End-tidal Carbon Dioxide Trajectory-based Prognostication of Out-of-hospital Cardiac Arrest

Author Affiliation
Chih-Hung Wang, MD, PhD National Taiwan University, College of Medicine, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Tsung-Chien Lu, MD, PhD National Taiwan University, College of Medicine, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Joyce Tay, MD National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Cheng-Yi Wu, MD National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Meng-Che Wu, MD National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Chun-Yen Huang, MD, MSc Far Eastern Memorial Hospital, Department of Emergency Medicine, New Taipei City, Taiwan
Chu-Lin Tsai, MD, ScD National Taiwan University, College of Medicine, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Chien-Hua Huang, MD, PhD National Taiwan University, College of Medicine, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan
Matthew Huei-Ming Ma, MD, PhD National Taiwan University, College of Medicine, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital Yunlin Branch, Department of Emergency Medicine, Yunlin County, Taiwan
Wen-Jone Chen, MD, PhD National Taiwan University, College of Medicine, Department of Emergency Medicine, Taipei, Taiwan;
National Taiwan University Hospital, Department of Emergency Medicine, Taipei, Taiwan

Introduction
Materials and methods
Results
Discussion
Limitations
Conclusion

ABSTRACT

Background

During cardiopulmonary resuscitation (CPR), end-tidal carbon dioxide (EtCO2) is primarily determined by pulmonary blood flow, thereby reflecting the blood flow generated by CPR. We aimed to develop an EtCO2 trajectory-based prediction model for prognostication at specific time points during CPR in patients with out-of-hospital cardiac arrest (OHCA).

Methods

We screened patients receiving CPR between 2015–2021 from a prospectively collected database of a tertiary-care medical center. The primary outcome was survival to hospital discharge. We used group-based trajectory modeling to identify the EtCO2 trajectories. Multivariable logistic regression analysis was used for model development and internally validated using bootstrapping. We assessed performance of the model using the area under the receiver operating characteristic curve (AUC).

Results

The primary analysis included 542 patients with a median age of 68.0 years. Three distinct EtCO2 trajectories were identified in patients resuscitated for 20 minutes (min): low (average EtCO2 10.0 millimeters of mercury [mm Hg]; intermediate (average EtCO2 26.5 mm Hg); and high (average EtCO2: 51.5 mm Hg). Twenty-min EtCO2 trajectory was fitted as an ordinal variable (low, intermediate, and high) and positively associated with survival (odds ratio 2.25, 95% confidence interval [CI] 1.07–4.74). When the 20-min EtCO2 trajectory was combined with other variables, including arrest location and arrest rhythms, the AUC of the 20-min prediction model for survival was 0.89 (95% CI 0.86–0.92). All predictors in the 20-min model remained statistically significant after bootstrapping.

Conclusion

Time-specific EtCO2 trajectory was a significant predictor of OHCA outcomes, which could be combined with other baseline variables for intra-arrest prognostication. For this purpose, the 20-min survival model achieved excellent discriminative performance in predicting survival to hospital discharge.

INTRODUCTION

The annual incidence of out-of-hospital cardiac arrest (OHCA) is estimated to be 28–44 cases per 100,000 population worldwide.1 The estimated proportion of survival to discharge in OHCA was 7.6% in Europe, 6.8% in North America, 3.0% in Asia, and 9.7% in Australia.1 High-quality cardiopulmonary resuscitation (CPR) is critical in improving OHCA outcomes.2,3 Capnography is recommended to monitor CPR quality in real time and adjust chest compression quality accordingly.2,3 During CPR, end-tidal carbon dioxide (EtCO2) is primarily determined by pulmonary blood flow, thereby reflecting the blood flow generated by CPR.4,5

The 2020 International Liaison Committee on Resuscitation (ILCOR) consensus6,7 recommended that EtCO2 ≥20 millimeters of mercury (mm Hg) measured after 20 minutes (min) of CPR may predict survival to discharge. Nonetheless, this weak recommendation was supported by only moderate-quality evidence. A 2018 ILCOR systematic review noticed that the measurement time points of EtCO2 were very heterogeneous across different studies.8 Accordingly, ILCOR6,7 suggested that instead of single EtCO2 values, the EtCO2 trend should be further explored in future studies for its prognostic performance.

The previous study noted that EtCO2 trajectory during CPR was associated with OHCA outcomes.9 However, the predictive ability of EtCO2 trajectory at a specific timing was not explored in the previous study.9 Whether EtCO2 can be combined with other metrics for intra-arrest prognostication was considered a critical knowledge gap by the 2020 American Heart Association (AHA) guidelines.2 In our recent study,10 we incorporated the minimum EtCO2 value into the return of spontaneous circulation after cardiac arrest (RACA) score and improved the performance of RACA score in predicting ROSC, suggesting that EtCO2 could potentially help intra-arrest prognostication.

In the current study, we further developed models that could predict survival at hospital discharge. Instead of a single EtCO2 value,10 we attempted to combine EtCO2 trajectory and other predictors in deriving prediction models. Moreover, these models were developed using time-specific windows to prognosticate patient outcomes during resuscitation, including 10- and 20 min6,7 after initiation of CPR.

