Amitriptyline for Prolonged Cutaneous Analgesia in the Rat

Both amitriptyline and bupivacaine hydrochloride salts were purchased from Sigma Chemical Co. (St. Louis, MO). Epinephrine was obtained from American Regent Laboratories (Shirley, NY) as a stock solution (1:1000). All drugs were freshly prepared and diluted in a solution of 0.9% NaCl, pH adjusted to 6.5–6.7, within 30 min of administering the injection.

Experiments were performed on conscious, unanesthetized, male Sprague-Dawley rats (weight, 250–300 g). The experimental protocol was approved by the Harvard Medical Area Committee on Animals. All rats were housed on a 12 h light–dark cycle with unlimited access to food and water.

Before the experiments were conducted, the animals were handled daily for up to 14 days to familiarize them with the behavioral investigator, the experimental environment, and the specific experimental procedures. This familiarization minimizes contamination from stress during the experiment and improves experimental performance. Criteria for sufficient handling have been described previously 11and included an absence of behavioral signs of stress (e.g. , frequent defecation, immobilization, and lack of exploratory behavior in an open environment) and an extinction of the initially present dorsal contractile response to the nonnoxious stroking of the area to be tested (accommodated by the end of the handling period), followed by a robust, distinctive response to noxious stimulation.

The cutaneous trunci muscle reflex (CTMR), which is characterized by reflex movement of the skin over the back produced by twitches of the lateral thoracispinal muscles in response to local dorsal cutaneous stimulation, was studied as a reaction to noxious pinprick. 26Because reactions to nonnoxious stroking were intentionally extinguished by repeated handling, it is presumed that responses to stimuli after LA injection were caused by noxious stimulation. The model of infiltration anesthesia used evaluated the inhibition of CTMR produced by subcutaneous injection of the given concentration of drug in 0.6 ml of solution.

After observing the animals normal reaction to six pinpricks applied outside the wheal raised by the drug injections and on the contralateral control side, six pinpricks (at a frequency of 0.5–1 Hz) were applied inside the wheal, and the number to which the rat failed to react was recorded. A Von Frey filament (20.9 g) to which the cut end of an 18-gauge needle was affixed, was used to standardize the stimulus intensity. Six pinpricks per test were sufficient to obtain reproducible results among the different rats within study groups but were few enough to avoid injury (redness, swelling) of the skin during repeated testing of the skin patches studied. The LA effectiveness of the drugs was evaluated quantitatively as the number of times the pinprick failed to elicit a response, with, for example, the complete absence of six responses defined as complete nociceptive block (i.e. , 100% of maximum possible effect [MPE]), the absence of three responses of six scored as 50% MPE, and a response identical to the control responses as 0% MPE. The test of six pinpricks was applied every 5–10 min for the first 30 min and then every 15 min to 2 h thereafter until the CTMR fully recovered from the block. The observer was blinded to the drugs and concentrations used for injections.

The drugs tested were injected (using a 30-gauge needle) subcutaneously in unanesthetized rats under the dorsal surface of the thoracolumbar region, from which hair had been mechanically removed 24 h before the experiments were conducted. The injections caused a circular elevation of the skin, a wheal, approximately 2 cm in diameter. This wheal was marked with ink within a minute after the injection. Each drug (with or without epinephrine) or combination of drugs tested was injected into a naïve area of the rat's shaved back. The back was divided into four quadrants (to clearly demarcate injection and control sites), and each animal was injected twice with the drug being tested, once without epinephrine and once with epinephrine, separated by an interval not less than 5 days. Animals used for the drug combination experiments were given a single injection of a particular drug combination. Because previous studies had shown very dilute solutions of epinephrine and saline to not have analgesic effects when injected alone, we did not include a saline control in our study. 26 

