cholesterol of man. Their observations w-ere limited to 4 patients, 3 of whom had initial

Size: px

Start display at page:

Download "cholesterol of man. Their observations w-ere limited to 4 patients, 3 of whom had initial"

Fay Sullivan
5 years ago
Views:

$SMITH, M.D., AND MARY E. DEIPSEY, 1\1.S. Previous studies in human beings on unrestricted diets have indicated that the plant sterols (beta or gamma sitosterol) cause decreases in serum cholesterol.$

1 The Effect of Beta Sitosterol on the Serum Lipids of Young Men with Arteriosclerotic Heart Disease Downloaded from by on October 2, 218 By JOHN W. FARQUHAR, M.D., RALPH E. SMITH, M.D., AND MARY E. DEIPSEY, 1\1.S. Previous studies in human beings on unrestricted diets have indicated that the plant sterols (beta or gamma sitosterol) cause decreases in serum cholesterol. In this study, 15 young men with previous mvocardial infarction were given 12 to 18 Gm./day of beta sitosterol with resultant sustained reductions of serum cholesterol and beta lipoprotein. Lipoprotein fractionation was performed by paper electrophoresis. These changes occurred irrespective of initial serum cholesterol or content of diet. Control observations of the effects of diet, weight maintenance, and the inclusion of placebos brought about increased confidence that the changes were due to the administeredl sitosterol. T HE search for agents capable of modifying the serum s of man was stimulated by the observation of Peterson in that the usual hypercholesterolemia of chicks fed cholesterol-enriched diets was prevented by the addition of plant sterols to the diet. In a similar study, Pollak2 described the same phenomenon in rabbits. Hernandez and Chaikoff3 found that the recovery of C14 labeled cholesterol from the thoracic duct of cholesterol-fed rats was substantially reduced by the addition of as little as 4 mg. of sitosterols to a meal containing 1 mg. of cholesterol. Subsequent investigators have reported decreases in serum cholesterol in normal and hypercholesterolemic human beings given varying doses of the plant sterols, gamma and beta sitosterols.4-6 More recently, Best and co-workers7 reported decreases in serum phospho, cholesterol, total, and Sf3-l lipoproteins after beta sitosterol administration. The reductions in phospho and lipoproteins were less consistent and less marked than the changes in serum cholesterol and total. The only negative report to date is that of Wilkinson and co-workers' who recently found that sitosterol was ineffective im lowering the serum From the Cardiovascular Disease Laboratory, Veterans Administration Hospital and University of Minnesota, Minneapolis, Minn. Presented in part at Western Section of American Federation for Clinical Research, Carmel, California, Jan. 26, cholesterol of man. Their observations w-ere limited to 4 patients, 3 of whom had initial serum cholesterols of 2 mg.j1 or less. No placebo periods were included in their study. This study was undertaken to evaluate the effects of -sitosterol on the serum s in patients with arteriosclerotic heart disease, with careful attention to the control of diet, weight, and activity variables. METHODS AND MATERIALS Fifteen normotensive, nondiabetic men between the ages of 26 and 45, who had sustained previous well documented myocardial infarction, were studied with serial serum determinations during a 3-phase study: (1) a premedication period of 6 to 12 weeks, (2) a period of sitosterol administration of 12 to 24 weeks, (3) a period of placebo administration of 1 to 16 weeks. The placebo preparation was similar in taste and appearance to the sitosterol and none of the patients were aware of the change. Prior to the first phase of the study all patients were placed on diets designed to maintain constant weight. The per cent of calories from fat was from 23 to 28 per cent in 9 patients and from 4 to 48 per cent in 6 patients. With the exception of 1 patient who lost 8 pounds during the first phase, the weights of all the patients were maintained to within 3 pounds of their initial values. Cholesterol intake was from 2 to 3 mg./day in the 9 patients on fat-restricted diets and 9 to 1 mg./day in the 6 patients on diets "normal" in fat content. The original diets, and all revisions necessary to maintain constant weight, were made up by a member of the dietetic staff. No effort was made to measure the percentage of fat from animal Circulation, Volume XIV, July, 19J56

$Paper strips after electrophoretic separation and staining. Upper strip stained for protein (bromphenol blue) with albumin and globulin fractions labeled. Lower strip stained for with Sudan black-b.$

