Image from MedPageToday.com

Published: Sep 8, 2014
By Nancy Walsh, Senior Staff Writer, MedPage Today

Diabetes was strongly associated with hand pain and joint tenderness in patients with erosive hand osteoarthritis (OA), Norwegian researchers reported.

In a stratified analysis, pain as measured on the Australian-Canadian (AUSCAN) hand pain scale was associated with diabetes among patients who had one or more joints with central erosions detected radiographically (beta=3.81, 95% CI 2.27-5.35), according to Karin Magnusson, who is a PhD student at Diakonhjemmet Hospital in Oslo, and colleagues.

There also was a strong association between diabetes and the number of hand joints that were tender on palpation, but again only in patients with erosive disease (beta=4.16, 95% CI 2.01-6.31), the researchers reported online in Arthritis Care & Research.

A number of factors have been shown to correlate with pain in knee OA, such as high body mass index and depression, but little is known about factors other than specific structural changes to the joint that might influence pain in hand OA.

“No previous studies have examined whether lifestyle and lifestyle-related factors, mental health, socioeconomic status, familial OA, and structural and inflammatory features are associated with pain in persons with hand OA, taking all other covariates into account,” Magnusson and colleagues wrote.

So they sent questionnaires asking about sociodemographic factors and medical history to participants in a population-based study who self-reported having OA, and provided clinical examinations and radiographic assessments for 530 individuals.

Factors included in the analysis included height and weight, physical activity, familial OA, smoking and alcohol consumption, hypertension and diabetes, mental health, and the presence of widespread pain.

The analyses were adjusted for age and sex, as well as for the use of analgesics and anti-inflammatory medications, and were stratified according to the two different phenotypes of erosive and non-erosive disease.

Hand pain was evaluated at the patient level both during activity and at risk with AUSCAN, and at joint level with palpation.

For the AUSCAN outcome, the researchers found that factors associated with greater pain on univariate analysis included female sex, familial OA, poor mental health, greater number of joints with radiographic abnormalities and synovitis, widespread pain, lower educational level, and low alcohol consumption.

On a final multivariate analysis, the same variables remained significant except familial OA. This model explained 35.4% of variance in the degree of AUSCAN pain, Magnusson and colleagues reported.

Then, on the stratified analysis of erosive and non-erosive OA, widespread pain and education level were associated with both phenotypes (beta= -1.15, 95% CI -1.82 to -0.48 and beta= 2.26, 95% CI 1.6-2.95, respectively).

Factors other than diabetes found to be associated with erosive disease and AUSCAN pain included female sex, radiographic OA, and severity of synovitis.

Factors associated with non-erosive disease and AUSCAN were poor mental health and low alcohol consumption.

For the joint-level assessment, factors associated with a greater number of tender joints included female sex, widespread pain, poor mental health, greater synovitis and radiographic changes, and diabetes.

This model explained 25.8% of variance in the number of tender joints, according to the researchers.

On the analysis of number of tender joints stratified according to erosive and non-erosive disease, widespread pain was associated with both.

The significant association number of tender joints and diabetes among patients with erosive disease remained after adjustment for body mass index. Synovitis also was associated with the number of tender joints only in erosive disease.

In non-erosive hand OA, factors that were associated with the number of tender joints included poor mental health and familial OA.

“An important finding of this study was the differences explaining hand OA pain in non-erosive versus erosive disease. The most remarkable difference was that diabetes was associated with both self-reported hand pain and number of tender joints in erosive OA, whereas no associations [with diabetes] were found in non-erosive OA,” the researchers wrote.

Why diabetes would be linked with pain in patients with hand OA remains to be fully explained.

One possible reason is that if the diabetes was uncontrolled, neuropathy may have been present. However, the researchers had no information about glycated hemoglobin levels for their study participants.

Another factor that may contribute is systemic inflammation associated with diabetes that may localize to the joints.

“Future longitudinal studies are needed to further look into the role of diabetes in OA pathogenesis as well as the association with pain in erosive hand OA,” the authors observed.

They also pointed out that widespread pain was associated with both phenotypes of OA, which suggested that central pain sensitization may contribute.

“In conclusion, the present study showed that except for having widespread pain, factors associated with hand OA pain depended on OA phenotype.”

A limitation of the study was its cross-sectional design.