The effect of low-dose daily aspirin intake on survival in the Finnish centenarians cohort.

Abstract

To the Editor: The benefit of using daily low-dose aspirin therapy in secondary prevention of cardiovascular morbidity and short-term mortality in patients with vascular diseases is well established in both middle-aged and older subjects.1 In contrast, the effect of low-dose aspirin on long-term survival in the general population is less certain. Although a good effect has been found in middle-aged patients,2 results from studies carried out in older subjects are still controversial.3, 4 Cardiovascular morbidity and mortality increase with increasing age, making cardiovascular diseases the main causes of death in older people; consequently, older subjects affected by these conditions may achieve substantial benefit from a prophylactic intervention. However, older subjects have decreased drug metabolism, greater tendency to develop side effects, and more comorbid illnesses that can interfere with pharmacological treatment than do younger subjects. Therefore, older patients are less likely than younger patients to have aspirin prescribed for prophylaxis.5 Centenarians, as examples of extreme longevity, are important natural models of age-related physiological changes. Thus, this study gives a unique chance to evaluate the effect of low-dose aspirin intake on survival, even in very advanced age. We have used data from a population-based survey of subjects ≥100 years, the “Finnish Centenarian Study.”6 Of the 224 centenarians who were identified from the Finnish National Population Register in 1991 from the whole of Finland, 39 died before clinical assessment and 6 refused examination or blood testing. Survival information for 179 centenarians (28 men and 151 women) was continuously updated from the Population Register during the subsequent 5 years. Participants, caregivers, and healthcare personnel were asked about all medications. Moreover, prescriptions and containers were inspected to confirm the information. Thirty subjects (17%) were taking low-dose aspirin daily during at least 3 months before assessment. The usual dose taken was 250 mg. Aspirin use on a need basis, as an analgesic or antipyretic medication, was not included in this definition. Kaplan-Meier plots and log rank tests were used to describe patterns of survival over 5 years of follow-up in aspirin users and nonusers. Cox proportional hazard models were constructed to quantify the association between aspirin use and 5-year mortality, after controlling for age, gender, documented history of cardiovascular disease and stroke, use of cardiovascular medicines, and cardiovascular findings. The cumulative 5-year mortality of this very old population was 98%. Subjects taking aspirin had an average survival of 2.4 years (95% confidence interval (CI) = 1.8–3.0), whereas subjects who did not take aspirin survived 1.7 years (95% CI = 1.4–1.8) (Figure 1). Kaplan-Meier survival plots comparing aspirin users versus nonaspirin users. Five years of follow-up of the Finnish centenarians study. Multiple logistic regression analyses indicated that younger age (odds ratio (OR) = 0.5, 95% CI = 0.4–0.8) and current use of cardiovascular drugs (OR = 3.2, 95% CI = 1.1–8.9) were the only factors associated with current use of aspirin. Because use of aspirin was related to younger age, Cox regression models were constructed for the whole population and for two age groups (100 years and> 100 years). The adjusted relative risk of 5-year mortality for aspirin users was 0.6 (95% CI = 0.3–0.9) in the whole population, 0.4 (95% CI = 0.2–0.8) in the younger age group, and 0.5 (95% CI = 0.3–1.0) in the older age group. Our findings indicate that centenarians who use low-dose aspirin daily in general tend to live longer, even after controlling for sociodemographic characteristics and health status, suggesting that there is an association between daily aspirin intake and long-term survival in extreme old ages. Several reasons may account for this finding. First, it may be that aspirin users were in better health. However, when we compared aspirin users and nonusers regarding health variables, we found that aspirin users had a significantly more extensive history of cardiovascular diseases (P < .05). Second, it may be that aspirin use was a reflection of other factors associated with longer survival, such as life-style habits, socioeconomic status, and education. Nevertheless, this population was quite homogeneous with regard to its sociodemographic characteristics.6 Third, the lower mortality associated with aspirin therapy may imply a real protective effect of aspirin. The main limitation of this study is the possible underestimation of the aspirin use, which should not, however, affect the main conclusion. Because these findings are derived from an observational study, clinical trials are needed to confirm our findings and to explore further the hypothesis that low-dose aspirin may increase longevity among older subjects.