Fentanyl Absorption, Distribution, Metabolism, and Excretion (ADME): Narrative Review and Clinical Significance Related to Illicitly-Manufactured Fentanyl

Abstract

Objectives:

This narrative review summarizes literature on pharmaceutical fentanyl’s absorption, distribution, metabolism, and excretion patterns to inform research on illicitly-manufactured fentanyl (IMF).

Results:

Fentanyl is highly lipophilic, lending itself to rapid absorption by highly-perfused tissues (including the brain) before redistributing from these tissues to muscle and fat. Fentanyl is eliminated primarily by metabolism and urinary excretion of metabolites (norfentanyl and other minor metabolites). Fentanyl has a long terminal elimination, with a documented secondary peaking phenomenon that can manifest as “fentanyl rebound”. Clinical implications in overdose (respiratory depression, muscle rigidity and “wooden chest syndrome”) and opioid use disorder treatment (subjective effects, withdrawal and buprenorphine-precipitated withdrawal) are discussed. The authors highlight research gaps derived from differences in medicinal fentanyl studies and IMF use patterns, including that medicinal fentanyl studies are largely conducted with persons who were opioid-naïve, anesthetized, or had severe chronic pain and that IMF use is characterized by supratherapeutic doses and frequent and sustained administration patterns, as well as adulteration with other substances and/or fentanyl analogs.

Conclusions:

This review reexamines information yielded from decades of medicinal fentanyl research and applies elements of the pharmacokinetic profile to persons with IMF exposure. In persons who use drugs, peripheral accumulation of fentanyl may be leading to prolonged exposure. More focused research on the pharmacology of fentanyl in persons using IMF is warranted.

1. Introduction

Fentanyl is a highly potent synthetic mu-opioid receptor agonist that was developed specifically for its efficacious analgesic profile.¹ Pharmaceutical fentanyl is routinely used for anesthesia and analgesia, however illicitly-manufactured fentanyl (IMF) has now proliferated – and in many places replaced – the illicit heroin supply in the United States. The proliferation of IMF in the United States shows no signs of slowing down; seizures of fentanyl-containing powders and pills have steadily risen from 2018 to 2021, with over a quarter of fentanyl seizures in pill form by the end of 2021.² The composition of IMF in seized samples varies tremendously, reflecting variability in exposure in persons who use drugs (PWUD), and there is reason to believe illicit fentanyl exposure exceeds medical exposures. Examining the fentanyl pharmacokinetic profile in the context of IMF is a crucial step towards identifying IMF-related harms and appropriate treatment strategies.

One major gap in knowledge is that data informing fentanyl effects are largely based upon medicinal studies of fentanyl. However, medicinal fentanyl studies are not directly translatable to persons exposed to IMF due to differences in population characteristics, drug composition, administration patterns, and dosage. Throughout this manuscript the authors use medicinal studies of pharmaceutical fentanyl to both explain observed phenomenon and generate hypotheses about IMF, but the distinctions between the two necessitate clarification of the terminology chosen to describe each. In this manuscript, the word ‘fentanyl’ refers to the active ingredient contained within pharmaceutical fentanyl: a regulated drug, often formulated as fentanyl citrate, and used in medical settings. IMF refers to an inconsistent mixture of fentanyl, fentanyl analogs, and adulterants produced in clandestine laboratories and subject to non-medically-directed use.

IMF is conferring tremendous consequences on PWUD, including unprecedented rates of morbidity and mortality. Pharmacologic properties may be contributing to enhanced lethality relative to other opioid agonists, and explain reports from patients that IMF causes a severe withdrawal syndrome³ that necessitates frequent daily administrations to manage emergent withdrawal.⁴ Fentanyl’s unique pharmacokinetic profile may explain a possible mechanism underlying the precipitated withdrawal syndrome reported when the mu-opioid partial agonist buprenorphine is administered within 48 hours of IMF exposure^5,6,7 (in contrast to heroin, which does not precipitate withdrawal after >12 hours of acute abstinence⁸).

