Chapter 7 Opioid Agonists and Antagonists PDF

7 Opioid Agonists and Antagonists C H A P TER Kenneth C. Cummings III • Mohamed A. Naguib† Opioids are a cornerstone of modern perioperative care and pain management. The modern word “opium” is derived from the Greek word opion (“poppy juice”); the opium poppy (Papaver somniferum) is the source of...

7 Opioid Agonists and Antagonists C H A P TER Kenneth C. Cummings III • Mohamed A. Naguib† Opioids are a cornerstone of modern perioperative care and pain management. The modern word “opium” is derived from the Greek word opion (“poppy juice”); the opium poppy (Papaver somniferum) is the source of 20 distinct alkaloids. Written mention of the medicinal use of poppy juice dates back to at least 300 BC, although religious use likely goes back much further.1 Drugs derived from opium are referred to as opiates. Morphine, the best known opiate, was isolated in 1803, followed by codeine in 1832 and papaverine in 1848. Morphine can be synthesized, but it is more easily derived from opium. The term narcotic is derived from the Greek word for stupor and traditionally has been used to refer to potent morphine-like analgesics with the potential to produce physical dependence. The development of synthetic drugs with morphine-like properties has led to the use of the term opioid to refer to all exogenous substances, natural and synthetic, that bind specifically to any of several subpopulations of opioid receptors and produce at least some agonist (morphine-like) effects. Opioids are unique in producing analgesia without loss of touch, proprioception, or consciousness. A convenient classification of opioids includes opioid agonists, opioid agonist-antagonists, and opioid antagonists (Table 7.1). TABLE 7.1 Classification of opioid agonists and antagonists Agonists Agonists-antagonists Antagonists Morphine Pentazocine Naloxone Morphine-6-glucuronide Butorphanol Naltrexone Meperidine Nalbuphine Nalmefene Sufentanil Buprenorphine Fentanyl Nalorphine Alfentanil Bremazocine Remifentanil Dezocine Carfentanil Meptazinol Codeine Hydromorphone Oxymorphone Oxycodone Hydrocodone Propoxyphene Methadone Tramadol Heroin (diacetylmorphine) Chemical Structure of Opium Alkaloids The active components of opium can be divided into two distinct chemical classes: phenanthrenes and benzylisoquinolines. The principal phenanthrene alkaloids present in opium are morphine, codeine, and thebaine (Figure 7.1). The principal benzylisoquinoline alkaloids present in opium, which lack analgesic activity, are papaverine and noscapine. FIGURE 7.1 Chemical structures of opium alkaloids. Phenanthrene (A) and benzylisoquinoline (B) alkaloids. The three rings of the phenanthrene core are composed of 14 carbon atoms. The fourth piperidine ring includes a tertiary amine nitrogen and is present in most opioid agonists. At pH 7.4, the tertiary amine nitrogen is highly ionized, making the molecule water soluble. These are chiral molecules, with levorotatory isomers being biologically active at opioid receptors. Semisynthetic Opioids Simple modification of the morphine molecule yields many derivative compounds with differing properties. For example, substitution of a methyl group for the hydroxyl group on carbon 3 results in methylmorphine (codeine). Substitution of acetyl groups on carbons 3 and 6 results in diacetylmorphine (heroin). Hydromorphone has a carbonyl group instead of hydroxyl at position 6 and lacks a double bond between carbons 7 and 8. Thebaine has insignificant analgesic activity but serves as the precursor for etorphine (analgesic potency >1,000 times morphine). Synthetic Opioids Synthetic opioids contain the phenanthrene nucleus of morphine but are manufactured by synthesis rather than chemical modification of morphine. Morphine derivatives (levorphanol), methadone derivatives, benzomorphan derivatives (pentazocine), and phenylpiperidine derivatives (meperidine, fentanyl) are examples of groups of synthetic opioids. There are similarities in the molecular weights (236-326) and pKs of phenylpiperidine derivatives and amide local anesthetics. Fentanyl, sufentanil, alfentanil, and remifentanil (Figure 7.2) are synthetic opioids that are widely used to supplement general anesthesia or as primary anesthetic drugs in very high doses. There are important clinical differences between these opioids.2–4 The major pharmacodynamic differences between these drugs are potency and rate of equilibration between the plasma and the site of drug effect (biophase). FIGURE 7.2 Synthetic opioid agonists. Opioid Receptors Opioid receptors are classified as μ, δ, and κ receptors (Table 7.2).5,6 The names of the three subtypes developed from the ligands originally found to bind to them or their tissue of origin (mu—morphine, kappa—ketocyclazocine, delta—isolated from mouse vas deferens). These opioid receptors belong to a superfamily of seven transmembrane-segment guanine (G) protein–coupled receptors that includes muscarinic, adrenergic, and somatostatin receptors. The opioid receptors have been cloned, and their amino acid sequences defined.7,8 A single μ-receptor gene has been identified, and six distinct μ receptors have been identified. TABLE 7.2 Classification of opioid receptorsa Effect μ1b μ2b Analgesia (supraspinal, spinal) κ δ Analgesia (spinal) Analgesia (supraspinal, spinal) Analgesia (supraspinal, spinal) Euphoria Depression of ventilation Dysphoria, sedation Depression of ventilation Low abuse potential Physical dependence Low abuse potential Physical dependence Miosis Miosis Constipation (marked) Constipation (minimal) Bradycardia Hypothermia Urinary retention Agonists Antagonists Diuresis Urinary retention Dynorphins Enkephalins Endorphinsc Endorphinsc Morphine Morphine Synthetic opioids Synthetic opioids Naloxone Naloxone Naloxone Naloxone Naltrexone Naltrexone Naltrexone Naltrexone Nalmefene Nalmefene Nalmefene Nalmefene aAdapted from Atcheson R, Lambert DG. Update on opioid receptors. Br J Anaesth. 1994;73(2):132-134. Copyright © 1994 Elsevier. With permission. bThe existence of specific μ and μ receptors is not supported based on cloning studies of μ receptors. 1 2 cμ Receptors seem to be a universal site of action for all endogenous opioid receptors. In the brain, opioid receptors are primarily found in the periaqueductal gray, locus ceruleus, and the rostral ventral medulla. In the spinal cord, opioid receptors are found both on interneurons and primary afferent neurons in the dorsal horn. Consequently, direct application of opioid agonists to the spinal cord can produce intense analgesia.9 Outside the central nervous system (CNS), opioid receptors are found on sensory neurons and immune cells. For example, intraarticular morphine is known to produce analgesia after knee surgery, presumably through action on peripheral nerves.10 Immune cells recruited to sites of inflammation also secrete opioid peptides to provide local analgesia.11 This may be useful in drug development. For example, lowering the pKa of an opioid compound should cause its preferential activity at areas of reduced pH such as inflammation.12 μ-Opioid receptors are principally responsible for supraspinal and spinal analgesia. Theoretically, activation of a subpopulation of μ receptors (μ1) is speculated to produce analgesia, whereas μ2 receptors are responsible for hypoventilation, bradycardia, and physical dependence. Nevertheless, cloning of the μ receptors does not support the