PARIS – Walking disability from hip or knee osteoarthritis puts patients with comorbid diabetes at increased risk for serious diabetes complications, according to a retrospective cohort study.
After a median follow-up of 6.6 years among 437 patients with moderately severe symptomatic osteoarthritis (OA) and type 1 or type 2 diabetes, 37% experienced the composite diabetes-specific primary outcome of hospitalization for hypo- or hyperglycemia, soft tissue infection, amputation, or initiation of chronic dialysis.
©iStockphoto.com
Diabetic patients who have limited walking functionality are at an increased risk for diabetic complications, researchers said.
Specifically, there were 51 hospitalizations for hypoglycemia, 11 for hyperglycemia, 127 for soft tissue infections, 10 amputations, and 4 patients who went on chronic dialysis.
After adjusting for age, sex, and preexisting cardiovascular disease, baseline walking disability was a significant independent predictor of risk for a non-CVD, diabetes-specific complication (Hazard ratio per unit increase in Health Assessment Questionnaire [HAQ] walking score, 1.26; P = .02), Dr. Gillian Hawker reported at the World Congress on Osteoarthritis.
In sensitivity analyses, the results were unchanged after additional adjustment for post-baseline receipt of a primary, elective hip or knee joint replacement or when retinopathy was included as an outcome.
Dr. Gillian A. Hawker
"In people with diabetes, OA functional limitations may impede their ability to carry out diabetes self-management activities, increasing their risk for diabetes complications," noted Dr. Hawker, professor of medicine at the University of Toronto and physician-in-chief of medicine at Women’s College Hospital, Toronto.
This is particularly worrisome since the coprevalence of diabetes with OA is high, with as many as 50%-60% of patients with longstanding diabetes also having clinically evident hand, hip, or knee OA.
In an oral presentation at the same meeting, Dr. Hawker stressed the need for early identification of walking disability after reporting that greater walking disability was an independent predictor of all-cause death and major cardiovascular events in the overall cohort of patients with symptomatic hip or knee OA alone.
The retrospective cohort study linked provincial health administrative databases with surveys from a population cohort of 2,156 patients with at least moderately severe symptomatic hip or knee OA who were recruited from 1996 to 1998 through a screening survey in Ontario.
At baseline, 457 patients self-reported diabetes or met criteria for inclusion in the Ontario Diabetes Database (sensitivity 86%, specificity 97%) and were without preexisting retinopathy or renal failure. After censoring patients who died, emigrated, or had insufficient data, 434 patients were included in the current analysis. The study could not differentiate type 1 from type 2 diabetes.
Their mean age was 71.6 years (all were 55 years or older), 41% were obese, and 53.2% had preexisting CVD. Their median HAQ walking disability score was 2 on a 3-point scale, median HAQ grip score was 1 on a 3-point scale, and mean Western Ontario and McMaster Osteoarthritis Index (WOMAC) summary score was 43.4. Overall, 55.5% used a walking aid, and 39% reported using NSAIDs. (A walking disability score of 2 corresponds with walking outdoors on flat ground with much difficulty, whereas a score of 3 means the person is unable to do it. A grip score of 1 corresponds with some difficulty in just one of the following tasks: opening car doors, opening previously open jars, or turning faucets on and off.)
In multivariable analyses, no effect was found for baseline grip strength (adjusted HR, 1.16; P = .07), Dr. Hawker reported at the meeting, sponsored by the Osteoarthritis Research Society International.
However, in sensitivity analyses that further adjusted for receipt of a diabetes drug prescription or NSAID in patients at least 66 years old at baseline and thus eligible for drug benefits, both HAQ walking and grip scores were significant predictors of risk for a diabetes complication, she reported. Specific data were not shown in the poster presentation, but the adjusted hazard ratios per unit increase in HAQ were 1.36 (P = .003) and 1.26 (P = .01), respectively, according to the abstract.
"Controlling for confounders, reduced grip strength and increased walking disability were identified as potentially modifiable risk factors for serious diabetes complications in people with both OA and diabetes," the authors concluded. "Additional studies are warranted to confirm or refute our findings, and if confirmed, to elucidate potential mechanisms."
Dr. Hawker reported having no financial disclosures.