Device-measured S3 showed a stronger stratification power than auscultation when assessed at follow-up visits

Background The third heart sound (S3) is caused by passive filling of blood into the ventricles and is a specific sign of worsening heart failure (HF). S3 is traditionally assessed using a stethoscope by experienced physicians. More recently, accelerometers embedded in implantable medical devices were used to quantify S3. We compared the stethoscope vs accelerometer for identifying patientu0027s risk of acute HF decompensation (ADHF). Methods The MultiSENSE study enrolled patients with cardiac resynchronization therapy defibrillators and followed them up for up to a year. Modified firmware collected heart sound data that were processed to obtain continuous S3 (device-measured S3, devS3). Auscultation was performed to assess the loudness of S3 (auscultated S3, audS3) while 7-day averaged accelerometer signals were used to assess S3 with a threshold of 1 mG. ADHF events were adjudicated by an independent committee blinded to sensor data. A Cox proportional hazard model was used to quantify the stratification power of the two S3 measures between FU visits and both measures were used to predict patients with ADHFs. Conclusion Device-measured S3 showed a stronger stratification power and better identified patients with ADHFs than auscultation between follow-up visits. Automatic, serial measurements of device-measured S3 may enable chronic monitoring of the risk of acute decompensations in HF patients. Result A total of 900 patients were enrolled in the MultiSENSE study. FU visits that were more than 98 days from previous visits were excluded. The remaining 5610 FU visits (mean interval 53 days +-25 days) were included in the analysis within which 189 ADHF events occurred. AudS3 had a hazard ratio of 1.8 between groups with and without audS3 (p=0.0055). DevS3 had a stronger stratification power with a hazard ratio of 5.6 (p