Correspondence

Acetaminophen, Aspirin, and Renal Failure

N Engl J Med 2002; 346:1588-1589May 16, 2002

Article

To the Editor:

There is strong experimental and epidemiologic evidence that the use of acetaminophen or aspirin is associated with a very small risk of analgesic nephropathy. However, only extensive and uncontrolled consumption has been proved to be dangerous. Because there is a higher risk of acute renal impairment associated with cyclooxygenase inhibition, current practice is to ban the use of nonsteroidal antiinflammatory drugs (including high-dose aspirin) in patients with chronic renal failure and to recommend instead acetaminophen or low-dose aspirin.

The case–control study by Fored et al. (Dec. 20 issue)1 apparently contradicts this view, but the data raise the question of recruitment bias. Because patients had quite advanced chronic renal failure, it is likely that they and their physicians had known about the underlying renal disease for years. During this “lag time,” it is likely that these patients used more aspirin for the prevention of cardiovascular events and used fewer nonsteroidal antiinflammatory drugs (and more acetaminophen) for pain and fever than the general population; such a treatment history would be a powerful confounding factor. Without correction for the timing of the diagnosis of renal disease and for coexisting conditions (which would be possible only in follow-up studies), the plausible hypothesis that the use of acetaminophen or aspirin is harmful in patients with renal failure remains unproved.

Andrea Campo, M.D.
S. Lazzaro Hospital, 12051 Alba, Italy
andrea_campo@hotmail.com

1 References

1
Fored CM, Ejerblad E, Lindblad P, et al. Acetaminophen, aspirin, and chronic renal failure. N Engl J Med 2001;345:1801-1808
Full Text | Web of Science | Medline

To the Editor:

Fored et al. found that the regular use of aspirin was associated with an increased cumulative risk of chronic renal failure from any cause. Their tables and bar graph show an increase in risk even with lifetime use in the range of 1 to 99 g. It is common practice for doctors to advise their patients to take (and to take themselves) one aspirin daily for decades for reasons unrelated to renal conditions. Even if patients take a “baby aspirin” (81 mg) daily, 1 g will be accumulated in 12.35 days, almost 30 g in one year, and 500 g in just under 17 years. This level of aspirin consumption places the user in a high-risk group, according to Figure 1 in the article by Fored et al. Persons who take one 325-mg tablet daily will accumulate aspirin four times as fast. Do the authors of this report believe we should continue to advise many of our patients to take one aspirin (whether “baby” or adult-sized) daily?

William Thurlow, M.D.
Summerside Medical Center, Summerside, PE C1N 3N9, Canada
wthurlow@isn.net

Author/Editor Response

The authors reply:

To the Editor: Contrary to Dr. Campo's assertion, protopathic bias — the possibility that the renal disease or its antecedents were responsible for the observed pattern of analgesic use — was our primary concern in the planning, analysis, and interpretation of our study. The main reason we chose to study risks in patients with early-stage disease was the assumption that drug use was less likely to be affected in these patients than in those with end-stage disease. It was, however, evident that the patients with chronic renal failure had more aches and pains than the disease-free controls. Although protopathic bias was not completely ruled out, we concluded that the observed excess risks could not be entirely explained by this bias.

First, lagged analyses, in which analgesic consumption during the most recent decade was disregarded, showed essentially the same relative risks as our primary analysis. Second, on direct questioning, few of the patients with chronic renal failure reported having received any advice about changing their use of analgesics, and reanalysis excluding the patients who had received such advice did not materially alter our results. Third, similar associations with analgesic use were found among patients with insidious renal failure that had not had clinical antecedents.

Dr. Thurlow asks whether we need to reconsider the practice of prescribing low-dose aspirin prophylactically. This question is best addressed by comparing the absolute effect of aspirin use on the risk of chronic renal failure with its absolute effect on the risk of major cardiovascular events. Since the size of our study base was known, we could estimate the sizes of our risk strata on the basis of the proportion of users among our controls. Since virtually all new cases in subjects in the study base were ascertained and the subjects could be assigned to a risk stratum, we were able to estimate the incidence in each risk stratum. Then we could compute the difference in incidence as a measurement of the excess incidence among the subjects with exposure to aspirin.

Although there is a small possibility of overestimation due to protopathic bias, we estimated that the excess among regular users of aspirin was 8.9 new cases of chronic renal failure per 100,000 person-years. In comparison, 150 myocardial infarctions were prevented by aspirin therapy during the same number of person-years, according to a recent meta-analysis of randomized, controlled trials.1 Hence, although some of the studies in that meta-analysis included subjects with hypertension and an increased base-line risk, it appears that the cardiovascular benefits of low-dose aspirin therapy far outweigh the renal hazards. Therefore, we see no good reason for changing the current practice of recommending low-dose aspirin to patients at risk for cardiovascular disease.

C. Michael Fored, M.D.
Olof Nyrén, M.D., Ph.D.
Karolinska Institutet, SE-171 77 Stockholm, Sweden
michael.fored@mep.ki.se

1 References

1
Sanmuganathan PS, Ghahramani P, Jackson PR, Wallis EJ, Ramsay LE. Aspirin for primary prevention of coronary heart disease: safety and absolute benefit related to coronary risk derived from meta-analysis of randomised trials. Heart 2001;85:265-271
CrossRef | Web of Science | Medline

Citing Articles (1)