existence of separate μ1 and μ2 receptor subtypes.6 It is possible that such subtypes result from posttranslational modification of a common precursor protein. Additionally, the response of μopioid receptors to agonists can be significantly affected by β-arrestins, a family of proteins that regulate the activity of G protein–coupled receptors. For example, βarrestins have been demonstrated to promote receptor desensitization (or resensitization) as well as to promote clathrin-mediated endocytosis.13 In principle, pharmacologic modulation of β-arrestin activity could improve the efficacy and tolerability of opioid agonists.14 Activation of κ receptors results in inhibition of neurotransmitter release via N-type calcium channels. Respiratory depression characteristic of μ-receptor activation is less prominent with κ-receptor activation, although dysphoria and diuresis may accompany activation of these receptors. A κ receptor–mediated analgesia may be less effective for high-intensity painful stimulation than μ opioid mediated. Opioid agonist-antagonists often act principally on κ receptors. The δ receptors respond to the endogenous ligands known as enkephalins, and these opioid receptors may serve to modulate the activity of the μ receptors. Functional and physical interactions between these receptor subtypes have been noted.15,16 Heteromerization between μ- and δ-opioid receptors leads to distinct receptor pharmacology in that doses of δ-receptor ligands (agonists and antagonists) too low to trigger signaling can potentiate the binding and signaling of μ-receptor agonists. Chronic, but not acute, morphine treatment results in an increase in the μ-δ heteromers in key areas of the CNS that are implicated in pain processing.17 Previously considered opioid receptors, σ receptors (types 1 and 2) are widespread in the CNS and peripheral tissues. They are now known not to be true opioid receptors but have diverse roles in intracellular signaling, metabolic regulation, mitochondrial metabolism, and other functions.18 Endogenous Pain-Modulating Mechanisms The logical reason for the existence of opioid receptors and endogenous opioid agonists is to function as an endogenous pain suppression system. Once pain is consciously perceived, it has served its purpose and it is reasonable to posit that the ability to dampen this perception would have a survival benefit. Opioid receptors are located in areas of the brain (periaqueductal gray, locus ceruleus, and the rostral ventral medulla ) and spinal cord (substantia gelatinosa) that are involved with pain perception, integration of pain impulses, and responses to pain (Figure 7.3).19 It is speculated that endorphins inhibit the release of excitatory neurotransmitters from terminals of nerves carrying nociceptive impulses. As a result, neurons are hyperpolarized, which suppresses spontaneous discharges and evoked responses. Analgesia induced by electrical stimulation of specific sites in the brain or mechanical stimulation of peripheral areas (acupuncture) most likely reflects release of endorphins.20 Even the analgesic response to a placebo may also involve the release of endorphins. Sustained pain and stress induce the regional release of endogenous opioids interacting with μ-opioid receptors in a number of cortical and subcortical brain regions. The activation of the μ-opioid receptor system is associated with reductions in the sensory and affective ratings of the pain experience, with distinct neuroanatomical involvements.21,22 FIGURE 7.3 Opioid-sensitive pain modulation system. Limbic system areas project to the periaqueductal gray (PAG). The PAG in turn controls afferent pain transmission in the rostroventral medulla. This action can be both inhibitory (red) or facilitatory (green). From Fields H. State-dependent opioid control of pain. Nat Rev Neurosci. 2004;5(7):565-575. In addition, one study demonstrated that positive treatment expectancy substantially enhanced (doubled) the analgesic benefit of remifentanil, whereas negative treatment expectancy abolished remifentanil analgesia.23 These subjective effects were substantiated by significant changes in the neural activity in brain regions involved with the coding of pain intensity. The positive expectancy effects were associated with activity in the endogenous pain modulatory system, and the negative expectancy effects with activity in the hippocampus.23 On the basis of subjective and objective evidence, we contend that an individual’s expectation of a drug’s effect critically influences its therapeutic efficacy and that regulatory brain mechanisms differ as a function of expectancy. Common Opioid Side Effects An ideal opioid agonist would have a high specificity for receptors, producing desirable responses (analgesia) and little or no specificity for receptors associated with side effects. To date, however, all opioids possess similar side effects that vary only in degree. Therefore, a focus on the effects of morphine provides a suitable starting point. Cardiovascular System Morphine, even in large doses, given to supine and normovolemic patients is unlikely to cause direct myocardial depression or hypotension. The same patients changing from a supine to a standing position, however, may manifest orthostatic hypotension and syncope, presumably reflecting morphine-induced impairment of compensatory sympathetic nervous system responses. For example, morphine decreases sympathetic nervous system tone to peripheral veins, resulting in venous pooling and subsequent decreases in venous return, cardiac output, and blood pressure.24 Morphine can also evoke decreases in systemic blood pressure due to druginduced bradycardia or histamine release. Morphine-induced bradycardia results from increased activity over the vagal nerves, which probably reflects stimulation of the vagal nuclei in the medulla. Morphine may also exert a direct depressant effect on the sinoatrial node and acts to slow conduction of cardiac impulses through the atrioventricular node. These actions, may, in part, explain decreased vulnerability to ventricular fibrillation in the presence of morphine. Administration of opioids (morphine) in the preoperative medication or before the induction of anesthesia (fentanyl) tends to slow heart rate during exposure to volatile anesthetics with or without surgical stimulation.25 Opioid-induced histamine release and associated hypotension are variable in both incidence and severity. The magnitude of morphine-induced histamine release and subsequent decrease in systemic blood pressure can be minimized by (1) limiting the rate of morphine infusion to 5 mg/min intravenously (IV), (2) maintaining the patient in a supine to slightly head-down position, and (3) optimizing intravascular fluid volume. Conversely, administration of morphine, 1 mg/kg IV, over a 10-minute period produces substantial increases in the plasma concentrations of histamine that are paralleled by significant decreases in systemic blood pressure and systemic vascular resistance (Figure 7.4).26 It is important to recognize, however, that not all patients respond to this rate of morphine infusion with the release of histamine, emphasizing the individual variability associated with the administration of this drug. In contrast to morphine, the infusion of fentanyl 50 μg/kg IV over a 10-minute period does not cause release of histamine in any patient (see Figure 7.4). Pretreatment of patients with H1- and H2-receptor antagonists does not alter release of histamine evoked by morphine but does prevent changes in systemic blood pressure and systemic vascular resistance.27 FIGURE 7.4 Morphine-induced decreases in systemic blood pressure (BP) and systemic vascular resistance (SVR) are accompanied by increases in the plasma concentration of histamine. Similar changes do not accompany the intravenous administration of fentanyl (*P < .05; **P < .005; mean ± standard error). Reprinted with permission from Rosow CE, Moss J, Philbin DM, et al. Histamine release during morphine and fentanyl anesthesia. Anesthesiology. 