High Carbohydrate Diets, Triglyceride-Rich Lipoproteins, and Coronary Heart

Disease RiskFahim Abbasi, MD, Tracey McLaughlin, MD, Cindy Lamendola, MSN, Hee Sun Kim, PhD,

Akira Tanaka, MD, Tao Wang, PhD, Katsuyuki Nakajima, PhD, andGerald M. Reaven, MD

In this study we compared the effects of variations indietary fat and carbohydrate (CHO) content on concen-trations of triglyceride-rich lipoproteins in 8, healthy,nondiabetic volunteers. The diets contained, as a per-centage of total calories, either 60% CHO, 25% fat, and15% protein, or 40% CHO, 45% fat, and 15% protein.They were consumed in random order for 2 weeks, witha 2-week washout period in between. Measurementswere obtained at the end of each dietary period ofplasma triglyceride, cholesterol, low-density lipoprotein(LDL) cholesterol, high-density lipoprotein (HDL) choles-terol, remnant lipoprotein (RLP) cholesterol, and RLP tri-glyceride concentrations, both after an overnight fastand throughout an 8-hour period (8 A.M. to 4 P.M.) inresponse to breakfast and lunch. The 60% CHO dietresulted in higher (mean 6 SEM) fasting plasma triglyc-erides (206 6 50 vs 113 6 19 mg/dl, p 5 0.03), RLP

cholesterol (15 6 6 vs 6 6 1 mg/dl, p 5 0.005), RLPtriglyceride (56 6 25 vs 16 6 3 mg/dl, p 5 0.003), andlower HDL cholesterol (39 6 3 vs 44 6 3 mg/dl, p 50.003) concentrations, without any change in LDL cho-lesterol concentration. Furthermore, the changes inplasma triglyceride, RLP cholesterol, and RLP triglyceridepersisted throughout the day in response to breakfastand lunch. These results indicate that the effects of low-fat diets on lipoprotein metabolism are not limited tohigher fasting plasma triglyceride and lower HDL cho-lesterol concentrations, but also include a persistent el-evation in RLPs. Given the atherogenic potential of thesechanges in lipoprotein metabolism, it seems appropriateto question the wisdom of recommending that all Amer-icans should replace dietary saturated fat with CHO.Q2000 by Excerpta Medica, Inc.

(Am J Cardiol 2000;85:45–48)

Amajor factor regulating the magnitude of the in-crease in triglyceride-rich lipoproteins after

meals is the fasting plasma triglyceride concentration;the higher the basal triglyceride concentration, thegreater the degree of postprandial lipemia.1,2 Giventhe relation between fasting and postprandial triglyc-eride concentrations, and evidence that high carbohy-drate (CHO) diets increase fasting plasma triglycerideconcentration,3–10 it is likely that the low fat, highCHO diets generally recommended to reduce coronaryheart disease risk would lead to day-long elevations intriglyceride-rich lipoproteins. If these particles are asatherogenic as they appear to be,11–15 the clinical useof low-fat, high CHO diets might be appropriatelyscrutinized. On the other hand, although high CHOdiets may increase fasting plasma triglyceride concen-trations, it has been argued that their lower fat contentwill lead to lower postprandial concentrations of tri-glyceride-rich lipoproteins. The view that magnitudeof postprandial lipemia is a direct function of dietary

fat content may be true after ingestion of fat loads, butrecent evidence suggests this is not the case whenmixed meals are consumed. Indeed, in both acute andchronic studies, we have shown that the increase inpostprandial triglyceride concentration is significantlymodulated by the CHO content of the diet.7,8,16,17Ifevidence continues to accumulate implicating the de-gree of postprandial lipemia as a coronary heart dis-ease risk factor, it seems reasonable to assess themetabolic effects of diets aimed at decreasing coro-nary heart disease in terms of both fasting and post-prandial lipemia. This study was initiated to accom-plish this goal. In particular, we wished to take ad-vantage of a recently described method18,19 to isolateremnant lipoprotein (RLP) cholesterol, and to deter-mine their cholesterol and triglyceride concentrations.In this study, we describe our results using thismethod to evaluate the effects of a CHO-enriched dieton fasting lipid and lipoprotein concentrations, as wellas day-long changes in RLP concentrations in re-sponse to mixed meals.

