Early endotracheal intubation is not associated with the rate of return of spontaneous circulation following cardiac arrest at the emergency department: an exploratory analysis

Cardiac arrest poses a significant global public health challenge, manifesting in approximately 550,000 cases annually within the United States.[1] In-hospital cardiac arrest (IHCA) is commonly attributed to airways and respiratory issues.[2] Recommendations emphasize the expertise of responders in airway management.[3] Various options exist, such as chest compression-only cardiopulmonary resuscitation (CPR), bag-mask ventilation (BMV), and advanced airways. The BMV and advanced airways are not deemed equivalent or superior based on previous evidence.[4] Achieving consistency in choosing and timing the optimal airway approach during IHCA is crucial. The current American Heart Association guidelines suggest an advanced airway strategy when endotracheal intubation (ETI) success rates are high, but the optimal time for advanced airway management remains unclear.[5] Wong et al[6] revealed that survival improved by less than 5 min with advanced airway management. According to a subgroup analysis of IHCA patients in emergency departments (EDs), early intubation was associated with a 1.5-fold greater rate of return of spontaneous circulation (ROSC) than in other locations.[7] ED patients’ constant monitoring and immediate management, with readily available intubation equipment, enhance early intubation and survival rates.[6] Nonetheless, IHCA patients intubated within the first 15 min had a lower ROSC rate.[8] Therefore, this study aimed to the impact of early and late intubation in ED patients with IHCA.

A retrospective cohort study was conducted at Chiang Mai University Hospital between June 2019 and May 2022, in accordance with the strengthening the reporting of observational studies in epidemiology (STROBE) statement.[9] The study included ED cardiac arrest (EDCA) patients older than 18 years who had not undergone advanced airway procedures during cardiac arrest. Exclusions comprised out-of-hospital cardiac arrest (OHCA) patients with specific conditions (Figure 1). The data included demographics, cardiac arrest details, comorbidities, treatments, and outcomes.

Figure 1.

Study flowchart. DNAR: do-not-attempt-resuscitation; ED: emergency department; OHCA: out-of-hospital cardiac arrest; ROSC: return of spontaneous circulation.

ETI time is the time from the start of chest compressions to the successful placement of the advanced airway. Early and late ETIs were defined as completing endotracheal tube placement within 5 min and over 5 min, respectively. In our study, late intubation was defined as 5 min based on the findings of a previous study which showed that patients with advanced airways of more than 5 min had low survival to hospital discharge.[6]

The primary outcome was the rate of ROSC; the secondary outcome was the 24-hour survival rate.

The sample size, determined by relevant literature,[10] was aimed at least 65 using a proportional-hazards regression model formula[11] for 80% power and a two-sided alpha error of 0.05.

Categorical variables were described using proportions and percentages. For continuous variables, means and standard deviations or medians with interquartile ranges were selected if appropriate. Standardized differences (STDs) between variables with more than 10% variation were considered significant, suggesting potential confounding effects.[12] We applied an inverse probability treatment weighting (IPTW) technique to adjust for imbalance in prognostic factors.[13] First, we developed a multivariable logistic regression model to predict the likelihood of being intubated during the EDCA. Pre-cardiac arrest factors, including age, sex, time of ED arrival, mechanism of cardiac arrest, shock index on ED arrival, previous airway and breathing procedures, and time from ED arrival to the onset of cardiac arrest, were all included in the propensity model. The model anticipated the probability of being intubated at more than 5 min after attempted resuscitation, defined as treatment weights for the late intubation group (>5 min). In contrast, the model predicted that the probability of being intubated at 5 min or less after attempted resuscitation was defined as the treatment weight for the early intubation group (≤5 min). We developed a balanced diagnostic plot to compare the differences in patients’ characteristics between weighted and unweighted samples. Pre-cardiac arrest factors with an STD of more than 10% after weighting, as well as intra-cardiac arrest factors (initial shockable rhythm and time from attempted resuscitation to the first dose of epinephrine administration), were adjusted for double-robustness in the weighted logistic regression model.[14]

We employed a multivariable logistic regression model to assess outcomes and a weighted Cox proportional hazards regression analysis to compare the association between intubation timing during the EDCA and the rate of sustained ROSC rates. Analyses were conducted using STATA16 (StataCorp., USA) with two-tailed tests and a significance set at P<0.05.

