Clonidine Administered as an Axillary Block Does Not Affect Postoperative Pain When Given as the Sole Analgesic

An intrinsic analgesic effect results from clonidine after its central or peripheral administration. In animal studies, clonidine depressed impulse conduction in isolated nerve fibers with some preference for C fibers ^[1,2]. Used as the sole analgesic, clonidine produced analgesia after intrathecal ^[3], epidural ^[4], and intraarticular ^[5] administration. Complete and long-lasting analgesia was obtained after one intrathecal injection of clonidine in women recovering from caesarean section ^[3]. De Kock et al. ^[4] reported dose-dependent postoperative analgesia after epidural clonidine administration in patients undergoing abdominal surgery. Intraarticular clonidine produced analgesia unrelated to vascular uptake of the drug ^[5]. Clonidine prolongs the duration of analgesia after axillary brachial plexus block when mixed with local anesthetics ^[6-9]; however, the exact importance of its intrinsic analgesic properties in this circumstance is not known. Therefore, we conducted this double-blinded, prospective, randomized study to determine whether clonidine has an analgesic effect on its own when administered into the brachial plexus sheath.

Methods

After informed consent and institutional approval, 45 ASA physical status I and II patients scheduled for elective surgery of the arm and forearm were included in the study. Exclusion criteria were age <18 or >60 yr, chronic use of clonidine, peripheral neuropathy, and an incomplete axillary plexus block. No premedication was given, and no additional medication was administered intraoperatively. On arrival in the anesthesia induction room, a visual analog pain scale (VAS) (0 = no pain to 10 = maximal pain) was clearly explained to the patients. After insertion of an IV cannula, axillary brachial plexus block was performed with 40 mL of 1.5% lidocaine with the aid of a neurostimulator, and a catheter was inserted into the axillary neurovascular sheath. The surgical procedure was not started until a complete loss of cold sensation, assessed with an ether-soaked swab, was noted for each major nerve distribution (median, ulnar, radial, and musculocutaneous nerves). The usual monitoring was used intraoperatively. At the end of surgery (time 0 = baseline), patients were divided into three groups of 15 each in a randomized, double-blinded fashion using a computer-generated randomization list and received one of the following solutions through the axillary catheter: in Group Saline (control group), 15 mL of an isotonic saline solution; in Group Clonidine, 150 [micro sign]g of clonidine diluted in saline solution to a total volume of 15 mL; in Group Bupivacaine, 15 mL of 0.25% bupivacaine. Arterial blood pressure, heart rate, oxygen saturation (SpO₂), and sedation score, evaluated on a four-point scale (0 = awake, 1 = drowsy, 2 = sleeping and arousable by verbal command, 3 = sleeping and arousable only by tactile stimulation), were checked every 60 min from baseline until the end of the 6-h study period. Hypotension and bradycardia, defined as a 20% decrease in relation to baseline value, were recorded. Postoperative pain was assessed every 60 min until the end of the study period (6 h) by an anesthesiologist unaware of the drug treatment administered, using the VAS scale previously explained to the patient. Analgesic therapy during the study period was standardized. Ketorolac 30 mg IM was given on request and was followed by morphine 5 mg IM if pain remained unchanged after 30 min. Time from baseline to onset of pain at the surgical site, time to the first pain medication request, and total dose of pain medication were recorded. The correct placement of the axillary catheter was confirmed at the end of the study period by effective analgesia afforded by 15 mL of 1% lidocaine when the patients required it. The patients were reviewed by the surgeon 1 mo after the procedure and were asked about any adverse effect or complications.

Demographic data, baseline hemodynamic values, time to report of first pain, and time to first pain medication request were analyzed using one-way analysis of variance with Student-Newman-Keuls multiple comparison test. VAS scores at each time, sedation scores, and total amount of ketorolac and of morphine were subjected to a Kruskal-Wallis test, followed by a Mann-Whitney U-test. Qualitative data were treated by the chi squared test. In all tests, P < 0.05 was considered statistically significant. Data are presented as mean +/- SD and numbers as appropriate. All analyses were performed using SIMSTAT statistical analysis package version 3.5 for DOS (Provalis Research, Montreal, Quebec, Canada).

