Is there a role for screening asymptomatic patients with diabetes?

Abstract

Coronary artery disease (CAD) remains a leading cause of death among patients with diabetes mellitus. However, many patients with diabetes and CAD are asymptomatic and may sustain a myocardial infarction as their presenting symptom of CAD. Non-invasive cardiovascular imaging offers an opportunity to detect the presence and severity of CAD, or its hemodynamic consequences. The Detection of Ischemia in Asymptomatic Diabetics study and the FACTOR-64 study examined the utility of non-invasive imaging tests to evaluate asymptomatic individuals with diabetes mellitus. The results of these trials may have been negative with regard to promoting CAD screening of asymptomatic diabetic patients, but they do strengthen the position of optimal medical management in reducing cardiovascular events. However, performing a trial to include true high-risk patients who have CAD and are more likely to have silent ischemia could lead to prognostically beneficial coronary revascularizations.

Keywords:

• asymptomatic
• calcium score
• coronary CT
• diabetes
• imaging

Coronary artery disease (CAD) remains a leading cause of death among patients with diabetes mellitus (DM). However, many patients with diabetes who have CAD are asymptomatic and may sustain a myocardial infarction as their presenting symptom of CAD [1]. In some patients, myocardial infarction may also occur without any symptoms. Consequently, in order to reduce the burden of cardiovascular disease in this cohort, there is an important role for early identification and treatment of CAD. Then again, some have argued whether we should bother identifying CAD, suggesting that perhaps a ‘treat all’ approach might be more effective. In light of recent studies in this field, it is now appropriate to ask: is there any role for screening asymptomatic patients with diabetes?

Non-invasive cardiovascular imaging offers an opportunity to detect the presence and severity of CAD, or its hemodynamic consequences. The underlying premise of such testing is that the identification of disease may lead to treatments (that are otherwise not being offered) that will improve cardiovascular outcomes. However, recent data have convincingly challenged the concept of widespread testing for CAD among asymptomatic patients with diabetes.

To date, two large trials have examined the utility of non-invasive imaging tests to evaluate asymptomatic individuals with DM. The first trial, the Detection of Ischemia in Asymptomatic Diabetics (DIAD) study, investigated whether routine screening for ischemia with adenosine-stress myocardial perfusion imaging (MPI) can improve cardiac outcomes [2]. This trial randomized 1123 participants with diabetes and no symptoms of CAD to myocardial perfusion imaging (MPI) versus no testing. Over a mean follow-up of nearly 5 years, there was no difference in the primary end point of cardiac death or nonfatal MI (HR: 0.88; p = 73). Notably, the authors noted a lower than expected cardiac event rate (2.9% over 4.8 years). These findings were thought to reflect the fact that patients in both groups benefitted from excellent medical management as verified by their lowered hemoglobin A_1C and blood pressure values as well as the fact that >40% of patients were on aspirin and at least 70% of patients were on a statin at baseline.

The more recent trial, the FACTOR-64 study, randomized 900 asymptomatic patients with diabetes to undergo screening with 64-slice CCTA or standard treatment [3]. An important difference between this trial and the DIAD study is that the CCTA results classified patients into low-, medium- and high-risk groups and accordingly different treatment algorithms. After a mean follow-up of 4 years, annual event rates in both the control and intervention groups were low (<2%) and the primary outcomes (death, MI or unstable angina) did not differ significantly (hazard ratio 0.80 [95% CI: 0.49–1.32]; p = 38). Similar to the DIAD study, the baseline characteristics of the patient population exhibited one that was well treated. Despite a mean duration of DM greater than 12 years, the mean HbA_1C, LDL cholesterol (LDL-C), systolic blood pressures were close to target goals at about 7.4%, 87 mg/dl and 130 mmHg, respectively. This study was conducted in the Intermountain Healthcare system, which initiated a Diabetes Prevention and Management team in 1997 that provided system-wide standards and therapeutic guidance for its practitioners. Therefore, the low-event rate and nonsignificant differences between the control and intervention groups has been ascribed to the excellent medical management received by all enrollees.

The results of these two negative studies advocating screening for CAD in asymptomatic diabetic patients were not necessarily unforeseeable. First, as previously mentioned, the patients in these studies are likely not the ‘high-risk’ diabetic patients that may have benefited from screening. Compared to prior studies of asymptomatic diabetic patients referred from non-invasive imaging tests that demonstrated an annual event rate of approximately 3–5%, the DIAD and FACTOR-64 studies displayed an annual event rate <2% [4,5]. In fact, in the DIAD study, 78% of patients had a normal MPI study. Moreover, in the FACTOR-64 study, of the patients enrolled into the screening arm, 31.3% of patients had a normal CCTA and 58% had non-obstructive CAD.