Rats in the respective groups (n = 8 for each group) were injected with a volume of 0.6 ml of the following drugs (weight/volume [%]): (1) 0.05% racemic bupivacaine and 0.05% racemic bupivacaine plus epinephrine (1:200,000); (2) 0.25% racemic bupivacaine and 0.25% racemic bupivacaine plus epinephrine (1:200,000); (3) 0.5% racemic bupivacaine and 0.5% racemic bupivacaine plus epinephrine (1:200,000); (4) 0.05% amitriptyline and 0.05% amitriptyline plus epinephrine (1:200,000); (5) 0.25% amitriptyline and 0.25% amitriptyline plus epinephrine (1:200,000); and (6) 0.5% amitriptyline and 0.5% amitriptyline plus epinephrine (1:200,000).

Each rat in the drug combination groups (n = 6 for each group) was injected with a volume of 0.6 ml of the following drugs via  a single injection (weight/volume [%]): (1) combined 0.05% racemic bupivacaine, 0.05% amitriptyline, and epinephrine (1:200,000); (2) combined 0.125% racemic bupivacaine, 0.125% amitriptyline, and epinephrine (1:200,000); and (3) combined 0.25% racemic bupivacaine, 0.25% amitriptyline, and epinephrine (1:200,000).

Drug combinations were compared in two distinct ways. First, the concentration of the individual drugs in the coinjected combination (durations, area under the curve [AUC], etc. ) were compared with the sum of the same concentration of the individual drugs injected separately with epinephrine. For example, animals injected with combined 0.05% amitriptyline and bupivacaine with epinephrine, and combined 0.25% amitriptyline and bupivacaine with epinephrine, were compared with the sum of the durations, or the sum of the AUCs of 0.05% amitriptyline with epinephrine and 0.05% bupivacaine with epinephrine injected separately, and with the sum of 0.25% amitriptyline with epinephrine and 0.25% bupivacaine with epinephrine injected separately, respectively. Second, we compared the sum of the individual drug concentrations (or total amount of drug) in the coinjected combinations with the equivalent total amount of individual drugs injected separately with epinephrine. For example, the 0.125% amitriptyline and bupivacaine with epinephrine combination was compared with 0.25% amitriptyline injected separately with epinephrine or with 0.25% bupivacaine injected separately with epinephrine. Similarly, the 0.25% amitriptyline and bupivacaine with epinephrine combination was compared with 0.5% amitriptyline injected separately with epinephrine, or with 0.5% bupivacaine injected separately with epinephrine.

A two-tailed Student t  test or chi-square test was used as appropriate for establishing significant differences between the values for the different drugs or combinations. Microcal Origin (Microcal Software, Northampton, MA) was used for statistical analysis, calculating areas under the curve, and creating figures. Data were reported as mean ± SE. Statistical significance was defined as P < 0.05.

Amitriptyline and bupivacaine elicited different durations and block densities (i.e. , %MPE) of reversible cutaneous analgesia depending on concentration and whether they were administered with or without epinephrine. The comparison of these two drugs or their combination considered their effect on (1) the duration of complete nociceptive blockade, (2) the fraction of animals within a group that had complete nociceptive block at a specified time after injection, (3) the duration until complete recovery of nociception, and (4) the AUC for the %MPE versus  time plots.

Amitriptyline elicited incomplete nociceptive blockade (< 100% MPE) in varying fractions of rats at specified times in each group tested. The density of block, averaged over all animals tested and expressed in terms of the maximum graded value, was less than 100% MPE in the amitriptyline groups because not all animals reached a complete nociceptive block for a sustained period (figs. 1A–C), despite there being some time points (especially at higher concentrations) at which block was complete in all animals. In contrast, bupivacaine elicited complete nociceptive blockade (table 1and figs. 1A–C). Epinephrine significantly (P < 0.05, t  test) increased the duration of complete nociceptive block at all three concentrations of bupivacaine tested (table 1). When amitriptyline, bupivacaine, and epinephrine were injected in combination (fig. 1D), the duration of complete block was significantly (P < 0.01) longer compared with that of bupivacaine with epinephrine alone and reached 24 h in the group that received a combination of 0.25% amitriptyline, 0.25% bupivacaine, and epinephrine (1:200,000) (table 2), whereas the duration was only 11.88 h in the group that received 0.5% bupivacaine plus epinephrine.

