Share this article on:

The Effect of Antinociceptive Drugs Tested at Different Times After Nerve Injury in Rats

Hama, Aldric T. PhD; Borsook, David MD, PhD

doi: 10.1213/01.ANE.0000155247.93604.62
Pain Medicine: Research Report

Given the evolving nature of anatomical and functional changes in the nervous system that are involved in the development of neuropathic pain, it is possible that the differing time course after injury underlies the inconsistent efficacy of drugs in neuropathic pain patients. In the current study, we evaluated the behavioral effects of two standard drugs used clinically for neuropathic pain, the anticonvulsant gabapentin and antidepressant imipramine, in rats at different times after peripheral nerve injury. Rats that underwent the spared nerve injury procedure responded to an innocuous mechanical stimulus (von Frey filament) 2, 4, and 8 wk after injury. Gabapentin dose-dependently suppressed mechanical sensitivity at all time points tested but the potency of gabapentin was three-fold less 4 wk postinjury (135 mg/kg) compared with 2 and 8 wk postinjury (41 and 44 mg/kg, respectively). In contrast, imipramine lacked significant efficacy at 2 and 8 wk postinjury but slightly attenuated mechanical hypersensitivity at 4 wk postinjury. The results show that drug effects may change over time in the neuropathic state, which should be an important consideration in the evaluation of drugs in preclinical animal pain models and has implications for temporal approaches to therapy in the clinic.

IMPLICATIONS: The efficacy of antineuropathic pain drugs varies from patient to patient. To understand the mechanism of this phenomenon, gabapentin and imipramine were tested in rats with neuropathic pain at various times after nerve injury. The results indicate that drug efficacy may depend on the length of time between injury and administration.

Descartes Therapeutics, Inc., Waltham, Massachusetts

Accepted for publication December 16, 2005.

Address correspondence and reprint requests to David Borsook, MD, PhD, P.A.I.N. Group Brain Imaging Center McLean Hospital, 115 Mill St., Belmont, MA 02478. Address e-mail to

Neuropathic pain occurs after disease or trauma to peripheral nerves and does not respond in a consistent manner to therapeutics (1,2). Although the etiology may be similar within a given group of neuropathic pain patients, the onset and duration of hypersensitivity and pain is very heterogeneous. The inconsistent efficacy of drugs in neuropathic patients is likely because of the fact that patients are often evaluated months or years after the initial injury and thus have been suffering from neuropathic pain for a period of time, which varies from patient to patient.

Two drug classes that have been used in patients with neuropathic pain include anticonvulsants and antidepressants (1,3). The clinical response to drugs may vary across individuals for a number of reasons, including genetic (4) or placebo-related effects (5,6). Indeed, even after acute surgical pain, the response to morphine varies considerably among patients (7). Drug efficacy may also relate to specific drug mechanisms of action or disease processes and is represented, for example, as numbers needed to treat (NNT). For example, for the anticonvulsant gabapentin the NNT for postherpetic neuralgia is 3.2 and 3.8 for diabetic neuropathy (8). Similar data on NNT for antidepressants have also been reported (1); the numbers indicate that pain relief is elusive for many patients.

After nerve injury, progressive changes occur at a functional level. Measurable hypersensitivity to stimuli in animal models of neuropathic pain may appear within days after surgery, and pain may persist for weeks or months, well beyond the time required for tissue healing. Clinically, this is seen as variations in times ranging from days to months that patients report neuropathic symptoms after injury (2). In addition, neuropathic pain seems to become more resistant to drugs (e.g., morphine) over time, which correlates with altered opioid receptor expression in the peripheral (PNS) and central nervous system (CNS) (9,10). A number of studies have indicated cellular and molecular changes in animal models of neuropathic pain over time, including altered expression of genes, channels or receptors, anatomical changes (11–13), and functional changes (14,15). Based on these factors, it is postulated that antinociceptive drugs may have different effects over the course of the evolution of neuropathic pain.