1982;56(2):93-96. Copyright © 1982 American Society of Anesthesiologists, Inc. Morphine does not sensitize the heart to catecholamines or otherwise predispose to cardiac dysrhythmias as long as hypercarbia or arterial hypoxemia does not result from ventilatory depression. Tachycardia and hypertension that occur during anesthesia with morphine are not pharmacologic effects of the opioid but rather are responses to painful surgical stimulation that are not suppressed by morphine. Both the sympathetic nervous system and the renin-angiotensin axis contribute to these cardiovascular responses. Large doses of morphine or other opioid agonists may decrease the likelihood that tachycardia and hypertension will occur in response to painful stimulation, but once this response has occurred, administration of additional opioid is unlikely to be effective. During anesthesia, however, opioids are commonly administered with inhaled or IV anesthetics to ensure amnesia. The combination of an opioid agonist such as morphine or fentanyl with nitrous oxide results in cardiovascular depression (decreased cardiac output and systemic blood pressure plus increased cardiac filling pressures), which does not occur when either drug is administered alone.28 Likewise, decreases in systemic vascular resistance and systemic blood pressure may accompany the combination of an opioid and a benzodiazepine, whereas these effects do not accompany the administration of either drug alone (Figure 7.5).29 FIGURE 7.5 Administration of fentanyl (50 μg/kg intravenously [IV] at 400 μg/minute) after injection of diazepam (0.125-0.50 mg/kg IV) is associated with significant decreases in mean arterial pressure (MAP) and systemic vascular resistance (SVR), whereas heart rate (HR) and cardiac index (CI) do not change. Administration of fentanyl in the absence of prior injection of diazepam (0 mg/kg) is devoid of circulatory effects. Reprinted with permission from Tomicheck RC, Rosow CE, Philbin DM, et al. Diazepam-fentanyl interaction: hemodynamic and hormonal effects in coronary artery surgery. Anesth Analg. 1983;62(10):881884. Copyright © 1983 International Anesthesia Research Society. Opioids have been increasingly recognized to play a role in protecting the myocardium from ischemia. Through several mechanisms, most prominently through σ and κ receptors, opioids enhance the resistance of the myocardium to oxidative and ischemic stresses. Mitochondrial adenosine triphosphate–regulated potassium channels appear to be central to this signaling pathway.30 Although once considered opioid receptors, σ receptors are now recognized to play diverse roles in cell signaling, metabolic regulation, and apoptosis and are not considered true opioid receptors.18 Ventilation Because analgesic and ventilatory effects of opioids occur by similar mechanisms, it is assumed that equianalgesic doses of all opioids will produce some degree of ventilatory depression and reversal of ventilatory depression with an opioid antagonist always involves some reversal of analgesia. Opioid-induced depression of ventilation is characterized by decreased responsiveness of these ventilation centers to carbon dioxide as reflected by an increase in the resting arterial partial pressure of CO2 and displacement of the carbon dioxide response curve to the right. Opioid agonists also interfere with pontine and medullary ventilatory centers that regulate the rhythm of breathing, leading to prolonged pauses between breaths and periodic breathing. It is possible that opioid agonists diminish sensitivity to carbon dioxide by decreasing the release of acetylcholine from neurons in the area of the medullary ventilatory center in response to hypercarbia. At the cellular level, there is evidence for β-arrestin mediation of opioid-induced ventilatory depression, which raises the possibility of using “biased ligands” at opioid receptors to mitigate this adverse effect.31 Depression of ventilation produced by opioid agonists is rapid and persists for several hours, as demonstrated by decreased ventilatory responses to carbon dioxide. High doses of opioids may result in apnea, but the patient remains conscious and able to initiate a breath if asked to do so. Death from an opioid overdose is almost invariably due to depression of ventilation. Clinically, depression of ventilation produced by opioids manifests as a decreased frequency of breathing that is often accompanied by a compensatory increase in tidal volume. The incompleteness of this compensatory increase in tidal volume is evidenced by predictable increases in the PaCO2. Many factors influence the magnitude and duration of depression of ventilation produced by opioid agonists. For example, advanced age and the occurrence of natural sleep increase the ventilatory depressant effects of opioids. Conversely, pain from surgical stimulation counteracts depression of ventilation produced by opioids. Likewise, the analgesic effect of opioids slows breathing that has been rapid and shallow due to pain. Opioids produce dose-dependent depression of ciliary activity in the airways. Increases in airway resistance after administration of an opioid are probably due to a direct effect on bronchial smooth muscle and an indirect action due to release of histamine (from some opioids). Cough Suppression Opioids depress cough by effects on the medullary cough centers that are distinct from the effects of opioids on ventilation. The greatest cough suppression occurs with opioids that have bulky substitutions at the number 3 carbon position (codeine). One useful property of dextrorotatory isomers (such as dextromethorphan) is that they can suppress cough but do not produce analgesia or depression of ventilation. They are not, however, devoid of abuse potential.32 Central Nervous System In the absence of hypoventilation, opioids decrease cerebral blood flow and possibly intracranial pressure (ICP). These drugs must be used with caution in patients with head injury because of their (1) associated effects on wakefulness, (2) production of miosis, and (3) depression of ventilation with associated increases in ICP if the PaCO2 becomes increased. Furthermore, head injury may impair the integrity of the blood–brain barrier, with resultant increased sensitivity to opioids. The effect of morphine on the electroencephalogram (EEG) resembles changes associated with sleep. For example, there is replacement of rapid α waves by slower δ waves. Recording of the EEG fails to reveal any evidence of seizure activity after administration of large doses of opioids (see “Fentanyl” section). Opioids do not alter the responses to neuromuscular-blocking drugs. Skeletal muscle rigidity, especially of the thoracic and abdominal muscles, is common when large doses of opioid agonists are administered rapidly and IV.33 Clonic skeletal muscle