Comparison Between Anatomical and Functional Imaging Modalities for Evaluation of Chest Pain in the Emergency Department

Role of Copeptin and hs-cTnT to Discriminate AHF from Uncomplicated NSTE-ACS with Baseline Elevated hs-cTnT-A Derivation and External Validation Study

BACKGROUND

In light of overlapping symptoms, discrimination between non-ST-elevation (NSTE) acute coronary syndrome (ACS) and acute heart failure (HF) is challenging, particularly in patients with equivocal clinical presentation for suspected ACS. We sought to evaluate the diagnostic and prognostic properties of copeptin in this scenario.

METHODS

Data from 1088 patients from a single-center observational registry were used to test the ability of serial high sensitivity cardiac troponin T (hs-cTnT)-compared to copeptin, or a combination of copeptin with hs-cTnT-to discriminate acute HF from uncomplicated non-ST-elevation myocardial infarction (NSTEMI) and to evaluate all-cause mortality after 365 days. Patients with STEMI, those with unstable angina and either normal or undetectable hs-cTnT concentrations were excluded. The findings were validated in an independent external NSTE-ACS cohort.

RESULTS

A total of 219 patients were included in the analysis. The final diagnosis was acute HF in 56 and NSTE-ACS in 163, with NSTEMI in 78 and unstable angina having stable elevation of hs-cTnT >ULN in 85. The rate of all-cause death at 1 year was 9.6% and occurred significantly more often in acute HF than in NSTE-ACS (15 vs. 6%, p < 0.001). In the test cohort, the area under the receiver operator curve (AUC) for the discrimination of acute HF vs. NSTE-ACS without HF was 0.725 (95% confidence interval [CI] 0.625-0.798) for copeptin and significantly higher than for hs-cTnT at 0 h (AUC = 0.460, 0.370-0.550) or at 3 h (AUC = 0.441, 0.343-0.538). Copeptin and hs-cTnT used either as continuous values or at cutoffs optimized to yield 90% specificity for acute HF were associated with significantly higher age- and sex-adjusted risk for all-cause mortality at 365 days. The findings from the test cohort were consistently replicated in the independent external NSTE-ACS validation cohort.

CONCLUSIONS

High concentrations of copeptin in patients with suspected NSTE-ACS and equivocal clinical presentation suggest the presence of acute HF compared to uncomplicated NSTE-ACS and are associated with higher rates of all-cause death at 365 days.

Transfer learning enables prediction of myocardial injury from continuous single-lead electrocardiography

OBJECTIVE

Chest pain is common, and current risk-stratification methods, requiring 12-lead electrocardiograms (ECGs) and serial biomarker assays, are static and restricted to highly resourced settings. Our objective was to predict myocardial injury using continuous single-lead ECG waveforms similar to those obtained from wearable devices and to evaluate the potential of transfer learning from labeled 12-lead ECGs to improve these predictions.

METHODS

We studied 10 874 Emergency Department (ED) patients who received continuous ECG monitoring and troponin testing from 2020 to 2021. We defined myocardial injury as newly elevated troponin in patients with chest pain or shortness of breath. We developed deep learning models of myocardial injury using continuous lead II ECG from bedside monitors as well as conventional 12-lead ECGs from triage. We pretrained single-lead models on a pre-existing corpus of labeled 12-lead ECGs. We compared model predictions to those of ED physicians.

RESULTS

A transfer learning strategy, whereby models for continuous single-lead ECGs were first pretrained on 12-lead ECGs from a separate cohort, predicted myocardial injury as accurately as models using patients' own 12-lead ECGs: area under the receiver operating characteristic curve 0.760 (95% confidence interval [CI], 0.721-0.799) and area under the precision-recall curve 0.321 (95% CI, 0.251-0.397). Models demonstrated a high negative predictive value for myocardial injury among patients with chest pain or shortness of breath, exceeding the predictive performance of ED physicians, while attending to known stigmata of myocardial injury.

CONCLUSIONS

Deep learning models pretrained on labeled 12-lead ECGs can predict myocardial injury from noisy, continuous monitor data early in a patient's presentation. The utility of continuous single-lead ECG in the risk stratification of chest pain has implications for wearable devices and preclinical settings, where external validation of the approach is needed.