MATERIALS AND METHODS

This observational study was a secondary analysis of a prospectively collected OHCA database registered in the emergency department (ED) of National Taiwan University Hospital (NTUH). The institutional review board approved this study (reference number: 201906082RINB) and waived the requirement for informed consent. The study was performed according to the recommendations from Worster et al11 regarding health record review studies in emergency medicine research with all elements followed. The results are reported according to the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) statement.12

Study Setting

The NTUH is a tertiary-care medical center with 2,600 beds, including 220 beds in intensive care units. Approximately 100,000 patients visit NTUH ED annually. Patients with OHCA are transported directly to the resuscitation bay of the critical care area in the ED for CPR, which is delivered according to resuscitation guidelines.2,3 Also, since 2013 ED staff have been trained with the A-C-L-S (airway-circulation-leadership-support) teamwork model9,13,14 to streamline the resuscitation process via both strengthened technical and non-technical skills.15,16 Any intervention, such as tracheal intubation performed during CPR, are timestamped by nurses with a specially designed mobile application. The EtCO2 is recorded every two min right before pulse check. The EtCO2 is monitored with devices attached to the advanced airways, including supraglottic airways and endotracheal tubes. For patients with OHCA who never achieve return of spontaneous circulation (ROSC), CPR is usually performed for at least 30 min in the ED, except for those with a documented do-not-resuscitate (DNR) order.

Study Population

Patients with OHCA sent to the NTUH ED between January 1, 2015–December 31, 2021 were screened. The inclusion criteria for the study were as follows: 1) non-traumatic arrest; 2) absence of ROSC before ED arrival; (3) absence of documented DNR order before CPR; 4) age ≥18 years; and 5) insertion of advanced airways during CPR. Based on the CPR duration, the included patients would be further selected for primary and secondary analyses. If the included patients received CPR ≥20 min and had EtCO2 measurements ≥3 times within 20 min of CPR, they would be selected into the 20-min group for the primary analysis. Similarly, if the included patients received CPR ≥10 min and had EtCO2 measurements ≥3 times within 10 min of CPR, they would be selected into the 10-min group for secondary analysis.

Data Collection, Variable Definitions, and Outcome Measures

In the NTUH database, OHCA events were recorded based on the Utstein template.17 Data requested for analysis included age, gender, variables derived from the Utstein template, advanced airway insertion timing, EtCO2 values with measurement timing, and outcomes. For ED resuscitation, the time point of the initial chest compression delivered in the ED was set as time zero for reference. Time to advanced airway use was defined as the interval between time zero and time for completing advanced airway insertion. If advanced airway devices were inserted before ED arrival, the time to advanced airway was recorded as zero. Duration of CPR in the ED referred to the time interval between time zero and the end of resuscitation, either due to ROSC or death. Time-specific EtCO2 referred to the EtCO2 level measured after the specific time elapsing following time zero.

The primary outcome was survival status at the time of hospital discharge. The secondary outcome was ROSC, defined as a palpable pulse for 20 seconds.18 Data abstraction for the current analysis was performed by trained researchers who were blinded to the study hypothesis.

Statistical Analysis

In the primary analysis, we used the 20-min group to build models for predicting survival (20-min survival model) and ROSC (20-min ROSC model). In the secondary analysis, similar procedures were applied to develop the 10-min survival model and 10-min ROSC model. We first performed group-based trajectory modeling (GBTM) to identify trajectory groups based on the EtCO2 level. The GBTM is an explanatory modeling technique to identify hidden groups of individuals with similar trajectories for a particular variable of interest.19 The GBTM performs better when longitudinal data is measured at least three times.

For descriptive statistics, categorical variables are presented as proportions, and continuous variables are presented as medians with interquartile ranges. We examined categorical variables using the chi-squared test, whereas continuous variables were compared using the Kruskal-Wallis test or Mann-Whitney test, as appropriate. We used multivariable logistic regression analyses to develop the prediction models. All available variables, including basic demographics, peri-CPR events, and EtCO2 trajectory were accounted for in the regression model via a stepwise, variable selection procedure. The EtCO2 trajectory would be tested as ordinal or categorical variables in the model-building process. We used generalized additive models (GAM)20 to identify the appropriate cutoff point(s) for dichotomization. The discriminative performance and calibration of the prediction model were assessed by area under the receiver operating characteristic curve (AUC) and the Hosmer-Lemeshow goodness-of-fit test, respectively. We internally validated the prediction model using the bootstrapping procedure with 1,000 repetitions to examine the robustness of the effect estimate of each variable in the prediction model.

We performed GBTM and bootstrapping using the traj package and bootstrap procedure of Stata software (StataCorp LLC, College Station, TX), respectively. We used the R 4.1.1 software (R Foundation for Statistical Computing, Vienna, Austria) for other analyses. A two-tailed P-value <0.05 was considered statistically significant.

RESULTS

The patient selection procedure resulted in 542 and 532 patients in the 20-min and 10-min groups, respectively (Figure 1). The two groups were not mutually exclusive. Because not all patients in the 20-min group had EtCO2 measurements ≥3 times within 10 mins, the 20-min group patients may not have been necessarily included in the 10-min group. Also, because some of the patients in the 10-min group would achieve ROSC within 20 min of CPR, the 10-min group patients would not necessarily have been included in the 20-min group. Therefore, there was an overlap of 385 patients between the 20-min and 10-min groups who met the selection criteria for both groups.

