STUDY OBJECTIVE Tramadol is metabolized by cytochrome P450 (CYP) 2D6 to form an active metabolite that exhibits its analgesic effect. Medications that inhibit this enzyme are used often in practice, yet the clinical impact of this interaction on the analgesic effects of tramadol has yet to be fully described. The objective was to determine whether a clinically relevant decrease in pain control is observed in patients taking scheduled tramadol concomitantly with a strong CYP2D6 inhibitor. DESIGN Retrospective cohort study. SETTING Large health care system. PATIENTS One hundred fifty-two adult inpatients who received scheduled tramadol for at least 24 hours with (76 patients) or without (76 patients) a strong CYP2D6 inhibitor between January 1, 2012, and February 28, 2017, were included in the analysis. Patients hospitalized for opioid use disorder or those receiving substandard dosing of tramadol were excluded. MEASUREMENTS AND MAIN RESULTS The primary outcome was mean breakthrough opiate consumption in the presence and absence of CYP2D6 inhibitors. Secondary outcomes included mean pain scores, length of hospital stay, tramadol discontinuation rates, and prespecified subgroup analyses based on patient sex, race, and specific CYP2D6 inhibitor administered. Patients receiving concurrent CYP2D6 inhibitors required significantly more breakthrough morphine milligram equivalents per day compared with patients receiving scheduled tramadol without CYP2D6 inhibitors (geometric mean AE SD 18.2 AE 6.3 vs 5.7 AE 6.7 mg morphine milligram equivalents, p<0.001). No significant differences existed between cohorts for mean pain score, length of hospital stay, or tramadol discontinuation rate. CONCLUSION This study demonstrated a clinically relevant decrease in the efficacy of tramadol when used for pain control in patients receiving a strong CYP2D6 inhibitor. These results should encourage clinicians to review medication lists for this interaction and adjust regimens accordingly to ensure adequate pain control.

What is known and objective: Current evidence-based guidelines for the treatment of acute low back pain (ALBP) recommend the use of opioid medications only after failure of nonpharmacological therapy, non-steroidal anti-inflammatory drugs and skeletal muscle relaxants and after thorough evaluation of risks and benefits. Despite this recommendation and the state of the opioid epidemic in the United States (US), opioids remain a common drug of choice for ALBP in the emergency department (ED). The purpose of this study was to quantify the prevalence and identify predictors of opioid prescribing for acute lower back pain (ALBP) in emergency departments (EDs) in the United States.Methods: This was a national, cross-sectional study of the National Hospital Ambulatory Care Survey from 2013-2016. ED visits for patients aged ≥18 years treated for ALBP were included. Patients presenting with specified reasons that an opioid may be indicated were excluded. The primary endpoint was frequency of opioids prescribed. A multivariate logistic regression model identified patient-and provider-level predictors of opioid use. Results and discussion:This analysis included 2260 visits for ALBP. Opioids were prescribed in 32.3% of visits. Positive predictors of opioid prescribing were pain score of 7-10 (OR 1.85; 95% CI 1.26-2.70), and patients seen in the Southern (OR 2.53; 95% CI 1.47-4.36) or Western US (OR 2.10; 95% CI 1.19-3.70). Opioids were prescribed less often to patients who received a NSAID or acetaminophen (OR 0.38; 95% CI 0.28-0.52 and OR 0.03; 95% CI 0.01-0.10, respectively). What is new and conclusion:Opioid prescribing rates for ALBP remain high and the predictors identified demonstrate that this prescribing pattern is not uniformly distributed across the patient and provider characteristics studied. K E Y W O R D S low back pain, opioid, opioid prescribing, pain | 699 WALKERLY Et AL. Received an NSAID a 0.36 0.25-0.51 Received acetaminophen a 0.03 0.01-0.10 History of alcohol use disorder 0.51 0.19-1.35a Indicates a statistically significant predictor of receiving an opioid via the logistic regression model.