Gulati, R

Abstract Aims Coronary angiography might contain clinically relevant information, beyond its traditional role in delineating coronary artery disease. We sought to explore the role of deep learning (DL) in detecting right ventricular dysfunction from routinely acquired cine images during coronary angiography. Methods and results 3D-convolutional neural networks (CNN) were trained to identify right ventricle (RV) dysfunction from diagnostic angiography of the right coronary artery (RCA). The model’s input was two cine angiograms from LAO and RAO projections. The model’s output was normal vs. abnormal RV function. Ground truth on RV function was obtained from the nearest transthoracic echocardiogram to the coronary angiogram. Two analyses were performed to identify (i) any RV dysfunction (≥mild) and (ii) significant RV dysfunction (≥mild–moderate). A total of 10 336 coronary angiograms from 9849 patients (mean age 66, 36% women) were included. The cohort was split into training (70%), validation (15%), and testing (15%) sets. In the testing sets, the models attained an area under the ROC curve (AUC), sensitivity, and specificity of 0.82 (95% CI: 0.80–0.84), 0.75, and 0.74, for detecting any RV dysfunction, and 0.83 (95% CI: 0.80–0.86), 0.82, and 0.70 for detecting significant (≥mild–moderate) RV dysfunction. Combining an ECG-driven AI model with the angio-driven model improved the AUC values to 0.83 and 0.87 for detecting any or advanced RV dysfunction, respectively. Conclusion A novel deep learning algorithm yielded acceptable classification accuracy in detecting RV dysfunction from routine cine angiography of the RCA. The model’s performance improved further by adding ECG to its inputs.

Abstract Background Elevated left ventricular filling pressure (LVFP) is associated with an increased risk of heart failure and mortality. Pulmonary artery wedge pressure (PAWP) is considered the gold standard for assessing LVFP. Recently, we validated an artificial intelligence-enabled electrocardiogram (AI-ECG) to assess LV diastolic function (LVDF) and LVFP. Herein, we compare AI-ECG with PAWP in predicting long-term mortality. Methods We included adults who had comprehensive echocardiography with available ECG within 14-days and right-heart-catheterization (RHC) within 30-days of echocardiography. Elevated-LVFP was defined as PAWP>15 mmHg by RHC, grades 2-3 LVDF by echocardiography, or as adjudicated by the AI-ECG algorithm. Prognosis by AI-ECG was compared to that by echocardiography and PAWP. Results Among 4,168 patients (median age 64 years, 47.7% female), 1,841 had normal-LVFP, and 2,327 had elevated-LVFP by RHC. Echocardiography classified 1,502 patients as normal-LVFP, 1,643 as elevated-LVFP, and 1,023 as indeterminate-LVFP. AI-ECG demonstrated an area-under-the-curve of 0.879 (95%-confidence-interval 0.867-0.891) for predicting elevated-LVFP as determined by echocardiography and 0.742 (95%-confidence-interval 0.724-0.759) for predicting elevated-LVFP as determined by RHC. Over a median follow-up of 7.7 years, 2,735 (34.4%) patients died. In multivariable models, elevated-LVFP by any modality was associated with an increased risk of death (AI-ECG: adjusted-hazard-ratio 1.80, 95% confidence-interval 1.55-2.09; echocardiography: adjusted-hazard-ratio 1.74, 95% confidence-interval 1.51-2.02; PAWP: adjusted-hazard-ratio 1.26, 95% confidence-interval 1.11-1.43). AI-ECG-derived LVFP provided incremental prognostic discrimination compared to echocardiography- or RHC-derived LVFP. Conclusions AI-ECG-derived LVFP offers an accessible, affordable, and noninvasive alternative for detecting elevated LVFP, with potential independent prognostic value beyond traditional echocardiographic and invasive assessments.Survival across LVFP concordance