 

Figure 1.
CPR, cardiopulmonary resuscitation; DNR, do-not-resuscitate; ED, emergency department; NTUH, National Taiwan University Hospital; OHCA, out-of-hospital cardiac arrest; ROSC, return of spontaneous circulation.

In the primary analysis, we identified and named three EtCO2 trajectories as low, intermediate, and high trajectories according to their respective average EtCO2 levels (Figure 2). The characteristics of the 20-min group and comparisons between these EtCO2 trajectories are presented in Table 1. The median CPR duration in the ED was 31.0 minutes, and the median number of EtCO2 measurements was eight. A total of 25 (4.6%) patients survived at hospital discharge. There seems to be an increasing trend of survival from low to high EtCO2 trajectory. The comparisons between patients stratified by survival are shown in Supplemental Table 1. During the model development, the 20-min EtCO2 trajectory was fitted as an ordinal variable by the logistic regression analysis and positively associated with survival (odds ratio [OR] 2.25, 95% confidence interval [CI] 1.07–4.74) and ROSC (OR 2.46, 95% CI 1.78–3.41) (Table 2). In other words, compared with the low EtCO2 trajectory, the intermediate trajectory had 2.25 times higher odds of survival to hospital discharge. Similarly, compared with the intermediate trajectory, the high EtCO2 trajectory also had 2.25 times higher odds of survival. When the 20-min EtCO2 trajectory was combined with other variables, the AUCs of the 20-min survival and ROSC models were 0.89 (95% CI 0.86–0.92) and 0.78 (95% CI 0.74–0.81), respectively.

 

Figure 2.

CPR, cardiopulmonary resuscitation; EtCO

2
, end-tidal carbon dioxide.

Table 1.Characteristics of patients included in the twenty-minute group stratified by end-tidal carbon dioxide trajectory group.

