Association Between Advanced Airway Management With Adrenaline Injection and Prognosis in Adult Patients With Asystole Asphyxia Out-of-hospital Cardiac Arrest

Background The neurological prognosis of asphyxia is poor and the effect of advanced airway management (AAM) in the prehospital setting remains unclear. This study aimed to evaluate the association between AAM with adrenaline injection and prognosis in adult patients with asystole asphyxia out-of-hospital cardiac arrest (OHCA). Methods This study assessed all-Japan Utstein cohort registry data between January 1, 2013 and December 31, 2019. We used propensity score matching analyses before logistic regression analysis to evaluate the effect of AAM on favorable neurological outcome. Results There were 879,057 OHCA cases, including 70,299 cases of asphyxia OHCAs. We extracted the data of 13,642 cases provided with adrenaline injection by emergency medical service. We divided 7,945 asphyxia OHCA cases in asystole into 5,592 and 2,353 with and without AAM, respectively. After 1:1 propensity score matching, 2,338 asphyxia OHCA cases with AAM were matched with 2,338 cases without AAM. Favorable neurological outcome was not significantly different between the AAM and no AAM groups (adjusted odds ratio [OR] 1.1; 95% confidence interval [CI], 0.5–2.5). However, the return of spontaneous circulation (ROSC) (adjusted OR 1.7; 95% CI, 1.5–1.9) and 1-month survival (adjusted OR 1.5; 95% CI, 1.1–1.9) were improved in the AAM group. Conclusion AAM with adrenaline injection for patients with asphyxia OHCA in asystole was associated with improved ROSC and 1-month survival rate but showed no differences in neurologically favorable outcome. Further prospective studies may comprehensively evaluate the effect of AAM for patients with asphyxia.


INTRODUCTION
2][3][4] Approximately 120,000 cases of out-of-hospital cardiac arrest (OHCA) have been reported annually in Japan, and the survival rates are <10% in cardiac cases and <5% in noncardiac cases. 5][8] In patients with asphyxia, the time to remove foreign bodies from the airway affects the survival rates and neurological prognoses. 7,9,10Igarashi et al reported in their recent multicenter retrospective study in Japan that prolonged airway obstruction was associated with poor neurological outcomes and that the goal should be to remove the airway foreign body within 4 minutes. 11n addition, previous reports in the United States have demon-strated that advanced airway management (AAM) and adrenaline injection by emergency medical service (EMS) were important to improve the OHCA outcomes. 12,135][26] Otomune et al reported in their subgroup analyses that AAM with adrenaline injection for patients with asphyxia due to FBAO has no effect on neurological outcomes but can improve survival. 24AAM and adrenaline injection can be performed by EMS only for patients with cardiac arrest under the instruction of a medical doctor according to a local protocol in Japan. 27Especially, the patient in whom the adrenaline injection was administered to was definitely determined to be in cardiac arrest at that time.
Asphyxia is a condition caused by interference with respiration, due to which the organs and tissues are deprived of oxygen, causing unconsciousness or death.When asphyxia occurs, tachycardia and elevated blood pressure are first observed under the influence of stimulation of the respiratory center.Then, hypotension and bradycardia may occur, eventually leading to pulseless electrical activity to asystole. 28Cardiac arrest due to asphyxia implies a long clinical time course, meaning that there was prolonged hypoxic exposure to the brain and organs prior to cardiac arrest. 29This clinical course explains why the prognosis is worse in asphyxia than in shockable cardiac arrests. 30In general, asystole represents the terminal rhythm of a cardiac arrest.Asystole has been reported to be a sign that a long time has passed since the cardiac arrest occurred and that the patient has a poor prognosis. 31Experimental reports in dogs and rabbits have reported a gradual progression of bradycardia after asphyxia, eventually leading to asystole approximately 1 hour later. 32,33An experimental report in rats showed that the QRS waveform on electrocardiogram disappeared within approximately 10 minutes after asphyxia, and it took approximately 30 minutes to reach asystole. 28n a clinical emergency setting, we often experience situations in which we must choose AAM for patients with asphyxia OHCA in asystole.As emergency physicians, it is important to be familiar with the effects of AAM for patients with asphyxia OHCA in asystole and to inform the family in advance of the outcome.However, there have been few reports on the effect of AAM for such patients.
In this study, we aimed to evaluate the effect of AAM with adrenaline injection for patients with asphyxia OHCA in asystole on neurological outcomes.

