For decades, scientists have been trying to find out what causes heart disease. Although we have gained some knowledge along the way, many questions remain unanswered. Instead of providing clear answers, scientific research often raises a set of new issues that have to be addressed.
The American physicist, Thomas S. Kuhn wrote that “to be scientific is among other things to be objective and open minded” (1). Throughout history, preconceived notions and conceptual misunderstanding have restrained scientific progress and, unfortunately, continue to do so in the 21st century.
Smoking, hypertension, diabetes and high blood levels of cholesterol are examples of risk factors that have been associated with increased risk of cardiovascular disease (CVD). Modification of these factors may affect the probability of having an event such as a heart attack (myocardial infarction) or stroke.
The risk of CVD increases with age. However, most intervention trials testing the efficacy of preventive measures have focused on middle-aged individuals. This may be misleading because risk factors may vary with age. Hence, a particular risk factor may have more impact on a young individual than an older person. In theory, this might, for example, mean that high cholesterol matters more when we’re young than when we’re older.
In a very interesting paper published recently in the European Heart Journal, Gränsbo and coworkers from Malmö, Sweden address the question whether risk factor exposure in individuals differs according to age at first myocardial infarction (MI) (2). In other words, are smoking, high cholesterol, diabetes and hypertension as bad for the heart health of a 40-year-old as a 75-year- old? Of course this is a very important scientific question and the study has several strengths. It is based on a large, solid database with virtually no loss to follow-up.
Interestingly, the paper is discussed in an editorial written by Dr. Eugene Braunwald, a world renowned and highly influential U.S. cardiologist (3). Although of much less quality than Gränsbo’s paper, and pretty much off target, Barunwald’s discussion of the study findings makes for an interesting read.
Risk Factor Exposure and Age at First Myocardial Infarction
The study by Gränsbo and coworkers addresses the presence of risk factors in individuals who later develop MI early and late in life. Their analysis is based on data from a population-based registry in Malmö, Sweden. They enrolled 33.346 individuals without cardiovascular disease and who were followed for an average of 22 years.
MI developed in 3.687 patients ranging in age from 37 to 84 years, each of whom was compared with 1-2 age-matched controls who did nod develop a cardiovascular event. The main findings of this analysis were that blood levels of cholesterol and family history of MI are much stronger risk factors in younger age than in older subjects.
On the other hand, diabetes, blood triglycerides, and hypertension were equally associated with risk of MI at all ages. Body mass index was weakly associated with risk and displayed no difference with regards to age. The authors conclude that obesity-mediated risk is primarily mediated through obesity-related risk factors such as diabetes and triglyceride levels.
“Furthermore, I consider that Carthage must be destroyed”
Dr. Eugene Braunwald’s editorial addressing the study by Gränsbo and coworkers is entitled “Reduction of LDL-Cholesterol: important at all ages.” This is interesting because the study did not address LDL cholesterol at all.
Dr. Braunwald underscores the importance of considering both the relative and absolute risk when discussing the association of risk factors with age. He cites earlier findings provided by the Prospective Studies Collaboration, which combined the results of 61 prospective observational studies relating total blood cholesterol to vascular mortality in almost 900.000 subjects (4).
According to Dr. Braunwald, the results from this large database confirm that cholesterol is a much more potent risk factor for coronary heart disease in younger than older subjects. However, the incidence of CVD is much lower in the younger compared to the older age groups. Hence, much more could be gained in absolute terms by lowering cholesterol in older than younger individuals.
Dr. Braunwald believes that the implications of the Prospective Studies Collaboration’s and Gränsbo’s analysis “is that efforts to reduce total cholesterol (and its important fraction, LDL cholesterol) should be undertaken as soon as an elevation is recognized, and these efforts should be lifelong”.
However, this is not at all what Gränsbo’s paper was about. It wasn’t about LDL cholesterol and the under-use of statins. It was about how the influence of risk factors varies with age.
Why is an MI in younger patients preceded by a different risk factor pattern than an MI in older subjects? Why does the impact of cholesterol soften with time? Is it possible that the role of other factors, such as inflammation and insulin resistance increase with age? Or, are the pathological mechanisms underlying MI in the young merely different from that in the old? Dr. Braunwald’s input on those issues would have been highly appreciated. But, unfortunately, he found it more important to stick with the old cliché.
Dr. Braunwalds editorial made me think of Cato the Elder’s words: “Ceterum autem censeo Carthaginem esse delendam” “Furthermore, I consider that Carthage must be destroyed.“ This oratorical phrase was in popular use in the Roman Republic in the 2nd Century BC during the latter years of the Punic Wars against Carthage. It was frequently uttered and persistently almost to the point of absurdity by the Roman senator Cato the Elder, as a part of his speeches.”
Cholesterol and the Elderly
Dr. Braunwald believes that “In the 2013 Guidelines of the American College of Cardiology and American Heart Association, the need for intensive therapy to reduce LDL-cholesterol for subjects older than 75 years in both primary and secondary prevention has been de-emphasized.”
He writes: “With respect, I voice my discomfort with this position on the elderly, who are the most rapidly growing segment of the population. The Cholesterol Treatment Trialists (CTT) Collaboration, in a meta-analysis of prospective clinical trials, reported a significant relative risk reduction of vascular events of 16% in patients over the age of 75 years receiving statin therapy”.
Let’s look closer at the CTT Collaboration’s study (5). Above is the data that Dr. Braunwald refers to.
Among those aged above 75, the risk of major vascular events per annum was 4.8% on statins compared with 5.4% in the control group. The relative risk reduction associated with statin therapy and a reduction of LDL cholesterol by 1.0 mmol/L is 11.1% (5.4 – 4.8/5.4 x 100), whereas the absolute risk reduction is 0.6% (5.4 – 4.8). The number needed to treat (NNT) is 167 (100/0.6). In other words, 167 individuals need to be treated for a year to prevent one major vascular event. But, importantly, no beneficial effect on mortality has been proven in this age group.
It is important to acknowledge that the above study addressed trials on primary and secondary prevention combined. In fact, the effect of statins becomes less impressive when looking solely at studies in primary prevention, e.g. patients without established cardiovascular disease.
A recent meta-analysis addressed studies on the effect of statins in elderly subjects (above age 65) without established CVD (6). The absolute risk reduction for one year was 0.34% for MI. The corresponding NNT for a year will be 294. It is important to acknowledge that the NNT was miscalculated by a factor of ten in the original publication (7).
Let’s take a closer look at the magnitude of this treatment effect. If treated with statins for a year, the probability of not having a heart attack will be increased from 98.9 percent to 99.2. percent. These numbers are hardly convincing, considering possible side effects of therapy.
Still, Dr. Braunwald concludes that “reductions of LDL cholesterol of this magnitude (1 mmol/L) can, of course, be readily obtained by intensive treatment with statins, most of which are now generic, inexpensive, and generally well tolerated. Frankly, I do not see the reason for recommending less intensive lipid-lowering therapy for the elderly, unless, of course, they have a life-threatening co-morbidity or are statin intolerant.”
Interestingly, Dr Braunwald does not mention lifestyle. Is urging doctors to prescribe statins more important than promoting a Mediterranean style diet and regular physical exercise?
Because of my deep and sincere respect for Dr. Braunwald and his achievements, it hurts to admit that I’m deeply disappointed. For me, his editorial appears off the mark, and the argumentation for more statin use among the elderly is contentious at best. It’s not all about the cholesterol anymore.
Maybe the German physicist, Max Planck was right when he said: “A new scientific truth is not usually presented in a way that convinces its opponents…; rather they gradually die of, and a rising generation is familiarized with the truth from the start.”