Search

Objectives: This study developed a neural network to perform automated pressure waveform analysis and allow real-time accurate identification of damping., Background: Damping of aortic pressure during coronary angiography must be identified to avoid serious complications and make accurate coronary physiology measurements. There are currently no automated methods to do this, and so identification of damping requires constant monitoring, which is prone to human error., Methods: The neural network was trained and tested versus core laboratory expert opinions derived from 2 separate datasets. A total of 5,709 aortic pressure waveforms of individual heart beats were extracted and classified. The study developed a recurrent convolutional neural network to classify beats as either normal, showing damping, or artifactual. Accuracies were reported using the opinions of 2 independent core laboratories., Results: The neural network was 99.4% accurate (95% confidence interval: 98.8% to 99.6%) at classifying beats from the testing dataset when judged against the opinions of the internal core laboratory. It was 98.7% accurate (95% confidence interval: 98.0% to 99.2%) when judged against the opinions of an external core laboratory not involved in neural network training. The neural network was 100% sensitive, with no beats classified as damped misclassified, with a specificity of 99.8%. The positive predictive and negative predictive values were 98.1% and 99.5%. The 2 core laboratories agreed more closely with the neural network than with each other., Conclusions: Arterial waveform analysis using neural networks allows rapid and accurate identification of damping. This demonstrates how machine learning can assist with patient safety and the quality control of procedures., (Copyright © 2019 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)

Background: Pressure-derived coronary flow reserve (CFR pres ) and pressure-bounded CFR (CFR pb ) enable simple estimation of CFR from routine pressure measurements, but have been inadequately validated. We sought to compare CFR pres and CFR pb against flow-derived CFR (CFR flow ) in terms of diagnostic accuracy, as well as regarding their comparative prognostic relevance., Methods: We evaluated 453 intermediate coronary lesions with intracoronary pressure and flow measurements. CFR was defined as hyperemic flow/baseline flow. The lower bound (CFR pres ) and upper bound of CFR pb were defined as √[(ΔP hyperemia ) / (ΔP rest )] and [(ΔP hyperemia ) / (ΔP rest )], respectively. Long-term follow-up (median: 11.8-years) was performed in 153 lesions deferred from treatment to document the occurrence of major adverse cardiac events (MACE) defined as a composite of cardiac death, myocardial infarction and target vessel revascularization. CFR < 2.0 was considered abnormal., Results: CFR pb was normal or abnormal in 56.7% of stenoses, and indeterminate in 43.3% of stenoses. There was a poor diagnostic agreement between CFR pres and CFR pb with CFR flow (overall agreement: 45.5% and 71.6% of vessels, respectively). There was equivalent risk for long-term MACE for lesions with abnormal versus normal CFR pres (Breslow p = 0.562), whereas vessels with abnormal CFR flow were significantly associated with increased long-term MACE (Breslow p < 0.001). For vessels where CFR pb was abnormal or normal, there was equivalent risk for long-term MACE for vessels with abnormal versus normal CFR pb (Breslow p = 0.194), whereas vessels with abnormal CFR flow were associated with increased MACE rates over time (Breslow p < 0.001)., Conclusions: Pressure-derived estimations of CFR poorly agree with flow-derived measurements of CFR, which may explain the inferior association with long-term MACE as compared to flow-derived CFR., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Objectives: The purpose of this study is to evaluate whether coronary flow capacity (CFC) improves discrimination of patients at risk for major adverse cardiac events (MACE) compared with coronary flow reserve (CFR) alone, and to study the diagnostic and prognostic implications of CFC in relation to contemporary diagnostic tests for ischemic heart disease (IHD), including fractional flow reserve (FFR)., Background: Although IHD results from a combination of focal obstructive, diffuse, and microcirculatory involvement of the coronary circulation, its diagnosis remains focused on focal obstructive causes. CFC comprehensively documents flow impairment in IHD, regardless of its origin, by interpreting CFR in relation to maximal flow (hyperemic average peak flow velocity [hAPV]), and overcomes the limitations of using CFR alone. This is governed by the understanding that ischemia occurs in vascular beds with substantially reduced hAPV and CFR, whereas ischemia is unlikely when hAPV or CFR is high., Methods: Intracoronary pressure and flow were measured in 299 vessels (228 patients), where revascularization was deferred in 154. Vessels were stratified as having normal, mildly reduced, moderately reduced, or severely reduced CFC using CFR thresholds derived from published data and corresponding hAPV percentiles. The occurrence of MACE after deferral of revascularization was recorded during 11.9 years of follow-up (quartile 1: 10.0 years, quartile 3: 13.4 years)., Results: Combining CFR and hAPV improved the prediction of MACE over CFR alone (p = 0.01). After stratification in CFC, MACE rates throughout follow-up were strongly associated with advancing impairment of CFC (p = 0.002). After multivariate adjustment, mildly and moderately reduced CFC were associated with a 2.1-fold (95% confidence interval: 1.1 to 4.0; p = 0.017), and 7.1-fold (95% confidence interval: 2.9 to 17.1; p < 0.001) increase in MACE hazard, respectively, compared with normal CFC. Severely reduced CFC was identified by FFR ≤0.80 in 90% of cases, although ≥40% of vessels with normal or mildly reduced CFC still had an FFR ≤0.80., Conclusions: CFC provides a cross-modality platform for the diagnosis and risk-stratification of IHD and enriches the interpretation of contemporary diagnostic tests in IHD., (Copyright © 2015 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)

Objectives: This study sought to understand the physiological basis of baseline distal-to-aortic pressure ratio (Pd/Pa) and fractional flow reserve (FFR) agreement and discordance, using coronary flow reserve (CFR), stenosis resistance, and microcirculatory resistance measurements, and form there, to investigate the potential value of combining Pd/Pa with FFR in the diagnostic rationale., Background: Pd/Pa is always available before FFR assessment, and emerging data supports the notion that baseline indices can determine the ischemic potential of coronary stenosis in selected subsets., Methods: A total of 467 stenosed vessels from 363 patients were investigated with pressure and flow sensors during baseline and hyperemia: 168 vessels (135 patients) with thermodilution-derived flow, and 299 vessels (228 patients) with Doppler-derived flow., Results: Pd/Pa correlated more strongly with CFR than FFR (ρ difference = 0.129; p for ρ comparison <0.001). Although Pd/Pa and FFR were closely correlated (ρ = 0.798; 95% confidence interval: 0.767 to 0.828), categorical discordance was observed in 19.3% of total vessels. Such discordance was associated with the patients' clinical profile and was characterized by contrastive changes in stenosis resistance, microcirculatory resistance, and the underlying CFR. Notably, all stenosis with Pd/Pa ≤0.83 (n = 74, 15.8%) progressed to FFR ≤0.80, and although no Pd/Pa cutoff was able to exclude the development of FFR ≤0.80 in the high end of values, only 15 (10.1%) vessels with Pd/Pa ≥0.96 (n = 149, 31.9%) developed FFR ≤0.80, from which none had definite ischemia, as defined by CFR ≤1.74., Conclusions: Combining baseline Pd/Pa with FFR seems to provide a more comprehensive physiological examination of stenosed coronary arteries and a closer pressure-based appraisal of the flow reserve of the downstream myocardial bed., (Copyright © 2015 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)