Short-Term

Effects of Substituting Protein for Carbohydrate in the Diets of Moderately Hypercholesterolemic Human Subjects Bernard

M. Wolfe and Patricia M. Giovannetti

The short-term effects on plasma lipoprotein lipids of substituting meat and dairy protein for carbohydrate in the diets of 10 free-living moderately hypercholesterolemic human subjects (four men, six women) were studied under closely supervised dietary control during the consumption of constant, low intakes of fat and cholesterol and the maintenance of stable body weight as well as constant fiber consumption. Subjects were randomly allocated to either the high or low protein diets (mean, 23% Y 11% of energy as protein, 24% as fat, and 53% Y 65% as carbohydrate) and then switched to the other diet for another 4 to 5 weeks. Mean fasting plasma high-density lipoprotein cholesterol (HDL-C) was significantly higher by 12% & 4% (0.97 -c 0.08 v 0.89 f 0.08 mmol/L, P < .Ol), whereas mean total cholesterol (TC) was lower by 6.5% ? 1.3% (5.7 2 0.3 v 6.1 2 0.3 mmol/L, P < .OOl), mean low-density lipoprotein-cholesterol (LDL-C) lower by 6.4% -+ 2.0% (4.5 -r- 0.2 v 4.8 r 0.2 mmol/L. P < .02), mean total triglycerides (TG) lower by 23% * 5% (1.7 2 0.1 v 2.4 Z?0.3 mmol/L, P < .02), and mean very-low-density lipoprotein triglycerides (VLDL-TG) lower by 31% 2 9% (0.96 2 0.15 v 1.48 2 0.26 mmol/L, P < .05) on the high versus low protein diet. Mean values for LDL-C were significantly lower during weeks 3 to 5 of the high protein diet than during either weeks 1 to 5 or weeks 1 to 2 of the high protein diet (4.3 rt 0.3, 4.5 2 0.2, and 4.7 f 0.3 mmol/L, respectively, P < .05) and 11% f 3% lower than on low protein diet, P < .005. The ratio of plasma LDL-C to HDL-C was consistently lower by 17% 2 3% during the high versus low protein diet (4.9 f 0.5~ 5.8 ? 0.5.P < .OOl). Lowering plasma TC and LDL-C and total TG and VLDL-TG and increasing HDL-C by chronic isocaloric substitution of dietary protein for carbohydrate may enhance the cardiovascular risk reduction obtained by restriction of dietary fat and cholesterol. Copyright 0 1991 by W.B. Saunders Company

B

ECAUSE OF convincing causal relationships between various lipoprotein disorders and the development of ischemic heart disease, such disorders are actively being sought out and vigorously treated.’ Ischemic heart disease is common in western industrialized populations, with an incidence based on electrocardiographic data in British men approaching 20% in the age range from 55 to 59 years.’ There is no longer any question about the atherogenic effects of severely elevated levels of serum low-density lipoproteins (LDL)?-5 The American Heart Association has made a vigorous effort to guide physicians and the public regarding the best diet for prevention of cardiovascular disease based on currently available evidence.617 Based on standard equations derived from metabolic studies, it was anticipated that the proposed fat-modified diets would decrease serum cholesterol by 0.8 to 1 mmol/L below current levels6 Somewhat smaller decreases have been attained with these diets in men with documented coronary heart disease.* Although the reduction of serum cholesterol through the exchange of complex carbohydrates for saturated fat is thought to be the most important aspect of this therapy,’ a clinical trial has shown heterogeneous cholesterol responses with decreases in cholesterol in only six of 10 subjects and increases in four.’ Larger hypocholesterolemic responses, approaching 2 mmol/L, have been re-

From the Departments of Medicine and Home Economics (Brescia College), The University of Western Ontario, London, Ontario, Canada. Supported by a grant from the Heart and Stroke Foundation of Ontario. Address reprint requests to Bernard M. Wolfe, FRCPC, University Hospital 5OF14, 339 Windennere Rd, London, Ontario N6A 5A5, Canada. Copyright 0 1991 by W.B. Saunders Company 0026-0495/9ll4004-0002$03.00/0

