Journal of electrocardiology最新文献

Background

Methods

Results

Conclusion

Introduction

Methods

Results

Conclusions

Introduction

Electrocardiography (ECG) remains a fundamental tool in cardiovascular diagnostics, frequently relying on System 1 thinking—rapid, intuitive pattern recognition (PR). However, this approach can be insufficient when dealing with complex cases where diagnostic precision is essential. This article emphasizes the importance of integrating System 2 thinking—a more deliberate, evidence-based approach—into ECG interpretation to enhance diagnostic accuracy and avoid clinical errors.

Methods

This review examines the distinction between findings that can be adequately managed through System 1 PR and those requiring System 2 reasoning supported by diagnostic accuracy studies.

Results

While System 1 PR is effective for recognizing routine ECG findings and self-evident truths, it falls short in conditions where the ECG serves as a mere surrogate marker for underlying pathology. Examples such as false-negative acute coronary occlusions illustrate the need for System 2 reasoning to account for the limitations of ECG's diagnostic precision. Relying solely on System 1 in these contexts risks treating the ECG as an infallible diagnostic tool and as a false gold standard for many diseases, which it is not.

Conclusion

To prevent diagnostic errors, ECG interpretation must distinguish between self-evident truths suited for PR and findings that require System 2 reasoning due to their association with actual pathology. Clinicians and educators should prioritize evidence-based methods, incorporating System 2 reasoning into practice to improve diagnostic precision and patient outcomes.

Background

Electrocardiographic diagnosis of acute myocardial infarction in the setting of cardiac pacing represents diagnostic challenge. There are no focusing data, neither reporting about diagnostic sensitivity of 12‑lead ECG with left bundle branch area pacing (LBBAP) during acute myocardial infarction (AMI).

Case summary

We present 12‑lead ECG morphology in a patient with permanent LBBAP during AMI.

Discussion

Abnormal repolarization changes induced by ventricular pacing can lead to delay in diagnosis in patients with AMI. LBBAP and overall conduction system pacing may facilitate a timely diagnosis providing additional, still underestimated, advantages of physiological pacing of the heart.

Background

The impact of P-wave abnormality in acute anterior MI, where the culprit vessel is the left anterior descending artery, remains undetermined. This study aimed to elucidate the impact of P-wave morphology on clinical outcomes in acute anterior MI.

Methods

Patients undergoing emergent percutaneous coronary intervention for acute anterior MI were enrolled between September 2014 and April 2019 (derivation cohort) and May 2019 through July 2023 (validation cohort). P-wave duration (Pd) and P-wave vector magnitude (Pvm) were measured. The Pvm was calculated as the square root of the sum of the squared P-wave magnitudes in leads II and V6 and one-half of the P-wave amplitude in V2. The patients were categorized into high and low Pd/Pvm groups using a statistically derived cut-off value. The endpoint comprised the composite of heart failure (HF) hospitalization and all-cause death.

Results

Consecutive 426 patients were enrolled in this study (derivation cohort, 213 patients; validation cohort, 216 patients). The calculated cut-off value of Pd/Pvm for predicting the clinical endpoint, determined through receiver operating curve analysis, was 793.5 ms/mV (area under the curve [AUC] = 0.85, sensitivity of 73.8 %, and specificity of 94.0 %) in the derivation cohort. Kaplan-Meier analyses revealed a significantly higher risk of the endpoint in patients with high Pd/Pvm than those with low Pd/Pvm in derivation and validation cohorts (Log-rank p < 0.001 and p < 0.001, respectively). Multivariate Cox proportional hazards analysis identified advanced age, elevated Pd/Pvm, and reduced left ventricular ejection fraction as independent and significant factors associated with the endpoint in the validation cohort (p = 0.008, p < 0.001, and p < 0.001, respectively).

Conclusion

High Pd/Pvm was significantly associated with the composite of HF hospitalization and all-cause death after acute anterior MI.

Introduction

Deep learning (DL) models offer improved performance in electrocardiogram (ECG)-based classification over rule-based methods. However, for widespread adoption by clinicians, explainability methods, like saliency maps, are essential.

Methods

On a subset of 100 ECGs from patients with chest pain, we generated saliency maps using a previously validated convolutional neural network for occlusion myocardial infarction (OMI) classification. Three clinicians reviewed ECG-saliency map dyads, first assessing the likelihood of OMI from standard ECGs and then evaluating clinical relevance and helpfulness of the saliency maps, as well as their confidence in the model's predictions. Questions were answered on a Likert scale ranging from +3 (most useful/relevant) to −3 (least useful/relevant).

Results

The adjudicated accuracy of the three clinicians matched the DL model when considering area under the receiver operating characteristics curve (AUC) and F1 score (AUC 0.855 vs. 0.872, F1 score = 0.789 vs. 0.747). On average, clinicians found saliency maps slightly clinically relevant (0.96 ± 0.92) and slightly helpful (0.66 ± 0.98) in identifying or ruling out OMI but had higher confidence in the model's predictions (1.71 ± 0.56). Clinicians noted that leads I and aVL were often emphasized, even when obvious ST changes were present in other leads.

Conclusion

In this clinical usability study, clinicians deemed saliency maps somewhat helpful in enhancing explainability of DL-based ECG models. The spatial convolutional layers across the 12 leads in these models appear to contribute to the discrepancy between ECG segments considered most relevant by clinicians and segments that drove DL model predictions.