C44 Podium Presentations stitutional review boards. OA menisci were obtained from 16 end-stage OA knees of 16 patients (age, 64-83 with a mean of 76) at time of prosthetic surgery. The knees were all medially involved in the disease. Control menisci were obtained from 10 non-arthritic knees of 5 donors (age, 72-88 with a mean of 79) at autopsy within 24 hours after death. For histological evalua- tion, hematoxylin and eosin (HE) and Masson’s trichrome (MT) stained sections were prepared and observed under a light mi- croscope. The presence of type III collagen was evaluated by immunostaining. In molecular biological evaluation, matrix gene expression was compared between OA and control menisci in a site-to-site manner, employing laser capture microdissection (LCM). For this, the evaluation was performed on the anterior segments of menisci because those segments were well preserved even in OA menisci. Cryosections were prepared and divided into three parts of the inner, surface, and deep regions by LCM. The ex- pression of COL1A1, COL1A2, COL2A1, COL3A1, and AGC1 were evaluated in respective regions quantitatively by real-time PCR. Results: In histology, the number of meniscus cells was moder- ately increased in all regions in OA menisci. Tissue degeneration was most obvious in the middle segments of medial OA menisci, where the integrity of collagen fibrils was severely disturbed. For- mation of cell clusters was not observed in human OA menisci. In molecular biological evaluation, mRNA expression of COL1A1 and COL1A2 was highly increased in OA menisci by 23- to 77- fold compared with the controls. The expression of COL2A1 was also strongly enhanced over 100-fold. The increase was most dramatic with COL3A1, which reached over 160-fold. On the con- trary, the expression of AGC1 was least increased, 7- to 22-fold elevation from that of the controls. The increase in mRNA ex- pression was similar between the medial and lateral OA menisci, although the medial menisci underwent severer degeneration. Although obvious regional differences have been reported with animal menisci, such differences were hardly recognized with human menisci in either OA or controls, by histological or molec- ular biological evaluation. The result of immunostaining indicated the presence of an increased amount of type III collagen in OA menisci. Conclusions: This is the first report, as far as we know, on the metabolic change of meniscus cells in human OA menisci. The metabolic activity of meniscus cells was highly enhanced in OA menisci, in accordance with previous animal studies. In the pathology of OA, it may be noteworthy that cellular metabolism is highly enhanced in meniscus cells, as well as in chondrocytes. 55 SITE AND WORSENING OF MENISCAL DAMAGE IN KNEES WITH AND WITHOUT RADIOGRAPHIC OA: THE MOST STUDY M. Javaid 1 , J.A. Lynch 1 , A. Guermazi 2 , F. Roemer 2 , I. Tolstykh 1 , C. McCulloch 1 , M. Englund 2 , D. Felson 2 , M.C. Nevitt 1 1 UCSF, San Francisco, CA; 2 BUSM, Boston, MA Purpose: Meniscal damage may alter joint loading and ac- celerate degeneration of the joint; recent reports suggest that meniscal damage is an important predictor of knee OA onset and progression. However, there is little research describing the sites of meniscal damage and meniscal degeneration over time in knees with varying degrees of OA. Therefore, we examined the site of meniscal damage and assessed worsening of meniscal damage over 30 months in knees with and without radiographic OA and evaluated factors associated with meniscal degeneration. Methods: Participants are from MOST, a community-based co- hort of persons with or at high risk of knee OA. Subjects had weight-bearing PA (fixed flexion) knee x-rays and full-limb films (for assessmenet of HKA angle) at baseline and sagittal proton density FSE fat suppressed and coronal STIR MRI of the knees (1.0 T OrthOne extremity scanner) at baseline and 30-months. MRIs were read paired and graded (0-4) for damage to the medial and lateral meniscus in the anterior, body and posterior locations (Grades: 1 = minor radial/parrot beak tear; 2 = non- displaced tear; 3= displaced tear/partial resection; 4 = complete maceration/resection) at each time point. We analyzed all knees with a K-L grade of 0-3 and aN MRI reading of baseline and 30- month meniscal damage. Medial and lateral menisci were each classified by location of damage: anterior, body, posterior and their combinations. Compartment-specific worsening of meniscal damage was defined as any increase of ≥1 grade in any location within a meniscus during follow-up. We used logistic GEE models to examine predictors of meniscal worsening. Results: MRIs of 397 knees (374 subjects) were read for menis- cal damage at baseline and 30 months; mean age was 62.6 yrs, 67% were female, 86% Caucasian with a median BMI of 29.6 kg/m 2 . 46% of knees were K/L grade =0, 24% grade =1, 15% grade = 2 and 15% grade 3. At baseline, 34.3% of medial and 9.8% of lateral menisci showed signs of damage (grade > 0) while 13.0% of medial and 3.0% of lateral menisci had severe damage (grade 3-4). 27% of knees with meniscal damage had a history of injury that limited walking vs. 18% of knees without meniscal damage (p=0.03). Medially, posterior damage was most common, followed by body then a small number with anterior damage; while laterally, damage was equally common by location. 13.2% of K-L 0 and 32.0% of K-L 1 knees had damage to the posterior region of the medial meniscus. The prevalence of damage in all locations of both menisci increased with higher KL grades (p<0.01), but patterns of damage by location were similar by K-L grade (p>0.2) (Fig. 1). Figure 1. Patterns of Baseline Meniscal damage (grade < 0) by KL grade in 379 knees. 1 For single sites, percentages include any damage to that region whether solitary or in combinations with other regions. Meniscal worsening was seen in 52 (12.3%) medial and 19 (4.5%) lateral menisci. Focusing on the medial meniscus, female gender and varus malalignment (HKA <177 degrees) (both p <0.001), but not age, injury nor BMI, were associated with baseline medial damage independently of K-L grade. None of these factors were significantly associated with worsening medial meniscal damage. K-L grade 1 knees had a significantly higher risk of worsening than grade 2-3 knees (p< 0.001). Conclusions: We found that the location of damage within the meniscus differed by compartment but was similar by radio-