Introduction

Elderly users of nonsteroidal anti-inflammatory drugs (NSAIDs) who have a history of heart disease are at higher risk for developing congestive heart failure (CHF) than non-users, according to the results of an Australian study summarized here. Though the use NSAIDs and the development of congestive heart failure in susceptible individuals has been documented through experimental studies, few epidemiological studies have been done to demonstrate the impact of this effect.

Study Design

Using a matched, case-controlled design, researchers compared the relative risk of a first admission to the hospital with CHF in recent users and non-users of NSAIDs.
They also investigated whether the relative risk increased for patients with a history of heart disease and the extent that the level of risk varied with the NSAID dosage and half-life of the drug.

Three hundred sixty-five patients ('case' patients) who were admitted to one of two hospitals in New South Wales with a primary diagnosis of CHF were compared with 658 controls. Controls were admitted to the same hospital, were the same age (+/- 5 years) and sex, and did not exhibit clinical or radiological evidence of CHF. All subjects were interviewed within one week of admission to obtain information on all drugs used in the week and month prior to admission, including use of aspirin and other NSAIDs; recent and past history of heart disease; hypertension; diabetes; demographic details; alcohol consumption and smoking history.

Results

Most case patients had moderate to severe CHF. Case patients were slightly older than controls, despite matching age to within 5 years (76.6 years +/- standard deviation 7.62 vs. 75.1 +/- 7.14, p <.001). Cases were more likely than controls to have a history of heart (p < .001) and respiratory (p = .005) disease. Of those with a history of heart disease, 90% had experienced a myocardial infarction or angina pectoris. The prevalence of other diseases was similar between the two groups. As expected, cardiac drug use was different between the groups.

Of the 365 cases, 149 were admitted for first-time CHF. Most baseline characteristics for first-time CHF subjects and their matched controls were similar to the whole group. However, a history of heart disease, use of angiotensin-converting enzyme (ACE) inhibitors and use of high-ceiling diuretics was lower for first-time cases than for the entire group of cases.

NSAIDs in Cases and Controls

Consumption of non-aspirin NSAIDs in the week prior to admission was higher among cases (17.3%) than controls (12.0%).

• Use of NSAIDs (other than low-dose aspirin) in the week prior to admission was associated with doubling of the odds ratio (OR) of hospital admission for CHF (adjusted OR 2.1; 95% CI, 1.2-3.3). Use of any NSAID (including low-dose aspirin) in the week prior to hospitalization was associated with an adjusted OR for first admission with CHF of 2.8 (95% CI, 1.5-5.1). Adjustment was made for the following potential cofounders: history of heart, renal, respiratory and peripheral arterial disease; the use of beta-blockers, calcium channel antagonists, anti-diabetic drugs and oral glucocorticosteroids; and age.
• Among first-time cases with a history of heart disease, use of non-aspirin NSAIDs in the week prior to admission was associated with an OR of 10.5 (95% CI, 2.5-44.9). The corresponding adjusted OR in cases without a history of heart disease was 1.6 (95% CI, 0.7-3.7). The test for interaction with heart disease was statistically significant (p = .01).

Dose and Half-Life

A dose-response relationship was apparent in first-time cases, and the dose-response relationship varied significantly with history of heart disease (p = .002).

• The OR for patients without preexisting heart disease was a non-significant 1.2 (95% CI, 0.7-2.0). However, there was an OR of 7.5 per unit increase in the daily NSAID dose for patients with preexisting heart disease (95% CI, 2.0-27.6).
• In patients with first-episode CHF, there was a significant relationship between the OR and the published half-life for the NSAID used in the week prior to hospitalization (P = .004 for interaction of drug half-life and heart disease).

Use of any type of NSAID in the week prior to admission among first-time CHF cases was 29.5%. Assuming that the relationships described are causal, the researchers estimate that approximately 19% of new cases of CHF are due to consumption of NSAIDs.

Conclusion

The authors conclude that use of NSAIDs doubles the odds for elderly individuals being admitted to hospital with an episode of CHF. The risk increases substantially in those with a history of heart disease. The level of risk is positively associated with the dose of the NSAID in the week prior to admission, and is increased to a greater degree with drugs having a longer half-life.

They also note that the "burden of illness resulting from NSAID-related CHF may exceed that resulting from gastrointestinal tract damage." This contention is sufficiently serious that further studies are called for to confirm or deny it. In the meantime, NSAID drugs should be used with caution in elderly patients with a history of heart disease.