Correspondence
Treatment of and Screening for Hyperlipidemia
N Engl J Med 1993; 329:1124-1128October 7, 1993
- Article
To the Editor:
The study by Hunninghake et al. (April 29 issue),1 “The Efficacy of Intensive Dietary Therapy Alone or Combined with Lovastatin in Outpatients with Hypercholesterolemia,” is very misleading. Despite the title of the article, a National Cholesterol Education Program Step 2 diet represents moderate dietary therapy, not “intensive” therapy. When dietary saturated fat and cholesterol are further restricted, patients tend to have much greater reductions in plasma low-density lipoprotein (LDL) cholesterol levels.
The level of plasma LDL is regulated by the LDL receptor, a cell-surface glycoprotein that removes LDL from plasma by receptor-mediated endocytosis2. Most patients with coronary atherosclerosis have plasma LDL cholesterol levels that are many times higher than the level necessary to saturate the LDL-receptor system3. For them, a Step 2 diet may still saturate and suppress the LDL-receptor system, thereby causing further progression of coronary atherosclerosis and little decrease in plasma cholesterol levels. However, further reduction of dietary fat and cholesterol may be sufficient to cause significant reductions in plasma LDL cholesterol and regression of coronary atherosclerosis.
In the Lifestyle Heart Trial, for example, ambulatory patients were able to adhere to a diet in which 10 percent of calories were in the form of fat and approximately 10 mg of cholesterol was provided per day, without lipid-lowering drugs4. After one year, the patients had a 37.4 percent decrease in plasma LDL cholesterol (from 151.4 to 94.8 mg per deciliter [3.92 to 2.46 mmol per liter])4. The patients were following a Step 2 diet at base line, so these reductions in LDL cholesterol during the trial represented additional decreases. Eighty-two percent of these patients had overall regression of coronary atherosclerosis. Patients reported a 91 percent reduction in angina within the first two weeks, and this was a strong motivator for dietary adherence. In addition, diet-induced reductions in plasma lipid and lipoprotein levels may result in a more favorable angiographic result than would be expected from a comparable change induced by drugs alone5.
Ambulatory patients can have diet-induced reductions in plasma LDL cholesterol levels comparable to those achieved with lovastatin, but without the costs and side effects of drug therapy. We must distinguish between what is optimal and what is easy. Whether or not patients wish to follow diets this low in fat and cholesterol is a personal choice. Statements such as those made by Dr. Sox in his accompanying editorial -- that “diet has a rather small effect on serum cholesterol levels” and “when we do treat, we should use drug therapy and not rely on diet alone”6 -- may be welcome news to drug companies, but they are both discouraging and misleading to the general public.
6 ReferencesDean Ornish, M.D.
Shirley Elizabeth Brown, M.D.
Preventive Medicine Research Institute, Sausalito, CA 949651
Hunninghake DB ,Stein EA ,Dujovne CA , et al. The efficacy of intensive dietary therapy alone or combined with lovastatin in outpatients with hypercholesterolemia. N Engl J Med 1993;328:1213-1219
Full Text | Web of Science | Medline2
Goldstein JL ,Brown MS . Regulation of low-density lipoprotein receptors: implications for pathogenesis and therapy of hypercholesterolemia and atherosclerosis. Circulation 1987;76:504-507
CrossRef | Web of Science | Medline3
Brown MS ,Goldstein JL . A receptor-mediated pathway for cholesterol homeostasis. Science 1986;232:34-37
CrossRef | Web of Science | Medline4
Ornish D ,Brown SE ,Scherwitz LW , et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-133
CrossRef | Web of Science | Medline5
Rossouw JE. Meta-analysis of randomized trials. Presented at the Workshop on Stabilization and Regression of Atherosclerosis, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md., June 20-21, 1991. abstract.
