Adherence to Sedative Weaning Opportunities and Length of Stay for Critically Ill Children

Importance: Prolonged sedative exposure in pediatric intensive care units (ICUs) is associated with adverse outcomes including delirium, withdrawal, and neurodevelopmental impairment. Clinical guidelines recommend timely weaning, but adherence to weaning opportunities and its association with clinical outcomes is unknown. Objectives: To evaluate adherence to sedative weaning opportunities and examine the association between weaning adherence and clinical outcomes among pediatric ICU patients. Design, Setting, and Participants: This multicenter cohort study included ICU patients at 4 pediatric academic medical centers admitted between January 1, 2011, and December 31, 2024, with greater than 48 hours of cumulative sedative infusion exposure. Data extraction and analyses were performed between February and November 2025. Neonatal ICU stays were excluded. Exposure: Weaning adherence, defined as the percentage of identified opportunities to wean from sedation that resulted in weaning actions (dose decreases or discontinuations), calculated at the ICU encounter level. An opportunity was defined as 0 or 1 as-needed doses of any sedative in a 12-hour period. Main Outcomes and Measures: The primary outcomes were ICU and hospital length of stay. Secondary outcomes included infusion duration and cumulative sedative dose of each drug class. Multivariable linear regression models adjusted for center, age, and sex examined associations between weaning adherence and outcomes. Results: The cohort included 15 747 ICU encounters from 12 737 patients (median age, 1.1 years [IQR, 0.2-6.5 years]; 56% of encounters among male patients). Median ICU length of stay was 13 days (IQR, 7-25 days) and median hospital length of stay was 26 days (IQR, 14-55 days). Among a median of 30 (IQR, 13-68) opportunities to wean per ICU encounter, overall adherence to weaning opportunities was 28% (IQR, 14%-49%), while missed weaning opportunities were common (median, 72%; IQR 51%-86%). Higher weaning adherence was associated with shorter median ICU length of stay (-2.3 days [IQR, -2.4 to -2.2 days] per 10% increase; P < .001) and shorter median total hospital length of stay (-3.6 days [IQR, -4.0 to -3.4 days] per 10% increase; P < .001). Higher weaning adherence was associated with shorter median sedative infusion duration for all sedative classes combined (-1.1 days [IQR, -1.2 to -1.0 days] per 10% increase in adherence; P < .001) and lower median cumulative exposure to opioids (-3.5 mg/kg [IQR, -3.8 to -3.2 mg/kg]; P < .001), benzodiazepines (-3.4 mg/kg [IQR, -3.8 to -3.0 mg/kg]; P < .001), and α2 agonists (-27.7 µg/kg [IQR, -30.2 to -25.3 µg/kg]; P < .001). Conclusions and Relevance: In this multicenter cohort, adherence to sedative weaning opportunities was low. Higher weaning adherence was independently associated with shorter ICU and hospital lengths of stay and reduced sedative exposure. Improving adherence to sedative weaning opportunities may represent a modifiable target for reducing length of stay and sedative exposure.