Background and aim: Coronary artery calcium score (CACs) is an established quantitative tool for assessing subclinical atherosclerosis. The aim of this study was to assess in a meta-analysis model the natural history and reproducibility of CACs measurements obtained from St Francis and EBEAT trials.

Methods: We analysed data from: 443 on placebo with 2 year follow up from St Francis trial (Study A) and 209 on 10 mg atorvastatin with 1 year follow up of EBEAT trial (Study B). Total CACs and that in the left coronary artery (LCA) branches, main stem (LMS), anterior descending (LAD), circumflex (Cx) and right coronary artery (RCA) were analysed.

Results: The overall agreement between the two measurements was fairly good, showing a small but significant increase in CAC: 68% of the group as a whole presented an increase in CACs, 23 % of the cohort had negligible change in CACs of <10% irrespective of the baseline CACs; and the remaining 10% showed a fall in CACs. Both studies showed similar patterns. The analysis of individual coronary arteries showed significantly higher variability of measurements in the RCA than that of the LCA. Males had higher baseline CAC, irrespective of age, than females but the rate of progression was not different between genders.

Conclusion: The higher variability in RCA measurements could be related to the low baseline CACs or exaggerated movement of the right side atrioventricular ring, whereas those for LCA branches are influenced by the branch allocation of the CACs.