The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
PreviousPrevious
Volume 333:276-282 August 3, 1995 Number 5
NextNext

Meta-Analysis of the Effects of Soy Protein Intake on Serum Lipids
James W. Anderson, M.D., Bryan M. Johnstone, Ph.D., and Margaret E. Cook-Newell, M.S., R.D.

 

This Article
-Abstract
- PDF

Commentary
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background In laboratory animals, the consumption of soy protein, rather than animal protein, decreases serum cholesterol concentrations, but studies in humans have been inconclusive. In this meta-analysis of 38 controlled clinical trials, we examined the relation between soy protein consumption and serum lipid concentrations in humans.

Methods We used a random-effects model to quantify the average effects of soy protein intake on serum lipids in the studies we examined and used hierarchical mixed-effects regression models to predict variation as a function of the characteristics of the studies.

Results In most of the studies, the intake of energy, fat, saturated fat, and cholesterol was similar when the subjects ingested control and soy-containing diets; soy protein intake averaged 47 g per day. Ingestion of soy protein was associated with the following net changes in serum lipid concentrations from the concentrations reached with the control diet: total cholesterol, a decrease of 23.2 mg per deciliter (0.60 mmol per liter; 95 percent confidence interval, 13.5 to 32.9 mg per deciliter [0.35 to 0.85 mmol per liter]), or 9.3 percent; low-density lipoprotein (LDL) cholesterol, a decrease of 21.7 mg per deciliter (0.56 mmol per liter; 95 percent confidence interval, 11.2 to 31.7 mg per deciliter [0.30 to 0.82 mmol per liter]), or 12.9 percent; and triglycerides, a decrease of 13.3 mg per deciliter (0.15 mmol per liter; 95 percent confidence interval, 0.3 to 25.7 mg per deciliter [0.003 to 0.29 mmol per liter]), or 10.5 percent. The changes in serum cholesterol and LDL cholesterol concentrations were directly related to the initial serum cholesterol concentration (P<0.001). The ingestion of soy protein was associated with a nonsignificant 2.4 percent increase in serum concentrations of high-density lipoprotein (HDL) cholesterol.

Conclusions We found that the consumption of soy protein rather than animal protein significantly decreased serum concentrations of total cholesterol, LDL cholesterol, and triglycerides.


Ingestion of vegetable protein in place of animal protein appears to be associated with a lower risk of coronary heart disease1,2; this effect may reflect decreases in serum cholesterol concentrations.3 The cholesterol-lowering effects of soy protein as compared with animal protein have been recognized in animals for more than 80 years.4 Carroll reviewed the evidence that soy protein produced less hypercholesterolemia and less atherosclerosis in laboratory animals than animal protein.5 Although many clinical investigators have examined the effects of soy protein on serum lipids in humans, the results have not been consistent6; consequently, the Nutrition Committee of the American Heart Association recently concluded that soy protein decreases serum cholesterol concentrations in rabbits but not in humans.7

Clinical investigators have used a variety of soy products, differing amounts of soy protein, differing criteria for selecting subjects, and a variety of protocols. We performed a meta-analysis of these studies, since combining the results of multiple studies of small or moderate size increases the statistical power brought to bear on the research question and thus greatly enhances the precision of estimates of effect. Our analysis indicated that the effects of soy protein in lowering serum cholesterol concentrations were significantly related to the initial serum cholesterol values. The substitution of soy protein for animal protein produced significant decreases in serum concentrations of total cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycerides without significantly affecting high-density lipoprotein (HDL) cholesterol concentrations.

Methods

Identification and Selection of Studies

We searched the medical literature for studies of the effects of soy protein on serum cholesterol concentrations in humans; 37 articles containing primary reports were identified.8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44 Studies were selected for analysis if they had used isolated soy protein or textured soy protein; if they were controlled and had either a crossover or a parallel design; and if they provided initial (base line) values so that changes for each study group could be calculated. Studies were excluded if there was no control group8,11,14,25,27; if they used several sources of vegetable protein26; if whole soybeans rather than soy protein were used30; or if base-line values were not provided.22 After these 8 articles were excluded, 29 articles remained in the analysis.

