Circulation assessment by automated external defibrillators during cardiopulmonary resuscitation

Machine learning model to predict evolution of pulseless electrical activity during in-hospital cardiac arrest

Background

During pulseless electrical activity (PEA) the cardiac mechanical and electrical functions are dissociated, a phenomenon occurring in 25-42% of in-hospital cardiac arrest (IHCA) cases. Accurate evaluation of the likelihood of a PEA patient transitioning to return of spontaneous circulation (ROSC) may be vital for the successful resuscitation.

The aim

We sought to develop a model to automatically discriminate between PEA rhythms with favorable and unfavorable evolution to ROSC.

Methods

A dataset of 190 patients, 120 with ROSC, were acquired with defibrillators from different vendors in three hospitals. The ECG and the transthoracic impedance (TTI) signal were processed to compute 16 waveform features. Logistic regression models where designed integrating both automated features and characteristics annotated in the QRS to identify PEAs with better prognosis leading to ROSC. Cross validation techniques were applied, both patient-specific and stratified, to evaluate the performance of the algorithm.

Results

The best model consisted in a three feature algorithm that exhibited median (interquartile range) Area Under the Curve/Balanced accuracy/Sensitivity/Specificity of 80.3(9.9)/75.6(8.0)/ 77.4(15.2)/72.3(16.4) %, respectively.

Conclusions

Information hidden in the waveforms of the ECG and TTI signals, along with QRS complex features, can predict the progression of PEA. Automated methods as the one proposed in this study, could contribute to assist in the targeted treatment of PEA in IHCA.

In-hospital cardiac arrest rhythm analysis by anesthesiologists: a diagnostic performance study

PURPOSE

In-hospital cardiac arrest is associated with high morbidity and mortality, with an overall survival rate at one year of approximately 13%. The first cardiac rhythm is often analyzed by anesthesiologist-intensivists. We aimed to determine the diagnostic performance of anesthesiologist-intensivists when distinguishing between shockable and nonshockable rhythms.

METHODS

We conducted a simulation-based, multicentre, prospective, observational study between May 2019 and March 2020. The responses of the participants were used to calculate individual sensitivity (defined as the proportion of decisions to shock for shockable rhythms) and individual specificity (defined as the proportion of decisions not to shock for nonshockable rhythms). The main outcome measure was the overall diagnostic performance, defined as the overall sensitivity and specificity. Secondary outcome measures were the sensitivity and specificity of participants' decisions for each type of cardiac arrest rhythm and their decision-making times.

RESULTS

Among the 267 physicians contacted, 179 (67%) completed the test. The median [interquartile range (IQR)] overall sensitivity was 88 [79-95]% and the median overall specificity was 86 [77-92]%. Among shockable rhythms, the median [IQR] sensitivity was 100 [100-100]% for ventricular tachycardia (VT), 100 [100-100]% for coarse ventricular fibrillation (VF), and 60 [20-100]% for fine VF. The median [IQR] specificities for nonshockable rhythms were 93 [86-100]% for asystole and 83 [72-86]% for pulseless electrical activity. The median decision times ranged from 2.0 to 3.5 sec.

CONCLUSION

Anesthesiologist-intensivists were quickly and effectively able to analyze rhythms in this simulation-based study. Participants' sensitivity in deciding to deliver shocks for VT and coarse VF was excellent, while specificity of their decisions for pulseless electrical activity was insufficient.

Contribution of chest compressions to end-tidal carbon dioxide levels generated during out-of-hospital cardiopulmonary resuscitation

AIM

Characterise how changes in chest compression depth and rate affect variations in end-tidal CO₂ (ETCO₂) during manual cardiopulmonary resuscitation (CPR) in out-of-hospital cardiac arrest (OHCA).

