A model combining rest-only ECG-gated SPECT myocardial perfusion imaging and cardiovascular risk factors can effectively predict obstructive coronary artery disease

Bao Liu; Wenji Yu; Jianfeng Wang; Xiaoliang Shao; Feifei Zhang; Mingge Zhou; Yunmei Shi; Bing Wang; Yiduo Xu; Yuetao Wang

doi:10.1186/s12872-022-02712-8

A model combining rest-only ECG-gated SPECT myocardial perfusion imaging and cardiovascular risk factors can effectively predict obstructive coronary artery disease

BMC Cardiovasc Disord. 2022 Jun 15;22(1):268. doi: 10.1186/s12872-022-02712-8.

Authors

Affiliations

¹ Department of Nuclear Medicine, The Third Affiliated Hospital of Soochow University, No.185, Juqian Street, Changzhou, 213003, Jiangsu Province, China.
² The Nuclear Medicine and Molecular Imaging Clinical Translation Institute of Soochow University, Changzhou, Jiangsu Province, China.
³ Department of Nuclear Medicine, The Third Affiliated Hospital of Soochow University, No.185, Juqian Street, Changzhou, 213003, Jiangsu Province, China. yuetao-w@163.com.

Abstract

Objective: The rest-only single photon emission computerized tomography (SPECT) myocardial perfusion imaging (MPI) had low sensitivity in diagnosing obstructive coronary artery disease (CAD). Improving the efficacy of resting MPI in diagnosing CAD has important clinical significance for patients with contraindications to stress. The purpose of this study was to develop and validate a model predicting obstructive CAD in suspected CAD patients, based on rest-only MPI and cardiovascular risk factors.

Methods: A consecutive retrospective cohort of 260 suspected CAD patients who underwent rest-only gated SPECT MPI and coronary angiography was constructed. All enrolled patients had stress MPI contraindications. Clinical data such as age and gender were collected. Automated quantitative analysis software QPS and QGS were used to evaluate myocardial perfusion and function parameters. The least absolute shrinkage and selection operator (LASSO) and multivariable logistic regression were used to select the variables and build the prediction model.

Results: Among the enrolled 260 patients with suspected CAD, there were 95 (36.5%, 95/260) patients with obstructive CAD. The prediction model was presented in the form of a nomogram and developed based on selected predictors, including age, sex, SRS ≥ 4, SMS ≥ 2, STS ≥ 2, hypertension, diabetes, and hyperlipidemia. The AUC of the prediction model was 0.795 (95% CI: 0.741-0.843), which was better than the traditional models. The AUC calculated by enhanced bootstrapping validation (500 bootstrap resamples) was 0.785. Subsequently, the calibration curve (intercept = - 0.106; slope = 0.843) showed a good calibration of the model. The decision curve analysis (DCA) shows that the constructed clinical prediction model had good clinical applications.

Conclusions: In patients with suspected CAD and contraindications to stress MPI, a prediction model based on rest-only ECG-gated SPECT MPI and cardiovascular risk factors have been developed and validated to predict obstructive CAD effectively.

Keywords: Coronary artery disease; Myocardial perfusion imaging; Prediction model; SPECT.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Cardiovascular Diseases*
Coronary Angiography / methods
Coronary Artery Disease* / diagnostic imaging
Coronary Artery Disease* / therapy
Electrocardiography
Heart Disease Risk Factors
Humans
Models, Statistical
Myocardial Perfusion Imaging* / methods
Prognosis
Retrospective Studies
Risk Factors
Tomography, Emission-Computed, Single-Photon / methods