METHODSThe study group consisted of 8 healthy, Caucasian

volunteers who had responded to an advertisement inthe local newspaper. Potential volunteers werescreened at the Stanford General Clinical ResearchCenter with a medical history, physical examination,blood count, urinalysis, and chemical screening bat-tery. The 4 men and 4 women who participated in the

From the Stanford University School of Medicine, Stanford, California;Tokyo Medical and Dental University, Tokyo, Japan; and OtsukaAmerica Pharmaceuticals, Inc., Rockville, Maryland. This study wassupported by Research Grants HL-08506 and RR-00070 from theNational Institutes of Health, Bethesda, Maryland. Manuscript re-ceived June 21, 1999; revised manuscript received August 16,1999, and accepted August 17, 1999.

Address for reprints: Gerald M. Reaven, MD, 213 East GrandAvenue, South San Francisco, California 94080. E-mail: greaven@shaman.com.

45©2000 by Excerpta Medica, Inc. All rights reserved. 0002-9149/00/$–see front matterThe American Journal of Cardiology Vol. 85 January 1, 2000 PII S0002-9149(99)00604-9

study had a mean (6 SEM) age of 576 3 years, abody mass index of 26.06 1.0 kg/m2, were free ofmajor medical problems and/or medications known toaffect lipid metabolism, and were nondiabetic.20 Base-line lipid values (mean6 SEM) were as follows:cholesterol (2166 13 mg/dl); triglycerides (1836 16mg/dl); low-density lioprotein (LDL) cholesterol(1236 12 mg/dl); and high-density lipoprotein (HDL)cholesterol (466 4 mg/dl).

Subjects were randomized to one of two, 14-day,eucaloric diet phases, varying in composition of CHOand fat. The macronutrient composition of 1 diet was40% CHO, 15% protein, and 45% fat; the alternatediet composition was 60% CHO, 15% protein, and25% fat. Saturated fat in both diets was,10% of totalcalories, and the ratio of polyunsaturated fat to mono-unsaturated fat was 0.9 in both diets. Dietary fiber was13.5 g/1,000 kcals in the high CHO diet, and 10.5g/1,000 kcals in the low CHO diet. The controlleddiets consisted of 3 rotating menus prepared in theStanford Clinical Research Center Kitchen. Subjectswere scheduled to visit the clinical research center 3times per week during the diet intervention phases. Ateach visit subjects met with the dietitian to discusscompliance, verify body weight maintenance, andpick up research meals. On the 15th day of eachdietary phase, subjects were admitted to the clinicalresearch center for metabolic testing. Subjects werestudied for an 8-hour period, during which test meals,with the same proportion of CHO, protein, and fat asthe study diet, were given at 8A.M. and noon, withbreakfast comprising 20% and lunch 40% of the esti-mated daily caloric requirement. Blood was drawnbefore breakfast, after an overnight fast, and thenevery 2 hours until 4P.M. Measurements were obtainedof plasma triglycerides, cholesterol, LDL cholesterol,and HDL cholesterol concentrations as previously de-scribed.5–8 Chylomicron and very low density lipopro-tein RLPs were isolated by an immunoseparation meth-od,21,22 using monoclonol antibodies to apolipoproteinA-1 and B-100 that recognize triglyceride-rich lipopro-teins containing apolipoprotein B-48 and a population ofapolipoprotein B-100 enriched in apolipoprotein E.Briefly, plasma (5ml) was added to an immunoaffinitygel suspension containing the 2 monoclonol antibodies,the reaction mixture incubated at room temperature for 2hours on an RLP mixer, and aliquots of the supernatanttaken for measurement of RLP cholesterol and RLPtriglyceride concentrations.18,19,21,22

Subjects then entered a 2-week washout phase,followed by randomization to the other diet. After 2weeks on the second diet, subjects were studied in theclinical research center as described above. Labora-tory analyses for both dietary periods for each patientwere performed in the same assay.

