AMSTERDAM — Among patients with knee osteoarthritis (OA), greater structural disease severity and worse functional limitations were associated with an increased risk of developing depression, a researcher reported here.

Compared with patients in the lowest quintile, patients in the highest quintile of disease structural severity as measured according to joint space width had an odds ratio for depressive symptoms of 2.25 (95% CI 1.27-3.99), according to Alan M. Rathbun, PhD, of the University of Maryland School of Medicine in Baltimore.

And for functional limitations as measured by performance on the 20-meter gait speed test, those in the highest quintile had an odds ratio for depression of 2.08 (95% CI 1.16-3.75), with both measures being statistically significant, he said during a press briefing at the annual European Congress of Rheumatologyy, sponsored by the European League Against Rheumatism (EULAR).

“The results of this study provide valuable insights into the components of OA disease severity that are related to depression onset,” commented Thomas Dörner, MD, of Charité University Hospital in Berlin, who chairs the EULAR abstract selection committee.

Knee OA affects more than 10% of men and women older than 60, and is a prominent cause of pain and disability.

“Depression occurs frequently in persons with OA, and meta-analytical evidence indicates that the prevalence of depressive symptoms in this population is approximately 20%, more than double that seen in the general population,” Rathbun said.

Moreover, depression in knee OA is associated with lower quality of life, greater healthcare utilization and expenditures, and higher mortality rates.

The progression of knee OA and symptom worsening over time may increase the risk of depression, but it is unclear what components of disease severity contribute to that risk in this population.

“We hypothesized that greater pain, functional limitation, and structural disease severity would be associated with an increased risk of depression onset in patients with knee OA,” he said.

To explore this, he and his colleagues analyzed data for 1,652 patients enrolled in the Osteoarthritis Initiative ages 45 to 79 who had radiographic disease but did not have depression at the time of enrollment and had complete data at baseline and three annual follow-up visits.

Marginal structural models, which use inverse probability weights to account for time-dependent confounding and loss to follow-up were the primary method of analysis. Potential confounders included sociodemographic and behavioral factors as well as clinical characteristics.

A third disease measure along with structural disease severity and functional limitation was pain, as rated on the Western Ontario and McMaster Universities OA Index. The presence of depressive symptoms was assessed at each annual visit on the 20-item Center for Epidemiologic Studies depression scale, with a cutoff off 16 or higher indicating the presence of depression.

On each of the three disease metrics, there was a nonlinear trend for probability of depression in the first through fourth disease severity quintiles, with nonsignificant marginal estimates of probability between 0.02 and 0.03. However, the likelihood of experiencing depressive symptoms was largest in the highest quintiles of each of the three predictor variables, although the odds ratio was not statistically significant for the pain metric (1.60, 95% CI 0.81-3.16).

“Our data imply that greater structural disease severity and decreased physical performance are associated with a statistically significant increase risk of depression. Structural progression is irreversible and likely precedes pain and functional limitations over time,” Rathbun said.

“We believe that to effectively treat individuals with radiographic knee OA and comorbid depressive disorder, it is necessary to use a combined treatment strategy of two interventions delivered in parallel to simultaneously target each condition,” he concluded.

“This study is a good example of a musculoskeletal disease that has psychosocial implications,” said Robert Landewé, MD, of the University of Amsterdam, who moderated the press briefing.