338

ported with high complex carbohydrate, low fat, low cholesterol diets when combined with exercise and weight 10~s.‘~ Substitution of monounsaturated fatty acids for saturated fat has been reported to have a more favorable effect on lipoprotein risk profiles than exchanging carbohydrate for saturated fat, because the former increases the ratio of apolipoprotein A-l to apolipoprotein B.” Dietary cholesterol restriction is not as effective in lowering cholesterol as restriction of saturated fat,” and there is substantial variation between hyporesponders and hyperresponders.‘3.‘4 Furthermore, foods that are rich in certain water-soluble fibers also lower cholesterol.15 In contrast to fat and carbohydrate, the proportion of energy derived from total dietary protein intake, when consumed in broadly physiological amounts, has not been thought to influence plasma concentrations of LDL or high-density lipoprotein (HDL).16 However, a direct relationship between simultaneous dietary intake of animal protein and mass of smaller LDL has recently been observed.” Lower levels of plasma total cholesterol” (TC) and lower rates of coronary artery disease in vegetarians versus nonvegetarians I9 have suggested that dietary plant protein may induce a lower plasma cholesterol than animal protein. However, it is not known whether sustained high intake of animal protein is hypercholesterolemic*O~z’ in man. The level of animal protein consumption has been positively correlated with atherosclerosis2z~23;however, this may not be important etiologically because of the primary role of fat, rather than protein, in determining the severity of atherosclerosis.Z Findings in Masai men are of particular interest in this respect, because they have low cholesterol levels (-3 mmo1/L)Z4and undetectable coronary heart diseasea despite consuming high protein diets consisting mainly of meat and milk (N 300 g/d) for decades. The present studies were undertaken to test the null hypothesis that short-term isocaloric substitution of animal protein for carbohydrate in the diets of moderately hypercholesterolemic

Metabolism,

Vol40. No 4 (April), 1991: pp 338-343

EFFECTS OF PROTEIN ON CHOLESTEROL

339

men and women receiving constant low fat, low cholesterol diets has no significant effect on lipoprotein lipids. METHODS

Subjects Participants were recruited from the Lipid Clinic at University Hospital, London, Ontario. Upon entry into the clinic, all subjects had fasting plasma TC levels above the 75th percentile” with values between 5.8 and 8.0 mmol/L; fasting plasma triglycerides (TG) were between 1 and 3 mmol/L, except for subjects no. 5 and 7 who had values above the 90th percentile,2h between 3 and 4 mmol/L. The latter may be considered to have Fredrickson type Ilb hyperlipoproteinemia. None of these moderately hypercholesterolemic subjects had tendon xanthomata, nor a history of any family member having an elevated plasma cholesterol in childhood. which would suggest that they had familial hypercholesterolemia. Subjects had been maintained on phase I American Heart Association fat-modified diets containing 30% fat and 300 mg cholesterol per day’ for at least 4 months prior to a further reduction of fat intake to 25% of calories during the 2 weeks before starting the study. All subjects were asymptomatic (subject no. 7 had undergone coronary artery bypass grafting 2 years before) and had no evidence of pancreatic, renal, or hepatic insufficiency. Apart from subject no. 2 who received 0.15 mg L-thyroxine daily for chronic primary hypothyroidism and subject no. 6 who received 0.625 mg conjugated equine estrogens daily for postmenopausal hormonal relacement, none of the subjects received medications known to influence lipid metabolism. Intake of ethanol was less than 30 mL/d during the month before the study and was virtually eliminated during the period of the study. Only one subject (no. 8) was a smoker (six cigarettes per day). All participated in moderate physical activity. The experimental protocol was approved by the Standing Committee on Human Research of The University of Western Ontario and informed written consent was obtained.

Experimental Design and Diets The protocol of the diets used was based on the conventional food habits and preferences of the individual subjects. Diet histories obtained by a qualified nutritionist were used in the advance planning of the dietary regimen of each subject. Dietary compliance was promoted by including the foods and food products normally procured, prepared, and eaten by each subject. Close liason was maintained between patients and the dietitian, nurse coordinator, and physician through frequent telephone contact and weekly clinic visits. Differences in mean body weight for any individual between the high and low protein diets did not exceed 1 kg (Table 1). The main differences between the food intake during the phase 1 American Heart Association fat-modified diet during which baseline lipid values were obtained (Table 1) and the food intake during the study were in the modest increases in mixed carbohydrates at the expense of fat and the subsequent provision of either an average of 23% or 11% of energy from protein instead of approximately 15%. High versus low protein diets tailored to each subject’s preferences were consumed daily for the 4- to S-week duration of each experimental dietary period. Using a crossover design, the subjects (who were already consuming diets containing 25% of energy as fat for 1 to 2 weeks) were allocated randomly to begin the study with either the low protein or high protein diets. All randomized subjects completed the study. Following 4 to 5 weeks on the first diet, they were switched to the alternative diet for a further 4 to 5 weeks. Menus were designed to supply essential nutrients and to maintain body weight, which was measured weekly.

