Postoperative analgesia using a low-dose, oral-transdermal clonidine combination: Lack of clinical efficacy

ELSEVIER

Postoperative Analgesia using a Low-Dose, Oral-Transdermal Clonidine Combination: Lack of Clinical Efficacy Medge D. Owen, MD,* E.E. Fibuch, MD,? Robert McQuillan, MD,* William R. Millington,

of Anesthesiology, Saint Luke’s Hospital of Kansas City, and the Division Biology and Biochemistry, University of Missouri-Kansas City School of

Department of Molecular

Medicine,

*Assistant Professor of Anesthesiology tProfessor Sciences

of Anesthesiology

IAssociate Professor and Biochemistry

and Biologic

of Molecular

Biology

Address reprint requests to Dr. Owen at the Department of Anesthesiology, Bowman Gray School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1009, USA Supported by a research grant from the Foundation for Education and Research, Saint Luke’s Hospital, Kansas City, MO. Received for publication October 10, 1995; revised manuscript accepted for publication April 16, 1996.

PhD$

Kansas City, MO.

Study Objective: To determine af a lower than previously reported oral-transdermal clonidine regimen could reduce postoperative mmphine requirements without producing systemic side effects. Design: Doubkblind, randomized, placebo-controlled study. Setting: University-afjliated hospital. Patients: 29 healthy, ASA physical status I and II&males undergoing elective abdominal hysterectomy. Interventions: Patients received preoperative oral clonidine 4 to 5 p/kg and a 7 cm2 transdermal clonidine patch (0.2 mg/24 hours) or a placebo tablet and patch. Measurements and Main Results: Postoperative patient-controlled analgesia pumps provided morphine during the 48-hour study period. Morphine use, hemodynamic changes, and nonhemodynamic side effects were recorded. Additionally, visual analog pain scales (VXPS) and plasma concentrations of morphine and clonidine were measured. Wefound that low-dose clonidine had no potentiating effect on morphine analgesia. Postoperative morphine use, VXPS, and morphine plasma levels were similar between the control and clonidine-treated groups. Nevertheless, patients in the clonidine group experienced a significantly greater incidence of intraoperative and postoperative hypotension and bradycardia than did the control group. No diflerences were noted in the incidence of nonhemodynamic side effects. Conclusions: The lowdose oral-transdermal clonidine regimen evaluated failed to reduce postojxrative morphine requirements, although patients who received clonidine were still at risk for developing hypotension. 0 1997 by Elsevier Science Inc. Keywords: Alpha-adrenergic agonists; analgesia, clonidine plasma levels; transdermal patch.

postoperative;

clonidine;

Introduction Opioids are widely used for postoperative pain management, but they can cause pruritus, nausea, vomiting, urinary retention, tolerance, and potentially lifethreatening respiratory depression.‘22 These side effects have prompted a search for nonopioid analgesic alternatives with better side effect profiles. Clonidine, which is a centrally acting o2 agonist, represents one of these alternative antinociceptive drugs.

Journal of Clinical Anesthesia 9:8-14, 1997 0 1997 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

