Association Between Prehospital Airway Management Methods and Neurologic Outcome in Out-Of-Hospital Cardiac Arrest (OHCA) with Respiratory Cause.


 Purpose: Cardiac arrests are resulted by various aetiology including respiratory cause. Advanced airway placement is an important prehospital intervention for oxygenation and ventilation in respiratory cardiac arrest. We evaluated the association between of advanced airway method and neurologic outcome in arrest with respiratory cause.Method: Adult witnessed non-traumatic OHCAs treated by emergency medical service (EMS) providers in 2013-2017 were enrolled in a nationwide OHCA database. The association between airway management methods (endotracheal intubation (ETI), supraglottic airway (SGA) and bag valve mask (BVM)) and outcome were evaluated according to the presumed cause of cardiac arrest (cardiac, respiratory or others). The primary outcome was good neurological recovery at discharge. Multivariable logistic regression models with interaction analysis was conducted.Result: Of 40,443 eligible OHCA patients, the cause of arrest of 90.0%, 7.5%, and 2.4% of patients were categorized as cardiac, respiratory and others, respectively. There were no statistically significant differences in the effect of the advanced airway type on good neurologic recovery in the total population (aOR 0.96 (0.81-1.14) for ETI; 1.01 (95% CI 0.93-1.11) for BVM). However, ETI was associated with better neurologic recovery than SGA or BVM in OHCA in cardiac arrest with suspected respiratory cause (aOR 3.12 (95% CI 1.24-7.80) for ETI; 0.99 (95% CI 0.51-1.91) for BVM).Conclusion: Prehospital ETI was associated with good neurologic outcome when the cause of arrest was respiratory. ETI may be considered initially when a respiratory cause is suspected on the scene.


Introduction
Various etiologies are responsible for the occurrence of out-of-hospital cardiac arrest (OHCA). (1,2) Although majority of OHCA is caused by cardiac cause, respiratory problems resulting from hypoxia and hypoventilation in conditions such as ARDS, pneumonia, COPD and asphyxia also cause cardiac arrest. (3)(4)(5) For cardiac arrest caused by respiratory aetiology, resuming and providing oxygenation and ventilation during cardiac arrest is especially important for correcting the cause of arrest and minimizing neurological damage in the brain. (6,7) Advanced airway placement is one of the key prehospital interventions performed by emergency medical service (EMS) providers during eld resuscitation of OHCA. (8) The main purpose of advanced airway placement is to provide more effective oxygenation and ventilation than basic airway such as bag-valve-mask (BVM). (9,10) Various studies have compared the effects of prehospital advanced airway management methods on outcomes in OHCA, but they are mostly tested in cardiac arrest with presumed cardiac cause the results are still controversial. (11)(12)(13)(14)(15) Difference in bystander response, prehospital epinephrine administration and clinical outcome between arrests caused by respiratory and cardiac cause have been reported in other studies. (16)(17)(18) However, little is known about the association of prehospital advanced airway management and outcome according to the cause of cardiac arrest.
In this study, we investigated the association between prehospital advanced airway method and neurologic outcome according to the presumed cause of cardiac arrest under the hypothesis that the effect of each advanced airway management type would be different when cause of arrest is suspected to be respiratory.

Study design and setting
A retrospective observational study was conducted using the prospectively collected nationwide EMS based OHCA registry of Korea. In Korea, a single national re agency exclusively operates the prehospital EMS system. In cases of OHCA, EMS providers perform basic life support (BLS), including the use of automatic de brillators for rhythm analysis and de brillation. EMS providers have no authority to declare death or stop CPR unless there is return of spontaneous circulation (ROSC). Prehospital EMS providers in Korea are mostly emergency medical technicians (EMTs) and nurses. The majority of EMS providers are EMTs, for whom there are two levels of national certi cation (Level-1 EMT and Level-2 EMT).
Level-1 EMTs provide services similar to EMT-intermediate services in EMS systems in United States. Advanced airway placement can only be performed by a Level-1 EMT or a nurse under direct medical control from a medical director on the phone. Endotracheal intubation or supraglottic airway insertion, such as I-gel and laryngeal mask airway (LMA), are also available. The decision to perform advanced airway placement and selection between airway techniques is usually at the discretion of an on-duty EMS provider.

