α2-Adrenergic Agonists for Regional Anesthesia

(Eisenach) Professor and Chair for Anesthesia Research, Wake Forest University Medical Center, Winston-Salem, North Carolina.(De Kock) Staff Anesthesiologist, Service d'Anesthesiologie, Universite Catholique de Louvain, Brussels, Belgium.(Klimscha) Staff Anesthesiologist, University of Vienna, Vienna, Austria.From the Department of Anesthesia, Wake Forest University Medical Center, Winston-Salem, North Carolina, and Universite Catholique de Louvain, Brussels, Belgium. Submitted for publication January 26, 1996. Accepted for publication May 10, 1996. Supported in part by National Institutes of Health grants GM35523 and GM48085. Dr. Eisenach is a consultant to Fujisawa, USA, which is developing clonidine for epidural use in the United States, and Abbott Laboratories, which is developing dexmedetomidine for intravenous use in the United States.Address reprint requests to Dr. Eisenach: Department of Anesthesia, Wake Forest University Medical Center, Medical Center Boulevard, Winston-Salem, North Carolina 27157-1009. Address electronic mail to: eisenach@bgsm.edu.Key words: Alpha2-adrenergic agonists, clonidine. Anesthetic techniques; epidural; spinal. Pain.Alpha2-ADRENERGIC mechanisms of analgesia have been exploited for more than 100 yr. Cocaine, the first spinal anesthetic, produces analgesia primarily by its local anesthetic action, but also inhibits norepinephrine re-uptake, and spinal cocaine produces analgesia, in part, by enhancing noradrenergic stimulation of alpha2-adrenoceptors. [1]Near the turn of the century, epinephrine was shown to produce spinal analgesia in animals, [2]an effect now recognized to be secondary to alpha2-adrenoceptor stimulation. [3,4]Nearly 50 yr ago, spinal epinephrine alone was shown to produce clinically useful analgesia, [5]although it is most commonly combined with local anesthetics for this purpose.Veterinarians have used alpha2-adrenergic agonists (xylazine, detomidine, medetomidine) for many years for regional analgesia, but experience with these agents in humans dates back only slightly more than 10 yr. In 1984, Tamsen and Gordh, [7]after testing for neurotoxicity in animals, [6]injected a parenteral preparation of the alpha2-adrenergic agonist, clonidine, epidurally in two patients with chronic pain. Since then, a complete toxicologic assessment (effects on spinal cord blood flow, [8-10]behavior after lumbar and cervical intrathecal injection in sheep and monkeys, [11,12]and histopathology in sheep, rats, and dogs [11,13,14]) has suggested that clonidine is safe for intraspinal use, and the vast majority of publications examining injectable clonidine for analgesia come from Europe, using their commercially available preparation. In the United States, a preservative-free preparation of clonidine has been submitted to the Food and Drug Administration for approval under the orphan drug indication of epidural treatment for intractable cancer pain.The goal of this review is to provide a clinically useful synthesis of published experience with clonidine for regional anesthesia, focusing on efficacy when administered alone and in combination with other analgesics in specific patient populations and appropriate monitoring and treatment of side effects. A reasonably comprehensive review of publications through 1995 includes 2,116 patients treated with clonidine via epidural, intrathecal, or peripheral injection (Table 1). Neuraxial administration reports consist primarily of single or multiple boluses for < 24 h (1,437 patients, 77% of total), followed by brief (< 48 h) continuous infusion (270 patients, 15% of total). Prolonged epidural infusion for > 1 week in the treatment of chronic pain represents 148 patients, only 8% of the total.Alpha2-Adrenoceptors are located on primary afferent terminals (both at peripheral and spinal endings), on neurons in the superficial laminae of the spinal cord, and within several brainstem nuclei implicated in analgesia, [15]supporting the possibility of analgesic action at peripheral, spinal, and brainstem sites. Experimental work in animals supports analgesic actions of alpha2-adrenergic agonists at all three sites, although their relative importance is controversial. Notably lacking are alpha2-adrenoceptors on axons of peripheral nerves. Clonidine does produce a minor degree of nerve conduction blockade at high concentrations, however, with some preference for C-fibers. [16-18]This conduction blockade may underlie, in part, the enhancement of peripheral nerve block when this agent is added to local anesthetics (see below).Several lines of evidence support a spinal action of clonidine in producing analgesia in humans. First, in a study in volunteers, [19]a single lumbar epidural bolus injection of clonidine produced analgesia in the lower, but not upper, extremity against a noxious cold stimulus, as would be anticipated from a spinal action. When clonidine was infused for 4 h in the lumbar epidural space in a subsequent study in volunteers, [20]analgesia spread to the upper extremity, suggesting that more extensive dermatomal distribution of analgesia is possible with continuous infusion.Second, pharmacokinetic and dynamic analysis supports a spinal site of action of clonidine in humans. After epidural administration in volunteers and patients, clonidine is rapidly absorbed, with peak