This measure was scored by recording the number of animals without any response to pinpricks between 20 and 40 min after injection (table 1). The addition of epinephrine significantly increased (P < 0.05, chi-square test) the fraction of animals achieving complete block in the amitriptyline groups from 0 to 50% at the 0.05% concentration and from 75 to 100% at the 0.25% concentration, respectively. The block achieved by the higher concentrations of amitriptyline (0.25 and 0.5%) was less reliable than that achieved with the corresponding bupivacaine concentrations; although all individual animals displayed a 100% MPE block at a given point in time between 20 and 40 min after injection (table 1), this effect was not sustained because the very next time point tested may have revealed, depending on the drug concentration, one or two responses to the six pinpricks applied. These “breakthrough” responses in the amitriptyline groups were characteristically in response to the first few pinpricks and were randomly scattered among all the animals tested but markedly decreased by the addition of epinephrine.

Amitriptyline produced a significantly longer duration of analgesia at all concentrations compared with bupivacaine (figs. 1A–C). Although the difference in duration until full recovery was highest at the 0.05% concentrations of amitriptyline and bupivacaine (at which the duration was several times longer for amitriptyline) and decreased progressively as the concentrations of both drugs increased, it still remained statistically significant (t  test, table 1). The ratio between the duration of complete block and full recovery was relatively small for the bupivacaine groups, as can be seen in figures 1A–C, whereas in the amitriptyline groups, this ratio was large because of amitriptyline's longer duration of less than 100% MPE block. Epinephrine significantly increased (P < 0.001, t  test) the duration until complete recovery in the 0.05% amitriptyline and bupivacaine groups and in the 0.5% amitriptyline group, but not in any of the other groups.

When animals were injected with the combination of 0.05% amitriptyline, 0.05% bupivacaine, and epinephrine, the time until full recovery (19.7 h;table 2) significantly outlasted the sum of the time until full recovery of each drug with epinephrine injected alone at the same concentration (17.3 h;P < 0.05). Similarly, but statistically not significantly, the combination of 0.25% amitriptyline, 0.25% bupivacaine, and epinephrine had a duration until complete recovery of 33.3 h, which outlasted the sum of the durations until complete recovery of 0.25% amitriptyline with epinephrine and 0.25% bupivacaine with epinephrine injected alone (31.0 h;P > 0.05, t  test). Note that here the comparisons are between the amount of each drug in the coinjected combinations and the sum of the durations of the same amount of each drug injected alone with epinephrine.

The time until full recovery in the groups that received 0.125% and 0.25% amitriptyline and bupivacaine combinations with epinephrine were significantly longer compared with either of the two drugs alone with epinephrine at the 0.25% and 0.5% concentrations, respectively (P < 0.001, t  test;fig. 2). Note that here, the 0.125% amitriptyline, bupivacaine, and epinephrine combination (i.e. , 0.25% total drugs) is being compared with 0.25% amitriptyline with epinephrine, and with 0.25% bupivacaine with epinephrine; and the combination of 0.25% amitriptyline and bupivacaine with epinephrine (i.e. , 0.5% total drug) is being compared with 0.5% amitriptyline with epinephrine, and with 0.5% bupivacaine with epinephrine.

The analgesic effectiveness of amitriptyline with and without epinephrine is significantly higher than bupivacaine at all the concentrations (with and without epinephrine) when compared with this integrated measure (P < 0.05;table 1). Furthermore, epinephrine significantly enhanced (P < 0.001) the analgesic effectiveness of amitriptyline at all the concentrations studied (table 1). The addition of epinephrine increased the AUC of 0.05% amitriptyline by approximately 300% (from 350 to 970% h), but increased the AUC of 0.5% amitriptyline by only 12% (from 1,578 to 1,769% h;table 1). Epinephrine had a significant effect on the analgesic effectiveness of bupivacaine only at the lowest (0.05%) concentration.

