Does coronary CT angiography improve risk stratification over coronary calcium scoring in symptomatic patients with suspected coronary artery disease? Results from the prospective multicenter international CONFIRM registry

Mouaz H. Al-Mallah, Waqas Qureshi, Fay Y. Lin, Stephan Achenbach, Daniel S. Berman, Matthew J. Budoff, Tracy Q. Callister, Hyuk Jae Chang, Filippo Cademartiri, Kavitha Chinnaiyan, Benjamin J.W. Chow, Victor Y. Cheng, Augustin Delago, Millie Gomez, Martin Hadamitzky, Joerg Hausleiter, Philipp A. Kaufmann, Jonathon Leipsic, Erica Maffei, Gilbert RaffLeslee J. Shaw, Todd C. Villines, Ricardo C. Cury, Gudrun Feuchtner, Fabian Plank, Yong Jin Kim, Allison M. Dunning, James K. Min

Research output: Contribution to journalArticle

71 Scopus citations

Abstract

AimsThe prognostic value of coronary artery calcium (CAC) scoring is well established and has been suggested for use to exclude significant coronary artery disease (CAD) for symptomatic individuals with CAD. Contrast-enhanced coronary computed tomographic angiography (CCTA) is an alternative modality that enables direct visualization of coronary stenosis severity, extent, and distribution. Whether CCTA findings of CAD add an incremental prognostic value over CAC in symptomatic individuals has not been extensively studied.Methods and resultsWe prospectively identified symptomatic patients with suspected but without known CAD who underwent both CAC and CCTA. Symptoms were defined by the presence of chest pain or dyspnoea, and pre-test likelihood of obstructive CAD was assessed by the method of Diamond and Forrester (D-F). CAC was measured by the method of Agatston. CCTAs were graded for obstructive CAD (>70% stenosis); and CAD plaque burden, distribution, and location. Plaque burden was determined by a segment stenosis score (SSS), which reflects the number of coronary segments with plaque, weighted for stenosis severity. Plaque distribution was established by a segment-involvement score (SIS), which reflects the number of segments with plaque irrespective of stenosis severity. Finally, a modified Duke prognostic index - accounting for stenosis severity, plaque distribution, and plaque location - was calculated. Nested Cox proportional hazard models for a composite endpoint of all-cause mortality and non-fatal myocardial infarction (D/MI) were employed to assess the incremental prognostic value of CCTA over CAC. A total of 8627 symptomatic patients (50% men, age 56 ± 12 years) followed for 25 months (interquartile range 17-40 months) comprised the study cohort. By CAC, 4860 (56%) and 713 (8.3%) patients had no evident calcium or a score of >400, respectively. By CCTA, 4294 (49.8%) and 749 (8.7%) had normal coronary arteries or obstructive CAD, respectively. At follow-up, 150 patients experienced D/MI. CAC improved discrimination beyond D-F and clinical variables (area under the receiver-operator characteristic curve 0.781 vs. 0.788, P = 0.004). When added sequentially to D-F, clinical variables, and CAC, all CCTA measures of CAD improved discrimination of patients at risk for D/MI: obstructive CAD (0.82, P < 0.001), SSS (0.81, P < 0.001), SIS (0.81, P = 0.003), and Duke CAD prognostic index (0.82, P < 0.0001).ConclusionIn symptomatic patients with suspected CAD, CCTA adds incremental discriminatory power over CAC for discrimination of individuals at risk of death or MI.

Original languageEnglish (US)
Pages (from-to)267-274
Number of pages8
JournalEuropean Heart Journal Cardiovascular Imaging
Volume15
Issue number3
DOIs
StatePublished - Mar 2014
Externally publishedYes

Keywords

  • coronary CT angiography
  • coronary artery calcium
  • coronary artery disease
  • prognosis
  • symptomatic

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Cardiology and Cardiovascular Medicine

Fingerprint Dive into the research topics of 'Does coronary CT angiography improve risk stratification over coronary calcium scoring in symptomatic patients with suspected coronary artery disease? Results from the prospective multicenter international CONFIRM registry'. Together they form a unique fingerprint.

Cite this