Data source
The Nationwide OHCA database, which includes all EMS-assessed OHCAs, was retrieved from the following four sources: the EMS run sheets for basic ambulance operation information, the EMS cardiac arrest registry, the dispatcher CPR registry for the Utstein factors, and the hospital medical record review registry for hospital care and outcomes. A detailed description of the data acquisition of each registry, as well as the training and quality of the medical record reviewers, have been described in previous studies. (19,20) Study population All adult witnessed OHCAs assessed and treated by EMS providers from January 2013 to December 2017 were initially enrolled. OHCAs witnessed by EMTs during transport (whose selecting airway management method could different from bystander witnessed OHCA patients), OHCAs treated only by EMS teams consisting of level 2 EMTs only (who could not perform advanced airway placement) and OHCAs with unknown survival and neurologic outcome were excluded. We also excluded OHCAs with traumatic cause.

Exposure and outcome variables
The primary exposure of interest was the type of airway management method. The leader of the EMS team at the scene recorded a detailed OHCA registry, including the method of airway management. The leader of the team recorded whether the advanced airway was attempted, if it was successful (and if successful, the time of advanced airway placement) and the type of advanced airway method attempted (endotracheal intubation or supraglottic airway (Combitube, LMA, King airway or I-gel)). Current guidelines for prehospital EMS management recommend that level-1 EMTs or nurses can perform advanced airway placement according to their preference under direct medical direction. The guidelines recommend the procedure should take less than 30 seconds and that attempts should be performed up to 2 times. The primary outcome of the study was good neurological recovery at hospital discharge, de ned as a cerebral performance category (CPC) 1 or 2.
Medical record reviewers assessed the CPC score based on discharge summary abstracts or documented medical records. Secondary outcome was survival to hospital discharge.

Cause of cardiac arrest
Pathogenesis of cardiac arrest in the database is determined by trained medical record reviewers after reviewing hospital records of patients according to Ustein recommendation. (21) In our analysis, we recategorized the cause of arrest into three groups after excluding traumatic arrest. Cause of arrest was presumed to be cardiac unless there was evidence to suggest a non-cardiac cause. Cardiac arrests caused by obvious aggravation of respiratory disease, asphyxia (including foreign body airway obstruction), drowning(submersion) or hanging were categorized as respiratory group. Lastly, remaining arrests by other etiologies such as poisoning or electrocution were categorized as others. We excluded traumatic arrest other than asphyxia, drowning and hanging from our study.

Statistical Analysis
Demographics, EMS management, hospital management and outcomes were compared by cause of arrest using the χ2 test for categorical variables and one-way analysis of variance (ANOVA) or the Kruskal-Wallis test for continuous variables.
Multivariable logistic regression was conducted to analyze the effect of airway management method on the outcome and is presented as the odds ratios (ORs) with 95% con dence intervals (CI) after adjusting for covariables. The variables included in the model were age, gender, location of arrest, bystander CPR, response time interval, hypertension and heart disease. We tested for multicollinearity between the covariables in the model, and there was no multicollinearity. To assess changes in the effect of airway management according to the cause of arrest, we added an interaction analysis between airway management methods and the cause. STATA 14.1 (StataCorp, College Station, TX, USA) was used for all analyses.

Results
In total, 145,003 OHCA patients in our registry were initially enrolled from January 2013 to December 2017. Patients aged less than 18 years (N = 3,170), patients with OHCA not witnessed (N = 79,208), patients not treated by EMS providers (N = 4,063), patients with cardiac arrests occurring during ambulance transport (N = 7,203), patients treated by EMS teams without any level-1 EMTs capable of advanced airway placement (N = 5,756), traumatic cause (N=5,062) and patients with unknown outcome (prehospital ROSC, survival and neurologic outcome) (N = 98) and were excluded ( Figure 1).
After serial exclusion, 40,443 OHCA patients were included in the nal analysis. Regarding the presumed cause of arrest, 36,404 (90.0%) cases were cardiac, 3,050 (7.5%) were respiratory and 989 (2.4%) were others. The demographic characteristics according to cause are summarized in Table 1. The proportion of advanced airway placement differed according to the group, and the rate of endotracheal intubation was highest in the respiratory cause group (6.7% vs. 8.1% vs. 5.1%, respectively, p<0.01). The neurologically favorable survival-to-discharge rates were 12.5% in the cardiac group, 9.6% in the respiratory group, and 12.1% in the others group (Table 1, p<0.01).
In the interaction analysis, the magnitude of the association between type of advanced airway and outcome measures was different according to the presumed cause of cardiac arrest. ETI was associated with signi cantly better neurological recovery than SGA (aOR 3.12 (95% CI 1.24-7.80)) and BVM(aOR 3.16 (95% CI 1.43-6.98)) when the cause of arrest was respiratory. Although the effect of ETI was nonsigni cant, BVM was associated with lower survival to discharge than SGA when the cause of arrest was respiratory (aOR 0.70 (95% CI 0.52-0.94)) ( Table 3 ).

