Related Experiment Videos
Evaluating changes in coronary artery calcium: an analytic method that accounts for interscan variability
John E Hokanson1, Todd MacKenzie, Gregory Kinney
1Department of Preventive Medicine and Biometrics, University of Colorado Health Sciences Center, 4200 E Ninth Ave, B 119, Denver, CO 80262, USA. john.hokanson@uchsc.edu
AJR. American Journal of Roentgenology
|April 22, 2004
Summary
To accurately track changes in coronary calcium, a square root transformation stabilizes variability, allowing for unbiased assessment of disease progression. This method improves the evaluation of therapeutic impacts on subclinical coronary artery disease.
Area of Science:
- Cardiovascular imaging
- Radiology
- Biostatistics
Background:
- Coronary calcium scoring via electron beam CT predicts cardiovascular events.
- Accurate assessment of coronary calcium changes requires addressing interscan variability.
- Variability in coronary calcium scores can obscure true progression or regression over time.
Purpose of the Study:
- To evaluate interscan variability in coronary calcium volume scores.
- To identify a transformation method that stabilizes variability for accurate change detection.
- To apply this method to assess coronary calcium progression in diabetic patients.
Main Methods:
- Electron beam CT performed on 1,074 participants with repeated scans 5 minutes apart.
- Bland-Altman plots and homogeneity tests used to examine interscan variability.
- Box-Cox transformations, including square root, applied to stabilize calcium volume scores.
Main Results:
- Interscan variability in coronary calcium scores increased with higher calcium levels.
- Square root transformation of coronary calcium volume scores effectively removed heterogeneity.
- A threshold of >= 2.5 mm³ difference in square root-transformed scores identified significant progression (10% of subjects).
Conclusions:
- Square root transformation of coronary calcium volume scores stabilizes interscan variability.
- This transformation provides an unbiased estimate of coronary calcium change, independent of baseline scores.
- The technique enhances the study of coronary calcium changes in relation to clinical outcomes and therapeutic interventions.