Background
Little is known about whether the choice of opioid influences long-term outcomes for critically ill patients.
Research Question
To determine whether initiation of IV morphine or hydromorphone during mechanical ventilation (MV) is associated with reduced opioid use after discharge relative to fentanyl.
Study Design and Methods
This was a retrospective cohort study of 14,197 veterans who underwent MV in 116 Veterans Administration hospitals (2014-2020) and who received fentanyl, morphine, or hydromorphone as the initial and only IV opioid during their first 2 days in the ICU. The primary outcome was persistent opioid use in the year after hospital discharge.
Results
Overall, 11,903 patients (83.8%) received fentanyl, 1,156 patients (8.1%) received morphine, and 1,138 patients (8.0%) received hydromorphone as the initial and only IV opioid during the first 2 days in the ICU. The median patient age was 67 years (interquartile range, 61-72 years). Persistent opioid use in the year after discharge was more common with hydromorphone (16.5%) vs fentanyl (12.0%; adjusted OR [aOR], 1.25; 95% CI, 1.00-1.56), but not with morphine (15.7%) vs fentanyl (aOR, 1.12; 95% CI, 0.91-1.39). Stratified by prior persistent opioid use, the association between opioid initially received in the ICU and an increased risk of persistent use in the following year was present only among individuals without this history for both morphine and hydromorphine compared with fentanyl (morphine: aOR, 1.44 [95% CI, 1.07-1.94]; hydromorphone: aOR, 1.51 [95% CI, 1.12-2.04]).
Interpretation
Among patients in the ICU who received MV, persistent opioid use in the year after hospital discharge was more frequent among patients initially exposed to IV morphine or hydromorphone compared with fentanyl, but only among those without a prior history of persistent opioid use. The choice of initial opioid may have long-term consequences for patients. Further research is needed to confirm these exploratory findings.