Analgesia and Sedation Use During Noninvasive Ventilation for Acute Respiratory Failure

Peter J Dunbar; Ryan Peterson; Max McGrath; Raymond Pomponio; Tyree H Kiser; P Michael Ho; R William Vandivier; Ellen L Burnham; Marc Moss; Peter D Sottile; Colorado Pulmonary Outcomes Research Group

doi:10.1097/CCM.0000000000006253

Analgesia and Sedation Use During Noninvasive Ventilation for Acute Respiratory Failure

Crit Care Med. 2024 Mar 20. doi: 10.1097/CCM.0000000000006253. Online ahead of print.

Affiliations

¹ Division of Pulmonary Sciences and Critical Care Medicine, Department of Medicine, University of Colorado, School of Medicine, Aurora, CO.
² Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado, Aurora, CO.
³ Department of Clinical Pharmacy, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences, Aurora, CO.
⁴ Division of Cardiology, Department of Medicine, University of Colorado School of Medicine, Aurora, CO.

PMID: 38506571
DOI: 10.1097/CCM.0000000000006253

Abstract

Objectives: To describe U.S. practice regarding administration of sedation and analgesia to patients on noninvasive ventilation (NIV) for acute respiratory failure (ARF) and to determine the association of this practice with odds of intubation or death.

Design: A retrospective multicenter cohort study.

Setting: A total of 1017 hospitals contributed data between January 2010 and September 2020 to the Premier Healthcare Database, a nationally representative healthcare database in the United States.

Patients: Adult (≥ 18 yr) patients admitted to U.S. hospitals requiring NIV for ARF.

Interventions: None.

Measurements and main results: We identified 433,357 patients on NIV of whom (26.7% [95% CI] 26.3%-27.0%) received sedation or analgesia. A total of 50,589 patients (11.7%) received opioids only, 40,646 (9.4%) received benzodiazepines only, 20,146 (4.6%) received opioids and benzodiazepines, 1.573 (0.4%) received dexmedetomidine only, and 2,639 (0.6%) received dexmedetomidine in addition to opioid and/or benzodiazepine. Of 433,357 patients receiving NIV, 50,413 (11.6%; 95% CI, 11.5-11.7%) patients underwent invasive mechanical ventilation on hospital days 2-5 or died on hospital days 2-30. Intubation was used in 32,301 patients (7.4%; 95% CI, 7.3-7.6%). Further, death occurred in 24,140 (5.6%; 95% CI, 5.5-5.7%). In multivariable analysis adjusting for relevant covariates, receipt of any medication studied was associated with increased odds of intubation or death. In inverse probability weighting, receipt of any study medication was also associated with increased odds of intubation or death (average treatment effect odds ratio 1.38; 95% CI, 1.35-1.40).

Conclusions: The use of sedation and analgesia during NIV is common. Medication exposure was associated with increased odds of intubation or death. Further investigation is needed to confirm this finding and determine whether any subpopulations are especially harmed by this practice.