Mu-opioid receptor blockade by naloxone administered for acute detoxification in patients addicted to opioids markedly increases catecholamine plasma concentrations, muscle sympathetic activity (MSA), and is associated with cardiovascular stimulation despite general anesthesia. The current authors tested the hypothesis that the alpha2-adrenoceptor agonist clonidine (1) attenuates increased MSA during mu-opioid receptor blockade for detoxification, and (2) prevents cardiovascular activation when given before detoxification.
Fourteen mono-opioid addicted patients received naloxone during propofol anesthesia. Clonidine (10 microg x kg(-1) administered over 5 min + 5 microg x kg(-1) x h(-1) intravenous) was infused either before (n = 6) or after (n = 6) naloxone administration. Two patients without immediate clonidine administration occurring after naloxone administration served as time controls. Muscle sympathetic activity (n = 8) in the peroneal nerve, catecholamine plasma concentrations (n = 14), arterial blood pressure, and heart rate were assessed in awake patients, during propofol anesthesia before and after mu-opioid receptor blockade, and after clonidine administration.
Mu-receptor blockade markedly increased MSA from a low activity (burst frequency: from 2 burst/min +/- 1 to 24 +/- 8, means +/- SD). Similarly, norepinephrine (41 pg/ml +/- 37 to 321 +/- 134) and epinephrine plasma concentration (13 pg/ml +/- 6 to 627 +/- 146) significantly increased, and were associated with, increased arterial blood pressure and heart rate. Clonidine immediately abolished both increased MSA (P < 0.001) and catecholamine plasma concentrations (P < 0.001). When clonidine was given before mu-opioid receptor blockade, catecholamine plasma concentrations and hemodynamic variables did not change.
Administration of the alpha2-adrenoceptor agonist clonidine decreases both increased MSA and catecholamine plasma concentrations observed after mu-opioid receptor blockade for detoxification. Furthermore, clonidine pretreatment prevents the increase in catecholamine plasma concentration that otherwise occurs during mu-opioid receptor blockade.
WITHDRAWAL symptoms in humans addicted to opioids markedly decrease when detoxification is performed by injection of naloxone during general anesthesia and coadministration of a variety of other drugs. 1–3However, it remains uncertain if withdrawal is attenuated by administration of large doses of μ receptor antagonists as indicated by animal experiments, 4maintenance of general anesthesia during μ-opioid receptor blockade, or by drugs reported to ease withdrawal during conventional detoxification procedures, e.g. , α2-adrenoceptor agonists. 5–7We previously demonstrated that μ-opioid receptor blockade by administration of a large dose of naloxone to patients addicted to opioids markedly increases efferent sympathetic neural outflow and catecholamine plasma concentrations, and evokes cardiovascular stimulation despite general anesthesia, apparently unmasking dampening effects of chronic opioid agonist administration on the sympathetic system. 8–10In the current study, we tested the hypothesis that the α2-adrenoceptor agonist clonidine abolishes the increase in muscle sympathetic activity (MSA) evoked by μ-opioid receptor blockade in patients addicted to opioids during detoxification under propofol anesthesia. Furthermore, we investigated the effects of clonidine treatment before opioid receptor blockade.
Materials and Methods
The study protocol was approved by the Ethics Committee of the Medical Faculty, University Essen, Germany, and is consistent with the Helsinki declarations. All patients were enrolled on a voluntary basis and gave written informed consent.
Fourteen young male patients (aged 30 yr ± 6, mean ± SD, range: 20–36 yr) were enrolled in the current study through a local methadone outpatient care unit to undergo rapid opioid detoxification during general anesthesia. All had a long history of mono-opioid addiction (6 yr ± 5, range: 2–18 yr) and received oral methadone substitution therapy (65 mg d−1± 38, range: 25–160 mg d−1, for 19 months ± 29, range: 1–69 months) to prevent heroin intake. Methadone therapy resulted in high urinary methadone concentrations (2911 μg ml−1± 1254). Besides methadone, the patients reported that they were otherwise drug-free. This was confirmed by weekly urine toxicology screens (sensitive for opioids, methadone, benzodiazepines, cocaine, amphetamine and metamphetamine, barbiturates, tetrahydrocannabinol, and tricyclic antidepressants). The last test took place the day before detoxification. The patients did not suffer from other overt diseases; however, three patients had serologic evidence of having been exposed to the hepatitis B or C virus. These three patients did not show current clinical or laboratory signs of abnormal liver function or active infection. The last dose of methadone was given 24 h before naloxone treatment. The patients were studied in the supine resting position the morning after an overnight fast.