Subgroup Analyses

Changes in serum lipid concentrations were analyzed in relation to the initial serum lipid values, the types of soy protein used (isolated soy protein, textured soy protein, or a combination), the amount of soy protein ingested (in grams per day), the type of diet (usual Western diet or low-fat and low-cholesterol diet), the age group of the subjects (adults or children), and the similarity of the control diet and the soy-containing diet (specifically, regarding weight change in the subjects and dietary intake of fat, saturated fat, and cholesterol). Study diets were considered to be similar in terms of weight change if the subjects' change in weight during consumption of the two diets did not differ significantly. Study diets were considered to be similar in terms of the intake of total dietary fat, saturated fat, and cholesterol if the reported values for the soy-containing and control diets differed by less than 10 percent. When values were not reported for weight change, dietary fat, or cholesterol intake, these variables were assumed not to be similar.

Meta-Analysis

Summary results of each clinical trial and selected characteristics of the study were tabulated for analysis. The estimate of the principal effect was defined as the mean difference (in milligrams per deciliter) between the change in lipid concentrations when the subjects ingested the soy-containing diet (final value minus initial value) and the change when they ingested the control diet (final value minus initial value). This difference is referred to as the net change. In additional meta-analyses we used only the mean difference attributed to the ingestion of soy protein. For the computation of pooled effects, each study was assigned a weight consisting of the reciprocal of its variance. When raw data were available, the variance for each study was calculated separately by computing the standard deviation of the differences between paired observations for the change during the soy-containing diet and the change during the control diet; the standard error of the differences was then calculated. When raw data were unavailable, the variances of the difference were based on the reported standard deviations for each measure and on either reported correlation coefficients or reported results of paired t-tests for the changes during the two diets.

Estimates of the average effect of soy protein on lipid values and 95 percent confidence intervals were calculated with models based on both fixed-effects and random-effects assumptions.45,46 Because substantial variability between observations was indicated by preliminary tests for homogeneity, we have presented the results of random-effects models calculated according to the method of DerSimonian and Laird.47 The assumption of heterogeneity implied by the use of random-effects models is plausible because of the diverse clinical settings and groups of subjects analyzed.

Predictive models were also developed to examine characteristics of the studies that were hypothesized to influence the observed treatment effects. For this purpose, two-stage mixed regression models (fixed-effects and random-effects) were used.48,49 Predictive models were estimated by hierarchical linear modeling.50 This approach models variation among studies as a function of the characteristics of the study that are hypothesized to affect the response to treatment and a two-stage random component. Both net and unadjusted effects of the substitution of soy protein for animal protein served as outcome variables in alternative models. The set of predictors used for testing hypotheses in regression models was defined at the outset and was based on a preliminary review of the literature. Several alternative coding strategies were evaluated in preliminary analyses. Our final models included the set of predictors specified above for subgroup analyses. To establish overall levels of variability in treatment effects, our regression analysis began with the estimation of unconditional random-effects regression models without predictors. In a second phase, predictors were entered into the models in bivariate and multiple-regression analyses. The degree of reduction in variance associated with each predictor was calculated by comparing the components of variance in unconditional models with those in conditional models containing predictors.50

Results

Characteristics of the Studies

Table 1 shows selected characteristics of the studies that met the criteria for analysis. The 29 articles chosen included the findings of 38 clinical studies; some articles reported data for different subgroups of subjects from one study (e.g., those with normal and those with high cholesterol concentrations); others reported on two different clinical studies. The 38 clinical studies were analyzed independently. in 4 studies the subjects were children, whereas in 34 they were adults. Most studies included both men and women, but the data necessary to analyze effects of soy protein according to sex were not available. Most studies used random assignment with crossover design. Twenty studies used isolated soy protein, 15 used textured soy protein, and 3 used a combination of the two. Soy protein intake averaged 47 g per day (range, 17 to 124); in 14 studies (37 percent) intake was <31 g per day.

View this table:
[in this window]
[in a new window]
 
Table 1. Characteristics of the 38 Studies.

 
In most studies the investigators attempted to provide similar amounts of total fat and saturated fat in the control and soy-containing diets. In 14 studies the diets were similar to conventional Western diets in fat and cholesterol content (these were termed "usual" diets), and in 18 studies the diets were low in fat content (<30 percent of energy) and low in cholesterol content (<200 mg per day). In 29 studies the amounts of total fat and saturated fat were similar in the control and soy-containing diets (i.e., they differed by less than 10 percent); 8 other studies were designed to provide similar total fat and saturated fat intake but the similarity of the diets was not documented. In 20 studies cholesterol intake was similar in the two diets; 9 other studies were designed to provide similar cholesterol intake but similarity was not documented.

