Warranty Period of a Calcium Score of Zero: Comprehensive Analysis From MESA

Omar Dzaye, Zeina A. Dardari, Miguel Cainzos-Achirica, Ron Blankstein, Arthur S. Agatston, Matthias Duebgen, Joseph Yeboah, Moyses Szklo, Matthew J. Budoff, Joao A.C. Lima, Roger S. Blumenthal, Khurram Nasir, Michael J. Blaha

Research output: Contribution to journalArticlepeer-review

62 Scopus citations


Objectives: This study sought to quantify and model conversion of a normal coronary artery calcium (CAC) scan to an abnormal CAC scan. Background: Although the absence of CAC is associated with excellent prognosis, progression to CAC >0 confers increased risk. The time interval for repeated scanning remains poorly defined. Methods: This study included 3,116 participants from the MESA (Multi-Ethnic Study of Atherosclerosis) with baseline CAC = 0 and follow-up scans over 10 years after baseline. Prevalence of incident CAC, defined by thresholds of CAC >0, CAC >10, or CAC >100, was calculated and time to progression was derived from a Weibull parametric survival model. Warranty periods were modeled as a function of sex, race/ethnicity, cardiovascular risk, and desired yield of repeated CAC testing. Further analysis was performed of the proportion of coronary events occurring in participants with baseline CAC = 0 that preceded and followed repeated CAC testing at different time intervals. Results: Mean participants’ age was 58 ± 9 years, with 63% women, and mean 10-year cardiovascular risk of 14%. Prevalence of CAC >0, CAC >10, and CAC >100 was 53%, 36%, and 8%, respectively, at 10 years. Using a 25% testing yield (number needed to scan [NNS] = 4), the estimated warranty period of CAC >0 varied from 3 to 7 years depending on sex and race/ethnicity. Approximately 15% of participants progressed to CAC >10 in 5 to 8 years, whereas 10-year progression to CAC >100 was rare. Presence of diabetes was associated with significantly shorter warranty period, whereas family history and smoking had small effects. A total of 19% of all 10-year coronary events occurred in CAC = 0 prior to performance of a subsequent scan at 3 to 5 years, whereas detection of new CAC >0 preceded 55% of future events and identified individuals at 3-fold higher risk of coronary events. Conclusions: In a large population of individuals with baseline CAC = 0, study data provide a robust estimation of the CAC = 0 warranty period, considering progression to CAC >0, CAC >10, and CAC >100 and its impact on missed versus detectable 10-year coronary heart disease events. Beyond age, sex, race/ethnicity, diabetes also has a significant impact on the warranty period. The study suggests that evidence-based guidance would be to consider rescanning in 3 to 7 years depending on individual demographics and risk profile.

Original languageEnglish (US)
Pages (from-to)990-1002
Number of pages13
JournalJACC: Cardiovascular Imaging
Issue number5
StatePublished - May 2021


  • coronary artery calcium
  • coronary artery disease
  • risk assessment

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'Warranty Period of a Calcium Score of Zero: Comprehensive Analysis From MESA'. Together they form a unique fingerprint.

Cite this