activity (myoclonus) occurring during administration of opioids may resemble grand mal seizures, but the EEG does not reflect seizure activity. Skeletal muscle rigidity may be related to actions at opioid receptors and involve interactions with dopaminergic and γ-aminobutyric acid–responsive neurons. Miosis is due to an excitatory action of opioids on the autonomic nervous system component of the Edinger-Westphal nucleus of the oculomotor nerve. Tolerance to the miotic effect of morphine is not prominent. Miosis can be antagonized by atropine, and profound arterial hypoxemia in the presence of morphine can still result in mydriasis. Rigidity Rapid IV administration of large doses of an opioid (particularly fentanyl and its derivatives as used in cardiac surgery) can lead to generalized skeletal muscle rigidity. This can be severe enough to interfere with manual ventilation. Although generally termed “chest wall” rigidity, evidence supports the conclusion that the majority of resistance to ventilation is due to laryngeal musculature contraction. Inhibition of striatal release of γ-aminobutyric acid and increased dopamine production are the likely explanations for opioid-induced increased skeletal muscle tone.34 The reported incidence of difficult ventilation after a moderate dose of sufentanil ranges from 84% to 100%.35 Treatment is muscle relaxation with neuromuscular-blocking drugs or opioid antagonism with naloxone. Sedation Postoperative titration of morphine frequently induces sedation that precedes the onset of analgesia.36 The usual recommendation for morphine titration includes a short interval between boluses (5-7 minutes) to allow evaluation of its clinical effect. Sedation occurs in up to 60% of patients during morphine titration and represents a common reason to discontinue morphine titration for postoperative analgesia. The assumption that sleep occurs when pain is relieved is not necessarily accurate, and morphine-induced sedation should not be considered as an indicator of appropriate analgesia during IV morphine titration. Biliary Tract Opioids can cause spasm of biliary smooth muscle, resulting in increases in biliary pressure that may be associated with epigastric distress or biliary colic. This pain may be confused with angina pectoris. Naloxone will relieve pain caused by biliary spasm but not myocardial ischemia. Conversely, nitroglycerin will relieve pain due to either biliary spasm or myocardial ischemia. Equal analgesic doses of fentanyl, morphine, meperidine, and pentazocine increase common bile duct pressure 99%, 53%, 61%, and 15% above predrug levels, respectively.37 During surgery, opioid-induced spasm of the sphincter of Oddi may appear radiologically as a sharp constriction at the distal end of the common bile duct and be misinterpreted as a common bile duct stone. It may be necessary to reverse opioidinduced biliary smooth muscle spasm with naloxone so as to correctly interpret the cholangiogram. Glucagon, 2 mg IV, also reverses opioid-induced biliary smooth muscle spasm and, unlike naloxone, does not antagonize the analgesic effects of the opioid.38 However, biliary muscle spasm does not occur in most patients who receive opioids. Indeed, the incidence of spasm of the sphincter of Oddi is about 3% in patients receiving fentanyl as a supplement to inhaled anesthetics.39 Contraction of the smooth muscles of the pancreatic ducts is probably responsible for increases in plasma amylase and lipase concentrations that may be present after the administration of morphine. Such increases may confuse the diagnosis when acute pancreatitis is a possibility. Gastrointestinal Tract Commonly used opioids such as morphine, meperidine, and fentanyl can produce spasm of the gastrointestinal smooth muscles, resulting in a variety of side effects including constipation, biliary colic, and delayed gastric emptying. Morphine decreases the propulsive peristaltic contractions of the small and large intestines and enhances the tone of the pyloric sphincter, ileocecal valve, and anal sphincter. The delayed passage of intestinal contents through the colon allows increased absorption of water. As a result, constipation often accompanies therapy with opioids and may become a debilitating problem in patients who require chronic opioid therapy, as little tolerance develops to this effect. Of interest, opium was used to treat diarrhea before its use as an analgesic was popularized. Increased biliary pressure occurs when the gallbladder contracts against a closed or narrowed sphincter of Oddi. Passage of gastric contents into the proximal duodenum is delayed because there is increased tone at the gastroduodenal junction. In this regard, preoperative medication that includes an opioid could slow gastric emptying (potentially increase the risk of aspiration) or delay the absorption of orally administered drugs. All these effects may be reversed or prevented by a peripheral-acting opioid antagonist (see “Opioid Antagonists” section). Nausea and Vomiting Opioid-induced nausea and vomiting are caused by direct stimulation of the chemoreceptor trigger zone in the floor of the fourth ventricle. This may reflect the role of opioid agonists as partial dopamine agonists at dopamine receptors in the chemoreceptor trigger zone. Indeed, apomorphine is a profound emetic and is also the most potent of the opioids at dopamine receptors. Stimulation of dopamine receptors as a mechanism for opioid-induced nausea and vomiting is consistent with the antiemetic efficacy of butyrophenones and phenothiazines. Morphine may also cause nausea and vomiting by increasing gastrointestinal secretions and delaying passage of intestinal contents toward the colon. Nausea and vomiting are relatively uncommon in recumbent patients given morphine, suggesting that a vestibular component may also contribute to opioid-induced nausea and vomiting. Genitourinary System Morphine can increase the tone and peristaltic activity of the ureter. In contrast to similar effects on biliary tract smooth muscle, the same opioid-induced effects on the ureter can be reversed by an anticholinergic drug such as atropine. Urinary urgency is produced by opioid-induced augmentation of detrusor muscle tone, but, at the same time, the tone of the urinary sphincter is enhanced, making voiding difficult. Antidiuresis that accompanies administration of morphine to animals has been attributed to opioid-induced release of arginine vasopressin hormone (antidiuretic hormone). In humans, however, administration of morphine in the absence of painful surgical stimulation does not evoke the release of this hormone.40 Furthermore, when morphine is administered in the presence of an adequate intravascular fluid volume, there is no change in urine output. Cutaneous Changes Morphine causes cutaneous blood vessels to dilate. The skin of the face, neck, and upper chest frequently becomes flushed and warm. These changes in cutaneous circulation are in part caused by the release of histamine. Histamine release probably accounts for urticaria and erythema commonly seen at the morphine injection site. In addition, morphine-induced histamine release probably accounts for conjunctival erythema and pruritus. Localized cutaneous evidence of histamine release, especially along the vein into which morphine is injected, does not represent an allergic reaction. Placental Transfer Opioids are readily transported across the placenta.41 