Cardiac CT in asymptomatic diabetes mellitus: role of non-invasive atherosclerosis imaging in high-risk asymptomatic individuals

The negative prognostic importance of diabetes mellitus (DM) on cardiovascular outcomes is well known. DM is considered as coronary artery disease (CAD) equivalent, which places an asymptomatic individual with DM in the high-risk group. Direct coronary imaging of CAD in asymptomatic high-risk individuals is controversial. In particular, the role of non-invasive cardiac computed tomography (CCT) in asymptomatic diabetics is less well understood.1–6 In this issue of the Journal, Kim et al.6 shed light on this topic with the results of the CRONOS-ADM prospective registry (CoROnary CT aNgiography evaluation for clinical OutcomeS in Asymptomatic patients with type 2 Diabetes Mellitus). These data provide insight into several aspects of the DM relationship with time and CAD extent using CCT in an asymptomatic population of individuals. The relevant findings of the study are that DM duration is associated with progression in the extent and severity of CAD which independently predicts an increased risk of major adverse cardiac events (MACE) beyond traditional risk factors. Therefore, as already demonstrated in general population of patients with suspected CAD assessed with CCT, CAD burden seems to be a strong predictor of events. In addition, the prevalence of normal coronary arteries on CCT was progressively reduced with increased DM duration (from 31% with DM duration <5 years to 16% with DM duration >10 years). For professionals dealing with cardiac imaging and CCT in particular, this is not an unexpected finding. However, the information provided strengthens the role of coronary imaging in the specific setting of asymptomatic DM individuals. Despite the increased incidence of myocardial ischaemia, often silent, in diabetics, routine screening for CAD is controversial although it may be considered in selected patients deemed to be at high risk. Based on the present findings, we can consider that a referral policy for exercise testing (or other functional tests) built on a standard cardiovascular risk assessment in patients with dysglycaemia may be overcome by performing a direct anatomical imaging approach particularly in DM individuals with >10 years of disease. This is reinforced by the consideration that the warranty period of a normal CCT is quite long.4 European Society of Cardiology (ESC) guidelines on DM state that there is a need for biomarkers and diagnostic strategies useful for the early detection of CAD in asymptomatic patients; the guideline discusses about the potential of coronary artery calcium score (CACS) and recommends screening for silent myocardial ischaemia in selected high-risk patients (Class IIb/C).7 ESC guidelines on prevention in clinical practice put invasive (coronary angiography) or non-invasive testing (nuclear imaging, stress echocardiography, and carotid plaque on ultrasound) as a priority only in very high-risk individuals with DM (type 1 or 2) with ≥1 CV risk factors and/or target organ damage.8 These guidelines do not refer to CCT for risk assessment; in this population, there is a lack of connection between studies on prevention in asymptomatic DM and novel coronary imaging techniques such as CCT. Bridging the gap is necessary to better understand appropriate endpoints for treatment management of DM patients. There are limitations in the CRONOS-ADM study. It is a prospective registry and therefore it may gather incomplete information (as acknowledged by authors) such as details about patients' medications and glycaemic control as reflected by missing HbA1C levels in the pre-CCTA period and during 5-year follow-up. Another observation is related to CCT indication. To date, asymptomatic DM is not a sharp indication for CCT and this may create a selection bias towards individuals more suitable for a CCT investigation (e.g. abnormal ECG or rest echocardiography). However, the present study further challenges the notion of DM as a known CAD equivalent. Nonetheless, due to the progressive nature of CAD and its relationship with DM duration and microvascular disease, the paper by Kim et al. indirectly raise concerns regarding a more aggressive treatment of risk factors early in the natural history of the disease. The recent FACTOR 64 trial showed that among asymptomatic patients with DM, the use of CCT to screen for CAD did not reduce the composite rate of all-cause mortality at 4 years; however, the authors acknowledged as a major limitation of the study the low incidence of adverse events, limiting power to exclude a clinically important difference between the study groups.9 Even the accompanying editorial by Dr Gibbons points out that negative results also may reflect the inability to successfully implement the more aggressive therapy recommended for patients with abnormal CCT findings and that trial was underpowered because patients in the control group did much better than anticipated in the sample size calculation because of their excellent baseline medical therapy.10 In a sub-study of the CONFIRM registry, CCT findings were incremental for the prediction of MACE in asymptomatic DM individuals beyond conventional risk factors and CACS, allowing for improved risk stratification, discrimination, and reclassification.11 DM is a systemic disease and multiple vascular districts may be affected with different distribution and progression rates. In this regard, more comprehensive prospective studies, including other biomarker and imaging modalities (IMT, ABI, CAN), integrated by phenotyping left ventricular structure and function, together with CACS/CCT analysis would be welcome. It seems clear that we need more information concerning the relationship between DM pharmacological control and CAD (presence, extent, severity, progression, etc.) in asymptomatic cohorts. To this end, an intervention study may need to be designed and performed for this purpose.