Computerized left ventricular regional ejection fraction analysis for detection of ischemic coronary artery disease with multidetector CT angiography

Irfan Zeb, Dong Li, Khurram Nasir, Mohit Gupta, Jigar Kadakia, Yanlin Gao, Eva Ma, Song Shou Mao, Matthew Budoff

Research output: Contribution to journalArticle

7 Scopus citations

Abstract

Regional ejection fraction (REF) provides important functional information of the left ventricular regional myocardium. We aimed to test the diagnostic accuracy of computerized REF analysis for detecting the ischemia and significant stenosis with multidetector CT angiography (MDCT). This is a retrospective study including 155 patients who underwent MDCT scans for evaluation of coronary artery disease. Among them, 83 patients also underwent SPECT imaging and invasive coronary angiography (ICA). Two groups of patients were defined: Control group with 0 coronary artery calcium and normal global and regional ventricular function, and comparison group. REF measurement was performed on all patients using computerized software. Control group REF measurements will be used as reference standard (mean-2SD REF/mean global ejection fraction) to define abnormal REF. The sensitivity, specificity, positive and negative predictive value of REF in detecting perfusion defects (fixed and reversible) was 73, 80, 75 and 79 % respectively, in a patient based analysis of comparison group. The diagnostic accuracy of REF in predicting significant stenosis (>50 %) on ICA compared with SPECT was 72 versus 61 % and 85 versus 79 % in patient and vessel based analysis of comparison group, respectively. ROC curve analysis showed REF to be a better predictor of perfusion defects on SPECT compared with significant stenosis (>50 %) alone or stenosis combined with REF (P < 0.05). The computerized assessment of REF analysis is comparable to SPECT in predicting ischemia and a better predictor of significant stenosis than SPECT. This study also provides reference standard to define abnormal values.

Original languageEnglish (US)
Pages (from-to)685-692
Number of pages8
JournalInternational Journal of Cardiovascular Imaging
Volume29
Issue number3
DOIs
StatePublished - Mar 2013

Keywords

  • Ischemia
  • Left ventricle
  • MDCT
  • Regional ejection fraction

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Cardiology and Cardiovascular Medicine

Fingerprint Dive into the research topics of 'Computerized left ventricular regional ejection fraction analysis for detection of ischemic coronary artery disease with multidetector CT angiography'. Together they form a unique fingerprint.

Cite this