All the studies except one37 were designed to maintain weight; 34 studies reported similar weight changes for subjects ingesting the control and soy-containing diets. In all, 19 studies had control and soy-containing diets that were similar with respect to intake of dietary fat (total and saturated), intake of dietary cholesterol, and weight change. These 19 studies are listed as "similar" in each of the last three columns of Table 1.

Changes in Serum Lipid Concentrations

The ingestion of diets containing soy protein, as compared with the control diets, was accompanied by a significant reduction in serum concentrations of total cholesterol, LDL cholesterol, and triglycerides (Table 2). The net change (change during the soy diet minus change during the control diet) in serum cholesterol concentrations was a decrease of 23.2 mg per deciliter (0.60 mmol per liter; 95 percent confidence interval for the decrease, 13.5 to 32.9 mg per deciliter [0.35 to 0.85 mmol per liter]), or 9.3 percent. Of 38 studies, 34 (89 percent) reported a net decrease and 4 (11 percent) reported a net increase in serum cholesterol concentrations.

View this table:
[in this window]
[in a new window]
 
Table 2. Net Change in Serum Lipids and Lipoprotein Concentrations in Subjects Ingesting the Soy-Containing Diets, as Compared with the Control Diets.

 
The net change in serum LDL cholesterol concentrations was a decrease of 21.7 mg per deciliter (0.56 mmol per liter; 95 percent confidence interval for the decrease, 11.2 to 31.7 mg per deciliter [0.30 to 0.82 mmol per liter]), or 12.9 percent. Figure 1 illustrates the net effects of the consumption of soy protein on serum LDL cholesterol concentrations as reported in 31 studies. Twenty-six studies (84 percent) reported a net reduction, four studies (13 percent) reported an increase, and one study (3 percent) reported no change.


View larger version (6K):
[in this window]
[in a new window]
 
Figure 1. Net Changes in Serum LDL Cholesterol Concentrations in 31 Clinical Trials of the Effects of Soy Protein on Serum Lipids.

These 31 trials presented data on LDL cholesterol for a total of 564 subjects. The values shown are the mean changes in LDL cholesterol concentrations while subjects received the diet containing soy protein minus the changes during the control diet, with 95 percent confidence intervals. A and B indicate separate studies reported in a single published article, listed here in the same order as in Table 1. To convert values to millimoles per liter, multiply by 0.02586.

 
Soy protein intake did not significantly affect serum HDL cholesterol concentrations, but the net change was an increase of 2.4 percent. Serum very-low-density lipoprotein (VLDL) cholesterol concentrations were not significantly altered by soy protein. The consumption of soy protein significantly decreased serum triglyceride concentrations, by 13.3 mg per deciliter (0.15 mmol per liter; 95 percent confidence interval for the decrease, 0.3 to 25.7 mg per deciliter [0.003 to 0.29 mmol per liter]), or 10.5 percent. Of 30 studies, 22 (73 percent) reported a net decrease in serum triglyceride concentrations, whereas 8 (27 percent) reported an increase.

Effect of Initial Serum Lipid Concentrations

Table 3 summarizes the effects of various factors on changes in serum cholesterol concentrations. In the complete regression model, the initial serum cholesterol concentration was the only significant predictor of the change in the serum cholesterol concentration (P<0.001). The relation between the initial serum cholesterol concentration and changes in serum cholesterol was modeled as a quadratic polynomial function. The proportion reduction in variance among studies between conditional and unconditional models indicated that the base-line cholesterol concentration accounted for approximately 77 percent of the overall variance. However, significant heterogeneity continued to be present in the model even after adjustment for hypothesized predictors of variation (variance component = 0.134, P<0.001).

View this table:
[in this window]
[in a new window]
 
Table 3. Fixed-Effects Estimates from the Regression Model Predicting Net Changes in Serum Cholesterol Concentrations as a Function of Characteristics of the Study.

