Nutrition Research, Vol. 20, No. 12, pp. 1707-1714, 2000 Copyright © 2000 Elsevier Science Inc. Printed in the USA. All rights reserved 0271-5317/00IS-see front matter , :~. ;i I ]ELSEVIER Plh S0271-5317(00)00276-1 PHYSICAL EXERCISE REVERSES DIET-INDUCED INCREASES IN LDL-CHOLESTEROL AND APO B LEVELS IN HEALTHY OVO-LACTOVEGETARIAN SUBJECTS Delgado M, Gonzhlez-Gross M, Cano MD, Guti6rrez A, Castillo MJ.I Departments of Physiology (School of Medicine), Physical Education (School of Sports Sciences), and Clinical Biochemistry (University Hospital), University of Granada, GRANADA, Spain. ABSTRACT The aim of this study was to investigate the effect of supplementing the daily intake of whole-fat dairy products with 750ml of milk and 250ml of yoghurt, and the influence of a concomitant program of physical exercise (30-45 min of running/cycling at 60% VO2 max every two-days) in young healthy ovolactovegetarian (n=14) and non-vegetarian control (n=10) athletes. Diet manipulation for a two-month period increased LDL-cholesterol (2.31±0.44 vs 2.51±0.42 mmol/1) and apo B (68±16 vs 75±15 mg/dl) levels as well as LDL/HDL-cholesterol (1.65±0.47 vs 1.94±0.68) and apo B/apo A1 (0.46-~.15 vs 0.66±0.30) ratios (all p<0.05), in vegetarian but not in nonvegetarian subjects. After two additional months of enhanced physical activity, in vegetarians, apo B levels and LDL-/HDL-cholesterol ratio returned to baseline values, and HDLcholesterol (1.42--~0.36 vs 1.76±0.31 mmol/i) significantly increased. The results show that increasing the intake of whole-fat dairy products impairs the plasma lipid profile in ovo-lactovegetarian subjects. This effect can be reversed by increased regular physical activity. © 2000ElsewerScienceInc KEY WORDS: Dairy products, Physical exercise, Vegetarian diet, Plasma lipids. INTRODUCTION Diet and physical exercise interact in the development and prevention of ischemic heart disease [1-4]. Both factors affect the plasma lipid profile [5,6]. The most potent dietary influence comes from cholesterol and fats rich in saturated fatty acids. This is well documented by crosscultural comparisons of saturated fat/cholesterol intakes, plasma cholesterol concentration, and ischemic heart disease rates [7]. Other components of the diet, such as soluble fiber [8-10], source of proteins [11] and other factors [4,5] may also be of influence. Consequently, there is a place for investigations on the effects of particular foods (not only some of their components) on risk factors for chronic diseases [5,12]. 1 Corresponding author: Prof Dr. M.J. CastiUo. Departamento de Fisiologia. Facultad de Medicina. Universidad de Granada. 18071 - Granada. Spain. Phone: +34 958 243540. Fax: +34 958 246179. E-mail: mcgarzon~gr.es 1707 1708 M. DELGADO et al. Vegetarian subjects present low fat and low animal protein intakes. For most of them, dairy products represent a major source of cholesterol, saturated fat and animal protein in their diet [ 13,14]. In fact, in vegetarian subjects, LDL-eholesterol concentrations are correlated with dairy products and eggs consumption [15]. In these subjects, an increase in the intake of whole-fat dairy products may have a negative effect on the plasma lipid profile. A physiological means of influencing diet-induced modifications of the plasma lipid profile is physical exercise. Physical activity favorably influences all three components of the atherogenic lipoprotein phenotype: the high density lipoprotein (HDL)-cholesterol concentration may increase, LDL-cholesterol may decrease, and serum triglycerides can also be reduced [4,6]. Therefore, the effects of dietary-induced hypercholesterolemia may be attenuated by endurance exercise [6,16]. The aim of this study was to investigate in healthy ovo-lactovegetarian subjects a) if