Implementation of a Myocardial Perfusion Imaging Risk Algorithm to Inform Appropriate Downstream Invasive Testing and Treatment

Incremental prognostic value of downstream positron emission tomography perfusion imaging after coronary computed tomography angiography: a study using machine learning

AIMS

To evaluate the incremental value of positron emission tomography (PET) myocardial perfusion imaging (MPI) over coronary computed tomography angiography (CCTA) in predicting short- and long-term outcome using machine learning (ML) approaches.

METHODS AND RESULTS

A total of 2411 patients with clinically suspected coronary artery disease (CAD) underwent CCTA, out of whom 891 patients were admitted to downstream PET MPI for haemodynamic evaluation of obstructive coronary stenosis. Two sets of Extreme Gradient Boosting (XGBoost) ML models were trained, one with all the clinical and imaging variables (including PET) and the other with only clinical and CCTA-based variables. Difference in the performance of the two sets was analysed by means of area under the receiver operating characteristic curve (AUC). After the removal of incomplete data entries, 2284 patients remained for further analysis. During the 8-year follow-up, 210 adverse events occurred including 59 myocardial infarctions, 35 unstable angina pectoris, and 116 deaths. The PET MPI data improved the outcome prediction over CCTA during the first 4 years of the observation time and the highest AUC was at the observation time of Year 1 (0.82, 95% confidence interval 0.804-0.827). After that, there was no significant incremental prognostic value by PET MPI.

CONCLUSION

PET MPI variables improve the prediction of adverse events beyond CCTA imaging alone for the first 4 years of follow-up. This illustrates the complementary nature of anatomic and functional information in predicting the outcome of patients with suspected CAD.

Impact of appropriateness in clinical practice: data from a single-centre nuclear cardiology laboratory

Aims

To verify the level of appropriateness of referral to our nuclear cardiology laboratory for stress myocardial perfusion imaging (MPI) and explore the correlation between test appropriateness patterns and ischaemia.

Methods and results

In 1870 consecutive patients (mean age 73 ± 12 years; 33% female) undergoing MPI, the level of imaging test appropriateness was evaluated according to the 2023 Appropriate Use Criteria (AUC) and the current European Society of Cardiology (ESC) guidelines for the management of chronic coronary syndromes. The evidence of moderate-to-severe ischaemia (i.e. summed difference score >7) was recorded. According to the AUC criteria, the MPI of 1638 (88%), 130 (7%), and 102 (5%) patients could be classified as 'appropriate', 'inappropriate', and 'uncertain', respectively. Similarly, in 1685 (90%) patients, the referral to MPI was adherent to ESC guidelines, while in 185 (10%), it was not. The majority of appropriate MPI tests showed the presence of moderate-to-severe ischaemia (55%), while only a limited number (10%; P < 0.05) of MPI tests with uncertain clinical appropriateness or clearly inappropriate indications did not. In patients managed adherently to ESC guidelines, invasive coronary angiography more frequently showed obstructive coronary artery disease (CAD) (93 vs. 47%, P < 0.001) and led to coronary revascularization (65 vs. 23%, P < 0.001) compared with patients managed non-adherently.

Conclusion

In a single-centre, single-national, single-modality population, the current rate of appropriate MPI tests is high. Appropriate referrals are associated with a higher probability of moderate-to-severe ischaemia and better downstream resource utilization than inappropriate ones.

Prognostic impact of patients' management based on anatomic/functional phenotype: a study in patients with chronic coronary syndromes

BACKGROUND

In stable coronary artery disease (CAD), the prognostic interaction between clinical variables and treatment appropriateness based on anatomic/functional phenotype needs to be evaluated.

METHODS

1585 consecutive patients underwent myocardial perfusion scintigraphy and coronary angiography within 90 days. Obstructive CAD (> 70% stenosis) with downstream moderate-to-severe ischemia (> 10%) was considered significant. Coronary revascularization was considered appropriate if all hemodynamically significant lesions were revascularized, while medical therapy only was deemed appropriate in the absence of hemodynamically significant CAD.

RESULTS

Obstructive CAD and moderate-to-severe ischemia were documented in 1184 (75%) and 466 (29%) patients, respectively. Over mean follow-up of 4.7 ± 2.5 years, the primary endpoint (cardiac death and non-fatal myocardial infarction) occurred in 132 (8.2%) patients. Of patients with obstructive CAD, 797 (67%) were managed appropriately. Patients' management was inappropriate in 389 patients, because either non-hemodynamically significant lesions were revascularized (50%, including 2 patients with non-obstructive lesions being inappropriately revascularized) or ischemia-causing CAD was left untreated (50%). At multivariate analysis, an inappropriate management (P < .001) was correlated with the primary endpoint, together with previous myocardial infarction (P = .009), lower ejection fraction (P < .001) and higher glucose levels (P < .001).

CONCLUSIONS

In stable CAD patients, management based on anatomic/functional phenotyping was correlated with a prognostic advantage at long-term follow-up. Correlation between treatment categories and patients' prognosis. A significantly higher event-rate was observed in patients where hemodynamically significant coronary lesions were left untreated-either because MT was not-adherently chosen or in the case of incomplete revascularization-than in those that were revascularized completely (17.6% vs 5.1%; P < .001). Conversely, the revascularization of non-hemodynamically significant CAD correlated with a higher event-rate than that of similar patients managed medically (13.8% vs 8.3%, P = .04). The event-rate of patients in whom coronary revascularization was performed in the presence of hemodynamically significant CAD ('appropriate revascularization') was similar to those with "No CAD/non-obstructive CAD" (5.1% vs 3.5%; P = NS).