6
Sox HC Jr . Screening for lipid disorders under health system reform. N Engl J Med 1993;328:1269-1271
Full Text | Web of Science | Medline
To the Editor:
In the patients studied by Hunninghake et al., the mean triglyceride levels at base line were approximately 150 mg per deciliter. Patients with triglyceride levels exceeding 300 mg per deciliter were excluded from the study. Thus, the patients studied had a Fredrickson type IIa phenotype.
It has been long recognized that patients with low triglyceride levels and a type IIa pattern do not respond very well to dietary therapy. In fact, the rate of response usually does not exceed 5 or 10 percent. Thus, the study by Hunninghake et al. is very much in agreement with this. The fact that the patients on the various diets, especially the low-fat diets, lost 2 kg of weight suggests that they were actually adhering to the diet. I believe the findings that the authors present are real, very important, and appropriate. It makes no sense to put enormous effort into treating these patients with diet and denying them drug treatment for six months, as currently recommended by the National Cholesterol Education Program. This is not true of patients with moderate elevations of serum triglycerides. These patients are far more responsive to dietary therapy, particularly weight reduction.
From a clinical standpoint, the lower the triglyceride level, the smaller the response one can expect to diet in the treatment of patients with hyperlipoproteinemia.
Bruce A. Kottke, M.D., Ph.D.
Mayo Clinic, Rochester, MN 55905To the Editor:
The question may be asked why the subjects' response to dietary treatment in the study by Hunninghake et al. differed from that of subjects in metabolic-unit experiments and clinical trials1,2. The regression lines developed by Keys et al.3 and others to predict the response of serum lipid levels to dietary manipulation describe mean group responses, and it is known that there is substantial individual variation in the magnitude of the response4. Thus, some patients are hyperresponsive to diet, and others are hyporesponsive5.
One possible explanation for the small responses to diet may be that the criteria for admission to the study tended to favor patients who were hyporesponsive to diet. The authors state that “the principal criterion for entry was an LDL cholesterol level between 160 and 200 mg per deciliter (4.14 and 5.17 mmol per liter) while the patient was following his or her current diet.” The subjects were referred by lipid clinics or recruited by advertisements and thus may have already tried dietary treatment unsuccessfully for mild-to-moderate hypercholesterolemia or may have been in the process of such unsuccessful treatment at the time of enrollment. Do the authors have data indicating whether such selection factors were operating?
5 ReferencesSteven Shea, M.D.
Columbia-Presbyterian Medical Center, New York, NY 100321
The Expert Panel
CrossRef | Web of Science | Medline2
Grundy SM ,Barrett-Connor E ,Rudel LL ,Miettinen T ,Spector AA . Workshop on the impact of dietary cholesterol on plasma lipoproteins and atherogenesis. Arteriosclerosis 1988;8:95-101
CrossRef | Medline3
Keys A ,Anderson JT ,Grande F . Prediction of serum-cholesterol responses of man to changes in fats in the diet. Lancet 1957;2:959-966
CrossRef | Web of Science4
Jacobs DR Jr ,Anderson JT ,Hannan P ,Keys A ,Blackburn H . Variability in individual serum cholesterol response to change in diet. Arteriosclerosis 1983;3:349-356
CrossRef | Medline5
Katan MB ,Beynen AC ,de Vries JH ,Nobels A . Existence of consistent hypo- and hyperresponders to dietary cholesterol in man. Am J Epidemiol 1986;123:221-234
Web of Science | Medline
To the Editor:
It may be misleading to imply that diet in combination with lovastatin is more effective than diet alone in reducing clinical events from coronary heart disease or even in inducing regression of atherosclerosis. In several trials, drug-induced changes in the lipoprotein profiles did not result in the same proportional reduction in clinical events and the same regression of disease as dietary interventions1,2. Dietary-intervention trials indicate that relatively small changes in LDL and high-density lipoprotein (HDL) cholesterol may reduce the number of events and promote the regression of atherosclerosis3-5.
