Precipitated Withdrawal in the Era of Street Fentanyl
William E. Soares,Elizabeth Schoenfeld,Peter D. Friedmann
DOI: https://doi.org/10.1001/jamanetworkopen.2024.35857
2024-09-29
JAMA Network Open
Abstract:The medication buprenorphine revolutionized the treatment of opioid use disorder (OUD). Whereas methadone involved complex dosing limited to highly regulated opioid treatment programs (ie, methadone clinics), buprenorphine, a partial opioid agonist with efficacy similar to that of methadone, is rapidly titrated, has a low risk of overdose, and can be prescribed by any clinician for home use. 1 The only unique consideration in starting buprenorphine has been the recognition and avoidance of precipitated withdrawal. Due to its high affinity for the μ-opioid receptor, buprenorphine given too soon after the use of a full agonist may precipitate opioid withdrawal symptoms, including restlessness, sweating, anxiety, cramps, and tremors. Historically, avoiding precipitating withdrawal was straightforward and involved waiting to administer buprenorphine until the patient began to experience withdrawal, measured by the Clinical Opiate Withdrawal Scale (COWS) score. However, reports of higher rates of buprenorphine-precipitated withdrawal have grown as fentanyl has replaced heroin in the street drug supply, even with high COWS scores and extended periods of abstinence. 2 Compared with heroin, fentanyl has a higher potency and shorter duration of action, resulting in patients using it more frequently and developing higher tolerance. Fentanyl is also lipophilic and may be retained in the body for a longer period. 3 Despite reports, 4 evidence confirming the association of buprenorphine-precipitated withdrawal with street fentanyl remains limited. The article by Thakrar et al 5 advances our understanding of buprenorphine-precipitated withdrawal and its association with street fentanyl. Their retrospective cohort study extracted data from the electronic health records of 226 adult patients with OUD (had a discharge diagnosis of OUD, were administered naloxone, or had a chief complaint consistent with OUD or overdose) who received buprenorphine (≥2 mg) at 3 urban academic hospitals in Philadelphia, Pennsylvania, between January 1, 2020, and December 31, 2021. Patients missing COWS documentation before or 4 hours after buprenorphine initiation and those with a low COWS score (<8) were excluded. The primary outcome was precipitated withdrawal, defined as an increase in COWS score of 5 points or greater from immediately before to within 4 hours after buprenorphine administration. The investigators also examined potential moderators including initial COWS score, initial buprenorphine dose, body mass index, and quantitative measurements of fentanyl in urine. Of the 226 patients in the study, 5 26 (11.5%) met criteria for precipitated withdrawal. In the subgroup of 123 patients who had a positive urine fentanyl test result, 20 (16.3%) experienced precipitated withdrawal. While there was no association between precipitated withdrawal and initial COWS score or buprenorphine dose, the authors noted that precipitated withdrawal was associated with higher urine fentanyl concentrations and higher body mass index. The study by Thakrar et al 5 has limitations that must be considered when interpreting the results. The investigators identified 1577 patients with OUD who received buprenorphine, but only 226 (14.3%) met the study criterion of having a recorded COWS score before and after buprenorphine, leading to concerns about selection bias. While the authors included demographic information from the excluded patients, we lack insight into the factors that health care practitioners used to inform clinical decisions, such as when to measure a COWS score, how to dose buprenorphine, and in which patients to obtain a urine fentanyl assay. Unfortunately, the small number of precipitated withdrawal events limits power to estimate this unmeasured bias. Additionally, without additional urine toxicology results, it is difficult to know whether fentanyl, other opioids (including methadone), or nonopioid substances contributed to or mimicked opioid withdrawal symptoms in this cohort. Finally, the finding of an association between precipitated withdrawal and urine fentanyl concentration is difficult to interpret given the maximum urine fentanyl concentration limitations as well as variability in fluid and kidney status. Nonetheless, the findings are consistent with the lived experience of people who use street drugs, and they support the complex and poorly understood relationship between fentanyl and buprenorphine-precipitated withdrawal. Thakrar et al 5 found a substantially higher rate of precipitated withdrawal than other studies. 6 ,7 Snyder et al 6 abstracted data from 492 emergency department patients with OUD who were part of California's CA Bridge program for OUD treatment. They found 8 patients (1.6%) had precipitated withdrawal documented in their medical records. While the rate -Abstract Truncated-
medicine, general & internal