Table 1. Characteristics of Subjects Weight (kg) Subject

Sex/Age (yr)

Height(cm)

Low P

Lipids* (mmol/L)

HighP

TC

TG 1.4

M/27

172

77

78

6.3

2

F/56

150

41

42

5.5

1.5

3

M/59

155

55

54

7.0

2.6

4

M/59

170

80

80

7.7

2.0

5

F/63

159

54

55

6.6

3.4

6

F/67

160

53

53

6.5

1.9

7

Ml46

168

85

86

7.7

3.8

8

F/53

167

85

85

6.2

2.6

9

F/42

161

58

59

6.3

1.4

10

F/24

156

50

51

5.2

2.4

50

162

64

64

6.5

2.3

5

2

5

5

0.3

0.3

Mean t SE

*Mean fasting baseline concentrations based on two to four determinations during phase

I American

Heart Association fat-modified diets

containing 30% of energy as fat and 300 mg cholesterol per day.’

The high protein diet differed from the low protein diet (Tables 2 and 3) by replacement of mixed carbohydrates (breads, raisins, pasta) with low fat, high protein foods (turkey. beef. ham, cottage cheese). The 3-day rotating menus included only those dishes that were readily accepted by individual subjects. The amount of dietary fiber found in the low protein diet was approximated to that of the high protein diet by adding wheat fiber. Hard-boiled egg yolk was added to the low protein diet to equalize cholesterol intake. The ratio of polyunsaturated to unsaturated fat was maintained at 1.0 (Table 2). All foods were prepared and proportioned by subjects in their own homes.

Analyses Specimens of blood for measurement of plasma TC, TG. very-low-density lipoprotein cholesterol (VLDL-C), VLDL-TG, low-density lipoprotein cholesterol (LDL-C), and high-density lipoproteiri cholesterol (HDL-C) were obtained from a forearm vein after a Q-hour fast at weekly intervals over 4 to 5 weeks during each dietary period in each subject, except subjects no. 1 to 4 in whom (1) values for VLDL-C were calculated using the Friedewald formula” for the first 3 to 4 weeks of 4 to 5 weeks, and (2) values for VLDL-TG were based on the fasting specimen of blood obtained at the end of each dietary period. The concentrations of TC, TG, and VLDL lipids in whole plasma, separated by the method of Have1 et al,‘* were determined as previously described.z9,3” The concentration of plasma HDL-C was assayed using a heparinmanganese precipitation method (0.092 mmol Mg2+/L).3’ LDL-C concentration was calculated from the formula LDL-C = TC (VLDL-C + HDL-C). Concentrations of TC, TG, and HDL-C were determined on plasma that had been frozen immediately after venesection and removal of blood cells. These determinations for each subject from both the high and low protein diets were done within the same assays. Mean intraassay coefficients of variation for cholesterol and triglycerides were 2.6% and 2.0%, respectively. The nutritive value of foods was estimated according to the latest available Canadian data.j’ Pectin intake was estimated from values found by Zyren et al.”

Statistical Analyses Subjects were assigned numbers and were entered to start with either the low protein or high protein diet according to their number in a series of randomly generated numbers. Variations in body weight and plasma lipid and lipoprotein concentrations

340

WOLFE AND GIOVANNE-ITI Table 2. Average Daily Intake of Nutrients Protein

Mean

Protein

Energy

Level

(kcal)

(91

2,011

52

147

3

Low

f SE Mean

High

Polyunsaturated:

Fat Cholesterol

1,909

108

2 SE

153

7

Mean difference

102

56’

90

5

-+ SE

Carbohydrate

(“4

1%)

(9)

11

344

0.3

65

25

23

0.7

Fiber

Saturated Ratio

Is)

W

(mg)

(9)

53

24

202

28

1.0

4

0.5

17

2

0.01 1.0

263

53

52

189

26

21

1

4

0.4

13

2

0.01

12*

81*

12’

2

0.3

13

2

0.01

1

17

1

3

0.2

8

1

0.01

0.8

24

NOTE. Mean values are based on the prescribed individualized menus (n = 10) as evaluated weekly by the dietitian. *Difference between means is significant, P < ,001.

