Real-time feedback during bag-valve-mask ventilation for out-of-hospital cardiac arrest: a before-and-after study (VEntilation during Cardiac Arrest Resuscitation Study, VECARS-2).

This study compared bag-valve-mask ventilation quality and adherence to the guideline-recommended ventilation volume range during 30:2 cardiopulmonary resuscitation for out-of-hospital cardiac arrest before and after the implementation of real-time feedback. We conducted a before-and-after observational study of two sequential adult cohorts with out-of-hospital cardiac arrest managed by basic life support teams from the Paris Fire Brigade. A bidirectional sensor measured insufflated volume (Vi) and expiratory tidal volume (Vt). The display was masked during Period-1 and provided real-time feedback during Period-2. We assessed adherence to the guideline-recommended range using mixed-effects regression; we analyzed Vt and leak volume using gamma log-link generalized linear mixed models. We analyzed 4,300 ventilations in 163 patients, 96 without feedback and 67 with feedback. Median Vi was lower with feedback (471 mL) than without (513 mL, p=0.03 for the first ventilation of each cycle) but median Vt was unchanged (235 vs. 243 mL, p=0.84). Vi dispersion was reduced by 35% (mean within-patient IQR ratio 0.65 [95%-CI, 0.53-0.77]). Vi was 400-600 mL in 68% of ventilations with feedback vs. 42% without feedback. In contrast, Vt was 400-600 mL in only 14% of ventilations with feedback vs. 15% without feedback. In multivariable analysis, leak volume was 12% lower with feedback (mean ratio, 0.88 [95%-CI, 0.77-0.99]). Real-time feedback was associated with improved adherence to the guideline-recommended range for Vi, less dispersed Vi,and lower leak volume. Causal inference, clinical significance, and generalizability warrant further evaluation.