 
Table 4 presents changes in serum cholesterol and LDL cholesterol concentrations according to quartiles of the initial cholesterol concentration. Subjects with normal cholesterol levels, who had initial values below 200 mg per deciliter, had nonsignificant reductions of 3.3 percent while receiving the soy protein diet. Those with mild hypercholesterolemia, who had initial values of 200 to 255 mg per deciliter (5.2 to 6.6 mmol per liter), had nonsignificant reductions of 4.4 percent. Subjects with moderate hypercholesterolemia, who had initial values of 259 to 333 mg per deciliter (6.70 to 8.61 mmol per liter), had significant decreases of 7.4 percent. Subjects with severe hypercholesterolemia, whose initial values were above 335 mg per deciliter (8.66 mmol per liter), had significant reductions of 19.6 percent.

View this table:
[in this window]
[in a new window]
 
Table 4. Changes in Serum Cholesterol and LDL Cholesterol Concentrations According to Quartiles of the Study Group for Initial Cholesterol Concentration.

 
The pattern of changes in serum LDL cholesterol concentrations, according to quartiles of the initial serum cholesterol values, was similar to the pattern for serum cholesterol concentrations: first quartile, a decrease of 7.7 percent; second quartile, a decrease of 6.8 percent; third quartile, a decrease of 9.8 percent; and fourth quartile, a decrease of 24.0 percent. Changes in serum HDL cholesterol concentrations were similar for all quartiles. Changes in serum triglyceride concentrations were significantly related to the initial serum triglyceride concentrations (P<0.05). However, changes in individual quartile groups were not statistically significant.

Effect of Other Variables

As shown in Table 3, the type of soy protein did not have a significant effect on the net change in serum cholesterol concentrations and accounted for only approximately 1.0 percent of the variance. The amount of soy protein in the diet was also not significant (P = 0.39) when net changes were assessed. The type of diet, although not statistically significant, accounted for approximately 12.6 percent of the variance (P = 0.07); larger changes tended to occur when the control diets were "usual" diets rather than low-fat and low-cholesterol diets. The results of studies of adult subjects did not differ significantly from those of the four studies of children; the age group of the subjects thus had a negligible effect on variance. The changes in the 19 studies with similar diets in terms of fat and cholesterol intake and weight change did not differ significantly from the changes in the remaining studies, in which the diets were not similar; this factor accounted for negligible variance.

To examine the effects of the type and amount of soy protein further, we performed a complete regression analysis using changes observed with the soy diet alone instead of net changes (soy diet minus control diet) as the outcome variable. In this model, significant effects were obtained for the initial serum cholesterol concentration (P<0.001; proportion of reduction accounted for, 0.69) and the amount of soy protein (P = 0.02; proportion of reduction, 0.13). This model predicted that soy protein intake would be associated with the following decreases in serum cholesterol concentrations, after adjustment for the initial values and other variables: 25 g per day of soy protein, a decrease of 8.9 mg per deciliter (0.23 mmol per liter); 50 g per day of soy protein, a decrease of 17.4 mg per deciliter (0.45 mmol per liter); and 75 g per day of soy protein, a decrease of 26.3 mg per deciliter (0.68 mmol per liter). The type of soy protein (P = 0.16), the type of diet (P = 0.11), the age group of the subjects (adults or children) (P = 0.39), and the similarity of the diets (P = 0.28) did not have significant effects on this model.

Discussion

This analysis of 38 controlled clinical studies reported in 29 scientific articles indicated that the replacement of animal protein in the diet with soy protein was associated with a significant decrease in serum cholesterol and LDL cholesterol concentrations. This was a fairly consistent finding, since decreases in serum cholesterol concentrations were reported in 34 of 38 studies; in the 4 studies15,20,23,36 that did not report such reductions, the subjects had fairly low initial serum cholesterol values (average, 185 mg per deciliter [4.78 mmol per liter]). Changes in serum lipid concentrations were independent of changes in body weight and dietary intake of total fat, saturated fat, and cholesterol.

The strength and consistency of these observations are surprising in the light of the conclusion of the Nutrition Committee of the American Heart Association that the "consumption of vegetable proteins leads to lower cholesterol levels than consumption of animal proteins in rabbits but not in humans."7 This comment was supported by only one study.51

Initial serum cholesterol concentrations had a powerful effect on changes in serum cholesterol and LDL cholesterol concentrations and accounted for approximately 77 percent of the variance among studies. The amount of soy protein ingested had a significant effect on serum cholesterol concentrations when the effects of the soy diet were examined alone, without the effects of the control diet. Soy protein intake averaged 47 g per day, and 37 percent of the studies used 31 g per day or less. These observations suggest that the daily consumption of 31 to 47 g of soy protein can significantly decrease serum cholesterol and LDL cholesterol concentrations. After adjustment for initial serum cholesterol concentrations and other variables, the ingestion of 25 or 50 g of soy protein per day was estimated to decrease serum cholesterol concentrations by 8.9 or 17.4 mg per deciliter, respectively. Persons with moderate or severe hypercholesterolemia (>250 mg per deciliter [6.46 mmol per liter]) should have even larger decreases in serum cholesterol concentrations when soy protein replaces animal protein in the diet.

