Abstract 2452: A New VF Detection Method During Cardio Pulmonary Resuscitation That Minimizes Hands-off Time
Background: Chest compressions (CC) and Defibrillation with Automated External Defibrillator (AED) is the optimum treatment for ventricular fibrillation (VF) patients. Minimum ‘hands-off’ intervals are required to improve the success rate of defibrillation. In support of such life-saving practice, we developed a two-stage shock advisory system, which takes a decision to pause cardiac massage in a first stage, and to prepare a shock in a second stage. We tested whether the system would meet the AHA recommendations for shock advisory decision in AEDs (specificity (Sp)>95 % for Non Shockable Rhythms (NS), sensitivity (Se)>90 % for VF).
Methods: Recordings of 168 OHCA interventions with AEDs (FredEasy, Schiller Medical SAS, France) were retrospectively processed. 825 ECG segments were identified for analysis, consisted of 10-seconds with CC artefacts followed by 3-seconds with no CC artefacts. Independent reviewers annotated the rhythm seen on the noise-free segment. The annotation is then accepted for the total 10s+3s ECG episode. Annotated ECG segments include: 670 NS (386 asystoles (ASYS), 284 other non shockable rythms (ONS)), 155 VF, with artifact and rhythm distributions reflecting those found in the 168 patients. AED analysis algorithm works in two stages. The first stage analyses 10 seconds of ECG with CC artefacts, the decision can be ‘Continue CC’ or ‘Prepare Shock’. The second stage relies on a short-time ECG analysis with no CC artifacts during the 3 seconds following the end of CC. The decision can be ‘Shock’ or ‘Cancel Shock’.
Results: The first stage of the algorithm correctly advises ‘Continue CC’ for 83.7% of ASYS (323/386) and 87% of ONS (247/284). The warning ‘Prepare Shock’ is correct for 94.2% of VF (146/155). The second stage advises ‘Cancel Shock’ for 97.4% of ASYS (376/386) and 99.7% of ONS (283/284). Shock is recommended in 99.4% of VF (154/155).
Conclusions: The first stage of the algorithm avoids CC pauses in about 85% of the non-shockables rhythms (NS), improving the chances of resuscitation. The sensitivity during CC (94%) is above AHA recommendations. The final decision for shock delivery taken by the second stage with short-time ECG analysis is exceeding the AHA recommendations for both Sp (98% for NS) and Se (99.4%).