Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): Academy of Finland, Turku University Hospital Background Coronary computed tomography angiography (CTA) and myocardial perfusion imaging (MPI) are powerful non-invasive tools to evaluate the patients with suspected coronary artery disease (CAD). Purpose The goal of this study was to evaluate the incremental value of these imaging methods in predicting short- and long-term cardiac events using machine learning (ML) approaches. Methods 2411 patients with clinically suspected CAD underwent coronary CTA and subsequent positron emission tomography (PET) MPI. Following our local routine, if obstructive CAD cannot be ruled-out by coronary CTA, PET MPI using 15O-water is performed. The incremental prognostic value of PET imaging was tested using several ML models of which XGBoost models consistently outperformed others and were chosen for further analyses. XGBoost models were trained by using clinical data from medical records and coronary imaging findings: one set of models used all the available variables, while a second set of models used only clinical and CTA-based variables. Differences in the performances of the models were then used to assess the incremental value of PET perfusion variables in outcome prediction. Results After the removal of incomplete data entries, data from 2284 patients was retained for further analysis. During 8-year follow-up period 210 patients had a major cardiac event (these endpoints correspond to 59 myocardial infarctions, 35 unstable angina pectoris, and 116 deaths). The PET perfusion imaging data improved the predictive power of CTA during the first 4 years of observation time. After that, no significant difference in the predictive power was observed between the considered sets of XGBoost models, which either included or did not include PET perfusion data in the input variables. The highest area under the receiver operating characteristic curve (AUC) was at the observation time of 2 years (0.81, 95% CI 0.805–0.823) when PET data were included. The corresponding AUC when PET data were not included was at the observation time of 2 years (0.79, 95% CI 0.785–0.802). Conclusions Based on ML approach PET perfusion imaging improves the power in predicting cardiac events over anatomical CTA imaging for the first 4 years. The results illustrate the differences and complementary nature of anatomic and perfusion information in predicting outcome of patients with suspected CAD.
Background Coronary CT angiography (CTA) combined with myocardial perfusion imaging accurately detects both non-obstructive and obstructive coronary artery disease (CAD). Lipid-lowering therapy is known to effectively reduce cardiovascular events, but the impact of non-invasive imaging findings on the usage of lipid-lowering therapy remains largely unknown. Purpose To assess the use of lipid-lowering medication in patients referred to coronary CTA and subsequent positron emission tomography (PET) myocardial perfusion imaging due to suspected obstructive CAD. Methods We retrospectively analyzed data on purchases of lipid-lowering drugs obtained from the Finnish national registry for a time period starting 1 year before and ending 2 years after the date of diagnostic imaging. This time period was divided into six 6-month intervals, for each of which the presence of any statin and/or ezetimibe purchase was recorded. According to the local routine, patients with suspected obstructive (≥50%) stenosis on coronary CTA entered downstream 15O-water PET myocardial perfusion imaging during adenosine stress to assess the hemodynamic significance of the coronary stenosis. The use of medication was compared among patients with normal coronary arteries, non-obstructive CAD, and obstructive CAD. Results During 2008–2016, a total of 1973 patients (41% male, median age 63 years) underwent coronary CTA and 33% of these had PET perfusion imaging. There were 9081 purchases of lipid-lowering drugs during the 3-year observation period. There were 676 (34%) patients with normal coronary arteries, 640 (32%) patients with non-obstructive atherosclerosis on CTA, 325 (16%) patients with suspected obstructive stenosis on CTA but normal PET perfusion, and 332 (17%) patients with obstructive stenosis and abnormal hyperemic PET perfusion. The proportion of patients buying lipid-lowering drugs was 24%, 51%, 72%, and 91%, respectively, as assessed within the 6-month interval following the CTA/PET imaging. After diagnostic testing, proportion of patients purchasing lipid-lowering medications increased in all groups except those with normal coronary arteries (Figure 1). However, there was a marked decrease in all patient groups in proportion of patients using lipid-lowering medication towards the end of the 2-year follow-up period. Conclusions In a real-world cohort of symptomatic patients with chest pain undergoing diagnostic imaging for suspected CAD, subsequent purchases of lipid-lowering drugs increase in relation to the severity of imaging findings. Although majority of patients with obstructive CAD initially purchased lipid-lowering medication, our results indicate a marked decrease in the use of these preventive medications towards the end of 2-year follow-up period. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): State Research Funding of Turku University Hospital; the Academy of Finland; Finnish Foundation for Cardiovascular Research.
Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): State Research Funding for Turku University Hospital. Background Coronary computed tomography angiography (CCTA) is recommended for diagnostic workup of coronary artery disease (CAD) and is known to provide prognostic information. However, it is unknown whether there is time-dependent variation in the prognostic strength of CCTA. Purpose To study the time-varying prognostic value of CCTA in patients with suspected CAD. Methods From two academic medical centers, we identified symptomatic patients who underwent CCTA for suspected CAD. A composite endpoint of all-cause mortality and myocardial infarction was recorded. The prognostic value of non-obstructive or obstructive (≥50%) CAD was assessed by Cox proportional hazard model. The prognostic power of the model according to the cumulative length of follow-up was evaluated by measuring area under the receiver operating characteristic (ROC) curve (AUC) as a function of follow-up time (time-dependent AUC). Results A total of 2662 patients (age 61 ± 10 years, 45% male) were included, with 247 adverse events (9%) during a median follow-up time of 6.9 years (25th–75th percentile 4.9–8.8 years; maximum 13.3 years). Based on CCTA, 804 (30%) patients had no CAD, 963 (36%) had non-obstructive CAD, and 895 (34%) had obstructive CAD. The presence of non-obstructive CAD (adjusted HR 2.24; p = 0.002) or obstructive CAD (adjusted HR 4.42; p<0.001) were independent predictors of adverse outcome. The predictive power of CCTA findings was relatively stable until 10 years of follow-up time, as demonstrated in the Figure showing time-dependent AUC of the adjusted model (solid line) with 95% confidence intervals (dashed lines). A model including only clinical variables is shown for comparison. Conclusion Coronary CTA findings retain their prognostic value in at least 10-year follow-up of symptomatic patients evaluated for suspected CAD.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.