After the patients were admitted to the intensive care unit, a peripheral venous cannula and radial arterial and central venous catheters were inserted under local anesthesia for pressure monitoring and fluid replacement. In eight patients a tungsten needle (University of Iowa, Medical Instrument Facility, Iowa City, IA) was successfully inserted into the common peroneal nerve behind the fibular head for microneurography of MSA with a reference electrode placed in the subcutaneous tissue a few centimeters away. In the other six patients, MSA could not be recorded.
After a resting period of 30 to 60 min general anesthesia was induced by 2 to 4 mg/kg propofol and 0.1 mg/kg of cisatracurium for muscle relaxation. The trachea was intubated and the patients were mechanically ventilated (FIO2: 0.21–0.3, positive end-expiratory pressure [PEEP]: 3 mbar). Normocarbia was established and repeatedly confirmed by arterial blood gas analysis. In addition, a gastric tube and urinary catheter were placed. Anesthesia was maintained by a continuous propofol infusion (169 μg kg−1min−1± 4) to abolish corneal and glabella reflexes as previously described. 8,9The infusion rate of Ringer's lactate was adjusted to keep central venous pressure near baseline values (7 mmHg ± 2). Potassium chloride was infused as required (8 mmol/h ± 4) to maintain a serum potassium concentration close to baseline.
Muscle Sympathetic Activity
Multiunit postganglionic efferent sympathetic activity to muscle was recorded in eight individuals by microneurography in the common peroneal nerve behind the fibular head as previously described. 10,11The nerve signal was amplified (× 50,000), filtered (bandpass, 500–2000 Hz), and fed through a discriminator for further noise reduction and audio-monitoring (662C-3 Nerve Traffic Analysis System, University of Iowa, Bioengineering, Iowa City, IA). A mean voltage (integrated) signal was obtained by passing the original signal through a resistance-capacitance circuit. Muscle sympathetic activity recording sites were accepted when burst amplitude was at least twice as high as baseline noise and reproducible responses were obtained to brief challenges (e.g. , apnea). Bursts of MSA were counted and expressed as MSA burst frequency [bursts/min] and MSA burst incidence [bursts/100 heart beats], the latter also accounting for changes in heart rate.
Catecholamine Plasma Concentrations
Norepinephrine and epinephrine plasma concentrations were determined using Beckmann System Gold HPLC (Beckmann, Munich, Germany) and electrochemical detection (Chromsystems #41,000, Munich, Germany) as described previously. 8–10The catecholamine detection kit (Chromsystems Cat.-No. 5000) included a probe preparation system, HPLC column, and all necessary chemicals and buffers. The lower detection limit was 10 pg ml−1for both epinephrine and norepinephrine with a coefficient of variation of 6.2% for norepinephrine and 6.8% for epinephrine, respectively.
Heart rate was determined from the electrocardiogram (lead II) of a 5-lead electrocardiography recording system including ST segment analysis. Arterial and central venous pressures were continuously measured by electromanometry relative to barometric pressure with transducers referenced to the mid-axillary line.
Data Recording and Processing
Analog variables (MSA, electrocardiograph, arterial and central venous pressures) were recorded on a thermoarray recorder and were stored on digital tape. Signals were simultaneously fed into a personal computer and digitized (sampling frequency: 200 Hz/channel). All analyses performed were computer-supported (off-line) using a dedicated program (Gunnar Wallin, M.D., and Tomas Karlsson, B.S., Department of Clinical Neurophysiology, Sahlgren University Hospital, Göteborg, Sweden).
The allocation of patients to study interventions depended on whether a stable MSA recording site could be established. In all fourteen patients, after achieving steady state conditions during anesthesia, naloxone administration was started for detoxification by injecting 0.4 mg intravenously. Four additional naloxone boluses of increasing dosage (0.8 mg, 1.6 mg, 3.2 mg, 6.4 mg) were injected at 15 min intervals. 8–10A total of 12.4 mg naloxone was administered over 60 min. Seventy-five minutes after the first naloxone administration naloxone was continuously infused in a dose of 0.8 mg h−1.
In six patients in whom recordings of MSA were successfully established, 10 μg kg−1clonidine was administered intravenously over 5 min + 5 μg kg−1h−1during steady state conditions 20 min after the last naloxone bolus was administered.
Muscle sympathetic activity (averages of 5 min periods) and cardiovascular variables were determined in the awake state, during propofol anesthesia both before and after 12.4 mg of naloxone was injected for μ-opioid receptor blockade, and during the first 15 min after clonidine was given. Simultaneously, arterial blood was collected for determination of norepinephrine and epinephrine plasma concentrations.
Two additional patients were treated in a similar manner, except that clonidine administration was omitted. These patients served as a time control with MSA recordings maintained for an additional 45 min after the last naloxone bolus administration.