Soy protein products are widely available in supermarkets, and lower-fat soy products are easily obtainable. Persons with hypercholesterolemia can achieve an intake of more than 30 g of soy protein per day by consuming two to three servings of soy products daily. The amount of soy protein in a single serving of various soy products is as follows: 8 oz (226 g) of soy milk contains 4 to 10 g of soy protein; 4 oz (113 g) of tofu, 8 to 13 g; 1 oz (28 g) of soy flour, 10 to 13 g; 1 oz (28 g) of isolated soy protein, 23 g; 1/2 cup (113 g) of textured soy protein, 11 g; and 3.2 oz (91 g) of meat analogue, 18 g.52 Thus, substituting two cups (473 ml) of soy milk for regular milk and consuming one serving of meat analogue would provide approximately 30 g of soy protein per day.

The mechanisms responsible for the effects of soy protein on serum lipoproteins are unknown6,44 and were not addressed in this study. Carroll6 recently reviewed and discussed various hypotheses. In experiments in animals, the amino acid composition of the diet affects serum cholesterol concentrations; increases in arginine are accompanied by decreases in serum cholesterol concentrations.6 Although some studies suggest that alterations in bile acid or cholesterol absorption may contribute to altered cholesterol homeostasis,6 Fumagalli et al.18 found no differences in the fecal excretion of bile acids or sterols by human subjects. Some observers, as discussed by Carroll,6 suggest that alterations in the ratio of serum glucagon to serum insulin may affect hepatic cholesterol synthesis; others6,44 suggest that serum free thyroxine concentrations may be higher when the diet contains soy protein. Huff and colleagues24 suggest that turnover of VLDL apoprotein B is increased in humans when soy protein is substituted for meat and dairy protein. Lovati and colleagues34 report that the LDL-receptor activity of monocytes is eight times greater in human subjects receiving soy protein than in those eating control diets.

Setchell53 suggests that soy estrogens may contribute to the cholesterol-lowering effects of soy protein. Most soy protein products contain soy estrogens (isoflavones or phytoestrogens),54 which have weak estrogenic effects under certain circumstances and antiestrogenic effects under others.55 The administration of oral estrogens56 or the synthetic antiestrogen tamoxifen57 decreases serum cholesterol and LDL cholesterol concentrations; soy estrogens may have similar actions. This suggestion is supported in studies by Anthony and colleagues.58,59,60 In three studies using cynomolgus or rhesus monkeys, soy protein rich in soy estrogens favorably affected serum lipids, whereas soy protein from which the soy estrogens had been extracted had a minimal effect. These primate studies suggest that soy estrogens may account for 60 to 70 percent of the hypocholesterolemic effects of soy protein.

In summary, this meta-analysis of 38 studies indicates that the consumption of soy protein is associated with significant decreases in serum cholesterol, LDL cholesterol, and triglyceride concentrations and with a nonsignificant increase in serum HDL cholesterol concentrations. The decreases in serum cholesterol and LDL cholesterol concentrations were strongly related to the subjects' initial serum cholesterol concentrations. Soy estrogens may be responsible for most of the hypocholesterolemic effects of soy protein.

Supported in part by Protein Technologies International, St. Louis. Dr. Anderson is a member of the Health and Nutrition Advisory Group of Protein Technologies International.


Source Information

From the Metabolic Research Group, Veterans Affairs Medical Center and Department of Medicine (J.W.A.), the Department of Behavioral Science, College of Medicine (B.M.J.), and the Department of Nutritional Sciences (M.E.C.-N.), University of Kentucky, Lexington.

Address reprint requests to Dr. Anderson at the Medical Service, 111C, Veterans Affairs Medical Center, Leestown Rd., Lexington, KY 40511.