The effect of coinjecting amitriptyline, bupivacaine, and epinephrine on integrated analgesic effectiveness is even more pronounced (table 2). Considering first the comparison of each drug injected in combination with the sum of the same amount of drug injected individually with epinephrine, the AUC for the combination of 0.05% amitriptyline, bupivacaine, and epinephrine is not significantly more than the sum of that of the two drugs injected alone with epinephrine at the 0.05% concentration. However, the AUC of the coinjected 0.25% amitriptyline, 0.25% bupivacaine, and epinephrine combination was significantly higher that the sum of the AUCs of 0.25% amitriptyline injected alone with epinephrine and 0.25% bupivacaine injected alone with epinephrine (tables 1 and 2).

Second, considering the comparison of the sum of the amount of the individual drugs (with epinephrine) in the coinjected combinations (i.e. , total amount of the two drugs injected) with the same amount of each individual drug injected separately with epinephrine, the 0.125% amitriptyline, 0.125% bupivacaine, and epinephrine combination had an AUC significantly more than the AUC of either 0.25% amitriptyline injected alone with epinephrine, or 0.25% bupivacaine injected alone with epinephrine (P < 0.05;tables 1 and 2). Similarly, the AUC of the 0.25% amitriptyline, 0.25% bupivacaine, and epinephrine combination had an AUC significantly more than either 0.5% amitriptyline injected alone with epinephrine or 0.5% bupivacaine injected alone with epinephrine (P < 0.05;tables 1 and 2). The dose–response relations of the integrated analgesic effectiveness of amitriptyline and bupivacaine with and without epinephrine, and in combination, are plotted versus  concentration in figure 3.

All rats in the amitriptyline, bupivacaine, and combination groups recovered completely and showed no signs of neurobehavioral impairment or of local skin toxicity over a 2-week follow-up period after completion of the experiments.

This study describes three principle findings: (1) amitriptyline is an effective and long-acting LA for cutaneous analgesia in our rat model; (2) the cutaneous analgesia elicited by amitriptyline is enhanced by epinephrine; and (3) amitriptyline and bupivacaine potentiate their analgesic effectiveness when subcutaneously coinjected with epinephrine. The significance of these findings is discussed in the following sections.

We have shown that the duration of cutaneous analgesia (i.e. , duration until full recovery of nociception) is longer for amitriptyline than for bupivacaine, either alone or with epinephrine. We also found that amitriptyline is a more effective LA than bupivacaine for tissue infiltration in our rat model when integrated analgesia (AUC) is compared. This is consistent with previous work showing that amitriptyline is a more potent LA than bupivacaine for sciatic nerve blockade in rats. 12Amitriptyline and desiprimine have also been shown to have peripheral antinociceptive actions in the rat paw formalin test, in which these drugs decreased the local pain response after formalin injection. 22,27Inhibition of adenosine reuptake was cited as the mechanism underlying amitriptyline's peripheral antinociceptive action in this study, but its LA actions could not be excluded. One of the limitations of our study is that the inhibition of a nocifensive reflex (CTMR) was studied in response to drug injections, and the degree of inhibition of the CTMR was assumed to be indicative of the degree of analgesia.

