The Interscan Variation of CT Coronary Artery Calcification Score. Analysis of the Calcium Acetate Renagel Comparison (CARE)-2 Study

Matthew J. Budoff, Paul Kessler, Yan L. Gao, Wajeh Qunibi, Moustafa Moustafa, Song S. Mao

Research output: Contribution to journalArticlepeer-review

25 Scopus citations


Rationale and Objectives: In the Calcium Acetate Renagel Evaluation (CARE)-2 study, the effects of calcium acetate plus atorvastatin (Lipitor) on the progression of coronary artery calcifications (CACs) are evaluated versus those of Renagel, monitored using dual electron beam tomography (EBT) scans (two scans at study initiation and two at follow up). The aim of this study is to estimate the interscan variation for the Agatston score and for the volume score determined in patients with end-stage renal disease (ESRD) in the CARE-2 study. Materials and Methods: CAC score and volume were measured at study initiation in 463 ESRD subjects (mean age: 59.4 ± 12.5 years, 48.3% female). All patients underwent dual scanning using an EBT, as first scan of two needed to measure the progression of CAC when treated with sevelamer (Renagel) compared with calcium acetate with or without atorvastatin. All scans in all participants were completed by using an EBT system (GE Imatron, South San Francisco, CA). Interscan variability was defined by the following formula: abs (scan A - scan B) ÷ (0.5 × scan A + 0.5 × scan B) × 100%, where A and B denote the first and second scan, respectively, of the dual scan procedure performed before treatment. We evaluated the reproducibility of the cutpoints commonly used for calcium scores clinically, namely 1-30, 31-100, 101-400, and >400. Results: The CAC interscan variability was 11.8% using the Agatston score and 10.3% using the volume score. The reproducibility was then assessed using cutpoints 1-30, 31-100, 101-400, and >400. Agatston score variability for the four subgroups was 61.3%, 23%, 16.1%, and 8.2%, respectively (mean variability, 11.8%). Volume score variability was 60.0%, 14.4%, 14.6%, and 7.7%, respectively (mean variability, 10.3%). The correlation coefficient for scan A to scan B goes up significantly with increasing calcium scores and reaches 0.99 for scores greater than 400 (P < .0001). Conclusion: Interscan variability was sufficiently small for patients with calcium scores greater than 30. Our study thus demonstrates a sufficient reproducibility of the calcium score using EBT. This score allows for accurate serial assessment of these patients and for comparing different therapies.

Original languageEnglish (US)
Pages (from-to)58-61
Number of pages4
JournalAcademic Radiology
Issue number1
StatePublished - Jan 2008


  • Atherosclerosis
  • cardiac CT
  • coronary calcium
  • electron beam tomography
  • reproducibility

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'The Interscan Variation of CT Coronary Artery Calcification Score. Analysis of the Calcium Acetate Renagel Comparison (CARE)-2 Study'. Together they form a unique fingerprint.

Cite this