In the six patients without successful recordings of MSA clonidine (10 μg kg−1administered intravenously over 5 min + 5 μg kg−1h−1) was given before naloxone administration after induction of anesthesia. Cardiovascular variables (averages of 5 min periods) were determined in the awake patient, during propofol anesthesia and during clonidine pretreatment, and during 90 min of naloxone administration for μ-opioid receptor blockade. At the same time, arterial blood was collected for determination of norepinephrine and epinephrine plasma concentrations.
All data are expressed as mean ± SD unless otherwise indicated. Differences in mean values of variables over time were determined by a one-way repeated measures analysis of variance (ANOVA) followed by Newman- Keuls post hoc test. Differences in mean values of catecholamine plasma concentrations and hemodynamic measurements over time and between groups (clonidine pretreatment: yes/no) were determined by two-way ANOVA followed by Newman-Keuls post hoc test. The following a priori null hypotheses were tested: There is no difference in means of variables when compared to observations (1) during anesthesia before μ-receptor blockade, (2) to μ-opioid receptor blockade by naloxone before administration of clonidine, and (3) between patients receiving clonidine pretreatment and those without pretreatment. A null hypothesis was rejected with an α-error of less than 0.05.
A representative recording demonstrates the effects of propofol anesthesia, naloxone, and clonidine on MSA and arterial blood pressure in a patient addicted to opioids (fig. 1).
μ-Opioid Receptor Blockade by Naloxone
Blockade of μ-opioid receptors by naloxone induced a 10-fold increase in MSA burst frequency (2 bursts/min ± 1 to 24 ± 8;P < 0.001) and MSA burst incidence (3 bursts/100 heart beats ± 1 to 30 ± 19;P < 0.001) despite general anesthesia (fig. 2). Similarly, norepinephrine plasma concentration increased sixfold (41 pg/ml ± 37 to 321 ± 134;P < 0.001), and epinephrine plasma concentration increased almost 40-fold (13 pg/ml ± 6 to 627 ± 146;P < 0.001;fig. 3). Significant increases in arterial pressure and heart rate were observed simultaneously (fig. 3).
Effects of the α2-Adrenoceptor Agonist Clonidine during μ-Opioid Receptor Blockade
Clonidine rapidly reversed increased MSA (P < 0.001), increased catecholamine plasma concentrations (P < 0.001, fig. 3), and heart rate (P = 0.01) to values close to those measured during propofol anesthesia before μ- receptor blockade (figs. 2 and 3). Systolic arterial pressure also decreased slightly, but significantly, after clonidine infusion (fig. 3). In contrast, in two additional patients where clonidine administration was omitted, MSA, catecholamine plasma concentrations, heart rate, and arterial pressure did not decrease during the 45 min period after the last naloxone dose.
Effects of the α2-Adrenoceptor Agonist Clonidine when Given before Naloxone
Administration of clonidine during propofol anesthesia decreased heart rate (73 beats/min ± 12 to 51 ± 11;P = 0.02) while arterial pressure was unchanged. Low catecholamine plasma concentrations were not altered (fig. 3). When μ-receptors were blocked after clonidine, catecholamine plasma concentrations did not change and were significantly (P < 0.001) lower when compared with corresponding values in patients without clonidine pretreatment. Although patients with clonidine pretreatment had significantly (P < 0.001) lower heart rates during detoxification, arterial pressure did not differ between groups during naloxone administration.
Muscle sympathetic activity and catecholamine plasma concentrations markedly increased during μ-opioid receptor blockade in patients undergoing opioid detoxification despite propofol anesthesia. The α2-adrenoceptor agonist clonidine abolished this increased sympathetic neural activity, decreasing MSA values to those observed before naloxone administration. Moreover, clonidine pretreatment before μ-receptor blockade abolished cardiovascular activation and the increase in catecholamine plasma concentrations evoked by μ-receptor blockade. Accordingly, clonidine prevents sympathetic neural activation and cardiovascular stimulation evoked by μ-receptor blockade and opioid detoxification.