an increase in whole-fat dairy products intake effectively modified the plasma lipid profile and b) if tiffs effect could be reversed by physical exercise. MATERIALS AND METHODS Subjects: Fourteen healthy ovo-lactovegetarian men were studied. Their age ranged between 18 and 25 years. They all were students or recent graduates in Physical Education. Informed written consent was obtained from all participants and the Ethical Committee for human investigations of our Institution approved the study. The diet of these subjects included dairy products and eggs but not other animal products. All subjects supplied a personal and familiar medical history devoid of metabolic or cardiovascular disease. For comparison, a matched control group of ten healthy nonvegetarians was also studied. Procedures: After a baseline study, all subjects from both groups accepted to have, for the next four months, an extra daily intake of whole fat milk (750ml) and yogurt (250ml). No specific recommendations were given concerning other nutrients. After two months, the subjects were evaluated and then an exercise program was proposed for the next two months. The exercise program had to be practiced in addition to their daily physical and training activities, and consisted of a session of aerobic exercise performed once every two days. Each session consisted in running or cycling for 30-45 rain at a minimal intensity of 60% maximal oxygen consumption (VO2 ~,). These training sessions were done in the laboratory, under the presence of one of the investigators. At the end of these two months the subjects were re-evaluated. Evaluations consisted of a) control of food intake, b) estimated energy expenditure, c) anthropometric measurements, d) assessment of aerobic capacity and e) analysis of the plasma lipid profile. a) The control of food intake was evaluated along the study by two different means: 1. Diary. In this diary, the subjects kept a 7-day dietary record during the week before the study (baseline), 8th week (diet) and 16th week (diet + exercise). Previously, they were instructed how to calculate in grams the food portions. These data were introduced in the computer program, analyzed and presented in a dally base. 2. Questionnaires. At entrance, the subjects completed a questionnaire indicating their dietary habits. In addition, during the study, another questionnaire was completed once a week, indicating the food consumed in the previous 24 hours with particular emphasis to dairy products. This intended to check if they were fulfilling the prescribed protocol. The dietary records were analyzed by a computer program and nutrient database adapted to the local diet [ 17,18]. Throughout the study period, weekly personal interviews were held with each subject to reinforce compliance. The increases in dairy products intake and physical activity were well tolerated by all subjects. b) The quantification of physical exercise as energy expenditure was done by two different means: 1. Diary. In this diary, the subjects wrote down all the physical and sport activities performed BLOOD LIPIDS IN VEGETARIANS 1709 before beginning with the study (baseline), during the 8th week (diet) and during the 16th week (diet + exercise). 