The purpose of the study was to compare three different methods of assessing the core temperature: by measuring the bladder temperature, the tympanic temperature, (Braun Thermoscan 3000) and the forehead skin temperature (Philips SensorTouch). Measuring of the bladder temperature and the tympanic temperature are wellknown methods whereas measuring of the forehead skin temperature by the use of Sensor Touch is a fairly new method by which the temperature at the warmest area of the forehead is measured.Forty-two randomly chosen adult patients who all had a bladder catheter with a thermometer and a normal urine output. The patients were admitted to an intensive care unit and none were intubated nasally. Otoscopy was performed prior to temperature measurement. The temperatures were measured almost simultaneously. At the same time a white blood cell count and the C-reactive-protein value was recorded.There was a good linear coherence between the bladder temperature and the tympanic temperature (r = 0.97). The linear coherence between the bladder temperature and the forehead skin temperature was poor (r = 0.59). We found a mean deviation of 0.07 degree C and a standard deviation of 0.3 degree C of the difference between the bladder temperature and the tympanic temperature (p = 0.19). The mean of the difference between the bladder temperature and the forehead skin temperature was 0.5 degree C and a standard deviation of 0.8 degree C (p = 0.0003). Furthermore the forehead skin thermometer was not able to measure temperatures < 35.0 degrees C. We found no statistically significant coherence between the core temperature and the white blood cell count or C-reactive-protein.Tympanic temperature was well coherent with bladder temperature--forehead skin temperature was poorly coherent with bladder temperature. By evaluating the core temperature in intensive care patients the tympanic temperature is a reliable alternative to the bladder temperature.