Improving cardiopulmonary resuscitation in the emergency department by real-time video recording and regular feedback learning

Aim of the study: Improvement in the quality of cardiopulmonary resuscitation (CPR) may improve the survival rate following cardiac arrest. The aims of our study were to describe how recording of CPR maneuvers performed in our emergency department with real-time video and regular feedback learning may improve CPR.

Methods: A digital video-recording system enabled us to record and analyze CPR procedures for adult patients from March 2007 to July 2008. Our resuscitation teams received video-recording feedback learning every week.

Results and conclusions: We analyzed 45 cases, divided into three groups of 15 consecutive patients. Instantaneous rates of chest compression showed variation with 75% exceeding 110 cpm. There was a significant difference in instantaneous rates among groups (135 [112-150] in group 1, 123 [110-136] in group 2 and 124 [111-137] cpm in group 3, P<0.001). Ratio of hands-off time to total manual compression time (%) significantly decreased over time (Spearman correlation=-0.30, P=0.04). There were significant differences in hands-off time per minute among the groups (11 [3-28], 6 [2-21] and 7 [2-19] s min(-1), P<0.001). There was a significant improvement in time delay to first chest compression (11 [5-50], 20 [8-68] and 0 [0-12] s, P=0.01), but not in time delay to first ventilation (91 [31-190], 65 [17-121] and 24 [9-64] s, P=0.08). Data are median [25-75% interquartile]. Regular feedback learning from real-time video recording may improve the quality of major CPR variables.