2. Questionnaires. At entrance, the subjects completed a questionnaire indicating their usual physical exercise pattern. In addition, during the study, another questionnaire was completed once a week, indicating the physical and sport activities during the previous 24 hours, including hours of slecla, hours of sitting (or no physical exercise) and hours of slight, moderate or high physical activity. This information was processed in the computer program that uses as database the energy expenditures of industrial, technical and leisure time activities [ 19]. Although some comments have been made about the use of these values [20,21], they can be considered still up to date. c) Anthropometric measurements included weight, height, skinfolds thickness, and circumference of arm and leg. Body composition was also determined by near-infrared technology (Futrex 5000). d) Aerobic capacity was determined using a Med-Graphics Cardiopulmonary Exercise System CPX (St Paul, MN). e) Blood samples for the plasma lipid profile analysis were drawn in the morning after a fast of 12 to 15 h. Plasma cholesterol and triglycerides were measured by enzymatic methods adapted to a Technicon Dax 72 Autoanalyzer (Terrytown, NY). HDL-cholesterol was determined after manganese and heparin precipitation of low and very low density lipoproteins. LDL-cholesterol was estimated by the Friedewald formula [22,23]. Apolipoprotein B and A1 levels were measured by immunonephelometry (Beckman Immunochemistry System, Brea, CA). The interassay coefficients of variation for these two assays were, respectively, 4.2% and 4.8% [24]. All the assays were performed in the General Clinical Laboratory of the University Hospital. No information was provided to the laboratory concerning the characteristics and eating patterns of the subjects. The Regional Health Authority guaranteed quality control procedures. Time-related variations in the assays were excluded by daily analysis of control sera. Statistics._ One-way analysis of variance was used to evaluate overall group differences. The Student-t test for either paired or non-paired data was employed to compare pairs of means. Results are expressed as mean+SD, p<0.05 was considered significant. Table 1. Mean ± SD daily intakes of several dietary components evaluated at baseline, after two months of supplemented dairy product intake (Diet) and after two additionalmonths of enhanced training (Diet+Exercise),in healthy ovo-lactovegetarian(n=14) and non-vegetarian(n=10) control subjects. Energy intake (kcal) Proteins (g) Carbohydrates(g) Fat (g) Saturated (g) Monounsaturated (g) Polyunsaturated(g) P/S ratio Cholesterol (g) Fiber (g) Ovo-lactovegetariansubjects BASELINE DIET DIET + EXERCISE 3035+754 3461+798' 3542_+569" 84±25 120±23' 114±17" 413±126 430+150 439+111 103-+34 142±39" 150±27" 31±14 63±20" 70±14" 51±17 57+19 58+14 20+11 22±13 22-+11 0.67±0.36 0.37-+0.24" 0.33+0.25" 257-+106 423+116' 441-+82" 22_+9 24±8 26±10 Non-vegetarian control subjects BASELINE DIET DIET ÷ EXERCISE 3598_+900 3860+894 3941_+866 124+25§ 145+17'§ 154±18"§ 441-+113 425+111' 432+120 164±58§ 1 7 5 ± 4 8 173±51 54±25 § 65+22' 75+22" 86±36 § 90-+32§ 75+37 22+7 21+6 23±6 0.44+0.13 0.34+0.10" 0.33+0.13" 839_+557 § 906+355~i 895±468§ 17±9 16_+6 16±8 * p<0.05 vs Baseline. 11p<0.05 vs Diet. §p<0.05 vs ovo-lactovegetariansubjects. 171 0 M. DELGADO et al. RESULTS Daily intakes of different dietary components for the two groups of subjects are given in Table 1. At baseline, ovo-lactovegetarian subjects presented lower protein, fat (saturated and monounsaturated) and cholesterol intake than non-vegetarian control subjects. The ratio polyunsaturated/saturated (P/S) fat tended to be (p=0.07) higher in the vegetarian, group. The diet manipulation increased the intakes of protein and saturated fat in both groups. P/S ratio decreased in both groups. Cholesterol intake increased significantly only in the vegetarian group, although it did not reach the high intakes of the non-vegetarian group. BMI was similar in both groups but percent of fat body mass and percent of lean body mass were respectively lower and higher in the vegetarian group (Table 2). After dietary manipulation, body weight increased slightly in the non-vegetarian control group but not in the vegetarian group. After the period of enhanced physical