Importance Individuals with genetic variation that results in absent (poor metabolizers) or reduced (intermediate metabolizers) cytochrome P450 2D6 (CYP2D6) enzyme activity have lower concentrations of highly potent active metabolites of tramadol, hydrocodone, and codeine and are thus at increased risk for inadequate pain control. Objective To determine the effect of CYP2D6-guided opioid prescribing on postoperative pain and opioid use. Design, Setting, and Participants This open-label randomized clinical trial enrolled participants from surgery clinics at 8 US health systems. Individuals undergoing a planned surgery anticipated to cause postoperative pain for at least 7 to 10 days were enrolled from March 2021 to September 2023, with follow-up concluded in March 2024. Intervention Participants in the CYP2D6-guided arm underwent CYP2D6 genotyping before surgery, with recommendations to avoid tramadol, hydrocodone, and codeine for postoperative pain in CYP2D6 poor and intermediate metabolizers, as defined by genotype and use of CYP2D6 inhibitors. Participants in the control arm had usual pain management. Main Outcomes and Measures The primary outcome was the 10-day Silverman integrated analgesic assessment (SIA) score, a rank-based composite measure of average pain intensity on a 10-point scale, with higher scores indicating greater pain and opioid use (in morphine milligram equivalents MMEs). Secondary end points included individual components of the primary outcome and concordance between metabolizer phenotype and prescribed opioid 10 days after surgery. The primary analysis compared outcomes between poor and intermediate metabolizers in the CYP2D6-guided arm vs the control study arm. The analytical population comprised the subset of intent-to-treat participants with an actionable phenotype who completed surgery. Results Of 1602 participants enrolled, 351 (mean SD age, 62 13 years; 237 68% female) had a CYP2D6 poor or intermediate metabolizer phenotype, proceeded to surgery, and were randomized to the CYP2D6-guided (n = 176) or control (n = 175) study arm. The most common procedures in the actionable population were total knee (177 50%) and total hip (97 28%) arthroplasties. Concordance between postsurgical opioid treatment and CYP2D6 phenotype was 64% (n = 112) in the CYP2D6-guided arm and 27% (n = 47) in the control arm (difference, 37 95% CI, 27-46 percentage points; P lt; .001). At 10 days, the mean (SD) SIA score was 1.4 (95.9) in the CYP2D6-guided arm and −1.4 (93.1) in the control arm (difference, 2.8 95% CI, −18.3 to 23.8; P = .80). Mean (SD) numeric pain intensity rating (5.2 2.2 and 5.1 2.3), overall opioid use (13.7 14.9 and 13.2 14.7 MME/d), and other secondary end points did not differ between the CYP2D6-guided arm and the control arm. Conclusions and Relevance In this randomized clinical trial of CYP2D6-guided postoperative opioid prescribing, there were significant prescribing changes in CYP2D6 poor and intermediate metabolizers but no differences in pain control compared with usual care. The data do not support a role for CYP2D6-guided opioid therapy in the contemporary postoperative setting of multimodal pain management. Trial Registration ClinicalTrials.gov Identifier: NCT05966129
Lee et al. (Fri,) studied this question.