Effects of medication-treated diabetes on incidence and progression of knee osteoarthritis: a longitudinal analysis of the Osteoarthritis Initiative data

• Published in: Rheumatology international Vol. 37; no. 6; pp. 983 - 991
• Main Authors: Shirinsky, Ivan V., Shirinsky, Valery S.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2017; Springer Nature B.V
• Subjects: Aged; Arthritis; Chi-Square Distribution; Diabetes; Diabetes Mellitus - diagnosis; Diabetes Mellitus - drug therapy; Diabetes Mellitus - epidemiology; Disease Progression; Female; Humans; Hypoglycemic Agents - therapeutic use; Incidence; Knee; Logistic Models; Longitudinal Studies; Male; Medicine; Medicine & Public Health; Middle Aged; Observational Research; Odds Ratio; Osteoarthritis, Knee - diagnostic imaging; Osteoarthritis, Knee - epidemiology; Prognosis; Prospective Studies; Protective Factors; Rheumatology; Risk Factors; Time Factors; United States - epidemiology; Progression; Knee osteoarthritis; Comorbidity; Diabetes mellitus; Incidence
• ISSN: 0172-8172; 1437-160X
• DOI: 10.1007/s00296-017-3676-7

Diabetes has been proposed as a factor involved in the pathogenesis of osteoarthritis (OA). Currently, there is a lack of research evaluating the prospective impact of diabetes on OA structural outcomes. In this study, we assessed the effects of medication-treated diabetes on incidence and progression of knee OA. We analysed longitudinal data from the multi-center, longitudinal, prospective observational Osteoarthritis Initiative (OAI) study. The main outcomes were radiographic OA incidence (development of Kellgren–Lawrence grade 2 with joint space narrowing, JSN) and progression (increase in semiquantitative JSN or a new knee replacement). For the study of incidence, we selected participants with KL <2 or /KL = 2 without JSN at baseline (incidence sample). To evaluate progression, we selected participants with baseline JSN <3 (progression sample). We used generalized estimating equations (GEE) logistic regression with adjustment for potential confounders to evaluate the effects of medication-treated diabetes on knee OA incidence and progression. We studied 3725 knees (3498 non-diabetic and 228 diabetic) in the incidence sample and 3594 knees (3335 non-diabetic and 259 diabetic) in the progression sample. Medication-treated diabetes did not have an effect on knee OA incidence (odds ratio, OR 0.53, 95% confidence interval, CI 0.23–1.5). There was an independent association between medication-treated diabetes and reduced progression of knee OA [OR 0.66, 95% CI (0.44–0.98)]. Medication-treated diabetes has no effect on knee OA incidence but reduces knee OA progression. The role of diabetes and anti-diabetic drugs in knee OA progression needs further exploration.