Therefore, depression of the neonate can occur as a consequence of administration of opioids to the mother during labor. However, their effects on neonates are quite variable: Maternal administration of morphine may produce greater neonatal depression than meperidine.42 Although fentanyl crosses the placenta, it does not produce significant neonatal depression unless large doses are used. Chronic maternal use of an opioid can result in the development of physical dependence in the fetus. Subsequent administration of naloxone to the neonate can precipitate a lifethreatening neonatal withdrawal syndrome. Drug Interactions The ventilatory depressant effects of some opioids may be exaggerated by amphetamines, phenothiazines, monoamine oxidase inhibitors, and tricyclic antidepressants. For example, patients receiving monoamine oxidase inhibitors may experience exaggerated CNS depression and hyperpyrexia after administration of an opioid agonist, especially meperidine.43 Hormonal Changes Prolonged opioid therapy may influence the hypothalamic-pituitary-adrenal axis and the hypothalamic-pituitary-gonadal axis, leading to endocrine and immune effects.44,45 Morphine may cause a progressive decrease in plasma cortisol concentrations. The main effects of opioids on the hypothalamic-pituitary-gonadal axis involve modulation of hormone release including increased prolactin and decreased luteinizing hormone, follicle-stimulating hormone, testosterone, and estrogen concentrations. Overdose The principal manifestation of opioid overdose is depression of ventilation manifesting as a slow breathing frequency, which may progress to apnea. Pupils are symmetric and miotic unless severe arterial hypoxemia is present, which results in mydriasis. Skeletal muscles are flaccid, and upper airway obstruction may occur. Pulmonary edema commonly occurs, but the mechanism is not known. Hypotension and seizures develop if arterial hypoxemia persists. The triad of miosis, hypoventilation, and coma should suggest overdose with an opioid. Treatment of opioid overdose is mechanical ventilation of the patient’s lungs with oxygen and administration of an opioid antagonist such as naloxone. Administration of an opioid antagonist to treat opioid overdose may precipitate acute withdrawal in dependent patients. With the recent increase in opioidoverdose deaths, the availability of intranasal or intramuscular (IM) naloxone for out-of-hospital treatment is critical. For patients being prescribed high-dose opioids, a prescription for naloxone as a rescue drug is appropriate. Extremely high doses of naloxone may be necessary for individuals using ultra-potent opioids such as carfentanil. Provocation of Coughing Paradoxically, preinduction administration of fentanyl, sufentanil, or alfentanil may be associated with significant reflex coughing.46 The exact cause of opioidinduced cough is unclear but is thought to be due to imbalance between sympathetic and vagal innervation of the airways and/or stimulation of juxtacapillary irritant receptors.47 Morphine and hydromorphone do not appear to cause this reaction. Pharmacodynamic Tolerance and Physical Dependence Pharmacodynamic tolerance and physical dependence with repeated opioid administration are characteristics of all opioid agonists and are among the major limitations of their clinical use. Cross-tolerance develops between all the opioids, although incomplete, as manifested by the reduced doses required with opioid rotation. Tolerance can occur without physical dependence, but the reverse does not seem to occur. Tolerance is the development of the requirement for increased doses of a drug (in this case, an opioid agonist) to achieve the same effect previously achieved with a lower dose. Such acquired tolerance usually takes 2 to 3 weeks to develop with analgesic doses of morphine, although acute tolerance can develop much more quickly with highly potent opioids.48 Tolerance develops to analgesic, euphoric, sedative, depression of ventilation, and emetic effects of opioids but not to their effects on miosis and bowel motility. The potential for physical dependence depends on the agonist effect of opioids. Indeed, physical dependence does not occur with opioid antagonists and is less likely with opioid agonistantagonists. When opioid agonist actions predominate, there often develops, with repeated use, both psychological and physiologic need for the drug. Physical dependence on morphine usually requires about 25 days to develop. Some degree of physical dependence, however, occurs after only 48 hours of continuous medication. When physical dependence is established, discontinuation of the opioid agonist produces a withdrawal abstinence syndrome that may vary depending on individual and environmental factors (Table 7.3).49 Initial symptoms of withdrawal include yawning, diaphoresis, lacrimation, or coryza. Insomnia and restlessness are prominent. Abdominal cramps, nausea, vomiting, and diarrhea reach their peak in 72 hours and then decline over the next 7 to 10 days. During withdrawal, tolerance to morphine is rapidly lost, and the syndrome can be terminated by a modest dose of opioid agonist. The longer the period of abstinence, the smaller the dose of opioid agonist that will be required. TABLE 7.3 Time course of opioid withdrawala Opioid Onset Peak intensity Duration Meperidine 2-6 hours 6-12 hours 4-5 days Fentanyl 2-6 hours 6-12 hours 4-5 days Morphine 6-18 hours 36-72 hours 7-10 days Heroin 6-18 hours 36-72 hours 7-10 days Methadone 24-48 hours 3-21 days 6-7 weeks aAdapted from Mitra S, Sinatra RS. Perioperative management of acute pain in the opioid-dependent patient. Anesthesiology. 2004;101(1):212-227. Pharmacodynamic tolerance has been related to neurologic changes that take place after long-term exposure to the opioid.49 The classic explanation for tolerance to a receptor agonist involved changes occurring at the level of the receptors and involve receptor desensitization. Opioid receptors on the cell membrane surfaces become gradually desensitized by reduced transcription and subsequent decreases in the absolute numbers of opioid receptors (downregulation). A second mechanism proposed to explain pharmacodynamic tolerance involves upregulation of the cyclic adenosine monophosphate (cAMP) system. Acutely, opioids inhibit functional activity of cAMP pathways by blocking adenylate cyclase, the enzyme that catalyzes the synthesis of cAMP. Long-term opioid exposure is associated with gradual recovery of cAMP pathways, and tolerance develops. Increased synthesis of cAMP may be responsible for physical dependence and physiologic changes associated with withdrawal. Upregulation of cAMP has been most clearly demonstrated in the locus ceruleus of the brain. Clonidine, a centrally acting α2-adrenergic agonist that diminishes transmission in sympathetic pathways in the CNS, is an effective drug in suppressing withdrawal signs in persons who are physically dependent on opioids. Tolerance is not due to enzyme induction because no increase in the rate of metabolism of opioid agonists occurs. Long-term pharmacodynamic tolerance characterized by opioid insensitivity may persist for months or years in some individuals and most likely represents persistent neural adaptation.49 In this regard, N-methyl-D-aspartate (NMDA) glutamate receptors are important in the development of opioid tolerance and increased pain sensitivity. Prolonged exposure to opioids activates