Untersuchung eines visuellen Feedbacksystems zur Verbesserung der Herzdruckmassage professioneller Helfer am Übungsmodell

Sudden cardiac arrest is one of the main causes of death in Europe, with about 700.000 cases each year. In spite of extensive efforts to improve the outcome of cardiac arrest, survival rates remain at a poor level. There is increasing evidence thatin the care of cardiac arrest , the effectiveness of external chest compressions is of crucial importance. However, the growing awareness of the significance of cardiac massage stands in sharp contrast to the often deficient quality of chest compressions as performed by laymen as well as healthcare professionals. The present study examined the influence of a new visual feedback system for external chest compressions on the quality of cardiopulmonal resuscitation applied by healthcare professionals. Primary endpoints were compression depth and frequency, secondary endpoints the percentage of appropriate decompressions and correct compressions and, finally, the sufficiency of each resuscitation effort. In addition, the acceptance of the feedback device by the participants of the study was evaluated. The tested feedback device (prototype by Laerdal, Stavanger, Norway) measures 152mm×64mm×28mm and is designed to be placed between the patient’s chest and rescuer´s palm. A 26mm×26mm colour display provides feedback information. The study was performed in a randomized crossover design. The experimental setup consisted of a training mannequin (Resusci® Anne, Laerdal, Stavanger, Norway) connected to a laptop. For data acquisition, the appropriate software (Laerdal PC Skill Reporting Systems Version 1.3.0) was used. At a two-sided level of significance of 5% and a power of 80%, a calculation based on former studies resulted in a sample size of 96 subjects. This calculation included a dropout rate of 10%. After a random assignment to one of two groups each subject passed twice through the testing assembly. For each test a standardized sequence including two minutes of continuous chest compressions was implemented. Group A was first tested applying continuous chest compressions using the device and then without it. Group B underwent the same testing, but in reverse order. Between the two tests there was a compulsory minimum break of 45 minutes. The statistical analysis of the crossover data was based on a modified logistic regression model for binary categorized endpoints and on a modified variance model for continuous endpoints. A carryover effect from the first to the second test had to be ruled out to ensure the quality of study design and statistical approach. Out of 94 potential subjects, 93 were finally included in the study. Only one subject was excluded because of impaired physical condition. In the end the sample was composed of 34 A&E practicioners, 14 EMTs and 45 paramedics. When the feedback device was used, all primary endpoints improved significantly. Furthermore, the 95%-confidence interval (95%-CI) could be found in a confined range for both criteria. Correct mean compression depth was achieved by 45.2% (95%-CI: 30.5–64.9mm) of subjects without vs. 73.1% (95%-CI: 40.3–57.4mm) with device (p < 0.001). Correct average compression rate could be observed in 62.4% (95%-CI: 80.1–147.8 min-1) without vs. 94.6% (95%-CI: 87.3–126.6 min-1) with device (p < 0.001). Applying all necessary criteria to each individual resuscitation effort, all those subjects were detected which had met the requirements of 80% accurate compressions and a rate of 90 to 120 compressions per minute. With feedback, 51.6% of the subjects comply with these criteria whereas without feedback, only 18.3% of the subjects performed a sufficient cardiac massage (p