Current issues of ACP Journal Club are published in Annals of Internal Medicine


Lipid-lowering diet for regression of atherosclerosis

ACP J Club. 1992 July-Aug;117:4. doi:10.7326/ACPJC-1992-117-1-004

Source Citation

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. Mar



To compare the effects of a practical lipid-lowering diet with or without cholestyramine on coronary atherosclerosis in men.


Randomized controlled trial with blinded assessment of end points after 3 years.


Not stated.


Men < 66 years of age, referred for coronary angiography on suspicion of coronary heart disease, were included if they had plasma cholesterol concentrations of 6 to 10 mmol/L. Reasons for exclusion were contraindications to cholestyramine, current treatment with diet or lipid-lowering drugs, fasting hyperglycemia or hypertriglyceridemia, recent cardiac events, previous or predicted coronary artery revascularization, accelerated hypertension, or other serious disease. Baseline and 3-year follow-up angiograms were available for 74 patients (82%).


Patients who showed at least 0.5 mmol/L decrease in cholesterol level during a 2-week open cholestyramine trial were randomized to lipid-lowering diet (D; n = 26), diet plus cholestyramine, 16 g/d (DC; n = 24), or usual care (U; n = 24). All patients received medical care and counseling about smoking and exercise.

Main outcome measures

Changes in mean absolute width of the coronary arteries (MAWS), measured with computer aid from selected, angiographic cine-frames (mean, 6.4 segments per patient) and averaged for each patient. Progression (regression) of atherosclerosis was defined as a mean MAWS decrease (increase) of 0.17 mm in diameter. Clinical events were recorded.

Main results

Progression of atherosclerosis occurred in 11 patients (46%) in group U, 4 (15%) in group D, and 3 (12%) in group DC (P < 0.02 for trend). Regression occurred in 1 patient (4%) in group U, 10 (38%) in group D, and 8 (33%) in group DC (P < 0.02 for trend). Mean change in MAWS was -0.201 mm in group U, +0.003 mm in group D, and +0.103 mm in group DC (P < 0.02). Total recorded cardiac events were more frequent in the U group than in the D or DC group (10 [36%],3 [11%], and 1 [4%], respectively; P < 0.05). Lipid levels were reduced in the intervention groups but remained stable in the U group.


A lipid-lowering diet was associated with regression of atherosclerosis in men with moderately elevated cholesterol levels. Treatment with cholestyramine increased the benefit of the diet.

Sources of funding: Unilever plc; St. Thomas' Hospital; Bristol-Myers Ltd.

Address for article reprint: Dr. G.F. Watts, Department of Endocrinology and Chemical Pathology, St. Thomas' Hospital, Lambeth Palace Road, London SE1 7EH, United Kingdom.


Secondary prevention trials have found that cholesterol reduction lessens the incidence of recurrent cardiac events and slows or reverses the progression of angiographically measured coronary atherosclerosis. The STARS trial, a 3-way comparison of usual care, diet alone, and diet with cholestyramine, affirms that cholesterol reduction diminishes the angiographic severity of coronary atherosclerosis and also shows that increases in the coronary segment width are inversely correlated with both the low density lipoprotein (LDL) cholesterol level and the LDL/HDL (high density lipoprotein) ratio. Like many angiographic studies, this trial is too small to detect changes in the rates of most clinical events, but both active treatment arms of the trial had significantly fewer total cardiac events. The patients treated with medication in combination with diet had the lowest rate of events. The selection criteria for the trial favored cholestyramine, however, because persons who did not tolerate or respond to cholestyramine during a brief test period were removed.

Although this study confirms that plaque regression may be one of the mechanisms by which cholesterol reduction lowers the incidence of recurrent coronary heart disease, it leaves several other important clinical questions unanswered: At what cholesterol level should treatment be instituted? What cholesterol level should be the goal of treatment? Should drugs be included as the initial therapy?

Drugs for primary prevention are only cost effective at very high cholesterol levels and in the presence of other cardiac risk factors, but presumably drug treatment of persons with pre-existing coronary disease should be initiated at lower cholesterol levels because the benefits of secondary prevention are much greater. Although this study suggests that drugs provide additional benefit in persons with coronary heart disease, larger trials are needed to evaluate alternative dietary and pharmacologic strategies.

Alan M. Garber, MD, PhD
Stanford University and Palo Alto Veterans Affairs Medical Center Stanford, California