NMDA receptors via second messenger mechanisms and also downregulates spinal glutamate transporters. The resultant high synaptic concentrations of glutamate and NMDA receptor activation contribute to opioid tolerance and abnormal pain sensitivity (pronociceptive or sensitization process). The observation that treatment with small doses of ketamine (an NMDA receptor antagonist) abolishes the acute opioid tolerance seen with remifentanil supports this hypothesis.50 Opioid Agonists Opioid agonists include but are not limited to morphine, meperidine, fentanyl, sufentanil, alfentanil, and remifentanil (see Table 7.1).51 The most notable feature of the clinical use of opioids is the extraordinary variation in dose requirements for pain management.52 This interindividual variation emphasizes that usual doses of opioids may produce inadequate or excessive opioid effects. Opioid rotation may be useful when dose escalation is not effective in treating pain. Morphine Isolated in 1806 and named for the Greek god of dreams, morphine is the prototype opioid agonist to which all other opioids are compared. In humans, morphine produces analgesia, euphoria, sedation, and a diminished ability to concentrate. Other sensations include nausea, a feeling of body warmth, heaviness of the extremities, dryness of the mouth, and pruritus, especially in the cutaneous areas around the nose. The cause of pain persists, but even low doses of morphine increase the threshold to pain and modify the perception of noxious stimulation such that it is no longer experienced as pain. Continuous, dull pain is relieved by morphine more effectively than is sharp, intermittent pain. Analgesia is most prominent when morphine is administered before the painful stimulus occurs.53 Pharmacokinetics Morphine is well absorbed after IM administration, with onset of effect in 15 to 30 minutes and a peak effect in 45 to 90 minutes. The clinical duration of action is about 4 hours. Morphine can be administered orally for treatment of chronic pain recognizing that absorption from the gastrointestinal tract may be limited. Morphine is usually administered IV in the perioperative period, thus eliminating the unpredictable influence of drug absorption. The peak effect (equilibration time between the blood and brain) after IV administration of morphine is delayed compared with opioids such as fentanyl and alfentanil, requiring about 15 to 30 minutes (Table 7.4). Morphine inhaled as an aerosol from a nebulizer may act on afferent nerve pathways in the airways to relieve dyspnea as associated with lung cancer and associated pleural effusion.54 However, profound depression of ventilation may follow aerosol administration of morphine.55 The onset and duration of the analgesic effects of morphine are similar after IV administration or inhalation via a pulmonary drug delivery system that produces a fine aerosol.56 TABLE 7.4 Pharmacokinetics of opioid agonists Contextsensitive halftime: 4Percent Protein Volume of Elimination hour nonionized binding Clearance distribution Partition half-time infusion pK (pH 7.4) (%) (mL/minute) (L) coefficient (hour) (minute) Effect-site (blood/brain) equilibration time (minute) Morphine 7.9 23 35 1,050 224 1 1.7-3.3 Meperidine 8.5 7 70 1,020 305 32 3-5 Fentanyl 8.4 8.5 84 1,530 335 955 3.1-6.6 260 6.8 Sufentanil 8.0 20 93 900 123 1,727 2.2-4.6 30 6.2 Alfentanil 6.5 89 92 238 27 129 1.4-1.5 60 1.4 Remifentanil 7.3 58 66-93 4,000 30 0.17-0.33 4 1.1 Plasma morphine concentrations after rapid IV injections do not correlate closely with the drug’s pharmacologic activity, likely due to the delay in transit of morphine across the blood–brain barrier. Cerebrospinal fluid (CSF) concentrations of morphine peak 15 to 30 minutes after IV injection and decay more slowly than plasma concentrations (Figure 7.6).57 As a result, the analgesic and ventilatory depressant effects of morphine may not be evident during the initial high plasma concentrations after IV administration of the opioid. Likewise, these same drug effects persist despite decreasing plasma concentrations of morphine. Moderate analgesia probably requires maintenance of plasma morphine concentrations of at least 0.05 μg/mL.58 FIGURE 7.6 Cerebrospinal fluid (CSF) concentrations following intravenous administration of morphine decay more slowly than plasma concentrations. The end-tidal CO2 concentration (PETCO2) remains increased despite a decreasing plasma concentration of morphine (mean ± standard error). Reprinted with permission from Murphy MR, Hug CC Jr. Pharmacokinetics of intravenous morphine in patients anesthetized with enflurane-nitrous oxide. Anesthesiology. 1981;54(3):187-192. Copyright © 1981 American Society of Anesthesiologists, Inc. Only a small amount of administered morphine gains access to the CNS. For example, it is estimated that <0.1% of morphine that is administered IV has entered the CNS at the time of peak plasma concentrations. Reasons for poor penetration of morphine into the CNS include (1) relatively poor lipid solubility, (2) high degree of ionization at physiologic pH, (3) protein binding, and (4) rapid conjugation with glucuronic acid. Alkalinization of the blood, as produced by hyperventilation of the patient’s lungs, will increase the nonionized fraction of morphine and thus enhance its passage into the CNS. Nevertheless, respiratory acidosis, which decreases the nonionized fraction of morphine, results in higher plasma and brain concentrations of morphine than are present during normocarbia (Figure 7.7).59 This suggests that carbon dioxide–induced increases in cerebral blood flow and enhanced delivery of morphine to the brain are more important than the fraction of drug that exists in either the ionized or nonionized fraction. In contrast to the CNS, morphine accumulates rapidly in the kidneys, liver, and skeletal muscles. Morphine, unlike fentanyl, does not undergo significant firstpass uptake into the lungs.60 FIGURE 7.7 Hypercarbia, which decreases the nonionized fraction of morphine, results in a higher brain concentration and longer elimination half-time (t1/2β) than occurs in the presence of normocarbia (*P < .05). Abbreviation: SEM, standard error of the measurement. Reprinted with permission from Finck AD, Ngai SH, Berkowitz BA. Antagonism of general anesthesia by naloxone in the rat. Anesthesiology. 1977;46(4):241-245. Copyright © 1977 American Society of Anesthesiologists, Inc. Metabolism Metabolism of morphine is primarily conjugation with glucuronic acid in hepatic and extrahepatic sites, especially the kidneys. About 75% to 85% of a dose of morphine appears as morphine-3-glucuronide, and 5% to 10% as morphine-6glucuronide (a ratio of 9:1). Morphine-3-glucuronide is detectable in the plasma within 1 minute after IV injection, and its concentration exceeds that of unchanged drug by almost 10-fold within 90 minutes (Figure 7.8).57 An estimated 5% of morphine is demethylated to normorphine, and a small amount of codeine (methylmorphine) may also be formed. Metabolites of morphine are eliminated principally in the urine, with only 7% to 10% undergoing biliary excretion. Morphine-3-glucuronide is detectable in the urine for up to 72 hours after the administration of morphine. A small fraction (1%-2%) of injected morphine is recovered unchanged in the urine. FIGURE 7.8 Morphine glucuronide is detectable in the plasma within 1 minute after intravenous injection, and its concentration exceeds that of unchanged morphine by almost 10-fold within 90 minutes (mean ± standard error). Reprinted with permission from Murphy MR, Hug CC Jr. Pharmacokinetics of intravenous morphine in patients anesthetized with enflurane-nitrous oxide. Anesthesiology. 