β-blocker Use is Associated with a Reduction in Opioid Use 30 Days After Total Knee Arthroplasty

Background: Total knee arthroplasty (TKA) can lead to chronic pain and prolonged postoperative opioid use. There are few evidence-based interventions to prevent these outcomes. Recently, β-blockers have emerged as possible novel analgesics. Objectives: The objective of this study was to determine whether perioperative β-blocker use is associated with reduced prolonged postoperative opioid use after TKA. Study Design: This study used a retrospective cohort design. Setting: The research took place within Department of Veterans Affairs hospitals in the United States between April 2012 and April 2016. Methods: Patients: IRB approval was obtained to examine the records of Veterans Affairs (VA) patients undergoing TKA. Patients using opioids 60 days before surgery were excluded. Intervention: The intervention being investigated was perioperative β-blocker use, overall and by class. Measurement: Oral morphine equivalent usage through postoperative day 1 and prescription opioid refills through 30, 90, and 365 days after TKA were recorded. Adjusted models were created controlling for relevant demographic and comorbidity covariates. A secondary analysis examined the same outcomes separated by β-blocker class. Results: The cohort was 93.8% male with a mean age of 66 years. Among the 11,614 TKAs that comprised the cohort, 2,604 (22.4%) were performed on patients using β-blockers. After adjustment, β-blocker use was associated with reduced opioid use through 30 days after surgery (odds ratio [OR] 0.89 [95% confidence interval (CI), 0.80-0.99], P = .026). Selective β-blockers were associated with reduced opioid use at 30 days (OR 0.88 [95% CI, 0.78-0.98], P = .021), and nonselective β-blockers were associated with reduced oral morphine equivalent usage through postoperative day 1 (β = -17.9 [95% CI, -29.9 to -5.8], P = .004). Limitations: Generalizability of these findings is uncertain, because this study was performed on a cohort of predominantly white, male VA patients. This study also measured opioid use, but opioid use is not a perfect surrogate for pain. Nevertheless, opioid use offers value as an objective measure of pain persistence in a national cohort for which patient-reported outcomes are otherwise unavailable. Conclusions: Perioperative β-blocker use was associated with reduced prescription opioid use at 30 days after surgery. Both selective and nonselective β-blockers were associated with reduced opioid use when analyzed individually.