Age, duration of mechanical ventilation, and outcomes of patients who are critically ill

Chest. 2009 Sep;136(3):759-764. doi: 10.1378/chest.09-0515.


Background: Age and duration of mechanical ventilation (MV) are strongly associated with mortality and hospital discharge disposition.

Methods: Electronic administrative records from a 425-bed community teaching hospital were obtained for 9,912 patients who were admitted to hospital ICUs between 2003 and 2008. Risk estimates of age and duration of MV for in-hospital mortality and discharge to home vs extended-care facilities (ECFs) also were obtained.

Results: Of 9,912 patients, 37 were discharged to hospice care, and 668 were < 18 years of age. Of the remaining 9,207 patients, 4,238 received invasive MV. Mortality or hospital discharge to ECFs increased consistently for each decade of age > 65 years and as the duration of MV increased. Although only 11.7% of patients < 65 years age who received MV for 1 or 2 days died during hospitalization, the mortality rate increased to 72.1% for patients > 85 years of age who had received MV for > 7 days. For patients requiring MV for >or= 7 days, < 10% of the >or= 65 years of age and < 5% of patients >or= 85 years of age survived to be discharged home from the hospital. Multivariate logistic regression analyses showed that age > 65 years and duration of MV remained significantly associated with outcomes, even after adjustment for hospital discharge diagnoses (Charlson scores).

Conclusions: This study suggests that age and duration of MV are strongly associated with mortality and posthospital disposition. If confirmed, the simple combination of age and duration of MV provides prognostic information that could be used with trajectory of illness and in the context of patients' values to inform end-of-life discussions with patients or their surrogates during a trial of critical care.

MeSH terms

  • Adolescent
  • Adult
  • Age Factors
  • Aged
  • Aged, 80 and over
  • Chi-Square Distribution
  • Comorbidity
  • Critical Illness / mortality*
  • Female
  • Hospital Mortality*
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Patient Discharge / statistics & numerical data
  • Prognosis
  • Respiration, Artificial / mortality*
  • Statistics, Nonparametric
  • Time Factors