The opioid analgesics comprise one of the most pharmacologically diverse drug classes in clinical medicine. Agents range from plant-derived alkaloids used for centuries to fully synthetic compounds engineered for specific pharmacokinetic profiles and receptor kinetics. Understanding the chemical classification, pharmacokinetic principles, and individual drug profiles of the opioid agonists is essential for rational prescribing: including agent selection, dose calculation, route optimization, equianalgesic conversion, and anticipation of drug-specific adverse effects. This module organizes the opioid agonists by chemical class, profiles the major agents within each class, covers partial agonists and mixed agonist-antagonists, and addresses equianalgesic dosing principles. Adverse effect management and clinical applications are covered in Modules 3 and 4.
Opioid analgesics are organized into chemical families based on structural scaffold, and this classification carries direct clinical relevance. Shared structural features predict metabolic pathways, active metabolite formation, cross-reactivity patterns in hypersensitivity reactions, and in some cases receptor selectivity profiles.2 The phenanthrenes are the oldest and largest chemical family, encompassing the naturally occurring opium alkaloids and their semisynthetic derivatives. The phenanthrene nucleus consists of three fused rings; modifications at key positions on this scaffold yield agents with dramatically different potency and pharmacokinetic properties. Naturally occurring phenanthrenes include morphine and codeine. Semisynthetic phenanthrenes derived from morphine or its close relative thebaine include hydromorphone, oxymorphone, oxycodone, hydrocodone, buprenorphine, nalbuphine, naloxone, and naltrexone. True IgE-mediated allergy to phenanthrenes is uncommon; most reactions attributed to "opioid allergy" within this class represent pharmacological histamine release rather than immune-mediated hypersensitivity. When genuine allergy is suspected, switching to a structurally unrelated class such as the phenylpiperidines or phenylheptylamines is a reasonable strategy.2
The phenylheptylamines include methadone and the withdrawn agent levomethadyl acetate (LAAM). Methadone is a racemic mixture with μ receptor agonist activity at both enantiomers and N-methyl-D-aspartate (NMDA) receptor antagonism at the R-enantiomer; this dual mechanism distinguishes it pharmacodynamically from all other opioid analgesics. LAAM was withdrawn from US markets due to serious QTc prolongation and torsades de pointes risk. The phenylpiperidines constitute a large and clinically important class that includes meperidine (pethidine) and the entire fentanyl family, fentanyl, sufentanil, alfentanil, remifentanil, and carfentanil, as well as the peripherally restricted antidiarrheal agents diphenoxylate and loperamide. High lipophilicity is a shared feature of the fentanyl congeners and accounts for their rapid CNS penetration, short duration of single-dose action, and tendency toward prolonged effect with continuous infusion due to peripheral tissue accumulation.2
The morphinans represent a bicyclic variant of the phenanthrene scaffold; levorphanol is the principal analgesic in this subclass, and like methadone it possesses NMDA receptor antagonist properties and a prolonged half-life. Butorphanol, a mixed agonist-antagonist, and dextromethorphan, an antitussive, are also morphinan derivatives. Among miscellaneous agents, tramadol combines weak μ receptor agonism with serotonin and norepinephrine reuptake inhibition (SNRI activity), and tapentadol combines μ agonism with selective norepinephrine reuptake inhibition with comparatively less serotonergic activity than tramadol.2
Oral bioavailability of opioids is highly variable and is the primary determinant of oral-to-parenteral potency ratios used in equianalgesic dosing.2 Most naturally occurring phenanthrenes undergo extensive first-pass hepatic metabolism, limiting oral bioavailability to 20–40% for morphine and approximately 50–70% for codeine, though codeine's analgesic activity depends on CYP2D6 (cytochrome P450 2D6)-mediated conversion to morphine. Semisynthetic opioids generally achieve higher oral bioavailability: oxycodone 60–87%, hydrocodone approximately 70–80%, and hydromorphone 30–60%.2 Sublingual and buccal routes bypass hepatic first-pass metabolism and are exploited by buprenorphine formulations (sublingual bioavailability approximately 30–50%) and fentanyl buccal and transmucosal products (approximately 50% for buccal fentanyl, approximately 25% for transmucosal lozenge due to variable swallowing). Transdermal fentanyl achieves continuous systemic delivery; after initial patch application, a subcutaneous depot builds over 12–24 hours, reaching steady-state plasma concentrations at approximately 48–72 hours, with a pharmacokinetic tail of 12–24 hours after patch removal due to continued absorption from the dermal depot.2
This extended-release behavior must be accounted for when converting to or from transdermal fentanyl; oral or parenteral breakthrough coverage is required during the initial 12–24 hours of patch application, and the analgesic effect persists for 12–24 hours after patch removal.
