OBJECTIVE
To identify parameters of non-invasively recorded cardiac biosignals associated with the development of adverse events in patients with chronic heart failure.
MATERIALS AND METHODS
A systematic literature review was carried out using the PRISMA methodology. PubMed and eLibrary databases were used. Search depth — from 1995 to 2025. For this study, 42 articles were selected for the final analysis.
RESULTS
The analyzed signals included electrocardiogram (ECG) (office or 24-hour Holter monitoring) in 28 studies, sphygmogram — in 1 study, acoustic cardiogram, impedance cardiogram, blood pressure (office measurement or 24-hour monitoring) — 4 studies for each biosignal, and pulse rate — 1 study (office measurement); in addition, 95% of the sample were cohort observational studies. The risk of bias in cohort studies was assessed using the Russian version of the Newcastle-Ottawa scale. It was shown that the standard deviation of NN intervals (SDNN, threshold values from 44 to 112 ms) calculated on the basis of 24-hour ECG monitoring data was the most common (in 10 studies) independent predictor of adverse events in a follow-up period of 6 months or more. A randomized controlled trial showed that treatment guided by EMAT% (target value <15%) and S3 score (target value <5) of the acoustic cardiogram was associated with improved outcomes. When combined with clinical status parameters (natriuretic peptide levels, NYHA functional class, peak oxygen consumption), the predictive value of some parameters increases. Information on threshold values of biosignals for scientific and practical application is summarized.
CONCLUSION
It is advisable to simultaneously record multiple biosignals and use a combination of indicators with clinical status parameters to improve the accuracy of predicting adverse events in patients with chronic heart failure.