2 78 BETA SITOSTEROL AND SERUM LIPIDS IN ARTERIOSCLEROSIS TABLE 1.-Comparison of Normal Control and Abnormal Patient Groups No. of Patients (all males) Normal (15) Abnormalt (15) Mean ± t a* Mean ± a Downloaded from by on October 2, 218 Atb. dx a? A v Protein FIG. 1. Paper strips after electrophoretic separation and staining. Upper strip stained for protein (bromphenol blue) with albumin and globulin fractions labeled. Lower strip stained for with Sudan black-b. The a-lipoprotein fraction appears in the region of albumin-a-1 globulin; the (3- lipoprotein in the region of (3-globulin; the origin at the point of application; and the -y within a 2 cm. area on the cathode side of the center of the strip. Serum was applied evenly across the width of the strip, using.15 for protein and.4 for. P Lipid =ar -8.5 cm - M x2 o Lipid =, rea E -ja are Migration toward anods c, a Point of application-j FIG. 2. Tracing of the strip after densitometry showing planimetric method of separation of,8-lipoprotein from origin. The peak of the,6 area is bisected and the area multiplied by 2 to obtain total,8-lipoprotein area. This "f3" area is then subtracted from the total area to obtain origin area. The mg./1 of stainable is then computed by comparison with the total value of the serum. or vegetable sources, but the proportions of these types of fat did not vary much since all patients followed their prescribed diets. Thren equal portions of f3-sitosterol* were given * Supplied by Eli Lilly and Company, Indianapolis, Ind. a A 9_ Age Weight (lb.) Total protein Gm./1 ml... Cholesterol ml... Total mg./1 ml... a Lipid mg./1.. (3 Lipid mg./1.. Origin mg./1 ml... y Lipid mg./1.. (3 + Origin mg./ 1... mg./1 213 ± i±.4 7. i i= 262 :1 419 i 248 i 36 ± 667 ± i 224 ± 557 :L < i i 15 > i 14 * Standard deviation, f = I X l t Measurements represent combined Means of premedication and placebo periods. t Probability that the differences between Means are due to chance alone. orally immediately before meals in total amounts of 12 to 18 Gm./day. During all phases of the study the patients were ambulatory, and efforts were made to achieve a "steady state" in respect to exercise, diet, and body weight. Fifteen apparently normal men matched for age with the 15 study patients had serum measurements for comparative baseline studies. Serum was obtained in the fasting state at weekly or biweekly intervals during all phases. Serum was analyzed for cholesterol by the method of Abell,9 total by the method of Swahn,1 and lipoprotein calculated by protein and staining of paper strips separated by electrophoresis. The electrophoresis was performed in an apparatus similar to that described by Durrum and associates,"1 using 6 35 cm. by 2.5 cm. Whatman 3 MM. t strips run for 12 hours at 6 C. in barbital buffer of ph 8.6 and ionic strength.5. Alternate strips were stained for protein with bromphenol blue and for with Sudan black-b (fig. 1). The various fractions of the stained strips were assigned proportional values by densitometric and planimetric measurement. The mg./1 of stainable was computed by comparison with the total value of each serum sample. A technic was devised that we believed allowed more precise measurement of -lipoprotein by effecting its separation from the predominantly neutral fat "origin" s (fig. 2). Refinet "Chromatography Grade."

Downloaded from http://ahajournals.org by on October 2, 218 TABLE 2.