0952.8180/97/$17.00 PI1 SO952-8180(96)00218-8

Oral-transdmmal

Most investigations of clonidine’s antinociceptive effcacy have employed subarachnoid or epidural routes of administration.‘,s5 A few recent studies, however, have assessed using clonidine parenterally (intramuscular or intravenous) for analgesia during the perioperative period. Bonnet et aL6 found similar profiles between intramuscular (IM) and epidural clonidine injections; the onset and quality of analgesia and plasma clonidine levels were nearly identical between groups. Carroll et al.,’ using a single dose, double-blind, double-dummy, crossover comparison between intravenous (IV) and epidural clonidine found that IV clonidine produced slightly better analgesia than did the same dose given by epidural injection. These results show no advantage of the epidural route over parenteral routes of administration. Two studies have been conducted using a combination of preoperative oral and transdermal clonidine. Ellis et nl.’ focused primarily on the sympatholysis produced by this administration route (7 cm2 transdermal clonidine patch applied the evening before surgery and 0.3 mg oral clonidine given the morning of surgery), and not analgesia. In oral-transdermal clonicontrast, Segal et aLg administered dine in high-dose (10.5 cm* transdermal clonidine patch and 4.5 pg/kg oral clonidine the evening before surgery; 6.0 pg/kg oral clonidine the morning of surgery) and low-dose (7 cm2 transdermal clonidine patch and 3 pg/kg oral clonidine the evening before surgery; 3.0 pg/kg oral clonidine the morning of surgery) regimens and found that both groups of patients receiving clonidine required significantly less patient-controlled analgesia (PCA) morphine for postoperative analgesia compared with controls. The high-dose group achieved higher plasma clonidine levels, as one might expect, but this finding was not associated with greater therapeutic efficacy. In fact, the lowdose group self-administered less morphine than did patients in the high-dose group. Segal et aLg concluded that the low-dose regimen may be preferable because, while providing equally efficacious analgesia, it was less likely to produce hypotension in the postoperative period than was the high-dose regimen. The present study was designed to evaluate a lower oraltransdermal clonidine dose than was used by Segal et abg to determine if a lower boundary of effectiveness (ie, acceptable analgesia without systemic side effects) exists, and to determine if acceptable blood levels could be achieved by oral-transdermal dosing the morning of surgery rather than initiating the dosing the evening before surgery, as was done in the Segal et al.’ and Ellis et al.” studies, because increasing numbers of patients are being admitted for surgery the morning of, rather than the evening before, surgery.

Materials

and Methods

After Institutional Review Board of Saint Luke’s Hospital of Kansas City protocol approval, 29 consenting ASA physical status I and II females undergoing elective abdominal hysterectomy were randomly allocated, via computerized

clonidine,

,bostop. analgesia:

Owen et al.

random numbers table, in a double-blinded fashion to receive either oral-transdermal clonidine or oral-transdermal placebo 30 minutes before surgery. The oral clonidine dose was 0.3 or 0.4 mg (based on weight, not to followed by the application of a exceed 4 to 5 pg/kg),“,i’ 7 cm2 clonidine transdermal patch (delivering 0.2 mg/24 hr), consistent with the Segal et al.’ dosing schedule. Exclusion criteria included patient refusal, allergy to the study drugs, hypertension, and the preoperative use of alcohol, benzodiazepines, opioids, B-adrenergic antagonists, a-methyldopa, or clonidine. Preoperative medication was limited to the tablet and patch combination. Intraoperative monitoring included noninvasive blood pressure, electrocardiogram (ECG) , Raman gas analysis, esophageal temperature, and pulse oximetry. General anesthesia was induced with thiopental sodium 4 to 6 mg/kg, and tracheal intubation was facilitated with succinylcholine 1 to 2 mg/kg. Anesthesia was maintained with either isoflurane 0.5% to 1.5% or enflurane 1% to 2% combined with nitrous oxide (N,O) (50%). Surgical relaxation was achieved with vecuronium, and intraoperative opioid was restricted to morphine. After surgery, muscle relaxation was reversed with neostigmine, and patients’ tracheas were extubated awake in the operating room. In the recovery room, at the first request for pain medication, patients rated their pain using a standard 10 cm visual analog pain scale (VAPS) before receiving IV morphine. Patients were instructed to use a morphine PCA pump (Baxter Healthcare Corp., North Deerfield, IL) set to deliver a 1.5 mg morphine bolus, with an 8-minute lockout interval. Postoperatively, blood pressure (BP) and heart rate (HR) were recorded hourly via an automated noninvasive BP monitor, and continuous pulse oximetry was used for 48 hours, at which time the transdermal patch was removed. The incidence of nausea, vomiting, dry mouth, pruritus, hypertension, hypotension, tachycardia, bradycardia, and oxygen (0,) desaturation was recorded. Intramuscular ketorolac (30 mg every 6 hours for 24 hours) was given as needed for postoperative pain that was not relieved with morphine. Ketorolac was chosen as an adjuvant analgesic because of its moderate analgesic and antiinflammatory profile. In addition, ketorolac would not affect the plasma levels of morphine. VAPS measurements were repeated at 24 and 48 hours, along with venous blood sampling, to determine plasma morphine and clonidine concentrations. The VAPS measurements and the blood sampling were performed at the same time on the first and second postoperative days. Plasma morphine concentrations were analyzed by radioimmunoassay using a commercial kit (Diagnostics Products Corporation, Los Angeles, CA) with a sensitivity of 0.8 rig/ml. Clonidine was assayed by TSI Mason Laboratories (Worcester, MA), using a radioimmunoassay developed by Boehringer Ingelheim Pharmaceuticals, Inc. (Ridgefield, CT), with a sensitivity of 0.05 ng/ml.‘a Data are presented as means * SD or means * SE, as indicated in each table. Confidence intervals of 95% were