Variables Twenty-min group (n = 542) Twenty-min low EtCO2 trajectory (n = 196) Twenty-min intermediate EtCO2 trajectory (n = 280) Twenty-min high EtCO2 trajectory (n = 66) P value
Basic demographics
 Age, year 68.0 (57.0–80.0) 70.0 (58.0–80.0) 67.0 (56.0–81.0) 66.0 (56.0–76.0) 0.45
 Male, n 354 (65.3) 111 (56.6) 199 (71.1) 44 (66.7) 0.005
Peri-CPR events
 Transported by EMS, n 507 (93.5) 179 (91.3) 263 (93.9) 65 (98.5) 0.11
 Arrest at home, n 296 (54.6) 113 (57.6) 149 (53.2) 34 (51.5) 0.55
 Witness by bystander, n 193 (35.6) 56 (28.5) 112 (40.0) 25 (37.9) 0.03
 Witness by EMS, n 28 (5.2) 11 (5.6) 12 (4.3) 5 (7.6) 0.52
 Witness by bystander or EMS, n 212 (39.1) 61 (31.1) 121 (43.2) 30 (45.4) 0.02
 Bystander CPR, n 269 (49.6) 93 (47.4) 140 (50.0) 36 (54.5) 0.60
 Prehospital defibrillation by EMS, n 117 (21.5) 17 (8.6) 82 (29.2) 18 (27.2) <0.001
 Initial shockable rhythms at ED arrival, n 37 (6.8) 8 (4.1) 25 (8.9) 4 (6.1) 0.43
 Duration of prehospital CPR performed by EMS, min 17.0 (12.0–21.0) 17.0 (10.5–21.0) 17.0 (12.0–21.0) 18.0 (12.0–22.0) 0.30
Procedures during CPR
 SGA use, n 376 (69.4) 134 (68.4) 196 (70.0) 46 (69.7) 0.93
 Time to SGA use, min 0 (0–0) (n = 376) 0 (0–0) (n = 134) 0 (0–0) (n = 196) 0 (0–0) (n = 46) 0.12
 ETT use, n 531 (98.0) 189 (96.4) 277 (98.9) 65 (98.5) 0.12
 Time to ETT use, min 3.0 (2.0–5.0) (n = 531) 3.0 (2.0–6.0) (n = 189) 3.0 (2.0–5.0) (n = 277) 3.0 (1.5–4.0) (n = 65) 0.30
Time-specific EtCO2 levels, mmHg
 0-min EtCO2 29.0 (20.3–36.0) (n = 39) 15.0 (12.5–20.5) (n = 8) 32.0 (25.3–36.0) (n = 27) 32.0 (25.0–50.5) (n = 4) 0.003
 1-min EtCO2 24.5 (15.0–38.5) (n = 56) 14.5 (10.5–19.0) (n = 20) 28.0 (22.0–38.3) (n = 25) 36.3 (24.0–68.3) (n = 11) <0.001
 2-min EtCO2 24.0 (5.8–33.0) (n = 113) 14.0 (9.0–23.3) (n = 37) 27.5 (20.0–33.0) (n = 62) 41.0 (24.0–54.0) (n = 14) <0.001
 3-min EtCO2 22.0 (13.5–36.0) (n = 120) 11.5 (6.0–20.0) (n = 46) 30.0 (21.0–39.5) (n = 60) 36.5 (23.0–43.0) (n = 14) <0.001
 4-min EtCO2 22.0 (12.0–33.0) (n = 231) 11.5 (7.0–18.0) (n = 78) 24.0 (18.0–34.8) (n = 123) 44.0 (30.0–52.0) (n = 30) <0.001
 5-min EtCO2 22.0 (12.0–33.0) (n = 121) 10.0 (3.0–14.8) (n = 43) 27.0 (21.0–35.0) (n = 62) 41.0 (30.5–60.0) (n = 16) <0.001
 6-min EtCO2 21.0 (12.0–31.0) (n = 245) 8.0 (3.0–12.0) (n = 75) 24.0 (18.0–31.0) (n = 141) 47.0 (37.8–60.8) (n = 29) <0.001
 7-min EtCO2 18.5 (10.0–32.0) (n = 142) 9.0 (4.0–13.5) (n = 61) 27.0 (18.0–34.0) (n = 62) 44.0 (30.0–62.3) (n = 19) <0.001
 8-min EtCO2 22.0 (11.0–34.0) (n = 282) 9.0 (3.0–12.0) (n = 94) 27.0 (19.0–35.3) (n = 157) 56.0 (45.0–60.8) (n = 31) <0.001
 9-min EtCO2 20.0 (10.0–34.0) (n = 147) 8.5 (3.0–12.0) (n = 58) 27.0 (19.0–36.0) (n = 70) 58.0 (45.0–72.8) (n = 19) <0.001
 10-min EtCO2 21.0 (12.0–33.0) (n = 296) 9.0 (3.3–13.0) (n = 103) 27.0 (20.0–34.8) (n = 163) 52.5 (48.0–68.0) (n = 30) <0.001
 11-min EtCO2 21.0 (11.0–36.5) (n = 144) 11.0 (5.0–15.0) (n = 58) 28.0 (21.0–36.0) (n = 63) 60.0 (45.0–65.8) (n = 23) <0.001
 12-min EtCO2 21.0 (12.0–31.8) (n = 331) 10.0 (5.0–14.0) (n = 122) 26.0 (21.0–33.0) (n = 176) 58.0 (43.8–71.3) (n = 33) <0.001
 13-min EtCO2 21.0 (12.0–33.8) (n = 123) 9.5 (7.5–13.5) (n = 48) 26.0 (20.8–33.0) (n = 57) 51.5 (44.0–65.0) (n = 18) <0.001
 14-min EtCO2 21.0 (12.0–33.0) (n = 324) 10.0 (5.0–15.0) (n = 117) 26.0 (21.0–34.0) (n = 173) 53.0 (45.0–69.0) (n = 34) <0.001
 15-min EtCO2 21.0 (11.0–32.0) (n = 143) 9.5 (4.0–14.0) (n = 58) 27.0 (21.0–35.0) (n = 65) 50.0 (43.0–58.5) (n = 20) <0.001
 16-min EtCO2 22.0 (12.0–33.0) (n = 329) 9.0 (6.0–14.0) (n = 114) 26.5 (21.0–33.0) (n = 180) 59.0 (47.3–68.8) (n = 35) <0.001
 17-min EtCO2wp 21.0 (12.0–36.0) (n = 139) 9.0 (5.0–13.5) (n = 52) 27.0 (21.0–33.8) (n = 63) 56.0 (45.5–66.5) (n = 24) <0.001
 18-min EtCO2 21.0 (10.8–32.0) (n = 333) 9.0 (3.0–14.0) (n = 125) 26.0 (20.0–33.0) (n = 173) 55.0 (43.0–69.0) (n = 35) <0.001
 19-min EtCO2 21.0 (10.0–34.0) (n = 137) 8.5 (3.0–13.0) (n = 50) 23.0 (20.0–34.0) (n = 68) 50.0 (44.3–62.0) (n = 19) <0.001
 20-min EtCO2 21.0 (11.0–33.3) (n = 329) 9.0 (4.5–14.0) (n = 123) 26.0 (20.0–34.0) (n = 171) 56.0 (50.0–64.5) (n = 35) <0.001
Available measurements of EtCO2 levels, times 8.0 (6.0–9.0) 8.0 (7.0–9.0) 8.0 (6.0–9.0) 8.0 (7.0–9.0) 0.64
EtCO2 summary parameters, mm Hg
 Initial 23.0 (14.0–36.0) 14.0 (7.0–20.5) 29.0 (20.0–41.0) 41.5 (28.0–61.0) <0.001
 Maximum 36.0 (22.0–50.0) 18.0 (12.0–24.0) 41.0 (34.0–49.0) 69.0 (63.0–79.0) <0.001
 Minimum 13.0 (5.0–21.0) 3.5 (2.0–9.0) 16.0 (12.0–22.0) 30.0 (23.0–41.0) <0.001
 Final 21.0 (11.0–35.0) 9.0 (4.0–14.0) 26.0 (20.0–34.5) 56.0 (46.0–65.0) <0.001
 Average 23.0 (14.0–33.0) 10.0 (6.5–14.0) 26.5 (22.0–33.0) 51.5 (37.0–58.0) <0.001
Duration of CPR performed in ED, min 31.0 (30.0–35.0) 31.0 (30.0–34.0) 31.0 (30.0–36.0) 31.0 (30.0–33.0) 0.50
Outcome, n
 ROSC 184 (33.9) 32 (16.3) 118 (42.1) 34 (51.5) <0.001
 Survival to hospital discharge 25 (4.6) 3 (1.5) 16 (5.7) 6 (9.1) 0.02

Data are presented as median (interquartile range) or counts (proportion).
CPR, cardiopulmonary resuscitation; ED, emergency department; EMS, emergency medical services; mm Hg, millimeters of mercury; ETT, endotracheal tube; ROSC, return of spontaneous circulation; SGA, supraglottic airway; min, minute.

