Phenotypic Clustering of Patients With Newly Diagnosed Coronary Artery Disease Using Cardiovascular Magnetic Resonance and Coronary Computed Tomography Angiography

Clinical phenotypes among patients with normal cardiac perfusion using unsupervised learning: a retrospective observational study

BACKGROUND

Myocardial perfusion imaging (MPI) is one of the most common cardiac scans and is used for diagnosis of coronary artery disease and assessment of cardiovascular risk. However, the large majority of MPI patients have normal results. We evaluated whether unsupervised machine learning could identify unique phenotypes among patients with normal scans and whether those phenotypes were associated with risk of death or myocardial infarction.

METHODS

Patients from a large international multicenter MPI registry (10 sites) with normal perfusion by expert visual interpretation were included in this cohort analysis. The training population included 9849 patients, and external testing population 12,528 patients. Unsupervised cluster analysis was performed, with separate training and external testing cohorts, to identify clusters, with four distinct phenotypes. We evaluated the clinical and imaging features of clusters and their associations with death or myocardial infarction.

FINDINGS

Patients in Clusters 1 and 2 almost exclusively underwent exercise stress, while patients in Clusters 3 and 4 mostly required pharmacologic stress. In external testing, the risk for Cluster 4 patients (20.2% of population, unadjusted hazard ratio [HR] 6.17, 95% confidence interval [CI] 4.64-8.20) was higher than the risk associated with pharmacologic stress (HR 3.03, 95% CI 2.53-3.63), or previous myocardial infarction (HR 1.82, 95% CI 1.40-2.36).

INTERPRETATION

Unsupervised learning identified four distinct phenotypes of patients with normal perfusion scans, with a significant proportion of patients at very high risk of myocardial infarction or death. Our results suggest a potential role for patient phenotyping to improve risk stratification of patients with normal imaging results.

FUNDING

This work was supported by the National Heart, Lung, and Blood Institute at the National Institutes of Health [R35HL161195 to PS]. The REFINE SPECT database was supported by the National Heart, Lung, and Blood Institute at the National Institutes of Health [R01HL089765 to PS]. MCW was supported by the British Heart Foundation [FS/ICRF/20/26002].

The Systemic Inflammation Index Predicts Poor Clinical Prognosis in Patients with Initially Diagnosed Acute Coronary Syndrome Undergoing Primary Coronary Angiography

Background

Systemic inflammation index (SII: neutrophil count * platelet count/lymphocyte count) is a new inflammatory marker that can reflect the degree of systemic inflammatory response after coronary artery disease (CAD). However, the predictive value of the SII for clinical prognosis in patients with initially diagnosed acute coronary syndrome (ACS) has yet to be thoroughly studied.

Patients and Methods

Patients with initially diagnosed ACS who underwent primary coronary angiography in our hospital from January 2019 to April 2021 were included in this study. 757 patients with ACS who underwent primary coronary angiography were enrolled. According to the baseline SII level, the patients were divided into a high SII group and a low SII group. The primary endpoint was major cardiovascular events (MACEs), defined as cardiac death, non-fatal myocardial infarction (MI), and non-fatal stroke.

Results

At a median follow-up of 33.9 months, 140 (18.5%) MACEs were recorded. Receiver operating characteristic (ROC) curve analysis showed that SII's best cut-off value for predicting MACEs was 713.9*109/L. Kaplan-Meier survival curve analysis showed that the survival rate of the low SII group was higher than the high SII group (P<0.001). Compared with the low SII group, the risk of MACEs was significantly increased in the high SII group (89 cases (33.3%) vs.51 patients (10.4%), P<0.001). Univariate and multivariate Cox regression analysis manifested that high SII level was independently associated with the occurrence of MACEs in patients with ACS undergoing primary coronary angiography (adjusted hazard ratio [HR]: 2.915, 95% confidence interval (CI%): 1.830-4.641, P<0.001). Adding SII to the conventional risk factor model improved the predictive value of MACEs.

Conclusion

This study showed that elevated SII was associated with adverse cardiovascular prognosis in patients with ACS undergoing primary coronary angiography, making SII a valuable predictor of poor prognosis in patients with ACS undergoing primary coronary angiography.

Phenotypic Clustering of Beta-Thalassemia Intermedia Patients Using Cardiovascular Magnetic Resonance

We employed an unsupervised clustering method that integrated demographic, clinical, and cardiac magnetic resonance (CMR) data to identify distinct phenogroups (PGs) of patients with beta-thalassemia intermedia (β-TI). We considered 138 β-TI patients consecutively enrolled in the Myocardial Iron Overload in Thalassemia (MIOT) Network who underwent MR for the quantification of hepatic and cardiac iron overload (T2* technique), the assessment of biventricular size and function and atrial dimensions (cine images), and the detection of replacement myocardial fibrosis (late gadolinium enhancement technique). Three mutually exclusive phenogroups were identified based on unsupervised hierarchical clustering of principal components: PG1, women; PG2, patients with replacement myocardial fibrosis, increased biventricular volumes and masses, and lower left ventricular ejection fraction; and PG3, men without replacement myocardial fibrosis, but with increased biventricular volumes and masses and lower left ventricular ejection fraction. The hematochemical parameters and the hepatic and cardiac iron levels did not contribute to the PG definition. PG2 exhibited a significantly higher risk of future cardiovascular events (heart failure, arrhythmias, and pulmonary hypertension) than PG1 (hazard ratio-HR = 10.5; p = 0.027) and PG3 (HR = 9.0; p = 0.038). Clustering emerged as a useful tool for risk stratification in TI, enabling the identification of three phenogroups with distinct clinical and prognostic characteristics.