References

  1. Hilleboe HE. Some epidemiologic aspects of coronary artery disease. J Chronic Dis 1957;6:210-228. [CrossRef][Medline]
  2. Terpstra AH, Hermus RJ, West CE. The role of dietary protein in cholesterol metabolism. World Rev Nutr Diet 1983;42:1-55. [Medline]
  3. Manson JE, Tosteson H, Ridker PM, et al. The primary prevention of myocardial infarction. N Engl J Med 1992;326:1406-1416. [Medline]
  4. Ignatowsky MA. Influence de la nourriture animale sur l'organisme des lapins. Arch Med Exp Anat Pathol 1908;20:1-20.
  5. Carroll KK. Hypercholesterolemia and atherosclerosis: effects of dietary protein. Fed Proc 1982;41:2792-2796. [Medline]
  6. Carroll KK. Review of clinical studies on cholesterol-lowering response to soy protein. J Am Diet Assoc 1991;91:820-827. [Medline]
  7. Chait A, Brunzell JD, Denke MA, et al. Rationale of the diet-heart statement of the American Heart Association: report of the Nutrition Committee. Circulation 1993;88:3008-3029. [Free Full Text]
  8. Hodges RE, Krehl WA, Stone DB, Lopez A. Dietary carbohydrates and low cholesterol diets: effects on serum lipids of man. Am J Clin Nutr 1967;20:198-208. [Abstract]
  9. Sirtori CR, Agradi E, Conti F, Mantero O, Gotti E. Soybean-protein diet in the treatment of type-II hyperlipoproteinaemia. Lancet 1977;1:275-277. [Medline]
  10. Carroll KK, Giovannetti PM, Huff MW, Moase O, Roberts DC, Wolfe BM. Hypocholesterolemic effect of substituting soybean protein for animal protein in the diet of healthy young women. Am J Clin Nutr 1978;31:1312-1321. [Free Full Text]
  11. Sirtori CR, Gatti E, Mantero O, et al. Clinical experience with the soybean protein diet in treatment of hypercholesterolemia. Am J Clin Nutr 1979;32:1645-1658. [Free Full Text]
  12. Descovich GC, Ceredi C, Gaddi A, et al. Multicentre study of soybean protein diet for outpatient hyper-cholesterolaemic patients. Lancet 1980;2:709-712. [Medline]
  13. Holmes WL, Rubel GB, Hood SS. Comparison of the effect of dietary meat versus dietary soybean protein on plasma lipids of hyperlipidemic individuals. Atherosclerosis 1980;36:379-387. [CrossRef]
  14. Schwandt P, Richter WO, Weisweiler P. Soybean protein and serum cholesterol. Atherosclerosis 1981;40:371-372. [CrossRef][Medline]
  15. Shorey RL, Bazan B, Lo GS, Steinke FH. Determinants of hypocholesterolemic response to soy and animal protein-based diets. Am J Clin Nutr 1981;34:1769-1778. [Free Full Text]
  16. van Raaij JM, Katan MB, Hautvast JG, Hermus RJ. Effects of casein versus soy protein diets on serum cholesterol and lipoproteins in young healthy volunteers. Am J Clin Nutr 1981;34:1261-1271. [Free Full Text]
  17. Wolfe BM, Giovannetti PM, Cheng DCH, Roberts DCK, Carroll KK. Hypolipidemic effect of substituting soybean protein isolate for all meat and dairy protein in the diets of hypercholesterolemic men. Nutr Rep Int 1981;24:1187-98.
  18. Fumagalli R, Soleri L, Farina R, et al. Fecal cholesterol excretion studies in type II hypercholesterolemic patients treated with the soybean protein diet. Atherosclerosis 1982;43:341-353. [CrossRef][Medline]
  19. Goldberg AP, Lim A, Kolar JB, Grundhauser JJ, Steinke FH, Schonfeld G. Soybean protein independently lowers plasma cholesterol levels in primary hypercholesterolemia. Atherosclerosis 1982;43:355-368. [CrossRef][Medline]
  20. van Raaij JM, Katan MB, West CE, Hautvast JG. Influence of diets containing casein, soy isolate, and soy concentrate on serum cholesterol and lipoproteins in middle-aged volunteers. Am J Clin Nutr 1982;35:925-934. [Free Full Text]
  21. Vessby B, Karlstrom B, Lithell H, Gustafsson IB, Werner I. The effects on lipid and carbohydrate metabolism of replacing some animal protein by soy-protein in a lipid-lowering diet for hypercholesterolaemic patients. Hum Nutr Appl Nutr 1982;36:179-189. [Medline]
