JMIR Publications

ABSTRACT

Background:

Although several biomarkers exist for patients with heart failure (HF), their use in routine clinical practice is often constrained by their cost and limited availability.

Objective:

We examined the utility of an artificial intelligence (AI) algorithm that analyses printed electrocardiograms (ECGs) for outcome prediction in patients with acute HF.

Methods:

We retrospectively analysed prospectively collected data of patients with acute HF in two tertiary centres. Baseline ECGs were analysed using a deep learning system called Quantitative ECG (QCG™) trained to detect several urgent clinical conditions, including shock, cardiac arrest, and reduced left ventricular ejection fraction (LVEF).

Results:

Among the 1,254 patients enrolled, in-hospital cardiac death (IHCD) occurred in 53 (4.2%) patients, and the QCG score for critical events (QCG-Critical) was significantly higher in these patients than in survivors (0.57±0.23 vs. 0.29±0.20, P<.001). QCG-Critical score was an independent predictor of IHCD after adjustment for age, sex, comorbidities, HF aetiology/type, atrial fibrillation, and QRS widening (adjusted odds ratio [OR], 1.68; 95% confidence interval [CI], 1.47–1.92; P<.001, per 0.1 increase), and even after additional adjustments for echocardiographic LVEF and N-terminal pro-B-type natriuretic peptide (adjusted OR, 1.59; 95% CI, 1.36–1.87; P<.001, per 0.1 increase). During long-term follow-up, patients with higher QCG-Critical scores (>0.5) had higher mortality rates than those with low QCG-Critical scores (<0.25) (adjusted hazard ratio, 2.69; 95% CI, 2.14–3.38; P<.001).

Conclusions:

Predicting outcomes in patients with acute HF using the QCG-Critical score is feasible, indicating that the AI-based ECG score may be a novel biomarker for these patients. Clinical Trial: The study design has been registered in ClinicalTrial.gov NCT01389843.