Volume of distribution is large for most opioids (1–10 L/kg or greater), reflecting extensive tissue binding and partitioning into lipid-rich compartments. Lipophilicity is the key determinant of CNS penetration rate and onset of action; fentanyl and its congeners are approximately 600–800 times more lipophilic than morphine, explaining fentanyl's IV onset of 1–2 minutes compared to morphine's 15–20 minutes.2 Morphine's comparatively lower lipophilicity has important neuraxial implications: after intrathecal administration, morphine spreads rostrally in the CSF more readily than fentanyl, producing broader segmental analgesia but also delayed respiratory depression (6–18 hours post-injection) from rostral spread to medullary respiratory centers; this is a critical monitoring consideration in neuraxial opioid use. Opioids are moderately to highly protein bound, primarily to albumin and α1-acid glycoprotein. Methadone has particularly high protein binding (approximately 85–90%). States that alter protein concentrations, hepatic disease, malnutrition, and critical illness with elevated acute-phase reactants including α1-acid glycoprotein, can produce unexpected alterations in free drug fraction and clinical effect.2
Hepatic metabolism is the primary route of elimination for virtually all opioids. Glucuronidation is the predominant metabolic pathway for morphine (via uridine diphosphate glucuronosyltransferase 2B7 (UGT2B7)), hydromorphone, oxymorphone, and buprenorphine. CYP3A4 (cytochrome P450 3A4) is the primary enzyme for fentanyl, methadone, buprenorphine (secondary pathway), alfentanil, and sufentanil. CYP2D6 is the principal enzyme for codeine, tramadol, and hydrocodone, converting these prodrugs or partially active agents to more potent metabolites. CYP2D6 genetic polymorphism creates clinically important pharmacokinetic variability: poor metabolizers (approximately 7–10% of white populations) cannot convert codeine to morphine and obtain no analgesia; ultrarapid metabolizers generate excess morphine rapidly, risking toxicity, particularly in breastfed infants of nursing mothers who are ultrarapid metabolizers.3 The FDA issued a black box warning against codeine use in breastfeeding mothers following fatalities from neonatal morphine toxicity in this setting. Renal excretion of parent drug is generally minor for most opioids, but several clinically critical active and neuroexcitatory metabolites are renally cleared, making this a major source of drug-specific toxicity in renal impairment; this point is developed in detail in the individual drug profiles below.2
Morphine is the prototypical opioid analgesic and the reference compound against which all others are compared. It is a naturally occurring phenanthrene alkaloid, the principal active constituent of opium, and remains among the most widely used analgesics in both acute and palliative care settings worldwide. Morphine is a full μ receptor agonist with additional δ receptor activity at higher concentrations. Its oral bioavailability is approximately 20–40% due to extensive first-pass glucuronidation.2 Following oral administration, peak plasma concentrations occur at 30–90 minutes for immediate-release formulations; extended-release formulations (MS Contin, Kadian, Avinza) provide steady plasma levels over 8–24 hours. Morphine is extensively distributed, crossing the blood-brain barrier more slowly than lipophilic opioids due to its polarity. It is metabolized by hepatic uridine diphosphate glucuronosyltransferase 2B7 (UGT2B7) to two primary glucuronide metabolites: morphine-3-glucuronide (M3G, approximately 50–60% of metabolites) and morphine-6-glucuronide (M6G, approximately 10–15%).2 M6G is a potent μ receptor agonist, more potent than morphine itself, that contributes substantially to analgesia, particularly after oral dosing when high M6G concentrations are generated; M3G is pharmacologically inactive at opioid receptors but neuroexcitatory through non-opioid mechanisms and is the primary driver of morphine-associated myoclonus, allodynia, and delirium in renal failure.
Both metabolites are renally excreted, and both accumulate dangerously in renal impairment; morphine should be used with extreme caution or avoided in patients with significant CKD (eGFR <30 mL/min/1.73m2) or acute kidney injury. Histamine release from mast cells occurs with parenteral morphine administration, particularly with rapid IV push, and can produce urticaria, pruritus, flushing, and hypotension; this is not an allergic reaction but a direct pharmacological effect that can be attenuated by slow infusion, premedication with antihistamines, or selection of a non-histamine-releasing opioid.2
Hydromorphone (Dilaudid) is a semisynthetic phenanthrene derived from morphine with approximately 5–7 times the potency of morphine on a milligram basis. It is a pure μ receptor agonist and is available in oral immediate-release, oral extended-release (Exalgo), rectal, IV, IM, and SC formulations.2 Oral bioavailability is approximately 40–60%. Hydromorphone is glucuronidated to hydromorphone-3-glucuronide (H3G), a neuroexcitatory metabolite analogous to M3G that accumulates in renal failure and can produce cognitive impairment, myoclonus, and allodynia. Of note, hydromorphone does not have an active analgesic metabolite analogous to M6G; its analgesic effect derives solely from the parent compound. In practice, hydromorphone is generally better tolerated than morphine in patients with moderate renal impairment (eGFR 30–60 mL/min/1.73m2), as H3G accumulates more slowly than M3G and M6G at this degree of renal dysfunction, but it is not renally safe in severe impairment. Hydromorphone does not cause the degree of histamine release associated with morphine and is often preferred in patients who experience histamine-mediated reactions with morphine.2
Oxycodone is a semisynthetic thebaine-derived phenanthrene with approximately 1.5 times the oral potency of morphine. It is metabolized by cytochrome P450 3A4 (CYP3A4) (to inactive noroxycodone) and CYP2D6 (cytochrome P450 2D6) (to oxymorphone, an active metabolite with higher μ receptor affinity, though it contributes modestly to clinical effect under normal CYP2D6 function).2 Oral bioavailability is 60–87%, significantly higher than morphine, which accounts for its robust and reliable oral analgesic effect. Available in immediate-release (oxycodone, Roxicodone) and extended-release (OxyContin) formulations, as well as fixed-dose combinations with acetaminophen (Percocet) and aspirin (Percodan). Oxycodone is not available in IV formulation in the United States. Its metabolites are renally excreted; accumulation of oxymorphone can occur in severe renal impairment, though clinically significant toxicity is less well characterized than with morphine metabolites. Drug interactions through CYP3A4 are clinically important: CYP3A4 inhibitors (azole antifungals, macrolide antibiotics, ritonavir, grapefruit juice) increase oxycodone plasma concentrations; CYP3A4 inducers (rifampin, carbamazepine, phenytoin) decrease oxycodone levels and can precipitate withdrawal in dependent patients.2
Hydrocodone is a semisynthetic phenanthrene used primarily for moderate pain and as an antitussive. Like oxycodone, it undergoes CYP2D6-mediated conversion to a more active metabolite (hydromorphone), which may contribute to its analgesic effect in extensive metabolizers; CYP2D6 poor metabolizers may have reduced analgesia.2 Extended-release hydrocodone formulations (Zohydro ER, Hysingla ER) provide around-the-clock dosing; immediate-release hydrocodone is available only in combination products with acetaminophen (Vicodin, Norco) or ibuprofen in the United States, a regulatory constraint based on abuse deterrence rationale. Oral bioavailability is approximately 70–80%.