Table 2.Multivariable logistic regression analysis for twenty-minute group to build end-tidal carbon dioxide trajectory-based prediction models.

Variables Odds ratio (95% confidence interval) P value
Twenty-min survival model
 Twenty-min EtCO2 trajectory 2.25 (1.07–4.74) 0.03
 Arrest at home 0.28 (0.10–0.77) 0.01
 Prehospital defibrillation by EMS 3.42 (1.34–8.77) 0.01
 Initial shockable rhythms at ED arrival 8.36 (3.13–22.31) <0.001
Twenty-min ROSC model
 Twenty-min EtCO2 trajectory 2.46 (1.78–3.41) <0.001
 Arrest at home 0.54 (0.34–0.85) 0.008
 Witness by bystander or EMS 1.72 (1.13–2.63) 0.01
 Prehospital defibrillation by EMS 2.72 (1.64–4.53) <0.001
 Initial shockable rhythms at ED arrival 4.97 (2.07–11.90) <0.001
 Duration of prehospital CPR performed by EMS 0.96 (0.93–0.99) 0.003

Twenty-min survival model: goodness-of-fit assessment: n = 542, adjusted generalized R
2 = 0.32, estimated area under the receiver operating characteristic curve = 0.89 (95% confidence interval: 0.86–0.92), and Hosmer and Lemeshow goodness-of-fit Chi-Squared test p = 0.64; Twenty-min ROSC model: goodness-of-fit assessment: n = 542, adjusted generalized R
2 = 0.30, estimated area under the receiver operating characteristic curve = 0.78 (95% confidence interval: 0.74–0.81), and Hosmer and Lemeshow goodness-of-fit Chi-Squared test p = 0.19.
CPR, cardiopulmonary resuscitation; ED, emergency department; EMS, emergency medical services; ROSC, return of spontaneous circulation; min, minute.

Similarly, in the secondary analysis we identified three EtCO2 trajectories (Figure 2 and Table 3). The median CPR duration in the ED was 30.0 min, and the median number of EtCO2 measurements was four. A total of 34 (6.4%) patients survived at hospital discharge. Significant survival differences were noted among the three EtCO2 trajectories; nonetheless, the survival of intermediate and high EtCO2 trajectories was similar. The survival-stratified comparisons are shown in Supplemental Table 2. During the model-fitting process, the 10-min EtCO2 trajectory was fitted as a categorical variable. As shown in Table 4, compared with the 10-min low EtCO2 trajectory, the 10-min intermediate or high EtCO2 trajectory was significantly associated with survival (OR 2.53, 95% CI 1.10–5.81). In addition, compared with the 10-min low EtCO2 trajectory, 10-min intermediate (OR 3.36, 95% CI 2.25–5.04) and high (OR 6.59, 95% CI 3.42–12.69) EtCO2 trajectories were significantly associated with ROSC, respectively. When the 10-min EtCO2 trajectory was combined with other variables, the AUC of the 10-min survival and ROSC models were 0.76 (95% CI 0.72–0.79) and 0.75 (95% CI 0.71–0.79), respectively.

Table 3.Characteristics of patients included in the ten-min group stratified by end-tidal carbon dioxide trajectory group.