METHODS

Retrospective analysis of adult OHCA monitor-defibrillator recordings having concurrent capnogram, compression depth, transthoracic impedance and ECG, and with atleast 1,000 compressions. Within each patient, during no spontaneous circulation, nearby segments with changes in chest compression depth and rate were identified. Average ETCO₂ within each segment was standardised to compensate for ventilation rate variability. Contributions of relative variations in depth and rate to relative variations in standardised ETCO₂ were characterised using linear and non-linear models. Normalisation between paired segments removed intra and inter-patient variation and made coefficients of the model independent of the scale of measurement and therefore directly comparable.

RESULTS

A total of 394 pairs of segments from 221 patients were analysed (33% female, median (IQR) age 66 (55-74) years). Chest compression depth and rate were 50.4 (43.2-57.0)mm and 111.1 (106.5-116.1)compressions per minute. ETCO₂ before and after standardization was 32.1 (23.0-41.4)mmHg and 28.5 (19.4-38.7)mmHg. Linear model coefficient of determination was 0.89. Variation in compression depth mainly explained ETCO₂ variation (coefficient 0.95, 95% confidence interval (CI): 0.93-0.98) while changes in compression rate did not (coefficient 0.04, 95% CI: 0.01-0.07). Non-linear trend analysis confirmed the results.

CONCLUSION

This study quantified the relative importance of chest compression characteristics in terms of their impact on CO₂ production during CPR. With ventilation rate standardised, variation in chest compression depth explained variations in ETCO₂ better than variation in chest compression rate.

A Machine Learning Model for the Prognosis of Pulseless Electrical Activity during Out-of-Hospital Cardiac Arrest

Pulseless electrical activity (PEA) is characterized by the disassociation of the mechanical and electrical activity of the heart and appears as the initial rhythm in 20–30% of out-of-hospital cardiac arrest (OHCA) cases. Predicting whether a patient in PEA will convert to return of spontaneous circulation (ROSC) is important because different therapeutic strategies are needed depending on the type of PEA. The aim of this study was to develop a machine learning model to differentiate PEA with unfavorable (unPEA) and favorable (faPEA) evolution to ROSC. An OHCA dataset of 1921 5s PEA signal segments from defibrillator files was used, 703 faPEA segments from 107 patients with ROSC and 1218 unPEA segments from 153 patients with no ROSC. The solution consisted of a signal-processing stage of the ECG and the thoracic impedance (TI) and the extraction of the TI circulation component (ICC), which is associated with ventricular wall movement. Then, a set of 17 features was obtained from the ECG and ICC signals, and a random forest classifier was used to differentiate faPEA from unPEA. All models were trained and tested using patientwise and stratified 10-fold cross-validation partitions. The best model showed a median (interquartile range) area under the curve (AUC) of 85.7(9.8)% and a balance accuracy of 78.8(9.8)%, improving the previously available solutions at more than four points in the AUC and three points in balanced accuracy. It was demonstrated that the evolution of PEA can be predicted using the ECG and TI signals, opening the possibility of targeted PEA treatment in OHCA.

Multimodal Algorithms for the Classification of Circulation States During Out-of-Hospital Cardiac Arrest

GOAL

Identifying the circulation state during out-of-hospital cardiac arrest (OHCA) is essential to determine what life-saving therapies to apply. Currently algorithms discriminate circulation (pulsed rhythms, PR) from no circulation (pulseless electrical activity, PEA), but PEA can be classified into true (TPEA) and pseudo (PPEA) depending on cardiac contractility. This study introduces multi-class algorithms to automatically determine circulation states during OHCA using the signals available in defibrillators.

METHODS

A cohort of 60 OHCA cases were used to extract a dataset of 2506 5-s segments, labeled as PR (1463), PPEA (364) and TPEA (679) using the invasive blood pressure, experimentally recorded through a radial/femoral cannulation. A multimodal algorithm using features obtained from the electrocardiogram, the thoracic impedance and the capnogram was designed. A random forest model was trained to discriminate three (TPEA/PPEA/PR) and two (PEA/PR) circulation states. The models were evaluated using repeated patient-wise 5-fold cross-validation, with the unweighted mean of sensitivities (UMS) and F ₁-score as performance metrics.