Plasma TG

between the low protein and high protein diets were assessed using the paired t test?’ Variance was expressed as SE.

The mean baseline values for fasting plasma TG (Table 1) were not significantly different between the subjects who started the high protein diet first versus those who started the low protein diet first (2.5 ‘_ 0.5 v 2.2 5 0.2 mmol/L, P > .5). Reversal of fasting plasma TG levels occurred within 1 week after switching the diets (Fig 1). Plasma TG concentration was consistently lower by 23% * 5% on the high versus low protein diet (mean values, 1.7 f 0.1 v 2.4 ? 0.3 mmol/L, Table 4). Values for fasting plasma VLDL-TG were 31% f 9% lower on the high versus low protein diet (0.96 2 0.15 v 1.48 2 0.26 mmol/L, P < .05).

RESULTS

Plasma TC

The mean baseline values for fasting plasma TC (Table 1) was not significantly different between the subjects who were randomized to commence the high protein diets first (no. 1,2,5,7, and 10) versus those (no. 3,4,6,8, and 9) who started the low protein diet first (6.3 & 0.4 v 6.7 2 0.3 mmol/L, respectively, P > .30). Eight of the 10 subjects had lower mean values for plasma fasting TC (mean overall decrease, 6.5% c 1.3%) and eight subjects also had lower LDL-C (mean overall decrease, 6.4% f 2.0%) on the high versus low protein diet (Tables 4 and 5). Mean values for LDL-C were significantly lower during weeks 3 to 5 of the high protein diet than during either the entire period of weeks 1 to 5, or the first 2 weeks of the high protein diet (4.3 5 0.3, 4.5 f 0.2, and 4.7 2 0.3 mmol/L, respectively, P < .05). Mean concentration of LDL-C during weeks 3 to 5 of the high protein diet was 11% c 3% higher than during weeks 1 to 5 of the low protein diet (4.3 f 0.3 v 4.8 ? 0.2 mmol/L, P < .OOS). Despite the somewhat higher initial mean plasma LDL-C levels in subjects starting the low protein diet first, reversal of plasma LDL-C levels occurred within 3 weeks after switching the diets (Fig 1). Mean fasting HDL-C was higher by 12% 2 4% on the high versus low protein diet (0.97 2 0.08 v 0.89 2 0.08 mmol/L, P < .Ol, Table 5) resulting in a consistent lower ratio of HDL-C to LDL-C (4.9 2 0.5 v 5.8 f 0.5, P < .OOl). The mean fasting concentration of plasma VLDL-C was 19% 2 10% lower during the high versus low protein diet (0.49 2 0.08 v 0.66 f 0.11 mmol/L, .05 < P < .l).

DISCUSSION

The reversal of plasma TC and TG levels after cross-over of the diets (Fig 1) provides substantial evidence for plasma lipid-lowering effects of dietary protein in individuals who are receiving low fat, low cholesterol diets. Other variables, reported to influence cholesterol concentration, such as fat composition and intake,9S35cholesterol intake,“.13 and dietary fiber,15 were kept similar during both dietary periods by means of frequent dietary evaluation and supervision. The low protein diet (mean, 52 * 3 g/d, Table 2) met or exceeded the needs of a safe maintenance level of protein intake considered to be the equivalent of 0.52 to 0.57 g of milk or egg protein per kilogram body weight per day,‘h since mature subjects have a lower requirement for essential amino acids in their protein.” The present finding of a consistent reduction of plasma TG during exchange of protein for carbohydrate contrasts with previous studies in which serum TG increasedA or remained unchanged (estimated as neutral lipids).39‘4’Although the higher carbohydrate intake during the low

Table 3. Average Daily Intake of Nutrients Animal

Plant Protein

Protein

Animal:

Protein

Mean Mean f SE Mean difference + SE

194

56

0.82

4.3

2

18

2

0.08

0.3

130

50

133

50

1 .Ol

3.7

0.27

12

2

11

2

0.07

0.3

57”

3.01*

19

5

62*

5

0.19

0.5

2

0.24

12

3

13

3

0.08

0.03

Low

15

29

37

70

2

3

3

83

77

6

High

(gl

(0~)

0.45

150

44

3

0.08

11

25

23

3.47

2

3

2

68’

48+

12’

5

2

2

1%)

19)

I%)

(9)

NOTE. Mean values are based on the prescribed individualized menus (n = 10) as evaluated weekly by the dietitian. *Difference between means is significant, P < .OOl.

