by
Shehabi, Yahya; Bellomo, Rinaldo; Kadiman, Suhaini; Ti, Lian Kah; Howe,
Belinda; Reade, Michael C.; Khoo, Tien Meng; Alias, Anita; Wong, Yu-Lin;
Mukhopadhyay, Amartya; McArthur, Colin; Seppelt, Ian; Webb, Steven A.; Green,
Maja; Bailey, Michael J.; for the Sedation Practice in Intensive Care
Evaluation (SPICE) Study Investigators and the Australian and New Zealand
Intensive Care Society Clinical
Objectives:
In the absence of a universal definition of light or deep sedation, the level
of sedation that conveys favorable outcomes is unknown. We quantified the
relationship between escalating intensity of sedation in the first 48 hours of
mechanical ventilation and 180-day survival, time to extubation, and delirium.
Design: Harmonized data from prospective multicenter international longitudinal
cohort studies Setting: Diverse mix of ICUs. Patients: Critically ill patients expected
to be ventilated for longer than 24 hours. Interventions: Richmond Agitation
Sedation Scale and pain were assessed every 4 hours. Delirium and mobilization
were assessed daily using the Confusion Assessment Method of ICU and a
standardized mobility assessment, respectively. Measurements and Main Results:
Sedation intensity was assessed using a Sedation Index, calculated as the sum
of negative Richmond Agitation Sedation Scale measurements divided by the total
number of assessments. We used multivariable Cox proportional hazard models to
adjust for relevant covariates. We performed subgroup and sensitivity analysis
accounting for immortal time bias using the same variables within 120 and 168
hours. The main outcome was 180-day survival. We assessed 703 patients in 42
ICUs with a mean (SD) Acute Physiology and Chronic Health Evaluation II score
of 22.2 (8.5) with 180-day mortality of 32.3% (227). The median (interquartile
range) ventilation time was 4.54 days (2.47–8.43 d). Delirium occurred in 273 (38.8%)
of patients. Sedation intensity, in an escalating dose-dependent relationship,
independently predicted increased risk of death (hazard ratio [95% CI], 1.29
[1.15–1.46]; p < 0.001, delirium hazard ratio [95% CI], 1.25 [1.10–1.43]), p
value equals to 0.001 and reduced chance of early extubation hazard ratio (95%
CI) 0.80 (0.73–0.87), p value of less than 0.001. Agitation level independently
predicted subsequent delirium hazard ratio [95% CI], of 1.25 (1.04–1.49), p
value equals to 0.02. Delirium or mobilization episodes within 168 hours,
adjusted for sedation intensity, were not associated with survival.
Conclusions: Sedation intensity independently, in an ascending relationship,
predicted increased risk of death, delirium, and delayed time to extubation.
These observations suggest that keeping sedation level equivalent to a Richmond
Agitation Sedation Scale 0 is a clinically desirable goal.
