Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement
Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.
Glucocorticoid Use Is Not Associated with an Increased Risk of Cerebrovascular Accidents in Patients with Rheumatoid Arthritis. A Population-Based Study.
Avina-Zubieta1, Antonio, Abrahamowicz2, Michal, Choi4, Hyon K., Rahman1, M. Mushfiqur, Sylvestre3, Marie-Pierre, Esdaile1, John, Lacaille1, Diane V.
Arthritis Res Centre of Canada, Vancouver, BC, Canada
Department of Epidemiology and Biostatistics, McGill University, Montreal, QC, Canada
Samuel Lunenfeld Research Institute
Univ of British Columbia, Vancouver, BC, Canada
While it has been shown that RA patients treated with glucocorticoids (GC) have an increased risk of coronary artery disease, the data available for cerebrovascular accidents (CVA) are scarce and controversial. We examined the effect of GC on risk of CVA in patients with rheumatoid arthritis (RA).
Using population-based administrative health data we assembled a cohort of 7,051 incident RA cases with disease onset between 1997 and 2001 with no history of CVA and no GC use prior to RA onset. Follow-up was until 2006. GC exposure was evaluated using four time dependent variables updated monthly: GC use, mean daily dose, total cumulative dose and total cumulative duration of use. CVA outcomes included acute thrombotic and hemorrhagic strokes, but not transient ischemic attacks, and were ascertained using ICD-9 codes (431, 434 and 436) from hospitalization and vital statistics data. To control for confounding by indication, wherein GC would be given to cases with more severe disease and/or less adverse cardiovascular profiles, we calculated propensity scores to control for the observed differences between GC users and non-users. We performed Cox-regression models to estimate the relative risk (RR) for CVA, adjusting for demographics, cardiovascular risk factors, RA characteristics, propensity scores and time-dependent prescription medications (methotrexate, Cox-2 inhibitors and other non-steroidal anti-inflammatory drugs).
The mean age of the cohort was 56 yrs (57 and 54 for GC users and non users respectively) and 66% were females (70% and 64% for GC users and non-users respectively). Over a mean of 6 years of follow-up (43,354 person-years), we identified 178 incident CVA cases (61 were fatal cases). The unadjusted stroke incidence rate was 4.1 per 1000 person-year in the RA cohort, 3.7 during GC exposure and 2.6 during non GC exposure. The directionality of the point estimates suggests a small non-significant effect in all adjusted models. The simplest time-dependent model that ignores dose and duration of exposure showed that GC use [yes/no] was not significantly associated with CVA (HR = 1.40, 95% CI; 0.83 2.37). Similarly, the model that accounts for current mean daily dose showed that current dose was not significantly associated with CVA (HR=1.01, 95% CI; 0.99 1.04, for each mg increase in the current mean daily dose). In addition, cumulative duration of of use and total past cumulative dose were not significantly associated with CVA (RR = 1.01, 95% CI; 0.99 1.02 per month of use, and RR= 1.001, 95% CI; 1.000 1.002 per each gram accumulated in the past respectively).
This large population-based study indicates that GC use is not significantly associated with an increased risk of CVA in cases with RA. Our results may have important implications for people with RA and their treating physicians, when weighting the risks and benefits of using GC to treat RA.
To cite this abstract, please use the following information:
Avina-Zubieta, Antonio, Abrahamowicz, Michal, Choi, Hyon K., Rahman, M. Mushfiqur, Sylvestre, Marie-Pierre, Esdaile, John, et al; Glucocorticoid Use Is Not Associated with an Increased Risk of Cerebrovascular Accidents in Patients with Rheumatoid Arthritis. A Population-Based Study. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1036