Subarachnoid injection of lidocaine produces anesthesia of relatively short duration. Analgesics, and anesthetics such as opioids and nitrous oxide, enhance the effect of the sensory block and may prolong analgesia (1–3).
Ondansetron, a selective 5-hydroxytryptamine3 (5-HT3) antagonist, exhibits local anesthetic properties (4). The 5-HT3 antagonists interfere with peripheral serotonin effects on nociception (5). More than one 5-HT3 receptor subtypes have been cloned: the 5-HT3A, expressed in central and peripheral neurons, and the 5-HT3B, which are expressed in peripheral neurons only (6). The effects of 5-HT3 inhibition on pain sensitivity are complex. Both analgesic and hyperalgesic actions have been reported.
Ondansetron is a powerful antiemetic, administered in the perioperative period to prevent or treat nausea and vomiting. Any interactions with anesthetics, including the effect on regional anesthesia, would be of interest to the anesthesiologist.
We hypothesized that ondansetron may enhance the effect of sensory subarachnoid block. The aim of this study was to investigate the effect of ondansetron on sensory and motor block after subarachnoid anesthesia with hyperbaric lidocaine.
After obtaining approval from the local ethics committee, 54 male, unpremedicated patients, ASA physical status II–III, and scheduled for transurethral excision of tumors of the bladder were recruited for the study. All patients were visited the day before surgery when the purpose of the study and the procedure of assessing the level of subarachnoid block were explained to them and written informed consent obtained. Exclusion criteria were hearing impairment, treatment for any type of chronic pain, nervous system disorder, and intake of any type of analgesics, such as acetaminophen or nonsteroidal antiinflammatory drugs with or without opioids, α2 agonists, or calcium channel blockers, during the last month.
Patients were randomly assigned to the control or ondansetron group using sealed envelopes. The ondansetron group received 4 mg of ondansetron by mouth the evening before surgery and 4 mg IV 15 min before the subarachnoid injection. The control group received placebo tablets and an equal volume of 0.9% normal saline solution IV the evening before and in the operating room (OR), respectively. Tablets and solutions were prepared by the hospital pharmacy and administered by an anesthesiologist who did not participate in the study.
Sensory block was assessed 15, 20, 25, and 30 min after subarachnoid injection, as in our previous studies (7,8). All measurements were performed with the patient in the lithotomy position.
In the OR, noninvasive monitoring (electrocardiogram, heart rate, Spo2, and noninvasive arterial blood pressure) was used. Each patient received 500 mL of lactated Ringer’s solution over 15 min before the subarachnoid anesthetic. Oxygen 35% via a Ventimask was administered during anesthesia and surgery. Subarachnoid anesthesia was performed at the L3-4 interspace using a Whitacre 25-gauge needle. Injection of 100 mg of 5% lidocaine with 8% dextrose (Xylocaine®, Astra, Södertälje, Sweden; specific gravity, 1030–1035) was performed at a rate of 1 mL per 10 s with no barbotage. The anesthesiologist who administered the subarachnoid lidocaine and assessed the sensory and motor block was blinded to the treatment given. The patient was placed in the sitting position for the subarachnoid injection and in lithotomy position for the duration of the study.
Fifteen minutes after the intrathecal injection, the level of sensory block was assessed unilaterally (left side) using a pressure palpator (Pressure FEELER 650g Sedatelec, Chemin des Muriers F-695040, Irigny, France), as described in previous studies (7,8). This manufactured device exerts a predetermined pressure of 650 g. Four lines along the posterior, middle, and anterior axillary lines of the left abdominal and thoracic wall and a line 5-cm medial to the anterior axillary line were drawn. The palpator was moved along these lines in a caudad to cephalad direction. The point on each line at which the patient felt the palpator (from no feeling to some feeling) was marked. The four points were joined to form a line, and the level of sensory block in dermatomes 15 min after subarachnoid injection was obtained 1–3,7,8). The level of the sensory block was similarly reassessed 20, 25, and 30 min after subarachnoid injection.
Motor block was assessed 30, 60, and 90 min after the subarachnoid injection of lidocaine, using the modified Bromage scale, and scored as 0 = no motor block, 1 = being unable to move the hip, 2 = being unable to move the knee, and 3 = being unable to move the ankle. All patients were questioned 24 h after surgery regarding the presence of back pain associated with dysesthesia radiating to the buttocks, thighs, hips, or calves.
Initial sample size estimation showed that 26 subjects should be included in each group to ensure a power of 0.80 for detecting a clinically meaningful difference of one dermatome in the regression of the sensory level between groups. Alpha error was assumed to be 0.05, and sd of approximately 1.9 was estimated from initial pilot observations. Demographics between the groups were compared with nonpaired, two-tailed Student’s t-test. The effect of treatment and time on the level of the sensory block (expressed in dermatomes) and the degree of motor block (Bromage scale) were assessed with nonparametric, two-way Friedman tests. Changes in dermatome values, as well as in Bromage scores, over time within the control group and the ondansetron group were statistically analyzed using individual repeated-measures design Friedman tests. Subsequent intragroup comparisons, when appropriate, were performed pair-wise using paired Wilcoxon’s signed ranks tests. For intergroup comparisons, the levels of sensory block expressed in dermatomes and the degree of motor block (Bromage scale values) were compared between groups at each time point with Mann-Whitney tests. P ≤ 0.05 was considered significant. The SPSS® 11 software for Macintosh was used for statistical analysis (SPSS® Base 10.0 for Macintosh; SPSS Inc, Chicago, IL).