3 Downloaded from by on October 2, 218 TABLE 2.-Mean Serum Cholesterol Changes in Patients with Arteriosclerotic Heart Disease During Beta Sitosterol Therapy Cholesterol Mean Period mg./1 p Value Diff. rng./loo % Mean Change Mean :1 a * Premedica- 293 i 48 tion ik (drop) Sitosterol 242 :i (rise) Placebo 288 :1 55J * Standard error of difference SEM1-M2 =V M11 + OrM2 ments including low temperature control allowed satisfactory reproducibility of results.* RESULTS Comparison of the 15 study patients with 15 "normals" matched for age revealed significantly lower mean values in the normal group for f3-lipoprotein and origin s, for total serum, for protein, and for cholesterol (table 1). A prompt fall in serum cholesterol occurred * Duplicate determinations agreed to within: :1:6.5 per cent for alpha, ±7.6 per cent for beta, i8.7 per cent for origin, and i3.7 per cent for total s. FARQUHAR, SMITH, AND DEMPSEY during the first 2 to 3 weeks of sitosterol. No tendency to return to pretreatment values was noted when sitosterol was taken for as long as 6 months. A similar prompt rise in serum cholesterol was noted within 3 weeks following substitution of placebo for sitosterol, the mean level of 288 mg./1 during this phase being only 6 mg./1 below the premedication mean of 293 mg./1 The mean serum cholesterol values of the subjects during each phase of the study are summarized in tables 2 and 3. Figure 3 expresses the changes in serum cholesterol in a different manner; the per cent deviation of each value from the mean of each patient is plotted on the ordinate, and the weeks of each phase of the study are plotted on the abeissa. A lowering of f,-lipoprotein proportional in magnitude to that of cholesterol was observed during the sitosterol period; in contrast, there was no change in origin or a-s. A slight rise in a-lipoprotein occurred during the sitosterol period, this change being statistically less significant than the change in f-lipoprotein. The 3 patients with control serum cholesterol levels below 25 mg./1 had de- TABLE 3.-Effect of Beta sitosterol on Serum Cholesterol of Fifteen Patients with Arteriosclerotic Heart Disease Patient 1. L.D) L.B. 3. R.T L.S. 5. J).L J.A. 7. J.E. 8. W.R. 9. M.M W.M J.L A.S. 13. J.P W.L. 15. R.L... Age A* Premedication period Mean 4 a 22 ± 2 22 ± ± ± ± ± ± ± 4 37 ± 1 37 ± ± ± 353 ± ± ± 38 B Sitosterol period Mean i a 187 ± ± ± ± ± ± ± ±t ± ± ± 9 21 ± ± ± ± 2 c Placebo period Mean i e 217 ± 4 22 ± ± ± ± ± ± ±t ±t ±t ± ±t ±t 25 mg./to Changet A to B Changet B to C p mg./to p <.5 <.2 <.3 <.2 * Mean number of determinations per patient: period A, 4.8; period B, 1.1; period C, 8.2. t See note table <.2 <.2 <.2 79

8 BETA SITOSTEROL AND SERUM LIPIDS IN ARTERIOSCLEROSIS Downloaded from http://ahajournals.org by on October 2, 218.4 +3 + -.1 - : 1 -,. i -1. & et W - -2-3 -4 >l 8 6 4 z Premedicotion Period i.!. '.

4 8 BETA SITOSTEROL AND SERUM LIPIDS IN ARTERIOSCLEROSIS Downloaded from by on October 2, : 1 -,. i -1. & et W >l z Premedicotion Period i.!. '. X -,- ---, r A ,..1 L 4 6. o Z 14 >16 Sitostarol Period WEEKS 1.. : a -JL.- -A;-- T a 6 lo12>14 Placebo Period FIG. 3. Serum cholesterol changes during the sitosterol and placebo periods. Each patient's cholesterol values are plotted as per cent deviation from their own control Mean cholesterol. Each dot represents 1 cholesterol value (for patients on a biweekly schedule) or the Mean of 2 cholesterol values (for patients on a weekly schedule). This graph best portrays the prompt and sustained cholesterol reduction occurring after sitosterol administration and the prompt rise on substitution of placebo. TABLE 4.-Summary of Mean Changes in Total Lipid and Lipoprotein Lipid Fractions Control Sitosterol Differperiod* period ence Lipoprotein mg./1 mg./1 mg./1 Mean i1: S.D. Mean + S.D. X2 Xl x2 -Xl Ra i i < i ± Origin i ±t79 +6 <.8 Y ± i <.8 ( + Origin t i Total Lipid i i * Premedication and placebo periods are combined as "control period." creases of similar magnitude to those with control cholesterol levels above 25 mg./1 No significant difference in response was noted in the 2 dietary groups. A slight but statistically significant drop occurred in total during the period of sitosterol administration. Table 4 summarizes the mean changes in total s and lipoprotein fractions. All values are included save for the 2-week period immediately following substitution of the placebo for sitosterol. In tables 4 and 5 the premedication and placebo periods are combined as the "control" period. No untoward effects of sitosterol were observed during the period of observation; the preparation was palatable and well tolerated by all patients. TABLE 5.-Summary of Individual Changes in Total Lipid and Lipoprotein Lipid Fractions after Beta Sitosterol Administration Values in mg./1 of stainable. Patient 2. L.B. 3. R.T J.L A.S W.L...I 14. R.L...l Mean p..... Total mg./1 U 4. L.S D.L J.A J.E... W.R...I M.M W.Li...l 1148 co..i U o._ A ~ l -1- I- -l l-- I <.2 a-lipoprotein mg.j1 P-lipoprotein mg./1 o oq cn Origin mg./1 o lo U n <.8 y- mg./1 oa o Q r uz U ch c;o <.8 DISCUSSION The known variability of serum cholesterol and f3-lipoproteinl2 makes it advisable to include a number of measurements of these values. Use of placebos and control of diet and body weight add further confidence that observed changes are due to the medication given. Under these circumstances fl-sitosterol effects prompt and sustained reductions of serum cholesterol, total, and :-lipoprotein of a moderate degree, with small increases occurring in a-lipoprotein. These changes occur in patients with serum cholesterols not in the range usually considered as hypercholesterolemic and occur despite previous moderate fat and cholesterol limitation. It has been adequately proved that sitosterols lower the serum cholesterol of man,5-7 and prevent increases in liver and plasma cholesterol of laboratory animals fed diets high in cholesterol." 2, However, the mechanisms of these actions are unknown.

