Real-Time Feedback Improves Quality of Chest Compression in Pediatric Cardiopulmonary Resuscitation Training Course

Background: High-quality cardiopulmonary resuscitation (CPR) is vital for pediatric survival, yet maintaining precision remains difficult even for specialists. This study evaluated the impact of real-time audiovisual feedback on pediatric CPR and identified independent factors influencing performance. Methods: This prospective study recruited 110 pediatric intensive care unit providers. Performance data were collected using an infant manikin and a defibrillator system with dual anterior–posterior sensors to eliminate mattress deflection artifacts. “On-target” quality was defined as the simultaneous achievement of American Heart Association-recommended rate (100–120/min) and depth (≥4.0 cm or one-third anterior-posterior diameter). Participants performed two 1-min sessions of compression-only CPR: A baseline session and a subsequent session with real-time visual feedback. Data were analyzed using nonparametric tests and multivariable logistic regression. Results: At baseline, performance was suboptimal (0% on-target), with female participants compressing too fast and certified providers performing significantly shallower compressions. With real-time feedback, the median rate was corrected to 110/min, and the median depth increased from 2.2 to 4.0 cm ( P < 0.001), raising the on-target rate to 49.2%. Feedback usage was the strongest independent predictor of rate adherence (adjusted odds ratios = 27.23). Notably, feedback enabled certified providers to reverse baseline trends and achieve greater depths than noncertified counterparts. Furthermore, participants with at least six real-world experiences achieved the highest on-target scores (median = 0.60). Conclusions: Real-time feedback effectively standardizes CPR quality by eliminating gender and training recency gaps. While it ensures a critical performance baseline, extensive clinical experience remains essential for achieving peak resuscitation quality.