Reduced coronary flow velocity reserve and blunted heart rate reserve identify a higher risk group in patients with chest pain and negative emergency department evaluation

To estimate the prognostic value of stress echo (SE) with the assessment of coronary flow velocity reserve (CFVR) and heart rate reserve (HRR) in patients admitted for chest pain with non-diagnostic EKG, negative troponin, and without inducible regional wall motion abnormalities (RWMA). 658 patients (age 67 ± 12 years) admitted to our Emergency Department with chest pain, non-diagnostic EKG, and negative serial troponin underwent dipyridamole (0.84 mg/kg in 6') SE with simultaneous assessment of RWMA, CFVR in the left anterior descending artery, and HRR as peak/rest heart rate. The outcome measure was all-cause mortality. Of the 658 patients initially enrolled, 20 (3%) showed RWMA during SE and were referred to ischemia-driven revascularization. In the remaining 638, CFVR was abnormal (≤ 2.0) in 148 patients (23%). HRR was abnormal (≤ 1.22 in patients in sinus rhythm, or ≤ 1.17 in patients with permanent atrial fibrillation) in 196 patients (31%). During a follow-up of 7.3 ± 4.3 years, 151 (24%) patients died. Survival at 8 years was 93% in patients with normal CFVR and HRR, 76% in patients with abnormal CFVR only, 73% in patients with abnormal HRR only, and 38% in those with abnormal CFVR and HRR (p < 0.0001). At multivariable analysis, abnormal CFVR (HR 1.49, 95% CI 1.05-2.10, p = 0.02) and abnormal HRR (HR 2.01, 95% CI 1.43-2.84, p < 0.0001) were independent predictors of survival. In admitted patients with non-ischemic EKG, negative serial troponin, and without RWMA during dipyridamole SE, a reduced CFVR and blunted HRR independently identify a subset with worse survival in the long term. Upper panel: Color and pulsed-wave Doppler with the electrocardiographic lead tracing of Four different response patterns (from left to right): normal CFVR and HRR; normal CFVR, abnormal HRR; abnormal CFVR, normal HRR; abnormal CFVR and HRR. Lower panel: The annualized death rate for each of the four groups with negative SE for RWMA and stratified according to the presence of CFVR and HRR: none, one, or two abnormalities.

Desempenho Diagnóstico da Angiotomografia Computadorizada e da Avaliação Seriada de Troponina Cardíaca Sensível em Pacientes com Dor Torácica e Risco Intermediário para Eventos Cardiovasculares

Fundamento

A angiotomografia coronária (ATC) tem sido usada para avaliação de dor torácica principalmente em pacientes de baixo risco, e poucos dados existem com pacientes em risco intermediário.

Objetivo

Avaliar o desempenho de medidas seriadas de troponinas sensíveis e de ATC em pacientes de risco intermediário.

Métodos

Um total de 100 pacientes com dor torácica, TIMI score 3 ou 4 e troponina negativa foram prospectivamente incluídos. Todos os pacientes foram submetidos à ATC, e aqueles com obstruções ≥ 50% foram encaminhados à cineangiocoronariografia. Pacientes com lesões < 50% recebiam alta hospitalar, receberam alta e foram contatados 30 dias depois por telefonema para avaliação dos desfechos clínicos. Os desfechos foram hospitalização, morte, e infarto agudo do miocárdio em 30 dias. A comparação entre os métodos foi realizada pelo teste de concordância kappa. O desempenho das medidas de troponina e da ATC na detecção de lesões coronárias significativas e desfechos clínicos foi calculado. Os resultados foram considerados estatisticamente significativos quando p <0,05.

Resultados

Estenose coronária ≥ 50% na ATC foi encontrada em 38% dos pacientes e lesões coronárias significativas na angiografia coronária foram encontradas em 31 pacientes. Dois eventos clínicos foram observados. A análise de concordância Kappa mostrou baixa concordância entre as medidas de troponina e ATC na detecção de lesões coronárias significativas (kappa = 0,022, p = 0,78). O desempenho da ATC para detectar lesões coronárias significativas na angiografia coronária ou para prever eventos clínicos em 30 dias foi melhor que as medidas de troponina sensível (acurácia de 91% versus 60%).

Conclusão

ATC teve melhor desempenho que as medidas seriadas de troponina na detecção de doença coronariana significativa em pacientes com dor torácica e risco intermediário para eventos cardiovasculares.