Intrathecal Acetyl Cholinesterase Inhibitors Produce Analgesia That Is Synergistic with Morphine and Clonidine in Rats

Abstract

Intrathecal (IT) administration of acetyl cholinesterase inhibitors produces analgesia through a muscarinic action.The addition of IT cholinergic agonists to IT opioids or alpha_2-adrenergic agonists results in enhanced analgesic effects, but it is not clear whether these interactions are synergistic, additive, or less than additive. Dose-response curves for hot plate and tail immersion tests were established for IT neostigmine, physostigmine, and echothiophate in rats. Dose-response curves for hot plate testing were established for IT morphine and clonidine. The effect of maximally effective doses of each of the three cholinergic drugs on hot plate testing was plotted over time. Isobolographic analysis was performed for hot plate testing using neostigmine-morphine and neostigmine-clonidine combinations. The three cholinesterase inhibitors produced profound analgesia on hot plate testing but incomplete analgesia using the tail immersion test. Duration of analgesia on hot plate testing ranged from 45 min for physostigmine to more than 24 h for echothiophate. IT administration of combinations of neostigmine plus morphine and neostigmine plus clonidine both produced significantly more profound analgesia than the calculated additive effects and are, therefore, synergistic in their actions.

(Anesth Analg 1995;81:501-7)

Intrathecal (IT) administration of acetyl cholinesterase inhibitors produces antinociception in animals ^[1-3]. It appears that these drugs exert their analgesic effect through a muscarinic action, since muscarinic but not nicotinic cholinergics are analgesic when administered IT ^[3-5], and atropine antagonizes the analgesia produced by IT cholinesterase inhibitors ^[2].

Cholinergics may have potential clinical applications in patients who exhibit inadequate analgesia or intolerable side effects from spinal or epidural opioids or clonidine. Combining cholinergic agonists with either opioids or adrenergic agonists may produce profound analgesia while minimizing side effects. Williams et al. ^[6] showed that combinations of neostigmine and clonidine administered intrathecally in sheep provided more stable blood pressures than IT clonidine alone. Gordh et al. ^[7] demonstrated that IT neostigmine plus clonidine provided more profound analgesia in rats than clonidine alone. Likewise, Dirksen and Nijhuis ^[8] showed that IT neostigmine plus morphine provided more intense analgesia than morphine alone. However, these studies failed to determine whether these drug interactions were synergistic, additive, or subadditive. Naguib and Yaksh ^[2], using an isobolographic analysis, documented synergy between IT neostigmine and clonidine and between IT edrophonium and morphine, but not between IT neostigmine and morphine.

We sought to document the IT analgesic effect of three cholinesterase inhibitors, neostigmine, physostigmine, and echothiophate, an irreversible inhibitor, on the hot-plate and tail immersion tests, and to reexamine the analgesic interactions of IT combinations of neostigmine plus morphine and neostigmine plus clonidine.

Methods

The following investigations were performed according to a protocol approved by the Animal Care Committee of the Zablocki VA Medical Center (Milwaukee, WI).

Male Sprague-Dawley rats weighing 250-350 g were implanted with lumbar intrathecal catheters introduced via an incision in the atlanto-occipital membrane under halothane anesthesia as previously described by Yaksh and Rudy ^[9]. Catheters were advanced 11 cm caudally and externalized through the anterior portion of the scalp. Animals showing neurologic deficits on emergence from anesthesia were killed by barbiturate overdose. To ascertain correct placement of the catheters, 20 micro Liter of 2% lidocaine was injected, followed by 10 micro Liter 0.9% saline to flush the catheter on the day of the operation. Only animals which developed transient bilateral motor and sensory blockade in the hind legs were included in the study. IT injection studies were performed at least 5 days postoperatively.

All IT drug doses were dissolved in normal saline and administered via micrometer driven injection device in a volume of 10 micro Liter, followed by 10 micro Liter saline to flush the catheter. The following drugs were used in the study: morphine (morphine sulfate, molecular weight [MW] = 334; Mallinckrodt, St. Louis, MO), clonidine (clonidine hydrochloride, MW = 230.1; Sigma, St. Louis, MO), neostigmine (neostigmine methylsulfate, MW = 209.3; ICN, Costa Mesa, CA) echothiophate (echothiophate iodide, MW = 383.22; Wyeth-Ayerst, Philadelphia, PA), physostigmine (physostigmine salicylate, MW = 413.5; Sigma).