activity, a decrease in fat body mass was observed in both groups. As could be expected, energy expenditure increased, although slightly, during this enhanced training period. Physical performance was similar in both groups and was not modified during the study. A slight increase in maximal aerobic power could be observed at the end of the study for the non-vegetarian group (Table 2). Table 2. Mean +_SD daily physical characteristics evaluated at baseline, after two months of supplemented dairy product intake (Diet) and after two additional months of enhanced training (Diet+Exercise), in healthy ovolactovegetarian (n=14) and non-vegetarian (n=10) control subjects. Height was similar in both groups (179±7 vs 177+_5cm). MAP = maximal aerobic power. Ovo-lactovegetarian subjects BASELINE DIET DIET + EXERCISE B.M.I. Body weight (kg) 22+0.5 69.3±7.9 22+0.5 22+0.5 69.5+_7.8 69.2+7.9 Fat body mass (%) 10.0±0.6 10.0_+0.5 Lean body mass (%) 49.8±2.4 50.5+-2.9 50.4+-3.0 Non-vegetarian control subjects BASELINE DIET DIET + EXERCISE 23+0.9 23±0.6 69.6+_7.2 70.6±7.5" 9.9-.0.611 10.7+-1.2§ 10.4-.1.0 55+6 293±37 56+_5 293-*37 56-*6 300-*46 70.1+7.7 10.3-.1.3' 47.2±1.1§ 48.3+_1.2§ 48.8±1.3 Energy expenditure (kcal/day) 4138+_491 4047+390 4231+_4901 4110+_389 4080±368 VO2,,~, (mL.kg-l.min 1) MAP (watts) 23±0.6 51_+8 270±42 50+7 270±42 4177+337 "1 53-*9 290-*46'11 ' p<0.05 vs Baseline. 11p<0.05 vs Diet. § p<0.05 vs ovo-lactovegetarian subjects. The plasma lipid profile during the different study phases for both groups of subjects are given in Table 3. At baseline, vegetarian subjects, compared to non-vegetarian, only presented significant lower ape B levels. After the first two months of dietary manipulation, the evolution of the plasma lipid profile differed in both groups. In the vegetarian group the increase in dairy products intake induced significant increases in LDL-cholesterol and ape B levels, as well as in LDL-/HDL-cholesterol and ape B/ape A1 ratios. In the non-vegetarian control group, the diet manipulation did not significantly affect the plasma lipid profile (Table 3). The increase in physical activity had a positive influence on both groups. In vegetarian subjects, physical exercise reduced LDL-cholesterol and ape B levels, and increased significantly HDL-cholesterol levels, although this was not accompanied of a similar effect on ape A1 plasma concentration. Consequently, LDL-/HDL-cholestcrol, but not ape B/ape A1 ratio, returned to baseline values. In the non-vegetarian control group, the training program also induced significant reductions in LDL-cholesterol and ape B levels (Table 3). BLOOD LIPIDS IN VEGETARIANS 1711 Table 3. Plasma lipid profile values (mean ± SD) measured at baseline, after two months of supplemented dairy product intake (Diet) and after two additional months of enhanced training (Diet+Exercise), in healthy ovolactovegetarian (n=14) and non-vegetarian (n=10) control subjects. Total cholesterol (mmol/L) LDL-cholesterol(mmol/L) HDL-cholesterol(mmol/L) LDL-/HDL-cholesterol Triglycerides(mmol/L) Ape B (mg/dL) ApoAl (mg/dL) Ape B/Ape AI Ovo-lactovegetarian subjects BASELINE DIET DIET ÷ EXERCISE Non-vegetarian control subjects BASELINE DIET DIET + EXERCISE 4.17±0.48 2.31±0.44 1,45±0.26 1.65±0.47 0.90+0.42 68±16 140_+41 0.46±0.15 4.43+0.45 4.44+0.77 4.24+0.55 2.52±0.35 2.47+0.53 2.30+0.32* 1.58÷0,40 1.59+0.42 1.60±0.42 1.67+0.41 1.67+0.57 1.51±0.37 0.71+0,18 0.84±0.33 0.72±0.15 85±14 § 86+18 78±12' 153+28 164+47 138±41 0.58±0.16 0.56±0.16 0.64±0.28 4.33±0.53 2.51±0.42' 1.42+0.36 1.94±0.68' 0.87±0.29 75±15' 132+49 0.66±0.30' 4.38±0.60" 2.40±0.51 1.76+0.31"t 1.60+0.501 0.85±0.23 68±131 125+34 0.57±0.11 "p<0.05 vs Baseline. ! p<0.05 vs Diet. § p<0.05 vs ovo-lactovegetarian subjects. DISCUSSION The initial premise of