Results

Demographics, type and duration of surgery, and baseline hemodynamics were not significantly different among the three groups (Table 1). Group Bupivacaine patients experienced a significantly greater duration of time to report of first pain and to first pain medication request compared with Group Saline and Group Clonidine patients (Table 2). Ketorolac consumption was significantly lower in Group Bupivacaine patients compared with the two other groups (Table 2). There were no significant differences between Group Saline and Group Clonidine at any time (Table 2). Group Bupivacaine patients had significantly lower VAS scores than those in the other groups at 2, 3, 4, and 5 h. The number of patients who did not require additional analgesia was significantly lower in Group Bupivacaine compared with Group Saline and Group Clonidine (Table 2). SpO₂ exceeded 96% in all patients at all times. No statistically significant difference in sedation score was noted among the groups. Six patients in Group Clonidine had bradycardia, compared with one in Group Saline and none in Group Bupivacaine (P < 0.01). One patient in Group Saline, four in Group Clonidine, and one in Group Bupivacaine had hypotension. Two patients in Group Clonidine had orthostatic hypotension. No postoperative sequelae were detected during the 1-mo follow-up.

Discussion

We found that administration of clonidine 150 [micro sign]g into the brachial plexus sheath did not produce postoperative analgesia after hand or forearm surgery. We decided to study only inpatients because clonidine can cause delayed sedation, bradycardia, and hypotension. We used clonidine 150 [micro sign]g because most experience in brachial plexus blocks is with this dose ^[6,7,10]. A short-lasting local anesthetic was used to make more evident any analgesic effect from clonidine administered at the end of the surgery. The failure to observe an analgesic effect produced by the administration of clonidine led us to limit the number of patients to 15 for each group.

The addition of clonidine to local anesthetic solutions extends the duration of brachial plexus block ^[6-10]. Neither a systemic action due to vascular uptake ^[5,6] nor local vasoconstriction mediated by an interaction with the postsynaptic adrenergic receptors ^[7] seems to be a valid explanation. A direct local anesthetic action of clonidine on neural transmission is supported by data from in vitro experiments. Butterworth and Strichartz ^[2] showed that clonidine inhibits action potentials of A alpha and C fibers in isolated, desheathed sciatic nerves. C fibers seem more sensitive to inhibition by clonidine than A alpha. Gaumann et al. ^[1] demonstrated a direct depressant property of clonidine on the C-fiber action potential and an enhancing effect on lidocaine-induced inhibition of C-fiber action potentials. Inhibition of noradrenaline release, mediated by an interaction with alpha_2-adrenergic presynaptic receptors, could be an alternative explanation for the enhancing effect of the peripheral administration of clonidine ^[5].

Under the conditions of the present study, we were not able to demonstrate an analgesic action of clonidine. This result might be explained by the inability of clonidine 150 [micro sign]g, administered into the axillary brachial plexus sheath, to cause a sufficient concentration of the drug at the level of the nerve fibers. Despite its lipophilic characteristics, clonidine may not have adequately penetrated tissue or nerve sheaths. This hypothesis is in agreement with electrophysiological studies conducted in isolated nerve fibers. Gaumann et al. ^[1] demonstrated that clonidine and lidocaine produced a comparable degree of C-fiber action potential inhibition at a high concentration (500 [micro sign]M). Concentrations of clonidine <500 [micro sign]M did not inhibit C-fiber action potential. However, when a low concentration of clonidine (500 nM), which itself had no effect on the action potential, was added to a perfusion of lidocaine, a significant enhancing effect on C-fiber blockade was obtained. Therefore, the administration of clonidine 150 [micro sign]g may have caused a concentration at the level of the nerve fibers sufficient to produce the enhancing effect but not the intrinsic local anesthetic action.

A more peripheral (intraarticular) administration of clonidine 150 [micro sign]g has an analgesic effect ^[5]. The difference of efficacy with the present study may be explained by the fact that the progressive reduction of the myelin layers that occurs toward the periphery might make the drug more accessible to the nerve fibers than at the axilla. Moreover, if part of the intrinsic analgesic action of clonidine is caused by an interaction with alpha_{2-adrenoceptors}, these are more numerous at the peripheral ending of primary afferent terminals than on axons of peripheral nerves ^[11]. Furthermore, analogous with the hypothesis of Stein ^[12] about opioid receptors, inflammation or tissue damage, such as that caused by an arthroscopic procedure, may increase the activity of peripheral alpha_{2-adrenoceptors} and their number by new synthesis and migration from the dorsal root ganglion ^[13]. In addition, the breakdown of the perineurium by inflammation may increase the accessibility of receptors to the drug ^[14]. Larger doses of clonidine may have revealed an analgesic effect, but larger doses may also have resulted in an increased incidence of systemic side effects. In the present study, the administration of 150 [micro sign]g produced a significantly more frequent incidence of bradycardia. Orthostatic hypotension was observed in two patients in Group Clonidine. In agreement with the results of recent studies ^[8,9], we believe that a reduction in the dosage of clonidine administered with local anesthetics for brachial plexus block might be advisable, especially in outpatients.

In conclusion, the administration of clonidine 150 [micro sign]g into the axillary brachial plexus sheath at the end of hand or forearm surgery did not prolong the onset of postoperative pain.