Complex: Simple Ratio

(9)

5 SE

Simple Carbohydrate

Complex Carbohydrate

(Xl

LW4

Plant Ratio

Pectin (9)

EFFECTS OF PROTEIN ON CHOLESTEROL

341

Table 4. Short-Term Effects of Substituting Protein for Carbohydrate

Triglycerides Total Cholesterol (mmol/L) Subject 1

Low P

5.6

Low P

1.7

6.0

1

VLDL High P

1.6

Low P

High P

0.80X

1.oo*

2

5.7

5.2

1.5

1.4

0.77’

0.68*

3

7.2

6.5

3.0

2.2

2.20’

2.03*

4

5.9

5.9

2.7

1.8

1.89*

0.67’

5

6.3

5.6

2.0

1.4

0.97

0.67

6

6.7

6.3

2.0

1.5

0.67

0.42

7

6.7

6.3

4.2

1.9

3.10

1.15

8

6.0

5.3

2.9

2.1

2.10

1.40 0.60

9

6.8

6.6

1.7

1.5

0.90

10

4.0

4.0

2.0

1.8

1.40

1.00

Maa”

6.1

5.7t

2.4

1.7*

1.48

0.961

2 SE

0.3

0.3

0.3

0.1

0.26

0.15

NOTE. Mean value based on four to five weekly samples. *Values for VLDL-TG were based on the sample obtained at the end

5.0

4.0

3-o

2.0

1 .o / 0I-

-

of each dietary period. tsignificantly different from low portion diet (P

< .OOl).

§Significantly different from low protein diet

(P

< ,051

versus high protein diet could have lead to a carbohydrateinduced increase in plasma TG,4.4’ recent studies indicate that the serum TG is not elevated by diets that are high in energy derived from complex carbohydrates.4’ The low protein diet differed from the high protein diet by containing both more complex carbohydrate (average, 150 v 130 g/d) and more simple carbohydrate (average, 194 v 133 g/d, Table 3). The lower mean plasma TC and LDL-C levels observed during the high versus low protein diet in the present hypercholesterolemic men and women (Tables 4 and 5) contrast with a previous report of high protein diets elevating TC in healthy younger women.4’ Beyond the Table 5. Short-Term Effects of Substituting Protein for Carbohydrate of Cholesterol Contained in Plasma

Lipoprotein Fractions Cholesterol VLDL Subject

Low P

(mmol/Ll

LDL High P

Low P

HDL High P

Low P

High P

1

0.37

0.46

5.4

4.7

0.92

0.95

2

0.39

0.39

4.9

4.4

0.99

0.99

3

1.00

1.11

5.8

5.2

0.83

0.77

4

0.70

0.25

5.2

5.0

0.62

0.81

5

0.56

0.44

4.8

4.1

1.01

1.12

6

0.36

0.26

5.1

4.5

1.39

1.54

7

1.34

0.54

4.9

5.1

0.61

0.69

8

0.95

0.60

4.2

3.8

0.83

0.93

9

0.50

0.30

5.1

5.2

1.10

1.20

10

0.40

0.50

3.0

2.8

0.55

0.73

Mean

0.66

0.49

4.8

4.5*

0.89

0.97*

_c SE

0.11

0.08

0.2

0.2

0.08

0.08

NOTE. Mean value based on four to five weekly samples for each dietary period. *Significantly different from low protein diet (P

< .Ol).

I

I

,

-3

-2

-1

I

1

I

0

1

2

2

3

4

WEEKS

SSignificantly different from low protein diet ( P < .02).

on the Concentrations

CROSSOVER

I-

(mmollL)

Total

High P

6.1

mmol/L

on Fasting Plasma Lipids

Fig 1. Effect of substituting high protein diet (-) for low protein diet (---) on serial mean fasting plasma LDL-C (circles) and TG (triangles) concentrations in subjects (n = 5) receiving the low protein diet first (0, A) and those (n = 5) receiving the high protein diet first (0. AI.