Abstract Background Diagnosing myocardial infarction (MI) in patients with chronic kidney disease (CKD) is difficult as these patients often have elevated baseline high-sensitivity cardiac troponin T (hs-cTnT) concentrations. Methods Observational U.S. cohort study of emergency department (ED) patients undergoing hs-cTnT measurement. Cases with >1 hs-cTnT increase >99th percentile were adjudicated following the Fourth Universal Definition of MI. Diagnostic performance of baseline and serial 2- and 6-hour hs-cTnT thresholds for ruling-in acute MI was compared between those without and with CKD (eGFR <60 ml/min/1.73m2). Results The study cohort included 1992 patients, amongst whom 501 (25%) had CKD. There were 701 (82%) patients with myocardial injury, 64 (7.5%) with type 1, and 91 (11%) with type 2 MI. In CKD patients with baseline hs-cTnT thresholds of >52, >100, >200 or >300 ng/L, PPVs for acute MI were 37% [95% CI 28-46], 53% (95% CI 39-67), 73% (95% CI 50-89) and 80% (95% CI 44-98), and in those without CKD, 64% (95% CI 50-77), 69% (95% CI 49-85), 59% (95% CI 33-82) and 54% (95% CI 25-81). In CKD patients with a 2-hour hs-cTnT delta of >10, >20 or >30 ng/L, the corresponding PPVs were 65% (95% CI 50-78), 86% (95% CI 68-96) and 88% (95% CI 68-97), and in those without CKD, 65% (95% CI 51-78), 73% (95% CI 57-86) and 75% (95% CI 58-88) (Figure). Conclusion Diagnostic performance of standard baseline and serial 2-hour hs-cTnT thresholds to rule-in MI is suboptimal in patients with CKD. It significantly improves when using higher baseline thresholds and delta values.

Abstract Background There are no robust data addressing the United States (US) experience with the 0/2 hours (h) European Society of Cardiology (ESC) algorithm using high-sensitivity cardiac troponin T (hs-cTnT). Purpose To evaluate the diagnostic performance of the 0/2h ESC-recommended absolute deltas in a US population. Methods Observational US cohort study of ED patients undergoing 0 and 2 h hs-cTnT measurements. The 99th percentiles upper reference limits (URL) used were 10 ng/L for women and 15 ng/L for men. Diagnostic outcomes were the adjudicated diagnoses of acute myocardial infarction (MI, type 1 or 2). Results A total of 1084 patients met inclusion criteria, amongst which 49% were women. Chest discomfort was reported in 521 (48%) patients. Among patients with a baseline hs-cTnT≤99th URL (n=434), 4 MIs occurred. To rule-out MI, an absolute 0-2h delta change <4 ng/L resulted in a sensitivity of 100% (95% CI: 39.8, 100) and a negative predictive value (NPV) of 100% (95% CI: 99.1, 100). To rule-in MI, an absolute 0-2h delta ≥10 ng/L had a specificity of 99.8 % (95% CI: 98.7, 99.9) and positive predictive value (PPV) of 75% (95% CI: 19.4, 99.4). Among patients with a baseline hs-cTnT >99th URL (n=650), 110 MIs occurred. In these patients, both rule-out (sensitivity 65%, NPV 91%) and rule-in (specificity 7.2%, PPV 11%) were poor using the ESC-recommended absolute deltas. Among patients with modest hs-cTnT increases (99th URL to 100 ng/L) neither the ESC-recommended absolute nor the Fourth Universal Definition of MI (UDMI)-recommended relative deltas provided acceptable performance. Among patients with marked hs-cTnT increases at presentation (>100 ng/L), a relative delta change >20% as recommended by the UDMI resulted in a superior rule-in performance, with a specificity of 92.0% (95% CI 74.0, 99.0) and PPV of 84.6% (57.4, 95.7). Conclusions In our unselected US-cohort of patients undergoing hs-cTnT, the 0/2h ESC algorithm had a good diagnostic performance in patients without myocardial injury at presentation. Conversely, in patients with myocardial injury at baseline, the ESC-recommended absolute deltas performed poorly to determine who had MI. In patients with marked hs-cTnT increases >100 ng/L at presentation, the UDMI recommended 20% relative change showed good rule-in performance.

