Objective: To assess the factual and delusional memories reported by intensive care unit survivors and its relationship with the development of Posttraumatic Stress Syndrome (PTSS).
Design: Multicenter observational cohort study.
Setting: Nine Portuguese intensive care units, as part of a multicenter study.
Methods and patients: Between January and June 2005, 1,174 patients were admitted across the nine intensive care units. Two hundred thirty-nine patients were excluded, 14 with < 18 yrs old and 225 with a length of intensive care stay < or = 48 hrs. Thus a total of 935 patients were included in the study. One hundred ninety (20%) patients died in the intensive care unit, 90 (12%) patients died on the ward (30% in-hospital mortality rate), and another 56 (9%) died in the next 6 months after intensive care unit discharge.
Results: From the 599 survivors at 6 months, 313 patients answered the questionnaires (52% response rate). From the 313 respondents, 58% (n = 183) were men, median age was 59. The median Simplified Acute Physiology Score II was 37, median intensive care unit length of stay was 8 days, 57% (n = 177) of the patients were admitted for medical reasons. Forty percent (n = 116) of the respondents did not remember their admission to hospital, 48% (n = 142) did not remember the time in the hospital before intensive care unit admission, 73% (n = 220) had factual memories and 39% (n = 118) had delusional memories. Twenty-three percent (n = 66) stated that they had had intrusive memories. A higher number of "adverse" experiences were significantly associated with a higher PTSS-14 score. Eighteen percent (n = 54) of patients had a PTSS-14 score > 49, indicating a higher risk of developing posttraumatic stress disorder. A PTSS-14 score > 49 was significantly associated with not remembering the hospital stay before intensive care unit admission.
Conclusion: Amnesia for the early period of critical illness (early amnesia) was positively associated with the level of posttraumatic stress disorder-related symptoms, which may be a proxy for severity of disease at the time of intensive care unit admission.