Diagnostic Yield and Model Prediction Using Wearable Patch Device in HFpEF

Stud Health Technol Inform. 2024 Jul 24:315:25-30. doi: 10.3233/SHTI240100.

Abstract

Heart failure (HF) is a prevalent global health issue projected to escalate, notably in aging populations. The study aimed to identify predictive markers for Heart Failure with preserved Ejection Fraction (HFpEF). We scrutinized vital parameters like age, BMI, eGFR, and comorbidities like atrial fibrillation, coronary artery disease (CAD), diabetes mellites (DM). Evaluating phonocardiogram indicators-third heart sound(S3) and Systolic Dysfunction Index (SDI)-our logistic regression revealed age (≥ 65years), BMI (≥ 25 kg/m2), eGFR (<60 mL/min/1.73m2), CAD, DM, S3 intensity ≥5, and SDI ≥5 as HFpEF predictors, with AUC = 0.816 (p < .001). ROC diagnosis curve showed that the sensitivity, specificity and Youden's index J of the model were 0.755, 0.673 and 0.838, respectively. Nonetheless, further exploration is crucial to delineate the clinical applicability and constraints of these markers.

Keywords: HFpEF; model; prognostic; wearable.

MeSH terms

  • Aged
  • Female
  • Heart Failure* / diagnosis
  • Heart Failure* / physiopathology
  • Humans
  • Male
  • Middle Aged
  • Phonocardiography
  • Sensitivity and Specificity
  • Stroke Volume
  • Wearable Electronic Devices*