Prognostic relevance of comprehensive non-invasive imaging approach in a diabetic and non-diabetic asymptomatic population

Backgroud: The aim of the study was to evaluate the incremental prognostic benefit of carotid artery disease and subclinical coronary artery disease (CAD) features in addition to clinical evaluation in a diabetic and non-diabetic asymptomatic population. Methods: Over a six-year period, 10-year-FRS together with carotid ultrasound (CUS) and coronary computed tomography angiography (CCTA) were evaluated for the prediction of major adverse cardiac events (MACE). Results: Five-hundred-seventeen consecutive patients were enrolled in the study, including 328 (63%) male with a mean age of 64±10 (SD). No diabetic patients were 426 (82.4%) and diabetic patients were 91 (17.6). Mean CACS was 16 [SD 0-88] in non-diabetic patients and 64 [SD 0-166] in diabetic patients (p<0.001). The patients with presence of CAD≥50% were 143 (27.7%), whom 105 (24.7%) non-diabetic and 38 (41.8%) diabetic (p=0.001). Over a median follow-up of 4.4 [3.4-5.1] years there were a total of 53 CHD events (10%) including 6 cardiac deaths (1.2%), 13 non-fatal myocardial infarction (2.5%), and 34 non ST elevation myocardial infarction (6.5%). Total events were 37 (7.1%) in non-diabetic population and 16 (17.6%) in diabetic population. The mean radiation dose during CCTA was 4.3±1.0 mSv with no difference between the two groups. The univariable analysis (Table 2) showed that hyperlipidemia, aspirin, carotid plaque, carotid disease, CAD ≥70%, % of segments with non calcific plaque, % of segments with mixed plaque, % of segments with remodeled plaque, and CACS was significant predictors of CHD events in non-diabetic population. Differently, in diabetic population only statins and % of segments with remodeled plaque were significant predictors of CHD events, while % of segments with mixed plaque did not reach statistical significance. On multivariable analysis in non-diabetic group, carotid disease was a significant independent predictor of CHD events when added to FRS (C-statistic, 95% CI: 0.62, 0.55-0.68; p=0.037). CACS were independent predictor when added to both FRS and carotid disease (C-statistic, 95% CI: 0.66, 0.60-0.73; p=0.016). CUS and CACS data were no more significant when CCTA parameters were included in the model, with the latter being the only significant independent predictors. In particular, % of segment with remodeled plaque was incremental independent predictor even when added to a model including the presence of CAD ≥70% (C-statistic, 95% CI: 0.84, 0.80-0.88; p<0.001). In diabetic group % of segments with remodeled plaque represented the only independent predictor of CHD events (C- statistic, 95% CI: 0.83, 0.77-0.87; p<0.001). Conclusions: In an asymptomatic at-risk population carotid disease assessment is able to predict MACE occurrence more accurately than traditional clinical scores and comparable to coronary calcium score. Coronary artery stenosis and plaque positive remodeling represent the most powerful tools of risk reclassification of this wide subset of patients. In the subgroup of diabetic subjects, the percentage of segments with remodeled plaque is the only.