| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Brigham and Women's Hospital, Harvard Medical School, Division of Preventive Medicine, 900 Commonwealth Avenue East, Boston, MA 02215, E-mail: simin.liu{at}channing.harvard.edu
Department of Nutrition, Harvard School of Public Health, Boston, MA 02215
Dear Sir:
We appreciate Katz's interest in our article. He suggested that because women who had coronary heart disease (CHD), angina, or overt CHD risk factors may have changed their diet to a high-carbohydrate one, the positive association between dietary glycemic load and CHD risk observed in our study may be spurious because of confounding by these high-risk conditions. This concern, although theoretically possible, was not supported by findings in the Nurses' Health Study. First, higher intake of dietary carbohydrate, total starch, or total grain was not significantly associated with higher CHD risk (1, 2); neither was intake of total fat (3). However, higher intake of trans fat (3) as well as lower intakes of polyunsaturated and monounsaturated fats (3), whole grains (1), and fruit and vegetables (4) were significantly associated with higher CHD risk. Second, nurses who had CHD, angina, or diabetes at baseline were excluded from the main analyses and higher dietary glycemic load was not associated with a less favorable CHD risk profile (Table 1 in reference 2). Third, the positive association between dietary glycemic load and CHD risk became even stronger after adjustment for conventional risk factors, arguing against the possibility of residual confounding.
Our findings that the types of carbohydrate as measured by the glycemic indexes are important for predicting CHD risk are consistent with an inverse relation between dietary fiber intake and CHD risk that we (5) and others (6) reported previously. Incorporating the concept of glycemic index in our assessment of the physiologic effect of carbohydrates enabled us to measure both the total amount of carbohydrate and the quality of carbohydrate intake. As shown in our article, the positive association between dietary glycemic load and CHD risk was independent of dietary fiber intake and was particularly strong in overweight women. This finding is supported by metabolic data showing that the adverse metabolic effects of high carbohydrate intake on blood HDL-cholesterol and triacylglycerol concentrations depend directly on the degree of insulin resistance, which is largely determined by excess body fat (7).
Although a prevalent nutritional recommendation has been that a low-fat, high-carbohydrate diet can prevent heart disease, few empirical data support such a recommendation (6). Results from the Nurses' Health Study add to the growing body of evidence suggesting that neither total fat nor total carbohydrate in the range typically eaten is related to CHD risk and that a more complex picture exists relating to different types of fats and carbohydrates. As in any field of scientific pursuit, new data often generate new hypotheses to be tested. Future investigations are thus needed to confirm our findings in high-quality prospective studies of different populations in whom the range and nature of carbohydrate intake and the degree of insulin resistance may be different. We do not advocate that dietary recommendations be based on epidemiologic data alone but do note that our findings are consistent with results from metabolic studies, including those cited by Katz. Additional basic and experimental studies are also warranted to achieve an understanding of how different types of carbohydrates may affect CHD risk. For example, why does a high-carbohydrate diet typically produce high plasma triacylglycerol and low HDL-cholesterol concentrations characteristic of the insulin resistance syndrome? How do different types of carbohydrates affect insulin and other hormonal responses, particularly in individuals who are already prone to insulin resistance? What other hemodynamic or inflammatory markers may also be related to such a diet?
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