Given its proliferation and unique effects, it is crucial to understand the pharmacology of IMF in persons with opioid use disorder (OUD) who are exposed regularly. This narrative review builds upon prior reviews of the pharmacodynamic effects of fentanyl^9,10 by discussing features of fentanyl’s pharmacokinetic profile from the perspective of the current opioid crisis. The authors highlight aspects of the absorption, distribution, metabolism, and excretion patterns (ADME) of fentanyl, and present the clinical significance and potential implications. The goal of this review is to provide a foundation from which investigations into IMF and associated treatments can be developed.

2. Absorption and Distribution

2.1. Absorption and Applications to IMF:

Fentanyl is highly lipophilic relative to other opioid agonists (like heroin) which, combined with its low molecular weight, facilitates swift penetration into biological membranes, including the central nervous system.¹¹ Nearly instantaneous absorption and onset is reflected by subjective experience, as persons using IMF frequently note the rapid, intense onset in qualitative studies.^4,12

Since fentanyl is readily-absorbed, it may be administered through a variety of routes to produce effects. This is reflected by the array of commercially-available dosage forms and associated routes of administration (e.g. transdermal, sublingual, buccal, and intranasal). Oral bioavailability is low due to extensive first-pass metabolism (~30% bioavailability for swallowed doses),¹³ and fentanyl is not recommended for oral administration. One study of intranasal fentanyl reported high bioavailability (89%) and a similar profile of effects relative to intravenous (IV) administration, with slightly (5 minute) delayed absorption from intranasal administration.¹⁴ A study of inhaled, aerosolized fentanyl showed comparable pharmacokinetic and pharmacodynamic profiles to an IV injection of the same dose.¹⁵ These properties of absorption are reflected in patterns of IMF use; drug effects are achieved through injecting or snorting IMF, and there are reports of PWUD smoking IMF to surmount poor venous access and infection risk.¹⁶

2.2. Plasma Protein Binding and Implications for IMF:

Experiments in seeded human plasma reveal that fentanyl becomes 86–89% protein-bound in blood, with approximately 78% bound to albumin and 12% bound to alpha-1 acid glycoprotein, which is somewhat unusual for a weakly basic drug (pKa 8.43).¹¹ Plasma protein binding is influenced by physiological factors such as plasma pH, disease states affecting protein levels, and other protein-bound drugs, leading to significant variation in proportion bound between individuals.^17,11 Since plasma protein binding and ionization are both affected by pH, changes in plasma pH (due to decreased respiration, for example) affect drug concentrations in plasma and tissues, including the brain.¹⁸ It is unknown if factors affecting plasma protein binding and pH have a clinically-meaningful impact on overdose or withdrawal in IMF-exposed PWUD.

2.3. Distribution and Implications for IMF:

Fentanyl has a large estimated volume of distribution, and tissue distribution occurs via both passive diffusion and carrier-mediated mechanisms.¹⁹ Consistent with highly lipophilic drugs, fentanyl is rapidly distributed into well-perfused tissues, followed by a slower redistribution into larger, less-perfused tissues.²⁰ In a study comparing plasma and tissue fentanyl concentrations in rats after a single IV injection, fentanyl reached its maximum concentration first in highly-perfused organs (brain, lungs, heart), followed by skeletal muscle, and finally subcutaneous fat.²⁰ Fentanyl concentrations in tissues were consistently higher than concentrations in plasma, reflecting its affinity for tissues, and plasma concentrations paralleled concentrations in the highly-perfused tissues.²⁰ A second study of the distribution of IV fentanyl and its metabolites to various organs in rats confirmed that fentanyl concentrations in the brain peak nearly instantaneously and are followed by rapid redistribution of fentanyl to less-perfused tissues like fat, which did not reach a maximum concentration until 90 minutes post-dose.²¹ This study showed no evidence of fentanyl re-entering the brain or metabolites accumulating in the brain after the initial peak.²¹ Fentanyl is a p-glycoprotein substrate, which contributes to this redistribution profile by transporting fentanyl out of the brain.¹⁹

Human drug administration studies align with these rodent studies. In humans, IV administration of 0.25–0.8 mg fentanyl for anesthesia produced detectable subjective effects within 30–90 seconds, followed immediately by heart rate and respiratory changes that were detectable within another 15–30 seconds.²² Anesthetic effects lasted approximately 40 minutes.²² A second pharmacokinetic study in healthy volunteers showed IV fentanyl distribution and redistribution to be complete by 60 minutes after administration (about the same as the duration of action), such that the terminal decay phase dominates by that time.¹⁷ Redistribution from the brain to peripheral tissues, in addition to metabolic conversion, is regarded as what terminates fentanyl’s central effects.

