The Association of Coronary Artery Calcium Score with Heart Failure – a Literature Review

Author	Year	Sample size	Mean age (years)	Study design	Imaging	Statistical analysis	Main findings
Naghavi et al.²⁶	2024	5,830	62 ± 10	Prospective cohort study (MESA)	Non-contrast CAC scans enhanced by Al-enabled automated cardiac chamber volumetry and calcified plaque characterization	Survival analysis using Cox proportional hazards regression Discrimination was assessed using time-dependent ROC area under the curve (AUC) and Uno’s C-statistic AUC difference calculated using inverse probability of censoring weighting (IPCW) estimator	The AI–CAC model significantly improved the prediction of all CVD events (including HF, AF, and stroke) compared to the Agatston score alone.
Sakuragi et al.²⁸	2016	487	69 ± 11	Prospective cohort study	MDCT	One-way AN0VA and the Kruskal–Wallis test (for continuous variables) Chi-squared test (for categorical variables) Kaplan–Meier analysis with the log-rank test Cox proportional hazards regression	Severe CAC is an independent determinant of high N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and a predictor of admission for HF in patients without a history of CAD or HF.
Ota et al.²⁵	2022	550	72.5 ± 13.5	Retrospective cohort study	Non-ECG-gated MDCT Invasive coronary angiography	Mann-Whitney U test (for continuous variables) Chi-squared test (for categorical variables) Kaplan–Meier analysis with the log-rank test Cox proportional hazards regression	Higher CAC score detected via non-contrast chest CT is significantly associated with all-cause mortality among patients with CHF.
Carr et al.²¹	2017	3,043	40.3 ± 3-6	Prospective cohort study (CARDIA)	Non-contrast CT	Cox proportional hazards regression adjusted for covariates Poisson regression Linear regression	In younger adults, a CAC score > 0 was associated with incident all CHD and all-cause death.
Kälsch et al.⁸	2010	4,230	59 ± 8	Prospective cohort study (Heinz Nixdorf Recall Study)	EBCT	Mann-Whitney U test (for continuous variables) Chi-squared test or Fisher’s exact test (for categorical variables) Multivariable analysis using logistic regression	CAC is associated with the presence of CHF. This association is partially diminished after adjusting for traditional risk factors.
Haddad et al.¹⁷	2022	2,082	50.6 ± 17.0	Prospective cohort study	Non-contrast CT	Multidimensional analysis using maximal information coefficient (MIC) to weight edges Multivariable regression analysis Stepwise linear regression	CAC score was independently associated with LV diastolic function.
de la Rosa et al.¹¹	2025	6,667	62 ± 10	Prospective cohort study (MESA)	Non-contrast CT EBCT	Cox proportional hazards models and time-dependent ROC analysis	CAC score was a significant predictor of new-onset HFrEF.
Nitta et al.¹⁸	2019	157	71 ± 8 years for patients with CAC 66 ±13 years for patients without CAC	Retrospective cohort study	Non-contrast CT Gated SPECT for myocardial perfusion TTE	Chi-squared test (for categorical variables) Mann-Whitney U test (for continuous variables) Pearson’s correlation Linear regression analysis (univariate and multivariate)	Log-transformed CAC score was significantly and independently associated with indices of impaired LV diastolic function (PFR and 1/3MFR).
Mhaimeed et al.¹⁰	2025	66,636	54.4 ± 10.6	Retrospective cohort study	Non-contrast CT EBCT	Student’s t test and Wilcoxon signed-rank test Chi-squared test Subdistribution hazard ratios (SHRs)	A higher CAC score was associated with an increasing incidence of HF-related mortality in primary prevention patients.
Kitjanukit et al.²⁹	2024	442	62.5 ± 10.4	Retrospective cohort study	Non-contrast CT	Extended Wilcoxon rank-sum test by Cuzick (for trending variables) Cox regression (to estimate HR) Weighted Cohen’s Kappa (to measure disagreement	Increasing CAC score categories were associated with a significantly increased HR for MACE.
Leening et al.⁵	2012	1,897	69.9 ± 6.5	Prospective populationbased cohort study (The Rotterdam Study)	EBCT	Cox proportional hazards models (implied by use of HR, C-statistic, and reclassification metrics) Measures included c-statistic, integrated discrimination index (IDI), and continuous net reclassification index (NRI).	CAC has a clear and graded association with the risk of developing HF in the elderly population, independent of traditional cardiovascular risk factors and incident overt CHD
Fathala et al.⁶	2019	204	48 ± 13	Retrospective cohort study	Contrast CT ICA	Descriptive statistics Pearson’s Chi-squared test Independent samples t-test or ANOVA Multivariate logistic regression	The mean CAC score was significantly higher in the CAD HF group compared to the non-CAD HF group.
Choi et al.¹⁹	2025	15,193	55.8 ± 8.6	Longitudinal cohort study	Non-contrast CT Echocardiography	Linear regression models Log-binomial regression Linear mixed models	CAC ≥ 100 significantly affects the progression of DD independently of other clinical factors.
