BOSTON, Massachusetts—Inflammation is a hallmark of rheumatoid arthritis (RA) and is also implicated in heart disease. Some research has suggested that inflammation may play a role in diabetes, but a new study in the March issue of the Journal of Rheumatology¹ found no link between diabetes and RA, which seems to debunk this theory.

Or does it?

Even the study authors point out that their study could not rule out exclusively a moderate association between the two conditions. A related study in the Annals of Rheumatic Diseases² shows that reducing

"The evidence regarding the link between RA and diabetes for both Type 1 and 2 is conflicting, and at this time I do not think there is enough evidence to support an association between these conditions; however, this has not been completely ruled out either," Michele Doran, MD, FRCPI, rheumatologist at St. James's Hospital, in Dublin, Ireland, told CIAOMed. Dr. Doran wrote an editorial accompanying the new report.³ "A large well-conducted prospective study addressing this question might provide a more definitive answer," Dr. Doran added.

No strong association seen

Several earlier studies have shown an association between diabetes and RA, but none were designed to pick up this association beforehand. In the new report, researchers led by Julia F. Simard, a doctoral candidate at the Harvard School of Public Health in Boston, used data from the National Health and Nutrition Examination Survey III (NHANES) that included 5302 participants aged >e;60 years, of whom 144 had RA and 24 (17%) of whom were found to have prevalent diabetes. Of the 5152 RA-free participants, 16% had diabetes.

After controlling for demographic factors, body mass index, physical inactivity, and steroid prescriptions, they found no association between RA and diabetes. The adjusted odds ratios for the cross-sectional association between RA and diabetes ranged from 1.1 to 1.5, but was not statistically significant, Simard et al report. The small numbers of patients with diabetes and RA limited the power of the study, they note.

"We can conclude that this study rules out a strong association, but cannot definitively exclude a modest non-null association," the study authors conclude. "Future longitudinal studies with larger numbers of participants with RA and diabetes are needed to examine temporality and consider the possibility of a modest association."

Limitations of new study leave question open

So is there or isn't there a connection between the two diseases? "I certainly think that it deserves further attention," Dr. Simard told CIAOMed. "Despite the limitations of our study, diabetes was more common amongst people with RA."

Exactly how RA and diabetes may be connected is unclear

"Inflammation is a likely possibility," she said. "Steroid use is another possibility; however, we saw a higher prevalence of diabetes among subjects with RA who did not use steroids."

Editorialist Dr. Doran adds that study patients with RA had very mild disease. Just 10% were talking disease modifying antirheumatic drugs and the mean C-reactive protein level was only 1.04 mg/dL. The proportion of rheumatoid factor positive patients was smaller than in other population-based studies, she writes.

Perhaps, she speculates, patients with severe RA did not participate in the NHANES study because of disability related to their RA. "Because diabetes may be expected to occur more often in patients with more severe RA, who have higher levels of inflammation, the noted prevalence of diabetes in these patients may be an underestimation."

TNF-inhibition restores insulin sensitivity in RA

In a related report, eight nondiabetic patients with inflammatory disorders who were receiving infliximab (Remicade®) had their blood sugar tested with hyperinsulinemic-euglycemic clamp technology at baseline and before each infusion. Two RA patients had a higher fasting glucose concentration and a slight decrease in insulin sensitivity at baseline, while four RA patients had low insulin sensitivity at baseline.

TNF-inhibition improved insulin sensitivity in seven patients (5.09 (6.24) at baseline versus 6.56 (8.97) after 6 weeks of therapy, P = .05).  Infliximab therapy restored insulin sensitivity within the normal ranges in three of these patients and worsened insulin sensitivity in one patient with ankylosing spondylitis and manifestations of peripheral arthritis.

"Our findings are in line with previous studies, which reported that antiTNF agents corrected disturbances in glucose metabolism in patients with RA," writes lead researcher Frank C. Huvers, MD, with Radboud University Nijmegen Medical Centre, in The Netherlands.

The researchers speculated that improvements in insulin sensitivity were due to suppressing the underlying inflammation, because therapy resulted in a decrease of disease activity score -28 (DAS-28) and erythrocyte sedimentation rate.

"However, no significant correlation between disease activity or other inflammatory markers and insulin sensitivity at both time points could be detected," they conclude.

References

1. Simard JF, Mittleman MA. Prevalent rheumatoid arthritis and diabetes among NHANES III participants aged 60 and older. J Rheumatol. 2007;34:469-473.
2. Huvers FC, Popa C, Netea MG, et al. Improved insulin sensitivity by anti-TNF-α antibody treatment in patients itch rheumatic diseases. Ann Rheum Dis. 2007;66:558-559.
3. Doran M. Rheumatoid arthritis and diabetes mellitus: evidence for an association? J Rheumatol. 2007;34:460-462.