5 Downloaded from by on October 2, 218 An attractive postulate is that sitosterol competes with cholesterol for available fatty acids, bile salts, and the enzyme used in cholesterol esterification, thereby interfering with cholesterol absorption.7' 14 Swell and co-workers14 have shown that soy sterols are esterified in vitro in the presence of pancreatic cholesterol esterase, the rate of esterification being only slightly less than that of cholesterol. In contrast to the action of free sitosterol, Best, Duncan, and Wathenl6 have found that sitosterol esters do not prevent increases in tissue s in myxedematous rats. This latter finding furnishes indirect support for the hypothesis that sitosterol acts by interfering with esterification of cholesterol. A second postulate of the action of sitosterol is that it combines with cholesterol in the gut to form a more stable, less absorbable complex.2 This union may be similar to that complex formed in the Liebermann-Burchard reaction between cholesterol and digitonin, another sterol. Either postulate assumes inhibition of absorption of both dietary cholesterol and that portion of endogenous cholesterol excreted in the bile. Since both dietary and biliary cholesterol are predominantly in the free form, there is no reason to believe sitosterol would differentiate between them. Although studies to date would indicate that sitosterol acts on cholesterol absorption, further work is needed to determine its relationship to the transport, metabolism, and excretion of cholesterol and lipoproteins. The possible role of sitosterol in the treatment of human atherosclerosis has not been established, nonetheless the lack of known side effects and the observed modification of serum s would indicate potential benefit from its use. SUMMARY Baseline serum values of 15 normotensive nondiabetic young men with arteriosclerotic heart disease are reported and compared with an apparently normal group matched for age. The abnormal group was characterized by significantly increased cholesterol, beta lipoprotein, and total. FARQUHAR, SMITH, AND DEMPSEY In this abnormal group, f-sitosterol administration resulted in highly significant reductions in serum cholesterol and A-lipoprotein, with a lesser reduction in total and a slight rise in a-lipoprotein. The apparent lack of deleterious effects, coupled with the observed changes, suggests potential benefits of sitosterol administration. Its role as a therapeutic agent in clinical medicine has not yet been established. SUMMARIO IN INTERLINGUA Es reportate valores de base pro o seral, observate in 15 normotensive e non-diabetic juvene adultos mascule con morbo cardiac arteriosclerotic. Le valores es comparate con illos obtenite ab un secunde e apparentemente normal gruppo de individuos de etates correspondente. Le gruppo anormal esseva characterisate per augmentos significative de cholesterol, o beta-lipoproteinic, e o total. In iste gruppo anormal, le administration de beta-sitosterol resultava in significativissime reductiones del cholesterol seral e del o beta-lipoproteinic, accompaniate de minus marcate reductiones de o total e un leve augmento del o alpha-lipoproteinic. Le apparente absentia de effectos noeive, insimul con le supra-mentionate alterationes, indica possibilemente que le administration de sitosterol resulta in effectos benefic. Le rolo de sitosterol como agente therapeutic in le medicine clinic es non ancora establite. ACKNOWLEDGMENT The authors wish to express their appreciation to Miss Ardelle Mead and Mr. Thomas Jenny for technical assistance, and to Mrs. Janet WAAesselman for preparing the diets. REFERENCES 1 PETERSON, D. WV.: Effect of soybean sterols in the diet on plasma and liver cholesterol in chicks. Proc. Soc. Exper. Biol. & Med. 78: 143, POLLAK,. J.: Successful prevention of experimental hypercholesterolemia and cholesterol atherosclerosis in the rabbit. Circulation 7: 696, HERNANDEZ, H. H., AND CHAIKOFF, I. L.: Do soy sterols interfere with absorption of cholesterol? Proc. Soc. Exper. Biol. & Mled. 87: 541,