this study was to include food pattern differences (higher intake of dairy products), as well as nutrient differences (higher intakes of cholesterol, saturated fat and animal protein), and to investigate the effect on the plasma lipid profile in a group of subjects presenting low intakes of those nutrients. At least in theory, vegetarian subjects could be more sensitive to this type of diet manipulation. Similarly, the positive effect of simultaneous physical exercise, if present, could be more easily appreciated. The results here presented show that in healthy young and physically active ovo-lactovegetarian subjects, an increase in dairy products intake, consisting in 750ml of whole fat milk and 250ml of whole fat yogurt, added to their daily diet, deteriorated the plasma lipid profile. This effect was reversed by a simultaneous increase in physical activity. Compared to non-vegetarian control subjects, the vegetarian group had a similar level of physical activity but tended (p=0.1) to have a lower total energy intake. Nevertheless, body weight was similar in both groups. This discrepancy could be obviously explained by under- or overreporting during dietary anamnesis, but seven days recall and food frequency questionnaires were highly concordant. Decreased lipid oxidation in vegetarians, which is associated to decreased fat stores [25], may be an alternative explanation. Although the diet manipulation increased total energy intake and the program of physical exercise increased energy expenses, body weight in the vegetarian group was unaffected. Several reasons can be argued. The adjustment of energy expenditure and substrate oxidation to energy and substrate intakes constitutes an essential process in the maintenance of whole-body homeostasis [26]. Variations in energy intakes are partly compensated by down-regulation of basal metabolic rate, and this is better related to the fat-free mass [23,27]. In addition, diet-induced thermogenesis is increased with higher protein and fat intake. Modifications of total fat intake and changes in the type of fat consumed may affect the plasma lipid profile, especially in subjects with low intake of saturated fat and cholesterol, such as vegetarian subjects. In these subjects, the restrictions of a vegetarian diet result in a smaller number of foods to be assessed for saturated fat and cholesterol content. Consequently, evaluation of 1712 M. DELGADO et al. dietary lipid consumption is easier and, likely, more accurate [15]. In addition, the low range of cholesterol and saturated fat intake of these subjects may be the range in which plasma lipids are more easily influenced by dietary changes. Diets providing a high percentage of energy derived from fat, as well as diets rich in saturated fat or with a low P/S ratio, generally induce elevations of both LDL- and HDLcholesterol levels [28]. Such diets increase LDL-cholesterol more than HDL-cholesterol levels. This results in a less favorable LDL-/HDL-cholesterol ratio. Our data show that in ovolactovagetarian subjects, increasing the intake of dairy products results in increases in both LDLcholesterol and apo B concentrations, with no change in HDL-cholestcrol. This cannot be attributed to a time-trend effect since the control group was studied in parallel and these variations were not observed. In strict vegetarians, it has been reported that plasma LDL-eholesterol levels are correlated with the consumption of dairy products [15]. ~ fat contains some of the most commonly consumed saturated fatty acids, such as laurie (C12:0) and myristie (C14:0) acids, that have a more pronounced influence on plasma cholesterol [28]. Fatty dairy products are also rich in cholesterol, which may increase LDL-cholesterol levels [29]. Our results agree with these