Variables Ten-min group (n = 532) Ten-min low EtCO2 trajectory (n = 234) Ten-min intermediate EtCO2 trajectory (n = 240) Ten-min high EtCO2 trajectory (n = 58) P value
Basic demographics
 Age, year 71.0 (59.5–82.0) 73.0 (60.0–84.0) 70.0 (60.0–81.0) 70.5 (56.0–79.0) 0.22
 Male, n 346 (65.0) 143 (61.1) 167 (69.6) 36 (62.1) 0.14
Peri-CPR events
 Transported by EMS, n 500 (94.0) 215 (91.9) 227 (94.6) 58 (100) 0.11
 Arrest at home, n 308 (57.9) 144 (61.5) 134 (55.8) 34 (51.7) 0.27
 Witness by bystander, n 192 (36.1) 78 (33.3) 89 (37.1) 25 (43.1) 0.35
 Witness by EMS, n 26 (4.9) 7 (3.0) 16 (6.7) 3 (5.2) 0.18
 Witness by bystander or EMS, n 207 (38.9) 79 (33.8) 101 (42.1) 27 (46.6) 0.08
 Bystander CPR, n 276 (51.9) 115 (49.1) 126 (52.5) 35 (60.3) 0.30
 Prehospital defibrillation by EMS, n 101 (19.0) 24 (10.2) 60 (25.0) 17 (29.3) <0.001
 Initial shockable rhythms at ED arrival, n 30 (5.6) 11 (4.7) 16 (6.7) 3 (5.2) 0.64
 Duration of prehospital CPR performed by EMS, min 17.0 (12.0–21.0) 18.0 (11.0–21.0) 17.0 (12.0–21.0) 18.0 (14.0–22.0) 0.44
Procedures during CPR
 SGA use, n 380 (71.4) 166 (70.9) 172 (71.7) 42 (72.4) 0.97
 Time to SGA use, min 0 (0–0) (n = 380) 0 (0–0) (n = 166) 0 (0–0) (n = 172) 0 (0–0) (n = 42) 0.24
 ETT use, n 508 (95.5) 219 (93.6) 234 (97.5) 55 (94.8) 0.12
 Time to ETT use, min 3.0 (2.0–4.0) (n = 508) 3.0 (2.0–4.0) (n = 219) 3.0 (2.0–4.0) (n = 234) 3.0 (1.3–4.0) (n = 55) 0.48
Time-specific EtCO2 levels, mm Hg
 0-min EtCO2 26.0 (18.0–36.0) (n = 48) 18.0 (14.5–20.5) (n = 16) 31.0 (25.3–38.3) (n = 27) 55.0 (28.5–60.8) (n = 5) <0.001
 1-min EtCO2 24.0 (12.0–38.3) (n = 73) 12.0 (7.0–17.0) (n = 30) 32.5 (22.0–41.0) (n = 34) 56.0 (40.5–69.3) (n = 9) <0.001
 2-min EtCO2 25.5 (17.0–37.5) (n = 148) 17.0 (11.0–23.0) (n = 62) 32.5 (23.0–42.0) (n = 70) 54.5 (45.0–61.5) (n = 16) <0.001
 3-min EtCO2 24.0 (14.0–36.0) (n = 158) 13.5 (9.0–21.0) (n = 70) 34.0 (25.3–43.8) (n = 71) 48.0 (22.8–54.0) (n = 17) <0.001
 4-min EtCO2 23.0 (13.3–35.8) (n = 299) 13.0 (9.0–19.8) (n = 131) 30.0 (22.0–38.0) (n = 132) 51.0 (45.5–62.5) (n = 36) <0.001
 5-min EtCO2 23.0 (12.0–34.0) (n = 153) 12.0 (3.5–17.0) (n = 63) 29.0 (23.0–36.0) (n = 74) 60.0 (42.0–66.5) (n = 16) <0.001
 6-min EtCO2 22.0 (13.0–34.0) (n = 326) 12.0 (7.0–18.0) (n = 142) 28.0 (21.3–38.0) (n = 147) 56.0 (42.8–63.3) (n = 37) <0.001
 7-min EtCO2 23.0 (10.0–36.0) (n = 154) 10.0 (5.5–15.5) (n = 68) 30.0 (24.8–37.0) (n = 69) 55.0 (45.8–64.8) (n = 17) <0.001
 8-min EtCO2 25.0 (13.0–38.0) (n = 343) 12.0 (7.8–17.3) (n = 149) 33.0 (26.0–40.0) (n = 159) 62.0 (56.3–72.5) (n = 35) <0.001
 9-min EtCO2 23.0 (11.0–37.0) (n = 142) 10.0 (4.0–15.8) (n = 63) 30.0 (23.0–37.0) (n = 60) 62.0 (54.5–78.0) (n = 19) <0.001
 10-min EtCO2 23.0 (14.0–39.8) (n = 339) 13.0 (6.0–18.0) (n = 150) 32.0 (24.0–43.0) (n = 154) 68.0 (58.0–79.5) (n = 35) <0.001
Available measurements of EtCO2 levels, times 4.0 (3.0–5.0) 4.0 (3.0–5.0) 4.0 (3.0–5.0) 4.0 (4.0–5.0) 0.41
EtCO2 summary parameters, mm Hg
 Initial 25.0 (15.0–40.0) 15.0 (10.0–22.0) 34.0 (25.0–43.5) 55.5 (45.0–65.0) <0.001
 Maximum 34.0 (22.0–50.0) 20.0 (13.0–26.0) 44.0 (36.0–51.0) 71.5 (63.0–89.0) <0.001
 Minimum 16.0 (9.0–24.5) 8.0 (3.0–12.0) 21.5 (17.0–27.0) 41.5 (33.0–54.0) <0.001
 Final 23.0 (13.0–39.0) 12.0 (6.0–18.0) 33.0 (24.0–42.5) 64.0 (57.0–78.0) <0.001
 Average 25.0 (15.0–36.0) 13.0 (8.0–19.0) 32.0 (26.0–37.0) 58.0 (51.0–64.0) <0.001
Duration of CPR performed in ED, min 30.0 (18.0–32.0) 30.0 (22.0–32.0) 30.0 (17.0–33.0) 20.0 (13.0–31.0) 0.008
Outcome, n
 ROSC 239 (44.9) 64 (27.4) 135 (56.3) 40 (69.0) <0.001
 Survival to hospital discharge 34 (6.4) 8 (3.4) 21 (8.8) 5 (8.6) 0.05

Data are presented as median (interquartile range) or counts (proportion).
CPR, cardiopulmonary resuscitation; ED, emergency department; EMS, emergency medical service; mm HG, millimeters of mercury; ETT, endotracheal tube; ROSC, return of spontaneous circulation; SGA, supraglottic airway.

Table 4.Multivariable logistic regression analysis for ten-minute group to build end-tidal carbon dioxide trajectory-based prediction models.

