Effects of Supervision by Attending Anesthesiologists on Complications of Emergency Tracheal Intubation

Emergency intubations of adult patients outside the operating room and emergency department are performed by anesthesia residents from the surgical intensive care unit at the Massachusetts General Hospital, Boston, Massachusetts. These residents have a minimum of 6 months of anesthesia experience. They performed a mean of 220 (range, 162–344) intubations before they performed emergent intubations outside the operating room. A small group of attending anesthesiologists, all subspecialized in critical care medicine, were assigned to supervise these residents. The attending anesthesiologist was actively supporting the residents with these emergent intubations. A registered respiratory therapist assisted with airway management and ventilation whether the attending anesthesiologist was present or not. That respiratory therapist prospectively recorded data from emergent intubations from November 1, 2006, to July 31, 2007. These data included the reason for intubation, ease of intubation, identification of the first intubator, presence or absence of the attending anesthesiologist, medications used to facilitate intubation, and immediate complications. Data were recorded on a standardized data collection form. Based on our previous study,4these complications were predefined as esophageal intubation, traumatic intubation, aspiration, dental injury, and endobronchial intubation. Our initial data focused on the relation between attending physician supervision and complications. To allow for an analysis of potential confounders, we added prospectively collected data from an additional 172 emergent intubations (December 1, 2007, to April 15, 2008). The initial and the additional data sets were similar regarding patient characteristics and intubation procedure (table 1).

Age and sex were identified from the medical record. Acute Physiologic and Chronic Health Evaluation II score of surviving patients on admission to the intensive care unit,7body mass index, and American Society of Anesthesiologists physical status classification8were calculated. Patient outcomes were extracted from the medical records. The study was approved by the institutional review board of the Massachusetts General Hospital.

All data analysis was performed using Stata 10 (StataCorp LP, College Station, TX). Continuous variables with a normal distribution are expressed as mean ± SD. Ordinal variables are expressed as median and interquartile range. Baseline characteristics, intubation data, and outcomes of the two groups were compared by unpaired t  test for normally distributed variables and by Wilcoxon rank sum test for variables that were not normally distributed. The chi-square test was used to compare absolute numbers and proportions. The Spearman rank correlation coefficient was used to measure the strength of association between nonnormally distributed variables. P  values of baseline characteristics, intubation data, complications, and outcomes were calculated using two-sided analysis, and the level of significance was set at less than 0.05. A multiple logistic regression analysis was performed to control for potential confounding variables in the association between complications and attending supervision. Opioid and muscle relaxant use was chosen for the model because use of different medication has been clinically attributed to different outcomes9and because of differences in frequency of use in the initial data set.

Of the 322 patients enrolled in the study, 115 were intubated with attending anesthesiologist supervision, whereas 207 where intubated by anesthesia residents in the absence of an attending anesthesiologist. Approximately half (n = 176, 54.6%) were intubated in intensive care units. More than half were medical patients (n = 219, 66%), and the others were surgical patients (n = 55, 17%) or neurologic patients (n = 48, 15%). There was no difference between groups (with or without an attending intensivist) for sex, Acute Physiologic and Chronic Health Evaluation II score, American Society of Anesthesiologists physical status classification, body mass index, or patient type (medical, surgical, or neurologic). Patients intubated with attending presence were slightly younger and were more likely to be located in the intensive care unit rather then the ward compared with patients intubated without an attending present (table 2).

Of the 322 patients, 208 (65%) were intubated for respiratory emergency, 40 (12%) for cardiac arrest, 41 (13%) for airway protection, and 8 (2%) for acute neurologic events. More patients intubated for airway protection or for miscellaneous reasons were intubated with attending supervision (table 3). Our residents were instructed in emergent airway management as outlined in the Critical Care Handbook of the Massachusetts General Hospital .10All patients were intubated orally. Most intubators used a MacIntosh 3 blade. Nearly all patients (95%) were intubated on the first or second attempt. There was no difference in number of intubation attempts between patients intubated with and without attending supervision. The number of attempts was higher in patients who had complications during emergent intubation. Patients intubated without complications required 1 (range, 1–3) intubation attempt, and patients intubated with complications required 2 (range, 1–6) attempts (P = 0.001).

Hypnotic, opioid, and muscle relaxant use is shown in table 3. No sedative medications or muscle relaxants were administered during cardiopulmonary resuscitation. There was greater use of muscle relaxants (46% vs.  17%; P = 0.001) in patients intubated with attending physician supervision.

Of the 322 patients, 266 (83%) were intubated by first- or second-year anesthesia residents. The remaining patients were intubated by third-year anesthesia residents (n = 7, 2%), anesthesia attending physicians (n = 13, 4%), or other physicians (n = 35, 11%) in the presence of the anesthesia resident. There was no difference in complication rates among first- and second-year anesthesia residents (25.4 vs.  19.8%; P = 0.46). Table 4tabulates the complication rates according to intubator experience.

Supervision of residents by attending anesthesiologists was associated with a significant decrease in complication rate (6.1% vs.  21.7%; P = 0.0001). Complications of emergent intubations are summarized in table 5. Attending supervision was associated with a decreased complication rate in all categories. Of 13 patients who experienced aspiration, only 4 survived.

Because there was higher use of muscle relaxants (P = 0.0001) and a trend for higher use of narcotics (P = 0.12) in the patients supervised by an attending anesthesiologist that may have contributed to the observed differences in complications, we performed a multiple regression analysis. Attending supervision was associated with a decreased rate of complication (odds ratio, 0.52; 95% confidence interval, 0.27–0.99; P = 0.047), use of muscle relaxants did not affect complication rate (odds ratio, 0.66; 95% confidence interval, 0.33–1.33; P = 0.248), and use of opioids was associated with an increased rate of complications (odds ratio, 2.17; 95% confidence interval, 1.22–3.86; P = 0.009).

