By Eric Walter, MD, MSc
Pulmonary and Critical Care Medicine, Northwest Permanente and Kaiser Sunnyside Medical Center, Portland, OR
Dr. Walter reports no financial relationships relevant to this field of study.
SYNOPSIS: In this large, propensity-matched cohort study of patients who experienced an in-hospital cardiac arrest, patients who underwent endotracheal intubation had worse survival to hospital discharge than patients who were not intubated.
SOURCE: Anderson LW, Granfeldt A, Callaway CW, et al. Association between tracheal intubation during adult in-hospital cardiac arrest and survival. JAMA 2017;317:494-506.
In 2010, the American Heart Association (AHA) changed from the classic “ABC” (airway-breathing-circulation) teaching of CPR to “CAB” (circulation-airway-breathing).1 The AHA made this change to focus efforts on chest compressions. It was also an acknowledgement that many laypersons either are not comfortable performing or not effective at rescue breathing. However, in the hospital, personnel trained in airway management are available immediately, and many wondered if the move away from airway first was appropriate. Time spent securing a definitive airway presents advantages and disadvantages. Endotracheal intubation provides a secure airway and a more effective way to oxygenate and ventilate. However, if performed incorrectly (e.g., esophageal intubation) or if it interrupts or delays effective circulation, it may be harmful.
Previous research has suggested that endotracheal intubation during in-hospital cardiac arrest may be associated with worse outcomes. However, these studies were unable to account for confounding by indication. Patients intubated may have been more severely ill, and severity of illness may have led to worse outcomes rather than intubation. Furthermore, previous studies could not account for the time-dependent nature of intubation during cardiac arrest.
To address these limitations, Andersen et al used data from the Get With The Guidelines-Resuscitation registry, a national, prospective registry sponsored by the AHA. Data are collected on all adult patients who experience in-hospital cardiac arrest. Detailed, minute-by-minute data are collected and controlled with reliability checks. The authors compared patients who were intubated during the first 15 minutes of the resuscitation to patients not intubated. The primary outcome was survival to hospital discharge. Patients with an invasive airway in place at the time of the arrest were excluded. To address confounding factors, multiple patient and hospital covariates were used to calculate a time-dependent propensity score. This score estimated the minute-by-minute “propensity to be intubated” following cardiac arrest. Patients who were intubated were matched to controls who had not yet been intubated at the same minute. For example, a patient with a “propensity to be intubated score” who was intubated at minute five would be matched with a patient at minute five who exhibited the same score, but was not intubated yet.
Data on 108,079 patients from 668 hospitals across the country were analyzed. The median age was 69, and 58% were male. Intubation was performed in 66.3% of patients within 15 minutes. During the study period (2000-2014), there was a decrease in the proportion of patients intubated within 15 minutes (70.0% in 2000 vs. 63.6% in 2014; P < 0.01 for trend). The median time to intubation in those intubated within 15 minutes was five minutes (interquartile range [IQR], 3-8 minutes).
Survival to hospital discharge was 22.4% among all patients. In an unadjusted analysis, survival was lower for patients intubated compared to patients not intubated (17.0% vs. 33.2%; relative risk [RR], 0.58; 95% confidence interval [CI]; P < 0.001). In the propensity score-adjusted analysis, intubation remained associated with worse survival (16.3% vs. 19.4%; RR, 0.84; P < 0.001). Intubation also was associated with lower rates of return of spontaneous circulation and good neurological outcome. Results were similar among many subgroup analyses, except for patients with preexisting respiratory sufficiency (hypoxia, hypercarbia, abnormal respiratory rate, or requiring non-invasive ventilation). In this subgroup, intubation was not associated with any difference in survival (RR, 0.97; 95% CI, 0.92-1.02).
COMMENTARY
Andersen et al reported a strong association between worse survival in patients intubated during in-hospital cardiac arrest compared to those who were not. Strengths of the study are the large size, data from a well-documented registry from hundreds of hospitals around the country, and detailed minute-by-minute data. Study results are generalizable and applicable to many patients with in-hospital cardiac arrest. Several sensitivity analyses were performed, and results were robust. The time-based propensity score allowed for matching patients down to the minute. This provided a far more detailed analysis than has been performed before.
The study still is limited by its observational nature. We are not able to identify what occurred at the time of intubation. Did intubation cause CPR to be held for prolonged periods of time? Were there significant complications with intubation? Did intubation take several minutes? Additionally, because patients were matched by the minute, some controls (non-intubated patients) became cases (intubated patients) minutes later. Finally, despite propensity matching, some degree of residual confounding likely remained. Despite these limitations, these results will be valuable to practitioners at the bedside faced with making split-second decisions to prioritize care. Clearly, the ABCs of resuscitation all are important for survival, but focusing on endotracheal intubation does not appear to be the most critical step.
REFERENCE
- Field JM, Hazinski MF, Sayre MR, et al. Part 1: Executive summary: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010;122:S640-S656.