RESULTS

The best model for 3-class had a median (interquartile range, IQR) UMS and F ₁ of 69.0% (68.0-70.1) and 61.7% (61.0-62.5), respectively. The best two class classifier had median (IQR) UMS and F ₁ of 83.9% (82.9-84.5) and 76.2% (75.0-76.9), outperforming all previous proposals in over 3-points in UMS.

CONCLUSIONS

The first multiclass OHCA circulation state classifier was demonstrated. The method improved previous algorithms for binary pulse/no-pulse decisions.

SIGNIFICANCE

Automatic multiclass circulation state classification during OHCA could contribute to improve cardiac arrest therapy and improve survival rates.

Assessment of the evolution of end-tidal carbon dioxide within chest compression pauses to detect restoration of spontaneous circulation

Background

Measurement of end-tidal CO₂ (ETCO₂) can help to monitor circulation during cardiopulmonary resuscitation (CPR). However, early detection of restoration of spontaneous circulation (ROSC) during CPR using waveform capnography remains a challenge. The aim of the study was to investigate if the assessment of ETCO₂ variation during chest compression pauses could allow for ROSC detection. We hypothesized that a decay in ETCO₂ during a compression pause indicates no ROSC while a constant or increasing ETCO2 indicates ROSC.

Methods

We conducted a retrospective analysis of adult out-of-hospital cardiac arrest (OHCA) episodes treated by the advanced life support (ALS). Continuous chest compressions and ventilations were provided manually. Segments of capnography signal during pauses in chest compressions were selected, including at least three ventilations and with durations less than 20 s. Segments were classified as ROSC or non-ROSC according to case chart annotation and examination of the ECG and transthoracic impedance signals. The percentage variation of ETCO₂ between consecutive ventilations was computed and its average value, ΔET_avg, was used as a single feature to discriminate between ROSC and non-ROSC segments.

Results

A total of 384 segments (130 ROSC, 254 non-ROSC) from 205 OHCA patients (30.7% female, median age 66) were analyzed. Median (IQR) duration was 16.3 (12.9,18.1) s. ΔET_avg was 0.0 (-0.7, 0.9)% for ROSC segments and -11.0 (-14.1, -8.0)% for non-ROSC segments (p < 0.0001). Best performance for ROSC detection yielded a sensitivity of 95.4% (95% CI: 90.1%, 98.1%) and a specificity of 94.9% (91.4%, 97.1%) for all ventilations in the segment. For the first 2 ventilations, duration was 7.7 (6.0, 10.2) s, and sensitivity and specificity were 90.0% (83.5%, 94.2%) and 89.4 (84.9%, 92.6%), respectively. Our method allowed for ROSC detection during the first compression pause in 95.4% of the patients.

Conclusion

Average percent variation of ETCO₂ during pauses in chest compressions allowed for ROSC discrimination. This metric could help confirm ROSC during compression pauses in ALS settings.

Monitoring chest compression rate in automated external defibrillators using the autocorrelation of the transthoracic impedance

Aim

High-quality chest compressions is challenging for bystanders and first responders to out-of-hospital cardiac arrest (OHCA). Long compression pauses and compression rates higher than recommended are common and detrimental to survival. Our aim was to design a simple and low computational cost algorithm for feedback on compression rate using the transthoracic impedance (TI) acquired by automated external defibrillators (AEDs).

Methods

ECG and TI signals from AED recordings of 242 OHCA patients treated by basic life support (BLS) ambulances were retrospectively analyzed. Beginning and end of chest compression series and each individual compression were annotated. The algorithm computed a biased estimate of the autocorrelation of the TI signal in consecutive non-overlapping 2-s analysis windows to detect the presence of chest compressions and estimate compression rate.

Results

A total of 237 episodes were included in the study, with a median (IQR) duration of 10 (6–16) min. The algorithm performed with a global sensitivity in the detection of chest compressions of 98.7%, positive predictive value of 98.7%, specificity of 97.1%, and negative predictive value of 97.1% (validation subset including 207 episodes). The unsigned error in the estimation of compression rate was 1.7 (1.3–2.9) compressions per minute.

