American Journal of Clinical Nutrition, Vol. 72, No. 4, 1061-1062, October 2000
© 2000 American Society for Clinical Nutrition

--------------------------------------------------------------------------------

Letter to the Editor

Reply to AE Hardman
Walter C Willett
Department of Nutrition, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115

Dear Sir:

Hardman's effort to expand the discussion of interactions between diet and exercise on blood lipids is appreciated. As I had commented, and Hardman agrees, there is no specific evidence to support Gibney's assertion that the HDL-lowering, triacylglycerol-raising effects of high-carbohydrate diets are totally negated by 30 min of brisk walking daily (1, 2). Hardman noted correctly that there are some data that bear indirectly on this issue. However, it is first important to clarify the question. Was Gibney implying that the effects of diet and exercise (30 min of walking) are simply additive (but in opposite directions), or that diet has no effect in combination with exercise (an interaction)? Gibney was unclear, but the latter seems to have been implied because, in an additive model, there would still be an effect of diet in persons who exercise, even if it were counterbalanced by the effect of exercise. In other words, the additive model would predict that the most favorable HDL and triacylglycerol concentrations would be seen in persons who exercised and did not consume high-carbohydrate, low-fat diets.

Hardman cited 2 earlier studies by Wood et al (2, 3) but, surprisingly, omitted more recent work of those authors (1) that is more informative because it included 4 groups (exercise alone, diet alone, both exercise and diet, and neither diet nor exercise). This study thus allowed the independent and interactive effects of diet and exercise to be estimated. As in the earlier studies by Wood et al, the exercise program included both brisk walking and jogging, which is probably somewhat more aerobic than is brisk walking alone. Also, the dietary intervention [National Cholesterol Education Program (NECP) Step II diet] was not purely a reduction in percentage of energy from fat, because the fatty acid composition was changed and the intervention included a modest weight reduction. Probably because of the reduction in weight, fat reduction and weight loss had a very small net effect on triacylglycerol [i.e. tryglycerides] in women and reduced triacylglycerol in men. Nevertheless, the relative effects of diet on blood lipids in the exercise and nonexercise groups can be evaluated. The findings related to triacylglycerol and HDL cholesterol are shown in Table 1.

TABLE 1. Effects of a high-carbohydrate, low-fat (high-CHO) diet and aerobic exercise on plasma triacylglycerol and HDL-cholesterol concentration

These data do not provide any support for Gibney's assertion that the effects of a high-carbohydrate, low-fat diet are less adverse in persons who exercise. In fact, in both men and women, the effect of this diet on triacylglycerol was relatively more adverse (ie, a more positive value) in persons who exercised than in those who did not. Similarly, in both sexes the effect of diet on HDL cholesterol was more adverse (ie, a more negative value) in persons who exercised than in those who did not. Importantly, in each case, the most favorable lipid values were in persons who exercised and did not consume the high-carbohydrate, low-fat diet. Many of these differences were small and would not have been statistically significant, but the consistent lack of a lesser effect of diet in people who exercise is notable. Although LDL cholesterol was lowest in the diet-plus-exercise group, because of the adverse effect of the diet on HDL cholesterol, in neither men nor women did the combination of diet plus exercise have a significant effect on the ratio of total to HDL cholesterol, which provides the best overall prediction of the risk of coronary heart disease.
Gibney and Hardman should not be satisfied that exercise merely compensates for the adverse effects of low-fat diets on blood lipids, because most people would not be happy to learn that, despite a major change in diet, regular aerobic exercise, and modest weight loss, they had not measurably altered their CHD risk. From numerous metabolic studies (4) we know that replacement of saturated and trans fats with nonhydrogenated unsaturated fats reduces LDL cholesterol without increasing triacylglycerol or reducing HDL cholesterol. Although further investigations of exercise in combination with diet are needed, the effects of this dietary change and exercise on blood lipids would both be in the beneficial direction, so the combination should provide the greatest benefit.


REFERENCES


Stefanick ML, Mackey S, Sheehan M, Ellsworth N, Haskell WL, Wood PD. Effects of diet and exercise in men and postmenopausal women with low levels of HDL cholesterol and high levels of LDL cholesterol. N Engl J Med 1998;339:12–20.[Abstract/Full Text]
Wood PD, Stefanick ML, Williams PT, Haskell WL. The effects on plasma lipoproteins of a prudent weight-reducing diet, with or without exercise, in overweight men and women. N Engl J Med 1991; 325:461–6.[Abstract]
Wood PD, Terry RB, Haskell WL. Metabolism of substrates: diet, lipoprotein metabolism and exercise. Fed Proc 1985;44:358–63.[Medline]
Mensink RP, Katan MB. Effect of dietary fatty acids on serum lipids and lipoproteins: a meta-analysis of 27 trials. Arterioscler Thromb1992;12:911–9.[Abstract]

Here's video of talk Dr. Willett gave in May 2007 at Univ of Wisconsin:

http://videos.med.wisc.edu/videoInfo.php?videoid=252