Atherosclerosis Imaging of Asymptomatic Individuals

Abstract

A LTHOUGH APPEALing, screening for subclinical vascular disease remains a controversial topic. In addition to exercise stress testing, which I will not discuss herein, atherosclerosis detection technologies are available as a potential means to refine risk prediction. Such procedures include computed tomography for coronary calcification, ultrasound for carotid intima media thickness, ankle-brachial index, and brachial artery reactivity. They are expensive, induce further expensive testing, and their use is burgeoning despite what I believe is insufficient evidence to support such practice. I present an argument that opposes the practice of screening for atherosclerosis in asymptomatic individuals at this time, consistent with the US Preventive Services Task Force review that there is insufficient evidence to support its routine use. It is intuitive to feel that somehow we need to do more, and do more earlier in the disease process, to better combat the menace of atherosclerotic disease. The scope of global illness burden attributable to cardiovascular disease is higher than any other single disease entity and is projected to continue to be so until at least 2020. The debate essentially revolves around whether our current strategies of predicting risk are good enough and whether newer tools improve on what we already do. Risk prediction tools explain up to 70% to 80% of the variance of cardiovascular disease using conventional risk factors, and almost all cardiovascular events occur in people with at least 1 conventional risk factor. Thus, regardless of whether new prediction technologies help improve risk, the best they can do is provide small incremental improvements in risk prediction because conventional risk factors are already providing the majority of clinically relevant information and are the main targets for interventions. Primary prevention of cardiovascular disease must continue to focus primarily on the detection and treatment of conventional cardiovascular risk factors. Nevertheless, risk prediction tools are not perfect. They are limited by their inability to accurately predict events beyond 10 years and are not useful for younger populations who are at low absolute risk but high relative risk and therefore high lifetime risk for cardiovascular disease. Most of the adult US population has at least 1 cardiovascular risk factor, yet the majority of US adults will never develop clinical cardiovascular disease in their lifetimes. This means that conventional risk factors are sensitive for the detection of cardiovascular risk but are not specific. I therefore agree with the rationale to identify tests, such as atherosclerosis imaging, that further refine the specificity of risk prediction. Although there is ample evidence to date, in limited populations, that the presence and extent of coronary calcification, carotid intima media thickness, ankle-brachial index, and brachial artery reactivity are associated with an incremental increased risk for cardiovascular events, there is still much work to be done in this area. There are several important prerequisites to consider before concluding that such testing would be ready for widespread use: (1) the test information must provide additive prognostic value to conventional risk prediction; (2) there should be improved outcomes associated with use of the technology; and (3) there must be reasonable costs associated with the test itself as well as the induced costs associated with abnormal findings. I will focus the following discussion primarily on coronary atherosclerosis imaging, but the principles of our argument apply to any method of atherosclerosis detection technologies.