The FACTOR-64 study projected an ambitious 40% reduction in major cardiovascular events based on a treatment algorithm, which included both risk factor modification and coronary revascularization. Although there was a significant difference in statin use among the intervention group versus the control group (6.6 vs 3.7%, p = 05), given that 76.5 vs 72% of patients, respectively, were already on a statin at baseline, it is not surprising that the rather modest improvements in the LDL-C values with a mean decrease of 2.64 mg/dl did not lead to decreased cardiac events. Furthermore, even while randomization to CCTA leads to more protocol-related revascularization procedures (17 percutaneous interventions and 7 coronary artery bypass graft surgeries), any effect of coronary revascularization for severe CAD on outcomes was likely diminished by its lower than expected rate of application as well as the fact that such procedures may not offer a significant improvement in outcomes in patients who are asymptomatic and who are on excellent medical therapy.

Although these trials may have been negative studies in promoting CAD screening of asymptomatic diabetic patients, they do strengthen the position of optimal medical management in reducing cardiovascular events. Along with the results of the BARI-2D and COURAGE trials, these large studies raise the question if coronary artery revascularization offers any improvement in optimal medical management [6,7].

Given these findings, should we ever screen individuals with diabetes….or should we just treat them all with aggressive medical therapy? We should also ask, among patients who may be candidates for screening, what is the best technique?

The answers to these questions depend on the population of patients being considered, the direct and indirect cost of testing and the values of patients and the healthcare system on widespread preventive therapies. While the above aforementioned studies could be used to advocate widespread preventive therapies for patients with diabetes, it is now recognized that some individuals with diabetes have a low risk of cardiovascular disease. For instance, in the Diabetes Heart Study, patients with diabetes with a coronary artery calcium (CAC) score of 0–9 had a 0.9% annual mortality and the risk of mortality increased with increasing levels of CAC scores [8]. Similarly, in the multiethnic study of atherosclerosis 38% of participants with diabetes had a calcium score of 0, a sub-group that experienced a significantly lower event rate with annual rate of coronary heart disease of 0.4% per year [9]. Therefore, it is plausible that among selected non-high risk diabetic patients who are considering initiation or intensification of preventive therapies, testing for coronary plaque may provide a useful risk assessment tool, whereby at least 1 out of every 4 individuals tested may have calcium score of 0, and may defer aggressive treatments while still focusing on lifestyle measures. Although a coronary artery calcium score (CACS) of 0 does not preclude the presence of non-calcified coronary artery plaques, it has been demonstrated that patients with a calcium score of 0 have a low likelihood of obstructive CAD as well as an excellent prognosis. The prevalence of noncalcified plaques and significant stenosis by noncalcified plaque is significantly higher in patients with a low CACS compared to those with 0 CACS [10–12]. Accordingly, the 2010 ACC/AHA guidelines (Class IIA, level of evidence B) advocate the measurement of CAC for cardiovascular risk assessment in asymptomatic adults with diabetes, 40 years of age and older [13–15]. Further data, however, are needed regarding the efficacy of such an approach of using calcium scoring to ‘de-risk’ individuals with diabetes. However, based on the available data, we propose that calcium scoring might represent the ideal risk assessment tool in such populations, as it is less costly than coronary CTA and, when used appropriately, is less likely to result in excess revascularizations.

On the other end of the spectrum, are the true high-risk individuals with diabetes. Such individuals should all be treated with aggressive medical therapies (the definition of which is continuously evolving). Among such patients, screening for silent ischemia (or high grade, multivessel obstructive CAD) could lead to prognostically beneficial coronary revascularizations. Given the null findings of the DIAD and FACTOR-64 studies, it seems that further trials in this field will be challenging to do, but it is conceivable that a trial designed in higher risk patients may be plausible. Such a trial would require a careful design to include true high-risk patients who have CAD and who are more likely to have silent ischemia. Criteria for inclusion for such a study may include blood biomarkers together with other risk factors (e.g., duration of having diabetes, need for insulin), and possibly CAC scoring (e.g., inclusion of only those with CAC > 100 would result in an event rate which is higher than observed in the above trials) [16,17].

Currently, there is no strong evidence to support screening asymptomatic diabetic patients for CAD, as the available data has not shown a significant reduction in major cardiovascular events associated nuclear stress testing or coronary CTA. However, we should be careful in generalizing these results across all populations of patients with diabetes, as it remains possible that larger studies that include high-risk patients could ultimately identify sub-groups of patients in whom screening for CAD could improve outcomes. In the meantime, aggressive medical and lifestyle therapies should remain the cornerstone of effective patient management.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.