Head-to-Head Comparison of CZT-SPECT and SPECT/CT Myocardial Perfusion Imaging: Interobserver and Intraobserver Agreement and Diagnostic Performance.
Life (Basel)
; 13(9)2023 Sep 07.
Article
em En
| MEDLINE
| ID: mdl-37763283
ABSTRACT
BACKGROUND:
Myocardial perfusion imaging (MPI) plays a crucial role in diagnosing coronary artery disease (CAD), with single-photon emission computed tomography (SPECT) being a widely accepted method. The accuracy of MPI relies on image quality and the expertise of physicians. While CZT-SPECT cameras offer advantages, they can be susceptible to attenuation artifacts. Therefore, our objective was to evaluate the diagnostic accuracy of CZT-SPECT and SPECT/CT in a clinical setting.METHOD:
We conducted a prospective single-center study involving patients with known or suspected stable ischemic heart disease who underwent SPECT-MPI using CZT-SPECT and SPECT/CT scanners, and the latter was equipped with cardiofocal collimation. Experienced physicians performed analysis and reporting based on automated quantification and visual image interpretation.RESULTS:
A total of 77 patients (32 women (41.6%) and 45 men (58.4%) with an average age of 71.9 ± 8.9 years) were included. The agreement between readers regarding the final conclusion based on imaging reporting using both devices was very high (Kappa 0.87-0.93). Per-vessel analysis revealed a trend suggesting that CZT-SPECT was superior to conventional SPECT/CT in terms of sensitivity, positive predictive value (PPV), negative predictive value (NPV), and accuracy, although the difference did not reach statistical significance.CONCLUSION:
Our study demonstrated that CZT-SPECT imaging offers comparable diagnostic accuracy, improved patient comfort, and eliminates CT-induced radiation compared to SPECT/CT. These findings suggest that cardiac CZT-SPECT imaging has the potential to become a valuable imaging modality in clinical practice.
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Tipo de estudo:
Diagnostic_studies
/
Prognostic_studies
Idioma:
En
Ano de publicação:
2023
Tipo de documento:
Article