Elsevier

Journal of Nuclear Cardiology

Volume 12, Issue 6, November–December 2005, Pages 676-686
Journal of Nuclear Cardiology

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

https://doi.org/10.1016/j.nuclcard.2005.08.006Get rights and content

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.

Section snippets

Phantoms

An anthropomorphic torso phantom with cardiac, lung, and spine inserts (ECT/TOR/P and ECT/CAR/I; Data Spectrum, Hillsborough, NC) was used to evaluate the effect of AC on the homogeneity of SPECT images of normal phantoms. The left ventricle (LV) was simulated by a cardiac insert consisting of 2 concentric cylinders. The inner cylinder (“cavity” of LV) was filled with water, and the space between the two cylinders (left ventricular “myocardium”) was filled with a uniform technetium-99m solution

Homogeneity of Cardiac Phantom Images With and Without AC

Figure 1 shows the mean percent variability of circumferential count profiles from short-axis phantom images acquired with NC and AC by use of 180° and 360° orbits. With AC, the percent variability of circumferential count profiles of normal phantoms was significantly reduced, indicating greater image homogeneity (11% ± 2% vs 5% ± 1% [P < .001] for 180° orbit and 13% ± 2% vs 6% ± 1% [P < .001] for 360° orbit).

Patient Lower Limits of Normal With and Without AC

Figure 2 shows examples of midventricular lower limit–of–normal curves for men, women,

Discussion

This study shows that CT-based AC improves the homogeneity of SPECT images of normal cardiac phantoms and of normal subjects. Moreover, AC consistently improved the diagnostic yield of SPECT imaging for detecting significant angiographic CAD (≥50% stenosis), as well as the normalcy rate. Our study also indicates that the degree of diagnostic gain by AC may vary among readers. This depends on a reader’s interpretive attitude (ie, reading style) and experience. The blinded readers in our study

Acknowledgment

We gratefully acknowledge the dedication and efforts of Jan Davey, MSN, APRN, in recruiting normal volunteers for this study.

Drs Wackers and Liu receive royalties from sales of WLCQ quantitative software. At the time of this study, Frank Anstett and Aharon Peretz were employed by GE Healthcare. The other authors have indicated they have no financial conflict of interest. This study was an investigator-initiated study. The sponsor had no role in the analysis and interpretation of data or in the

References (40)

  • H. Almquist et al.

    Clinical implication of down-scatter in attenuation-corrected myocardial SPECT

    J Nucl Cardiol

    (1999)
  • P.H. Pretorius et al.

    The influence of attenuation and scatter compensation on the apparent distribution of Tc-99m sestamibi in cardiac slices

    J Nucl Cardiol

    (2001)
  • F. Harel et al.

    Clinical impact of combination of scatter attenuation correction, and depth-dependent resolution recovery for 201-Tl studies

    J Nucl Med

    (2001)
  • Wackers FJTh. Coronary artery disease: exercise stress. Zoghbi GJ, Iskandrian AE. Coronary artery disease:...
  • G. DePuey

    How to detect and avoid myocardial perfusion SPECT artifacts

    J Nucl Med

    (1994)
  • P. Tan et al.

    A scanning line source for simultaneous emission and transmission measurements in SPECT

    J Nucl Med

    (1993)
  • D.L. Bailey et al.

    Improved SPECT using simultaneous emission and transmission tomography

    J Nucl Med

    (1987)
  • E.P. Ficaro et al.

    Simultaneous transmission/emission myocardial perfusion tomographydiagnostic accuracy of attenuation-corrected 99mTc-sestamibi single photon emission computed tomography

    Circulation

    (1996)
  • R. Kluge et al.

    Attenuation correction by simultaneous emission-transmission myocardial single-photon emission tomography using a technetium-99m-labelled radiotracerimpact on diagnostic accuracy

    Eur J Nucl Med

    (1997)
  • R.C. Hendel et al.

    Multicenter clinical trial to evaluate the efficacy of correction for photon attenuation and scatter in SPECT myocardial perfusion imaging

    Circulation

    (1999)
  • Cited by (160)

    • Cardiovascular Imaging in Women

      2024, Seminars in Nuclear Medicine
    • SPECT in Ischemic Heart Diseases

      2022, Nuclear Medicine and Molecular Imaging: Volume 1-4
    View all citing articles on Scopus

    This study was supported in part by an educational grant from GE Healthcare.

    View full text