Poster Presentations / Osteoarthritis and Cartilage 18, Supplement 2 (2010) S45–S256 S151 Abstract 340 Scenarios (Proportion developing OA Number of new OA cases likely Number of new OA cases likely Proportion of all new cases after arthroscopic surgery) due to arthroscopic surgery due to arthroscopic surgery aged 5-54 “due to” arthroscopy (5-54 years) (all ages) (proportion of new cases - all ages) 10% 2265 6267 1.9% (1.4%) 20% 4531 12534 3.9% (2.9%) 30% 6796 18801 5.8% (4.3%) 40% 9062 25068 7.8% (5.8%) OA over time, and a linear risk of developing OA after having arthroscopic surgery. Results: Examination of arthroscopic rates showed no significant fluctu- ation from 2000 to 2007. Overall, the rate of arthroscopic knee surgery was 188 per 100,000 population in Ontario, 127 per 100,000 for ages 5-44 years. Men aged 5 to 44 years had twice as many surgeries as women the same age (166/100,000 persons and 88/100,000 persons respectively). Rates increased with increasing age peaking in the 45-54 years age group in men and in the 54-64 years age group in women. The findings from applying the rates for those aged 5 to 44 years at surgery to the population and the four scenarios for the proportion developing OA over 10 years are shown in the table below. Conclusions: A calculation juxtaposing different sources of data suggests that arthroscopic surgery associated with knee injury likely accounts for between 1.9% and 7.8% of new cases of OA in the younger population and between 1.4% and 5.8% of all cases. Given population rates of arthroscopic surgery and assuming outcomes similar to those reported in the literature, this simple estimation highlights the importance of injury prevention particularly in relation to OA in the younger population. 341 EROSIVE HAND OSTEOARTHRITIS: ITS PREVALENCE AND CLINICAL IMPACT IN THE GENERAL POPULATION AND SYMPTOMATIC HAND OSTEOARTHRITIS W.-Y. Kwok 1 , M. Kloppenburg 1 , F.R. Rosendaal 2 , J.B. van Meurs 3 , A. Hofman 4 , S.M. Bierma-Zeinstra 5 1 Leiden Univ. Med. Ctr., Dept. of Rheumatology, Leiden, Netherlands; 2 Leiden Univ. Med. Ctr., Dept. of Clinical Epidemiology, Leiden, Netherlands; 3 Erasmus Med. Ctr., Dept. of Internal Med., Rotterdam, Netherlands; 4 Erasmus Med. Ctr., Dept. of Epidemiology, Rotterdam, Netherlands; 5 Erasmus Med. Ctr., Dept. of Gen. Practice, Rotterdam, Netherlands Purpose: Osteoarthritis (OA) is a heterogeneous disease that is prevalent in hands. Different subsets within hand OA, such as nodular hand OA and erosive hand OA (EOA), are recognized. EOA is a subset, defined as having erosions in interphalangeal joints on radiographs, that is suggested to have an aggressive clinical course. No clear data about prevalence and clinical burden of EOA are available. This study assesses the prevalences of erosive hand OA in the general population and symptomatic hand OA, and its relation to hand pain and disability. Methods: Cross-sectional data of participants from a population based study (age ≥ 55 years) in the Rotterdam area were used. Radiographs of the hands were scored previously by the Kellgren-Lawrence (K/L) scoring method. In the present study, erosive lesions (defined as an erosive or remodeled phase) in the distal, proximal and thumb interphalangeal joints (DIPJs, PIPJs and IPJs) were scored according by the Verbruggen-Veys scoring method. Radiographic hand OA was defined as the presence of K/L grade ≥ 2 in two out of three groups of hand joints (DIPJs/IPJs, PIPJs and first carpometacarpal joint (1st CMCJs)) of each hand. Symptomatic hand OA was defined as hand pain and presence of criteria for radiographic hand OA. Hand pain was assessed by a standard question by trained interviewers. Hand disability was measured by 8 hand questions