By Samuel Nadler, MD, PhD

Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University of Washington, Seattle

Dr. Nadler reports no financial relationships relevant to this field of study.

SYNOPSIS: Delaying intubation until aspiration or cardiopulmonary decompensation did not affect mortality but increased the incidence of pneumonia and length of stay.

SOURCE: Stewart R, Perez R, Musial B, et al. Outcomes of patients with alcohol withdrawal syndrome treated with high-dose sedatives and deferred intubation. Annals Am Thorac Soc 2016;13:248-252.

Alcohol withdrawal is a very common cause of hospital admission and complicates many otherwise routine hospitalizations. Catecholamine storm and agitation that requires sedative medication administration characterize this syndrome. However, both the underlying withdrawal and sedative treatments may precipitate aspiration and cardiopulmonary compromise, necessitating endotracheal intubation. The timing of this inherently risky procedure often is based on clinicians’ experiences alone. Both premature and delayed intubation may have significant effects on patient outcomes.

This study examined whether delaying intubation led to worsening outcomes. This was a single center, observational, cohort study of 188 patients admitted between 2008 and 2012 with alcohol withdrawal. Per protocol, all patients received continuous infusions of lorazepam (up to 1.2 mg/hour) then titrated to a Clinical Institute Withdrawal Assessment score of 6 or less. Patients could be admitted to either the ICU or floor-level care. The decision to intubate was deferred until clinically apparent aspiration or cardiopulmonary decompensation occurred. Patients were overwhelmingly male (92.6%) with a median age of 50.8 ± 9 years. Most were admitted to the ward (76.1%) with a mean Acute Physiology and Chronic Health Evaluation II (APACHE II) score of 6.2 ± 3.4. In this cohort, 12.8% experienced seizures, 16% were diagnosed with pneumonia, and the mean length of stay was 9.6 ± 11.7 days.

Several variables were statistically different in intubated vs. non-intubated patients. Intubated patients had higher admission APACHE II scores (7.4 vs. 5.9; P = 0.01), a higher rate of cirrhosis (10.5% vs. 2.7%; P = 0.05), and congestive heart failure (13.2% vs. 1.3%; P = 0.004). Interestingly, there was not a statistically significant difference in the history of delirium tremens, ICU admission, or seizure disorder between the two groups. Although there was no difference in overall mortality in intubated vs. non-intubated patients (2.6% vs. 0%), the rate of pneumonia was significantly higher in intubated patients (55.3% vs. 6%; P < 0.0001) as was length of stay (14.7 vs. 6 days; P < 0.0001). As would be expected, intubated patients required much higher total doses of benzodiazepines (761 vs. 229 mg lorazepam equivalents; P < 0.001) and daily doses of benzodiazepines (64.9 vs. 41.7 mg lorazepam equivalents; P = 0.01).


The authors concluded that deferring endotracheal intubation was not associated with excess morbidity or mortality. However, the data reported here paint a more complicated picture. Clearly, if earlier intubation could have prevented aspiration and the development of pneumonia, there would have been a benefit. This study showed certain variables retrospectively were associated with intubation, but these may not prospectively help clinicians understand the optimal timing of intubation. In the study, intubation was associated with pneumonia (odds radio [OR] = 23.54; 95% confidence interval [CI], 7.97-69.46; P < 0.001) and APACHE II > 10 (OR = 5.26; 95% CI, 1.79-15.5; P = 0.003). Length of stay was significantly longer in intubated patients. Age > 60 years, average benzodiazepine dose > 50 mg, and seizures did not have statistically significant ORs. It is unclear if this is intrinsic to intubation or related to delayed intubation leading to a more complicated hospitalization.

Although no statistical difference occurred for mortality, this study was not powered to detect such a change. Other studies of alcohol withdrawal show similar or higher mortality rates.1 This study adds to the knowledge base regarding alcohol withdrawal. It remains a prevalent cause of hospital admission and develops during many routine hospital admissions. Treatment of alcohol withdrawal via a protocol with infusions of benzodiazepines even up to 20 mg/hour without intubation until aspiration or cardiopulmonary compromise did not show excess mortality when compared with other studies. However, it is debatable whether there is not excess morbidity in waiting until aspiration occurs before intubation. A prospective, randomized trial of this protocol would better address this question. Furthermore, understanding prospectively the factors associated with intubation may enable higher vigilance in these patients and prompt earlier intubation that may reduce excess morbidity shown in this trial.


  1. Monte R, Rabunal R, Cadariego E, et al. Analysis of the factors determining survival of alcoholic withdrawal syndrome patients in a general hospital. Alcohol Alcohol 2010;45:151-158.