Unsupervised learning to characterize patients with known coronary artery disease undergoing myocardial perfusion imaging

PURPOSE

Patients with known coronary artery disease (CAD) comprise a heterogenous population with varied clinical and imaging characteristics. Unsupervised machine learning can identify new risk phenotypes in an unbiased fashion. We use cluster analysis to risk-stratify patients with known CAD undergoing single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI).

METHODS

From 37,298 patients in the REFINE SPECT registry, we identified 9221 patients with known coronary artery disease. Unsupervised machine learning was performed using clinical (23), acquisition (17), and image analysis (24) parameters from 4774 patients (internal cohort) and validated with 4447 patients (external cohort). Risk stratification for all-cause mortality was compared to stress total perfusion deficit (< 5%, 5-10%, ≥10%).

RESULTS

Three clusters were identified, with patients in Cluster 3 having a higher body mass index, more diabetes mellitus and hypertension, and less likely to be male, have dyslipidemia, or undergo exercise stress imaging (p < 0.001 for all). In the external cohort, during median follow-up of 2.6 [0.14, 3.3] years, all-cause mortality occurred in 312 patients (7%). Cluster analysis provided better risk stratification for all-cause mortality (Cluster 3: hazard ratio (HR) 5.9, 95% confidence interval (CI) 4.0, 8.6, p < 0.001; Cluster 2: HR 3.3, 95% CI 2.5, 4.5, p < 0.001; Cluster 1, reference) compared to stress total perfusion deficit (≥10%: HR 1.9, 95% CI 1.5, 2.5 p < 0.001; < 5%: reference).

CONCLUSIONS

Our unsupervised cluster analysis in patients with known CAD undergoing SPECT MPI identified three distinct phenotypic clusters and predicted all-cause mortality better than ischemia alone.

The long-term prognostic value provided by Coronary CT angiography

BACKGROUND

Risk-stratification of patients has a major role in the prevention and treatment of cardiovascular disease. The aim was to find the most informative predictors of cardiovascular events in patients undergoing Coronary CT Angiography.

METHODS

We carried out a secondary analysis of a large registry dataset. The included population comprises adults aged 18 or older who underwent Coronary CT Angiography of 64-detector rows or greater. We clustered patients based on their characteristics and compared the risk for poor clinical outcomes between the two clusters.

RESULTS

There were two clusters of patients having different risks for all-cause death, myocardial infarction, and late revascularization [hazard ratios (95%CI) = 2.28 (2.02, 2.57), 1.63 (1.40, 1.89), and 2.46 (2.1, 2.88), all P < 0.001]. The severity of stenosis in the left main coronary artery adjusted for age and sex was the most significant predictor of the high-risk cluster [adjusted odds ratio (95%CI) = 3.35 (2.98, 3.77), P < 0.001]. The severity of stenosis in the first obtuse marginal branch of the left circumflex, distal left circumflex, distal left anterior descending, posterior descending, the first diagonal branch of the left anterior descending, and proximal right coronary artery were important as well (all adjusted odds ratios ≥ 2.52). Cluster profiling showed a higher performance for CT Angiography features (sensitivity = 97.4%, specificity = 85.7%, C-statistic = 98.7%) than calcium, Framingham, and Duke scores in identifying high-risk patients (C-statistic = 82.1, 77.0, and 88.2%, respectively).

CONCLUSION

Coronary CT Angiography can accurately risk-stratify patients concerning poor clinical outcomes.

Clustering of patients with inconclusive non-invasive stress testing referred for vasodilator stress cardiovascular magnetic resonance

BACKGROUND

Inconclusive non-invasive stress testing is associated with impaired outcome. This population is very heterogeneous, and its characteristics are not well depicted by conventional methods.

AIMS

To identify patient subgroups by phenotypic unsupervised clustering, integrating clinical and cardiovascular magnetic resonance data to unveil pathophysiological differences between subgroups of patients with inconclusive stress tests.

METHODS

Between 2008 and 2020, consecutive patients with a first inconclusive non-invasive stress test referred for stress cardiovascular magnetic resonance were followed for the occurrence of major adverse cardiovascular events (defined as cardiovascular death or myocardial infarction). A cluster analysis was performed on clinical and cardiovascular magnetic resonance variables.

RESULTS

Of 1402 patients (67% male; mean age 70±11years) who completed the follow-up (median 6.5years, interquartile range 5.6-7.5years), 197 experienced major adverse cardiovascular events (14.1%). Three distinct phenogroups were identified based upon unsupervised hierarchical clustering of principal components: phenogroup 1=history of percutaneous coronary intervention with viable myocardial infarction and preserved left ventricular ejection fraction; phenogroup 2=atrial fibrillation with preserved left ventricular ejection fraction; and phenogroup 3=coronary artery bypass graft with non-viable myocardial scar and reduced left ventricular ejection fraction. Using survival analysis, the occurrence of major adverse cardiovascular events (P=0.007), cardiovascular mortality (P=0.002) and all-cause mortality (P<0.001) differed among the three phenogroups. Phenogroup 3 presented the worse prognosis. In each phenogroup, ischaemia was associated with major adverse cardiovascular events (phenogroup 1: hazard ratio 2.79, 95% confidence interval 1.61-4.84; phenogroup 2: hazard ratio 2.59, 95% confidence interval 1.69-3.97; phenogroup 3: hazard ratio 3.16, 95% confidence interval 1.82-5.49; all P<0.001).

CONCLUSIONS

Cluster analysis of clinical and cardiovascular magnetic resonance variables identified three phenogroups of patients with inconclusive stress testing, with distinct prognostic profiles.