1981;54(3):187-192. Copyright © 1981 American Society of Anesthesiologists, Inc. Morphine-3-glucuronide is pharmacologically inactive, whereas morphine-6glucuronide produces analgesia and depression of ventilation via its agonist actions at μ receptors.61 In fact, the ventilatory response to carbon dioxide is impacted similarly by morphine and morphine-6-glucuronide (Figure 7.9).62 The duration of action of morphine-6-glucuronide is greater than that of morphine, and it is possible that the majority of analgesic activity attributed to morphine is actually due to morphine-6-glucuronide, especially with long-term administration of morphine.63 Morphine and morphine-6-glucuronide bind to μ-opioid receptors FIGURE 7.10 Plasma concentrations of unchanged morphine (closed circles) and morphine metabolites (open circles) in normal and renal failure patients. Reprinted with permission from Chauvin M, Sandouk P, Scherrmann JM, et al. Morphine pharmacokinetics in renal failure. Anesthesiology. 1987;66(3):327-331. Copyright © 1987 American Society of Anesthesiologists, Inc. Elimination Half-Time After IV administration of morphine, the elimination of morphine-3-glucuronide is somewhat longer than for morphine (see Table 7.4 and Figure 7.8).57 The decrease in the plasma concentration of morphine after initial distribution of the drug is principally due to metabolism because only a small amount of unchanged opioid is excreted in the urine. Plasma morphine concentrations are higher in the elderly than in young adults (Figure 7.11).58 In the first 4 days of life, the clearance of morphine is decreased and its elimination half-time is prolonged compared with that found in older infants.68 This is consistent with the observation that neonates are more sensitive than older children to the respiratory depressant effects of morphine. Patients with renal failure exhibit higher plasma and CSF concentrations of morphine and morphine metabolites than do normal patients, reflecting a smaller volume of distribution (Vd).69 Possible accumulation of morphine-6-glucuronide suggests the need for caution when administering morphine to patients with significant renal dysfunction. Concentrations of morphine in colostrum of parturients receiving patient-controlled analgesia (PCA) with morphine are low, and it is unlikely that significant amounts of drug will be transferred to the breast-fed neonate.70 FIGURE 7.11 The plasma (serum) concentration of morphine increases progressively with advancing age. From Berkowitz BA, Ngai SH, Yang JC, et al. The disposition of morphine in surgical patients. Clin Pharmacol Ther. 1975;17(6):629-635. Copyright © 1975 American Society for Clinical Pharmacology and Therapeutics. Reprinted by permission of John Wiley & Sons, Inc. Sex Sex may affect opioid analgesia, but the direction and magnitude of these differences depend on many interacting variables including the opioid used.71 Morphine exhibits greater analgesic potency and slower speed of offset in women than men.72 This observation is consistent with higher postoperative opioid consumption in men compared with women. Morphine has also been demonstrated to cause greater respiratory depression in women.73 Making broad assertions about the clinical significance of these differences is difficult because pain perception, opioid-related side effects, and sensitivity to opioids all affect opioid dosing requirements and may lead to differences in some contexts but not others.74 Side Effects Side effects described for morphine are also characteristic of other opioid agonists, although the incidence and magnitude may vary. Meperidine First synthesized in 1939, meperidine (also referred to as pethidine) is a synthetic opioid agonist at μ- and κ-opioid receptors and is derived from phenylpiperidine (see Figure 7.1). There are several analogues of meperidine, including fentanyl, sufentanil, alfentanil, and remifentanil. Meperidine shares several structural features that are present in local anesthetics including a tertiary amine, an ester group, and a lipophilic phenyl group. Indeed, meperidine administered intrathecally blocks sodium channels to a degree comparable with lidocaine and can be used to provide surgical anesthesia.75 Structurally, meperidine is similar to atropine, and it possesses a mild atropine-like antispasmodic effect on smooth muscle. Pharmacokinetics Meperidine is about one-tenth as potent as morphine. The duration of action of meperidine is 2 to 4 hours, making it a shorter acting opioid agonist than morphine. In equal analgesic doses, meperidine produces equivalent sedation, euphoria, nausea, vomiting, and depression of ventilation to morphine. Meperidine is absorbed from the gastrointestinal tract, but extensive first-pass hepatic metabolism (up to 80%) limits its oral usefulness. Metabolism Hepatic metabolism of meperidine is extensive, with about 90% of the drug initially undergoing demethylation to normeperidine and hydrolysis to meperidinic acid.76 Normeperidine subsequently undergoes hydrolysis to normeperidinic acid. Urinary excretion is the principal elimination route and is pH dependent. For example, if the urinary pH is <5, as much as 25% of meperidine is excreted unchanged. Indeed, acidification of the urine can be considered in an attempt to speed elimination of meperidine. Decreased renal function can predispose to accumulation of normeperidine. Normeperidine has an elimination half-time of 15 hours (35 hours in patients in renal failure) and can be detected in urine for as long as 3 days after administration of meperidine. This metabolite is about one-half as active as meperidine as an analgesic. In addition, normeperidine produces CNS stimulation. Normeperidine toxicity manifesting as myoclonus and seizures is most likely during prolonged administration of meperidine as during PCA, especially in the presence of impaired renal function.76 Normeperidine may also be important in meperidine-induced delirium (confusion, hallucinations), which has been observed in patients receiving the drug for longer than 3 days, corresponding to accumulation of this active metabolite. Elimination Half-Time The elimination half-time of meperidine is 3 to 5 hours (see Table 7.4). Because clearance of meperidine primarily depends on hepatic metabolism, it is possible that large doses of opioid would saturate enzyme systems and result in prolonged elimination half-times. Nevertheless, elimination half-time is not altered by doses of meperidine up to 5 mg/kg IV. About 60% of meperidine is bound to plasma proteins. Elderly patients manifest decreased plasma protein binding of meperidine, resulting in increased plasma concentrations of free drug and an apparent increased sensitivity to the opioid. The increased tolerance of alcoholics to meperidine and other opioids presumably reflects an increased Vd, resulting in lower plasma concentrations of meperidine for a given dose. Clinical Uses The clinical use of meperidine has declined greatly in recent years. Meperidine is the only opioid considered adequate for surgery when administered intrathecally.77 An IM injection of meperidine for postoperative analgesia results in peak plasma concentrations that