J. Clin.

Anesth.,

vol. 9, February

1997

9

Original

Contributions

determined between groups for PCA morphine use and plasma morphine concentrations. Statistical analysis included the two-sample t-test, chi square analysis, and Fisher’s exact test. A pvalue less than 0.05 was required for statistical significance.

Results Twenty-nine patients were enrolled in the study. One patient withdrew from the clonidine group because of a rash caused by morphine sulfate. The remaining patients were similar in age, height, weight, admission mean arterial pressure (MAP), HR, surgical duration, and intraoperative morphine usage (Table I). The intraoperative hemodynamic profiles were not statistically different (Tu6Z.e 2) when MAE’ or HR was compared between the clonidine and placebo groups. However, when the incidence of hypotension and bradycardia were combined, there was a statistically significant difference between the groups, suggesting that clonidine was having a suppressive hemodynamic effect. Nine patients in the placebo group and 8 patients in the clonidine group became briefly hypertensive and/or tachycardic during intubation (MAP or HR 20% above baseline) but this resolved with deepening anesthesia. One patient in the placebo group and five patients in the clonidine group were hypotensive intraoperatively (MAP 20% below baseline) but quickly responded to IV fluid and/or ephedrine. Four patients in the placebo group and 8 patients in the clonidine group became bradycardic (HR 20% below baseline) but this spontaneously resolved or was treated with atropine. All patients arrived in the recovery room extubated, awake, and responsive to verbal command. There was no statistically significant difference in the time for first analgesic request in the recovery room; however, patients in the clonidine group waited longer before seeking analgesia than did the placebo group (25 minutes uersus 18 minutes) . Patients completed a VAPS in the recovery room (Figure I) and on the first and second postoperative days. Cloni-

Table

1.

Demographics Clonidine (n = 14)

Age (~4 Height (cm) Weight (kg) Admit MAF’ (mmHg) Admit HR (beats/min) Surgical duration (min) Intraoperative morphine use bdkg)

42.9 e 6.2 166 + 6 71.8 t 20.0

89.0 + 6.9 79.2 * 9.6

Placebo (II = 14) 45.3 + 8.1 161+ 5 72.1 i 16.5 86.4 & 8.5

80.9 f 8.1

112.6 + 50.2

97.8 f 36.0

0.18 z!z0.1

0.21 -c 0.1

Note: There were no significant demographic differences bemeen the clonidine and placebo treatment groups using the rwo-sample t-test. Data are reported

as means ? SD.

MAP = mean arterial pressure; HR = heart rate.

10

J. Clin. Anesth., vol. 9, February 1997

dine treatment did not improve analgesia, as noted by the VAPS scores, which did not differ significantly between the clonidine and placebo-treated patients. Clonidine also failed to reduce the morphine dose required to control postoperative pain. Neither the amount of morphine administered by PCA nor the plasma morphine concentrations differed significantly between clonidine and placebotreated patients (Tubb 3). More patients in the placebo group required supplemental IM ketorolac for pain not responding to PCA morphine than did the clonidine group (36% vs. 14%). However, even when the ketorolac doses were converted into “morphine equivalents” (30 mg ketorolac = 10 mg morphine sulfate)13 and added to the actual amount of PCA morphine used, there was no statistically significant difference between the groups. Plasma clonidine levels were also measured postoperatively and compared with those reported by Ellis et aZ.s and Segal et al.’ (Table 4). Patients in the clonidine group had significantly lower BP levels postoperatively than the placebo group (Table 5). Nine (64%) clonidine patients versus three (21%) placebo patients experienced a 20% or greater decrease in MAP from preoperative baseline during the 48-hour postoperative period (1-,< 0.05). These episodes were transient, and no treatment was required other than increasing IV fluid administration. There were no statistically significant differences between the groups in the incidence of postoperative bradycardia (Table 5). However, when the incidence of bradycardia and hypotension were combined and compared between groups, there was a statistically significant difference noted (p < 0.05)) suggesting that clonidine was having a suppressive hemodynamic effect. All patients who reported feeling sleepy were easily aroused with verbal stimulation. Oxygen desaturation below 90% (breathing room air) was common in both groups and was treated with supplemental 0,.