  22. Grundy SM, Abrams JJ. Comparison of actions of soy protein and casein on metabolism of plasma lipoproteins and cholesterol in humans. Am J Clin Nutr 1983;38:245-252. [Free Full Text]
  23. Sacks FM, Breslow JL, Wood PG, Kass EH. 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 1983;24:1012-1020. [Abstract]
  24. Huff MW, Giovannetti 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 1984;39:888-897. [Free Full Text]
  25. Kolb S, Sailer D. Soybean protein crispbread as additional dietetic measure in hypercholesterolaemia. Nutr Rep Int 1984;30:719-24.
  26. Wiebe SL, Bruce VM, McDonald BE. A comparison of the effect of diets containing beef protein and plant proteins on blood lipids of healthy young men. Am J Clin Nutr 1984;40:982-989. [Free Full Text]
  27. Sanchez A, Horning MC, Shavlik GW, Wingeleth DC, Hubbard RW. Changes in levels of cholesterol associated with plasma amino acids in humans fed plant proteins. Nutr Rep Int 1985;32:1047-56.
  28. Sirtori CR, Zucchi-Dentone C, Sirtori M, et al. Cholesterol-lowering and HDL-raising properties of lecithinated soy proteins in type II hyperlipidemic patients. Ann Nutr Metab 1985;29:348-357. [Medline]
  29. Verrillo A, de Teresa A, Giarrusso PC, La Rocca S. Soybean protein diets in the management of type II hyperlipoproteinaemia. Atherosclerosis 1985;54:321-331. [CrossRef][Medline]
  30. Wolfe BM, Giovannetti PM. Elevation of VLDL-cholesterol during substitution of soy protein for animal protein in diets of hypercholesterolemic Canadians. Nutr Rep Int 1985;32:1057-65.
  31. Giovannetti PM, Carroll KK, Wolfe BM. Constancy of fasting serum cholesterol of healthy young women upon substitution of soy protein isolate for meat and dairy protein in medium and low fat diets. Nutr Res 1986;6:609-18.
  32. Widhalm K. Effect of diet on serum lipids and lipoprotein in hyperlipoproteinemic children. In: Beynen AC, ed. Nutritional effects on cholesterol metabolism. Voorthuizen, the Netherlands: Transmondial, 1986:133-40.
  33. Gaddi A, Descovich GC, Noseda G, et al. Hypercholesterolaemia treated by soybean protein diet. Arch Dis Child 1987;62:274-278. [Free Full Text]
  34. Lovati MR, Manzoni C, Canavesi A, et al. Soybean protein diet increases low density lipoprotein receptor activity in mononuclear cells from hypercholesterolemic patients. J Clin Invest 1987;80:1498-1502.
  35. Mercer NJH, Carroll KK, Giovannetti PM, Steinke FH, Wolfe BM. Effects on human plasma lipids of substituting soybean protein isolate for milk protein in the diet. Nutr Rep Int 1987;35:279-87.
  36. Meinertz H, Faergeman O, Nilausen K, Chapman MJ, Goldstein S, Laplaud PM. Effects of soy protein and casein in low cholesterol diets on plasma lipoproteins in normolipidemic subjects. Atherosclerosis 1988;72:63-70. [CrossRef][Medline]
  37. Jenkins DJ, Wolever TM, Spiller G, et al. Hypocholesterolemic effect of vegetable protein in a hypocaloric diet. Atherosclerosis 1989;78:99-107. [CrossRef][Medline]
  38. Meinertz H, Nilausen K, Faergeman O. Soy protein and casein in cholesterol-enriched diets: effects on plasma lipoproteins in normolipidemic subjects. Am J Clin Nutr 1989;50:786-793. [Free Full Text]
  39. Gaddi A, Ciarrocchi A, Matteucci A, et al. Dietary treatment for familial hypercholesterolemia -- differential effects of dietary soy protein according to the apolipoprotein E phenotypes. Am J Clin Nutr 1991;53:1191-1196. [Free Full Text]
  40. Laurin D, Jacques H, Moorjani S, et al. Effects of a soy-protein beverage on plasma lipoproteins in children with familial hypercholesterolemia. Am J Clin Nutr 1991;54:98-103. [Free Full Text]
  41. Steele MG. The effect on serum cholesterol levels of substituting milk with a soya beverage. Aust J Nutr Diet 1992;49:24-8.