Conclusion

Our algorithm is accurate and robust for real-time guidance on chest compression rate using AEDs. The algorithm is simple and easy to implement with minimal software modifications. Deployment of AEDs with this capability could potentially contribute to enhancing the quality of chest compressions in the first minutes from collapse.

The impact of ventilation rate on end-tidal carbon dioxide level during manual cardiopulmonary resuscitation

AIM

Ventilation rate is a confounding factor for interpretation of end-tidal carbon dioxide (ETCO₂) during cardiopulmonary resuscitation (CPR). The aim of our study was to model the effect of ventilation rate on ETCO₂ during manual CPR in adult out-of-hospital cardiac arrest (OHCA).

METHODS

We conducted a retrospective analysis of OHCA monitor-defibrillator files with concurrent capnogram, compression depth, transthoracic impedance and ECG. We annotated pairs of capnogram segments presenting differences in average ventilation rate and average ETCO₂ value but with other influencing factors (e.g. compression rate and depth) presenting similar values within the pair. ETCO₂ variation as a function of ventilation rate was adjusted through curve fitting using non-linear least squares as a measure of goodness of fit.

RESULTS

A total of 141 pairs of segments from 102 patients were annotated. Each pair provided a single data point for curve fitting. The best goodness of fit yielded a coefficient of determination R² of 0.93. Our model described that ETCO₂ decays exponentially with increasing ventilation rate. The model showed no differences attributable to the airway type (endotracheal tube or supraglottic King-LT-D).

CONCLUSION

Capnogram interpretation during CPR is challenging since many factors influence ETCO₂. For adequate interpretation, we need to know the effect of each factor on ETCO₂. Our model allows quantifying the effect of ventilation rate on ETCO₂ variation. Our findings could contribute to better interpretation of ETCO₂ during CPR.

Study on the Improvement of Electrical Facility System of Automated External Defibrillators by Real-Time Measurement of Thoracic Impedance

Sudden Cardiac Arrest (SCA) is a serious emergency disease that has increased steadily every year. To this end, an Automated External Defibrillator (AED) is placed in a public place so that even non-professional medical personnel can respond to SCA. However, the thoracic impedance of patients changes due to CardioPulmonary Resuscitation (CPR) and artificial respiration during first aid treatment. In addition, changes in chest statues due to gender, age, and accidents cause changes in thoracic impedance in real time. The change in thoracic impedance caused by this has a negative effect on the intended electrical energy of the automatic heart shocker to the emergency patient. To prove this, we divided it into adult and pediatric modes and experimented with the energy error of the AED according to the same impedance change. When the first peak current was up to 56.4 (A) and at least 8.4 (A) in the adult mode, the first peak current was up to 32.2 (A) and at least 4.8 (A), respectively, when the impedance changed, the error of the current figure occurred. In this paper, the inverse relationship between thoracic impedance and electric shock energy according to the state of the cardiac arrest patient is demonstrated through the results of the experiment, and the need for an electric facility system that can revise for changes in thoracic impedance of the cardiac arrest patient by reflecting them on electric shock energy in real time is presented.

Modeling the impact of ventilations on the capnogram in out-of-hospital cardiac arrest

Aim

Current resuscitation guidelines recommend waveform capnography as an indirect indicator of perfusion during cardiopulmonary resuscitation (CPR). Chest compressions (CCs) and ventilations during CPR have opposing effects on the exhaled carbon dioxide (CO₂) concentration, which need to be better characterized. The purpose of this study was to model the impact of ventilations in the exhaled CO₂ measured from capnograms collected during out-of-hospital cardiac arrest (OHCA) resuscitation.