Animal models of neuropathic pain have been previously described, which exhibit many of the relevant symptoms that are present in humans. An advantage of using animal models is that variability such as time postinjury may be standardized, avoiding temporal differences in CNS changes that may affect the efficacy of drugs. Despite this advantage, preclinical drug evaluation is inconsistent within studies, ranging anywhere from days to months after injury (e.g., 16). It is possible that drug effects, based on particular CNS or PNS targets, may change over time in parallel with changes in the relevant drug target. Thus, drugs that apparently have no efficacy in these rats at a particular time may have efficacy at some later time after injury. The goal of the current study was to evaluate the behavioral effects of drugs at different time points in rats with a peripheral nerve injury.

Back to Top | Article Outline


Procedures were approved by the internal institutional animal care and use committee and followed guidelines of the National Institutes of Health. A unilateral spared nerve injury (SNI) was performed in male Sprague-Dawley rats (100–150 g; Harland, Indianapolis, IN), as previously described (17,18). At specific time points after nerve injury (2, 4, and 8 wk), sensitivity to a von Frey filament (5.18 to 15 g) was evaluated in these rats similar to a procedure previously described (19). Briefly, rats were placed in individual plexiglas containers with a wire mesh floor. After an acclimation period, the filament was pressed to the lateral left (nerve-injured) hindpaw. A positive response was a brisk hindpaw withdrawal away from the filament, and the total number of responses out of five presentations was recorded. The total number of responses was divided by 5, and a response frequency (%) was calculated. To be included in the study, rats required a minimum of three positive responses. After baseline response frequency determination, rats were injected with either vehicle (saline), gabapentin (10, 30, and 100 mg/kg; Toronto Research Chemicals, Canada), or imipramine (20 mg/kg; Sigma-Aldrich, St. Louis, MO). Rats were tested every 30 min for up to 180 min after injection. Rats were injected and tested once and were killed by CO2 overdose.

A two-way analysis of variance for repeated measures was used to analyze treatment effect over time. Where significant drug and time interactions were observed, the data were analyzed with Student-Newman-Keuls test as the post hoc test. Statistical significance was taken at P < 0.05. To determine the dose at which drug efficacy was 50% of maximal (ED50), a modified computer program adapted from Tallarida and Murray (20) was used to calculate this value.

Back to Top | Article Outline


Before SNI surgery, rats displayed minimal responsiveness to the von Frey filament (0% response frequency). After surgery but before injection, the mean baseline response frequencies (percentage ± sem) were 77% ± 3%, 91% ± 2%, and 93% ± 2% at 2, 4, and 8 wk, respectively; response frequencies were significantly increased at 4 and 8 wk after nerve injury (P < 0.05 versus 2 wk).

Gabapentin dose-dependently decreased mechanical hypersensitivity 2, 4, and 8 wk after nerve injury (Fig. 1; P < 0.05 versus vehicle and baseline). In contrast, vehicle did not affect the response frequency. The maximal efficacy of gabapentin was similar whether it was tested at 2, 4, or 8 wk after injury. The peak efficacy of gabapentin seemed to be at 120 min postinjection; thus, the ED50 was calculated at that time point. The ED50 of gabapentin was 41 and 44 mg/kg at 2 and 8 wk postinjury, respectively, whereas the ED50 at 4 wk was 135 mg/kg (P < 0.05 versus 2 and 8 wk).

Figure 1

Figure 1

Rats injected with 100 mg/kg of gabapentin showed decreased exploratory behavior compared with vehicle-treated rats. However, rats did move when prodded and were not ataxic (16). It should be noted that all rats, whether drug- or vehicle-injected, were alert and resting on all four paws during filament probing.

Imipramine had no significant effect on mechanical hypersensitivity at 2 and 8 wk postinjury. Although there was no statistically significant interaction between time and treatment, at 4 wk postinjury imipramine reduced mechanical hypersensitivity by approximately 30%–40% (P < 0.05 versus vehicle at 30, 60, 120, and 180 min postinjection).