Oxymorphone is a semisynthetic morphine derivative with approximately twice the potency of oxycodone and six times the potency of oral morphine. Available as oral immediate-release and extended-release (Opana ER) formulations and for parenteral use.2 It does not undergo significant CYP metabolism and is primarily glucuronidated; it is a substrate of UGT2B7 like morphine but produces fewer neuroexcitatory metabolites. Oxymorphone extended-release was withdrawn from US markets by the manufacturer following FDA requests due to abuse concerns related to its formulation, but the immediate-release formulation remains available. Hydromorphone and the high-potency phenanthrene etorphine deserve mention for context. Etorphine has potency approximately 1,000 to 3,000 times that of morphine and is used exclusively for large animal immobilization; even skin contact with undiluted solution can produce life-threatening opioid effects in humans. Carfentanil, a fentanyl congener, is similarly used in veterinary practice for large animal sedation and has been detected as an adulterant in illicit opioid supplies, where nanogram quantities can produce fatal respiratory depression.2
Methadone (Dolophine, Methadose) is a synthetic phenylheptylamine with a pharmacological and pharmacokinetic profile that distinguishes it from other opioid analgesics and demands specialized clinical knowledge before prescribing.4 It is a full μ receptor agonist at both enantiomers, and the R-enantiomer additionally functions as an N-methyl-D-aspartate (NMDA) receptor antagonist; this property may confer advantages in neuropathic pain syndromes and may help attenuate opioid tolerance through NMDA receptor-dependent central sensitization mechanisms. Methadone is highly lipophilic with extensive tissue distribution (volume of distribution (Vd) approximately 4–7 L/kg) and very high protein binding (approximately 85–90%). Its most clinically critical pharmacokinetic feature is its profoundly variable and unpredictable elimination half-life, ranging from 8 to 80 hours or longer in individual patients.4
This variability arises from differences in tissue distribution, genetic variation in CYP3A4 (cytochrome P450 3A4) and CYP2D6 (cytochrome P450 2D6) activity (the primary hepatic enzymes responsible for N-demethylation to the inactive metabolite 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP), the primary inactive metabolite of methadone), enterohepatic recirculation, and lipid solubility. The prolonged and variable half-life creates a major clinical hazard: drug accumulates with repeated dosing and plasma concentrations continue to rise for 3–5 days or more before reaching steady state, meaning that dose titrations based on short-term response can result in delayed overdose as drug continues to accumulate. The clinical principle for methadone titration is to increase dose no more frequently than every 5–7 days to allow steady state to be established and toxicity to manifest before the next adjustment. Methadone has no renally excreted active metabolites, making it pharmacokinetically safe in renal impairment; it undergoes primarily fecal excretion.4
However, methadone uniquely prolongs the QTc interval through blockade of cardiac hERG (IKr) potassium channels; this effect is dose-dependent and significantly amplified by inhibitors of CYP3A4 (azole antifungals, macrolides, HIV protease inhibitors), which raise methadone plasma concentrations.5 The risk of torsades de pointes is substantially elevated when methadone is combined with other QTc-prolonging agents. Guidelines from the American Pain Society and College on Problems of Drug Dependence recommend baseline ECG and repeat ECGs at dose thresholds and dose increases for all patients initiated on methadone for pain.4 Methadone's complex pharmacology makes it appropriate only for clinicians experienced in its use, though for selected patients, particularly those with neuropathic pain, renal impairment, or histories of intolerance to other opioids, it offers substantial advantages.
Levorphanol (Levo-Dromoran) is a synthetic morphinan opioid with approximately four to eight times the potency of oral morphine. Like methadone, it has NMDA receptor antagonist properties and a prolonged half-life (11–16 hours), requiring careful titration to avoid drug accumulation with repeated dosing.2 It is glucuronidated to an inactive metabolite and has no active neuroexcitatory metabolites, making it relatively safer in renal impairment than morphine. Levorphanol is available in the United States but infrequently used due to limited familiarity and the complexity of managing its prolonged half-life. It may have a role in patients with neuropathic pain or renal impairment who are intolerant of more commonly used agents.
