Prevention of Cardiovascular Disease: Time for a Course Correction

Allan D. Sniderman; George Thanassoulis; Michael Pencina


Eur Heart J. 2020;41(31):3016-3017. 

To prevent cardiovascular disease, all guidelines recommend better health habits for everyone but recommend statin therapy only for those at high risk of a cardiovascular event. High risk includes those with markedly elevated levels of LDL-cholesterol (LDL-C) or diabetes and, the largest group by far, those with a calculated 10-year risk of a cardiovascular event above a threshold level. Ten-year risk, therefore, is the primary tool to select subjects for primary prevention with statins. As confidence in the safety and the efficacy of statins increased and as the cost of statins decreased, the major guideline groups broadened their indications for statin eligibility. The latest ESC/EAS guidelines[1] extend the age of those eligible from 65 to 75 years, and lower selection based just on LDL-C from >6 mmol/L (232 mg/dL) to ≥4.9 mmol/L (190 mg/dL) and for selection based on risk from LDL-C ≥4.0 mmol/L (155 mg/dL) to ≥2.6 mmol/L (100 mg/dL).

Guidelines come with costs and consequences for society. Estimates of these costs and consequences compared with current practice are often not available in guideline reports. Fortunately, in this issue of the European Heart Journal, Mortensen and Nordestgaard compare application of the 2019 with the 2016 ESC/EAS guidelines within the 45 750 subjects of the Copenhagen General Population Study.[2] The results are not generalizable to all of Europe, but the general features should be relevant. Statin eligibility increased from 15.4% [95% confidence interval (CI) 15.1–15.7] to 32.3% (95% CI 31.8–32.7). Of the increase, 33.2% was due to lower LDL-C thresholds, 49.4% to extending the age range from 65 to 70, 14.7% to a combination of these, and 2.8% to a change in the SCORE algorithm. If perfectly implemented—obviously an unachievable outcome—25% rather than 11% of cardiovascular disease (CVD) events could be prevented, with a number needed to treat (NNT) of 19 rather than 20. The bottom line is: the number eligible for statin therapy would be doubled and so would the reduction in clinical events. Treat more. Save more.

But is this enough? Is selection based on 10-year risk all we need to do? We argue that it is not. This analysis of the EAS/ESC guidelines confirms a previous analysis of the ACC/AHA guidelines[3] that demonstrated that prevention based principally on 10-year risk is succeeding in older adults but is not succeeding nearly as well in younger adults. Table 1 integrates Supplementary tables S1 and S2 of Mortensen and Nordestgaard.[2] In Copenhagen, those under 65 make up ~80% of the population, of whom 26% of men and 16% of women would be eligible for prevention. Those over 65 make up ~20% of the total, but 91% of men and 75% of women over 65 would be eligible. In both sexes, the average age of those who are statin eligible who are less than 65 years old is ~60. In those over 65, 36% of events would be prevented by full implementation of high-intensity statin therapy whereas in those under 65, only 17% would.[2]

How does this happen? The conventional risk strategy integrates the effects of the most potent factors that determine risk and then selects for statin preventive therapy those most likely to suffer a CVD event. Creating the 10-year risk algorithm was a brilliant achievement, but the limited period of time (10 years) does not capture the true risk for younger adults over the midlife period. Age dominates CVD risk algorithms. In this study, the average SCORE risk is six times greater in those over 65 compared with those under 65. However, the younger group is much larger than the older group and, therefore, the younger group accounts for the majority of cases. This effect is particularly dramatic in this analysis, but in the USA, as a whole, one-half of all CVD events in men and one-third in women occur before age 65.[4] Not only will the 10-year risk approach miss many cases in younger subjects, but the chance to intervene in many younger subjects in whom atherosclerosis has begun slowly, but inexorably, to destroy the normal architecture of their arterial walls, creating the pathology, which will eventually kill or injure them, will be missed as well. CVD events after 65 are often, perhaps even largely, the outcome of a disease process, which began well before 65.

How can this be fixed? Taken together with previous results,[3,4] this analysis demonstrates that to move forward we must change the model for prevention in those who are younger. Almost all of us are at significant lifetime risk of CVD.[5] If this was the standard, the outcome would be a 'statin for all' strategy from an early age. This seems excessive. However, a strategy based on longer term risk and the benefit of therapy offers a more measured and productive approach to improve prevention in those under 65.[6,7] Thus, those with high levels of the apoB lipoproteins at age 30–40 may be at low 10-year risk, but are at substantial 30-year risk of a CVD event over the next 30 years,[8] arguably the most critical personal and professionally important period of their lives.

The potential benefit of statin preventive therapy is determined by the risk of an event over the period of interest plus the baseline level of the causal factor producing the risk. Based on this approach, we[6] and others[7] have shown that younger individuals with higher levels of LDL-C, but lower 10-year risk, receive the same benefit from therapy even within the shorter term as older individuals at higher risk but lower levels of LDL-C with similar NNT. Importantly, these analyses do not take into account the longer term potential benefits from statin therapy. In summary, the present analysis of the most recent ESC/EAS guidelines[2] adds to the evidence that moving from a 10-year risk strategy for everyone to a longer term benefit strategy for those who are younger will improve the overall prevention of CVD.