concentrations in arterial blood within 10 min and in venous blood within 30-45 min (Table 2). Elimination from blood is slow (Table 2) compared with the relatively brief duration of analgesia after epidural clonidine administration (3-5 h), arguing against an action by systemic absorption and redistribution to peripheral or central sites. As expected from these divergent time courses, the correlation between blood clonidine concentration and analgesia within individuals is relatively poor.In contrast to blood, there is a strong correlation between clonidine concentration in cerebrospinal fluid (CSF) and analgesia after epidural clonidine administration. Clonidine is rapidly and extensively absorbed into the spinal CSF compartment after epidural administration (Table 2), with concentrations peaking 30-60 min after injection. This coincides closely with attainment of near-maximal analgesia. In volunteers, [19]there is a close correlation between lumbar CSF clonidine concentration and analgesia to a noxious stimulus to the lower extremity, with a concentration producing a 95% maximal effect (EC95) of 130 ng/ml (Figure 1). This agrees with the observation in patients after surgery that rescue pain medication usage by patient-controlled analgesia (PCA) approaches zero when calculated CSF clonidine concentrations approach 130 ng/ml (Figure 1). [21]Similarly, the duration of effective analgesia from epidural bolus injection of clonidine, 100-900 micro gram, [22]is in keeping with clonidine's rapid elimination from CSF (Table 2). Duration of complete analgesia in those patients corresponds to the time for CSF clonidine concentrations to decline to 97 plus/minus 52 ng/ml (here and throughout the manuscript, variability is given as plus/minus SD). Finally, computer-controlled infusion of epidural clonidine to steady state CSF concentrations in volunteers [20]yields a concentration-response for analgesia similar to that obtained in volunteers after bolus administration or in patients after continuous infusion (Figure 1).Cerebrospinal fluid is clearly not the site of action of alpha2-adrenergic agonists for analgesia, and the drug can reach sites producing analgesia in the spinal cord and elsewhere. As with lipophilic opioids, it is possible to achieve analgesia from systemic, epidural, or intrathecal administration of clonidine. However, clonidine is more potent after neuraxial than systemic administration, indicating a spinal site of action and favoring neuraxial administration. This is typified in two types of experiments. In the first, one can compare analgesia from an equal dose of clonidine administered by each of these routes. Therefore, intrathecal injection of a small clonidine dose, 150 micro gram, after cesarean section or minor orthopedic surgery yields analgesia for 4-6 h, but injection of this same dose by intramuscular or epidural routes produces no more analgesia than a placebo (Figure 2). [23,24]Comparing epidural and intravenous administration of a larger clonidine dose, epidural administration produces better analgesia, accompanied by a 50% reduction in rescue morphine requirements. [25]These results are in accordance with a spinal site of action.In the second type of study, one can allow patients to titrate drug to similar degrees of pain relief via PCA to compare the relative potency of the drug by different routes of administration. Using this paradigm for opioids, for example, fentanyl is equipotent by intravenous or epidural administration, whereas hydromorphone is approximately twice as potent when given epidurally. [26,27]Using this method, Bernard and colleagues [28]recently demonstrated that clonidine is also approximately twice as potent given epidurally as intravenously.Clonidine produces analgesia by actions on alpha2-adrenoceptors, as shown by partial reversal in humans of epidural clonidine analgesia and sedation, by the alpha2-adrenergic antagonist, yohimbine, although clonidine's effects on blood pressure and heart rate were not reversed. [29]In animals, intraspinal alpha2-adrenergic agonists cause analgesia, in part, by spinal cholinergic activation (Figure 3), [30,31]and two observations suggest this may also occur in humans. First, epidural clonidine, either by bolus [19]or computer-controlled infusion, [20]increases acetylcholine concentrations in lumbar CSF (Figure 3). More precise experiments in animals have demonstrated that this increase is due to release of acetylcholine in the dorsal, but not the ventral, horn. [32]Second, epidural clonidine analgesia in volunteers is enhanced by intrathecal injection of the cholinesterase inhibitor, neostigmine (Figure 3). [33]Whereas this interaction is only additive in humans compared with its synergy observed in animals, it nonetheless supports a reliance on cholinergic mechanisms in spinal analgesia from clonidine.Clonidine enhances both sensory and motor blockade from epidural or peripheral nerve block injection of local anesthetics (see below). Three possible mechanisms for this interaction have been suggested. First, clonidine blocks conduction of C and A delta fibers [17]and increases potassium conductance [34]in isolated neurons in vitro and intensifies conduction block of local anesthetics. [16]Because systemic pharmacokinetics are not a factor in these in vitro experiments, these data support a direct effect of clonidine on neural