Skip navigation

Food for Thought: Cholesterol Lowering in Patients with Coronary Heart Disease

Does lowering serum cholesterol improve survival in patients with coronary heart disease (CHD)? This loaded question may now be answered following the results of an in-depth study from Scandinavia, the 4S study [1].


Men and women aged 35-70 years with a history of angina or acute myocardial infarction (MI) were screened for eligibility. If none of the exclusion criteria applied and the patient consented, fasting serum cholesterol and triglyceride were measured. If, after 8 weeks of dietary advice cholesterol was 5.5 - 8.0 mmol/L or triglyceride over 2.5 mmol/L, patients were randomised to simvastatin 20 mg or placebo to be taken before the evening meal.

4444 patients were so randomised between May 1988 and August 1989, and were followed up for a median of 5.4 years with frequent clinic visits for lipid and other measurements.


The primary endpoint was mortality.

The secondary endpoint was a major coronary event - coronary death, non-fatal MI, resuscitated cardiac arrest and silent MI confirmed by electrocardiogram.

The tertiary endpoints were any coronary event (secondary endpoint plus cardiac revascularisation procedures), non-coronary atherosclerotic events, incidence of heart operations and hospital admission for acute CHD without a diagnosis of MI.


Lipid concentrations showed little change in the placebo group over three years except for triglycerides, which rose by 7%. In the simvastatin group over the whole course of the study there were substantial (25-35%) falls in total and LDL cholesterol, a small (10%) fall in triglycerides, and a small but useful increase of 8% in HDL cholesterol. Doses needed to be increased to 40 mg a day in 37% of patients.

There were fewer deaths, and especially coronary deaths, in the treated patients. The NNT for simvastatin to prevent a coronary death was 29.

There were fewer non-fatal coronary events in the treated patients. The NNT to prevent a major coronary event was 15. Fewer treated patients had coronary surgery or angioplasty. The NNT to prevent one episode of coronary surgery was 17. Overall secondary endpoints were much reduced in treated patients. The NNT to prevent any secondary endpoint was 12.

The overall frequency of adverse effects was similar in the two groups, with 6% in each group discontinuing treatment because of adverse effects. A single case of rhabdomyolysis occurred in one woman taking simvastatin. It resolved when treatment stopped.


This is an important study, well conducted over many years, collecting a very considerable amount of information on many patients. The question now is whether cholesterol lowering should be a target in all patients with coronary heart disease and serum cholesterol >5.5 mmol/L?

Addition of simvastatin to the treatment regimens of 100 CHD patients would yield the following benefits over the first six years: preservation of the lives of four of the nine patients who would otherwise die of CHD; prevention of non-fatal MI in seven of an expected 21 patients; avoidance of revascularisation procedures in six of the 19 patients in which it would otherwise be done.

The 4S study will take a little while to digest, but will aid considerably those working on guidelines.

Economic analysis

The economic analysis remains to be published. The report contains all the details needed for a quick "back of envelope" estimate of whether there are any economic benefits from introducing this expensive treatment. What NNT is needed for economic benefits to accrue?


1: Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994 344: 1383-9.

previous or next story in this issue