The general behavior of all of the rats was carefully observed and tested. The following tests of motor function and coordination were performed: observation of gait; placing/stepping reflex (the dorsum of either hindpaw was drawn across the edge of a table, which normally results in the animal lifting the paw and placing it on the Table surface; and righting reflex. The presence of allodynia was examined by looking for agitation (escape, aggression, or vocalization) evoked by lightly stroking the fur.

Five to seven animals were used for each dose of each drug. Animals were placed on a 52 degrees C hot plate surrounded by a Plexiglas cage. Latency to a behavioral response to pain (licking a hindpaw) was recorded. Cutoff time was 60 s. Animals not responding within the cutoff period were assigned a latency of 60 s. Baseline testing was done twice, at least 10 min apart, and the mean was recorded. Testing was repeated 10, 20, 30, 45, and 60 min after drug injection. The percent of maximum possible effect (%MPE) for each test of response latency was calculated as: Equation 1 For neostigmine and echothiophate, calculation of the area under the time-response curve from 10 to 60 min was calculated by a trapezoidal rule, and doseresponse curves were constructed from the area under the %MPE curves. For echothiophate, testing was repeated at 2, 3, 24, and 48 h after injection, although these values were not used in the calculation of MPE. Since the effect of physostigmine was receding by 45 min, only the 10, 20, and 30 min values were used in the calculation of the MPE.

A mark was placed on the tail 10 cm from the tip. With the animal wrapped in a towel, the tail was immersed to the 10-cm line in a thermostatically controlled 52 degrees C water bath. Latency to withdrawal of the tail was recorded. Cutoff time was 10 s. Animals not responding within the cutoff period were assigned a latency of 10 s. In all cases, the tail immersion test was performed immediately after hot plate testing at the intervals noted above. Area under the curve over the testing period was calculated in the same manner as for hot plate testing and was expressed as %MPE.

Dose-response curves for both hot plate and tail immersion tests were determined for each of the cholinesterase inhibitors. Dose-response curves for morphine and clonidine were determined only for the hot plate test. Five to six animals were used for each dose of each drug. The following doses were used: morphine 1, 3, and 10 micro gram; clonidine 3, 10, and 30 micro gram; neostigmine 1, 5, and 10 micro gram; echothiophate 0.1, 0.2, and 1 micro gram; and physostigmine 15, 30, and 60 micro gram.

Isobolographic analysis was performed using the hot plate data to determine the degree of interaction between neostigmine and morphine and neostigmine and clonidine. Dose-response curves were constructed from the %MPE data, and the ED₅₀ was calculated for each drug. Fractions of the ED₅₀ of each drug (1/2, 1/4, 1/8, 1/16 ED₅₀ for neostigmine/clonidine study and 1, 1/2, 1/4, 1/8 ED₅₀ for neostigmine/morphine study) were then administered concurrently, and the ED₅₀ of the total dose of the mixture was determined. The ED₅₀ doses of the individual drugs given in combination were calculated from the dose ratios (2:1 for morphine:neostigmine, 7.5:1 for clonidine:neostigmine), and these values were used for plotting the isobologram.

The isobolograms were constructed as follows, based on the technique described by Malmberg and Yaksh ^[10]: The ED₅₀ values were plotted on the x and y axes. The theoretical additive dose combination was calculated by the method described by Tallarida ^[11,12]. The experimentally derived values were determined as described above. The theoretical additive point lies on a line connected the ED (₅₀) values of the individual drugs. Experimental values that lie on or near this line are considered to have additive interactions. Values that lie below and to the left of this additive line are considered to be synergistic, whereas values that lie above and to the right of the line demonstrate a less than additive interaction.

Calculations of ED₅₀ values and their 95% confidence intervals were obtained from pharmacologic software programs of Tallarida and Murray ^[12]. Statistical analysis of drug interactions was conducted according to procedures of Tallarida and Murray ^[12]. The variances for the individual drugs in each mixture were calculated from the variance of the total dose. The difference between the theoretical additive point and the experimentally derived ED₅₀ was compared using the Student's t-test. For experimental values that were lower than theoretical additive values, a P value <0.05 for the differences in both the x and y directions was interpreted as a significant synergistic interaction.

