Skin conductance variability and stressful exposures in critical care

Abstract: Numerous stressors affect the intensive care unit (ICU) patient. The ICU environment is busy and noisy, with monitoring and treatment around the clock. ICU patients may have problems in getting some sleep, due to lighting and noise. Many ICU survivors report having pain during their ICU stay. Stressful experiences from the ICU contribute to later development of posttraumatic stress symptoms. The comfort and well-being of ICU patients is today an important goal and methods and strategies to achieve this are developing. In this thesis, two areas related to ICU patient exposures were evaluated: noise and pain. A method of monitoring skin conductance variability (SCV) was evaluated. In Paper I, we compared sound pressure levels in three different ICU room types and three different shifts, as well as analysed the sources of disruptive sounds in the different room types. We found that sound pressure levels were similar in the different room types, with a trend towards lower night-time levels. Levels were well above international recommendations. Disruptive sounds were more frequent in three-bed rooms than in singlebed rooms. Main disruptive ICU sounds were from machine alarms and from noisy chatting. In Paper II, we studied SCV as a measure of pain in 40 ICU patients. Increasing levels of stimulation was associated with elevation of SCV. In non-intubated patients, there was an interaction effect between pain and agitation on SCV. In Paper III, we monitored 18 volunteers with SCV and exposed them to a standardised pain stimulus, to pictures with varying emotional content and to an ICU sound recording, in random combinations. SCV was significantly elevated by pain stimulation and to a lesser extent affected by emotion-inducing pictures or ICU sound. In Paper IV, 30 recently discharged ICU patients were monitored with SCV and exposed to the same ICU sound recording as in study III. During SCV monitoring, patients were also asked questions regarding traumatic experiences from the ICU. SCV was significantly elevated in most patients in response to both ICU sounds and questions. There was, however no correlation with stress symptoms assessed with a specific questionnaire for ICU survivors. In conclusion, the studies of this thesis show that a) sound levels preclude normal sleep and can potentially be modifying machine alarms and behaviour b) skin conductance variability may be difficult to interpret in awake patients but potentially has a room for monitoring pain in poorly communicable patients. Further studies in poorly communicable ICU patients during interventions may further elucidate the role of such monitoring