Statistical analyses were performed using Systat7.0.1 for Windows (SPSS, Inc., Chicago, Illinois). Forpurposes of analysis, metabolic response to dietaryintervention was compared for all patients while theywere taking the 60% CHO diet versus the 40% CHOdiet. Data are expressed as mean6 SEM. Student’spairedt test was used to compare changes in fasting

concentrations of the experimental variables usinglog-transformed data. Day-long responses to breakfastand lunch were compared between the 2 diets by2-way analysis of variance.

RESULTSFasting plasma triglyceride, cholesterol, LDL cho-

lesterol, HDL cholesterol, RLP cholesterol, and RLPtriglyceride concentrations with the 40% and 60%CHO diets are given in Table I. It is apparent fromthese results that the 60% CHO diet was associatedwith significantly higher plasma triglyceride, RLPcholesterol, and RLP triglyceride concentrations. Inaddition, HDL cholesterol concentrations were signif-icantly lower with the 60% CHO diet. Plasma choles-terol and LDL cholesterol concentrations were essen-tially identical for the 2 diets.

Comparison of the day-long concentrations of tri-glyceride, RLP cholesterol, and RLP triglyceride inresponse to the 2 experimental diets is seen in Figure1. These data again show that fasting triglyceride,RLP cholesterol, and RLP triglyceride concentrationswere higher with the 60% CHO diet. Although therewas some variation from person to person, the generalresponse to breakfast and lunch was for plasma tri-glyceride, RLP cholesterol, and RLP triglyceride con-centrations to increase progressively throughout the 8hours of observation. It is also clear, despite the lowerfat content in the 60% CHO diet, that concentrationsof triglyceride, RLP cholesterol, and RLP triglycerideremained elevated throughout the day.

DISCUSSIONResults show that isocaloric diets enriched with

CHO increase fasting plasma triglyceride concentra-tion.3–8 Results of recent meta-analyses9,10 haveshown that the increase in plasma triglyceride concen-tration after ingestion of CHO-enriched diets is asso-ciated with lower HDL cholesterol concentrations asshown in this study. Furthermore, as emphasized inthe meta-analyses cited above, replacement of satu-rated fat with monounsaturated and/or polyunsatu-rated fat, as was done in the 40% CHO diet, does notincrease total and LDL cholesterol, despite the highertotal fat content. Finally the results replicate recentobservations7,8 that elevated fasting triglyceride con-centrations, induced by high CHO diets, persistthroughout the day in response to meals, despite thedecrease in the fat content of the meals. To put it most

TABLE I Fasting Plasma Cholesterol, Triglyceride, RemnantLipoprotein Cholesterol, and Triglyceride Concentrations

Variable 40% CHO 60% CHO p Value

Triglyceride (mg/dl) 113 6 19 206 6 50 0.03Total cholesterol (mg/dl) 191 6 12 198 6 9 0.27LDL cholesterol (mg/dl) 124 6 11 123 6 11 0.95HDL cholesterol (mg/dl) 44 6 3 39 6 3 0.003RLP cholesterol (mg/dl) 6 6 1 15 6 6 0.005RLP triglyceride (mg/dl) 16 6 3 56 6 25 0.003

46 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 85 JANUARY 1, 2000

explicitly, high CHO diets are associated with in-creases in both fasting and postprandial triglycerideconcentrations.

A somewhat unexpected finding was that overnightfasting concentrations of both RLP cholesterol andRLP triglyceride were significantly higher after con-sumption of the 60% CHO diet. Because only 1 per-son had fasting triglyceride concentration.300 mg/dlon the 60% CHO diet, we were surprised to findevidence of a substantial increase in the fasting con-centration of RLPs.

To address the clinical significance of this study,several nettlesome issues must be addressed. First, theargument that induction of fasting hypertriglyceride-mia with high CHO diets will be compensated bydecreases in postprandial lipemia is not tenable. Sec-ond, even if it is maintained that higher fasting plasmatriglyceride concentrations do not increase the risk ofcoronary heart disease,23,24is this also true of the othereffects of the 60% CHO diet on lipoprotein metabo-lism observed in this study? There is substantial evi-dence that low-fat, high CHO diets decrease HDLcholesterol concentrations,9,10 an observation repro-duced in the current study. In addition to showing that

the increase in plasma triglyceride concentrations in-duced by a CHO-enriched diet persists throughout theday, we have now demonstrated that fasting concen-trations of RLP cholesterol and RLP triglyceride arehigher in volunteers eating low-fat, high CHO diets,and this difference also persists throughout the day.

