ABSTRACT 63

Abstract

Background and aims: Sedation and analgesia are priority issues in the management of critically ill children, however lacks strong evidence to guide practice. Aims: To identify PICU management practices of sedation, and their relationship with relevant clinical outcomes. Methods: We conducted a multicentre, prospective longitudinal cohort study of ventilated and sedated (>24hrs) children (n= 231). Patients were assessed for administration of sedatives, ventilation, sedation depth using State Behaviour Scale (SBS, q4h), delirium (daily), and PICU and hospital mortality. Early deep sedation was defined as a continuous score, whereby SBS was -2 to -3 at each q4h interval in the first 48 hours. Results: The median age was 2.75 months (IQR 0.40,14), with PICU and hospital mortality of 13 (5.6%) and 21/231 (9.1%). Deep sedation occurred in 62% (142/228) patients at first assessment and in 74.8% patients at 48 hours. Delirium occurred in 123 (55.2%) patients during their PICU admission. Midazolam was predominantly prescribed compared to Dexmedetomidine and Clonidine, given to 166 (74.4%) versus 71 (31.8%) and 63 (28.3%) patients, respectively. Early deep sedation was associated with length of mechanical ventilation (r=0.2, p=.002) and PICU length of stay (r=0.16, p=.08). Conclusions: Considerable variation in sedation practice continues across PICUs. Midazolam is the most commonly prescribed sedative, with individual centre and age influencing the first-line sedative choice. Early deep sedation is common and is associated with increased morbidity. Incidence of delirium appears consistent with adult ICU data, however must be interpreted with caution due to the subjective nature of assessments and insufficient validation in infants.