The interaction between endotracheal intubation and cardiopulmonary resuscitation

The establishment of airway to maintain gas exchange is one of the most basic elements of life-sustaining in physiology. In the past few decades, the importance of tracheal intubation for resuscitation has never been questioned by physicians, and because there is a definite possibility of great injury to the control group, it is rashly to run randomized controlled clinical trials. However, many traditional concepts of cardiopulmonary resuscitation have changed rapidly since 2000. More and more clinical reports showed that persistent high quality chest compression is the first factor leading to the success of cardiopulmonary resuscitation. All such operations that may result in interruption of medical treatment of chest compression, including endotracheal intubation, defibrillation, and observation of heart rhythm, could reduce the success rate of cardiopulmonary resuscitation^[1]. Due to the high degree of difficulty in practicing traditional tracheal intubation during resuscitation, many observational studies indicated that the first tracheal intubation during cardiopulmonary resuscitation has a failure rate of nearly 15%, and children's tracheal intubation has a failure rate of up to 40%, leading to the interruption of chest compression for 1-5 minutes^[2-5]. In addition, even if tracheal intubation is successful, excessive ventilation may lead to failure of cardiopulmonary resuscitation. Excessive pulmonary ventilation may increase intrapulmonary pressure which could block coronary perfusion pressure, leading to no restoration of cardiac perfusion. Therefore, scientists in the field of cardiopulmonary resuscitation put forward a bold hypothesis: what if the technical or low invasive methods, like balloon mask ventilation and laryngeal mask ventilation replace traditional tracheal intubation, it may reduce the interruption of chest compressions or avoid hyperventilation, therefore it is possible to improve the efficiency of resuscitation^[6-8].

Observational research results

Before conducting a large-scale clinical trial to test this hypothesis, high quality observational research is bound to be performed to support it. The American Heart Association has registered the quality data of the home and hospital first aid in the United States since 2000, and established a large and high quality Get With The Guidelines database, which provides the possibility of analyzing the relationship between first-aid prognosis and endotracheal intubation. The emergency team at the Affiliated Hospital of Harvard University in the United States first performed a strict matched analysis of first-aid data in children and adults. Both analyses indicate that no matter for adults or children, the live discharge ratio (Survival to discharge) of patients undergoing endotracheal intubation during resuscitation was significantly lower than those without endotracheal intubate, it overturns the iron law of textbooks for the past fifty years. The two analysis reports were published in the Journal of the American Medical Association (JAMA) in 2016 and 2017^{[9, 10]}. The data in adult groups showed that 71615 (66.3%) cases of cardiac arrest in 108079 were intubated in 15 minutes, among which 43314 (60.5%) patients had a matched control group. The results showed that the patients with tracheal intubation were on a disadvantage condition, regardless of the recovery rate of autonomic rhythm (57.8% Vs. 59.3%; P < 0.001), the survival rate of discharge (16.3% Vs. 19.4%; P < 0.001), or the neurological prognosis (10.6% Vs. 13.6%; P < 0.001)^[11]. Due to differences in the causes of cardiac arrest in children and in adults, another children group were performed indecently, among which 1555 (68%) out of 2294 children were performed tracheal intubation within 15 minutes before resuscitation, and the survival rate of the children (36% Vs. 41%; P=0.03) was poor, but there was no statistical significance either in the recovery rate of autonomic heart rhythm (68% Vs. 68%; P= 0.96) nor in neurological prognosis (10.6% Vs. 13.6%; P < 0.001).

Design and analytical bias

To this conclusion, clinicians would usually doubt at first sight whether it comes from the fallacy of "take the cause as the result". Only severe patients need tracheal intubation, and those who do not need the tracheal intubation tend to be able to regain heart beats simply by pressing their chests. There are typical indications of adaptive disturbance (confounding by indication) in the observational study of non-randomized clinical trials. And it's difficult to control such interference even when many factors such as patient age, comorbidity, diagnosis, and resuscitation drugs are included in the correction model. JAMA, a long historical and influential medical journal in the United States, is known to all the judges and editorial members, and the findings may have great influence on the health of the world and even the health of human beings. The results cannot be published until after rigorous and repeated test by top clinical medicine, statistics, and epidemiologists. Of course, to overcome the rapid changes of the disease and the disturbance of indications during the emergency resuscitation, the author used the time-dependent propensity score to make a strict matching analysis. The application of this statistical analysis in this study can be interpreted directly as follows: the author used one minute as a single case, all the factors that may affect the prognosis of the patients who underwent endotracheal intubation in 15 minutes before resuscitation were analyzed in 15 stratified match, which was equivalent to 15 different randomized clinical trials. The results still showed that tracheal intubation patients had poor prognosis. The study was also externally validated by the study from the world's largest sample size at present, which has collected 640, 000 samples from all Japan pre-hospital sudden cardiac death, and the study also showed that patients with tracheal intubation had lower resuscitation survival rate than those with non-tracheal intubation patients.

Although it is an observational study, the study still has many advantages that cannot be surpassed in a short period. In addition to the above rigorous statistical analysis, the database had a prospective collection of large-scale samples from 668 hospitals in the United States for 15 years. Even if a randomized trial is conducted at present, sufficient samples cannot be accumulated for conclusions. In addition, under ethical considerations, it is foreseeable that it is difficult to carry out a randomized clinical trial of endotracheal intubation during resuscitation in short period. Henry Wang, an emergency cardiopulmonary resuscitation scholar at University of Alabama, has started designing a randomized clinical trial involved in 9000 samples, which takes laryngeal mask airway as the control group. Due to the large scale sample size, the results of this study cannot be drawn until years later^{[12, 13]}. At the same time, the study cannot merely testify the feasibility of using balloon mask ventilation before hospitalization. Therefore, the existing evidence is rich enough to enable us to review the current first-aid treatments of endotracheal intubation and make sure whether its' surely harmless to patients.

Clinical enlightenment

Although no final conclusion can be drawn from the present study whether endotracheal intubation should be performed for cardiopulmonary resuscitation, we cannot ignore the clinical enlightenment from this high-quality observational study. And therefore some adjustment in clinical resuscitation measures should be made. Firstly, five cycles of cardiopulmonary resuscitation should first be performed for cardiac arrest patients, unless there are clear evidences of airway obstruction. secondly, if the first endotracheal intubation fails, the clinicians should not persist in tracheal intubation so as to save time for the chest compression. Laryngeal mask airway may be considered for patients who failed in the first endotracheal intubation attempt; finally, hyperventilation should be avoided if tracheal intubation is successful.

Conclusion

Due to the limitation of unpredictability and rarity of cardiac arrest events, research on cardiopulmonary resuscitation remains difficult. However, with the establishment of large registry database, the progress in resuscitation research field will be gradually accelerated by large-scale data analysis. And the clinical practice in resuscitation will also become more and more accurate and precise.