Based on these results in 12 healthy volunteers, aswell as the wealth of previously published data,3–10 itseems apparent that substituting CHO for saturated fatleads to higher plasma triglyceride and lower HDLcholesterol concentrations, associated with day-longincreases in circulating triglyceride and RLP concen-trations. All of these changes have been shown to beassociated with enhanced atherogenesis.11–15,25–29Fur-thermore, there is now substantial evidence that ef-forts to lower LDL cholesterol do not depend onsubstitution of CHO for saturated fat, but are equallyeffective when the saturated fat is replaced withmonounsaturated and/or polyunsaturated fat.9,10Giventhis information, it is reasonable to suggest that theclinical use of current dietary guidelines aimed atdecreasing risk of coronary heart disease be reexam-ined.

1. Nestel PJ. Relationship between plasma triglycerides and removal of chylo-microns.J Clin Invest1964;43:943–949.2. Wilson DE, Chan I-F, Buchi KN, Horton SC. Postchallenge plasma lipoproteinretinoids: chylomicron remnants in endogenous hypertriglyceridemia.Metabo-lism 1985;34:551–558.3. Ahrens EH, Hirsch J Jr, Oette K, Farquhar JW, Stein Y. Carbohydrate-inducedand fat-induced lipemia.Trans Assoc Am Physicians1961;74:134.4. Farquhar JW, Frank A, Gross RC, Reaven GM, Glucose, insulin and trigly-ceride responses to high and low carbohydrate diets in man.J Clin Invest1966;45:1648–1656.5. Coulston AM, Liu GC, Reaven GM. Plasma glucose, insulin, and lipidresponses to high-carbohydrate low-fat diets in normal humans.Metabolism1983;32:52–56.6. Chen YDI, Swami S, Skowronski R, Coulston AM, Reaven GM. Effect ofvariations in dietary fat and carbohydrate intake on postprandial lipemia inpatients with non-insulin dependent diabetes mellitus.J Clin Endocrinol Metab1993;76:347–351.7. Chen YDI, Coulston AM, Zhou MY, Wang T, Reaven GM. Why do low-fathigh-carbohydrate diets accentuate postprandial lipemia in patients withNIDDM? Diabetes Care1995;18:10.8. Jeppesen J, Schaaf P, Jones C, Zhou MY, Chen YDI, Reaven GM. Effects oflow-fat, high-carbohydrate diets on risk factors for ischemic heart disease inpost-menopausal women.Am J Clin Nutr1997;65:1027–1033.9. Mensink RP, Katan MB. Effect of dietary fatty acids on serum lipids andlipoproteins.Arterioscler Thromb1992;12:911–919.10. Gardner CD, Kraemer HC. Monounsaturated versus polyunsaturated dietaryfat and serum lipids-meta-analysis.Arterioscler Thromb Vasc Biol1995;15:1917–1925.11. Simons LA, Dyer T, Simons J, Bernstein L, Mock P, Poonia NS, Balasu-bramaniam S, Baron D, Branson J, Morgan J, Roy P. Chylomicrons and chylo-micron remnants in coronary artery disease: a case-controlled study.Atheroscle-rosis 1987;65:181–189.12. Simpson HS, Williamson CM, Olivercrona T, Pringle S, Maclean J, LorimerAR, Bonnefous F, Bogaievsky Y, Packard CJ, Shepherd J. Postprandial lipemia,fenofibrate and coronary artery disease.Atherosclerosis1990;85:193–202.13. Groot PH, van Stiphout WA, Krauss XH, Jansen H, van Tol A, vanRamshorst E, Chin-On S, Hofman A, Cresswell SR, Havekes L. Postprandiallipoprotein metabolism in normolipidemic men with and without coronary arterydisease. Arterioscler Thromb1991;11:653–662.14. Patsch JR, Miesenbock G, Hopferweiser T, Muhlberger V, Knapp E, DunnJK, Gotto AM Jr, Patsch W. Relation of triglyceride metabolism and coronaryartery disease studies in the postprandial state.Arterioscler Thromb1992;12:1336–1345.15. Karpe F, Bard JM, Steiner G, Carlson LA, Fruchart JC, Hamsten A. HDLsand alimentary lipemia: studies in men with previous myocardial infarction atyoung age.Arterioscler Thromb1993;13:11–22.16. Jeppesen J, Chen YDI, Zhou MY, Schaaf P, Coulston A, Reaven GM.