Study design and setting
We conducted a retrospective observational cohort study using all-Japan Utstein registry data of OHCA cases recorded by the Fire and Disaster Management Agency (FDMA) between January 1, 2013 and December 31, 2019.The details of this registry were previously reported. 34These data include almost all OHCAs that occurred in Japan, which are recorded in Utstein-style reporting templates by the FDMA. 35Moreover, these data are available to anyone with permission from the FDMA.Cardiac arrest was defined as the condition, in which the heart stopped mechanical beating and could no longer deliver effective circulation to the entire body and brain. 36e included patients with asphyxia OHCA who received adrenaline injection by EMS.We excluded those aged <18 and >100 years; those who did not receive adrenaline injection; those with time from call to cardiopulmonary resuscitation (CPR) or to hospital arrival of >60 minutes; and those who had missing data.We divided patients into two groups, namely the asystole and non-asystole groups, according to the initial ECGs because asystole suggested that a long period of time had passed in asphyxia OHCA.For patients with OHCA in Japan, EMS first performs chest compression and ventilation using a bag valve mask (BVM) and uses an automated external defibrillator (AED) as basic life support (BLS) according to the Japanese resuscitation guidelines. 27If patients cannot be adequately ventilated with BVM, the next option is to secure the airway with AAM, such as SGA devices and ETI; however, only specially trained EMS can perform ETI under medical control direction.Moreover, ETI is the treatment of choice in the following situations: (1) asphyxia due to FBAO, (2) absence of ventilation using an SGA device, and (3) requirement of an intubation per medical control doctor's discretion.If necessary AAM and adrenaline injection are selected under medical control direction, which is permitted only for patients in cardiac arrest and not for those with ROSC.The requirement for informed consent was waived because all-Japan Utstein cohort registry data are public anonymized data and are publicly allowed to be analyzed.According to the Japanese ethical guidelines for medical and biological research involving human participants, the Ethics Review Committee of Hokkaido University Hospital determined that ethics review for this study was not required because the Utstein data were anonymous and had already been created. 37

Data collection and outcomes
The data prospectively collected from the all-Japan Utstein cohort registry of OHCA included age, sex, time of call for EMS, regional divisions, type of person witnessed, dispatcher's instruction for CPR, BLS (chest compression, respiratory support, AED), initial ECG, AAM (SGA devices and ETI), adrenaline injection, time from call to CPR, time from call to hospital arrive, ROSC, 1-month survival, cerebral performance category (CPC) (1, good cerebral performance; 2, moderate cerebral disability; 3, severe cerebral disability; 4, coma or vegetative state; 5, death), 15,34 and overall performance category (OPC) (1, normal; 2, mild disability; 3, moderate disability; 4, severe disability; 5, death).CPC was determined by the treating physician at 1 month after OHCA.In this study, the primary outcome was favorable neurological outcomes at 1 month after OHCA, which were defined as CPC 1 or 2. The secondary outcomes were ROSC and 1-month survival rates and OPC 1 or 2.

Statistical analyses
We used propensity score (PS) matching analyses before logistic regression analysis to evaluate the effect of AAM on outcomes.We included the following variables, which were selected as important confounders in previous reports to estimate the PS to evaluate the effect of AAM: age, sex (male, female), dispatcher's instruction for CPR (yes, no), bystander CPR of a healthcare provider (yes, no), time of call for EMS (0:00-5:59, 6:00-11:59, 12:00-17:59, 18:00-23:59), regional divisions (Hokkaido, Tohoku, Kanto, Chubu, Kinki, Chugoku and Shikoku, and Kyusyu and Okinawa), bystander BLS (push, respiratory, AED), and time from call to CPR.][8][9][10][11] To evaluate propensity analyses, we used standardized mean differences (SMDs), which can be used to evaluate a balanced matching if it is <0.1. 38We calculated crude and multivariable adjusted odds ratios (ORs) and 95% confidence intervals (CIs) in logistic regression analyses to estimate the effect of AAM for primary and secondary outcomes.Continuous variables are presented as means and standard deviations (SDs) and as medians and interquartile ranges.Categorical variables are presented as numbers and percentages (%).We also analyzed patients with non-asystole OHCA using PS matching and logistic regressions to evaluate the effect of AAM for outcomes in non-asystole.
We used the EZR (version 1.55; Saitama Medical Center, Jichi Medical University) for all statistical analyses.