Results

The dose-response curves for hot plate and tail immersion tests for the three cholinergic agents are shown in Figure 1. The ED₅₀ values for these drugs on the hot plate test (followed by the 95% confidence limits) are neostigmine 2.1 micro gram (0.9-5.0); echothiophate 0.3 micro gram (0.2-0.4); and physostigmine 36.6 micro gram (16.8-79.5). As can be seen from the dose-response curves for the tail immersion test, the effect at the highest dose of each drug was below 50% MPE. Therefore ED₅₀ values were not calculated. Larger doses were not used because of the presence of side effects at the largest doses given. At the largest doses administered, analgesia on the hot plate test was less than 50% at 30 min for physostigmine and 180 min for neostigmine. For echothiophate, analgesia was near maximal at 180 min, was 47% at 24 h, and returned to near baseline by 48 h Figure 2.

Figure 1: Dose-response curves for the intrathecal acetyl cholinesterase inhibitors for the hot plate (A) and tail immersion (B) tests. ECHO = echothiophate; NEO = neostigmine; PHYSO = physostigmine; %MPE = percent maximum possible effect.

Figure 2: The percent maximum possible effect (%MPE) of 1.0 micro gram intrathecal (IT) echothiophate (ECHO), 10 micro gram IT neostigmine (NEO), and 60 micro gram IT physostigmine (PHYSO) versus time for the hot plate test.

Dose-response curves for morphine and clonidine for the hot plate test are shown in Figure 3. The ED₅₀ values (followed by the 95% confidence limits) were morphine 4.0 micro gram (2.2-7.2) and clonidine 15.2 (10.1-22.9).

Figure 3: A, Dose-response curves for intrathecal (IT) morphine (MS) and the morphine-neostigmine (MS-NEO) combination. B, Dose-response curves for IT clonidine (CLON) and the clonidine-neostigmine (CLON-NEO) combination. %MPE = percent maximum possible effect.

The dose-response curves for the morphine-neostigmine and clonidine-neostigmine fixed-ratio combinations are shown in Figure 3. For both the morphine-neostigmine and clonidine-neostigmine combinations, the combined effect of the drugs was significantly greater then the calculated additive effect Figure 4. Experimental values that lie inside the line of additivity which connects the ED₅₀ values of the two drugs are considered to have greater-than-additive, or synergistic effects. For the neostigmine-morphine combination, the experimental ED₅₀ values (followed by 95% confidence intervals) were morphine 0.89 micro gram (0.79-0.99) and neostigmine 0.46 micro gram (0.44-0.48). The expected additive values for this combination are morphine 2.04 micro gram (1.14-2.94) and neostigmine 1.04 micro gram (0.84-1.24). For the neostigmine-clonidine combination, the experimental ED₅₀ values were clonidine 2.24 (1.23-3.25) and neostigmine 0.31 (0.29-0.33). The expected additive values for this combination are clonidine 7.6 micro gram (3.33-11.87) and neostigmine 1.04 micro gram (0.96-1.12). Although the combined effect is clearly greater for the neostigmine-clonidine interaction, the experimental doses are significantly less (P < 0.05, Student's t-test) than the calculated additive doses for both drug interactions Figure 4.

Figure 4: ED₅₀ isobologram for the analgesic interaction of intrathecal neostigmine and morphine (A) and neostigmine and clonidine (B) coadministered in a fixed-dose ratio. The ED₅₀ dose of neostigmine (+/- SEM) is shown on the x axes, while the ED₅₀ doses of morphine and clonidine (+/- SEM) are shown on the y axes. The line connecting the ED₅₀ points is the theoretical additive line, and the theoretical additive points for the drug combinations (+/- SEM) are shown on the additive lines. The experimental points for both the neostigmine-morphine and neostigmine-clonidine combinations decreased significantly (P < 0.05) below the lines of additivity, indicating a synergistic effect.