differences in gender, age, and lipid levels from the former, the present study had (1) a larger differential in the percent of energy derived from protein that was substituted for carbohydrate (12% v 7%), (2) a higher ratio of polyunsaturated to saturated fat (1.0 v 0.45), and (3) similar intake of cholesterol on both the high versus low protein diet compared with a cholesterol intake approximately 100 mg higher during the high protein diet in the previous study. The significant overall lowering of plasma TC (mean decrease, 0.38 +- 0.08 mmol/L) and LDL-C (mean decrease, 0.36 2 0.10 mmol/L), not to mention the even greater reduction of LDL-C (0.53 2 0.13 mmol/L) during weeks 3 to 5, in response to exchanging protein for carbohydrate compares favorably with the decreases in plasma TC and LDL-C (0.31 and 0.36 mmol/L, respectively) reported in response to substituting 10% of energy as carbohydrate for saturated fat.” Eight of the 10 subjects in the present study showed reductions of LDL-C and all had reductions in the ratio of LDL-C to HDL-C, whereas only six of 10 subjects in the previous study lowered their LDL-C levels and only four of 10 subjects reduced their ratio of plasma LDL-C to HDL-C. HDL-C increased in eight of the 10 subjects in the present study (mean increase, 0.09 2 0.03 mmol/‘L), whereas it was reported to decrease significantly during substitution of carbohydrate for saturated fat.” Wide individual differences in plasma LDL-C responses (hyperresponders and hyporesponders) to dietary fat and cholesterol have been reported.44 Similarly, genetic factors, such as apolipoprotein E polymorphism4’ or other determinants or defects of lipoprotein transport, may underlie differences in plasma lipoprotein responses by dietary protein. However, the basis is unknown for the different responses in TC in a

WOLFE AND GIOVANNETTI

342

minority of the present subjects (two of 10, Table 4) to exchange of dietary protein for carbohydrate. Similarly, the mechanisms by which exchange of protein for carbohydrate influences plasma TC, VLDL-C, LDL-C, HDL-C, and VLDL-TG levels in man are unknown. Changes in fractional catabolic rate of VLDL, intermediate-density lipoprotein (IDL), and LDL apolipoprotein B in response to dietary perturbations have been observed even in the absence of changes in lipoprotein pool size or lipid concentrations.46 Although studies in rats suggest that the dairy protein casein can reduce hepatic recognition and binding of VLDL resulting in higher levels in blood plasma (as compared with soy protein),47 the present study found that increased consumption of animal protein (including dairy protein) by human subjects led to decreases in the concentration of both plasma VLDL and LDL lipids. The extent to which altered lipoprotein secretion and/or removal underlay the observed changes in plasma lipids remains to be determined. In summary, prolonged isocaloric substitution of protein for carbohydrate over 4 to 5 weeks induced an unexpected significant increase in plasma HDL-C and a consistent significant decrease the ratio of LDL-C to HDL-C, along

with significant decreases in fasting plasma TC, LDL-C, TG, and VLDL-TG. The observed changes in lipoprotein concentration resulting from the substitution dietary protein for carbohydrate could be useful in preventing or retarding progression of coronary heart disease in view of the reported prophylactic effects of raising HDL-C4*-50and lowering LDL-C50~S’and triglyceridess2 Furthermore, the Cholesterol Lowering Atherosclerosis Study Group has recently reported that subjects developing new coronary artery lesions showed a larger mean decrease in the percent of energy intake from protein, while those subjects without new lesions increased protein intake.53 The substitution of low fat meats and dairy products for high fat meats and dairy products may be an acceptable, effective approach to reducing cardiovascular risk associated with moderate hypercholesterolemia?4 ACKNOWLEDGMENT

The authors thank Louise White and Nancy Boyd for dietary planning and supervision, Shirley Koenig for expert technical assistance, Mary Cann for coordinating patient visits and taking blood, and Elene Wolfe for assistance in preparation of the manuscript and graphic art.

