Memory in relation to depth of sedation in adult mechanically ventilated intensive care patients

Objective: To investigate the relationship between memory and intensive care sedation. Design and setting: Prospective cohort study over 18 months in two general intensive care units (ICUs) in district university hospitals. Patients: 313 intubated mechanically ventilated adults admitted for more than 24 h, 250 of whom completed the study. Measurements: Patients (n = 250) were interviewed in the ward 5 days after discharge from the ICU using the ICU Memory Tool. Patient characteristics, doses of sedative and analgesic agents, and sedation scores as measured by the Motor Activity Assessment Scale (MAAS) were collected from hospital records after the interview. Results: Patients with no recall (18%) were significantly older, had higher baseline severity of illness, and experienced fewer periods of wakefulness (median proportion of MAAS score 3; 0.37 vs. 0.70) than those who had memories of the ICU (82%). Multivariate analyses showed that increasing proportion of MAAS 0-2 and older age were significantly associated with having no recall. Patients with delusional memories (34%) had significantly longer ICU stay (median 6.6 vs. 2.2 days), higher baseline severity of illness, higher proportions of MAAS scores 4-6, and more administration of midazolam than those with recall of the ICU without delusional memories. Conclusions: Heavy sedation increases the risk of having no recall, and longer ICU stay increases the risk of delusional memories. The depth of sedation during total ICU stay as recorded with the MAAS may predict the probability of having memories of the ICU.