Of the 322 patients emergently intubated, 54% survived and 18% were discharged home. There was no difference in 28-day mortality, ventilator-free days, or patient disposition whether or not the patients were intubated with attending supervision (table 6).

The most important finding of our study is that supervision of emergent intubations by an attending anesthesiologist was associated with a decreased complication rate.

Most of the intubations were performed for respiratory emergencies. The most commonly used sedative was propofol. In other centers, etomidate5is the preferred hypnotic agent because of its favorable hemodynamic profile.11,12However, reports of adrenal insufficiency associated with the use of etomidate13may influence clinicians' choice. Opioids were used in 38% of our patients and were associated with an increased risk for complications. Although this is difficult to explain, opioids have been associated with an increased risk for aspiration and regurgitation.14Variable choices of hypnotic agents also reflect the lack of randomized controlled studies evaluating approaches to emergent airway management outside the operating room. Similarly, there is wide variability in the use of muscle relaxants for emergent intubation, ranging between 5%4and 80%.2The use of muscle relaxants can cause severe hypoxia if the trachea cannot be intubated and the patient cannot be ventilated. In our study, muscle relaxants were used more commonly in the presence of an attending anesthesiologist (46% vs.  17%; P = 0.001). Perhaps this reflects that attending physicians were more confident in their ability to intubate.

Complication rates for emergent intubations outside the operating room have remained high during the past decade.2,3,5,15In our study, the presence of an attending anesthesiologist decreased the complication rate significantly (6.1% vs.  21.7%). Similarly, multivariate analysis of an observational study of complications of emergent intubations of intensive care unit patients revealed that supervision by a senior intensivist may decrease the risk of complications.5It is unknown whether this decrease in complication rate was due to the experience of the supervising anesthesiologist or the result of additional help by the anesthesiologist. In our institution, the resident physician performing the intubation has assistance from an experienced senior respiratory therapist assisting with airway management.4Schwartz et al.  2reported that the experience of the physician performing the intubation did not influence the complication rate for emergent intubation. In contrast, Tayal et al.  16reported that success of intubation by emergency medicine residents increased with experience. In our study, the majority of intubations were performed by first- and second-year residents. Only a few patients (n = 20) were intubated by third-year anesthesia residents or attending physicians as first intubators. Residents for the first time involved in emergent airway management had an average experience of 220 intubations. This is higher than the published learning curves for orotracheal intubations, which report between 19 and 57 orotracheal intubations to reach good success.17–19There was no difference in complication rate between first-year anesthesia residents and the more experienced second-year residents. Based on this, we do not know how many emergent intubations residents need to perform to reach competency.

In the presence of an attending anesthesiologist, a significantly higher percentage of patients received muscle relaxants. The supervising attending anesthesiologist was actively involved in the care of the patient, bringing his expertise as a consultant to the bedside. This includes appropriate use of medications. The multiregression analysis revealed that attending presence was associated with a decreased rate of complications after controlling for the effects of muscle relaxants and opioids. In the emergency department, rapid sequence intubation has decreased the complication rate of emergency intubation and is considered safe.9,16,20In addition, the use of muscle relaxants was associated with a lower complication rate of tracheal intubation in critically ill patients.5To our knowledge, there is no published study testing the safety and efficacy of muscle relaxations for emergent intubations of hospitalized patients.

Our observed mortality of 46% is comparable to that in previous reports of patients requiring urgent and emergent intubation.4,21–23Despite a decrease in complications, there was no difference in mortality whether patients were intubated in the presence or absence of an attending anesthesiologist.

This was a single-center study in an academic tertiary care center, which limits the ability to generalize the findings to different settings. For example, emergent intubations in hospital wards in the United States are often performed by nonanesthesia care providers such as respiratory therapists.24 

The study design as a prospective observational study limits the ability to control variables of the study. The attending was actively involved in planning and performing these intubations and may have introduced different techniques and drugs that influenced the complication rate. Although this decreases the ability to detect advantages or disadvantages of different approaches to emergent airway management, it reflects the reality of emergent intubations.

In our study, the supervising attending physicians were anesthesia trained. It is not known whether similar results would have been achieved with different training backgrounds of the supervising attending physicians. Jaber et al.  5reported no difference in complications of emergent intubations whether the supervising attending physician was an anesthesiologist or not. Similarly, Bushra et al.  25reported no difference in success and complications whether emergent intubations were supervised by attending anesthesiologists or attending emergency physicians.

In our study, complications were limited to predefined categories that could be reliably charted by an observer who did not perform the intubation. We cannot exclude observer bias. However, there were more than 100 residents and respiratory therapists who took part in the study, and the respiratory therapists charted objective endpoints limiting potential bias. For practical reasons, following our previous study4we concentrated on four complications that our respiratory therapists have the expertise to report. Emergent intubations may lead to many more complications that may cause unfortunate outcomes.5,14,26The current study was not intended or powered to address outcomes of patients with complications. We can only assume but do not know that attending presence decreased other complications as well.

The number of patients enrolled (n = 322) was too small to detect differences in mortality between patients intubated in the presence or absence of an attending anesthesiologist. A large multicenter trial might be necessary to assess the impact on survival.

Emergent intubation is associated with a high complication rate. Supervision of anesthesia residents by an attending anesthesiologist is associated with a decreased incidence of complications during emergent intubations.