Variables Odds ratio (95% confidence interval) P value
Ten-min survival model
 Ten-min intermediate or high EtCO2 trajectory 2.53 (1.10–5.81) 0.03
 Witness by bystander 3.00 (1.42–6.33) 0.004
 Initial shockable rhythms at ED arrival 5.21 (2.03–13.33) <0.001
Ten-min ROSC model
 Ten-min intermediate EtCO2 trajectory 3.36 (2.25–5.04) <0.001
 Ten-min high EtCO2 trajectory 6.59 (3.42–12.69) <0.001
 Age between 37 and 69 (year) 1.49 (1.02–2.20) 0.04
 Witness by bystander or EMS 1.92 (1.31–2.84) 0.001
 Initial shockable rhythms at ED arrival 5.29 (2.04–13.71) <0.001
 Duration of prehospital CPR performed by EMS (min) 0.96 (0.93–0.98) <0.001

Ten-min survival model: goodness-of-fit assessment: n = 532, adjusted generalized R
2 = 0.14, estimated area under the receiver operating characteristic curve = 0.76 (95% confidence interval: 0.72–0.79), and Hosmer and Lemeshow goodness-of-fit chi-squared test P = 0.79; ten-min ROSC model: goodness-of-fit assessment: n = 532, adjusted generalized R
2 = 0.25, estimated area under the receiver operating characteristic curve = 0.75 (95% confidence interval: 0.71–0.79), and Hosmer and Lemeshow goodness-of-fit chi-squared test P = 0.65.
CPR, cardiopulmonary resuscitation; ED, emergency department; EMS, emergency medical service; ROSC, return of spontaneous circulation.

For the 20- and 10-min models, all the predictors remained significantly associated with outcomes after the bootstrapping procedure, indicating the robustness of these models (Supplemental Table 3).

DISCUSSION

Main Findings

By using a prospectively collected database, we identified that the time-specific EtCO2 trajectory was a significant intra-arrest outcome predictor. Time-specific EtCO2 trajectory could be combined with other predictors to assist in intra-arrest prognostication at different time points during CPR. Among all the prediction models, the 20-min EtCO2 trajectory-based survival model achieved the highest discriminative performance (AUC 0.89).

Comparison with Previous Studies

For outcome prediction in OHCA, most models were developed for patients who had already achieved ROSC.21 There were few, if any, models available for patients who were still undergoing CPR. For predicting ROSC before CPR was performed, the RACA score18 was one of the most well-validated models, demonstrating AUC ranging from 0.71 to 0.76.2224 All the predictors included in the RACA score were baseline variables, such as arrest location and arrest rhythms, which did not consider the treatment effects of CPR. Nonetheless, it was possible that even though the RACA score-predicted ROSC probabilities were similar, the actual outcomes may differ because of different CPR qualities and durations delivered by rescuers. To make individualized intra-arrest prognostication, variables specific to the patient and resuscitation process, such as EtCO2, may be necessary,.

The 2018 ILCOR systematic review8 indicated that EtCO2 was associated with ROSC probability. Nonetheless, the optimal parameter of EtCO2 for prognostication is still debated.8 For example, despite its convenience in statistical analysis, average EtCO2 could not differentiate between different EtCO2 trajectories. Ascending and descending EtCO2 trajectories may have similar average EtCO2, but their prognoses may be very different.25,26 Moreover, the term “initial” EtCO2 may not accurately reflect the EtCO2 level during the early phase of CPR, as the endotracheal tube could potentially be introduced later during the resuscitation. It was reported that the specificity of EtCO2 in predicting ROSC would increase progressively from 50% at 0 min to 60%, 98%, and 100% at 10, 15, and 20 min, respectively.27 Therefore, for EtCO2 to be a valid predictor, the timing of prognostication should be specified, and its trend during CPR, instead of a single value, should be adopted.

Interpretation of Current Analysis

The 2020 ILCOR consensus6,7 recommends that EtCO2 measured after 20 min of CPR may be a predictor of survival to discharge. Rosman et al28 indicated that when higher EtCO2 levels were reached beyond 20 min of CPR they may not lead to ROSC. Progressively worsening ischemia may cause refractoriness to CPR during the metabolic phase of cardiac arrest,29 and EtCO2 trajectories beyond 20 min may not be prognostic of outcomes. Therefore, CPR for 20 min was used to select the 20-min cohort and identify the 20-min EtCO2 trajectory. The advantage of employing GBTM was that it offered an efficient method to unravel the hidden trajectories that may not be readily recognizable from the baseline characteristics or initial EtCO2 values. The significantly different EtCO2 levels among EtCO2 trajectories indicated the success of GBTM in distinguishing these hidden clusters (Table 1). Also, in an unbiased manner, GBTM identifies the hidden EtCO2 trajectories only by examining the repeatedly measured EtCO2 without considering baseline variables or outcomes. Whether the identified trajectories were associated with outcomes should be further investigated. For example, compared with patients with low 20-min EtCO2 trajectory, those with intermediate or high 20-min EtCO2 trajectory had higher proportions of bystander-witnessed arrest (Table 1), which may also explain better outcomes in the latter.

In the 20-min survival model, the multivariable logistical regression analysis indicated that the 20-min EtCO2 trajectory was positively associated with survival, demonstrating the trend of a higher EtCO2 trajectory with increased survival. Studies revealed that for every 10 mm increase in chest compression depth, EtCO2 would increase by 1.4 mm Hg30 or 4.0%.31 Higher EtCO2 trajectory may suggest better CPR quality, which may explain the positive association between EtCO2 trajectory and chances of survival. In contrast, arrest etiology may also be a confounding factor in explaining the associations between favorable outcomes and intermediate or high EtCO2 trajectory. Studies have shown that patients with asphyxial arrest32 or suspected respiratory etiology33 may have higher EtCO2 levels than those with initial shockable rhythms 32 or suspected cardiac etiology,33 respectively. Nonetheless, in our cohort, patients of intermediate or high EtCO2 trajectory had higher proportions of prehospital defibrillation by emergency medical services (EMS) (Table 1). Therefore, instead of the arrest etiology, the CPR quality may account for the positive association between 20-min EtCO2 trajectory and survival.