Fifty-four patients were enrolled in the study, 26 assigned to the ondansetron group and 28 assigned to the control group. In the control group, subarachnoid block in one patient was not complete, and anesthesia was supplemented with fentanyl and nitrous oxide. Sensory and motor block were not assessed in this patient. Another patient could feel the pressure palpator everywhere, and the level of sensory block could not be defined. However, the transurethral procedure was feasible under subarachnoid anesthesia. Thus, 24 patients were included in the analysis for sensory block, and 27 were included in the motor block analysis. In the ondansetron group, one patient received general anesthesia because of failure of the subarachnoid block (sensory and motor block not assessed), and four patients had incomplete sensory block because they could feel the pressure palpator everywhere, and no level of sensory block could be determined. As a result, 21 patients were included in the analysis of sensory block and 24 in the analysis of motor block. In one patient, follow-up to assess motor block was also lost. Entire failure or incomplete sensory block occurred in 5 of 26 patients (19%) in the ondansetron versus 2 of 28 patients (7%) in the control group (P = 0.181).
The mean ± sd values of age, body weight, and height were 69 ± 9 versus 69 ± 8 yr, 73 ± 11 versus 74 ± 12 kg, and 167 ± 9 versus 167 ± 8 cm in the control and ondansetron groups, respectively, and did not differ significantly between groups.
The levels of sensory block (median values) in each group 15, 20, 25, and 30 min after subarachnoid lidocaine are shown in Table 1. We found significant differences among the dermatome values of the level of sensory block in the two groups, with regard to the effect of the type of treatment and to the effect of time (χ2 = 164; df = 2; P < 0.001). The level of sensory block did not differ within the control group during the 30-min study period (χ2 = 3.74; df = 2; P = 0.154), but it differed significantly over time within the ondansetron group (χ2 = 6.07; df = 2; P = 0.048). Sensory block was significantly lower in the ondansetron versus the control group 30 min after subarachnoid anesthesia (P = 0.019) (Table 1).
We found significant intergroup differences in Bromage scale values with regard to the effect of treatment and to the effect of time (χ2 = 247; df = 2; P < 0.001). Intragroup comparisons of the Bromage scale values over time showed significant differences within the control group (χ2 = 157; df = 2; P < 0.001) and within the ondansetron group (χ2 = 141; df = 2; P < 0.001). The motor block expressed in Bromage scale values differed significantly between the 30- and 60-min, 30- and 90-min, and 60- and 90-min time points within each group (P ≤ 0.001). The motor block did not differ between groups during the study intervals (Table 2).
Six of the patients in the control group and four in the ondansetron group reported back pain alone not associated with dysesthesia radiated to the buttocks, thighs, hips, or calves.
Our results demonstrate that pretreatment with ondansetron, a 5-HT3 selective antagonist, enhances the regression of subarachnoid block produced by lidocaine, although the mechanism is unknown. However, ondansetron seems to interact with drugs having antinociceptive effects. In rats, the selective 5-HT3 receptor agonist 2-methyl-serotonin, administered intrathecally, exhibits antinociceptive activity measured by increased latency in tail response and in the hotplate test. These effects are antagonized by the potent and selective antagonists of the 5-HT3 receptors (5). Likewise, ondansetron given subcutaneously in rats reduces or abolishes the antinociceptive effect of acetylsalicylic acid in the hotplate test in a dose-dependent manner (9). The drug has been shown to block the nifedipine analgesia assessed by tail-flick latency in rats (10). However, these results were not consistent with a study in humans, where ondansetron did not inhibit the analgesic effect of alfentanil (11).
In the clinical setting, ondansetron seems to antagonize the effect of analgesics like tramadol. Pretreatment with 4 mg of ondansetron one minute before the induction of anesthesia reduced the analgesic effect of tramadol after surgery by 25% (12). Also, a continuous infusion of ondansetron 1 mg/h for 24 hours after surgery was associated with increased tramadol requirements by threefold at 4 hours and by twofold at 8 and 12 hours, respectively (13).
Analgesics like opioids, when given IV, enhance the level of sensory block produced by intrathecal local anesthetic (1–3,7,8). On the contrary, ondansetron, a drug frequently used in anesthetic practice to prevent or to treat nausea and vomiting associated with anesthesia or surgery, antagonizes the block produced by subarachnoid lidocaine.
The antinociceptive action in rats mediated via spinal 5-HT3 receptors has been associated with possible release of γ-aminobutyric acid (GABA). The inhibition of responses to noxious stimuli observed by the selective 5-HT3 agonist 2-methyl 5-HT is blocked by both 5-HT3 antagonists and GABA antagonists. Thus, 5-HT3 receptors may evoke release of GABA and indirectly inhibit nociceptive transmission (14).
