Skip to main content

30-06-2011 | Cardiology | Article

CTCA accuracy in clinical care varies across centers


Free abstract

MedWire News: The diagnostic accuracy of computed tomographic coronary angiography (CTCA) varies with the clinical care setting, researchers report.

Although CTCA has been clinically accepted for the detection of obstructive coronary artery disease (CAD) and has demonstrated "excellent accuracy" in single-center studies, multicenter studies have yielded variable results, explain Ron Goeree (McMaster University, Hamilton, Ontario, Canada) and colleagues.

"Factors affecting institutional variability need to be better understood before CTCA is universally adopted," they write in the Archives of Internal Medicine.

The researchers conducted a multicenter cohort study between September 2006 and June 2009, in which they compared multidetector CTCA with invasive coronary angiography (ICA) to probe CTCA's diagnostic accuracy in a real-world setting. The study comprised 169 patients with a mean age of 61 years and a mean pretest likelihood for obstructive CAD of 46.8%. All patients underwent CTCA and ICA within 10 days of inclusion.

The findings showed that the overall prevalence of obstructive CAD (defined as ≥50% stenosis by ICA) was 53%, with a prevalence of 21% in a group of 52 patients primarily referred for valvular heart disease, cardiomyopathy, congenital heart disease, or aortic disease (p<0.001). Among the 117 asymptomatic patients referred to ICA for intermediate pretest probability for CAD , the prevalence of angiographic disease was 61% (p<0.001).

The overall patient-based sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of CTCA for detecting obstructive CAD were 81.3%, 93.3%, 91.6%, and 84.7%, respectively. The area under the receiver-operating characteristic curve was 0.873.

CTCA's diagnostic accuracy varied across centers (p<0.001), with a sensitivity, specificity, PPV, and NPV ranging from 50.0% to 93.2%, 92.0% to 100%, 84.6% to 100%, and 42.9% to 94.7%, respectively.

"Our real-world field evaluation of the diagnostic accuracy of CTCA suggests that the operating characteristics of CTCA are good, but implementation into clinical practice may result in a decline in sensitivity and NPV," comment Goeree et al.

In a related commentary, George Diamond (Cedars-Sinai Medical Center, Los Angeles, California, USA) and Sanjay Kaul (University of California, Los Angeles, USA) point out that practicing clinicians need to know if and when to use CTCA as an alternative to initial medical management or other screening methods.

"Logistical availability, local expertise, and financial self-interest will likely play a greater role than comparative effectiveness research in making this choice," they state.

Diamond and Kaul conclude: "In the meantime, we would be well advised to more openly acknowledge the 'real-world' limitations of every diagnostic test."

MedWire ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2011

By Piriya Mahendra

Related topics