Background: Unplanned extubation (self-extubation or accidental extuba
tion) occurs commonly in mechanically ventilated patients, and many pa
tients do not receive mechanical ventilation indefinitely. Unfortunate
ly, weaning parameters are often unavailable in the setting of unplann
ed extubation, and it would be useful to define pre-extubation respira
tory and ventilatory parameters that predict which patients require re
intubation. Methods: The medical records of all patients who experienc
ed unplanned extubation for the 2-year period of July 1989 to July 199
1 were reviewed. Pre-extubation values of respiratory rate, tidal volu
me (VT), fraction Of inspired oxygen (FIO2), PEEP, ventilatory mode, a
nd ventilator-delivered minute volume (VVE, ventilator rate multiplied
by set VT) were recorded. In addition, the following data were obtain
ed: age, gender, respiratory failure diagnosis, duration of intubation
, amount, and type of sedative agents in the 24 h before extubation. C
omparisons of these values among patients who ultimately required rein
tubation and those who were not reintubated were made using the Mann-W
hitney U two-sample test. Results: During this period, there were 23 u
nplanned extubations involving 22 patients. Reintubation was required
for 18 episodes of unplanned extubation, but was not required for 5 ep
isodes. There were no significant differences between the two groups f
or any of the parameters except VVE and FIO2. The mean pre-extubation
FIO2 of the reintubated group (0.49) was significantly higher than tha
t of the patients who were not reintubated (0.35) (p=0.021); all of th
e patients who remained extubated were receiving an FIO2 less than or
equal to 0.40. The VVE was also higher in the reintubated group (9.73
L/min) than in the patients who were not reintubated (1.40 L/min); all
patients who remained extubated were receiving less than or equal to
7.0 L/min of ventilator-delivered minute ventilation. Conclusions: Rei
ntubation after unplanned extubation should not be considered mandator
y. Patients who require reintubation have significantly higher preextu
bation FIO2 and ventilatory requirements than patients who remain extu
bated.