| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTER TO THE EDITOR |
Children’s Hospital Oakland Research Institute
5700 Martin Luther King, Jr Way
Oakland, CA 94609
Life Science Division
Lawrence Berkeley National Laboratory
Berkeley, CA
E-mail: rkrauss{at}chori.org
Life Science Division
Lawrence Berkeley National Laboratory
Berkeley, CA
Dear Sir:
Equivalent amounts of weight loss achieved by various interventions, including diets with high carbohydrate compared with diets with high monounsaturated fat (1) and physical activity compared with caloric restriction (2), have been shown to result in similar changes in LDL particle profiles, which were a major focus of our study (3). The studies referred to by Westman et al (4-6) failed, as did other representative reports that we cited—including one coauthored by an author of the letter from Westman et al (7)—to distinguish the effects of carbohydrate restriction on lipoprotein concentrations from those of the weight loss observed on such diets. A major conclusion from our study, and one that could not be drawn from earlier studies, is that carbohydrate limitation without weight loss can produce similar results.
We are at a loss to understand the concerns of Westman et al with respect to the data presentation and interpretations in our report (1). In that article, Table 2 presents all of the mean changes in weight and lipoproteins, including ANOVA, for overall significance and between-group comparisons using Scheffe’s approach for multiple group comparisons. The table clearly shows that, compared with men who remained on the 54% carbohydrate diet, those who switched to a 26% carbohydrate diet reduced total weight, total cholesterol, apolipoprotein B, triacylglycerols, and the ratio of total to HDL cholesterol and increased the LDL peak particle diameter. Although a legitimate dialogue could be conducted as to whether Scheffe’s approach is too restrictive, we have chosen a conservative approach. The claim by Westman et al that HDL cholesterol was "increased significantly by weight loss after carbohydrate restriction" appears to be based on their interpretations of the within-group mean ± SE changes, which, although suggestive of a beneficial change with caloric restriction, did not attain statistical significance for our a piori hypothesis of between-group differences. In a randomized controlled clinical trial, the strength of these findings is based on group differences among men randomly assigned to different carbohydrate intakes. Although we agree that the benefits of reduced carbohydrate intake are suggested by these within-group HDL differences, we are required to point out they were not significant at the conventional threshold of P < 0.05.
It is true that, adhering to the a priori hypotheses, we did not comment directly on comparisons of LDL particles between the high-carbohydrate diet and the low-carbohydrate diet that was higher in saturated fat. However, we showed in both Table 2 and Figure 2 and clearly pointed out in the text that the increase in LDL peak particle diameter and the related conversion of LDL subclass pattern B to A induced by the low-carbohydrate diet was the same on the high- and low-saturated fat diets. Nevertheless, neither the earlier studies cited nor ours points to "the substantial advantage of low carbohydrate ... over low fat" if comparable amounts of weight loss are achieved.
The "tired concerns" regarding nutritional adequacy of some very-low-carbohydrate diets and the evidence for benefits of diets high in fiber and of increased physical activity are hardly unscientific. It is critical to base dietary guidelines on a careful and balanced evaluation of the evidence. In this regard, one may expect that future guidelines will reflect growing evidence that diets high in carbohydrate, particularly refined carbohydrates (8), can increase the risk of cardiovascular disease and that the substitution of fat, and perhaps protein, for carbohydrates, along with an avoidance of caloric excess, can reduce this risk.
ACKNOWLEDGMENTS
Supported by a grant from the National Dairy Council.
None of the authors had a personal or financial conflict of interest.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
