Aims: Cardiopulmonary resuscitation quality (CPRQ) parameters can be derived from electric signals obtained during resuscitation, such as chest compression depth (CD) and thoracic impedance (TI). A patient's response to CPR can be estimated by analysing the cardiac rhythm transitions during cardiac arrest by assessing the electrocardiogram (ECG). We propose to model a probabilistic relationship between CPRQ parameters and the physiological response as judged by ECG-features, to guide therapy in a clinical context.
Methods: A total of 821 compression sequences were extracted from 394 out-of-hospital resuscitation episodes. Sequences were categorized as positive if the post-sequence cardiac rhythm had better prognosis than the pre-sequence rhythm, otherwise as negative. A total of 467 positive and 354 negative sequences were considered. For each sequence, six CPRQ parameters (mean and standard deviation) related to depth and rate were calculated from 3 second windows of CD and TI. Three alternative classification approaches were employed for the binary (positive/negative) decision making based on the QCPR parameters: quadratic discriminant analysis (QDA), logistic regression (LR), and artificial neural networks (ANN). The positive class discriminant function defined the probability of effective compressions (Pec).
Results: The discriminative accuracies for the three types of classifiers were 0.59 (QDA), 0.63 (ANN) and 0.63 (LR). Three compression depth (2-4,4-6,6-8 cm) and rate intervals (70-100,100-130,130-160 min-1) were analysed, corresponding to "less than recommended", "recommended", "more than recommended" CPRQ values, respectively. For the LR model the median (interquartile range) Pec were 0.3 (0.1-0.6), 0.7 (0.6-0.8), 0.2 (0.1-0.3) for the depth, and 0.5 (0.3-0.6), 0.6 (0.4-0.7), 0.2 (0.0-0.4) for the rate intervals, respectively.
Conclusion: We have proposed a novel method to relate the quality of chest compressions to the physiologic response to CPR. The highest probability estimates of effective chest compressions corresponded to the depth (5-6 cm) and rate (100-120 min-1) currently recommended in the CPR guidelines.