RESULTS
There were 879,057 OHCA cases, including 70,299 asphyxia OHCA cases documented in the all-Japan OHCA Utstein registry between January 1, 2013 and December 31, 2019 (Figure 1).After exclusion, 60,924 patients with asphyxia OHCA remained, and we extracted the data of 13,642 patients with adrenaline injection by EMS.We divided these patients into 7,945 with asystole and 5,697 with non-asystole.Finally, we divided patients with asphyxia OHCA with asystole into 5,592 and 2,353 patients with AAM and without AAM, respectively.After 1:1 PS matching, 2,338 patients with asphyxia OHCA with AAM and 2,338 such patients without AAM were included.
Table 1 shows the baseline characteristics of patients with asphyxia OHCA according to asystole on initial ECG.AAM was performed to 5,592 patients (70.4%) in asystole and 3,972 patients (69.7%) in non-asystole.Time from call to CPR was longer in the asystole than in the non-asystole group (9.8; SD, 3.5 min vs 9.3; SD, 3.5 min).Chest compressions were performed on 5,022 patients (63.2%) in asystole and 3,424 patients (60.1%) in nonasystole.Bystander CPR of a healthcare provider was performed to 110 patients (1.4%) in asystole and 370 patients (6.5%) in nonasystole.Favorable neurological outcomes were observed in 37 patients (0.5%) in asystole and 68 patients (1.2%) in non-asystole.
Table 2 shows the baseline characteristics of asphyxia OHCA in asystole according to AAM before and after PS matching analysis.After PS matching, 2,338 patients were matched for each group, and the baseline characteristics were well balanced between the AAM and no AAM groups except regional divisions (SMD <0.1).
Table 3 shows the crude and multivariable adjusted ORs of AAM in asystole for the primary and secondary outcomes after PS matching analysis.Regarding the primary outcome, favorable neurological outcome (CPC 1-2) was not significantly different between the AAM and no AAM groups (adjusted OR 1.1; 95% CI, 0.5-2.5).ROSC (adjusted OR 1.7; 95% CI, 1.5-1.9)and the 1-month survival rates (adjusted OR 1.5; 95% CI, 1.1-1.9)were significantly higher in the AAM than in the no AAM group, but the favorable functional outcomes (OPC 1-2) were not significantly different between the two groups (adjusted OR 1.0; 95% CI, 0.4-2.1).
Table 4 shows the crude and multivariable adjusted ORs of AAM in non-asystole for the primary and secondary outcomes after PS matching analysis.Regarding the primary outcome, the favorable neurological outcomes (CPC 1-2) was not significantly different between the AAM and no AAM groups (adjusted OR 1.5; 95% CI, 0.8-2.9).ROSC (adjusted OR 1.6; 95% CI, 1.4-1.9)and 1month survival (adjusted OR 1.7; 95% CI, 1.4-2.1)were significantly higher in the AAM than in the no AAM group, but favorable functional outcome (OPC 1-2) was not significantly different between the two groups (adjusted OR 1.6; 95% CI, 0.8-3.0).