Interpretation of Results
We demonstrated that μ-opioid receptor blockade by naloxone for acute detoxification of mono-opioid addicts evoked increased catecholamine plasma concentrations for at least 180 min and marked cardiovascular stimulation during both barbiturate and propofol general anesthesia. 8,9Furthermore, an increased sympathetic neural outflow to muscle during μ-receptor blockade suggests that increased sympathetic drive is at least partly responsible for increased norepinephrine and epinephrine plasma concentrations. 10This indicates that μ-opioid receptor blockade evokes increased neural sympathetic outflow, apparently unmasking tonic inhibitory effects of chronic μ-receptor agonist stimulation on cardiovascular regulation. 10The observed increases in norepinephrine and epinephrine plasma concentrations in response to μ-receptor blockade are consistent with our previous results. 8,9
Within a few minutes, intravenous clonidine reversed both increased efferent sympathetic nerve activity to muscle and increased catecholamine plasma concentrations evoked by μ-receptor blockade. In contrast, in two patients in whom clonidine was omitted we observed a sustained increase in MSA and catecholamine plasma concentrations for at least 105 min after initiation of opioid receptor blockade. Furthermore, pretreatment with clonidine completely abolished the increase in catecholamine plasma concentrations, heart rate, and arterial blood pressure otherwise evoked by μ-receptor blockade.
Intravenous administration of the α2-adrenoceptor agonist clonidine decreased renal sympathetic nerve activity and norepinephrine spillover in anesthetized rats by a central mechanism as well as muscle and skin sympathetic neural activity at supine rest in awake human volunteers. 12–14When clonidine was given epidurally, however, no additional effects on muscle and skin sympathetic activity, hand and foot laser doppler blood flow, and on arterial blood pressure and heart rate were observed when compared with intramuscular injection achieving similar clonidine plasma concentrations. 15During naloxone-precipitated withdrawal in morphine-dependent rats, intravenous clonidine reversed neuronal discharge and increased activity of adenylate cyclase and cAMP-dependent protein kinase in the locus coeruleus. 16,17Furthermore, injection of clonidine in the vicinity of the locus coeruleus attenuated both the local concentration of noradrenergic metabolites and behavioral signs associated with opioid withdrawal in rats. 18In contrast, systemic administration of α2-adrenoceptor agonists that cannot penetrate the blood brain barrier did not evoke significant effects in the same experiment. 18This suggests that clonidine's effects on efferent sympathetic nerve activity are mainly mediated by a supraspinal mechanism. Moreover, these central sympatholytic effects are augmented by inhibition of ganglionic transmission of α2-adrenoceptor agonists. 19Finally, an inhibition of norepinephrine release by stimulation of presynaptic α2-adrenoceptors contributes to the decrease in norepinephrine plasma concentrations after clonidine administration. 20
Taken together, these data indicate that the increase in MSA evoked by μ-opioid receptor blockade, abolished by clonidine, are mediated by effects on μ-opioid receptors in the brain rather than in the peripheral sympathetic nervous system, although the latter may contribute.
Control of sympathetic activation during withdrawal from opioids is of clinical interest because serious side-effects of opioid receptor antagonists have been reported when administered during detoxification. 21–26Although the mechanism for the observed arrhythmias, pulmonary edema, or sudden deaths have not been clarified sympathetic activation may play a role. Thus, control of sympathetic activation may increase the margin of safety for antagonist-supported opioid detoxification.
Finally, administration of lower dosages of clonidine reportedly decreases subjective withdrawal symptoms during opioid detoxification, irrespective of general anesthesia. 5–7Our data suggest a potential mechanism for the observed decrease in withdrawal symptoms because clonidine apparently abolishes sympathetic neural activation and decreases catecholamine plasma concentrations, supplying a rationale for treatment. Thus, clonidine treatment can be recommended for treatment and prevention of sympathetic activation during withdrawal from opioids.
Recordings of MSA are available from patients who received clonidine after opioid receptor blockade, but not from those treated with clonidine before naloxone. Although we attempted to record MSA in all patients, the procedure was discontinued to avoid injuring the peroneal nerve when an acceptable recording site was not obtained within 60 min of beginning the search. 27Thus, we do not know whether clonidine pretreatment, apart from preventing an increase in catecholamine plasma concentrations, also prevents the increase in MSA evoked by μ-opioid receptor blockade. However, administration of clonidine before naloxone allowed us to test the hypothesis (previously demonstrated by our group) that clonidine prevents sympathoadrenal and cardiovascular activation during opioid withdrawal. 8,9
To assess the full pharmacologic response, clonidine was administered over a period of 5 min in a large single dose (10 μg kg−1); twice as much as that reported for relief of subjective opioid withdrawal symptoms. 5,6This was not associated with untoward effects; therefore, we cannot exclude the possibility that similar effects might be achieved by clonidine when given in smaller dosages.
In summary, μ-opioid receptor blockade by naloxone in patients addicted to opioids evoked an increase in MSA and in catecholamine plasma concentrations that was abolished by intravenous clonidine. Furthermore, clonidine pretreatment prevented the increase in norepinephrine and epinephrine plasma concentrations evoked by μ-opioid receptor blockade. These findings may explain the beneficial effects of clonidine during withdrawal procedures.