Brent Ershoff, MD, MS: Intraoperative hydromorphone decreases postoperative pain: an instrumental variable analysis

Brent D. Ershoff, MD, MS

Published on April 12, 2023

Using an instrumental variable analysis treating hydromorphone presentation dose (i.e. 2mg versus 1mg vial) as the instrumental variable, Brent Ershoff, MD, MS, shows that increased intraoperative hydromorphone administration causes decreased postoperative pain.


"Background: Intraoperative administration of short-acting opioids might lead to increased postoperative pain and opioid requirements. There are few data describing the effects of intermediate-duration opioids such as hydromorphone on these outcomes. We have previously shown that a switch from a 2 mg to a 1 mg vial of hydromorphone was associated with decreased intraoperative dose administration. As presentation dose affected intraoperative hydromorphone administration and was unrelated to other policy changes, it could serve as an instrumental variable, assuming significant secular trends were not present during the study period.

Methods: In this observational cohort study of patients who received intraoperative hydromorphone (n=6750), an instrumental variable analysis was used to evaluate whether intraoperative hydromorphone administration affected postoperative pain scores and opioid administration. Before July 2017, hydromorphone was available as a 2-mg unit dose. From July 1, 2017 to November 20, 2017, hydromorphone was only available in a 1-mg unit dose. A two-stage least squares regression analysis was used to estimate causal effects.

Results: A 0.2-mg increase in intraoperative hydromorphone administration caused a decrease in admission PACU pain scores (mean difference, –0.8; 95% confidence interval, –1.2 to –0.4; P<0.001) and decreased maximum and time-weighted mean pain scores over 2 days postoperatively, without increased opioid administration.

Conclusions: This study suggests that intraoperative administration of intermediate-duration opioids does not cause the same effects as short-acting opioids with respect to postoperative pain. Instrumental variables can be used to estimate causal effects using observation data when unmeasured confounding is present."

Read more in the British Journal of Anaesthesia.