Article

Journal of Nuclear Cardiology

, Volume 12, Issue 6, pp 676-686

Clinical validation of SPECT attenuation correction using x-ray computed tomography—derived attenuation maps: Multicenter clinical trial with angiographic correlation

  • Yasmin MasoodAffiliated withCardiovascular Nuclear Imaging Laboratory, Yale University School of Medicine
  • , Yi -Hwa LiuAffiliated withCardiovascular Nuclear Imaging Laboratory, Yale University School of Medicine
  • , Gordon DePueyAffiliated withDepartment of Nuclear Medicine, St. Lukes-Roosevelt Hospital
  • , Raymond TailleferAffiliated withDepartment of Nuclear Medicine, Hôtel Dieu de Montréal, Montréal
  • , Luis I. AraujoAffiliated withDepartment of Nuclear Medicine, University of Pennsylvania
  • , Steven AllenAffiliated withDepartment of Nuclear Medicine, Mercy Medical Center
  • , Dominique DelbekeAffiliated withDepartment of Nuclear Medicine, Vanderbilt University
  • , Frank AnstettAffiliated withGE Healthcare
  • , Aharon PeretzAffiliated withGE Healthcare
    • , Mary -Jo ZitoAffiliated withCardiovascular Nuclear Imaging Laboratory, Yale University School of Medicine
    • , Vera TsatkinAffiliated withCardiovascular Nuclear Imaging Laboratory, Yale University School of Medicine
    • , Frans J. Th. WackersAffiliated withCardiovascular Nuclear Imaging Laboratory, Yale University School of Medicine Email author 

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Background

Nonuniform attenuation artifacts cause suboptimal specificity of stress single photon emission computed tomography (SPECT) myocardial perfusion images. In phantoms, normal subjects, and patients suspected of having coronary artery disease (CAD), we evaluated a new hybrid attenuation correction (AC) system that combines x-ray computed tomography (CT) with conventional stress SPECT imaging.

Methods and Results

The effect of CT-based AC was evaluated in phantoms by assessing homogeneity of normal cardiac inserts. AC improved homogeneity of normal cardiac phantoms from 11% ± 2% to 5% ± 1% (P < .001). Attenuation-corrected normal patient files were created from 37 normal subjects with a low likelihood (<3%) of CAD. The diagnostic performance of AC for detection of CAD was evaluated in 118 patients who had stress technetium 99m sestamibi or tetrofosmin stress SPECT imaging and coronary angiography. SPECT images with and without AC were interpreted by 4 blinded readers with different interpretative attitudes. Overall, AC improved the diagnostic performance of all readers, particularly the normalcy rate. The degree of improvement depended on interpretative attitude. Readers prone to high sensitivity or with less experience had the greatest gain in the normalcy rate, whereas a reader prone to higher specificity had improvements in sensitivity and specificity but not the normalcy rate. Importantly, improvement of one diagnostic variable was not associated with worsening of other variables.

Conclusion

CT-based AC of SPECT images consistently improved overall diagnostic performance of readers with different interpretive attitudes and experience. CT-based AC is well suited for routine use in clinical practice. (J Nucl Cardiol 2005;12:676-86.)

Key Words

Single photon emission computed tomography computed x-ray tomography attenuation correction coronary artery disease