Meperidine (Demerol, pethidine) is a synthetic phenylpiperidine that was once one of the most widely prescribed opioid analgesics but has fallen sharply out of favor due to a well-characterized toxicity profile that makes it inferior to alternatives in virtually every clinical setting.6 It is a moderate-potency μ receptor agonist with approximately one-tenth the potency of morphine; it also has weak antimuscarinic properties (contributing to tachycardia, dry mouth, and urinary retention) and local anesthetic activity. Oral bioavailability is approximately 50–60%. Meperidine is hepatically metabolized by N-demethylation (via CYP3A4 (cytochrome P450 3A4) and CYP2C19 (cytochrome P450 2C19)) to normeperidine, a metabolite with a prolonged half-life of 15–30 hours compared to meperidine's 2–4 hours.6 Normeperidine has no analgesic activity but is a potent CNS stimulant that lowers the seizure threshold; at high concentrations it produces CNS excitability, tremors, myoclonus, agitation, and frank seizures; these effects are not reversed by naloxone. Normeperidine accumulates with repeated dosing, in renal impairment, and in the elderly; the ISMP (Institute for Safe Medication Practices) and most major pain societies identify meperidine as a high-alert medication with limited indications.
Its only well-supported current clinical indication is short-term treatment of drug-induced rigors (e.g., from amphotericin B infusion or post-anesthesia shivering), where its spasmolytic properties at kappa receptor and N-methyl-D-aspartate (NMDA) receptor level are useful at single low doses.6 Meperidine should not be used for routine pain management, should never be used in renal impairment, should not be used in patients receiving monoamine oxidase (MAO) inhibitors (risk of potentially fatal serotonin syndrome or excitatory reaction), and should not be used for repeated dosing in any patient. Despite this well-established evidence base, meperidine continues to be prescribed in some settings, representing an opportunity for prescriber education.
Fentanyl (Sublimaze) is a fully synthetic phenylpiperidine opioid approximately 100 times more potent than morphine on a weight basis. It is a highly selective and pure μ receptor agonist with no active metabolites; hepatic CYP3A4-mediated N-dealkylation yields inactive norfentanyl, and renal excretion of unchanged drug is minimal.2 This absence of active or neuroexcitatory renally excreted metabolites makes fentanyl one of the safest opioids in renal impairment. Its extreme lipophilicity (octanol-water partition coefficient approximately 800 times that of morphine) produces rapid CNS penetration and onset of action within 1–2 minutes of IV administration, but also rapid redistribution out of the CNS into peripheral tissues, producing a short duration of effect of 30–60 minutes after a single IV bolus.
With repeated boluses or continuous infusion, fentanyl accumulates extensively in fat and muscle, and the context-sensitive half-time, the time required for plasma concentration to fall by 50% after terminating an infusion, increases dramatically with infusion duration, from approximately 20 minutes after a short infusion to several hours after a prolonged infusion.2 This context-sensitive accumulation is clinically important in the ICU, where patients on prolonged fentanyl infusions may have delayed awakening and prolonged respiratory depression after the infusion is discontinued. Fentanyl is available in more delivery formulations than virtually any other opioid: intravenous bolus and infusion; epidural and intrathecal; transdermal (Duragesic, 25–100 mcg/hr patches); transmucosal immediate-release formulations including the oral transmucosal fentanyl citrate lozenge (Actiq, for cancer breakthrough pain), sublingual tablet (Abstral), buccal tablet (Fentora), buccal soluble film (Onsolis), and intranasal spray (Lazanda). The transmucosal products are approved specifically for breakthrough pain in opioid-tolerant cancer patients and require patients to already be receiving at least 60 mg/day oral morphine equivalent; they are not dose-interchangeable across formulations, and prescribers must dose each product individually based on titration.2
Sufentanil (Sufenta) is approximately 5–10 times more potent than fentanyl (500–1,000 times more potent than morphine), making it the most potent opioid available for clinical use in humans.2 It is used primarily in anesthesia for cardiac surgery and other high-risk procedures requiring profound intraoperative analgesia, and for epidural analgesia in labor. Its highly lipophilic profile provides rapid onset similar to fentanyl. Sufentanil is metabolized by CYP3A4 to inactive metabolites. A sublingual sufentanil tablet system (Zalviso) is approved in Europe for acute pain management in medically supervised settings; FDA approval in the United States has been sought. Alfentanil (Alfenta) is approximately 10–25 times less potent than fentanyl and has a shorter context-sensitive half-time due to a smaller volume of distribution and higher plasma protein binding, which limits tissue sequestration.2 Onset is slightly slower than fentanyl but duration after a bolus dose is shorter (5–10 minutes). Alfentanil is used primarily as an induction and maintenance agent in anesthesia and for short painful procedures requiring brief intense analgesia. CYP3A4 inhibitors significantly prolong its effect.
