Letter by Palatini Regarding Article, “Habitual Coffee Consumption and Risk of Heart Failure: A Dose-Response Meta-Analysis”

To the Editor:

In a dose-response meta-analysis of prospective studies, Mostofsky et al¹ report a J-shaped relationship between coffee consumption and risk of heart failure. However, as the authors themselves point out, this study missed important information about potential modifiers of this association. Mostofsky et al¹ put emphasis only on caffeine when discussing the relationship of coffee with heart failure. However, coffee is a blend of a vast number of bioactive chemicals, with different effects on the cardiovascular system. Among these, the polyphenols chlorogenic acid and dihydrocaffeic acid seem to act as protective antioxidants and have other beneficial actions on the cardiovascular system.² The overall action of coffee on the cardiovascular system is regulated by the CYP1A2 gene.² Ninety-five percent of caffeine is detoxified through N³-demethylation catalyzed by CYP1A2.² However, this enzyme has a wide interindividual variability in activity, which is regulated by a genetic polymorphism. In subjects with the fast CYP1A2 *1A/*1A genotype, the detrimental cardiovascular effects of caffeine are negligible and outweighed by the favorable action of polyphenols or other bioactives.^2–5 In contrast, in carriers of CYP1A2*1F allele, who are slow caffeine metabolizers, the harmful effects of caffeine seem to prevail.^2–5 In a population study of 1107 individuals, we showed that coffee consumption had a positive nonlinear relationship with the risk of hypertension.³ However, in a more recent analysis of 553 participants in whom the CYP1A2 polymorphism was studied, we demonstrated that an increased risk of hypertension was present only among carriers of slow *1F allele, whereas among individuals with *1A/*1A genotype there was a reduction in risk.⁴ Similar results were found by Cornelis et al⁵ in patients with myocardial infarction. Also in this clinical setting, carriers of *1F allele had an increased risk of myocardial infarction with increasing coffee consumption, and subjects who were homozygous for *1A allele had a decrease in risk. These differential effects of coffee according to CYP1A2 genotype are likely to occur also in subjects at risk of heart failure. Thus, studying the average risk of coffee drinking in a population without taking into account the genetic background, as done in previous investigations, is of limited value, and the message may be even misleading. The distribution of CYP1A2 genotypes may vary in different populations,⁴ and the overall results of a study will largely depend on the frequency of carriers of the *1F allele in that population. This variability likely accounts for the conflicting results on the relationship between coffee use and heart failure found by different investigators.¹ At variance with the conclusion of Mostofsky et al,¹ carriers of CYP1A2*1F allele may be at increased risk of heart failure with increasing coffee consumption. In individual subjects, the CYP1A2 genotype should be determined to establish whether they can safely drink coffee. Future investigations on the cardiovascular effects of coffee should include CYP1A2 polymorphism determination, and the association of coffee consumption with heart failure should be studied in subjects stratified by CYP1A2 genotype.

Footnotes