Importantly, the distribution of fentanyl from highly perfused tissues into muscle is hypothesized as contributing to the skeletal muscle rigidity that has been observed within 5 minutes of rapidly-administered doses.²² Rigidity in some IMF overdose presentations differs from ‘slouching’ heroin overdose presentations, making it difficult for bystanders to recognize.¹² Muscle rigidity also contributes to fentanyl’s extremely lethal overdose profile by contributing to chest wall rigidity known as “wooden chest syndrome.” ¹⁰ Understanding fentanyl-associated vocal cord closure and wooden chest syndrome has clear implications in overdose response and treatment.²³

Fentanyl distribution into peripheral tissues may cause drug accumulation following multiple administrations, which might be clinically-meaningful. One study in rats found the analgesic properties of fentanyl became enhanced (or sensitized) following several injections of high (but not low) fentanyl doses.²⁴ In qualitative studies of PWUD, persons note the short duration of IMF prompts them to use it more frequently,¹² and more frequent use may increase the cumulative effects of peripherally-sequestered fentanyl, but further studies are warranted to examine this effect. Additionally, peripheral accumulation of fentanyl may contribute to prolonged clearance and secondary peaking (discussed below).

3. Metabolism and Excretion:

Fentanyl is eliminated from the body primarily as metabolites in the urine. Metabolism occurs quickly, as metabolites can be detected in the plasma as early as 90 seconds after IV administration (representing 19% of a radiolabeled dose).¹⁷ Metabolite levels peak in plasma around 90 minutes (68.5% of total dose), declining slowly thereafter with a mean terminal half-life of 375 minutes.¹⁷ After 72 hours most of the parent fentanyl product is metabolized, with only 6.4% of the dose detected unchanged in urine and 80% appearing as metabolites in urine (76%) or feces (8%).¹⁷ Fentanyl terminal elimination half-life estimates vary from 219–853 minutes (~3.6–14.2 hours).²⁵

3.1. Norfentanyl:

Fentanyl elimination is facilitated by metabolism, which is heavily dependent on hepatic phase 1 cytochrome-P450 (CYP)-mediated reactions.¹³ The primary biotransformation is N-dealkylation at the piperidine ring to norfentanyl, which is mediated by CYP3A-enzymes.¹³ Conversion of fentanyl to norfentanyl can be inhibited by co-administration of CYP3A4 inhibitors.¹³ Urinary excretion of norfentanyl 24 hours after continuous infusion of fentanyl for patients accounts for 8–25% of the parent administration.²⁶ Norfentanyl can be detected in the urine of rats up to 72 hours after injection²⁷ and in the urine of post-surgical patients over 96 hours later.²⁸ Norfentanyl has been found in various biological media and tissues in higher concentrations than parent fentanyl following fentanyl exposure (excluding the brain), and there are no published data that support norfentanyl activity.²¹ Norfentanyl is also the byproduct of some fentanyl analogs.²⁷ Very little norfentanyl is bound in plasma (<8% to either albumin or alpha-1 acid glycoprotein in seeded human plasma), thus it is unlikely that norfentanyl outcompetes fentanyl for binding sites, and norfentanyl binding should not affect free concentrations of fentanyl to exert effects or be removed from the body.¹¹