Remifentanil (Ultiva) has a unique pharmacokinetic profile that sets it apart from all other opioids.2 It is metabolized not by hepatic CYP enzymes but by nonspecific plasma and tissue esterases that cleave its ester bond, producing a metabolite (remifentanil acid) with negligible opioid activity. This esterase-mediated metabolism is organ-independent and produces an ultra-short and highly predictable context-sensitive half-time of approximately 3–5 minutes regardless of infusion duration; the half-time does not increase with prolonged infusion, in contrast to all other fentanyl congeners.2 Remifentanil must be administered as a continuous IV infusion and never as a bolus for analgesia during monitored procedures, because its extreme potency combined with rapid delivery can produce apnea within 60–90 seconds. Due to its ultra-short action, remifentanil does not provide any residual postoperative analgesia; transition to a longer-acting opioid or non-opioid analgesic must be planned before its infusion is terminated. Remifentanil is associated with a higher incidence of opioid-induced hyperalgesia (OIH) than other opioids, likely due to the intense, rapid μ receptor activation and NMDA receptor sensitization it produces; this contributes to acute postoperative pain that may be paradoxically worse following remifentanil-based anesthesia.7
Codeine is a naturally occurring phenanthrene alkaloid that is a prodrug requiring CYP2D6 (cytochrome P450 2D6)-mediated conversion to morphine for approximately 10–15% of its analgesic efficacy; it also has modest direct opioid activity at μ receptors.3 At recommended doses it produces mild-to-moderate analgesia, roughly equivalent to acetaminophen for many pain types. Its primary utility is as an antitussive, though even this indication is increasingly questioned by evidence showing limited efficacy and availability of safer alternatives. Codeine is available in combination products with acetaminophen and is also used as an antidiarrheal. The critical clinical issue with codeine is its CYP2D6 pharmacogenomics: CYP2D6 poor metabolizers (7–10% of white populations, higher in some Asian populations) obtain no meaningful analgesia and are at lower risk for opioid adverse effects; CYP2D6 ultrarapid metabolizers generate excess morphine rapidly, creating toxicity risk that is especially dangerous in breastfeeding, where neonatal exposure to high morphine concentrations in breast milk has caused neonatal death.3
The FDA contraindicated codeine in breastfeeding women and in pediatric patients under 12 years of age following these safety signals, and many guidelines now recommend avoiding codeine in children entirely. Codeine is approximately 10% as potent as morphine on a weight basis; 30 mg of codeine is roughly equianalgesic to 3 mg of parenteral morphine.
Tramadol (Ultram) is a centrally acting analgesic with a dual mechanism: weak μ receptor agonism of the O-desmethyltramadol (M1) metabolite, generated by CYP2D6, and inhibition of serotonin and norepinephrine reuptake.2 The SNRI activity of tramadol produces an analgesic effect independent of opioid receptors, which is why tramadol has historically been classified outside the controlled substance scheduling in some contexts, though it is now Schedule IV in the United States. Tramadol's pharmacology creates several important clinical hazards. It lowers the seizure threshold, both through its serotonergic activity and through a direct mechanism, and seizures have occurred at therapeutic doses as well as in overdose. It carries significant risk of serotonin syndrome when combined with SSRIs, SNRIs, monoamine oxidase (MAO) inhibitors, linezolid, or other serotonergic agents; this risk is underappreciated in practice given the frequency with which these drug combinations occur.
CYP2D6 poor metabolizers generate less M1 and obtain less analgesia; CYP2D6 ultrarapid metabolizers generate excess M1, creating overdose risk.3 Tramadol's opioid partial-dependence profile means that abrupt discontinuation can precipitate a withdrawal syndrome with both opioid and SNRI features (including flu-like symptoms, dysphoria, paresthesias, and rebound pain). Tapentadol (Nucynta) combines μ agonism with norepinephrine reuptake inhibition without significant serotonin reuptake inhibition, potentially offering a more favorable interaction profile and lower seizure risk than tramadol, though clinical data on comparative safety are limited.2
Buprenorphine (Buprenex, Butrans, Suboxone, Subutex, Belbuca, Sublocade) is a thebaine-derived semisynthetic phenanthrene with unique receptor pharmacology that places it in a category of its own among clinically used opioids.8 It is a partial agonist at μ receptors, a partial agonist or antagonist at κ receptors, and an antagonist at δ receptors and opioid receptor-like 1 receptor (ORL1)/nociceptin opioid peptide receptor (NOP) receptors. Its two defining pharmacological characteristics are an extremely high μ receptor binding affinity (Ki approximately 0.1–1 nM, among the highest of any opioid) and slow receptor dissociation kinetics; it dissociates from mu-opioid receptor (MOR) far more slowly than full agonists.
These properties produce several critical clinical consequences. First, buprenorphine's high-affinity, slow-dissociation receptor occupancy means it effectively blocks the binding and effect of subsequently administered full agonists at typical doses; this is the basis of its use in opioid use disorder treatment as a maintenance agent that attenuates the euphoric effect of illicit opioid use. Second, its partial agonism produces a ceiling effect for respiratory depression, at high occupancy, further dose escalation does not increase respiratory depression, providing a substantially wider therapeutic index for overdose than full agonists. Third, its extremely high receptor affinity means it can displace full agonists from receptors, precipitating acute withdrawal in physically dependent patients. Standard clinical practice requires patients to be in moderate opioid withdrawal (Clinical Opiate Withdrawal Scale (COWS) score ≥8–12) before initiating buprenorphine to minimize this risk.8
Oral bioavailability of sublingual buprenorphine is approximately 30–50%; transdermal bioavailability is approximately 15%. The extended-release subcutaneous injectable formulation (Sublocade) provides stable plasma concentrations for one month after a single injection. Hepatic glucuronidation and CYP3A4 (cytochrome P450 3A4)-mediated N-dealkylation to norbuprenorphine are the primary metabolic pathways; neither parent drug nor metabolites accumulate significantly in renal impairment, making buprenorphine one of the safest opioids in CKD. An important management issue has emerged regarding surgical patients on buprenorphine maintenance therapy: the historical practice of discontinuing buprenorphine preoperatively is now generally discouraged, as evidence suggests patients who continue buprenorphine through surgery have better pain management outcomes than those in whom it is discontinued; postoperative analgesia is achieved by dose escalation of buprenorphine itself or by adding full agonists at higher doses sufficient to compete for receptor occupancy.8
Nalbuphine (Nubain) is a synthetic phenanthrene derivative with κ receptor agonism and μ receptor antagonism at higher doses.2 It produces analgesia primarily through κ receptor activation, which is less potent and qualitatively different from μ-mediated analgesia. Like other κ agonists, nalbuphine can produce dysphoria and psychotomimetic effects, though less consistently than pentazocine. Its mixed agonist-antagonist profile means it can precipitate withdrawal in μ-receptor-dependent patients and has a ceiling effect for analgesia analogous to buprenorphine's ceiling for respiratory depression. Nalbuphine is available only parenterally in the United States; it is not orally bioavailable. It is used occasionally in anesthesia for its opioid-sparing and antipruritic properties (κ agonism attenuates μ-mediated pruritus without fully reversing analgesia) and has some utility in the management of opioid-induced pruritus from neuraxial administration.
