Automated assessment of invasive hemodynamics during TAVR showed excellent agreement with manual assessment and comparable discriminatory value for relevant PVR at 30 days (AUC 0.81 vs 0.80).
Observational (n=77)
Does an automated rule-based algorithm accurately assess invasively-measured hemodynamics and predict relevant paravalvular regurgitation compared to manual assessment in patients undergoing TAVR?
An automated algorithm for assessing invasive hemodynamics during TAVR shows excellent agreement with manual assessment and strong discriminatory value for 30-day CMR-derived paravalvular regurgitation, supporting its use for real-time assessment.
Absolute Event Rate: 0.81% vs 0.8%
Abstract Introduction Paravalvular regurgitation (PVR) is a frequent complication following Transcatheter Aortic Valve Replacement (TAVR). Invasive monitoring of aortic and left ventricular hemodynamics following TAVR has shown promise for diagnosis of PVR, but automated software options to allow real-time assessment are lacking. Aim To develop a rule-based algorithm for automated assessment of invasively-measured hemodynamic indices of PVR, and evaluate its construct validity and discriminatory value for CMR-derived relevant PVR compared to standard manual assessment of hemodynamics. Methods As part of the APPOSE study, left ventricular and aortic pressure signals were invasively measured before and after valve replacement using two fluid-filled pressure catheters. To evaluate construct validity of automated versus manual assessment of invasive hemodynamics, we compared 1) degree of cardiac cycles affected by arrythmias or noise, 2) transvalvular pressure gradients and 3) indices of PVR. In addition, we compared the discriminatory value of automatically and manually-determined hemodynamic indices of PVR for CMR-determined relevant PVR at 30-days. Results In total, 77 patients were enrolled (664 cardiac cycles). Automated filtering of cardiac cycles affected by cardiac arrhythmias or noise demonstrated excellent sensitivity (95.2%) and specificity (86.4%) compared to manual assessment. In addition, excellent agreement was observed between algorithm and manual computation of mean gradients (39.3±12.1 vs 37.5±11.9 mmHg, intraclass correlation coefficient (ICC): 0.916; 1.92±5.87 vs 1.14±5.89, ICC: 0.957, respectively), and indices of PVR (diastolic delta (DD): 41.7±12.4 vs 40.6±12.3 mmHg, ICC: 0.982). Algorithm- and manual assessment of DD showed comparable discriminatory value for relevant PVR (area under the curve (AUC): 0.81 vs 0.80, respectively). Conclusion Rule-based, automated assessment of invasive hemodynamics showed excellent agreement with manual assessment of hemodynamics before and following TAVR, and automatically-derived indices of PVR had strong discriminatory value for relevant PVR at 30-days post-TAVR. Our findings support integrating automated algorithms to allow real-time assessment of invasive hemodynamics in patients undergoing TAVR. Fig1.ROC - indices of PVR
Stens et al. (Sat,) conducted a observational in Paravalvular regurgitation following Transcatheter Aortic Valve Replacement (TAVR) (n=77). Automated rule-based algorithm for assessment of invasive hemodynamics vs. Standard manual assessment was evaluated on Discriminatory value for CMR-determined relevant paravalvular regurgitation at 30-days (AUC). Automated assessment of invasive hemodynamics during TAVR showed excellent agreement with manual assessment and comparable discriminatory value for relevant PVR at 30 days (AUC 0.81 vs 0.80).