Influence of coronary artery disease prevalence on predictive values of coronary CT angiography: a meta-regression analysis
- First Online:
- Cite this article as:
- Schlattmann, P., Schuetz, G.M. & Dewey, M. Eur Radiol (2011) 21: 1904. doi:10.1007/s00330-011-2142-2
- 226 Downloads
To evaluate the impact of coronary artery disease (CAD) prevalence on the predictive values of coronary CT angiography.
We performed a meta-regression based on a generalised linear mixed model using the binomial distribution and a logit link to analyse the influence of the prevalence of CAD in published studies on the per-patient negative and positive predictive values of CT in comparison to conventional coronary angiography as the reference standard. A prevalence range in which the negative predictive value was higher than 90%, while at the same time the positive predictive value was higher than 70% was considered appropriate.
The summary negative and positive predictive values of coronary CT angiography were 93.7% (95% confidence interval [CI] 92.8–94.5%) and 87.5% (95% CI, 86.5–88.5%), respectively. With 95% confidence, negative and positive predictive values higher than 90% and 70% were available with CT for a CAD prevalence of 18–63%. CT systems with >16 detector rows met these requirements for the positive (P < 0.01) and negative (P < 0.05) predictive values in a significantly broader range than systems with ≤16 detector rows.
It is reasonable to perform coronary CT angiography as a rule-out test in patients with a low-to-intermediate likelihood of disease.