FIGURE 1. Plasma concentration of triglyceride (TG) (upper pan-el), (RLP cholesterol (middle panel), and RLP triglyceride (lowerpanel) measured before and at 2-hour intervals after breakfast(8 A.M.) and lunch (noon).

PREVENTIVE CARDIOLOGY/HIGH CARBOHYDRATE DIETS AND CHD RISK 47

Postprandial triglyceride and retinal ester responses to oral fat: effects of fructose.Am J Clin Nutr1995;61:787–791.17. Jeppesen J, Chen YDI, Zhou MY, Wang T, Reaven GM. Effect of variationsin oral fat and carbohydrate load on postprandial lipemia.Am J Clin Nutr1995;62:1201–1205.18. Nakajima K, Saito T, Tamura A, Suzuki M, Nakano T, Adachi M, Tanaka A,Tada N, Nakamura H, Campos E, Havel RJ. Cholesterol in remnant-like lipopro-teins in human serum monoclonal anti-apoB-100 and anti apoA-1 immunoaffinitymixed gels.Clin Chim Acta1993;223:53–71.19. Watanabe N, Taniguichi T, Taketoh H, Kitagawa Y, Namura H, Yoneda N,Kurimoto Y, Yamada S, Ishikawa Y. Elevated remnant-like lipoprotein particlesin impaired glucose tolerance and type 2 diabetic patients. Diabetes Care1999;22:152–156.20. The Expert Committee on the Diagnosis and Classification of DiabetesMellitus. Report of the expert committee on the diagnosis and classification ofdiabetes millitus.Diabetes Care1998;21:S5–S22.21. Nakajima K, Okazaki M, Tanaka A, et al. Separation and determination ofremnant-like particles in serum from diabetes patients using monocional antibod-ies to apoB-100 and apo A-1.J Clin Ligand Assay1996;19:177–183.22. Leary ET, Wang T, Baker DJ, Cilla DD, Zhong J, Warnick GR, Nakajima K,Havel RJ. Evaluation of an immunoseparation method for quantitative measure-

ment of remnant-like particle-cholesterol in serum and plasma.Clin Chem1998;44:2490–8.23. Hulley SB, Rosenman RH, Bawol RD, Brand RJ. Epidemiology as a guide toclinical decisions. The association between triglyceride and coronary heart dis-ease.N Engl J Med1980;302:1383–1389.24. Criqui MH, Heiss G, Cohn RR, Cowan LD, Suchindran CM, Bangdiwala S,Kritchevsky S, Jacobs DR, O’Grady HK, Davis CE. Plasma triglyceride level andmortality from coronary heart disease.N Engl J Med1993;328:1220–1225.25. Miller GJ, Miller NE. Plasma-high-density-lipoprotein concentration anddevelopment of ischaemic heart-disease.Lancet1975;1:16–19.26. Gordon T, Kannel WB, Castelli WP, Dawber TR. Lipoproteins, cardiovas-cular disease, and death: the Framingham Study.Arch Intern Med1981;141:1128–1131.27. Assmann G, Schultz H,von Eckardstein A. Hypertriglyceridemia and elevatedlipoprotein (a) are risk factors for major coronary events in middle-aged men.Am J Cardiol1996;77:1179–1184.28. Austin MA, Hokanson JE, Edwards KL. Hypertriglyceridemia as a cardio-vascular risk factor.Am J Cardiol1998;81:(suppl 4A):7B–12B.29. Zilversmit DB. Atherogenesis: a postprandial phenomenon.Circulation1979;60:473–485.

48 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 85 JANUARY 1, 2000