End-tidal carbon dioxide (ETCO2) and ventricular fibrillation amplitude spectral area (AMSA) for shock outcome prediction in out-of-hospital cardiac arrest. Are they two sides of the same coin?

An acoustic method (Spectral Flux) to analyze ECG signals for optimizing timing for defibrillation in a porcine model of ventricular fibrillation

Aim

Spectral Flux (SF), which is based on common algorithms in the audio processing field, was applied to quantitatively analyze ECG signals to optimize the timing of defibrillation. With the aim of proving the performance in optimizing the timing of defibrillation, SF was compared with Amplitude Spectrum Area (AMSA) in a porcine model of ventricular fibrillation (VF) in a retrospective analysis experiment.

Methods

A total of 56 male domestic pigs, weighing 40 ± 5 kg, were induced to undergo VF. Animals were then left untreated for 10 min, and after 6 min of cardiopulmonary resuscitation (CPR) defibrillation was performed. The respective SF and AMSA values were calculated every minute during VF and CPR. Comparisons were made through receiver operating characteristic (ROC) curves, one-way analyses of variance (one-way ANOVA), and scatterplots for the successful initial defibrillation sample (positive samples, Group R) and the failed initial defibrillation sample (negative samples, Group N) to illustrate the performance in optimizing the timing of defibrillation for the AMSA and SF methods.

Result

Values of SF and AMSA gradually decreased during the 10 min VF period and increased in during the 6 min CPR period. The scatterplots showed that both metrics had the ability to distinguish positive and negative samples (p < .001). Meanwhile, ROC curves showed that SF (area under the curve, AUC = 0.798, p < .001) had the same ability as AMSA (AUC = 0.737, p < .001) to predict the successful defibrillation (Z = 1.35, p = 0.177). Moreover, when comparing the values for AMSA and SF between the successful initial defibrillation samples (Group R) and the failed initial defibrillation samples (Group N), the results showed that the values of both AMSA and SF in Group R were significantly higher than those in Group N (p < .001).

Conclusion

In the present study, SF method had the same ability as AMSA to predict successful defibrillation with significantly higher values in cases of successful defibrillation than the instances in which defibrillation failed. Additionally, SF method might be more stable than AMSA for filtering out the higher frequency interference signals due to the narrower frequency range and had higher specificity and predictive accuracy than AMSA. So SF method had high clinical potential to optimize the timing of defibrillation. Nevertheless, further animal and clinical studies are still needed to confirm the effectiveness and practicality of SF as a predictive module for defibrillators in clinical practice.

Amplitude spectral area of ventricular fibrillation can discriminate survival of patients with out-of-hospital cardiac arrest

Background

Evidence of the association between AMplitude Spectral Area (AMSA) of ventricular fibrillation and outcome after out-of-hospital cardiac arrest (OHCA) is limited to short-term follow-up. In this study, we assess whether AMSA can stratify the risk of death or poor neurological outcome at 30 days and 1 year after OHCA in patients with an initial shockable rhythm or with an initial non-shockable rhythm converted to a shockable one.

Methods

This is a multicentre retrospective study of prospectively collected data in two European Utstein-based OHCA registries. We included all cases of OHCAs with at least one manual defibrillation. AMSA values were calculated after data extraction from the monitors/defibrillators used in the field by using a 2-s pre-shock electrocardiogram interval. The first detected AMSA value, the maximum value, the average value, and the minimum value were computed, and their outcome prediction accuracy was compared. Multivariable Cox regression models were run for both 30-day and 1-year deaths or poor neurological outcomes. Neurological cerebral performance category 1-2 was considered a good neurological outcome.

Results

Out of the 578 patients included, 494 (85%) died and 10 (2%) had a poor neurological outcome at 30 days. All the AMSA values considered (first value, maximum, average, and minimum) were significantly higher in survivors with good neurological outcome at 30 days. The average AMSA showed the highest area under the receiver operating characteristic curve (0.778, 95% CI: 0.7-0.8, p < 0.001). After correction for confounders, the highest tertiles of average AMSA (T3 and T2) were significantly associated with a lower risk of death or poor neurological outcome compared with T1 both at 30 days (T2: HR 0.6, 95% CI: 0.4-0.9, p = 0.01; T3: HR 0.6, 95% CI: 0.4-0.9, p = 0.02) and at 1 year (T2: HR 0.6, 95% CI: 0.4-0.9, p = 0.01; T3: HR 0.6, 95% CI: 0.4-0.9, p = 0.01). Among survivors at 30 days, a higher AMSA was associated with a lower risk of mortality or poor neurological outcome at 1 year (T3: HR 0.03, 95% CI: 0-0.3, p = 0.02).