Discussion Clonidine induces analgesia primarily by stimulating presynaptic and postsynaptic au,-adrenergic receptors in the dorsal horn of the spinal cord, 14*15which inhibit substance P release16 and decrease dorsal horn neuronal firing.17 Clonidine is considered to act at the spinal cord level, because following systemic administratio; in animals, the analgesia produced by clonidine’s is sustained even when the spinal cord is transected. Although ol,-adrenergic agonists produce analgesia, they are not without significant side effects. Clonidine exerts its hemodynamic effects by acting at several sites, both centrally and peripherally.‘Y Stimulation of brainstem c+-adrenergic receptors reduces central nervous system

*Comparative radioimmunoassay sensitivities: Owen et aZ., 0.05 rig/ml (coeffkient variation 10% to 13%) This paper. 0 Ellis etal.,” 0.05 rig/ml (coefficient variation 10% to 13%) 0 Segal et aL,9 0.10 rig/ml (coefficient variations < 15%). l

Oral-transdenaal

Table 2.

Intraoperative

postop. analgesia:

Owen et al.

Hemodynamics (% Change from Preoperative Baseline) Mean Arterial

Clonidine (n = 14) Placebo (n = 14)

clonidine,

Pressure

Heart

(MAP)

Rate (HR)

-20%

Baseline

>20%

20%

5 (36%) 1 (7%)

4 (29%) 7 (50%)

5 (36%) 6 (43%)

8 (57%) 4 (29%)

3 (21%) 7 (50%)

3 (21%) 3 (21%)

p = NS between clonidine and placebo groups for MAP or HR; Fisher’s exact test. P < 0.05 when HR and MAP are combined and compared between clonidine and placebo groups. The numbers in parentheses indicate the i ercentage of patients in each hemodynamic range group. sympathetic outflow, resulting in decreased peripheral vascular resistance, HR, and BP.” In the spinal cord, a,-adrenergic agonists inhibit firing of preganglionic sympathetic neurons, which results in decreased sympathetic o~tflow.~~ In contrast to these central neuraxial effects, the peripheral effects of a,-adrenergic agonists occur directly at postsynaptic receptors to produce vasoconstriction and hypertension, which can overcome the central sympatholytic effects. ” Clonidine can also produce significant bradycardia by slowing sinus and nodal conduction.** These hemodynamic changes are generally mild and well tolerated in healthy patients,‘,” and they can be reversed by treating the patient with IV fluid, vasopressors, and anticholinergic drugs.” Clonidine-induced analgesia has been demonstrated after intrathecal,+ epidural,5 IM,6 and IV” injection, but few studies have examined using an oral or oral-transdermal clonidine route of administration for postoperative analgesia. Preoperative oral clonidine 5 pg/kg decreases intraoperative inhalational and opioid requirements,‘O~‘l but these analgesic effects are not sustained in the postoperative period. 23 Although clonidine is highly lipid-soluble, with oral bioavailability ranging between 70% and go%,** most patients have oral intake restrictions for several days postoperatively, making oral clonidine impractical for providing prolonged analgesia. In addition, oral clonidine

may not reach peak plasma levels for 3 to 5 hours, and sustained plasma levels cannot be guaranteed.25 In theory, combined oral (preoperatively) and tran+ dermal (intraoperatively and postoperatively) clonidine may circumvent shortcomings associated with oral administration alone. The time-released transdermal patch delivers clonidine at a constant rate, but it requires at least 24 to 48 hours before therapeutic blood levels are achieved.26 This delay in achieving therapeutic blood levels can be overcome by co-administering a preoperative oral loading dose of clonidine. Other advantages include simplicity of application and low cost of treatment.8 clonidine is Segal et aLg suggested that oral-transdermal effective in reducing postoperative morphine sulfate requirements. However, in their study, plasma morphine concentrations were not measured, making it impossible to determine whether the reduction in use of PCA morphine postoperatively was due to the effect of clonidine. In that study, alfentanil was used intraoperatively instead of morphine and, unexpectedly, the authors found that the

Clonidine Treatment Does not Reduce Patient-Controlled Analgesia (PCA) Morphine Administration or Plasma Morphine Concentration

Table 3.

