[Methohexital for analgosedation of ventilated intensive care patients : prospective nonrandomized single center observational study on incidence of delirium]

Delirium is defined by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM-IV-TR) as a potentially reversible disturbance of consciousness and a change of cognition caused by a medical condition, drug intoxication, or medication side effect. Delirium affects up to 80 % of intensive care unit (ICU) patients and is associated with increased morbidity and mortality. One risk factor for development of delirium in ventilated intensive care unit patients is sedation. The German S3 guidelines on "Analgesie, Sedierung und Delirmanagement in der Intensivmedizin" (analgesia, sedation and delirium management in intensive care medicine) of the DGAI (German Society for Anesthesiology and Intensive Care Medicine) and the DIVI (German Interdisciplinary Association for Intensive Care and Emergency Medicine) recommend midazolam and propofol for sedation, although both drugs are associated with a high incidence of delirium.

Within the framework of this study the question arose whether the barbiturate methohexital could be associated with a lower incidence of delirium in comparison to midazolam or propofol in analgosedated and ventilated ICU patients.

This was a prospective nonrandomized observational cohort study in a mixed medical surgical intensive care unit. Patients ventilated within 72 h after admittance were consecutively allocated to either propofol/remifentanil versus methohexital/remifentanil (expected ventilation duration ≤ 7 days) or midazolam/fentanyl versus methohexital/fentanyl (expected ventilation duration > 7 days) by the attending senior consultant anesthetist at the time of admission and/or intubation. Primary endpoint was delirium at any time during the ICU stay. Delirium was checked every 8 h by ICU nurses using the intensive care delirium screening checklist (ICDSC), with delirium defined as ICDSC ≥ 4 points. Before evaluation of the ICDSC the Richmond agitation sedation scale (RASS) score and the visual analogue scale for pain (VAS target ≤ 4) were measured. To assure reliable evaluation of the ICDSC, the RASS score of the patient at the time of evaluation had to be ≥ - 2. Assuming an incidence of delirium in the midazolam group of 70 % and in the methohexital group of 35 %, 16 patients were needed each in the midazolam/fentanyl and the methohexital/fentanyl cohorts (p = 0.05, β = 0.1). Assuming an incidence of delirium in the propofol group of 50 % and in the methohexital group again of 35 %, 94 patients were needed in the propofol/remifentanil and methohexital/remifentanil groups, respectively (p = 0.05, β = 0.1).

A total of 222 patients were evaluated, 34 in the methohexital vs. midazolam group and 188 in the methohexital vs. propofol group. Out of 16 patients sedated with midazolam, 15 developed delirium (94 %) in contrast to only 5 out of 18 patients sedated with methohexital (28 %). Thus compared to midazolam the sedation with methohexital reduced the incidence of delirium by 66 % (p < 0.001) corresponding to a number needed to treat (NNT) of 1.5. Out of 94 patients in the propofol/remifentanil group, 64 developed delirium (68 %) in contrast to only 23 out of 94 in the methohexital/remifentanil group (24 %). Thus compared to propofol the sedation with methohexital reduced the incidence of delirium by 44 % (p < 0.001), corresponding to an NNT of 2.5.

Sedation with methohexital compared to midazolam or propofol reduced the incidence of delirium by more than 50 % in ventilated ICU patients.