Abstract Background Bioprosthetic valve thrombosis is currently a well-recognized cause of bioprosthetic valve dysfunction. It was found to be associated with accelerated degeneration of the bioprosthesis with higher rates of valve re-replacement, even after treatment with anticoagulation. We hypothesized that the use of warfarin for three months after transcatheter aortic valve replacement (TAVR) protects against accelerated valve degeneration and is therefore associated with better outcomes compared to dual antiplatelet therapy (DAPT). Methods Consecutive adult patients who underwent TAVR in our clinic between 2012 and 2019 were identified retrospectively. Patients with atrial fibrillation were excluded. Subsequently, patients who received DAPT were propensity matched to up to 2 patients who received three months of warfarin as part of their anti-thrombotic regimen. Matching was performed for variables that were significantly different at baseline between the two groups and included diabetes mellitus, prior myocardial infarctions, chronic lung disease, peripheral arterial disease, hemoglobin at time of TAVR, kidney function [creatinine>2], use of angiotensin-converting enzyme inhibitors / angiotensin II receptor blockers, beta blockers, the Society of Thoracic Surgeons (STS) score [STS ≥8, STS 4–8, STS<4], and valve size. The two groups were then compared for outcomes of ischemic stroke, death, valve re-replacement/intervention, the composite endpoint of the aforementioned three outcomes, as well as the three-month outcome of hemorrhagic strokes. Kaplan Meier was used for outcome analysis, and discharge date was considered time zero. Patients who had their anti-thrombotic therapy interrupted were censored at that time point. Results A total of 1,373 patients who underwent TAVR were identified. Of these, 576 patients with atrial fibrillation were excluded. Baseline characteristics were compared between 633 patients who received three months of warfarin and 164 patients who received DAPT after TAVR. After matching the two groups, 435 patients were included in the final analysis [warfarin in 281, DAPT in 154; median time to last follow up 2.61 years], Table 1. There was no difference in matched (Figure 1) or unmatched analysis (not shown) in outcomes of ischemic stroke, death, valve re-replacement/intervention, their composite endpoint, or hemorrhagic strokes (p>0.05 for all). Conclusion Antithrombotic regimen including three months of warfarin after TAVR was not associated with better outcomes of ischemic strokes, deaths, and valve re-replacement/intervention or with increased risk of hemorrhagic strokes compared to DAPT. Funding Acknowledgement Type of funding sources: None.

Abstract Background Artificial intelligence-augmented electrocardiogram (AI-ECG) algorithms have been developed from the standard 12-lead ECG and validated for the recognition of left ventricular systolic dysfunction (LVSD), defined as LV ejection fraction (LVEF)≤35%. Whether AI-ECG facilitates identification of LVSD and is associated with adverse outcomes in emergency department (ED) patients undergoing high-sensitivity cardiac troponin (hs-cTnT) testing is uncertain. Purpose To investigate the diagnostic and prognostic performance of AI-ECG in ED patients undergoing hs-cTnT measurement. Methods Observational US cohort study of ED patients undergoing hs-cTnT measurement. Cases with hs-cTnT increases >sex-specific 99th percentiles were adjudicated following the Fourth Universal Definition of Myocardial Infarction (MI). Post-discharge major adverse cardiac events (MACE) included death, MI, heart failure (HF) hospitalization, stroke or transient ischemic attack, and new onset atrial fibrillation/flutter during 2-years follow-up. The AI-ECG network output, which is a continuous number between 0–1, that provides a probability of LVSD, was obtained for each patient from the first ECG during the index presentation. An AI-ECG threshold of ≥0.256 indicates a positive screen that correlates with a high probability of LVSD. Results Among 1977 patients, 1729 (87%) had a negative AI-ECG screen, while 248 (13%) had a positive AI-ECG screen. Patients with a positive AI-ECG screen were older and had more comorbidities. As compared to patients with hs-cTnT≤99th percentile in whom AI-ECG was positive in 5.8%, those with hs-cTnT>99th percentile had a positive AI-ECG in 22% of cases (p<0.0001). Based on adjudicated diagnoses, the frequency of a positive AI-ECG was 20% in myocardial injury, 38% in type 1 MI, and 20% in type 2 MI. At 2-years follow-up, as compared to patients with a negative AI-ECG, those with a positive AI-ECG had a higher risk for MACE (48% vs. 21%, p<0.0001, adjusted HR 1.39, 95% CI 1.11–1.75) (Figure 1), mainly because of more deaths (43% vs. 30%, p=0.004) and HF hospitalizations (36% vs. 13%, p<0.0001). A positive AI-ECG was associated with a higher risk for MACE (60% vs. 41%, p<0.0001, adjusted HR 1.30, 95% CI 1.02–1.64) in those with hs-cTnT increases >99th percentile, but not in those without hs-cTnT increases. Among patients with an echocardiogram during index presentation or within 30-days (n=452), the diagnostic accuracy of AI-ECG for LVEF ≤35% was 81.4% (95% CI 77.5, 84.9) with a negative predictive value of 96.5% (95% CI 94.0, 98.2). A normal LVEF (>50%) was observed in 87% of those with a negative AI ECG, whereas in those with a positive AI-ECG LVEF was reduced (<50%) in 60%. Conclusions Among ED patients evaluated with hs-cTnT, a positive AI-ECG screen for LVSD identifies patients at high risk of MACE. These findings are largely because of more deaths and HF hospitalizations in those with hs-cTnT increases >sex-specific 99th percentiles. Funding Acknowledgement Type of funding sources: None.