5 ReferencesJacques D. Barth, M.D., Ph.D.
University of British Columbia, Vancouver, BC V6T 2B5, Canada1
Brown G ,Albers JJ ,Fisher LD , et al. Regression of coronary artery disease as a result of intensive lipid-lowering therapy in men with high levels of apolipoprotein B. N Engl J Med 1990;323:1289-1298
Full Text | Web of Science | Medline2
Higgins G, Blankenhorn DH. Should treatment of atherosclerosis start early? Lessons from the Monitored Atherosclerosis Regression Study. Inpharma. May 23, 1992:4-5.
3
Arntzenius AC ,Kromhout D ,Barth JD , et al. Diet, lipoproteins, and the progression of coronary atherosclerosis: the Leiden Intervention Trial. N Engl J Med 1985;312:805-811
Full Text | Web of Science | Medline4
Barth JD ,Arntzenius AC ,Kromhout D . Follow-up on Leiden Trial. N Engl J Med 1987;316:881-882
Web of Science | Medline5
Watts GF ,Lewis B ,Brunt JN , et al. Effects on coronary artery disease of lipid-lowering diet, or diet plus cholestyramine, in the St Thomas' Atherosclerosis Regression Study (STARS). Lancet 1992;339:563-569
CrossRef | Web of Science | Medline
To the Editor:
In the study by Hunninghake et al., the finding that the cholesterol-lowering effect of diet was much less than that seen with comparable dietary restriction in a metabolic-unit setting calls into question the compliance of the subjects and the validity of their self-reporting. In addition, there is tremendous individual variation in response to dietary manipulation. A trial of dietary therapy is therefore warranted in every person, even if clinical trials show that the mean benefit is small.
Levels of HDL cholesterol can certainly be lowered by a reduced fat intake. Yet the ratio of LDL cholesterol to HDL cholesterol has significantly improved in other trials of dietary fat restriction1,2. In addition, the prognostic importance of reduced HDL cholesterol in the setting of a low-fat diet is unclear; populations consuming a low-fat diet have low HDL cholesterol levels, but also have low rates of coronary heart disease. Low HDL cholesterol levels due to reduced fat intake are the result of a decreased transport rate rather than the increased catabolism that is responsible for most cases of low HDL cholesterol levels in persons consuming a typical Western diet3. Given that we do not yet understand how HDL cholesterol exerts its protective effect, a given value of HDL cholesterol may have very different implications depending on the underlying physiology behind it.
Most important, we should not forget that our ultimate goal in treating hypercholesterolemia is to decrease morbidity and mortality from coronary heart disease, not to achieve the ideal lipid profile. The switch from a typical Western diet to a low-fat diet with increased intake of fruits, vegetables, and complex carbohydrates will probably lower the risk of coronary heart disease through multiple mechanisms, of which cholesterol lowering is only one.
3 ReferencesGregory K. Bryan, M.D.
Portland Cardiovascular Institute, Portland, OR 972161
Ornish D ,Brown SE ,Scherwitz LW , et al. Can lifestyle changes reverse coroanry heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-133
CrossRef | Web of Science | Medline2
Watts GF ,Lewis B ,Brunt JN , et al. Effects on coronary artery disease of lipid-lowering diet, or diet plus cholestyramine, in the St Thomas' Atherosclerosis Regression Study (STARS). Lancet 1992;339:563-569
CrossRef | Web of Science | Medline3
Brinton EA ,Eisenberg S ,Breslow JL . A low-fat diet decreases high density lipoprotein (HDL) cholesterol levels by decreasing HDL apolipoprotein transport rates. J Clin Invest 1990;85:144-151
CrossRef | Web of Science | Medline
To the Editor:
In their article (April 29 issue), Criqui et al.1 found that, overall, the plasma triglyceride level showed no independent association with coronary mortality, but a small association was observed in certain subgroups of subjects. In particular, the addition of glucose to multivariate models and the stratification of risk according to the levels of LDL and HDL cholesterol provide useful information regarding the risk of mortality from coronary heart disease. It is important, however, to be cautious in generalizing these results beyond the study population.
