NEW YORK (Reuters Health) – In elderly patients with rheumatoid arthritis, treatment with tumor necrosis factor (TNF)-alpha antagonists may increase their risk of new-onset heart failure, and exacerbate pre-existing heart failure.
In a study reported in the American Heart Journal for August, anti-TNF treatment appeared to quadruple the risk of mortality due to established heart failure compared with methotrexate treatment.
Previous research among relatively young patients with rheumatoid arthritis indicated that TNF blockers may reduce the risk of heart failure, the investigators at Harvard Medical School note in their paper. Other studies, however, demonstrated potential harm to patients with symptomatic heart failure.
To clarify these issues, Dr. Soko Setoguchi and colleagues conducted a cohort study using information from Medicare databases for 1994-2004.
The cohort included patients age 65 and older who had at least one prescription for a TNF antagonist (etanercept or infliximab, n = 1002) or methotrexate (n = 5593) following a diagnosis of rheumatoid arthritis. Among those subjects, 1033 already had heart failure.
Compared to methotrexate treatment, TNF blocker therapy conferred an adjusted hazard ratio for hospitalization for heart failure of 1.70, regardless of heart failure history.
Among patients with pre-existing heart failure, the adjusted hazard ratio for death among those using TNF antagonists was 4.2.
The Boston-based team estimates that, for elderly patients, the number-needed-to-harm in terms of hospitalization for heart failure from anti-TNF treatment is 19 for those with prior heart failure and 105 for those without heart failure when treatment is initiated. The number-needed-to-harm for death could be as low as 5, among patients with pre-existing heart failure.
“Our results do not rule out the hypothesis that blocking TNF-alpha may reduce the risk of heart failure in younger patients with rheumatoid arthritis,” Dr. Setoguchi and her associates note. They suggest that there could be a differential effect of TNF-alpha in patients with and without cardiovascular conditions.
They call for “larger and more detailed studies” to more clearly define the risk-benefit ratio of TNF blockade among elderly patients with rheumatoid arthritis.
Am Heart J 2008;156:336-341.