POD #I

PACU

Figure

POD #l

1. There were no significant

tive pain between reported POD #l day two.

•d

Cbnidine

Cl

Placebo

1

POD X2

differences in postoperascores using a 10 cm visual analog pain scale (VAPS) clonidine and placebo treatment groups. Data are as means + SD. PACU = postanesthesia care unit; = postoperative day one; POD #2 = postoperative

PCA Morphine Used (mg/day) Actual Clonidine 53.4 + 5.7 Placebo 49.9 f 6.2 95% CI of difference [-13.8, 20.81 Adjusted for ketoralac Clonidine 55.6 + 6.9 Placebo 57.7 + 8.3 95% CI of difference [-24.2, 20.01 Plasma Morphine Concentration (rig/ml) Clonidine 21.1 + 4.7 Placebo 24.8 + 3.6 95% CI of difference [-15.9, 8.51

POD #2

32.3 f 5.2 26.1 * 7.0 [-11.7,24.1] 34.4 f 5.3 27.6 + 6.9 [-11.2,24.8] 18.5 f 4.5 8.6 + 2.6 [-0.8, 20.61

Note: There were no significant differences in PCA morphine used by patients (actual and adjusted) or in plasma morphine concentrations when evaluated by the two-sample t-test. Adjusted ketorolac doses were derived by converting ketorolac into “morphine equivalents” (30 mg ketorolac = 10 mg morphine) and adding this to the actual amount of PC4 morphine used. Data are reported as means i SE. POD #I = postoperative day one; POD #2 = postoperative day two; CI = confidence intervals. J. Clin. Anesth., vol. 9, February 1997

11

Table 4. The Effect of Oral-Transdermal Clonidine Administration on Plasma Clonidine Concentrations: Comparison of Three Studies Plasma Clonidine @g/ml)

Segal et aZ.* High-dose Lowdose Ellis et al.? Owen et al.1 Dose

N

POD #l

POD #2

14 14 30 14

1.70 + 0.4 1.21 * 0.6 0.75 (0.625, 1.09)

1.83 f 0.4 1.32 zt 0.5 not drawn

0.69 f 0.2

0.95 f 0.5

*Segal et al. Anesthesiology 74:220-5,1991-Hiihdose

group: 10.5 cm*

transdermal clonidine patch and 4.5 pg/kg oral clonidine evening before surgery; 6.0 pg/kg oral clonidine morning of surgery. Lowdose group: 7 cm* transdermal clonidine patch and 3 pg/kg oral clonidine evening before surgery; 3 pg/kg oral clonidine morning of surgery. tEllis et al. An&h Andg 79:1133-40, 1994-7 cm* transdermal clonidine patch evening before surgery and 0.3 mg oral clonidine morning of surgery. Reported as mean (25%, 75%) interquartile range. IOwen et al. 7 cm* transdermal clonidine patch and 4 to 5 pg/kg oral clonidine morning of surgery (this study). Data are reported as means f SD (except Ellis et al.). POD #l postoperative day one; POD #2 = postoperative day two.

plasma

=

alfentanil

concenhations in the clonidine treattwice that of the placebo group, may inhibit alfentanil’s metaboOn the second postoperative day, the plasma cloni-

ment groups were nearly suggesting that clonidine lism.

dine levels of Segal et aLg were 1.93 times higher in their “high-dose” group and 1.39 times higher in their “lowdose” group than the plasma levels of clonidine measured in this study. In addition, despite what appeared to be an antinociceptive effect from the low-dose clonidine regi-

Table 5.

Postoperative

Hypotension Bradycardia Sedation Dry mouth Nausea/vomiting Visual change Pruritus Arterial oxygen

saturation

Side Effects