The Lipid Research Clinics Follow-up Study did not involve a population-based sample, but a sample enriched with subjects with hyperlipidemia. Specifically, 43 percent of the subjects had hyperlipidemia1. Seventy-three percent of the subjects with hyperlipidemia from the original sample had high total cholesterol levels on visit 1 of the Lipid Research Clinics Prevalence Study2. Because cholesterol is itself a well-established risk factor for coronary disease, the enrichment with subjects with hypercholesterolemia will increase the incidence of coronary disease irrespective of the triglyceride levels. In fact, in the Lipid Research Clinics Follow-up Study, the men with hyperlipidemia had a 60 percent increase in mortality from coronary heart disease as compared with the men with normolipidemia (19.7 per 10,000 person-years vs. 12.2 per 10,000 person-years)3.
Weiss and Liff4 have shown that when the rate of disease is increased because of the presence of alternative causal pathways, the ability to detect relations between the exposure of interest and the disease is diminished. Because the base-line incidence of coronary heart disease in this population is high, the proportional increase due to triglyceride levels will be small. Thus, the selection criteria employed by Criqui et al. may attenuate the strength of the association between the triglyceride level and coronary heart disease.
The overall relative risk reported in this study may not reflect the risk of hypertriglyceridemia in the general population. We disagree with the editorial by Sox, which extends the interpretation of these results to argue against screening apparently healthy persons for hypertriglyceridemia.
4 ReferencesJohn E. Hokanson, M.P.H.
Melissa A. Austin, Ph.D.
University of Washington, Seattle, WA 981951
Criqui MH ,Heiss G ,Cohn R , et al. Plasma triglyceride level and mortality from coronary heart disease. N Engl J Med 1993;328:1220-1225
Full Text | Web of Science | Medline2
Williams OD ,Stinnett S ,Chambless LE , et al. Populations and methods for assessing dyslipoproteinemia and its correlates: the Lipid Research Clinics Program Prevalence Study. Circulation 1986;73:Suppl I:I-4
CrossRef3
Gordon DJ ,Ekelund L-G ,Karon JM , et al. Predictive value of the exercise tolerance test for mortality in North American men: the Lipid Research Clinics Mortality Follow-up Study. Circulation 1986;74:252-261
CrossRef | Web of Science | Medline4
Weiss NS ,Liff JM . Accounting for the multicausal nature of disease in the design and analysis of epidemiologic studies. Am J Epidemiol 1983;117:14-18
Web of Science | Medline
To the Editor:
In his editorial, Dr. Sox concludes from the article by Criqui et al. that physicians should not screen for high plasma triglyceride levels. The study actually demonstrates that elevated triglyceride levels are predictive of an increased risk of death in a subgroup of subjects. Although triglyceride levels do not always remain an independent predictor after multivariate analysis,1,2 the results indicate that more information than the total cholesterol level may be needed to identify some persons at high risk. If not triglyceride levels, maybe HDL cholesterol, glucose, or insulin levels should be measured.
Dr. Sox then follows the lead of Hulley et al.3 and suggests a reduction in the scope of cholesterol screening. This approach, based on mortality data and recent intervention trials, is misdirected. Physicians are interested not only in preventing death, but also in preventing all the adverse effects of coronary artery disease. Aggressive interventions3 are extremely expensive and may have side effects. Are they a cost-effective alternative to early identification and risk reduction?
Dr. Sox also discards the idea of diet therapy. The study by Hunninghake et al.1 was designed to test the independence of diet-induced and drug-induced reductions in plasma cholesterol; it was not a test between diet and drugs. Appropriate drug therapy is almost always more potent than diet therapy. Again, public health policy and the approach to the treatment of high-risk patients or those with disease are confused. Lowering plasma cholesterol levels by 5 to 7 percent with diet alone should achieve a 10 to 15 percent reduction in cardiovascular disease. That would mean 100,000 to 150,000 fewer new cases of coronary artery disease each year as a result of an inexpensive, safe, natural approach.
