The use of an antimalarial medication may prevent the development of diabetes in patients with rheumatoid arthritis, according to research presented this week at the American College of Rheumatology Annual Scientific Meeting in San Francisco, Calif.

Diabetes affects nearly eight percent of the population and is a major cause of morbidity and death due to complications that include coronary heart disease, kidney failure and blindness. Rheumatoid Arthritis is a systemic autoimmune disease affecting about 1 percent of the population that causes pain, stiffness, swelling and reduced mobility and function of multiple joints. Despite multiple risk factors for diabetes, including treatment with steroids and sedentary lifestyle, people with RA are not affected by diabetes more than the general population. The use of hydroxychloroquine—a medication developed to treat malaria and is currently used to treat some autoimmune diseases—may be the reason for the unexpectedly low rate of diabetes in RA.

A 2007 study published in the Journal of the American Medical Association showed a 77 percent reduction in new cases of diabetes in patients with RA who were taking hydroxychloroquine for over four years. A limitation of the study, however, was that all data was collected by mailed questionnaires to patients with RA.

Researchers recently used an electronic health records database with physician-coded diagnoses and laboratory measures to corroborate the results of the 2007 study. They used the database to identify 1,824 patients with RA without diabetes and collected data such as demographics, body mass index, laboratory results and medication use. Of the participants, 74 percent were women and 97 percent were Caucasian; they had a mean age of 62.4 years and a body mass index of 29.3.

Participants were divided into groups based on their previous use of hydroxychloroquine and rates of new cases of diabetes were compared for the hydroxychloroquine users and non-users.

Of the 1,824 participants, 489 had previously taken hydroxychloroquine and 1,335 had not. During observation, the rate of newly diagnosed diabetes among hydroxychloroquine users was about half the rate noted in the non-users (17.2 vs. 33.8 new cases of diabetes per 1,000 people per year). In further analysis—adjusted for demographics, body mass index, use of steroids, methotrexate and tumor necrosis factor inhibitors, and disease severity—having ever used hydroxychloroquine was associated with a 53 percent reduction in the development of new cases of diabetes.

These results confirm, and further strengthen, the previously reported protective relationship between hydroxychloroquine use and the risk of developing diabetes.

“In patients with rheumatoid arthritis, use of hydroxychloroquine may reduce the risk of developing diabetes by 53 percent, explains Androniki Bili, MD, MPH; rheumatology, Geisinger Health System, Danville, Pa., and lead investigator in the study. “Given the favorable safety profile and low cost of this generic medication, these findings may have implications for the use hydroxychloroquine for prevention of diabetes in other high risk groups.”

Patients should talk to their rheumatologists to determine their best course of treatment.

The ACR is an organization of and for physicians, health professionals, and scientists that advances rheumatology through programs of education, research, advocacy and practice support that foster excellence in the care of people with or at risk for arthritis and rheumatic and musculoskeletal diseases. For more information on the ACR’s annual meeting, see http://www.rheumatology.org/annual.

Editor’s Notes: Dr. Bili will present this research during the ACR Annual Scientific Meeting at the Moscone Center from 9:00 – 11:00 AM on Monday, October 27, in Hall A. Dr. Bili will be available for media questions and briefing at 1:30 PM on Monday, October, 27 in the on-site press conference room, 114.

Presentation Number: 780

Hydroxychloroquine Use Reduces Risk Of Diabetes In Rheumatoid Arthritis Patients

Androniki Bili1, Eric D. Newman1, H. Lester Kirchner2, Jonathan A. Cook2, Lindsay J. Ledwich1, Azadeh Stark2, Robert D. Langer2, Mary Chester M. Wasko3. 1Geisinger Medical Center, Danville, PA; 2Geisinger Center for Health Research, Danville, PA; 3University of Pittsburgh, Pittsburgh, PA

Purpose: A recent self-report study showed a decreased incidence of diabetes mellitus (DM) in rheumatoid arthritis (RA) patients taking hydroxychloroquine (HCQ) (Wasko, JAMA 2007.) The aim of this study was to verify these findings using our health system Electronic Health Records (EHR - Epic) database with physician-coded diagnoses and laboratory measures.

Methods: Patients with RA (ICD-9 code 714.0 at >2 office visits with a rheumatologist Sep 00-Feb 08, n=2093) were identified through the EHR. Prevalent cases of DM were excluded (n=269). Data on demographics, body mass index (BMI), laboratory values and medications were collected. Outcome: Incident cases of DM defined as either CD-9 250 (diabetes) or random serum glucose ≥200 mg/dl or Hemoglobin A1c ≥7 or hypoglycemic drug use were identified in EHR. Statistical Analysis: Patients were stratified by ever and never use of HCQ. Incidence rates were computed as the number of DM per 1000 person-y and compared between HCQ groups using a Poisson regression model. Potential risk and protective factors for DM were evaluated using a Cox Proportional Hazard regression model. Drug use was considered as both a binary (ever/never) and a cumulative exposure, allowing patients to go on and off each medication. Interactions between medications were evaluated.

Results: Patients were 74% women, 97% Caucasian, with mean age of 62.4 y and BMI 29.3 kg/m2. Rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP) antibodies were positive in 79% and 52%, respectively. There were 525 ever and 1299 never HCQ users, with mean follow-up 36.2 and 36.8 months, respectively. During observation, incident DM was recorded in 16 ever and 154 never HCQ users, with incidence rates of 17.2 per 1000 patient-y vs. 33.8 per 1000 patient-y (p=0.01.) In analysis adjusted for gender, age, BMI, positive RF and anti-CCP, use of steroids, methotrexate and anti-tumor necrosis factor drugs, the hazard ratio for incident DM was 0.47 (95% confidence interval 0.26-0.82, p=0.008.) Inclusion of erythrocyte sedimentation rate in the model did not alter the results. Duration of HCQ use was not significantly associated with the outcome. No interactions between medications were identified.

Conclusions: Use of HCQ in RA patients in our health system was associated with 53% reduction in risk of DM. Incidence rates of DM in our population are higher than reported elsewhere likely due the broad definition of DM and high mean BMI in our patients. Duration of HCQ use did not affect risk, perhaps due to the short observation period. These results verify and strengthen the protective association between HCQ use and incident DM by using physician diagnoses and laboratory measures. Given the relative safety and low cost of this generic drug, HCQ may be useful in the prevention and treatment of DM in the general population.

Disclosure Block: A. Bili, None; E.D. Newman, None; H.L. Kirchner, None; J.A. Cook, None; L.J. Ledwich, None; A. Stark, None; R.D. Langer, None; M.M. Wasko, None.

Source: American College of Rheumatology (ACR)