transmission in high local concentrations, such as may occur after local injection. Second, clonidine may cause local vasoconstriction in the clinical setting, thereby reducing vascular removal of local anesthetic surrounding neural structures. Although clonidine and other alpha2-adrenergic agonists can vasoconstrict in high concentrations, there is little evidence for this mechanism with clinically used concentrations. For example, plasma lidocaine concentrations are similar whether or not clonidine is combined with lidocaine for epidural anesthesia. [35]In contrast, combining lidocaine with epidural epinephrine does reduce systemic absorption, as reflected in reduced lidocaine concentrations in blood. [35]Finally, it has become evident that analgesics, whether administered systemically or with local anesthetics, can enhance peripheral or spinal blockade. For example, intravenous or intrathecal fentanyl both enhance intrathecal lidocaine anesthesia, [36,37]and the same is observed with clonidine. [38,39]Alpha2-Adrenergic agonists also enhance analgesia from intraspinal opioids. In animals, this interaction occurs both pre- and postsynaptic to the primary afferent synapse in the spinal cord, and is clearly synergistic when both drugs are administered intrathecally. [12,40,41]In contrast, epidural clonidine and fentanyl interact in an additive or only mildly synergistic manner after bolus administration in humans. [42]Nonetheless, the dose of each component can be reduced by more than 60% when epidural clonidine and fentanyl are combined for postoperative analgesia. The type of interaction between clonidine and opioids after intrathecal administration has not been quantified.Mechanistic studies support a primary spinal site of action of alpha2-adrenergic agonists for analgesia and a multifactorial mechanism of action in enhancing peripheral or intraspinal blockade from local anesthetics. Pharmacokinetic studies support an EC95of 130 ng/ml clonidine in CSF for analgesia after intraspinal administration and help to clarify the dose responses observed in clinical studies (see below). Opioids and neostigmine enhance intraspinal alpha2-adrenergic agonist analgesia, and it is likely that a combination of all three classes of agents could result in dramatic reductions in the dose of each.Because systemic absorption of clonidine and other lipophilic alpha2-adrenergic agonists after spinal administration is rapid and extensive, their hemodynamic effects are due, in part, to actions in the brain and the periphery.Clonidine affects blood pressure in a complex fashion after neuraxial or systemic administration because of opposing actions at multiple sites (Figure 4). [43,44]In the nucleus tractus solitarius and locus coeruleus of the brainstem, activation of postsynaptic alpha2-adrenoceptors reduces sympathetic drive. In addition, clonidine is not a pure alpha2/alpha1adrenergic agonist; it also activates nonadrenergic imidazoline-preferring binding sites in the lateral reticular nucleus, thereby producing hypotension and an antiarrythmogenic action. [45,46]Blood pressure typically decreases more in hypertensive than in normotensive patients after systemic or epidural clonidine administration, [22]perhaps reflecting increased tonic sympathetic drive in some patients with chronic hypertension. In the periphery, activation of presynaptic alpha sub 2 -adrenoceptors at sympathetic terminals reduces their release of norepinephrine by the sympathetic nerve terminals, which could cause vasorelaxation and reduced chronotropic drive. These brainstem and peripheral effects of alpha2-adrenoceptor stimulation are counter-balanced by direct peripheral vasoconstriction from circulating concentrations of the alpha2/alpha1adrenergic agonist, clonidine. [47]As a result, the dose response for clonidine by neuraxial or systemic administration is U-shaped, with peripheral vasoconstriction from circulating drug concentrations at high doses opposing central sympatholysis (Figure 4).In addition to brainstem and peripheral sites of actions, neuraxial administration of clonidine directly inhibits sympathetic preganglionic neurons in the spinal cord. [48]As a result, the degree of clonidine-induced hypotension is related to the spinal level of injection. At low thoracic or lumbar levels of injection, epidural clonidine is not associated with an increased incidence of hemodynamic side effects when compared with intravenous injection. [25]In contrast, more profound hypotension occurs with thoracic epidural injection (Figure 4), [49,50]perhaps reflecting the rostrocaudal gradient of noradrenergic innervation of sympathetic preganglionic neurons. [51]Alternatively, direct inhibition of sympathetic preganglionic neurons in the upper thoracic dermatomes, which supply the heart, may also have a more profound impact on resting blood pressure than does the inhibition of sympathetic preganglionic neurons elsewhere.The action of alpha2-adrenergic agonists on myocardial performance is complex. Clonidine reduces heart rate partly by a presynaptically mediated inhibition of norepinephrine release at the neuroreceptor junction and partly by a vagomimetic effect. Although clonidine depresses atrioventricular nodal conduction, severe bradyarrhythmias are rare with chronic clonidine use. [52]By reducing afterload, cardiac output may increase after clonidine