and other [30] studies. Regular physical exercise may offset the effect of high intakes of cholesterol, saturated fat or other dietary components on the plasma lipid profile [4,6]. It has been demonstrated that aerobic conditioning prevented the rise in plasma cholesterol during short-term egg feeding [31]. On the other hand, moderate physical conditioning, in initially sedentary middle aged men, induced significant increases in HDL-cholesterol levels [32,33]. A small number of studies have analyzed the opposite effects of both dietary modifications and training programs on the plasma lipid profile of well-trained subjects. Kiens et al [34] found no modification in HDL-cholesterol in healthy, welltrained men who consumed a fat-rich diet for 4 weeks, followed by 4 weeks of a fat-poor diet. In physically active subjects, dietary modifications probably require longer periods to induce significant changes in the plasma lipid profile [35]. In addition, the threshold of exercise necessary to modify the plasma lipid profile may differ in sedentary and non-sedentary subjects [36,37]. In any case, regular physical exercise of moderate intensity helps to prevent dietary-induced changes in the plasma lipid profile. Our results dearly show the ability of exercise to reduce serum lipids of vegetarians consuming high levels of whole-fat dairy products. In conclusion, a supplemented intake of whole-fat dairy products in healthy young ovolactovagetarian subjects increases LDL-cholesterol and apo B levels. This effect can be reversed by moderately intense regular physical exercise, which, in addition, increases HDL-eholesterol levels. REFERENCES 1. Leon AS. Physiological interactions between diet and exercise in the etiology and prevention of ischaemie heart disease. Ann Clin Res 1988;20:114-1201 ' 2. Colt EWD, Hashim S. Effect of exercise and diet on lipids and cardiovascular disease. Curr Concepts Nutr 1986;15:117-143. 3. Kannel WB, Wilson PWF. Efficacy of lipid profiles in prediction of coronary disease. Am Heart J 1992;124:768-774. 4. Ordovas JM. The genetics of serum lipid responsiveness to dietary interventions. Proc Nutr Soc 1999; 58:171-87. 5. Fraser GE. Diet and coronary heart disease: beyond dietary fats and low-density-lipoprotein cholesterol. Am J Clin Nutr 1994;59(suppl): 1117S-23S. 6. Berg A, Frey I, Baumstark MW, Halle M, Keul J. Physical activity and lipoprotein lipid disorders. Sports Med 1994;17:6-21. 7. Stamler J, Shekelle R. Dietary cholesterol and human coronary heart disease. The BLOOD LIPIDS IN VEGETARIANS 1713 epidemiologic evidence. Arch Pathol Lab Med 1988; 112:1032-40. 8. Singh RB, Rastogi SS, Niaz MA, Ghosh S, Singh R, Gupta S. Effect of fat-modified and fruit- and vegetable-enriched diets on blood lipids in the Indian diet heart study. Am J Cardiol 1992;70:869-874. 9. JenkinsDJA, Wolever TMS, Venketeshwer R et al. Effect on blood lipids of very high intakes of fiber in diets low in saturated fat and cholesterol. N Eng J Med 1993;329:21-6. 10. Lairon D. Dietary fibers: effects on lipid metabolism and mechanisms of action. Eur J Clin Nutr 1996; 50:125-33. 11. Lovati MR, Manzoni C, Sirtori CR. Animal versus vegetable proteins as risk factors for atherosclerotic disease. Cardiovasc Risk Factors 1991; 1:300-306. 12. Judd JT, Baer DJ, Clevidence BA, et al. Effects of margarine compared with those of butter on blood lipid profiles related to cardiovascular disease risk factors in normolipemie adults fed controlled diets. Am J Clin Nutr 1998; 68(4): 768-77. 13. Millet P, Guilland JC, Fuehs F, Klepping J. Nutrient intake and vitamin status of healthy French vegetarians and nonvegetarians. Am J Clin Nutr 1989;50:718-727. 14. Clark N. Nancy Clark's sport nutrition guidebook. 2nd ed. Champaign (IL): Human Kinetic Publishers; 1997. 