vary three- to fivefold as well as a time required to achieve peak concentrations that varies three- to sevenfold among patients.78 The minimum analgesic plasma concentration of meperidine is highly variable among patients; however, in the same patient, differences in concentrations as small as 0.05 μg/mL can represent a margin between no relief and complete analgesia. A plasma meperidine concentration of 0.7 μg/mL would be expected to provide postoperative analgesia in about 95% of patients.79 Normeperidine toxicity has been described in patients receiving meperidine for PCA.76 Therefore, because there are other effective agents, PCA with meperidine cannot be recommended. Meperidine may be effective in suppressing postoperative shivering that may result in detrimental increases in metabolic oxygen consumption. The antishivering effects of meperidine may reflect stimulation of κ receptors (estimated to represent 10% of its activity) and a drug-induced decrease in the shivering threshold (not present with alfentanil, clonidine, propofol, or volatile anesthetics).80–82 In addition, meperidine is a potent agonist at α2-adrenergic receptors, which might contribute to antishivering effects.83 Indeed, clonidine is even more effective than meperidine in reducing postoperative shivering. Butorphanol (a κ-receptor agonist-antagonist) stops shivering more effectively than opioids with a predominant μ-opioid receptor agonist effect. Evidence for a role of κ receptors in the antishivering effects of meperidine and butorphanol is the failure of naloxone to completely inhibit this drug-induced effect. Unlike morphine, meperidine is not useful for the treatment of diarrhea and is not an effective cough suppressant. Meperidine is not used in high doses because of significant negative cardiac inotropic effects plus histamine release in a substantial number of patients.84 Side Effects The side effects of meperidine generally resemble those described for morphine. Meperidine, in contrast to morphine, rarely causes bradycardia but instead may increase heart rate, reflecting its modest atropine-like qualities. Large doses of meperidine result in decreases in myocardial contractility, which, among opioids, is unique for this drug. Delirium and seizures, when they occur, presumably reflect accumulation of normeperidine, which has stimulating effects on the CNS. Serotonin syndrome (autonomic instability with hypertension, tachycardia, diaphoresis, hyperthermia, behavioral changes including confusion and agitation, and neuromuscular changes manifesting as hyperreflexia) occurs when drugs capable of increasing serotonin administration are administered. In severe cases, coma, seizures, coagulopathy, and metabolic acidosis may develop. Administration of meperidine to patients receiving antidepressant drugs (monoamine oxidase inhibitors, fluoxetine) may elicit this syndrome.85 Meperidine readily impairs ventilation and may be even more of a ventilatory depressant than morphine. This opioid promptly crosses the placenta, and concentrations of meperidine in umbilical cord blood at birth may exceed maternal plasma concentrations.42 After equal analgesic doses, biliary tract spasm is less after meperidine injection than after morphine injection but greater than that caused by codeine.37 Meperidine does not cause miosis but rather tends to cause mydriasis, reflecting its modest atropine-like actions. A dry mouth and an increase in heart rate are further evidence of the atropine-like effects of meperidine. Transient neurologic symptoms have been described following the administration of intrathecal meperidine for surgical anesthesia.86 The pattern of withdrawal symptoms after abrupt discontinuation of meperidine differs from that of morphine in that there are few autonomic nervous system effects. In addition, symptoms of withdrawal develop more rapidly and are of a shorter duration compared with those of morphine. Fentanyl Fentanyl is a phenylpiperidine-derivative synthetic opioid agonist that is structurally related to meperidine (see Figure 7.1). As an analgesic, fentanyl is 75 to 125 times more potent than morphine. It was first synthesized by Janssen Pharmaceutica in 1960 during an assay of meperidine derivatives and subsequently released as the citrate salt under the trade name Sublimaze.87 Pharmacokinetics A single dose of fentanyl administered IV has a more rapid onset and shorter duration of action than morphine. Despite the clinical impression that fentanyl produces a rapid onset, there is a distinct time lag between the peak plasma fentanyl concentration and peak slowing on the EEG. This delay reflects the effect-site equilibration time between blood and the brain for fentanyl, which is 6.4 minutes. The greater potency and more rapid onset of action reflect the greater lipid solubility of fentanyl compared with that of morphine, which facilitates its passage across the blood–brain barrier. Consequently, plasma concentrations of fentanyl (unlike morphine) correlate well with CSF concentrations. Likewise, the short duration of action of a single dose of fentanyl reflects its rapid redistribution to inactive tissue sites such as fat and skeletal muscles, with an associated decrease in the plasma concentration of the drug (Figure 7.12).88 FIGURE 7.12 The short duration of action of a single intravenous dose of fentanyl reflects its rapid redistribution to inactive tissue sites such as fat and skeletal muscles, with associated decreases in the plasma concentration of drug (mean ± standard error). Reprinted with permission from Hug CC Jr, Murphy MR. Tissue redistribution of fentanyl and termination of its effects in rats. Anesthesiology. 1981;55(4):369-375. Copyright © 1981 American Society of Anesthesiologists, Inc. The lungs also serve as a large inactive storage site, with an estimated 75% of the initial fentanyl dose undergoing first-pass pulmonary uptake.60 This nonrespiratory function of the lungs limits the initial amount of drug that reaches the systemic circulation and may play an important role in determining the pharmacokinetic profile of fentanyl. When multiple IV doses of fentanyl are administered or when there is continuous infusion of the drug, progressive saturation of these inactive tissue sites occurs. As a result, the plasma concentration of fentanyl does not decrease rapidly, and the duration of analgesia, as well as depression of ventilation, may be prolonged. Cardiopulmonary bypass causes clinically insignificant effects on the pharmacokinetics of fentanyl despite associated hemodilution, hypothermia, nonphysiologic blood flow, and cardiopulmonary bypass– induced systemic inflammatory responses.89 Metabolism Fentanyl is extensively metabolized by N-demethylation, producing norfentanyl, hydroxyproprionyl-fentanyl, and hydroxyproprionylnorfentanyl. Norfentanyl is structurally similar to normeperidine and is the principal metabolite of fentanyl in humans. It is excreted by the kidneys and can be detected in the urine for 72 hours after a single IV dose of fentanyl. Less than 10% of fentanyl is excreted unchanged in the urine. The pharmacologic activity of fentanyl metabolites is believed to be minimal.90 Fentanyl is a substrate for hepatic P-450 enzymes (CYP3A) and is susceptible to drug interactions that reflect interference with enzyme activity (less likely than with alfentanil).91 Elimination Half-Time Despite the clinical impression that fentanyl has a short