Rebecca Piland, Andrea D Hill, Lisa Burry, David N Juurlink, Nicholas A Bosch, Bijan Teja, Henry T Stelfox, Theodore J Iwashyna, Allan Walkey, Hannah Wunsch
Across 20 ICUs in Alberta, Canada, there were temporal changes and significant inter-ICU variability in opioid-prescribing practices for patients receiving mechanical ventilation. Further research is warranted to explore drivers of such substantial inter-ICU variability and associations between opioid selection and clinical outcomes to support evidence-based guidelines.
PURPOSE: Limited evidence exists to inform opioid selection in patients in the intensive care unit (ICU) receiving mechanical ventilation, and current guidelines treat opioids as interchangeable. We sought to describe temporal trends in, and factors associated with, the choice of first opioid infusion in critically ill patients.
METHODS: We conducted a historical cohort study of patients receiving mechanical ventilation (MV) in ICUs in Alberta, Canada, 2013-2019. We included patients who received a first continuous infusion of fentanyl, hydromorphone, or morphine. We assessed temporal trends and used a multilevel multinomial logistic regression model to identify patient-level factors associated with opioid choice, as well as ICU-level variability using the adjusted median odds ratio (aMOR).
RESULTS: Among 17,898 admissions in 20 ICUs who met inclusion criteria, fentanyl was the most frequent initial opioid infusion (78.6%; n = 14,060), followed by hydromorphone (16.3%, n = 2,917) and morphine (5.1%; n = 921). Over time, fentanyl use decreased (89.0% [n = 1,468/1,650] in 2013; 75.0% in 2019 [n = 1,763/2,351]; P < 0.001), as did morphine use (7.7% [n = 127/1,650] to 3.2% [n = 75/2,351]; P < 0.001), while hydromorphone use increased (3.3% [n = 55/1,650] to 21.8% [n = 513/2,351]; P < 0.001). Surgical patients were more likely to receive hydromorphone than fentanyl (adjusted odds ratio [aOR], 1.52; 95% confidence interval [CI], 1.32 to 1.74) and less likely to receive morphine (aOR, 0.81; 95% CI, 0.68 to 0.96). We observed substantial inter-ICU variability in opioid use: adjusted median odds ratio (aMOR), 14.5 (95% CI, 7.3 to 25.7) for hydromorphone compared with fentanyl; aMOR, 5.2 (95% CI, 3.11 to 9.2) for morphine.
CONCLUSIONS: Across 20 ICUs in Alberta, Canada, there were temporal changes and significant inter-ICU variability in opioid-prescribing practices for patients receiving mechanical ventilation. Further research is warranted to explore drivers of such substantial inter-ICU variability and associations between opioid selection and clinical outcomes to support evidence-based guidelines.