Prevalence of Aortic Valve Calcium and the Long-Term Risk of Incident Severe Aortic Stenosis.

BACKGROUND: Aortic valve calcification (AVC) is a principal mechanism underlying aortic stenosis (AS). OBJECTIVES: This study sought to determine the prevalence of AVC and its association with the long-term risk for severe AS. METHODS: Noncontrast cardiac computed tomography was performed among 6,814 participants free of known cardiovascular disease at MESA (Multi-Ethnic Study of Atherosclerosis) visit 1. AVC was quantified using the Agatston method, and normative age-, sex-, and race/ethnicity-specific AVC percentiles were derived. The adjudication of severe AS was performed via chart review of all hospital visits and supplemented with visit 6 echocardiographic data. The association between AVC and long-term incident severe AS was evaluated using multivariable Cox HRs. RESULTS: AVC was present in 913 participants (13.4%). The probability of AVC >0 and AVC scores increased with age and were generally highest among men and White participants. In general, the probability of AVC >0 among women was equivalent to men of the same race/ethnicity who were approximately 10 years younger. Incident adjudicated severe AS occurred in 84 participants over a median follow-up of 16.7 years. Higher AVC scores were exponentially associated with the absolute risk and relative risk of severe AS with adjusted HRs of 12.9 (95% CI: 5.6-29.7), 76.4 (95% CI: 34.3-170.2), and 380.9 (95% CI: 169.7-855.0) for AVC groups 1 to 99, 100 to 299, and ≥300 compared with AVC = 0. CONCLUSIONS: The probability of AVC >0 varied significantly by age, sex, and race/ethnicity. The risk of severe AS was exponentially higher with higher AVC scores, whereas AVC = 0 was associated with an extremely low long-term risk of severe AS. The measurement of AVC provides clinically relevant information to assess an individual's long-term risk for severe AS.

Coronary Calcium Detection using 3D Attention Identical Dual Deep Network Based on Weakly Supervised Learning.

Coronary artery calcium (CAC) is biomarker of advanced subclinical coronary artery disease and predicts myocardial infarction and death prior to age 60 years. The slice-wise manual delineation has been regarded as the gold standard of coronary calcium detection. However, manual efforts are time and resource consuming and even impracticable to be applied on large-scale cohorts. In this paper, we propose the attention identical dual network (AID-Net) to perform CAC detection using scan-rescan longitudinal non-contrast CT scans with weakly supervised attention by only using per scan level labels. To leverage the performance, 3D attention mechanisms were integrated into the AID-Net to provide complementary information for classification tasks. Moreover, the 3D Gradient-weighted Class Activation Mapping (Grad-CAM) was also proposed at the testing stage to interpret the behaviors of the deep neural network. 5075 non-contrast chest CT scans were used as training, validation and testing datasets. Baseline performance was assessed on the same cohort. From the results, the proposed AID-Net achieved the superior performance on classification accuracy (0.9272) and AUC (0.9627).

Thoracic aortic calcification and coronary heart disease events: the multi-ethnic study of atherosclerosis (MESA).

BACKGROUND: The presence and extent of coronary artery calcium (CAC) is an independent predictor of coronary heart disease (CHD) morbidity and mortality. Few studies have evaluated interactions or independent incremental risk for coronary and thoracic aortic calcification (TAC). The independent predictive value of TAC for CHD events is not well-established. METHODS: This study used risk factor and computed tomography scan data from 6807 participants in the multi-ethnic study of atherosclerosis (MESA). Using the same images for each participant, TAC and CAC were each computed using the Agatston method. The study subjects were free of incident CHD at entry into the study. RESULTS: The mean age of the study population (n=6807) was 62±10 years (47% males). At baseline, the prevalence of TAC and CAC was 28% (1904/6809) and 50% (3393/6809), respectively. Over 4.5±0.9 years, a total of 232 participants (3.41%) had CHD events, of which 132 (1.94%) had a hard event (myocardial infarction, resuscitated cardiac arrest, or CHD death). There was a significant interaction between gender and TAC for CHD events (p<0.05). Specifically, in women, the risk of all CHD event was nearly 3-fold greater among those with any TAC (hazard ratio: 3.04, 95% CI: 1.60-5.76). After further adjustment for increasing CAC score, this risk was attenuated but remained robust (HR: 2.15, 95% CI: 1.10-4.17). Conversely, there was no significant association between TAC and incident CHD in men. In women, the likelihood ratio chi square statistics indicate that the addition of TAC contributed significantly to predicting incident CHD event above that provided by traditional risk factors alone (chi square=12.44, p=0.0004) as well as risk factors+CAC scores (chi square=5.33, p=0.02). On the other hand, addition of TAC only contributed in the prediction of hard CHD events to traditional risk factors (chi-square=4.33, p=0.04) in women, without contributing to the model containing both risk factors and CAC scores (chi square=1.55, p=0.21). CONCLUSION: Our study indicates that TAC is a significant predictor of future coronary events only in women, independent of CAC. On studies obtained for either cardiac or lung applications, determination of TAC may provide modest supplementary prognostic information in women with no extra cost or radiation.