The enhancement of cutaneous analgesia elicited by amitriptyline by the addition of epinephrine supports, in part, the vascular uptake hypothesis as a cause of the less dense block of amitriptyline than that of bupivacaine. The antagonism of the vasodilatory effects of amitriptyline by epinephrine 28would slow vascular uptake and increase the concentration locally in the tissues, thus increasing both the duration of the analgesia and the density of the block, as seen in figures 1A–C. The greater impact of epinephrine on the integrated analgesia (AUC) at lower concentrations of amitriptyline probably results from the greater antagonism of the epinephrine-induced vasoconstriction at higher concentrations of amitriptyline. The pharmacokinetics of subcutaneously or intradermally administered bupivacaine and its interactions with vasoconstrictors has been studied extensively and has been discussed elsewhere. 26,29Our results are consistent with previous studies showing that epinephrine prolongs the duration of bupivacaine block, depending on the site of administration and the dose of bupivacaine used. Our study showed that epinephrine enhances the cutaneous analgesic effectiveness of amitriptyline to a greater extent than that of bupivacaine, especially at lower concentrations, and that the effect of epinephrine decreases as the concentration of amitriptyline increases. We chose the epinephrine concentration of 5 μg/ml (1:200,000) because it is the most common clinically used concentration. A study of the effect of varying the concentration of epinephrine on the cutaneous analgesic effectiveness of amitriptyline would be worthwhile.

The analgesic effectiveness of amitriptyline and bupivacaine is potentiated by their coinjection in combination with epinephrine. All three parameters—duration of complete nociceptive block, duration until complete recovery of nociception, and the integrated analgesia (AUC)—were significantly greater for the amitriptyline–bupivacaine combinations than for the equivalent amounts of the individual drugs injected separately with epinephrine (tables 1 and 2). In addition, the duration of complete recovery for the 0.05% amitriptyline, bupivacaine, and epinephrine combination was significantly longer than the sum of the durations until complete recovery of 0.05% amitriptyline injected separately with epinephrine, and 0.05% bupivacaine injected separately with epinephrine (fig. 2). Similarly, the AUC of the 0.25% amitroptyline, bupivacaine, and epinephrine combination was significantly more than the sum of the AUCs of 0.25% amitriptyline injected separately with epinephrine and the 0.25% bupivacaine injected separately with epinephrine (tables 1 and 2). This supraadditive analgesic effect of peripherally administered amitriptyline and bupivacaine has not been reported previously. The exact mechanism underlying this effect is not known, but it also was seen in a study comparing the cutaneous analgesic effectiveness of low doses of bupivacaine enantiomers, i.e. , the racemic mixture was more effective than the equivalent dose of each enantiomer injected individually. 26Peripheral potentiation of the analgesic effect of infiltrated bupivacaine and lidocaine by the addition of ketamine 30and fentanyl, 31respectively, but not of clonidine, 32has also been reported. In these studies, the additives had transient LA properties when injected individually.

Possible explanations for this finding include both pharmacokinetic and pharmacodynamic phenomena. Amitriptyline and bupivacaine could facilitate the transport of one another through diffusion barriers; compete for and displace each other from local protein binding sites, thus increasing the amount of drug available for diffusion into nerves; or retard the dissipation of the coadministered drug into tissues or the bloodstream. Interestingly, the development of tachyphylaxis to the prolonged effects of LAs has been attributed to similar pharmacokinetic factors rather than to pharmacodynamic factors. 33The prevention of tachyphylaxis to the prolonged LA effects of amitriptyline and bupivacaine also could account for their potentiating effects after coadministration.

The advantages and disadvantages of combining LAs (e.g. , reduced or increased toxicity, rapid onset, prolonged duration, or potentiation) have been discussed elsewhere. 34–36Briefly, by using half the concentration of bupivacaine and amitriptyline in the coinjected mixture, and obtaining a significantly longer duration of analgesia than the equivalent total amount of bupivacaine alone with epinephrine (as shown in this study), the concerns about bupivacaine toxicity may be mitigated by using it in combination with amitriptyline and epinephrine. The reduced toxicity, accompanied by the prolonged duration of cutaneous analgesia, could make this combination clinically useful for infiltration anesthesia and postoperative analgesia.

The authors thank Gary R. Strichartz, Ph.D. (Professor of Anesthesiology [Pharmacology], Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women's Hospital, Boston, MA), for invaluable assistance and advice during this study.