The descending serotoninergic systems may inhibit nociception by altering the response of dorsal horn neurons to noxious stimuli. Stimulation of the periaqueductal gray increases the release of 5-HT in the dorsal horns of the spinal cord that, along with other neurotransmitters, may inhibit the nociception of dorsal horn neurons (15). Levels of 5-HT in the cerebrospinal fluid are also increased after subarachnoid injection of bupivacaine (16). This increase is not associated with a 5-HT increase in blood levels. Whether this increase contributes to the antinociception produced by subarachnoid anesthesia remains unknown. However, the specific 5-HT3 antagonist ondansetron antagonizes the subarachnoid block produced by lidocaine. Although the mechanism of the antianalgesic action of ondansetron is not known, the clinical implications are important for the following reasons: First, in patients receiving ondansetron, an otherwise successful subarachnoid block may prove to be insufficient or of short duration. Second, patients with malignancies, who experience intractable pain and receive ondansetron as antiemetic, may exhibit resistance to analgesic techniques.
The short duration of recording sensory block and the lack of data on the maximal dermatomal spread and on block duration are limitations of our methodology. We assessed sensory block at 15, 20, 25, and 30 minutes; all measurements were performed with the patient in the lithotomy position. Changing from the lithotomy to the supine position and changes in blood circulation might affect the regression of sensory block. We did not aim to define the peak sensory block but rather the course of the block in the ondansetron-treated group versus the control group. Another limitation is that we evaluated only 5% hyperbaric lidocaine. More studies are required to confirm or refute the antagonism between ondansetron and other local anesthetics and to assess the block for longer periods.
In conclusion, subarachnoid lidocaine anesthesia in patients receiving ondansetron will be associated with a more rapid regression of sensory block. This may affect patients treated with ondansetron and undergoing surgery under subarachnoid anesthesia because the block may be less effective.
1. Fassoulaki A, Sarantopoulos C, Chondreli S. Systemic fentanyl enhances the spread of spinal analgesia produced by lidocaine. Br J Anaesth 1991;67:437–9.
2. Sarantopoulos C, Fassoulaki A. Systemic opioids enhance the spread of sensory analgesia produced by intrathecal lidocaine. Anesth Analg 1994;79:94–7.
3. Fassoulaki A, Sarantopoulos C, Zotou M. Nitrous oxide enhances the level of sensory block produced by intrathecal lidocaine. Anesth Analg 1997;85:1108–11.
4. Ye JH, Mui WC, Ren J, Hunt EE. Ondansetron exhibits the properties of a local anesthetic. Anesth Analg 1997;85:1116–21.
5. Glaum SR, Proudfit HK, Anderson EG. Reversal of the antinociceptive effects of intrathecally administered serotonin in the rat by a selective 5-HT3 receptor antagonist. Neurosci Lett 1988;95:313–7.
6. Morales M, De Wang S. Differential composition of 5-hydroxytryptamine3 receptors synthesized in the rat CNS and peripheral nervous system. J Neurosci 2002;22:6732–41.
7. Fassoulaki A, Zotou M, Sarantopoulos C. Effect of nimodipine on regression of spinal analgesia. Br J Anaesth 1998;81:358–60.
8. Fassoulaki A, Sarantopoulos C, Zotou M, Karabinis G. Assessment of the level of sensory block after subarachnoid anesthesia using a pressure palpator. Anesth Analg 1999;88:398–401.
9. Sandrina M, Vitalea G, Pinib LA. Central antinociceptive activity of acetylsalicylic acid is modulated by brain serotonin receptor subtypes. Pharmacology 2002;65:193–7.
10. Hunt TE, Wu W, Zbuzej VK. Ondansetron blocks nifedipine-induced analgesia in rats. Anesth Analg 1996;82:498–500.
11. Petersen-Felix S, Arendt-Nielsen P, Bak P, et al. Ondansetron does not inhibit the analgesic effect of alfentanil. Br J Anaesth 1994;73:326–30.
12. De Witte JL, Schoenmaekers B, Sessler DL, Deloof T. The analgesic efficacy of tramadol is impaired by concurrent administration of ondansetron. Anesth Analg 2001;92:1319–21.
13. della Rocca M, Romano S, Romano R, et al. Ondansetron inhibits the analgesic effects of tramadol: a possible 5-HT3 spinal receptor involvement in acute pain in humans. Anesth Analg 2002;94:1553–7.
14. Alhaider AA, Lei SZ, Wilcox GL. Spinal 5-HT3 receptor mediated antinociception: possible release of GABA. J Neurosci 1991;11:1881–8.
15. Cui M, Feng Y, McAdoo DJ, Willis WD. Periaqueductal gray stimulation-induced inhibition of nociceptive dorsal horn neurons in rats is associated with the release of norepinephrine, serotonin, and aminoacids. J Pharmacol Exp Ther 1999;289:868–76.
16. Naesh O, Hindberg I, Christiansen C. Subarachnoid bupivacaine increases human cerebrospinal fluid concentration of serotonin. Reg Anesth 1996;21:446–50.