Abstract
IMPORTANCE: Cytochrome P450 2D6 (CYP2D6) bioactivates hydrocodone, tramadol, codeine, and oxycodone to active metabolites that primarily provide analgesic activity. Reduced CYP2D6 activity may be associated with poor pain control. OBJECTIVE: To evaluate associations of impaired CYP2D6 activity based on genotype or CYP2D6 inhibitors, alone and together, with analgesic activity of CYP2D6-metabolized opioids among patients with pain. DESIGN, SETTING, AND PARTICIPANTS: This retrospective national, community-based cohort study used electronic health records and genetics data from the All of Us Research Program. Participants included adults prescribed at least 1 CYP2D6-metabolized opioid for more than 7 days between January 1, 2014, and December 31, 2022, with whole-genome sequencing data available. Analysis groups were defined by CYP2D6 phenotype, which was determined based on CYP2D6 genotype or CYP2D6 inhibitor-mediated phenoconversion. Statistical analysis was performed from July 1, 2023, to January 15, 2025. EXPOSURES: CYP2D6-metabolized opioids, with or without concomitant CYP2D6 inhibitor exposure, based on prescription records and overlap with opioids. MAIN OUTCOMES AND MEASURES: The primary outcome was occurrence of any pain-related emergency department (ED) visits during opioid treatment, up to 60 days after opioid initiation. The association between ED visits and CYP2D6 phenotype was assessed using inverse probability treatment weighting-adjusted logistic regression. Additional analyses were conducted by drug and isolating CYP2D6 genotype and inhibitors. RESULTS: Among 31 669 patients (mean [SD] age, 51.2 [15.4] years; 66.5% women) prescribed CYP2D6-metabolized opioids, 15 960 had reduced CYP2D6 activity, and 15 709 had normal or high CYP2D6 activity based on genotype and inhibitors. A higher percentage of patients with reduced CYP2D6 activity (hereafter referred to as phenotypic intermediate metabolizers [pIMs] or phenotypic poor metabolizers [pPMs]) had experienced pain-related ED visits compared with patients with normal or high CYP2D6 activity (phenotypic normal metabolizers [pNMs] and phenotypic ultrarapid metabolizers [pUMs]) (2.1% vs 1.8%; inverse probability-weighted odds ratio, 1.19; 95% CI, 1.06-1.33). There were no significant differences in ED visits among CYP2D6 genotypic IMs or PMs vs NMs or UMs when testing all 4 drugs together. Among genotypic NMs, ED visits were more frequent among the individuals prescribed CYP2D6 inhibitors (inverse probability-weighted odds ratio, 1.49; 95% CI, 1.32-1.68). In analyses by medication, drug interactions were important for all 4 medications, while genotype associations were significant only for hydrocodone, tramadol, and codeine. CONCLUSIONS AND RELEVANCE: In this cohort study, reduced CYP2D6 activity was associated with increased ED visits among individuals treated with CYP2D6-metabolized opioids. This finding suggests that incorporating data on CYP2D6 genotype and accounting for drug interactions in opioid prescribing may improve pain management and reduce ED visits.