The approach outlined by Dr. Sox fails to distinguish between the safe, inexpensive approaches of primary prevention, based on early identification of many persons at moderate risk, and the more expensive and potentially dangerous approaches of traditional medicine, which treat patients after they become sick.
3 ReferencesHenry N. Ginsberg, M.D.
Alan R. Tall, M.D.
Richard J. Deckelbaum, M.D.
Columbia University College of Physicians and Surgeons, New York, NY 100321
Hunninghake DB ,Stein EA ,Dujovne CA , et al. The efficacy of intensive dietary therapy alone or combined with lovastatin in outpatients with hypercholesterolemia. N Engl J Med 1993;328:1213-1219
Full Text | Web of Science | Medline2
Austin MA . Plasma triglyceride and coronary heart disease. Arterioscler Thromb 1991;11:2-14
CrossRef | Medline3
Hulley SB ,Newman TB ,Grady D ,Garber AM ,Baron RB ,Browner WS . Should we be measuring blood cholesterol levels in young adults? JAMA 1993;269:1416-1419
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: In reply to Dr. Shea, we stated that there was no interaction between previous treatment for hypercholesterolemia (i.e., whether the patient had previously received professionally prescribed dietary therapy) and the response to the National Cholesterol Education Program Step 2 diet. Specifically, the mean reduction in LDL cholesterol was 5 percent in the 61 patients who had received previous treatment and 4 percent in the 35 patients who had not. We thank Dr. Kottke for his comments and agree that patients with substantial hypertriglyceridemia, who were excluded from our study, may well be more responsive to dietary therapy.
In response to Drs. Ornish and Brown, although the dietary intervention in our study was not intensive as compared with the extremely-low-fat vegetarian diet used successfully in their Lifestyle Heart Trial,1 it was certainly intensive relative to current practice, in terms of both fat restriction and the frequency of dietary counseling and monitoring. Few patients can adhere to a diet as extreme as that used in their study, the point of which “was to determine what is true, not what is practicable”1. Drs. Ornish and Brown, and Dr. Barth, propose that a given change in lipoprotein levels has a greater beneficial effect on coronary disease if produced by a diet rather than by a drug. This hypothesis has yet to be tested in a randomized, controlled clinical trial and is not supported by the totality of the evidence from the intervention trials performed to date2.
We agree with Dr. Bryan that all patients being treated for hypercholesterolemia should receive dietary advice, but we also share the view of Dr. Kottke that undue delay in initiating drug therapy in patients whose levels cannot be controlled by diet alone should be avoided. We also agree with Dr. Bryan that our patients probably had difficulty adhering to the prescribed diets. However, patients with hypercholesterolemia are not treated on metabolic units; the typical practitioner, without the considerable resources available to us to apply and monitor dietary therapy, may also find good dietary adherence difficult to achieve. Like Dr. Bryan, we believe that the clinical importance of the dietary reduction of HDL cholesterol is unclear and that a low-fat diet may well have beneficial effects not mediated by changes in lipoprotein levels.
2 ReferencesDonald B. Hunninghake, M.D.
University of Minnesota, Minneapolis, MN 554551
Ornish D ,Brown SE ,Scherwitz LW , et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-133
CrossRef | Web of Science | Medline2
Law MR, Wald NJ, Thompson SG. By how much and how quickly does serum cholesterol reduction lower the risk of ischaemic heart disease? BMJ (in press).
Author/Editor Response
Mr. Hokanson and Dr. Austin are concerned that the enrichment of the Lipid Research Clinics Follow-up Study population with subjects with hyperlipidemia attenuated the association between the triglyceride level and death from coronary heart disease, but they note that the triglyceride level nonetheless added useful information in specific subgroups. They also disagree with Dr. Sox's editorial proscribing the measurement of triglycerides. The effect of increased hyperlipidemia in our population was evaluated with weighted models (unpublished data), which showed a small increase in the predictive value of the triglyceride level. Of greater relevance is the examination of the predictive value of triglyceride levels according to LDL cholesterol levels. As shown in our study, the triglyceride level was a stronger predictor in men and women with LDL cholesterol levels below 160 mg per deciliter than in those with higher LDL cholesterol levels. We concur that the triglyceride level adds information about risk in selected patients, and we agree with the Adult Treatment Panel II of the National Cholesterol Education Program that triglycerides should be measured in persons free of coronary heart disease if they have high total cholesterol levels, low HDL cholesterol levels, or other risk factors for coronary heart disease and in all patients with known atherosclerotic disease1.