At the largest dose of each of the cholinesterase inhibitors, some animals showed irritability (vocalization when handled) and abnormal posturing (back arched), and two animals in the high-dose neostigmine group demonstrated increased salivation. Motor activity was normal. These side effects lasted less than 30 min in all animals. Although there was some vocalization of animals in the high-dose cholinesterase inhibitor groups with handling, there was no apparent allodynia, as the animals could be stroked lightly without reaction. None of the animals in the morphine, clonidine, morphine/neostigmine, or clonidine/neostigmine groups showed any abnormal behavior. No abnormalities of righting or placing/stepping reflex were noted in any animal tested.

Animals that received 10 or 30 micro gram clonidine all exhibited some sedation. All animals were arousable with stimulation, but tended to sleep when left alone. All animals that received 10 or 30 micro gram clonidine, and some animals that received 3 micro gram showed a substantial diuresis. They were not incontinent, but would urinate frequently, in some cases as often as every 2-3 min. Some diuresis but no sedation was seen in the animals that received 1/2 or 1/4 the ED₅₀ doses of neostigmine plus clonidine (these animals received 7.5 and 3.75 micro gram clonidine, respectively).

Discussion

Analgesic Properties of Cholinesterase Inhibitors

All three anticholinesterase drugs tested produced dose-related analgesia but, at the largest doses, were associated with irritability and abnormal posturing. It is likely that the analgesic effect of these drugs is related to agonist effects at muscarinic receptors. Evidence for this is provided by the observations that muscarinic cholinergic drugs are analgesic when injected intrathecally ^[1,2,4], and that atropine is capable of blocking the analgesic effect of IT cholinesterase inhibitors ^[2,7]. The irritability seen at the larger doses may be related to the spinal nicotinic effects of these drugs; as evidenced by the appearance of gnawing, vocalization, and hyperactivity after IT nicotinic agonist administration ^[5]. In addition, Naguib and Yaksh ^[2] observed irritability, vocalization, and truncal rigidity in animals given IT neostigmine after atropine pretreatment. Such behavior was absent among animals given the same dose of neostigmine alone. The use of selective muscarinic receptor agonists, therefore, may prove to be a better therapeutic option than cholinesterase inhibitors, as such irritability is not generally reported after IT doses that are analgesic ^{[1,2,4,13,14]}.

The irreversible cholinesterase inhibitor echothiophate is a very potent antinociceptive drug, as evidenced by hot plate testing, when administered IT. Profound analgesia persisted for several hours, was considerably attenuated 24 h after administration, and was essentially gone by 48 h. Since there was attenuation of the antinociceptive response during this relatively short interval, we postulate that tolerance may have developed to the increased acetylcholine levels associated with acetylcholinesterase inhibition. Svensson et al. ^[4] demonstrated the development of complete tolerance to the antinociceptive effects of IT carbachol over a 5-day period.

The relative lack of analgesic potency of IT cholinergic drugs on the tail immersion test compared to the hot plate test suggests that the intrinsic activity of the cholinesterase inhibitors, may be fairly low. If this is the case, it implies that a large fraction of receptor occupancy (FRO) is required to produce a given effect. Drugs with a high intrinsic activity (i.e., producing profound effects with a low FRO), such as sufentanil, are capable of producing antinociception for a high intensity noxious stimulus at doses only moderately larger than those required for analgesia for a low intensity noxious stimulus, whereas drugs with a lower intrinsic activity (requiring a high FRO to produce an effect), such as morphine, show a much greater discrepancy in antinociceptive effect for highversus low-intensity stimuli ^[15]. There is probably considerable difference in stimulus intensity between the hot plate and tail immersion tests used in this study, as the control response latency for tail immersion is considerably shorter. With the higher stimulus intensity of the tail immersion test, none of the cholinesterase inhibitors effectively suppress the response to the noxious stimulus. Since all of the cholinergic drugs tested act to increase the availability of acetylcholine, as opposed to acting directly on the receptor, it is not surprising that they demonstrate similar response characteristics for the two tests. Another characteristic of drugs with low intrinsic activity is a profound loss of effect with continuous or repetitive administration, i.e., rapid development of tolerance ^[16], which appears to be the case for the antinociceptive effect of cholinergic agonists ^[4].

Drug Interaction Studies

Our findings with respect to the interactions between IT neostigmine and clonidine are in agreement with those of Naguib and Yaksh ^[2], who also documented a synergistic interaction between these drugs using a noxious radiant heat stimulus. On the other hand, their study failed to show a synergistic effect between IT neostigmine and morphine, although edrophonium and morphine were shown to be synergistic ^[2].