Methods

We retrospectively analyzed OHCA monitor-defibrillator files with concurrent capnogram, compression depth, transthoracic impedance and ECG signals. Segments with CC pauses, two or more ventilations, and with no pulse-generating rhythm were selected. Thus, only ventilations should have caused the decrease in CO₂ concentration. The variation in the exhaled CO₂ concentration with each ventilation was modeled with an exponential decay function using non-linear-least-squares curve fitting.

Results

Out of the original 1002 OHCA dataset (one per patient), 377 episodes had the required signals, and 196 segments from 96 patients met the inclusion criteria. Airway type was endotracheal tube in 64.8% of the segments, supraglottic King LT-D^™ in 30.1%, and unknown in 5.1%. Median (IQR) decay factor of the exhaled CO₂ concentration was 10.0% (7.8 − 12.9) with R² = 0.98(0.95 − 0.99). Differences in decay factor with airway type were not statistically significant (p = 0.17). From these results, we propose a model for estimating the contribution of CCs to the end-tidal CO₂ level between consecutive ventilations and for estimating the end-tidal CO₂ variation as a function of ventilation rate.

Conclusion

We have modeled the decrease in exhaled CO₂ concentration with ventilations during chest compression pauses in CPR. This finding allowed us to hypothesize a mathematical model for explaining the effect of chest compressions on ETCO₂ compensating for the influence of ventilation rate during CPR. However, further work is required to confirm the validity of this model during ongoing chest compressions.

Convolutional Recurrent Neural Networks to Characterize the Circulation Component in the Thoracic Impedance during Out-of-Hospital Cardiac Arrest

Pulse detection during out-of-hospital cardiac arrest remains challenging for both novel and expert rescuers because current methods are inaccurate and time-consuming. There is still a need to develop automatic methods for pulse detection, where the most challenging scenario is the discrimination between pulsed rhythms (PR, pulse) and pulseless electrical activity (PEA, no pulse). Thoracic impedance (TI) acquired through defibrillation pads has been proven useful for detecting pulse as it shows small fluctuations with every heart beat. In this study we analyse the use of deep learning techniques to detect pulse using only the TI signal. The proposed neural network, composed by convolutional and recurrent layers, outperformed state of the art methods, and achieved a balanced accuracy of 90% for segments as short as 3 s.

Capnography: A support tool for the detection of return of spontaneous circulation in out-of-hospital cardiac arrest

BACKGROUND

Automated detection of return of spontaneous circulation (ROSC) is still an unsolved problem during cardiac arrest. Current guidelines recommend the use of capnography, but most automatic methods are based on the analysis of the ECG and thoracic impedance (TI) signals. This study analysed the added value of EtCO₂ for discriminating pulsed (PR) and pulseless (PEA) rhythms and its potential to detect ROSC.

MATERIALS AND METHODS

A total of 426 out-of-hospital cardiac arrest cases, 117 with ROSC and 309 without ROSC, were analysed. First, EtCO₂ values were compared for ROSC and no ROSC cases. Second, 5098 artefact free 3-s long segments were automatically extracted and labelled as PR (3639) or PEA (1459) using the instant of ROSC annotated by the clinician on scene as gold standard. Machine learning classifiers were designed using features obtained from the ECG, TI and the EtCO₂ value. Third, the cases were retrospectively analysed using the classifier to discriminate cases with and without ROSC.

RESULTS

EtCO₂ values increased significantly from 41 mmHg 3-min before ROSC to 57 mmHg 1-min after ROSC, and EtCO₂ was significantly larger for PR than for PEA, 46 mmHg/20 mmHg (p < 0.05). Adding EtCO₂ to the machine learning models increased their area under the curve (AUC) by over 2 percentage points. The combination of ECG, TI and EtCO₂ had an AUC for the detection of pulse of 0.92. Finally, the retrospective analysis showed a sensitivity and specificity of 96.6% and 94.5% for the detection of ROSC and no-ROSC cases, respectively.

CONCLUSION

Adding EtCO₂ improves the performance of automatic algorithms for pulse detection based on ECG and TI. These algorithms can be used to identify pulse on site, and to retrospectively identify cases with ROSC.