In the intensive care units (ICUs) sedation strategies have changed in the past decade towards less sedation and daily wake-up calls. Recent studies indicate that no sedation (after intubation) is most beneficial for patients. A smaller number of these patients have been assessed for post-traumatic stress disorder (PTSD) after ICU discharge, but none of them were assessed for delirium while in the ICU. In other studies, delirium in the ICU is described as distressing for the patients and increasing morbidity, i.e. dementia after discharge and mortality. The associations between sedation, delirium, and PTSD have not previous been described. The aim of this PhD study was to investigate: 1) how sedation is associated with delirium in the ICU, 2) the consequences of delirium in relation to PTSD, anxiety, and depression, 3) the consequences of delirium for the patients' memories from ICU and the health-related quality of life after discharge. In a prospective observation study with patients admitted a minimum of 48 hours to the ICUs in Aarhus or Hillerød, we included all patients aged > 17 years. Non-Danish-speaking, patients transferred from other ICUs and patients with brain injury that made delirium-assessment impossible were excluded. Patients were interviewed face-to-face after 1 week, and at 2 months and 6 months by telephone using six different questionnaires. Among 3,066 patients admitted to the ICUs, 942 fulfilled the inclusion criteria. Primarily due to the inability to test for delirium, 302 patients were later excluded. Of the remaining 640 patients, 65% were delirious on 1 or more days. Fluctuations in sedation levels increased the risk of delirium statistically significantly with or without adjustments for age, gender, severity of illness, surgical/medical patient, or ICU site. After 2 months vs. 6 months, 297 patients vs. 248 patients were interviewed. PTSD was found in 7% vs. 5%, anxiety in 6% vs. 4%, and depression in 10% at both interviews. Delirium had no association with any of the psychometric results. Memories of delusion and memories of feelings were statistically significantly associated with delirium and with the psychometric outcomes, whereas memories of facts had no association with the psychometric outcomes. Health-related quality of life (SF-36) was statistically significantly decreased in most of the domains if patients had PTSD, anxiety, or depression but was not associated with delirium or the type of memories.
Conclusion: Fluctuations in the level of sedation of patients in the ICU increased the incidence of delirium, but the delirium did not affect the risk of PTSD, anxiety, or depression. These were, however, affected by the type of memories the patients had. Health-related quality of life (SF-36) was decreased if patients had PTSD, anxiety, or depression but was unaffected by memories of the ICU and the presence of delirium while in the ICU.