  42. Potter SM, Bakhit RM, Essex-Sorlie DL, et al. Depression of plasma cholesterol in men by consumption of baked products containing soy protein. Am J Clin Nutr 1993;58:501-506. [Free Full Text]
  43. Widhalm K, Brazda G, Schneider B, Kohl S. Effect of soy protein diet versus standard low fat, low cholesterol diet on lipid and lipoprotein levels in children with familial or polygenic hypercholesterolemia. J Pediatr 1993;123:30-34. [CrossRef][Medline]
  44. Bakhit RM, Klein BP, Essex-Sorlie D, Ham JO, Erdman JW Jr, Potter SM. Intake of 25 g of soybean protein with or without soybean fiber alters plasma lipids in men with elevated cholesterol concentrations. J Nutr 1994;124:213-222.
  45. Laird NM, Mosteller F. Some statistical methods for combining experimental results. Int J Technol Assess Health Care 1990;6:5-30. [Medline]
  46. National Research Council. Combining information: statistical issues and opportunities for research. Washington, D.C.: National Academy Press, 1992.
  47. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177-188. [CrossRef][Medline]
  48. Morris CN, Normand SL. Hierarchical models for combining information and for meta-analyses. In: Bernardo JM, Berger JO, Dawid AP, Smith AFM, eds. Bayesian statistics 4. Oxford, England: Oxford University Press, 1992:321-44.
  49. Bryk AS, Raudenbush SW. Hierarchical linear models: applications and data analysis methods. Newbury Park, Calif.: Sage, 1992.
  50. Bryk AS, Raudenbush SW, Congdon RT Jr. HLM23. Chicago: Scientific Software International, 1994.
  51. Huff MW, Hamilton RM, Carroll KK. Plasma cholesterol levels in rabbits fed low fat, cholesterol-free, semipurified diets: effects of dietary proteins, protein hydrolysates and amino acid mixtures. Atherosclerosis 1977;28:187-195. [CrossRef][Medline]
  52. Department of Agriculture. Composition of foods. Handbook 8-16: legumes and legume products. Washington, D.C.: Department of Agriculture, 1986.
  53. Setchell KDR. Naturally occurring non-steroidal estrogens of dietary origin. In McLachlan JA, ed. Estrogens in the environment II: influences on development. New York: Elsevier, 1985:69-85.
  54. Dwyer JT, Goldin BR, Saul N, Gaultieri L, Barakat S, Adlercreutz H. Tofu and soy drinks contain phytoestrogens. J Am Diet Assoc 1994;94:739-743. [CrossRef][Medline]
  55. Cassidy A, Bingham S, Setchell KD. Biological effects of a diet of soy protein rich in isoflavones on the menstrual cycle of premenopausal women. Am J Clin Nutr 1994;60:333-340. [Free Full Text]
  56. Walsh BW, Schiff I, Rosner B, Greenberg L, Ravnikar V, Sacks FM. Effects of postmenopausal estrogen replacement on the concentrations and metabolism of plasma lipoproteins. N Engl J Med 1991;325:1196-1204. [Abstract]
  57. Love RR, Wiebe DA, Newcomb PA, et al. Effects of tamoxifen on cardiovascular risk factors in postmenopausal women. Ann Intern Med 1991;115:860-864.
  58. Anthony MS, Clarkson TB, Hughes CL. Plant and mammalian estrogen effects on plasma lipids of female monkeys. Circulation 1994;90:Suppl:I-235.abstract 
  59. Anthony MS, Clarkson TB, Weddle DL, Wolfe MS. Effects of soy protein phytoestrogens on cardiovascular risk factors in rhesus monkeys. J Nutr 1995;125:Suppl 3S:803S-804S.abstract 
  60. Anthony MS, Burke GL, Hughes CL Jr, Clarkson TB. Does soy supplementation improve coronary heart disease (CHD) risk? Circulation 1995;91:925-925.abstract 

 

This Article
-Abstract
- PDF

Commentary
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

Related Letters:

Soy Protein and Serum Lipids
Krauss R. M., Chait A., Stone N. J., Anderson J. W., Johnstone B. M., Cook-Newell M. E.
Extract | Full Text  
N Engl J Med 1995; 333:1715-1716, Dec 21, 1995. Correspondence

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  TERMS OF USE  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved.