Coronary Computed Tomography Angiography for Risk Assessment in Asymptomatic Type 2 Diabetics

Original article 


David A. Halon MB ChB, FACC, FESC

Address for correspondence

Department of Cardiovascular Medicine,

Lady Davis Carmel Medical Center

7, Michal Street, Haifa

Israel 3436212

E- mail:





Type 2 diabetes mellitus (DM) is a metabolic condition initiated by resistance to the action of insulin in promoting absorption of blood glucose by body cells and is characterized by an elevated blood glucose level. Widespread large vessel and microvascular complications of this condition have been extensively described. Coronary artery disease is prominent in DM and CAD outcomes have often been reported to be twice as common as in non-DM individuals. Computed tomography  (CT) angiography is a non-invasive, radiographic method of obtaining detailed assessment of the site, extent and character of atherosclerotic plaques in the coronary tree (Figure 1) potentially leading to vessel obstruction and myocardial infarction. We prospectively initiated a CT angiographic based survey of DM in subjects with no clinical evidence of coronary artery disease who have now been followed clinically for 8-10 years to assess the value of CT findings, in addition to clinical data, in identification of a cohort at high risk for coronary artery disease related events who may benefit from intensified preventive or interventional therapy and serve as a high risk cohort for focused study of new preventive or therapeutic modalities. In addition we examined the relation of baseline clinical and CT angiographic findings to late non-coronary vascular events.



Figure 1. CT angiogram of the proximal portion of a coronary artery with a partially calcified atherosclerotic plaque causing mild luminal stenosis. Insert: A cross-section through the calcified portion of the plaque.


Subjects and CTA findings at study entry

Subjects included in the study were aged 55-74 years and about half were women. They had been diagnosed with diabetes for a mean of 10 years and two thirds had hypertension. Only 14% were current smokers but a further 30% had smoked in the past and 22% were receiving insulin as part of their treatment for DM. The baseline CTA findings demonstrated a wide diversity of prevalence and extent of CAD (Figure 2) demonstrating that despite having DM 20% of subjects had no coronary artery plaque whatsoever whereas 39% had involvement of all 3 coronary arteries. Obstructive disease (luminal stenosis ≥50% by visual assessment) was however far less prevalent (Figure 2). A clinical risk score for coronary heart disease events in DM, the United Kingdom Prospective Diabetes Study risk score, based on standard risk factors, glycated hemoglobin levels and duration of DM, correlated with the extent of coronary artery disease on CTA.


Outcome prediction

A combined coronary heart disease outcome including cardiovascular death, myocardial infarction, unstable angina or new onset angina requiring revascularization was assessed over a follow-up period of 6.6 years. Both clinical and angiographic factors were predictors of events. Clinical outcome predictors included duration of DM, glycated hemoglobin level and blood lipids specifically the ratio of total cholesterol to high density lipoprotein cholesterol. The coronary artery calcium score, derived from a low radiation cardiac CT scan without contrast injection, had additional predictive value independently from the clinically derived risk score. Information obtained from the full CT angiogram, following intravenous contrast injection and reconstruction of the coronary arteries, showed that measures of total coronary arterial atherosclerotic plaque burden (Figure 3) and an overall coronary angiographic assessment based on the degree and functional significance of luminal stenoses, added significant further information assisting in prediction of outcomes. The final predictive model from the clinical and CT angiographic data identified a cohort of patients in the upper quintile of risk with a 22% risk for an outcome event vs a 6.5% risk in the overall study cohort. This cohort included 68% of all patients with a coronary outcome event.


Microvascular events (renal failure and intervention for retinovascular disease) were predicted by baseline clinical variables related to the duration and treatment status of diabetes as well as laboratory evidence of microvascular disease (albuminuria) at study entry and had a univariate association with coronary arterial stenosis on the CT angiogram but the coronary CT angiogram did not add significant independent information for prediction of microvascular related events.


Although CT angiography is not recommended as a routine test in all asymptomatic diabetics many may present with atypical complaints often suggestive of coronary artery disease and in these patients the CT angiogram may be a useful test to assess risk of future coronary events. In addition it might be used to assist in defining a high risk patient cohort for the study of new preventive agents in the setting of a clinical trial.



Figure 2.  Left: Prevalence of 0-3 vessel (V) coronary artery atheroma at time of baseline CT scan at study entry. Right: Prevalence of 0-3 vessel coronary artery luminal stenosis (≥50% by visual assessment). There is a wide diversity of plaque findings but multivessel stenosis is uncommon.


Figure 3. Kaplan-Meier event-free survival curves for study subjects in relation to quartiles of total coronary arterial plaque volume. Event-free survival is significantly worse in subjects in quartiles 3 and 4.