3.2. Minor Metabolites:

Norfentanyl is not the only metabolite of fentanyl; other minor metabolic pathways (representing an estimated <1% of metabolism²⁵) include hydroxylation and hydrolysis.^21,29 Metabolite structures discussed here are presented in Figure 1. Studies of incubated human cells with fentanyl (10 microM) found the cells produced norfentanyl and several other hydroxylated metabolites, including 4’-hydroxyfentanyl and beta-hydroxyfentanyl.²⁶ Fentanyl hydroxylation can occur at several molecular sites, and the metabolites can undergo additional reactions into further metabolites, such as both hydroxylation and N-dealkylation of fentanyl to form hydroxynorfentanyl.¹³ The metabolite 4’-hydroxyfentanyl produced opioid-like effects in guinea pig ileum, with effects between those of morphine and meperidine,²¹ and is likely further metabolized via a second hydroxylation and O-methylation by catechol-O-methyltransferase (COMT) to yield 4’-hydroxy-3’-methoxyfentanyl.²⁶

Figure 1.

Select Metabolic Pathways and Metabolites of Fentanyl

Beta-hydroxyfentanyl has been identified in the blood of postmortem IMF overdose cases, and in smaller quantities in patients receiving fentanyl, and has been correlated with fentanyl and norfentanyl concentrations.³⁰ In rat tissue studies, beta-hydroxyfentanyl has also been detected in small quantities, and is hypothesized to exhibit opioid activity consistent with a competitive antagonist versus a pure agonist, perhaps also displaying differing affinities for mu and delta opioid receptors.²¹ Beta-hydroxyfentanyl is metabolized into norfentanyl by liver microsomes in vitro.²⁷ Beta-hydroxyfentanyl is also an illicit fentanyl analog^{30, 27} and has been described as such in toxicology surveillance reports.³¹ Therefore, determining the source of beta-hydroxyfentanyl in persons using IMF is complicated by the fact that beta-hydroxyfentanyl could be both a metabolite of fentanyl or an exogenously-administered illicit fentanyl analog.³⁰

Fentanyl amide hydrolysis produces despropionylfentanyl, also known as 4-anilino-N-phenethylpiperidine (4-ANPP).¹³ 4-ANPP has been identified in seized fentanyl powders and the biosamples of PWUD, though it is not clear whether its presence represents a fentanyl metabolite, a metabolite of fentanyl analogs (such as acetylfentanyl), or an IMF precursor and contaminant.³⁰ More research is needed to determine if these and other minor metabolites are clinically-relevant, including effects with respect to endogenous versus exogenous origin.

3.3. Clinical Implications of Metabolism:

Fentanyl’s extensive metabolism has unclear implications for PWUD. The contribution by CYP3A may lead to drug-drug interactions between IMF and other substances, as well as genetic differences in pharmacokinetics.²⁵ While fentanyl metabolites are currently considered inactive or inconsequential (due to small quantities produced by medicinal dosing), it remains possible that some metabolites (or fentanyl analog metabolites) have activity that have not yet been studied. An IMF metabolite may exert meaningful effects, similar to how heroin/morphine metabolites have been proposed to have neuroexcitatory effects.³² There is a dearth of data regarding the presence and physiological implications of metabolites in persons consuming IMF, which is characterized by supratherapeutic exposures that exceed the data available on medicinal fentanyl metabolism.

4. Secondary Peaking and “Fentanyl Rebound”:

Pharmacokinetic studies of fentanyl have reliably produced a plasma concentration-time profile characterized by an initial rapid distribution phase and a slow terminal decay phase.¹⁷ However, individual concentration-time curves may demonstrate multiple peaking (sudden increases in fentanyl plasma concentrations). A thorough overview of the multiple peaking phenomenon, which is evident in other drugs as well, is presented elsewhere.³³

The earliest fentanyl peak in plasma can be observed within five minutes of IV administration.³⁴ Studies of patients receiving IV fentanyl in hospitals showed secondary peaks in fentanyl plasma concentrations occurred approximately 45–60 minutes after administration,^35,36 though some patients experienced a secondary plasma peak much later (210 minutes³⁷). These effects are clinically-meaningful; studies have reported new onset respiratory distress during the secondary peak, sometimes severe enough to warrant naloxone rescue.^36,38 One examination of patients receiving fentanyl for anesthesia found changes in carbon dioxide consistent with respiratory depression were evident immediately following fentanyl administration, but returned to baseline before reoccurring in 90% of patients hours later during surgical recovery.³⁹ A second examination of cardiac surgery patients found delayed chest rigidity and respiratory depression (as well as increased carbon dioxide and decreased pH) 2–6 hours after surgery associated with increased plasma fentanyl.⁴⁰ This recurrence of respiratory depression following surgical recovery was termed “fentanyl rebound” by some groups.²¹