REFERENCES

1. The Expert Panel, National Cholesterol Education Program: Report of the National Cholesterol Education Program Expert Panel on detection, evaluation and treatment of high blood cholesterol in adults. Arch Intern Med 148:36-69,1988 2. Shaper AG, Cook DG, Walker M, et al: Prevalence of ischaemic heart disease in middle aged British men. Br Heart J 51:595-6051984 3. Stone NJ: Primary type II hyperlipoproteinemia, in Rifkind BM, Levy RI (eds): Hyperlipidemia, Diagnosis and Therapy. Philadelphia, PA, Grune & Stratton, 1977, pp 113-136 4. Slack J: Risks of ischaemic heart-disease in familial hyperlipoproteinemic states. Lancet 2:1380-1382,1969 5. Grundy SM: Cholesterol and coronary heart disease. JAMA 256:2849-2858,1986 6. Grundy SM, Bilheimer D, Blackburn H, et al: Rationale of the diet-heart statement of the American Heart Association. Report of the Nutrition Committee. Circulation 65:839A-854A, 1982 7. Nutrition Committee, American Heart Association: Position statement. Dietary guidelines for healthy American adults. Circulation 74:1465A-1468A, 1986 8. Herbert PN, Flynn MM, Nugent AM, et al: Efficacy of the American Heart Association Diets in men with coronary heart disease. Circulation 76:IV-292,1987 (abstr 1164) 9. Wolf RN, Grundy SM: Influence of exchanging carbohydrate for saturated fatty acids on plasma lipids and lipoproteins in men. J Nutr 113:1521-1528,1983 10. Rosenthal MB, Barnard RJ, Rose DP, et al: Effects of a high-complex-carbohydrate, low-fat, low-cholesterol diet on levels of serum lipids and estradiol. Am J Med 78:23-27,1985 11. Mensink RP, de Groot MJM, van den Broeke LT, et al: Effects of monounsaturated fatty acids v complex carbohydrates on serum lipoproteins and apoproteins in healthy men and women. Metabolism 38:172-178,1989 12. Keys A: Serum cholesterol response to dietary cholesterol. Am J Clin Nutr 40:351-359,1984

13. Mattson FH, Erickson BA, Kligman AM: Effect of dietary cholesterol on serum cholesterol in man. Am J Clin Nutr 25:589594,1972 14. Katan M, Beynen A, de Vries J, et al: Characteristics of human hypo- and hyper-responders to dietary cholesterol. Circulation 72:III118, 1985 (abstr 471) 15. Anderson JW, Story L, Sieling B, et al: Hypocholesterolemic effects of oat-bran or bean intake for hypercholesterolemic men. Am J Clin Nutr 40:1146-1155,1984 16. Sacks FM, Breslow JL, Wood PG, et al: Lack of an effect of dairy protein (casein) and soy protein on plasma cholesterol of strict vegetarians. An experiment and a critical review. J Lipid Res 24:1012-1020,1983 17. Williams PT, Krauss RM, Kindel-Joyce S, et al: Relationship of dietary fat, protein, cholesterol, and fiber intake to atherogenic lipoproteins in men. Am J Clin Nutr 44:7&g-797, 1986 18. Sacks FM, Castelli WP, Donner A, et al: Plasma lipids and lipoproteins in vegetarians and controls. N Engl J Med 292:11481151.1975 19. Phillips RL, Lemon FR, Beeson WL, et al: Coronary heart disease mortality among Seventh-Day Adventists with differing dietary habits: A preliminary report. Am J Clin Nutr 31:S191-S198, 1978 20. Wolfe BM, Grace DM: Substitution of mixed amino acids resembling soy protein for mixed amino acids resembling casein in the diet reduces plasma cholesterol in slowly, but not rapidly fed nor fasted baboons. Metabolism 36:223-229,1987 21. Barth CA, Pfeuffer M: Dietary protein and atherogenesis. Klin Wochenschr 66:135-143,1988 22. Stamler J: Population studies, in Levy R, Rifkind B, Dennis B, et al (eds): Nutrition, Lipids, and Coronary Heart Disease. New York, NY, Raven, 1979, pp 25-88 23. McGill HC Jr (ed): Geographic Pathology of Atherosclerosis. Baltimore, MD, Williams &Wilkins, 1968 24. Mann GV, Shaffer RD, Anderson RS, et al: Cardiovascular disease in the Masai. J Atheroscler Res 4:289-312,1964

EFFECTS OF PROTEIN ON CHOLESTEROL

343

25. Orr JB, Gilks JL: Studies of nutrition, The physique and health of two African tribes. Rep Med Res Council 155:15-66,193l

26. Rifkind BM, Segal P: Lipid reference values for hyperlipidemia 250: 1869-1872, 1983

Research Clinics and hypolipidemia.

Program JAMA

27. Friedewald WT. Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 18:499502,1972 28. Have1 RJ, Eder HA, Bragdon JH: The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest 34:1345-1353, 1955 29. Sperry WM. Webb M: Revision of the Schoenheimer-Sperry method for cholesterol determination. J Biol Chem 1X7:97-106, 1950 30. Carlson LA: Determination scler Res 3:334-336, 1963

of serum triglycerides.