The ability of spinally administered cholinergic drugs to produce greater than additive analgesic effects when combined with IT morphine or clonidine suggests that clinical use of such combinations may provide a wider therapeutic window, producing analgesia at substantially smaller doses of each drug. In the case of the clonidine-neostigmine combination, there is the added benefit of opposite effects of the two drugs on arterial blood pressure ^[6].

The combination of cholinesterase inhibitors with other analgesics in relatively small doses is particularly appealing because of the potential for cholinergic drugs to produce side effects. At the larger IT doses of all three cholinesterase inhibitors, animals exhibited irritability, abnormal posturing, and vocalization when handled. With combined doses of neostigmine plus morphine or clonidine that produced substantial analgesic effects, no such side effects were seen. In a recent study by Hood et al. ^[17], IT neostigmine produced troublesome side effects at doses that were antinociceptive to a noxious cold stimulus in volunteers. Nausea and vomiting and reversible lower extremity weakness were seen at doses between 50 and 750 micro gram IT, whereas antinociception increased in a dose-dependent fashion throughout this dose range. At the largest dose, increased heart rate and arterial blood pressure, respiratory stimulation, sedation, and anxiety were noted ^[17].

Synergy between neostigmine and both clonidine and morphine may be useful in limiting the side effects of alpha_2-adrenergic agonists and opioids. The ability to achieve analgesia at lower doses may prevent the development of respiratory depression, itching, and urinary retention after spinal opioid administration, and sedation, hypotension, and bradycardia from spinal clonidine. It is not clear whether nausea will continue to be a problem using combinations of spinal opioids and cholinesterase inhibitors.

Clinical Significance

The synergistic effect between neostigmine and morphine provides some encouragement to add neostigmine to IT morphine infusions in cancer patients who have become opiate tolerant. On the other hand, the relative inability of the cholinergic drugs to provide antinociception for the more intense stimulus (tail immersion) and the evidence for tolerance development to cholinergic agonists ^[4] suggests that these drugs may not be very helpful for pain in advanced terminal cancer, when intense nociception and tolerance development are common issues ^[18]. It must also be kept in mind that antinociception in animal and human studies does not necessarily imply analgesia in the clinical situation of established pain.

In the area of postoperative pain management, coadministration of spinal morphine plus IT cholinergic agonists may allow substantial reductions in opioid dose, and may reduce the likelihood of respiratory depression and other side effects of neuraxial opioids. However, the relatively shor duration of action of IT neostigmine in humans (3-4 h) ^[17] suggests that single-shot IT administration would not result in more than a few hours of added benefit. A long-acting drug such as echothiophate may be a better candidate for single-shot coadministration with morphine. It seems unlikely that epidural administration of cholinergic drugs will be feasable, as there would probably be systemic side effects associated with the higher doses required. The incidence of side effects associated with such combinations remains to be determined.

The potential hemodynamic benefit to combining IT neostigmine and clonidine has been demonstrated in animals ^[6]. It is not clear whether such a combination would be beneficial in humans. In human volunteer studies of IT neostigmine, increases in heart rate and arterial blood pressure were seen only at the highest (750 micro gram) doses ^[17], suggesting that, at clinically effective doses, there may not be enough hemodynamic effect of neostigmine to offset the bradycardic and arterial hypotensive effect of IT clonidine.

Conclusions

Cholinesterase inhibitors, injected intrathecally in rats, provide dose-dependent antinociception on hot plate testing, but minimal antinociception on the tail immersion test. This finding may be an indication that these drugs are relatively ineffective when noxious stimulus intensity is high. IT neostigmine provides synergistic antinociceptive effects when combined with either morphine or clonidine. Some side effects were noted at analgesic doses of the cholinesterase inhibitors (irritability, abnormal posturing) and clonidine (sedation, diuresis) but no side effects were noted with any of the drug combinations. It is possible, therefore, that, by limiting the dose of each drug used in these combinations, the side effects of these drugs may be minimized when administered IT in humans.

The authors thank Dr. James Fujimoto, Jody Rady, and Blythe Holmes for their assistance and advice.