Discussion
In this analysis of a nationwide, population-based OHCA registry, we evaluated the association between prehospital advanced airway methods and clinical outcome of cardiac arrest. The magnitude of the association was different according to the presumed cause of cardiac arrest. There was no statistically signi cant difference in the effect according to the advanced airway type in total population and arrest with cardiac origin. However endotracheal intubation was associated with better neurologic outcome when the cause of arrest was respiratory. We believe more e cient oxygenation and ventilation through prehospital endotracheal intubation provides a rapid and effective solution to respiratory cause of cardiac arrest, as an antidote for drug intoxication. Although most of cardiac arrest cases receive endotracheal intubation eventually after hospital transport, faster reversal of hypoxia and hypoventilation by prehospital endotracheal intubation in the eld might show an effect on minimizing hypoxic brain injury. Based on the results of our study, we suggest that endotracheal intubation should be considered as a rst choice of method in OHCA patients whose cause of arrest is suspected to be respiratory, such as respiratory disease, asphyxia and strangulation.
Even after recent prospective clinical trials, de ning the optimal choice of prehospital advanced airway methods in OHCA is controversial. Wang (12) Overall, the results showed no de nite bene t of ETI over BVM or SGA. In a nationwide cohort study conducted in Japan, advanced airway management (ETI and SGA) was associated with better survival and more favorable neurologic outcomes compared with no advanced airway management in non-shockable (pulseless electrical activity and asystole) OHCA patients. (23) However, these studies did not consider the etiologies of arrest.
The supraglottic airway can be inserted blindly and therefore has a higher success rate than endotracheal intubation, even for those with low skills, such as EMS providers in Korea. However, compression of important structures such as carotid vessels has been suggested in animal studies in some SGAs due to its morphologic characteristics and mechanism. (24,25) In severe hypoxia in respiratory arrest, carotid blood ow reduction caused by SGA insertion might be more critical than for other etiologies of cardiac arrest. One case reported ventilation failure of a drowned patient with supraglottic airways due to leakage from high pressure. (26) Several previous studies have reported negative effects of prehospital advanced airway placement on the outcome of respiratory cardiac arrest. (15,18) However, these studies included only limited number of patients with medical intrinsic respiratory disease, excluding external causes of arrest, such as asphyxia or hanging. Additionally, the studies only tested the overall effect of prehospital advanced airway rather than the effect of each advanced airway method according to different etiologies, as in our study.
Only witnessed cardiac arrests were enrolled for analysis in our study. We opted not to include patients with arrest not witnessed by initial bystanders. Although our study was a retrospective observational study and there are currently no guidelines for EMS providers to consider cause of arrest when selecting the airway management method, we supposed it might be di cult for EMS providers to estimate the cause of arrest in unwitnessed cases. We believe that, for witnessed cardiac arrest, EMS providers can presume the cause of cardiac arrest based on a history taken from the bystander who witnessed the arrest or could at least determine whether the cause was respiratory or not. Based on the result of our study, we suggest that EMS providers might consider estimated cause when deciding the airway management method in the eld.
There are some limitations to our study. First, this was a retrospective observational study; therefore, careful interpretation is needed considering the unique characteristics of our EMS system. Second, we extracted cause of arrest by medical record review. For patients who did not achieve any ROSC and were pronounced dead after initial resuscitation, there might be few clues to estimate the cause of arrest.
Third, the number of patients with favorable neurological outcome in the ETI group with respiratory arrest were very small (9 patients), which leading to wide con dence intervals and making results di cult to interpret. Furthermore, detailed diagnosis of respiratory disease or severity of trauma were not included in our database. Last, there might be unmeasured confounders not included in our registry that could affect both airway management method and outcome.

Conclusion
The association between prehospital advanced airway and the neurological outcomes of patients with witnessed nontraumatic OHCA was not signi cant. However, endotracheal intubation was associated with better neurologic recovery in cardiac arrest with respiratory cause. Endotracheal intubation may be considered initially by EMS providers when a respiratory cause is suspected for the pathogenesis of cardiac arrest.

Declarations CONFLICT OF INTEREST
The Authors declare that there are no con icts of interest.

Funding Acknowledgement
None.
Author Contributions: Drs. Choi and Kim had full access to all of the data in the study and take responsibility for the integrity of the data as well as the accuracy of the data analysis.       Figure 1 Enrollment ow of study population ; OHCA: out-of-hospital cardiac arrest; EMS: emergency medical service ; EMT; emergency medical technician