IV Opioids in the Hospital: A Controversial Lifeline for Patients in Fentanyl Withdrawal

A patient with bacteria teeming in his blood began packing his belongings to leave the hospital. He had done this more than a dozen times in the past year, and his physicians knew that if he walked out again, he might not come back. The reason was not the infection itself but what would follow it: untreated opioid withdrawal. In a viewpoint published in the Journal of General Internal Medicine, Dr. Michael R. Rose of Johns Hopkins University School of Medicine argues that this clinical scenario, increasingly common in the fentanyl era, demands a radical reframing of how hospitals manage opioid withdrawal, including the controversial use of intravenous opioids as a bridge to stabilizing care.

The clinical stakes are laid bare in the opening case. Mr. Warren, hospitalized with bacteremia caused by resistant gram-negative bacteria, repeatedly requested discharge because standard withdrawal treatments were not controlling his symptoms. His care team started with methadone, his preferred medication for opioid use disorder, along with clonidine and symptomatic adjuncts. When those measures failed and he again moved to leave, the team turned to high doses of intravenous hydromorphone, delivered through a patient-controlled analgesia pump that allowed smaller, more frequent doses with lockout periods. The objective signs of withdrawal, including elevated heart rate, dilated pupils, yawning, tremors, goosebumps, vomiting, diarrhea, and sweating, all ceased. What remained were cravings, which the author identifies as a cornerstone symptom of withdrawal and the predominant driver of the patient’s urge to leave.

The central argument of the viewpoint is that opioid withdrawal, including uncontrolled cravings, should be treated as an acute, life-threatening illness requiring prompt management, just as sepsis or diabetic ketoacidosis would be. Rose contends that the historical framing of withdrawal as uncomfortable but benign has bred complacency. In reality, untreated or undertreated withdrawal is a common cause of premature hospital discharge and a rapid return to fentanyl use, which acutely raises mortality both through overdose and through complications of foregone medical care. Avoidance of withdrawal, he notes, is also a major reason patients with opioid use disorder continue to use opioids in the first place.

Fentanyl has magnified these dangers. As the dominant non-prescribed opioid across the United States, it carries a markedly higher overdose risk than heroin or prescription opioids. Its brief duration of action increases injection frequency, compounding risks of infectious complications such as endocarditis and viral hepatitis. Its potency also raises patient tolerance, and its accumulation in tissues prolongs withdrawal, often making traditional methadone titrations unbearable and increasing the risk of precipitated withdrawal when buprenorphine is initiated. Clinicians are therefore increasingly confronting hospitalized patients with severe illness whose withdrawal and cravings cannot be controlled by conventional protocols alone.

The pharmacological rationale for intravenous therapy follows established practice in other acute conditions. For pathologies ranging from sepsis to heart failure to diabetic ketoacidosis, intravenous medications are initially required to achieve faster clinical improvement and offer superior titratability. The same logic applies to other withdrawal syndromes, such as alcohol withdrawal, and to opioid use for acute pain. Intravenous opioids produce a rapid rise in drug concentrations that facilitates swift symptomatic relief, while their short half-lives reduce the risk of accumulation. Patient-controlled analgesia pumps, which require active patient participation and enforce lockout intervals, can safely deliver the large morphine-equivalent doses demanded by patients with especially high fentanyl-driven tolerances while granting them a measure of autonomy over their own treatment.

Short-acting opioids, whether intravenous or oral, already have a recognized adjunctive role in the literature. They can fill residual opioid agonism deficits during methadone titrations. During buprenorphine initiation, they can be used for symptom control while fentanyl clears from the body, alongside buprenorphine during low-dose initiations, or to treat precipitated withdrawal if it occurs. This adjunctive use is described as an increasingly accepted harm reduction practice. What remains undefined, and what the viewpoint seeks to address, is the specific role of the intravenous route, whose absence from formal guidance has produced uncertain and inconsistent use across hospitals.

A crucial element of the argument is the assessment of risk. Clinicians rightly worry that the association between receiving opioids and relief from distress or cravings could reinforce, rather than reverse, the learned behavior that opioids are a solution to stressors. Faster time to peak effect, a hallmark of intravenous administration, is known to increase this reinforcing potential. However, Rose argues that for hospitalized patients with severe opioid use disorder, fentanyl use, and uncontrolled withdrawal, the risk of worsening the disorder is low while the benefit of controlling symptoms, thereby facilitating ongoing medical care including addiction treatment, is high. As time and treatment suppress withdrawal and cravings and the acute medical illness improves, the risk-benefit calculus reverses, and short-acting opioids should be discontinued.

The viewpoint is emphatic that intravenous opioids are never a substitute for evidence-based treatment. Medications for opioid use disorder should be recommended to all patients and serve as the backbone of withdrawal management, with buprenorphine or methadone identified as the most effective treatments even for patients uninterested in continuing them after discharge. Alpha-2 agonists such as clonidine or lofexidine should be first-line adjunctive therapies unless contraindicated. Symptomatic treatment with anxiolytics like hydroxyzine, antiemetics like ondansetron, antispasmodics like dicyclomine, antidiarrheals like loperamide, and non-opioid pain medications such as NSAIDs should also be deployed as needed. Short-acting opioids enter the picture only for patients with residual symptoms despite these measures, or in rare instances where patients decline buprenorphine and methadone entirely and the medications are used to facilitate hospital-based care.

Route selection and expectation setting form the practical core of the proposed guidance. Because the reinforcing risk of rapid-onset opioids is amplified with prolonged use, oral opioids should be prioritized once initial withdrawal symptoms have improved and effective doses have been identified. This transition also smooths drug exposures, offering more consistent symptom control throughout the day. Plans to move from intravenous to oral therapy and eventually to discontinue short-acting opioids altogether should be established with patients early. Once patients reach therapeutic buprenorphine doses, typically 16 to 24 milligrams daily, short-acting opioids should no longer be required. Methadone initiation is typically slower, and patients should understand that short-acting opioids will be tapered as methadone is increased, a process that growing evidence suggests can be expedited through more rapid titration schedules.

The case concludes with hard-won but partial success. Mr. Warren’s symptoms improved on the patient-controlled pump, which was transitioned to oral therapy over subsequent days. As methadone was uptitrated, the short-acting opioids were tapered and eventually discontinued. He remained hospitalized for fourteen days of intravenous antibiotics, his first full treatment course in years. On the eve of transfer to rehabilitation, he again packed his bag and requested discharge, and this time no argument could dissuade him. He left with naloxone, a month’s supply of his medications, and a handshake. His infection was cured and he was alive, but not yet ready for recovery. That outcome, the viewpoint suggests, is precisely the point: intravenous opioids did not cure addiction, but they kept a critically ill patient in the hospital long enough to cure the infection that would otherwise have killed him, and they bought time that conventional withdrawal management alone could not.

Subject of Research: Use of intravenous opioids to manage opioid withdrawal in hospitalized patients with opioid use disorder

Article Title: The Role of IV Opioids for Opioid Withdrawal in Hospitalized Patients

Article References: Rose, M. R. (2026). The Role of IV Opioids for Opioid Withdrawal in Hospitalized Patients. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10856-y

Image Credits: AI Generated

DOI: 10.1007/s11606-026-10856-y

Keywords: opioid withdrawal, intravenous opioids, opioid use disorder, fentanyl, methadone, buprenorphine, hospitalized patients, harm reduction, patient-controlled analgesia, medications for opioid use disorder, withdrawal management, internal medicine