Ginsberg et al. also note the predictive value of triglyceride levels in defined subgroups. The remainder of their letter addresses primary as compared with secondary prevention of coronary heart disease and actually concerns Dr. Sox's editorial. We have previously emphasized that lipid modification as secondary prevention, as compared with primary prevention, provides a much larger absolute benefit for a given relative benefit2.
2 ReferencesMichael H. Criqui, M.D., M.P.H.
University of California, San Diego, La Jolla, CA 92093-0607Gerardo Heiss, M.D., Ph.D.
University of North Carolina, Chapel Hill, NC 275141
Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults
CrossRef | Web of Science2
Criqui MH . Cholesterol, primary and secondary prevention, and all-cause mortality. Ann Intern Med 1991;115:973-976
Web of Science | Medline
Author/Editor Response
My response to these thoughtful letters reflects the principle that physicians should offer preventive services only if there is good evidence that the benefit exceeds the harm.
Drs. Ornish and Brown cite their Lifestyle Heart Trial as evidence that ambulatory patients can adhere to diets that are free of animal products and therefore far lower in cholesterol and total fat than the National Cholesterol Education Program Step 2 diet1. Their study population had symptomatic coronary artery disease. We need a clinical trial of a strict vegetarian diet in asymptomatic patients with moderately severe hypercholesterolemia. If this intervention leads to large, sustained reductions in total cholesterol and a lower incidence of coronary heart disease, physicians will have to rethink their advice about diet in patients with moderately severe hypercholesterolemia.
Ginsberg et al. criticize the proposal to postpone screening people with a very low risk of coronary heart disease until their risk is high enough for treatment to reduce substantially the number of deaths from this disease. There are no randomized trials of treating young, low-risk adults for 25 to 30 years with drugs that have come under suspicion as a result of relatively short-term trials that showed increased mortality from causes other than coronary heart disease2. Given the potential for harm from treatment,3,4 proponents of screening young adults with a low risk of coronary heart disease bear the burden of proving that their approach confers net benefits.
Showing that an abnormality is a risk factor for a disease is only the first step in building a case that the benefits of screening exceed the drawbacks. Suppose that Hokanson and Austin are correct and Criqui et al. missed an important, independent effect of hypertriglyceridemia on the incidence of coronary heart disease. Proponents of screening for hypertriglyceridemia must still prove that early detection leads to better outcomes. Their next step would be a randomized trial in which subjects with hypertriglyceridemia are observed for outcomes related to coronary heart disease after having been randomly assigned to a treatment or a control group.
The road to wide acceptance of a screening test is usually long and difficult. Typically, several randomized trials of screening provide the high-quality evidence that an expert panel needs to recommend screening. This process protects an American public that can respond when it is convinced that there is benefit to changing its behavior.
4 ReferencesHarold C. Sox
Dartmouth-Hitchcock Medical Center, Lebanon, NH 037561
Ornish D ,Brown SE ,Scherwitz LW , et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-133
CrossRef | Web of Science | Medline2
Muldoon MF ,Manuck SB ,Matthews KA . Lowering cholesterol concentrations and mortality: a quantitative review of primary prevention trials. BMJ 1990;301:309-314
CrossRef | Web of Science | Medline3
Hulley SB ,Newman TB ,Grady D ,Garber AM ,Baron RB ,Browner WS . Should we be measuring blood cholesterol levels in young adults? JAMA 1993;269:1416-1419
CrossRef | Web of Science | Medline4
Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults
CrossRef | Web of Science
