点击显示 收起
Medical University of South Carolina
Laboratory of Drug Disposition and Pharmacogenetics
Department of Psychiatry, Room 246N
67 President Street
Charleston, SC 29425
E-mail: donovanj{at}musc.edu
Dear Sir:
In a prospective study of 38 445 women, Sesso et al (1) reported that flavonoid intake was not strongly associated with reduced mortality from cardiovascular disease (CVD). Significant inverse correlations with CVD were observed for broccoli, apples, and tea, although Sesso et al concluded that the observed relations were not mediated by flavonoids. I am concerned that the terminology used in their article was not specific enough and that their conclusions may be premature. It may be useful for readers of the Journal to properly understand the terminology used to describe the types of flavonoids in foods.
Hundreds of flavonoids are consumed regularly within the food supply. They are divided into subclasses with different physiochemical and biological properties. There are 5 major subclasses of flavonoids: flavonols, flavones, catechins (flavanols), anthocyanins, and flavanones (2). Sesso et al determined the dietary intakes of 2 of the major flavonoid subclasses: flavonols and flavones. Although these 2 subclasses have received more attention than have the other subclasses because of reports of specific biological activities, these 2 subclasses are the least abundant subclasses of flavonoids in the diet (3). Flavonols are ubiquitous in fruit and vegetables but are present in fairly low concentrations in common foods, with the notable exception of yellow onions. Flavones are abundant in only a few select foods, such as parsley, celery, and certain types of sweet peppers.
Sesso et al concluded that the inverse relations between CVD and the consumption of broccoli, apples, and tea were not due to the flavonoids present in these foods (1). Although broccoli, apples, and tea are sources of at least one of the subclasses measured in the study, apples and tea contain significantly higher amounts of catechins than of flavonols or flavones (Table 1). Catechins and other important subclasses of flavonoids were not measured in Sesso et al’s study, so the conclusion that the inverse relations observed for tea and apples were not mediated by flavonoids may be incorrect. In a prospective cohort study, Arts et al (6) showed that the intake of monomeric catechins is inversely associated with ischemic heart disease mortality, and the major dietary sources of catechins in that study were tea and apples. Dietary intervention studies indicate that catechins modulate platelet activity and other risk factors for CVD (7, 8).
View this table:
TABLE 1. Flavonoid concentrations in foods that have been shown to be inversely correlated with cardiovascular disease mortality in women1
Sesso et al also concluded that "total flavonoid" intake was not related to the incidence of CVD in women. Although the term total flavonoids was defined in the Subjects and Methods section as the sum of the major compounds from the flavonol and flavone subclasses, the term was also used in the abstract and throughout the text. Because the Journal attracts a diverse group of readers in all areas of nutrition and the health sciences, it is possible that some readers will take the use of the term total flavonoids literally. I am not aware of any studies that have measured the consumption of total flavonoids including all subclasses. On the basis of the amounts of flavonoids in common foods, the average dietary intake of total flavonoids from a typical Western diet was estimated to be as high as 650 mg/d (3). Although total flavonoid intake may arguably be much lower than 650 mg/d in the United States, Sesso et al based their conclusions on mean consumption values of only 25 mg/d, a small proportion of the intake of total flavonoids.
Because this important research is likely to stimulate further investigation, it is necessary for readers to understand that Sesso et al have shown that the intake of flavonols and flavones is not correlated with the incidence of CVD in women. We do not know whether other abundant flavonoid subclasses have effects or whether there are additive effects of consuming several or all flavonoid subclasses together. The inverse relations for tea and apples that were observed in the study by Sesso et al may have been due to catechins, a subclass of flavonoids that were not measured. Future epidemiologic reports must be specific when referring only to particular flavonoid subclasses to avoid the possibility of misleading readers who are not familiar with this diverse group of biologically active compounds.
REFERENCES