82 BETA SITOSTEROL AND SERUM LIPIDS IN ARTERIOSCLEROSIS 4POLLAK,. J.: Reduction of blood cholesterol in man. Circulation 7: 72, 1953. 5BEST, M. M., DU

6 82 BETA SITOSTEROL AND SERUM LIPIDS IN ARTERIOSCLEROSIS 4POLLAK,. J.: Reduction of blood cholesterol in man. Circulation 7: 72, BEST, M. M., DUNCAN, C. H., VAN LOON, E. J., AND WATHEN, J. D.: Lowering of serum cholesterol by the administration of a plant sterol. Circulation 1: 21, BARBER, J. M., AND GRANT, A. P.: The serum cholesterol and other s after administration of sitosterol. Brit. Heart J. 17: 296, 1955.,BEST, M. M., DUNCAN, C. H., VAN LOON, E. J., AND WATHEN, J. D.: The effects of sitosterol on serum s. Am. J. Med. 19: 61, WILKINSON, C. F., BOYLE, E., JACKSON, R. S., AND BENJAMIN, M. R.: Effect of varying the intake of dietary fat and the ingestion of sitosterol on and lipoprotein fractions of human serum. Metabolism 4: 32, ABELL, L. L., LEVY, B. B., BRODIE, B. B., AND KENDALL, F. E.: A simplified method for the estimation of total cholesterol in serum and demonstration of its specificity. J. Biol. Chem. 195: 357, SWAHN, B.: Studies on blood s. Scandinav. J. Clin. & Lab. Invest., Suppl. 8, 5: WILLIAMS, F. G., JR., PICKELS, E. G., DURRUM, E. L.: Improved hanging-strip paper-electrophoresis. Science 121: 829, WATKIN, D. M., LAWRY, E. Y., MANN, G. V., AND HALPERIN, M.: A study of serum beta lipoprotein and total cholesterol variability and its relation to age and serum levels in adult human subjects. J. Clin Invest. 33: 874, PETERSON, D. W., SHNEOUR, E. A., AND PEEK, N. F.: Effects of dietary sterols and sterol esters on plasma and liver cholesterol in the chick. J. Nutrition 53: 451, SWELL, L., BOITER, T. A., FIELD, H., AND TREAD- WELL, C. R.: Esterification of soybean sterols in vitro and their influence on blood cholesterol level. Proc. Soc. Exper. Biol. & Med. 86: 295, BURKE, K. A., MCCANDLESS, R. F. J., AND KRIT- CHEVSKY, D.: Effect of soybean sterols on liver deposition of cholesterol-c'4. Proc. Soc. Exper. Biol. & Med. 87: 87, BEST, M. M., DUNCAN, C. H., AND WATHEN, J. D.: Effect of sitosterol on the hyperlipemia of myxedema. Abstracted, Circulation 12: 482, Downloaded from by on October 2, Enos, W. F., Jr., Beyer, J. C., and Holmes, R. H.: Pathogenesis of Coronary Disease in American Soldiers Killed in Korea. J. A. M. A. 158: 912 (July 16), The authors have previously reported a series of 3 American soldiers killed in Korea, in 77.3 per cent of whom, some gross evidence of coronary disease was demonstrated. The present paper discusses all available clinical data and reviews the histopathology of the lesions. Complete records were available on 2 men. The average age was noted to be 22.1 years, average height 5 ft. 73 in. and average weight lb. The gross and microscopic studies indicate that the coronary lesions are due in part to intravascular stress caused by the hemodynamics of the coronary circulation, as modified by anatomic factors. The stress results in subendothelial fibroblastic proliferation, deposition of a mucoid ground substance, and fragmentation of the internal elastic membrane. It is of interest to compare the lesions found in the American soldiers with the lesions found in a small series of Japanese natives now being studied. Thirty Japanese men, approximating the age group of the American soliders, were found in material from 114 cases studied in Japan. The amount and distribution of the s on the elastica and within the parenchyma of the plaques found in the Americans, as compared to the Japanese, indicate that certain plasma s, as modified by diet, are another important agent in the development of coronary disease in young males. KITCHELL

The Effect of Beta-Sitosterol Upon Intestinal Absorption of Cholesterol in the Rat

The Effect of Beta-Sitosterol Upon Intestinal Absorption of Cholesterol in the Rat By MEYER FRIEDMAN, M.D., RAY H. ROSENMAN, M.D. AND SANFORD. BYERS, PH.D. Thoracic duct lymph was collected from groups