Butorphanol (Stadol) is a morphinan derivative with κ agonist and partial μ agonist or antagonist activity.2 It is available as an intranasal spray and parenterally. Like nalbuphine, it can produce dysphoria, and its mixed receptor profile means it can precipitate withdrawal. Butorphanol has a ceiling effect for respiratory depression but also a ceiling effect for analgesia, limiting its utility in severe pain. It has been used for migraine treatment (nasal formulation) and as an analgesic adjunct in anesthesia, but its relatively unfavorable adverse effect profile and availability of better alternatives have diminished its clinical role.
Pentazocine (Talwin) is the oldest mixed agonist-antagonist in clinical use and was originally developed with the explicit goal of creating a non-addictive opioid; that goal it did not achieve.2 It is a κ agonist and partial μ antagonist with a particularly high propensity for dysphoria and psychotomimetic effects (depersonalization, visual hallucinations, mood disturbances), which significantly limits its tolerability. The oral formulation is combined with naloxone (Talwin NX) to deter parenteral abuse; the naloxone is inactive orally but precipitates withdrawal if the tablet is dissolved and injected. Pentazocine has largely been displaced from practice by better-tolerated agents. Dezocine is a synthetic aminotetrahydrobenzazocine with mixed agonist-antagonist properties similar to pentazocine and nalbuphine. It is not approved in the United States for general use but remains available in China, where it is widely used parenterally. Its κ agonist-μ partial antagonist profile produces moderate analgesia with a ceiling on respiratory depression; like other agents in this class, it can precipitate withdrawal.2 Dezocine also has norepinephrine reuptake inhibition activity, which may contribute to its analgesic effect through descending monoaminergic pathways.
Equianalgesic dosing refers to the identification of doses of different opioids or different routes of the same opioid that produce approximately equivalent analgesic effect. Equianalgesic tables are derived primarily from single-dose crossover studies in opioid-naive subjects and represent estimates, not precise equivalences; individual pharmacokinetic and pharmacodynamic variability means that equianalgesic conversions are starting points for clinical titration, not precise dose prescriptions.1 The most clinically important equianalgesic relationships to know are: oral morphine 30 mg ≈ IV/IM morphine 10 mg (3:1 oral-to-parenteral ratio); oral oxycodone 20 mg ≈ oral morphine 30 mg (oxycodone approximately 1.5 times the oral potency of morphine); IV hydromorphone 1.5 mg ≈ IV morphine 10 mg (approximately 6.7:1 ratio); oral hydromorphone 7.5 mg ≈ oral morphine 30 mg; transdermal fentanyl 25 mcg/hr ≈ oral morphine 45–90 mg/day (wide variation in published conversion ratios, conservative estimates recommended).
Methadone is uniquely dose-dependent in its equianalgesic relationship with morphine: the ratio increases as the prior morphine dose increases, ranging from approximately 4:1 (morphine:methadone) at low morphine doses to 12:1 or higher at high morphine doses; this non-linear relationship reflects receptor-level differences in partial tolerance and should prompt extra caution and dose reduction when converting from high morphine doses to methadone.4
Opioid rotation, switching from one opioid to another, is indicated when a patient experiences inadequate analgesia despite dose escalation, intolerable adverse effects, development of drug-specific toxic metabolite accumulation (as with M3G/M6G in renal failure), or changes in route availability.1 The key principle underlying safe opioid rotation is incomplete cross-tolerance: patients tolerant to one opioid are not fully cross-tolerant to another, because different opioids have different intrinsic efficacies, receptor binding kinetics, and interactions with tolerance mechanisms.
As a result, the equianalgesic dose of the new opioid will be more potent relative to the patient's actual level of tolerance than it would be in a fully opioid-naive patient. The standard clinical approach is to calculate the equianalgesic dose of the new opioid from the current total daily dose of the existing opioid, then reduce the calculated dose by 25–50% to account for incomplete cross-tolerance.1 The reduction is larger (closer to 50%) when rotating from very high doses of the prior opioid or when the new opioid is methadone, and smaller (25%) when rotating due to inadequate analgesia (since some additional effect from the new opioid may be desirable). The new opioid should then be titrated to effect, with careful reassessment in the 24–72 hours following the rotation.
Opioid analgesics participate in a broad range of drug interactions that are responsible for a disproportionate share of opioid-related adverse events, including overdose deaths. Understanding these interactions requires distinguishing pharmacokinetic interactions, which alter opioid plasma concentrations through effects on absorption, metabolism, or excretion, from pharmacodynamic interactions, which alter the clinical effect of opioids without changing plasma concentrations. Both categories carry serious clinical risk and must be systematically considered whenever opioids are prescribed alongside other medications. The most dangerous pharmacodynamic interactions involve co-administration of opioids with other central nervous system (CNS) depressants. Benzodiazepines and opioids together produce synergistic, not merely additive, respiratory depression through complementary mechanisms: opioids blunt the hypercapnic ventilatory drive through mu-opioid receptor (MOR) activation in medullary respiratory centers, while benzodiazepines potentiate gamma-aminobutyric acid type A (GABA-A) receptor-mediated inhibition throughout the CNS, including respiratory control circuits.910 The combination of an opioid and a benzodiazepine was implicated in approximately 30% of prescription opioid overdose deaths in the United States before regulatory restrictions on co-prescribing were introduced, and the 2022 Centers for Disease Control and Prevention (CDC) Clinical Practice Guideline explicitly recommends avoiding this combination when possible.