Discussion

Lower AMSA values were significantly and independently associated with the risk of death or poor neurological outcome at 30 days and at 1 year in OHCA patients with either an initial shockable rhythm or a conversion rhythm from non-shockable to shockable. The average AMSA value had the strongest association with prognosis.

Amplitude spectral area of ventricular fibrillation and defibrillation success at low energy in out-of-hospital cardiac arrest

The optimal energy for defibrillation has not yet been identified and very often the maximum energy is delivered. We sought to assess whether amplitude spectral area (AMSA) of ventricular fibrillation (VF) could predict low energy level defibrillation success in out-of-hospital cardiac arrest (OHCA) patients. This is a multicentre international study based on retrospective analysis of prospectively collected data. We included all OHCAs with at least one manual defibrillation. AMSA values were calculated by analyzing the data collected by the monitors/defibrillators used in the field (Corpuls 3 and Lifepak 12/15) and using a 2-s-pre-shock electrocardiogram interval. We run two different analyses dividing the shocks into three tertiles (T1, T2, T3) based on AMSA values. 629 OHCAs were included and 2095 shocks delivered (energy ranging from 100 to 360 J; median 200 J). Both in the "extremes analysis" and in the "by site analysis", the AMSA values of the effective shocks at low energy were significantly higher than those at high energy (p = 0.01). The likelihood of shock success increased significantly from the lowest to the highest tertile. After correction for age, call to shock time, use of mechanical CPR, presence of bystander CPR, sex and energy level, high AMSA value was directly associated with the probability of shock success [T2 vs T1 OR 3.8 (95% CI 2.5-6) p < 0.001; T3 vs T1 OR 12.7 (95% CI 8.2-19.2), p < 0.001]. AMSA values are associated with the probability of low-energy shock success so that they could guide energy optimization in shockable cardiac arrest patients.

aMplitude spectral area of ventricular fibrillation and amiOdarone Study in patients with out-of-hospital cArdIaC arrest. The MOSAIC study

Objective

Antiarrhythmic drugs are recommended for out of hospital cardiac arrest (OHCA) with shock-refractory ventricular fibrillation (VF). Amplitude Spectral Area (AMSA) of VF is a quantitative waveform measure that describes the amplitude-weighted mean frequency of VF, it correlates with intramyocardial adenosine triphosphate (ATP) concentration, it is a predictor of shock efficacy and an emerging indicator to guide defibrillation and resuscitation efforts. How AMSA might be influenced by amiodarone administration is unknown.

Methods

In this international multicentre observational study, all OHCAs receiving at least one shock were included. AMSA values were calculated by retrospectively analysing the pre-shock ECG interval of 2 s. Multivariable models were run and a propensity score based on the probability of receiving amiodarone was created to compare two randomly matched samples.

Results

2,077 shocks were included: 1,407 in the amiodarone group and 670 in the non-amiodarone group. AMSA values were lower in the amiodarone group [8.8 (6-12.7) mV·Hz vs. 9.8 (6-14) mV·Hz, p = 0.035]. In two randomly matched propensity score-based groups of 261 shocks, AMSA was lower in the amiodarone group [8.2 (5.8-13.5) mV·Hz vs. 9.6 (5.6-11.6), p = 0.042]. AMSA was a predictor of shock success in both groups but the predictive power was lower in the amiodarone group [Area Under the Curve (AUC) non-amiodarone group 0.812, 95%CI: 0.78-0.841 vs. AUC amiodarone group 0.706, 95%CI: 0.68-0.73; p < 0.001].

Conclusions

Amiodarone administration was independently associated with the probability of recording lower values of AMSA. In patients who have received amiodarone during cardiac arrest the predictive value of AMSA for shock success is significantly lower, but still statistically significant.

The ability of machine learning algorithms to predict defibrillation success during cardiac arrest: A systematic review

OBJECTIVE

To evaluate the existing knowledge on the effectiveness of machine learning (ML) algorithms inpredicting defibrillation success during in- and out-of-hospital cardiac arrest.

