Ever since CHD (coronary heart disease) moved into first place as a major killer in the West, scientists have once again been busy concocting and testing prevention theories. In the absence of an understanding of the cause, the risk factor concept has been involved to deal with the problem: the main risk factors identified being smoking, high blood pressure, raised serum cholesterol, diabetes etc.
However, reduction of even those risk factors presumed to be the most important—such as raised serum cholesterol—has not been an effective means of preventing the disease, as shown for instance by the results of the $115 million Multiple Risk Factor Intervention Study. (MRFIT).
This comprehensive study, involving 12,500 men who smoked and had raised blood cholesterol and blood pressure values, was designed to test the effect of risk factor modification on the incidence of CHD over a period of 6 years.
The conclusion reached was that it was not possible to control CHD by the simultaneous modification of several risk factors ( drug-induced reduction of cholesterol and substitution by polyunsaturated fatty acids of saturated fats in the diet, cessation of smoking as well as several other risk factors.)( JAMA 1982; 248)
The editorial in the British Medical Journal (1993; 306) concluded that “the lack of evidence of significant reductions in all causes of mortality by lipid-lowering drugs, particularly in primary prevention trials, indicates that any favourable effect of lipid-lowering drugs on all causes of mortality is as best small and of only marginal benefit to the individual.”
This mounting scepticism among doctors was virtually swept aside in late 1994 by the publication and subsequent publicity of a single trial, the Scandinavian Simvastatin Survival Study (4S), which appeared to vindicate a new group of cholesterol-lowering drugs, called statins, at least among patients with a heart condition and high cholesterol levels.
After 5 and a half years the group given cholesterol-lowering drugs had a one third reduction (RRR)* in heart disease and a 41% lower rate (RRR)* of heart attack compared to the placebo (inactive pill) group. (Lancet 1994; 3440)
Hard on the heels of the 4S research was the West of Scotland Coronary Prevention Study (WOSCOPS) This study purported to show that Pravastatin, another statin cholesterol-lowering drug, could prevent heart attacks by 28% (RRR)* in men with high levels of cholesterol but no history of heart disease. (NEJM 1995; 333) The effect that these conclusions had on the medical rank-and-file, was galvanic. All patients with higher cholesterol levels, irrespective of age or sex, were being placed on cholesterol-lowering drugs for life. (Lancet 1996; 347)
The 4S study contained a number of basic flaws. In the drug treatment group, 38 people had previous bypass surgery or angioplasty, and were therefore less likely to die. There were also 54 more smokers in the control group, placing them at a higher risk for CHD. (Lancet 1995; 345)
Although not statistically significant, slightly more women died from other causes in the 4S study. In the WOSCOP study, deaths from heart disease among those not given the cholesterol-lowering drugs, were more than that of the general population—closer to an average death rate in people at least 10 years older- suggesting that particular people selected to represent the “average citizen” happened to be more ill than usual. (Lancet 1996;347). The trial also only showed a 0.9% drop in absolute total mortality over 5 years in people taking the statins.
Absolute total mortality (all-cause death rate) and relative risk reduction (RRR)*
Before considering the effectiveness and safety of any prescription drug, these statistical definitions must first be understood. Absolute total mortality refers to the actual difference in risk reduction between the treated and non treated groups.
Using the absolute total mortality rate to measure effectiveness, ensures that while a drug might prevent the targeted disease, it does not accidentally kill you from another deadly disease.
Relative risk reduction—the percentage of the decrease achieved by the treated vs. non treated group—exaggerates benefits.
The supposed benefits of the statins are usually given as lowered relative risk, of mostly non-fatal heart attacks without the slightest indication of the low magnitude of the more meaningful reduction of absolute risk of all-cause death rates. (Gigerenzer 2002). Lowering cholesterol and LDL levels are also highlighted, despite existing knowledge that the beneficial effects of statins on CHD are independent of either initial or achieved levels, and thus a useless surrogate endpoint. (Nielsen J V et al BMJ 2001 323)
Statin Drug Trials
A meta-analysis of 5 major statin drug trials, these being PROSPER, ALLHAT-LLT, ASCOT-LLA, AFCAPS and WOSCOPS failed to show any significant benefit to using statins. In the pooled data of these trials, statin drugs provided an absolute risk reduction in total mortality of 0.3% among those who showed no signs of having cardiovascular disease (primary prevention). (Therapeutics Initiative 2003) With respect to preventing heart attack and stroke, the five combined studies showed that statins prevented these events by a mere 1.4%.
Using LIPS, PROSPER, GREASE and HPS, a meta-analysis showed that statin use provided an absolute risk reduction in total mortality of 1.8% among those who showed signs of having cardiovascular disease. (secondary prevention.)
A meta analysis (2003) of 44 trials involving almost 10,000 patients, showed an identical death rate of 1% in each of three groups—those taking atorvastatin (Lipitor), those taking other statins and those taking nothing.. (Hecht et al 2003; Am. J. of Cardiol. 92)
In the ASCOTT-LLA trial (Sever P S et al 2003; Lancet; 361) on hypertensive patients with average or lower total cholesterol, there was no change in the all-cause death rate between atorvastatin (Lipitor) treatment vs. placebo after 3.5 years. However, there was a higher incidence in the arbitrarily defined primary endpoint of non-fatal heart attack plus fatal cardiovascular disease among the 19% women in the trial.
The most favourable trial, with minimal financial conflict of interest and seemingly impeccable reporting, the Heart Protection Study (2002), yielded users, on simvastatin for 5 years, an all-cause death rate drop of 0.38% per year. (Lancet 2002; 360)
Gender, Age and Ethnic bias
Statin drug trials, especially those conducted between 1990 and 1999, suffer from age and gender bias. These drug trials were mainly conducted using white middle-aged men, and did not study the effects among women, the elderly and ethnic groups. (Bandyopadhyay S et al 2001; Quarterly J. of Med: 94)
Consumers have filed (28 September 2005) the first-of-its-kind nationwide lawsuit against Pfizer, the manufacturer of atorvastatin. (www.lipitor-classaction.com)
The lawsuit alleges that Pfizer engaged in a massive campaign to convince both doctors and patients that Lipitor is a beneficial treatment for nearly everyone with elevated cholesterol, even though no studies have shown it to be effective for those over 65, and women of any age who do not already have heart disease or diabetes.
The dangers of statin drugs are rarely discussed. The British Medical Journal has reported that of 164 statin drug trial reviewed, only 48 reported the number of participants with one or more negative side-effects caused by the drug. (Law et al 2003. BMJ June 28)
While the benefits have been hyped beyond belief, statins have many more unpleasant side-effect than have been admitted.
Gigerenzer G, 2002 Calculated risks: How to Know when Numbers Deceive you.. New York. NY Simon & Schuster
Ravnskov U, 2000 The Cholesterol Myths, Washington, DC, New Trends
Therapeutics Initiative: Evidence Based Drug Therapy. Do statins have a role in Primary Prevention? Apr-May-June 2003. The Univ of British Columbia. www.ti.ubc.ca