Singh et al.³⁰	2024	1,988	DM cohort: 63.4 ± 8.9 Non-DM cohort: 60.3 ± 9.6	Prospective cohort study	Contrast CT	t-tests and p-tests (for categorical variables) Survival analysis metrics including c-index and hazard ratio (HR) Cox proportional hazard models with elastic net regularization 5-f0ld cross-validation with bootstrap iterations Kaplan–Meier curves	Models leveraging radiomic features of calcium-omics and epicardial adipose tissue (fat-omics) extracted via deep learning from CTCS scans showed competitive or superior performance in predicting incident HF compared to traditional models.
Wakaki et al.²²	2025	353	68.6 ± 12.7	Retrospective cohort study	Non-ECG-gated CT ICA	One-way ANOVA Kruskal–Wallis test Chi-squared test Log-rank test Kaplan–Meier method Univariable and multivariable analysis	The CAC score significantly predicted cardiovascular events (including CHF).
Hashimoto et al.²³	2021	108	74 ±13	Retrospective observational study	Non-contrast CT I-BMIPP SPECT	Chi-squared test Kaplan–Meier method Univariable and multivariable analysis	Increased CAC scores were associated with all-cause mortality in patients with CHF.
Tian et al.²⁷	2025	1,310	56.5 ± 11.8	Prospective cohort study (Chronic Renal Insufficiency Cohort)		t test and chi-squared test Multiple imputation Comparison of models using Akaike’s information criterion Schoenfeld residuals to confirm proportional hazard assumptions Survival analysis using methods appropriate for competing risks	Progression of CAC is associated with a higher risk of atherosclerotic CVD and all-cause mortality, but not with CHF.
Ahmad et at.¹⁵	2025	6,592	62 ± 10	Prospective cohort study (MESA)	Non-contrast CT EBCT	Chi-squared test ANOVA Kruskal–Wallis test Kaplan–Meier plot and log-rank test Multivariable Cox proportional hazard models Proportional hazards assumption evaluated using Schoenfeld residuals and formal testing Benjamin-Hochberg (BH) adjusted p value.	The combination of elevated IL-6 and CAC > 0 was associated with a higher risk of HFrEF, showing significant additive and multiplicative interactions. There was an antagonistic interaction for HFpEF.
Wada et al.²⁰	2023	982	64.7 ± 6.6	Prospective cohort study (NADESICO)	Non-contrast CT	Univariable and multivariable analysis Cox proportional hazard models C-statistic analysis	Adding the CAC score to the conventional cardiovascular risk factors (Suita score) significantly improved the predictive ability for future MACE in Japanese patients.
Elnagar et al.³¹	2024	435	48.04 ± 7.19 (Group I: CAC <400)and 49·77 ± 7.15 (Group II: CAC >400).	Retrospective cohort study	Contrast CT	Independent sample t-test Chi-squared test Multivariate logistic regression analysis	CAC scores ≥ 400 predict MACE.
Sharma et al.¹²	2017	5,282	62 ± 10	Prospective cohort study (MESA)	Non-contrast CT EBCT	Cox proportional hazard models Proportional hazard assumption Test for trend	The CAC score was positively associated with incident HFpEF risk in women, but not in men.
Abunassar et al.³²	2011	4,394	No prior HF/LV dysfunction: 51.7 ± 10.3 HF and abnormal EF: 58.8 ± 11.7 High-risk CAD: 62.1 ± 9.7	Retrospective cohort study	Contrast CT	Wilcoxon rank-sum test Fisher’s exact test	CAC = 0 excludes ischemic cardiomyopathy in patients presenting with HF.
Lehmann et al.³³	2018	3,281	58.7 ± 7·5	Prospective cohort study (Heinz Nixdorf Recall study)	EBCT	Chi-squared test Mann-Whitney U test Univariate and multivariable logistic regression Multivariable Cox proportional hazards regression Net reclassification index Integrated discrimination index	The CAC score is associated with CHF in persons without clinically overt CAD, but in longitudinal progression analysis, the predictive value of CAC progression has reduced risk prediction for CVDE when compared to the most recent CAC value.

1 EBCT, electron-beam computed tomography; ICA, invasive coronary angiography; MDCT, multidetector computed tomography; TTE, transthoracic echocardiography

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.2478/jce-2025-0023 | Journal eISSN: 2457-5518 | Journal ISSN: 2457-550X

Journal RSS Feed

Language: English

Page range: 137 - 147

Submitted on: Oct 1, 2025

Accepted on: Dec 13, 2025

Published on: Dec 31, 2025

Published by: Asociatia Transilvana de Terapie Transvasculara si Transplant KARDIOMED

In partnership with: Paradigm Publishing Services

Publication frequency: 4 issues per year

Keywords:

coronary artery calcium score,

Related subjects:

Emergency medicine and intensive-care medicine,

Radiology

© 2025 Daniel Stirbu, Grigore Tinica, Liviu Moraru, Klara Brînzaniuc, Bogdan Suciu, Raluca Ozana Chistol, Raluca Moraru, Cristina Furnica, published by Asociatia Transilvana de Terapie Transvasculara si Transplant KARDIOMED
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Volume 11 (2025): Issue 4 (December 2025)