Whether EtCO2, along with other factors, can be used for intra-arrest prognostication was listed by AHA guidelines2 as an important knowledge gap. In the 20-min survival model, besides EtCO2 trajectory, other baseline variables, including arrest at home, prehospital defibrillation by EMS, and initial shockable rhythms on ED arrival, were also selected as significant predictors. These baseline variables had been well-validated for their predictive performance in previous studies.18 The 20-min survival model achieved excellent discriminative performance and may first answer the question presented by the AHA.2 Moreover, we further tested whether the 20-min EtCO2 trajectory could facilitate predicting ROSC. However, the AUC of the 20-min ROSC model was 0.78, lower than that of the 20-min survival model. In our study, ROSC was defined as a palpable pulse for 20 seconds, as used by RACA score.18 The swift nature of this secondary outcome may render it difficult to be predicted, even though the 20-min ROSC model included more variables than the 20-min survival model.

Finally, we developed the 10-min prediction models to explore whether outcomes could be predicted at an earlier time point during CPR. Nevertheless, the AUCs of both 10-min models were respectively lower than their counterparts of 20-min models. As shown in Figure 2, the 10-min EtCO2 trajectory was slightly different from the 20-min EtCO2 trajectory in the trend pattern. For example, the high EtCO2 trajectory continued to rise within 10 min; it was only evident later in the 20-min window that the trajectory had plateaued. Taken together, these time-specific models varied over time in terms of trajectory shapes and model performance. Earlier trajectories may still be evolving with moderate model performance, while late trajectories may have improved model performance at the cost of more medical recourses consumed. Our data suggested that 20 min after CPR may be the earliest point in time with excellent model performance to predict distant, clinically important outcomes, such as survival to hospital discharge.

Future Applications

For OHCA patients transported to the ED for continuous CPR, emergency clinicians are faced with the problem of balancing the probability of a favorable outcome with the utilization of current and future resources when making important decisions, such as termination of resuscitation or implementation of invasive extracorporeal CPR.34 Most of these advanced interventions are reserved for patients receiving CPR within a certain duration.34 Despite the fact that CPR duration is known to be inversely associated with favorable outcomes,35 it may not be the sole prognostic factor. Quality CPR may facilitate maintaining patients’ potential for favorable outcomes and lengthen the time window for advanced interventions to be implemented. Our prediction models demonstrated that time-specific EtCO2 trajectory, taking into account both the CPR duration and quality, could be a significant intra-arrest prognostic factor. In the future, time-specific EtCO2 may be transmitted instantaneously from EtCO2 monitors to mobile devices with the assistance of advanced information and communication technology. The predicted outcomes could be updated instantaneously minute by minute for each individual patient and may not be restricted to a certain time point during CPR, such as 20 min or 10 min, as used in our study.

LIMITATIONS

First, while we had internally validated the prediction models by using the bootstrap method, further external validation in other datasets should be performed. Second, the analyzed EtCO2 dataset was derived from a prospectively collected database of a single ED with a specialized training model for CPR. Further studies are needed to investigate whether these models could be generalized to other EDs or prehospital resuscitation.

CONCLUSION

Time-specific EtCO2 trajectory was a significant predictor of OHCA outcomes, which could be combined with other baseline variables for intra-arrest prognostication. For this purpose, the 20-min survival model achieved the highest discriminative performance in predicting survival to hospital discharge.

Footnotes

Section Editor: Eric Melnychuk, MD

Full text available through open access at http://escholarship.org/uc/uciem_westjem

Address for Correspondence: Chun-Yen Huang, MD, MSc, Far Eastern Memorial Hospital, Department of Emergency Medicine, No. 21, Sec. 2, Nanya S. Rd., Banqiao Dist., New Taipei City 220, Taiwan. Email: meureka@gmail.com. Chu-Lin Tsai, MD, ScD, National Taiwan University, College of Medicine, Department of Emergency Medicine, No.7, Zhongshan S. Rd., Zhongzheng Dist., Taipei City 100, Taiwan. Email: chulintsai@ntu.edu.tw
07 / 2024; 25:521 – 532

Submission history: Revision received July 18, 2023; Submitted October 25, 2023; Accepted January 19, 2024

Conflicts of Interest: By the WestJEM article submission agreement, all authors are required to disclose all affiliations, funding sources and financial or management relationships that could be perceived as potential sources of bias. Chih-Hung Wang received a grant (111-FTN0003) from the Far Eastern Memorial Hospital and National Taiwan University Hospital Joint Research Program. Chun-Yen Huang received a grant (111-FTN0003) from the Far Eastern Memorial Hospital and National Taiwan University Hospital Joint Research Program. Both Far Eastern Memorial Hospital and National Taiwan University Hospital had no involvement in designing the study, collecting, analysing or interpreting the data, writing the manuscript, or deciding whether to submit the manuscript for publication. There are no other conflicts of interest or sources of funding to declare.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.