Monitoring rectal temperatures (RTs) in orally intubated, critically ill patients is common practice, as they are generally believed to be more accurate than oral temperatures (OTs). This study was designed to compare OTs of critically ill patients, with and without an endotracheal tube, using patients as their own control. Axillary (AT), tympanic membrane (TMT) and RTs of patients were also studied. Sixty-five orally intubated patients ready for extubation, ranging in age from 17 to 96 years (17 female and 48 male), were included in the study. OT, AT, RT and TMTs were taken before and after extubation. Using a t-test for paired samples, significant differences were found between both ATs and OTs measured with and without an oral endotracheal tube (P < 0.01); the differences (-0.12 and 0.08 degrees C, respectively) were not considered clinically important. Further t-tests for paired samples revealed no significant difference between TMTs and RTs pre and post extubation. Respiratory rate, presence or absence of teeth, and mouth position were shown to have no effect on OT (t-tests, P < 0.01). Analysis of variance showed no difference in OT dependent on room temperature or temperature within the endotracheal tube. It was concluded that OTs are accurate when taken in patients who are orally intubated and ready for extubation.