DISCUSSION
In this Japanese nationwide Utstein registry cohort study with PS matching, we revealed that AAM with adrenaline injection for patients with asphyxia OHCA was associated with better 1-month survival and ROSC in both asystole and non-asystole patients.We also found that AAM with adrenaline injection for patients with asphyxia OHCA was not associated with poor neurological and functional outcomes.Especially in elderly patients, it is important to understand the effect of AAM for asphyxia OHCA in asystole, explain the prognosis to the family, and confirm their advance care planning before introducing AAM.In previous reports, AAM for patients with asphyxia was associated with poor neurological outcomes. 24,25Unlike in previous reports, we did not find that performing AAM for patients with asphyxia OHCA in asystole had an opposite effect on favorable neurologic prognosis.One possible reason for this difference could be the different inclusion criteria for the group of patients for whom AAM was not performed.Especially, patients who were in cardiopulmonary arrest at the time of call but got ROSC before arrival of EMS and those who might have had a favorable outcome, may have been included in the previous studies.Sakurai et al reported that confirmed cardiac output on EMS arrival should be considered as a confounding factor, as indicated in observational studies of AAM. 39In the Japanese Utstein registry, the reasons for performing or not performing AAM for patients with OHCA were not recorded.Japanese EMS can only perform AAM and adrenaline injection for patients in cardiac arrest. 27If EMS determined that the patient was in cardiac arrest and adrenaline injection was performed, the patient was definitely in cardiac arrest at that time.Thus, in this study, we included only patients who had cardiac arrest, as determined by EMS, and received adrenaline injection.
Whether prehospital AAM is better than BVM for patients with OHCA remains unclear. 15,16,40According to Jabre et al in their randomized controlled trial of resuscitation for OHCA, there was no difference in neurological prognosis between ETI and BVM. 17 Such reports have indicated the possibility of bias in the difficulty and success rate of the intubation technique, but no report has examined this issue in detail. 41According to Benger et al in their randomized controlled trial, in patients with OHCA, the type of AAM, such as ETI or SGA devices, did not change the outcome of neurological prognosis. 18esuscitation time bias is common in Utstein registry analysis, which means that the resuscitation time affects the frequency of resuscitation interventions.Longer resuscitation duration is associated with worse outcomes and the later intervention is biased toward harmful. 42In this Utstein registry data, the exact onset time of the FBAO was not recorded.We believed that if we focused on asystole patients, we could focus on patients with severe asphyxia to reduce the influence of time bias as much as possible.Okubo et al reported that the timing of AAM was not statistically associated with improved 1-month survival for shockable rhythms, but AAM within 15 minutes after EMS-    initiated CPR was associated with improved 1-month survival rates for non-shockable rhythms. 43Fukuda et al reported that delay in AAM was associated with a decreased chance of 1-month neurologically favorable survival among patients with OHCA. 44Izawa et al reported that in the time-dependent PS sequential matching for OHCA in adults, AAM was not associated with survival among patients with shockable rhythm, but AAM was associated with better survival among patients with non-shockable rhythm. 14ur study has several strengths.First, using nationwide Utstein registry data, this study could be conducted with a large sample size.Second, there were no studies concerning the effects of AAM with adrenaline injection for asphyxia OHCA focused on asystole, which indicated that a long time had passed and the condition was severe.
However, this study had some limitations.First, as the Utstein data were recorded only for determined items, we could not control many confounders in our study, such as medical histories, living conditions, activities of daily living, quality of BLS or ALS, detailed time course of resuscitation, detailed information of foreign bodies that obstruct patient's airway, and medical treatment after hospital arrival.Second, this study used the Utstein data from Japan, and the results may not be simply applicable to other countries because of differences in the emergency medical care and EMS systems in these countries. 40hird, our study did not examine cases, in which intubation was technically impossible or difficult because the Japanese Utstein data do not provide detailed information on the reasons for the decision to intubate or not intubate.An increasing number of intubation attempts during OHCA resuscitation was associated with a lower likelihood of favorable neurological outcomes. 45ourth, there is no information in the Ustein registry data to confirm whether the foreign body was removed from the airway.This may have implications for the effect of AAM in this study.Fifth, this study is a retrospective analysis of the Utstein data.Thus, there is a potential risk of power to detect the difference in the neurological outcome in this study.Finally, as the exact time of AAM was not recorded in the Ustein registry data, we could not analyze the resuscitation time bias.Further detailed studies, such as prospective randomized studies, on the effect of AAM for asphyxia are required.
In conclusion, AAM for asphyxia OHCA in asystole was associated with improved ROSC and 1-month survival rates, but it made no difference in the neurologically favorable outcomes.This result was different from that of previous studies, in which AAM had negative effects on the neurological outcomes of patients with asphyxia.Further prospective studies are expected to comprehensively evaluate the effect of AAM for patients with asphyxia.

Figure 1 .
Figure 1.Flowchart of the study data documented in all-Japan Utstein out-of-hospital cardiac arrest registry data between January 1, 2013, and December 31, 2019.AAM, advanced airway management; CPR, cardiopulmonary resuscitation; OHCA, out-of-hospital cardiac arrest; PS, propensity score.

Table 1 .
Baseline characteristics of patients with asphyxia OHCA with adrenaline injection according to asystole

Table 2 .
Baseline characteristics of patients with asphyxia OHCA in asystole according to AAM before and after PS matching analysis AED, automated external defibrillator; BLS, basic life support; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; EMS, emergency medical service; IQR, interquartile range; OPC, overall performance category; ROSC, return of spontaneous circulation; SD, standard deviation.

Table 3 .
Crude and adjusted odds ratios of AAM for primary and secondary outcomes among asphyxia OHCA cases in asystole BLS, basic life support; CI, confidence interval; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; EMS, emergency medical service; OPC, overall performance category; OR, odds ratio; ROSC, return of spontaneous circulation.adjusted factors: age, sex, time of call for EMS, regional divisions, bystander CPR of health care provider, dispatcher's instruction for CPR, bystander BLS, time from call to CPR.

Table 4 .
Crude and adjusted odds ratios of AAM for primary and secondary outcomes among asphyxia OHCA cases in non-asystole AAM, advanced airway management; BLS, basic life support; CI, confidence interval; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; EMS, emergency medical service; OPC, overall performance category; OR, odds ratio; ROSC, return of spontaneous circulation.adjusted factors: age, sex, time of call for EMS, regional divisions, bystander CPR of health care provider, dispatcher's instruction for CPR, bystander BLS, time from call to CPR.Katabami K, et al.J Epidemiol 2024;34(1):31-37 j 35