Among 519 individuals who received CPR at the ED from June 2019 to May 2022, 70 who experienced EDCA without in-situ ETI were included in the analysis. Of these, 46 (65.7%) were intubated within 5 min, while 24 (34.3%) were intubated over 5 min. Table 1 shows the differences in baseline characteristics between the early and late ETI groups. The early group had a mean ETI time of 2.8 min, compared to 7.5 min in the late group. Nearly all the baseline characteristics, except for age, arrests during the night shift, and some comorbidities such as asthma and chronic obstructive pulmonary disease (COPD), diabetes, and malignancy, differed between the two groups (Supplementary Table 1).

Supplementary Figure 1 displays the STD between unweighted and weighted samples. Initially, imbalanced, most variables achieved balance after IPTW, except for age and previous airway and breathing procedures, with absolute STDs <0.1.

The ROSC rate was higher in the late intubation group than in the early intubation group, but the difference was not statistically significant (adjusted OR [aOR] 2.24, 95% confidence interval [CI] 0.76–6.58; P=0.14). The late intubation group had a significantly lower 24-hour survival rate than the early intubation group (aOR 0.27, 95%CI 0.09–0.77; P=0.02). The late intubation group did not have a greater ROSC rate than the early intubation group (hazard ratio 1.12; 95%CI 0.75–1.67; P=0.59, Figure 2).

Figure 2.

Cumulative chances of achieving a return of spontaneous circulation versus minutes after intubation during resuscitation and corresponding analytical approximation with Cox proportional hazards regression model. EDCA: emergency department cardiac arrest; ROSC: return of spontaneous circulation.

This study included 70 patients with EDCA without ETI, and observed a 65.7% intubation rate within 5 min. While no significant ROSC rate difference emerged between the early and late intubation groups, the late intubation group exhibited a significantly lower 24-hour survival rate. The optimal option for airway management during cardiac arrests has still been inconsistent.[5] Notably, our study contradicts prior findings favoring early ETI, emphasizing the unique focus of EDCA.[15,16]

One potential explanation for similar ROSC rates despite timing differences is the ED team’s effective CPR maintenance, which mitigates the impact of ETI interruptions. CPR quality has been linked to survival, with ETI causing longer interruptions.[17] We proposed that the care team in the ED was able to minimize the impact of these interruptions by maintaining high-quality CPR throughout the resuscitation process. Despite the potential advantages of early intubation for IHCA patients,[6] there are several issues with ETI during CPR. Concerns about ETI during CPR include a high incidence of esophageal intubation and significant interruptions, impacting coronary perfusion pressure.[18] Our findings are consistent with a previous article showing that early ETI may not improve IHCA outcomes.[6] Although the included samples were slightly different, these findings are similar, showing that early ETI may not improve resuscitation outcomes. Confounding factors, unaccounted for in the analysis, might contribute to negative outcomes. Patients in the early ETI group may have had severe underlying conditions impacting the ROSC. The unusual discrepancy between ROSC and 24-hour survival outcomes could be influenced by the causes of EDCA, post-ED management, or sample size. The lack of comprehensive data on EDCA causes and post-ED interventions, along with sample size limitations, calls for cautious interpretation.

To our knowledge, this is the first article investigating the relationship between the timing of ETI in EDCA patients using the time-to-event analysis. This study adds to the growing literature supporting the idea that early ETI might not improve resuscitation outcomes, especially in the ED. However, strong recommendations may not be made due to the study design and sample size. In the meantime, physicians should prioritize providing high-quality CPR and use clinical judgment to determine the most appropriate airway management technique for individual patients.

Our study has several limitations. The single-center nature of the study limits generalizability, and excluding non-intubated CPR cases introduces selection bias. The absence of a control group hinders the assessment of the relative effects of early ETI versus alternative techniques. Unaddressed confounders, such as comorbidities and CPR quality, further complicate outcome interpretation. The timing of the study during the COVID-19 pandemic adds complexity, although no positive cases were identified. However, the long-term effects, survival to discharge, and neurological function remain unexplored.

In conclusion, early ETI was not associated with the rate of ROSC following EDCA, but was associated with improved 24-hour survival in EDCA. The timing of ETI might not be a crucial factor in determining patient outcomes in this setting. However, given its retrospective design, a prospective, well-controlled trial is imperative to provide definitive insights into this critical issue.