Aspiration pneumonitis in an overdose population: frequency, predictors, and outcomes

Crit Care Med. 2004 Jan;32(1):88-93. doi: 10.1097/01.CCM.0000104207.42729.E4.


Objective: To characterize the frequency of aspiration pneumonitis in an unselected population of overdose patients and, further, to identify factors that predispose to aspiration pneumonitis and the outcomes of patients with aspiration pneumonitis compared with those without.

Design: Retrospective cohort study.

Setting: Toxicology unit of a tertiary referral hospital.

Patients: All poisoning admissions.

Measurements and main results: A total of 71 of 4,562 poisoning admissions to the Hunter Area Toxicology Service between January 1997 and October 2002 had definite aspiration pneumonitis (1.6%; 95% confidence interval, 1.2-2.0). Older age, Glasgow Coma Score of <15, spontaneous emesis, seizures, delayed presentation to hospital, and ingestion of tricyclic antidepressants were associated with an increased risk of aspiration pneumonitis. Paracetamol poisoning and female sex were associated with a decreased risk of aspiration pneumonitis with univariate analysis. Ingestion of alcohol, benzodiazepines, antipsychotics, and administration of activated charcoal were not associated with aspiration pneumonitis. A logistic regression model for predicting aspiration pneumonitis contained seven predictors: age, sex, Glasgow Coma Score of <15 (odds ratio, 3.14; 95% confidence interval, 1.87-5.27), emesis (odds ratio, 4.17; 95% confidence interval, 2.44-7.13), seizure, tricyclic antidepressant ingestion, and time from ingestion to presentation (delay of >24 hrs [odds ratio, 4.42; 95% confidence interval, 2.42-8.10]). The mortality for patients with aspiration pneumonitis was 8.5% compared with 0.4% for those without (odds ratio, 23; 95% confidence interval, 9-60; p <.0001), and they had a significantly higher intensive care unit admission rate. The median length of stay of patients with aspiration pneumonitis was 126 hrs (interquartile range, 62-210 hrs) compared with 14.7 hrs (interquartile range, 7-23 hrs) in patients without (p <.0001).

Conclusions: Our study has shown a number of risk factors in overdose patients that are associated with aspiration pneumonitis that may allow the early identification of these patients for appropriate observation and management. Patients with aspiration pneumonitis have a significantly increased mortality and length of stay in the hospital.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Age Distribution
  • Aged
  • Analysis of Variance
  • Australia / epidemiology
  • Cohort Studies
  • Drug Overdose / complications
  • Drug Overdose / diagnosis
  • Female
  • Glasgow Coma Scale
  • Humans
  • Intensive Care Units
  • Logistic Models
  • Male
  • Middle Aged
  • Pneumonia, Aspiration / epidemiology*
  • Pneumonia, Aspiration / etiology
  • Pneumonia, Aspiration / therapy*
  • Predictive Value of Tests
  • Prevalence
  • Probability
  • Registries
  • Retrospective Studies
  • Severity of Illness Index
  • Sex Distribution
  • Survival Analysis