treatment in some patients, including those with heart failure, whereas by reducing heart rate, it may reduce cardiac output in other patients. [53]Clonidine may reduce myocardial oxygen demand and has been shown to reduce infarct size when administered to patients in the acute phase of myocardial infarction. [54]Hemodynamic effects of clonidine after neuraxial or systemic administration begin within 30 min, reach maximum within 1-2 h, and last approximately 6-8 h after a single injection. Delayed onset of hypotension has not been observed with use of clonidine for analgesia alone or in combination (see below).Combination of an alpha2-adrenergic agonist with neuraxially administered local anesthetic could increase the degree of sympatholysis and resulting hypotension. However, in clinical studies in which local anesthetic alone was compared with that anesthetic and clonidine infrequently report significant reduction in arterial blood pressure or heart rate in patients having received the combination therapy (see below). Clonidine has minor or no effects on responses to vasoconstrictors or atropine given to treat hypotension or bradycardia that may occur with neuraxial anesthesia. [55-58]Clonidine pretreatment delays the central nervous system and cardiovascular toxic manifestations of bupivacaine overdose in animals, without accentuating the subsequent hypotension. [59]Treatment with an alpha2-adrenoceptor agonist during bupivacaine overdose improves the ventricular electrophysiologic parameters in dogs. [60]This is not to imply that clonidine should be used as treatment for bupivacaine overdose, but rather to emphasize that, should such overdose occur, inclusion of clonidine is unlikely to exacerbate the problem.Spinal neostigmine counteracts the hypotension produced by clonidine, [61]likely due to a cholinergically mediated increase in preganglionic sympathetic nervous system activity. [62]Because neostigmine also enhances clonidine-induced analgesia, [33]this combination may be clinically useful.Sedation commonly accompanies the use of clonidine for regional anesthesia, consistent with the known sedative/anesthetic-sparing properties of alpha2-adrenergic agonists by actions in the locus coeruleus. [63]This brainstem nucleus is associated with a wide variety of physiologic regulatory processes, including regulation of sleep and wakefulness, and is inhibited by alpha2-adrenergic agonists via a G-protein mediated mechanism that involves inhibition of adenylate cyclase. [63]Sedation after epidural administration of clonidine likely reflects systemic absorption and vascular redistribution to higher centers. Although it is conceivable that cephalad migration of clonidine in CSF could result in delayed onset of sedation, such delayed onset sedation has not been observed, nor has delayed-onset hypotension, as described earlier. The more profound depression of electroencephalographic measure of cerebral activity during enflurane/N sub 2 O anesthesia in patients having received epidural compared with intravenous administration [64]could be construed as indicating more profound sedation. However, this more likely represents reduced noxious afferent input to the central sites from a regional spinal effect. Sedation from epidural clonidine represents an alpha2-adrenergic effect, as witnessed in its reversal by the relatively specific antagonist, yohimbine, in postoperative patients. [29]Clonidine produces dose-dependent sedation over the dose range 50-900 micro gram of rapid onset (< 20 min) regardless of route of administration. After a large epidural bolus dose (700 micro gram), sedation is intense for 4-6 h. In many cases, sedation is a desired property, and several studies have demonstrated the reduced need for other sedatives and anxiolytic medications when clonidine is administered intraoperatively. With continuous infusion, as much as 40 micro gram/h epidural clonidine produces no more sedation than epidural placebo plus PCA intravenous morphine for postoperative pain, [21,65]nor does 30 micro gram/h epidural clonidine produce more sedation than epidural placebo plus PCA epidural morphine for cancer pain. [66]Although there is some evidence that implicates a noradrenergic mechanism of opioid-induced respiratory depression, alpha sub 2 -adrenergic agonists alone do not induce profound respiratory depression, even after massive overdose, [67]nor do they potentiate respiratory depression from opioids. [68,69]A few studies indicate greater respiratory depression from epidural than from systemic clonidine, but lack of true control groups calls these results into question. [70,71]A study conducted on human volunteers failed to demonstrate any important effect of epidural clonidine on resting respiratory control. [19]However, when considering the respiratory effects of clonidine, it must be considered that drugs acting on the central nervous system to alleviate pain, relieve anxiety, and produce sedation are almost always accompanied by some reduction in alveolar ventilation. As such, it is conceivable that clonidine therapy, by causing pain relief, could unmask respiratory depression from other drugs administered concurrently. Occasional reports of intermittent upper airway obstruction during deep sedation with clonidine, accompanied by transient oxyhemoglobin desaturation, suggest that monitoring with pulse