There are two prevailing hypotheses regarding the mechanism underlying fentanyl’s multiple plasma peaking phenomenon. The first is enterogastric reabsorption, a process whereby drugs are removed from the blood by gastric cells, secreted and held in the stomach lumen, and then reabsorbed via the intestine after stomach contents are emptied.⁴¹ One study found that 3–4% of a 0.5 mg IV fentanyl dose was secreted into the gastric juice, and an estimated 16% of the dose was present in the stomach wall 10 minutes after injection in surgical patients.⁴² Co-administration of fentanyl with the anti-histamine cimetidine, which binds to gastric cells to stop production of stomach acid, reduces secondary peaks of fentanyl.³³ Additional evidence supporting enterogastric reabsorption is that patients who are anesthetized – and as a result have inhibited gastric emptying and intestinal reabsorption – do not demonstrate secondary peaks.³⁷

A second potential mechanism is an altered redistribution pattern from peripheral tissues. This is supported by studies conducted in persons who received fentanyl for anesthesia, which have reported that secondary peaks coincide with changes in blood flow and exercising of muscles, which may prompt elution of stored fentanyl from muscle and subsequent peaks in plasma concentrations.^17,37 Additional evidence for this theory are reports that fentanyl administration prior to applying a tourniquet in surgery can result in secondary plasma concentration peaks following tourniquet deflation and resumption of circulation.³³ Muscle rigidity-induced muscle contractions could also cause efflux of fentanyl from skeletal and lung muscles that is redeposited back into plasma,³⁴ causing a secondary peak. A large dose of fentanyl can produce secondary peaks as soon as 3 minutes after administration (too early for enterogastric reabsorption), suggesting both mechanisms likely contribute to this phenomenon.³⁴

4.1. Clinical Implications of Secondary Peaking:

The clinical implications of secondary peaking in PWUD are unknown, but could be profound. Medicinal studies indicate that meaningful levels of fentanyl (capable of exerting opioid agonist effects strong enough to warrant antagonist intervention) are reintroduced into the blood hours after initial administration occurs and the subjective effects of fentanyl have remitted. This secondary-peaking process may also play a role in the anecdotal reports of non-linear withdrawal reported by persons with OUD who use IMF regularly, which complicates patient treatment strategies.

It is not currently possible to predict the occurrence of secondary plasma peaks, and it is not known if secondary peaking in the plasma affects IMF-related withdrawal presentation or contributes to late-onset overdose risk via respiratory depression.

5. Discussion

Fentanyl has a rapid onset of subjective and physiological effects and short duration of centrally-mediated action, which has led to the consensus that medicinal fentanyl is a short-acting opioid. However, lipophilicity-driven distribution of fentanyl into muscle and adipose tissue results in a long terminal elimination. Since some PWUD administer IMF frequently over a prolonged time, it is plausible that fentanyl is accumulating in peripheral tissues. This is further supported by evidence that persons entering treatment for IMF tested positive for fentanyl and norfentanyl in urine for an average of 7 and 13 days, respectively (up to 26 days).⁴³ A second study confirmed the finding that persons entering treatment for IMF tested positive for fentanyl in their urine for about 7 days, with overweight people testing positive for a statistically-significantly longer time than people with a healthy weight.⁴⁴

It is possible that the pharmacokinetic profile of fentanyl plays a role in the challenges IMF is placing on the OUD treatment system. If fentanyl accumulates in the periphery and is released from tissue stores for extended periods, it is possible that fentanyl is acting like a long-acting opioid. This may explain how fentanyl interacts with buprenorphine during the buprenorphine induction process in a manner similar to long-acting opioids like methadone.⁴⁵ Although this is a plausible rationale for buprenorphine-precipitated withdrawal in the context of fentanyl exposure, the degree to which varying pharmacodynamic profiles may also play a role is unclear and warrants inquiry.