When co-prescribing is clinically unavoidable, for example, in patients with both chronic pain and anxiety disorders, or in patients requiring procedural sedation, both agents should be prescribed at the lowest effective doses, and patients and caregivers should receive naloxone and specific education on overdose recognition. Alcohol produces respiratory depression through GABA-A and N-methyl-D-aspartate (NMDA) receptor mechanisms that are additive with opioid-induced respiratory depression; patients on opioid therapy must be counseled that alcohol substantially raises their overdose risk even at amounts that would ordinarily be safe for that individual. Gabapentinoids, specifically gabapentin and pregabalin, have emerged as an important contributor to opioid-related respiratory depression through their inhibition of voltage-gated calcium channels in brainstem respiratory neurons; observational studies have shown that the combination of opioids with gabapentinoids increases the risk of opioid-related mortality, and this combination warrants the same caution as the opioid-benzodiazepine combination in high-risk patients.910 Muscle relaxants, sedating antihistamines (particularly diphenhydramine and promethazine), and first-generation antipsychotics all add to CNS depression and respiratory suppression risk in patients receiving opioids.
Serotonin syndrome is the most important pharmacodynamic toxicity from opioid combinations with serotonergic agents, and it is both underrecognized and potentially fatal.1011 Tramadol is the opioid with the highest intrinsic serotonergic activity, inhibiting neuronal serotonin reuptake through a mechanism analogous to a selective serotonin reuptake inhibitor (SSRI). When tramadol is combined with SSRIs, serotonin-norepinephrine reuptake inhibitors (SNRIs), monoamine oxidase inhibitors (MAOIs), linezolid (which has MAOI activity), methylene blue, tricyclic antidepressants (TCAs), triptans, lithium, or other serotonergic agents, the resulting serotonin excess can precipitate the serotonin syndrome triad of mental status changes, autonomic instability, and neuromuscular abnormalities.1011 The neuromuscular findings, clonus (particularly ankle and inducible clonus), hyperreflexia, and diaphoresis, differentiate serotonin syndrome from neuroleptic malignant syndrome and from anticholinergic toxicity.
Tramadol-MAOI combinations carry the most severe risk and are absolutely contraindicated; tramadol combined with SSRIs or SNRIs carries a lower but clinically meaningful risk that is dose-dependent and substantially increased in CYP2D6 (cytochrome P450 2D6) ultrarapid metabolizers who generate excess tramadol active metabolite. Meperidine shares tramadol's serotonin toxicity risk with MAOIs through a different mechanism (serotonin reuptake inhibition); co-administration of meperidine with any MAOI is absolutely contraindicated and has produced fatal excitatory reactions. Fentanyl and methadone have weak serotonin reuptake inhibitory properties, and clinical cases of serotonin syndrome with these agents in combination with serotonergic drugs have been reported, though the risk is substantially lower than with tramadol or meperidine at standard analgesic doses.1011
Pharmacokinetic drug interactions with opioids are dominated by the cytochrome P450 (CYP) enzyme system, particularly CYP3A4 (cytochrome P450 3A4) and CYP2D6. CYP3A4 is responsible for the primary hepatic metabolism of fentanyl, methadone, buprenorphine (partial), alfentanil, sufentanil, and oxycodone.1012 Strong CYP3A4 inhibitors, including azole antifungals (fluconazole, itraconazole, voriconazole, ketoconazole), macrolide antibiotics (clarithromycin, erythromycin, but not azithromycin), human immunodeficiency virus (HIV) protease inhibitors (ritonavir, lopinavir, atazanavir), and grapefruit juice, reduce CYP3A4 activity and can substantially elevate plasma concentrations of CYP3A4-dependent opioids, converting a stable therapeutic dose into a potentially toxic one. The clinical consequence is greatest for agents with narrow therapeutic windows and steep dose-response curves at the upper end; fentanyl and methadone in particular carry this risk. When a patient on a stable fentanyl regimen is initiated on fluconazole for a fungal infection, for example, the resulting increase in fentanyl plasma concentration may be sufficient to produce sedation and respiratory depression within days; the same interaction applies to methadone, where CYP3A4 (cytochrome P450 3A4) inhibition can not only raise methadone plasma concentrations but also prolong the QTc interval through higher systemic methadone exposure.1012
Strong CYP3A4 inducers, rifampin (rifampicin), carbamazepine, phenytoin, phenobarbital, St. John's wort, and efavirenz, dramatically accelerate CYP3A4-dependent opioid metabolism, reducing plasma concentrations and potentially precipitating opioid withdrawal in physically dependent patients or inadequate analgesia in patients with pain. Rifampin co-administration with methadone has produced opioid withdrawal requiring methadone dose increases of 50% or more; patients on opioid maintenance therapy being treated for tuberculosis (TB) with rifampin-containing regimens require close monitoring and often substantial dose escalation. CYP2D6 interactions primarily affect codeine, tramadol, and hydrocodone, which depend on CYP2D6 for conversion to their more active metabolites. CYP2D6 inhibitors, fluoxetine, paroxetine, bupropion, duloxetine, and quinidine among others, reduce this conversion and can produce analgesic failure in patients relying on codeine or tramadol, while simultaneously increasing accumulation of the parent compound; this interaction has particular clinical significance in patients prescribed tramadol or codeine while also receiving fluoxetine or paroxetine for depression.1012
Opioid interactions with anticoagulants deserve specific mention because they are clinically relevant and incompletely appreciated. Tramadol inhibits platelet aggregation through serotonergic mechanisms and may potentiate the anticoagulant effect of warfarin through CYP2C9 (cytochrome P450 2C9) inhibition, raising the international normalized ratio (INR) in patients on warfarin; INR monitoring is appropriate when tramadol is initiated or discontinued in warfarin-treated patients.912 Aspirin and non-steroidal anti-inflammatory drugs (NSAIDs) are frequently combined with opioids in multimodal analgesic regimens; while the combination enhances analgesia through complementary mechanisms, NSAIDs increase gastrointestinal bleeding risk that may be potentiated in the context of opioid-induced constipation, straining, and fecal impaction.