oximetry may be indicated for 30 min to 2 h after large bolus doses. Two human volunteer studies specifically addressed the question of the potentiation of the respiratory depressant effects of opiates by the alpha2-adrenoceptor agonists. Absolutely no potentiation of this effect could be demonstrated. [68,72]As will be noted later, oxyhemoglobin desaturation is less likely with an epidural clonidine-opioid combination than with opioid alone.Clonidine is a potent sympatholytic agent, as discussed earlier. In stress situations, it reduces, but does not suppress, the neurohormonal secretion (norepinephrine, epinephrine, adrenocorticotrophic hormone, cortisol) secondary to sympathoadrenal hyperactivation. [73,74]alpha2-Adrenergic agonists promote the release of growth hormone, but this effect is short lived. also the release of by a direct action on the of the of although this effect is minor and of clinical the experience with clonidine for regional anesthesia is with epidural administration. is a review of published reports of epidural clonidine treatment for chronic pain, and postoperative pain, and in and patients, followed by suggested in the use of epidural for the use of epidural clonidine in patients with chronic pain are of opioids because of over in patients with or when opioids cause side efficacy in of reduced potency of opioids due to of or in pain to opioids, and specific of or clonidine is indicated in the treatment of intractable pain, which is the for approval of clonidine in the United In the this patients with severe cancer pain to doses of or epidural opioids were to epidural 30 micro gram/h clonidine or a in a patients received rescue pain medication by and was as a reduction in either epidural morphine use or pain, with the other not was more in patients clonidine than placebo and pain were reduced in patients clonidine (Figure Clonidine was effective in patients with pain patients with clonidine with of the study, patients were to continuous epidural clonidine infusion in an patients received clonidine for an of their of pain in a patient with pain received epidural clonidine in the and of the study, by the between the study the dramatic response observed in many patients (Figure studies of patients with intractable pain support the results of this example, epidural clonidine bolus produced dose-dependent analgesia in patients with cancer 100-900 micro gram, and continuous infusion of micro gram/h for as as in analgesia during the of only 1 of the patients was epidural clonidine considered the vast majority of patients in the of epidural clonidine for cancer pain side there was no in the incidence of side effects between clonidine and blood pressure by but hypotension to of the drug in < of Although sedation was noted on of clonidine therapy, this rapidly and there were no in sedation between clonidine and placebo during the was less in patients clonidine plus rescue morphine than in those placebo plus rescue These observations of hypotension and transient sedation from epidural clonidine are in the are few studies of epidural clonidine for chronic pain, and all but one single boluses of small doses micro and colleagues a to compare 150 micro gram epidural clonidine to epidural morphine in patients with chronic low back pain or pain from Clonidine pain in of 20 patients, was as or better than epidural and produced side effects than These results were in their studies of patients with pain after spinal 20 patients with chronic pain, primarily of the lower however, be with because this failed to demonstrate efficacy from epidural clonidine in a of patients with low back pain analgesia from clonidine in a pain relief from thoracic epidural injection of small doses micro of clonidine in 10 patients with this reflects of this pain to clonidine or is an of the some patients with chronic pain relief from regional and the of pain is considered to consist of neural at three to norepinephrine at peripheral primary afferent terminals, of large fibers in pain at sites, and enhanced sympathetic from the spinal cord. agonists could reduce pain in such by actions at all three reduction in peripheral norepinephrine release by stimulation of alpha sub 2 inhibition of noxious neural transmission in the by both pre- and postsynaptic and direct inhibition of spinal preganglionic sympathetic experience that epidural clonidine is effective in patients with pain. clonidine, and micro gram, produced analgesia as by and in patients with the clinical of sympathetic in a sedation nor hypotension from epidural clonidine was more severe with compared with injection. patients in this study received continuous epidural clonidine infusion in an manner for plus/minus Clonidine usage plus/minus micro gram micro and pain during this phase was reduced plus/minus compared with their pain the study plus/minus has been suggested that chronic pain such as pain after are due to that occur in the spinal cord in response to peripheral neural and and that block of afferent input by neuraxial analgesia may block of such pain. This is by a small clinical study in which 24 patients to lower were to no epidural treatment or 24 h of epidural infusion of a combination of and clonidine, for h after incidence of pain 1 yr after surgery was reduced from in the control to 8% in the epidural treatment The relative of clonidine to this