Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Mar 5, 2025
Date Accepted: Jun 16, 2025
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Wearable Device-Based Respiratory Complexity Analysis for Detecting Pulmonary Congestion in Acute Decompensated Heart Failure: A Prospective Observational Study
ABSTRACT
Background:
Acute decompensated heart failure (ADHF) decompensation is closely associated with pulmonary congestion (PC), which triggers abnormal breathing patterns. Early detection of PC-driven respiratory changes via wearable devices could enable timely intervention and reduce hospitalizations. However, specific respiratory features linked to PC remain unclear.
Objective:
This study employed a wearable device to analyze nocturnal respiratory signals in hospitalized ADHF patients, comparing those with and without PC.
Methods:
This prospective trial investigated breathing pattern characteristics in ADHF patients hospitalized. PC was assessed via lung ultrasound (LUS) in 28 standardized zones at admission, with patients stratified by LUS-defined severityusing >5 B-lines as the threshold for significant PC. Concurrently, wearable devices were deployed to continuously capture chest and abdominal movement signals for respiratory waveform analysis. Breathing patterns were quantitatively characterized through three dimensions: respiratory cycle, respiratory amplitude and multiscale entropy (MSE). Logistic regression analysis and the receiver operating characteristic (ROC) curve were used to identify risk factors associated with PC and evaluate the ability of respiratory pattern parameters to identify HF combined with PC.
Results:
A total of 62 patients with ADHF were included in the study, 44 of whom had more than 5 B-lines. PC patients exhibited longer mean expiratory time (TE_mean), smaller mean ratio of expiratory time (TE_ratio_mean), and greater MSE values in respiratory amplitude (RA). RA_area_1_5 and RA_area_6_20 were identified as risk factors for PC after adjusting for clinical variables. The established logistic regression model could accurately distinguish whether HF patients complicated with PC, the AUC of the multivariable model constructed using respiratory complexity parameters was 0.910(95% CI: 0.837～0.984, P<.001).
Conclusions:
The study highlights the potential of wearable devices combined with MSE algorithms for monitoring of ADHF patients' respiratory complexity. The identified respiratory complexity parameters, particularly RA_area_1_5 and RA_area_6_20, could serve as an early warning tool for PC exacerbation.
Citation