METHODS

MEDLINE, Embase, CINAHL and Scopus were searched from inception to August 30, 2022. Studies were included that utilized ML algorithms for prediction of successful defibrillation, observed as return of spontaneous circulation (ROSC), survival to hospital or discharge, or neurological status at discharge.Studies were excluded if involving a trauma, an unknown underlying rhythm, an implanted cardiac defibrillator or if focused on the prediction or onset of cardiac arrest. Risk of bias was assessed using the PROBAST tool.

RESULTS

There were 2399 studies identified, of which 107 full text articles were reviewed and 15 observational studies (n = 5680) were included for final analysis. 29 ECG waveform features were fed into 15 different ML combinations. The best performing ML model had an accuracy of 98.6 (98.5 - 98.7)%, with 4 second ECG intervals. An algorithm incorporating end-tidal CO2 reported an accuracy of 83.3% (no CI reported). Meta-analysis was not performed due to heterogeneity in study design, ROSC definitions, and characteristics.

CONCLUSION

Machine learning algorithms, specifically Neural Networks, have been shown to have potential to predict defibrillation success for cardiac arrest with high sensitivity and specificity.Due to heterogeneity, inconsistent reporting, and high risk of bias, it is difficult to conclude which, if any, algorithm is optimal. Further clinical studies with standardized reporting of patient characteristics, outcomes, and appropriate algorithm validation are still required to elucidate this. PROSPERO 2020 CRD42020148912.

Predefibrillation end-tidal CO2 and defibrillation success in out-of-hospital cardiac arrest: an observational cohort study

BACKGROUND

Predefibrillation end-tidal CO₂ (ETCO₂) may predict defibrillation success and could guide defibrillation timing in ventricular fibrillation (VF) cardiac arrest. This relationship has only been studied using advanced airways. Our aim was to evaluate this relationship using both basic (bag-valve-mask (BVM)) and advanced airways (supraglottic airways and endotracheal tubes).

METHODS

Prehospital patient records and defibrillator files were abstracted for patients with out-of-hospital cardiac arrest in Ontario, Canada, with initial VF cardiac rhythms between 1 January 2018, and 31 December 2019. Analyses assessed the relationship between each predefibrillation ETCO₂ reading and defibrillation outcomes at the subsequent 2 min pulse check (ie, VF, asystole, pulseless electrical activity (PEA) or return of spontaneous circulation (ROSC)), accounting for airway types used during resuscitation. Multivariable logistic regression evaluated the association between the first documented predefibrillation ETCO₂ and postshock VF termination or ROSC.

RESULTS

Of 269 cases abstracted, 153 had predefibrillation ETCO₂ measurements and were included in the study. Among these cases, 904 shocks were delivered and 44.4% (n=401) had predefibrillation ETCO₂ measured. The first ETCO₂ reading was more often from BVM (n=134) than advanced airways (n=19). ETCO₂ readings were lower when measured through BVM versus advanced airways (30.5 mm Hg (4.06 kPa) (±14.4 mm Hg (1.92 kPa)) vs 42.1 mm Hg (5.61 kPa) (±22.5 mm Hg (3.00 kPa)), _adjANOVA p<0.01). Of all shocks with ETCO₂ reading (n=401), no difference in preshock ETCO₂ was found for subsequent shocks that resulted in persistent VF (32.2 mm Hg (4.29 kPa) (±15.8 mm Hg (2.11 kPa))), PEA (32.8 mm Hg (4.37 kPa) (±17.1 mm Hg (2.30 kPa))), asystole (32.4 mm Hg (4.32 kPa) (±20.6 mm Hg (2.75 kPa))) or ROSC (32.5 mm Hg (4.33 kPa) (±15.3 mm Hg (2.04 kPa))), analysis of variance p=0.99. In the multivariate analysis using the initial predefibrillation ETCO₂, there was no association with VF termination on the subsequent shock (adjusted OR (_adjOR) 0.99, 95% CI 0.97 to 1.02, p=0.57) or ROSC (_adjOR 1.00, 95% CI 0.97 to 1.03, p=0.94) when evaluated as a continuous or categorical variable.

CONCLUSION

Predefibrillation ETCO₂ measurement is not associated with VF termination or ROSC when basic and advanced airways are included in the analysis. The role of predefibrillation ETCO₂ requires careful consideration of the type of airway used during resuscitation.