The interaction between opioids and monoamine oxidase inhibitors is among the most dangerous in all of clinical pharmacology and merits systematic attention.1011 MAOIs, including phenelzine, tranylcypromine, selegiline (at higher doses), isocarboxazid, and the reversible MAOI moclobemide, interact with opioids through two distinct mechanisms. The excitatory or serotonergic reaction occurs with meperidine and tramadol (and to a lesser extent fentanyl and methadone): MAOI-mediated elevation of synaptic serotonin interacts with opioid serotonin reuptake inhibition to produce an acute, life-threatening serotonin toxicity syndrome characterized by hyperthermia, agitation, seizures, and cardiovascular instability.
The depressive or potentiation reaction occurs with morphine and most other full MOR agonists: MAOIs inhibit hepatic oxidative metabolism of these agents, producing dramatically elevated opioid plasma concentrations with resultant excessive CNS and respiratory depression. Meperidine is absolutely contraindicated with MAOIs regardless of timing; other opioids should be used with extreme caution or avoided within 14 days of stopping an irreversible MAOI (2 days for reversible MAOIs). The clinical reality is that MAOIs are rarely prescribed in contemporary practice, but patients may be taking selegiline for Parkinson's disease at doses that confer significant MAOI activity, and linezolid, a widely used antibiotic with reversible MAOI properties, carries the same interaction risk when combined with serotonergic opioids.1013
Brant JM. Opioid equianalgesic conversion: the right dose. Clin J Oncol Nurs. 2001;5(4):163–165
doi:10.1188/01.CJON.163-165Trescot AM, Datta S, Lee M, Hansen H. Opioid pharmacology. Pain Physician. 2008;11(2 Suppl):S133–S153. PMID: 18443637
Crews KR, Gaedigk A, Dunnenberger HM, et al. Clinical Pharmacogenetics Implementation Consortium guidelines for cytochrome P450 2D6 genotype and codeine therapy: 2014 update. Clin Pharmacol Ther. 2014;95(4):376–382
doi:10.1038/clpt.2013.254Chou R, Cruciani RA, Fiellin DA, et al. Methadone safety: a clinical practice guideline from the American Pain Society and College on Problems of Drug Dependence, in collaboration with the Heart Rhythm Society. J Pain. 2014;15(4):321–337
doi:10.1016/j.jpain.2014.01.494Krantz MJ, Martin J, Stimmel B, Mehta D, Haigney MC. QTc interval screening in methadone treatment. Ann Intern Med. 2009;150(6):387–395
doi:10.7326/0003-4819-150-6-200903170-00103Latta KS, Ginsberg B, Barkin RL. Meperidine: a critical review. Am J Ther. 2002;9(1):53–68
doi:10.1097/00045391-200201000-00010Fletcher D, Martinez V. Opioid-induced hyperalgesia in patients after surgery: a systematic review and a meta-analysis. Br J Anaesth. 2014;112(6):991–1004
doi:10.1093/bja/aeu137Walsh SL, Eissenberg T. The clinical pharmacology of buprenorphine: extrapolating from the laboratory to the clinic. Drug Alcohol Depend. 2003;70(2 Suppl):S13–S27
doi:10.1016/s0376-8716(03)00056-xPark TW, Saitz R, Ganoczy D, Ilgen MA, Bohnert AS. Benzodiazepine prescribing patterns and deaths from drug overdose among US veterans receiving opioid analgesics: case-cohort study. BMJ. 2015;350:h2698
doi:10.1136/bmj.h2698Dowell D, Ragan KR, Jones CM, Baldwin GT, Chou R. CDC clinical practice guideline for prescribing opioids for pain — United States, 2022. MMWR Recomm Rep. 2022;71(3):1–95
doi:10.15585/mmwr.rr7103a1Boyer EW, Shannon M. The serotonin syndrome. N Engl J Med. 2005;352(11):1112–1120
doi:10.1056/NEJMra041867Overholser BR, Foster DR. Opioid pharmacokinetic drug-drug interactions. Am J Manag Care. 2011;17(Suppl 11):S276–S287. PMID: 21999760
Trescot AM. Genetics and implications in perioperative analgesia. Best Pract Res Clin Anaesthesiol. 2014;28(2):153–166
doi:10.1016/j.bpa.2014.03.004