TY - JOUR T1 - Effects of hypnosis versus enhanced standard of care on postoperative opioid use after total knee arthroplasty: the HYPNO-TKA randomized clinical trial JF - Regional Anesthesia & Pain Medicine JO - Reg Anesth Pain Med DO - 10.1136/rapm-2022-103493 SP - rapm-2022-103493 AU - Jessie Markovits AU - Ondrej Blaha AU - Emma Zhao AU - David Spiegel Y1 - 2022/06/16 UR - http://rapm.bmj.com/content/early/2022/06/16/rapm-2022-103493.abstract N2 - Background Hypnosis decreases perioperative pain and has opioid-sparing potential but has not been rigorously studied in knee arthroplasty. This trial investigates the impact of perioperative hypnosis on inpatient opioid use following total knee arthroplasty.Methods This prospective randomized controlled trial was conducted at a single academic medical center. The hypnosis arm underwent a scripted 10 min hypnosis session prior to surgery and had access to the recorded script. The control arm received hypnosis education only. The primary outcome was opioid use in milligram oral morphine equivalents per 24 hours during hospital admission. A secondary analysis was performed for patients taking opioids preoperatively.Results 64 primary knee arthroplasty patients were randomized 1:1 to hypnosis (n=31) versus control (n=33) and included in the intent-to-treat analysis. The mean (SD) postoperative opioid use in oral morphine equivalents per 24 hours was 70.5 (48.4) in the hypnosis versus 90.7 (74.4) in the control arm, a difference that was not statistically significant (difference −20.1; 95% CI −51.8 to 11.4; p=0.20). In the subgroup analysis of the opioid-experienced patients, there was a 54% daily reduction in opioid use in the hypnosis group (82.4 (56.2) vs 179.1 (74.5) difference of −96.7; 95% CI -164.4 to –29.0; p=<0.01), equivalent to sparing 65 mg of oxycodone per day.Conclusion Perioperative hypnosis significantly reduced inpatient opioid use among opioid-experienced patients only. A larger study examining these findings is warranted.Trial registration number NCT03308071.Data are available upon reasonable request. De-identified data are available upon reasonable request at the discretion of the authors beginning two years after the publication of the manuscript, by request to corresponding author, jkittle@stanford.edu. ER -