End-tidal carbon dioxide (ETCO2) at intubation and its increase after 10 minutes resuscitation predicts survival with good neurological outcome in out-of-hospital cardiac arrest patients

AIM

To evaluate whether end-tidal carbon dioxide (ETCO2) value at intubation and its early increase (10 min) after intubation predict both the survival to hospital admission and the survival at hospital discharge, including good neurological outcome (CPC 1-2), in patients with out-of-hospital cardiac arrest (OHCA).

METHODS

All consecutive OHCA patients of any etiology between 2015 and 2018 in Pavia Province (Italy) and Ticino Region (Switzerland) were considered. Patients died before ambulance arrival, with a "do-not-resuscitate" order, without ETCO2 value or with incomplete data were excluded.

RESULTS

The study population consisted of 668 patients. An ETCO2 value at intubation > 20 mmHg and its increase 10 min after intubation were independent predictors (after correction for known predictors of OHCA outcome) of survival to hospital admission and survival at hospital discharge. Relative to hospital discharge with good neurological outcome, ETCO2 at intubation and its 10-min change were confirmed predictors both individually and in a bivariable analysis (OR 1.83, 95 %CI 1.02-3.3; p = 0.04 and OR 3.9, 95 %CI 1.97-7.74; p < 0.001, respectively). This was confirmed also when accounting for gender, age, etiology and location. After further adjustment for bystander and CPR status, presenting rhythm and EMS arrival time, the ETCO2 change remained an independent predictor.

CONCLUSIONS

ETCO2 value > 20 mmHg at intubation and its increase during resuscitation improve the prediction of survival at hospital discharge with good neurological outcome of OHCA patients. ETCO2 increase during resuscitation is a more powerful predictor than ETCO2 at intubation. A larger prospective study to confirm this finding appears warranted.

Prehospital Cardiac Arrest Airway Management: An NAEMSP Position Statement and Resource Document

Airway management is a critical component of out-of-hospital cardiac arrest (OHCA) resuscitation. Multiple cardiac arrest airway management techniques are available to EMS clinicians including bag-valve-mask (BVM) ventilation, supraglottic airways (SGAs), and endotracheal intubation (ETI). Important goals include achieving optimal oxygenation and ventilation while minimizing negative effects on physiology and interference with other resuscitation interventions. NAEMSP recommends:Based on the skill of the clinician and available resources, BVM, SGA, or ETI may be considered as airway management strategies in OHCA.Airway management should not interfere with other key resuscitation interventions such as high-quality chest compressions, rapid defibrillation, and treatment of reversible causes of the cardiac arrest.EMS clinicians should take measures to avoid hyperventilation during cardiac arrest resuscitation.Where available for clinician use, capnography should be used to guide ventilation and chest compressions, confirm and monitor advanced airway placement, identify return of spontaneous circulation (ROSC), and assist in the decision to terminate resuscitation.

Optimizing defibrillation during cardiac arrest

PURPOSE OF REVIEW

Current cardiac arrest guidelines are based on a fixed, time-based defibrillation strategy. Rhythm analysis and shock delivery (if indicated) are repeated every 2 min requiring cyclical interruptions of chest compressions. This approach has several downsides, such as the need to temporarily stop cardiopulmonary resuscitation (CPR) for a variable amount of time, thus reducing myocardial perfusion and decreasing the chance of successful defibrillation. A tailored defibrillation strategy should identify treatment priority for each patient, that is chest compressions (CCS) or defibrillation, minimize CCs interruptions, speed up the delivery of early effective defibrillation and reduce the number of ineffective shocks.

RECENT FINDINGS

Real-time ECG analysis (using adaptive filters, new algorithms robust to chest compressions artifacts and shock-advisory algorithms) is an effective strategy to correctly identify heart rhythm during CPR and reduce the hands-off time preceding a shock. Similarly, ventricular fibrillation waveform analysis, that is amplitude spectrum area (AMSA) represents a well established approach to reserve defibrillation in patients with high chance of shock success and postpone it when ventricular fibrillation termination is unlikely. Both approaches demonstrated valuable results in improving cardiac arrest outcomes in experimental and observational study.

SUMMARY

Real-time ECG analysis and AMSA have the potential to predict ventricular fibrillation termination, return of spontaneous circulation and even survival, with discretely high confidence. Prospective studies are now necessary to validate these new approaches in the clinical scenario.