in the Stanford Health Assessment Questionnaire (HAQ). A mean HAQ-score of ≥ 0.5 was classified as disability. Point prevalence was calculated as to divide the number of subjects with one erosive lesion by the total population. Multivariate logistic regression analyses were used to estimate the effect of erosive hand OA on pain and disability between erosive and non-erosive patients. Results were presented as odds ratios (OR) with a 95% confidence interval (95%CI), adjusted for age and sex. Results: Of the 3459 participants, 56% was female, with a mean age (SD) of 66 years (7.0) and mean Body Mass Index (SD) of 26.3 (3.6) kg/m 2 . Radiographic hand OA was seen in 27% (n=941) and symptomatic hand OA in 6% (n=207). One erosion only was seen in 96 subjects and minimal two erosions or more were seen in 44 persons (46% of erosive patients). The overall prevalence of EOA in the general population was 2.8% and in individuals with symptomatic hand OA 15.8%. Presence of erosive hand OA led to an adjusted OR for pain of 3.6 (95%CI 2.4-5.6) and for disability 2.4 (95% CI 1.1-5.4) in the general population. In subjects with radiographic hand OA, participants with erosion were more likely to experience hand pain (adjusted OR 3.1, 95%CI 1.9-5.2) and more likely to report hand disability (adjusted OR 3.1, 95%CI 1.3-7.6). A dose-response relationship between the number of erosions per participant and presence of pain was seen. If participants had two or more joints with erosions, they were five times more likely to have pain in their hands than individuals without erosions in the general population (adjusted OR 5.3, 95%CI 2.9-9.9). Conclusions: The prevalence of erosive hand osteoarthritis is 2.8% in the general population and rises to 15.8% in individuals with symptomatic hand OA. It has a substantial impact on clinical burden. Further studies should focus on treatment options for this hand OA subset. 342 ALL-CAUSE AND DISEASE-SPECIFIC MORTALITY IN PATIENTS WITH KNEE OR HIP OSTEOARTHRITIS: POPULATION-BASED COHORT STUDY E. Nüesch 1 , P. Dieppe 2 , S. Reichenbach 1 , S. Iff 1 , P. Jüni 1 1 Univ. of Bern, Bern, Switzerland; 2 Peninsula Coll. of Med. and Dentistry, Plymouth, United Kingdom Purpose: Older age, presence of comorbidities and adverse effects of anal- gesics have been suggested to contribute to an excess mortality in patients with osteoarthritis, but available data are limited. Therefore, we examined determinants of mortality in a large population based cohort of patients with osteoarthritis of the knee or hip. Methods: We analysed all-cause and disease-specific mortality over a median of 14 years in 1163 patients with symptomatic and radiologically confirmed osteoarthritis of the knee or hip from the population-based Somerset and Avon Survey of Health (SASH) using mortality data from Characteristic Hazard ratio P-value (95% confidence interval) Age at baseline <0.001 35 to 54 years 1.00 (reference) 55 to 74 years 12.1 (5.38 to 27.4) ≥75 years 40.7 (17.7 to 93.5) Male gender 1.56 (1.27 to 1.90) <0.001 Lower social class 1.13 (0.93 to 1.38) 0.21 Smoking at baseline 1.26 (0.97 to 1.63) 0.08 Type of osteoarthritis 0.66 Knee only 1.00 (reference) Hip only 1.12 (0.87 to 1.45) Knee and hip 1.08 (0.86 to 1.35) Previous joint replacement 1.16 (0.83 to 1.61) 0.38 Obesity 0.84 (0.67 to 1.05) 0.13 COPD 1.20 (0.93 to 1.55) 0.16 Diabetes 1.97 (1.33 to 2.93) 0.001 Depression 0.95 (0.68 to 1.34) 0.79 Arterial hypertension 1.14 (0.92 to 1.40) 0.24 Cancer 2.32 (1.53 to 3.52) <0.001 Chronic inflammatory disease 1.01 (0.83 to 1.24) 0.91 Eye disease 1.07 (0.81 to 1.42) 0.65 Cardiovascular disease 1.40 (1.14 to 1.73) 0.002 Knee or hip pain 0.90 (0.73 to 1.10) 0.30 Walking disability 1.53 (1.23 to 1.91) <0.001