Oral glucocorticoid therapy increases the risk of serious infection in elderly patients with rheumatoid arthritis, even when the drugs were stopped as long as 2.5 years earlier, according to findings from a recently published study.
The researchers used a novel, weighted cumulative dose model to assess serious infection risk associated with glucocorticoid exposure, providing greater accuracy over conventional models.
Conducted in Quebec, the case-control analysis assessed the infection risk in 16,207 patients who were at least 65 years of age and who had rheumatoid arthritis (RA). The findings showed that taking a 5 mg/day dose of prednisolone for 12 months increases risk of serious infection by 55%, compared to non-users. Furthermore, this risk was found to increase cumulatively to 100% (odds ratio, 2.00; 95% confidence interval, 1.69 to 2.26) in patients who took prednisolone for 3 years (Ann. Rheum. Dis. 2012 Jan. 12 [doi:10.1136/annrheumdis-2011-200702]).
"This is important as many physicians consider a 5 mg [daily prednisolone equivalent dosage] a ‘physiological dose’ that may not have important harms," according to Dr. William G. Dixon, Medical Research Council Clinician Scientist at the University of Manchester (England).
The investigators noted that 1,927 patients developed a first serious infection after an average 3.8 years of follow up. Patients who are currently or recently taking glucocorticoid therapy had the greatest risk of serious infection. The analysis showed that a current user on 5 mg/day prednisolone had an increased risk over non-users of 11% (OR, 1.11), 30% (OR, 1.30), and 55% (OR, 1.46) after 1 month, 3 months, and 12 months of use respectively.
"The risk associated with 5 mg prednisolone taken for the last 3 years was similar to that associated with 30 mg taken for the last month," noted Dr. Dixon and his associates.
The impact of discontinuation of glucocorticoid therapy was also investigated. Discontinuing a 2-year course of a 10-mg daily dose of prednisolone for 6 months was found to halve risk compared to continued use (OR, 1.87; CI, 1.32 to 2.63).
Dr. Dixon used a weighted, cumulative-dose model to assess the risk of serious infection associated with glucocorticoid exposure. This enabled doses to be given a weighting determined by how recently the patient took that dose. The authors note that this model "provides a far superior fit than all conventional models."
Dr. Jeffrey Curtis commented in an interview that the weighted cumulative dose model represents an advance. "The advance here is that there may be preferable ways to model steroid dose that allow for different estimations of risk, and presumably more accurate estimations. In the model used here, the researchers take into account the patients’ history of steroid therapy."
However, the study did not resolve the issue of confounding by indication, in which the effects of more active and severe disease can be difficult to disentangle from the effects of drugs, noted Dr. Curtis, who is director of the Arthritis Clinical Intervention Program Division of Clinical Immunology and Rheumatology at the University of Alabama at Birmingham.
"I think the risks with steroid therapy may produce different estimations of risk depending on the data source used, and the pattern of use in any particular disease. Analysis needs to be done by disease."
Addressing the issue of whether serious infection risk in patients with autoimmune diseases was associated with glucocorticoids, biologics, or non-biologic treatments, Dr. Curtis referred to a recently published paper he coauthored (JAMA 2011;306:2331-9). "We found a significantly increased risk of serious infection for steroid therapy but no greater risk overall for tumor necrosis factor (TNF)–alpha antagonists compared to patients receiving non-biologic treatments in a range of autoimmune diseases."
The study was funded by the Canadian Institutes of Health Research and the Canadian Foundation for Innovation. The lead author received support from a Medical Research Council Clinician Scientist Fellowship. The authors did not report having any other financial disclosures. Dr. Curtis is a consultant to various pharmaceutical companies that manufacture biologic therapies but he has no disclosures of relevance to glucocorticoids.