JAthero-

31. Warnick GR, Albers JJ: A comprehensive evaluation of the heparin-manganese procedure for estimating high density lipoprotein cholesterol. J Lipid Res 19:65-76, 1978 32. Health and Welfare Common Footis. Canadian tawa, Canada, 1987

Canada: Nutrient Values for Some Government Publishing Center, Ot-

33. Zyren J, Elkins ER, Dudek JA, et al: Fibre contents selected raw and processed vegetables, fruits and fruit juices served. J Food Sci 48:600-603, 1983 34. Snedecor GW, Cochran WG: Statistical Ames, IA, Iowa State University, 1967, p 92

Methods

of as

(ed 6).

35. Connor WE, Connor SL: The key role of nutritional factors in the prevention of coronary heart disease. Prev Med 1:49-83,1972 36. Passmore R, Nicol BM, Narayana Rao M, et al: Handbook of Human Nutritional Requirements. Geneva, Switzerland. World Health Organization, 1974 37. Report of a Joint FAO/WHO Ad Hoc Expert Committee: Energy and protein requirements. Geneva, Switzerland, World Health Organization, 1973, Technical Report Series No. 522, p 118 38. Leveille GA, Sauberlich HE, Powell RC, et al: The influence of dietary protein on plasma lipids in human subjects. J Clin Invest 41:1007-1012. 1962 39. Wilcox EB, Galloway LS, Taylor F: Studies of serum lipids and nitrogen excretion. Effect of protein, milk intake, and exercise on athletes. J Am Diet Assoc 44:95-99, 1964 40. Lutz RN, Barnes RH, Kwong E, et al: Effect of dietary protein on blood serum cholesterol in men consuming mixed diets. Fed Proc 18:534, 1959 (abstr)

41. Prather ES: Effect of protein on plasma lipids of young women. J Am Diet Assoc 47:187-191,196s 42. Hatch FT, Abell LL, Kendall FE: Effects of restriction of dietary fat and cholesterol upon serum lipids and lipoproteins in patients with hypertension. Am J Med 19:48-60, 1955 43. Cominacini L, Zocca I, Garbin U, et al: Long-term effect of a low-fat, high-carbohydrate diet on plasma lipids of patients affected by familial endogenous hypertriglyceridemia. Am J Clin Nutr 48:57-65, 1988 44. Katan MB, Berns AM, Glatz JFC, et al: Congruence of individual responsiveness to dietary cholesterol and to saturated fat in humans. J Lipid Res 29:883-892, 1988 45. Kesaeniemi YA, Enholm E, Miettinen TA: Intestinal cholesterol absorption is related to apoprotein phenotype. J Clin Invest 80:578-581, 1987 46. Huff MW, Giovanneti PM, Wolfe BM: Turnover of very low-density lipoprotein-apoprotein B is increased by substitution of soybean protein for meat and dairy protein in the diets of hypercholesterolemic men. Am J Clin Nutr 39:888-897, 1984 47. Cohn JS, Nestel PJ: Hepatic lipoprotein receptor activity in rats fed casein and soy protein. Atherosclerosis 56:247-250, 1985 (letter) 48. Manninen V, Elo 0, Frick MH, et al: Lipid alterations and decline in the incidence of coronary heart disease in the Helsinki Heart Study. JAMA 260:641-651,1988 49. Blankenhorn DH, Nessim SA, Johnson RL, et al: Beneficial effects of combined colestipol-niacin therapy on coronary atherosclerosis and coronary venous bypass grafts. JAMA 257:3233-3240, 1987 50. Brown BG, Lin JT, Schaefer SM, et al: Niacin or lovastatin. combined with colestipol, regress coronary atherosclerosis and prevent clinical events in men with elevated apolipoprotein B. Circulation 80:11-266, 1989 (abstr 1061) 5 1. Lipid Research Clinics Program: The Lipid Research Clinics Coronary Primary Prevention Trial Results. 1. Reduction in incidence of coronary heart disease. JAMA 251:351-364,1984 52. Carlson LA, Rosenhamer G: Reduction of mortality in the Stockholm ischemic heart disease secondary prevention study by combined treatment with clofibrate and nicotinic acid. Acta Med Stand 223:405-418,1988 53. Blankenhorn DH, Johnson RL, Mack WJ, et al: The influence of diet on the appearance of new lesions in human coronary arteries. JAMA 263:1646-1652,199O 54. Canadian Liproprotein Conference Ad Hoc Committee on Guidelines for Dyslipoproteinemias: Guidelines for the detection of high-risk lipoprotein profiles and the treatment of dyslipoproteinemias. Can Med Assoc J 142:1371-1382, 1990