Accumulation of fentanyl in peripheral stores is a distinctive element of IMF exposure necessitating prospective examination. The “fentanyl rebound” effect during secondary plasma peaks in post-surgical recovery suggests that fentanyl stores are continuing to exert active physiological effects. This is further compounded by fentanyl’s metabolic profile, which is characterized by a major metabolite (norfentanyl) and an array of minor metabolites that may themselves be able to confer opioid effects, which is important to consider in future research with persons who use IMF regularly.

The pharmacokinetic data that form the basis for these assumptions derive from medicinal studies of fentanyl. The pharmacokinetics of fentanyl are notoriously variable between individuals, and a more thorough review of person-specific factors (e.g. genetics, age, renal and hepatic function) contributing to variability in fentanyl pharmacokinetics is presented elsewhere.²⁵ Even though medicinal studies can help form hypotheses about IMF, there are four major limitations to their applicability. First, the medicinal studies are conducted in either a) opioid-naïve, healthy volunteers or b) anesthetized (often intubated) patients undergoing surgery or c) patients with severe chronic pain (e.g. cancer-related pain). The degree to which population-related factors and tolerance impacted fentanyl pharmacology compared to PWUD is uncertain.¹⁰ Second, as mentioned in the introduction, these studies examined pharmaceutical fentanyl, while IMF is a mixture of fentanyl, fentanyl structural analogs, and adulterants (e.g. heroin, stimulants, depressants). Differential pharmacology in active fentanyl analogs, fentanyl analog metabolites with unstudied activity, and active adulterants (with pharmacodynamic interactions) complicate IMF pharmacology. The pharmacology of fentanyl analogs (“fentalogs”) is reviewed elsewhere.⁴⁶

Third, fentanyl exposure through IMF is more frequent and sustained than the studies cited, many of which were conducted as single-administration studies. Fourth, fentanyl exposure through IMF is likely orders of magnitudes larger than exposure to doses studied medicinally. One ‘high-dose’ fentanyl study cited here administered 60 mcg/kg (calculated range 3,180–5,220 mcg) IV bolus fentanyl to patients undergoing major cardiac surgery as the sole anesthetic agent (with some surgeries lasting >2 hours).³⁴ A case report from 2016 found fentanyl doses up to 6,900 mcg per pill pressed to look identical to hydrocodone/acetaminophen tablets.⁴⁷ In 2022, the DEA reported that seized counterfeit pills contained as much as 5.1 mg (5,100 mcg) of fentanyl, with 42% of pills tested containing at least 2 mg (2,000 mcg) fentanyl (a potentially lethal dose).⁴⁸ Therefore, a person administering multiple doses of IMF daily may be exposing themselves to considerably more fentanyl more frequently than the cited literature, and research needs to be done on the pharmacokinetic consequences of chronic high-dose fentanyl exposure.

Fentanyl has been studied for almost five decades, and it is likely that pivotal studies in this area were not included in this review. In that regard, this review should be considered an initial introduction to the nuances of fentanyl ADME to support more focused and prospective examination of these trends in IMF exposure.

6. Conclusions

Although fentanyl used for anesthesia and analgesia is considered a short-acting opioid, there is evidence of peripheral sequestration and accumulation that may contribute to a profile more akin to a long-acting opioid in IMF-exposed PWUD. Accumulation would suggest it may be possible for persons who use IMF regularly to continue to be fentanyl-exposed for a prolonged period after their last IMF use. The degree to which this unique pharmacokinetic profile contributes to extended elimination, buprenorphine-IMF interactions, or overdose risk and presentation remain important empirical questions. Additionally, the degree to which the pharmacokinetics of fentanyl might impact euphorigenic and reinforcing effects or opioid withdrawal presentation in the context of IMF and OUD are unknown. The information reviewed here provide a foundation to support focused and prospective research on the role of ADME in persons with chronic exposure to IMF.