Photoplethysmogram-derived pulse arrival time for acute assessment of cardiac resynchronization therapy response
H Luo, M Witso, J M Aalen, H M Kristiansen, M Flattum, H Skulstad, E Kongsgard, C Tronstad, H H Odland, T Holm, E W RemmeAbstract
Background
About one third of patients receiving cardiac resynchronization therapy (CRT) show limited clinical benefit without left ventricular reverse remodelling. Identifying better tools for assessing the acute hemodynamic effect could help improve overall response rates. Changes in pulse arrival time (PAT) may offer a simple noninvasive marker because PAT incorporates the duration of the pre-ejection interval, which tends to shorten when more synchronous and effective ventricular contractions are achieved after successful CRT.
Purpose
This study aimed to investigate whether PAT derived from photoplethysmogram (PPG) can serve as a noninvasive indicator of acute CRT response.
Methods
Heart failure patients with left bundle branch block were included at routine 6-month follow-up after CRT. PPG signals were obtained using an ear-clip sensor attached to the left earlobe, while a single-lead electrocardiogram (ECG) was simultaneously recorded during biventricular (Biv) and AAI pacing. PAT was measured as the time interval from the onset of the QRS complex on ECG to the upstroke of the PPG waveform (Figure, left panel), averaged over 10 consecutive heartbeats for each pacing mode. CRT response was defined as ≥15% reduction in left ventricular end-systolic volume (ESV) by echocardiography at follow-up.
Results
A total of 31 patients were included. The mean ESV reduction after 6 months was −36±20%. Five patients were classified as non-responders, with an average ESV reduction of −4±7%. In responders, Biv pacing significantly shortened PAT compared with AAI pacing (213±25 ms vs. 237±26 ms; P<0.001) during acute adjustments. In contrast, no significant difference was observed in non-responders (212±29 ms vs. 210±27 ms; P=0.83) (Figure, left panel). PAT was shortened in 24 of 26 responders and prolonged in 4 of 5 non-responders (Figure, right panel). Receiver-operating curve analysis showed a sensitivity of 0.80 and a specificity of 0.96 for detecting CRT response when applying an optimal cut-off of 3.5 ms PAT prolongation from AAI to Biv pacing.
Conclusion
These preliminary findings suggest that PAT may capture the acute hemodynamic effect of CRT in a simple noninvasive manner. PAT could therefore assist intra-procedural evaluation during device implantation, support CRT optimization, and help verify treatment effects during follow-up.