria; 2 Clinic for Orthpaedics and Traumatology, Technical Univ. of Munich, Munich, Germany, Munich, Germany; 3 Dept. of Orthopedics, Clinical Sci., Lund Univ., Lund, Sweden, Lund, Sweden; 4 Paracelsus Med. Univ., Inst. of Anatomy and musculoskeletal Res. and Chondrometrics GmbH, Salzburg, Austria Purpose: Symptomatic knee osteoarthritis is known to be more common in women than in men. Also, it is suspected that meniscus extrusion increases the peak loads of the knee, which in turn increase the incidence of knee OA. Recently, an MR imaging-based method for quantitative analysis of meniscus size, shape and position (extrusion) has been introduced, which can be applied in vivo. Since men are, however, larger (and heavier) than women, a comparison of size-related morphometric meniscus measures between both sexes is not easily made. Our goals were: (1) To test whether meniscus size, shape or position differ between healthy men and women, before the onset of knee OA; (2) To examine to what extent the size of the normal meniscus correlates with body height, weight, and bone size, in men and women. (3) To identify whether a relative measure of meniscus size can be derived for a direct comparison between sexes. Methods: Knees of the healthy reference cohort of the Osteoarthritis Initiative (47 men; 70 women) without radiographic signs, symptoms or risk factors for osteoarthritis were studied. Three-dimensional multi-planar reconstructed coronal DESSwe MR images were used. Segmentation of the tibial plateaus and menisci was performed medially and laterally, and 3D measures of meniscal size and position were computed. Differences in these measures between men and women (non-paired t-test) and Pearson correlation coefficients with body height, weight, ipsi-compartimental, contra-compartimental, and total tibial plateau size were calculated. Finally, the ratio between total meniscus surface and the ipsi-compartimental tibial bone size was compared between men and women (non-paired t-test). Results: The total medial and lateral tibial plateau area was 26% greater in men (22.8 cm 2 ) than in women (18.0cm 2 ). The total medial meniscal surface area (17.8cm 2 in men, 14.7 cm 2 in women; +18%; p < 0.001) and the total lateral meniscal surface area (18.5cm 2 in men, 14.5 cm 2 in women; +22%; p < 0.001) were significantly different between both sexes. Ipsi-compartimental tibial bone size was more strongly correlated with medial/lateral meniscus size in men (r = 0.71/0.67) and women (r = 0.63/0.68) than the contra-compartimental or total tibial bone size, body height or body weight. No significant correlation was observed with age. The ratio between the total meniscus surface areas and the ipsi- compartimental tibial bone size was similar between men and women (medial meniscus: 1.52±0.16 in men and 1.55±0.16 in women; p = 0.37; lateral meniscus: 1.69±0.19 in men and 1.71±0.16 in women; p = 0.57). The within-sex, inter-subject variability for the above ratio was considerably smaller than that for total meniscus surface area without normalization (medial meniscus: CV% 10.3% vs. 14.8% in men / 10.5% vs. 13.3% in women; lateral meniscus 11.1% vs. 14.0% in men and 9.5% vs. 12.0% in women). Tibial coverage was similar in men and women (50% medially; 58% laterally). However, medial meniscal extrusion was significantly