Clinical value of attenuation correction in stress-only Tc-99m sestamibi SPECT imaging. Academic Article uri icon

Overview

abstract

  • BACKGROUND: Attenuation artifact remains a substantial limitation to confident interpretation of images and reduces laboratory efficiency by requiring comparison of stress and rest image sets. Attenuation-corrected stress-only imaging has the potential to ameliorate these limitations. METHODS AND RESULTS: Ten experienced nuclear cardiologists independently interpreted 90 stress-only electrocardiography (ECG)-gated technetium 99m sestamibi images in a sequential fashion: myocardial perfusion imaging (MPI) alone, MPI plus ECG-gated data, and attenuation-corrected MPI with ECG-gated data. Images were interpreted for diagnostic certainty (normal, probably normal, equivocal, probably abnormal, abnormal, and perceived need for rest imaging). With stress MPI data alone, only 37% of studies were interpreted as definitely normal or abnormal, with a very high perceived need for rest imaging (77%). The addition of gated data did not alter the interpretations. However, attenuation-corrected data significantly increased the number of studies characterized as definitely normal or abnormal (84%, P <.005) and significantly reduced the perceived need for rest imaging (43%, P <.005). These results were confirmed by use of a nonsequential consensus interpretation of three readers. CONCLUSION: Attenuation correction applied to studies with stress-only Tc-99m ECG-gated single photon emission computed tomography images significantly increases the ability to interpret studies as definitely normal or abnormal and reduces the need for rest imaging. These findings may improve laboratory efficiency and diagnostic accuracy.

publication date

  • January 1, 2004

Research

keywords

  • Coronary Artery Disease
  • Exercise Test
  • Gated Blood-Pool Imaging
  • Image Interpretation, Computer-Assisted
  • Obesity
  • Technetium Tc 99m Sestamibi
  • Tomography, Emission-Computed, Single-Photon

Identity

Scopus Document Identifier

  • 3242673497

Digital Object Identifier (DOI)

  • 10.1016/j.nuclcard.2004.03.005

PubMed ID

  • 15173774

Additional Document Info

volume

  • 11

issue

  • 3