15. Sacks FM, Omish D, Rosner B, McLanahan S, Castelli WP, Kass EH. Plasma lipoprotein levels in vegetarians. The effect of ingestion of fats from dairy products. JAMA 1985;254:1337-1341. 16. Wood PD, Stefanick ML, Exercise, fitness and atherosclerosis. In: Bouchard et al. Exercise, fitness and health. A consensus of current knowlegde. Human Kinetics Publishers, 1990:409-423. 17. Moreiras O., A. Carbajal, M.L. Cabrera. La composition de los alimentos. Madrid: Ed. Piramide, S.A. 1995. 18. Villegas JA. Alimentaci6n y deporte. Aplicaci6n de control diet6tico. (Food and Sport. Application of dietary control). Murcia: General Asde, 1988. 19. McArdle WD, Katch FI, Katch VL. Exercise Physiology. Philadelphia: Lea and Febirger, 1986. 20. American College of Sports Medicine. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy subjects. Med Sci Sports Exert 1998; 30:975-991. 21. Shephard ILl. How much physical activity is needed for good health? Int J Sports Med 1999;20:23-27. 22. Friedewald WT, Levy RI, Fredickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-505. 23. Gonzalvo C, Rodriguez-Aleman F, Cano MD, Castillo MJ. Comparacion de tres m6todos aplicables al laboratorio clinico para determinaci6n de colesterol en las lipoproteinas de baja densidad (cLDL). Rev Diagn Biol 1992;41:363-367. 24. Carlo MD, Gonzalvo C, Seheen AJ, Castillo MJ. Measurement of apolipoproteins B and A by radial immunodiffusion: methodological assessment and clinical applications. Ann Biol Clin (Pads) 1994;52:657-661. 25. Tremblay A, Buemann. Exercise-training, macronutrient balance and body weight control. Int J Obesity 1995;19:79-86. 26. Schultz Y. The adjustment of energy expenditure and oxidation to energy intake: the role of carbohydrate and fat balance. Int J Obesity 1993;17(Supp13):23-27. 27. Luke A, Schoeller DA. Basal metabolic rate, fat-free mass, and body cell mass during energy restriction. Metabolism 1992;41:450-456. 28. Nestel PJ. Nutritional control of cardiovascular risk factors. Cardiovasc Risk Factors 1991;1:259-264. 1714 M. DELGADO et al. 29. Tan MH, Dickinson MA, Albers JJ, Havel RJ, Cheung MC, Vigne JL. The effect of a high cholesterol and saturated fat diet on serum high density lipoprotein-cholesterol, apoprotein A-1 and apoprotein E levels in normolipidemic humans. Am J Clin Nutr 1980;33:2559-65. 30. Fraser GE, Dysinger W, Best C, Chan R. Ischemic heart disease risk factors in middle-aged Seventh-Day Adventist men and their neighbors. Am J Epidemiol 1987;126:638-646. 31. Faber M, Benade AJS, Van Eck M. Dietary intake, anthropometric measurements, and blood lipid values in weight training athletes (body builders). Int J SportsMed 1986;7:342-346. 32. Tran ZV, Weltman A, Glass GV, Mood DP. The effects of exercise on blood lipids and lipoproteins: A meta-analysis of studies. Med Sci Sports Exerc 1983;15:393-402. 33. Lokey EA, Tran ZV. Effects of exercise training on serum lipid and lipoprotein concentrations in women: A meta-analysis. Int J Sports Med 1989;10:424-429. 34. Kiens B, Gad P, Lithell H, Vessby B. Minor dietary effects on HDL in physically active men. Eur J Clin Invest 1981;11: 265-271. 35. Kiens B, Essen-Gustavsson B, Gad P, Lithell H. Lipoprotein lipase activity and intramuscular triglyceride stores after long-term high-fat and high-carbogydrate diets in physically trained men. Clin Physiol 1987;7:1-9. 36. Danner SA, Wieling W, Havekes L, Leuven JG, Smith EM, Dunning AJ. Effect of physical exercise on blood lipids and adipose tissue composition in young healthy men. Atherosclerosis 1984;53:83-90. 37. Thompson PD, Cullinane EM, Sady SP, et al. Modest changes in high-density lipoprotein concentration and metabolism with prolonged exercise training. Circulation 1988;78:25-34. Accepted for publication September 20, 2000.