Imipramine 20 mg/kg did not seem to impair exploratory behavior in rats. Although we were curious to evaluate the effect of a larger dose of imipramine (50 mg/kg), it was not tested because of possible hypotension (whitening of the ears) and an obvious decrease in activity.

Back to Top | Article Outline


In the current study, a SNI led to a robust and persistent mechanical hypersensitivity. Gabapentin dose-dependently attenuated mechanical hypersensitivity, regardless of duration of nerve injury, but the potency was reduced four weeks after injury. By contrast, imipramine seemed to have efficacy only at a particular time after injury. These data suggest a significant time-related component of neuropathic pain that may influence the efficacy of antinociceptive drugs.

A group of patients with neuropathic pain with a specific cause (e.g., trigeminal neuralgia) may not have the same severity of symptoms or may differentially respond to medications (7). Part of the reason for the heterogeneity may be due to variances in disease onset and disease mechanism within the group. Data obtained from rats indicate changes in CNS and PNS pain-related molecules and brain function over time after nerve injury (11–15). Thus, patients in drug trials are randomized based on duration of pain in the hope of normalizing possible PNS or CNS functional changes between the groups. Although, such a strategy invites excessive variability and will not definitively determine if a drug is effective or not. In contrast, animal models lend themselves to better temporal as well as behavioral consistency.

Based on the literature, however, there is wide divergence of testing times postinjury (e.g., 21 and 22). One extreme example of this is that drugs have been evaluated in SNI rats between one week and seven months after injury (16). The lack of effect of gabapentin on mechanical hyperalgesia and cold allodynia observed by Erichsen and Blackburn-Munro (16) may have been related to their broad postinjury testing time point. Because drug data were collected from a variety of time points, it is possible that efficacy at one point may have been obscured by the lack of efficacy at others. The current results—that drug effects differ depending on when the drug is administered after injury and the fact that CNS processes related to neuropathic pain change over time (12,13,15)—point out the need for consistency in the evaluation of drugs. Also, evaluating novel drugs at only one time point after surgery may not be an adequate way to define potency. There are of course other factors to consider that may lead to a negative effect of a drug (e.g., model and testing method), but one that may be easily controlled is time after injury.

Gabapentin was chosen in the current study because it has been consistently reported to alleviate mechanical allodynia/hypersensitivity in neuropathic rats (21,22). The current data support the existing preclinical data with respect to observing a significant antinociceptive effect of systemic gabapentin. However, the current data demonstrate almost full reversal of mechanical hypersensitivity with a dose of 100 mg/kg, whereas De Vry et al. (22) did not attain even 50% reversal with this dose. It is possible that the difference may be caused by fundamental mechanistic differences in models (chronic constriction injury versus SNI) or that sensitivity to gabapentin is time-dependent. De Vry et al. (22) tested gabapentin 27–31 days after surgery, corresponding with the time point at which gabapentin potency was lowest in the current study. Our data suggest that, in the SNI model, gabapentin may have a differential potency depending on when it is administered after nerve injury. Although Luo et al. (21) demonstrated marked differences in expression levels of the putative binding site of gabapentin (the α2δ subunit of the voltage-gated calcium channel) between the spinal nerve ligation and chronic constriction nerve injury models, the efficacy of gabapentin in these two models was similar. Because a correlation was not observed between gabapentin efficacy and levels of α2δ subunit expression, it is possible that this subunit may play other nongabapentin-related roles in the nerve-injured state (21). Although gabapentin in the SNI model has reduced potency at four weeks after injury, it is not known whether this reduction is related to the α2δ subunit or to some other molecule.

Tricyclic antidepressants (TCAs), in addition to gabapentin, have also been used in neuropathic pain patients (1,2). The lack of robust efficacy in the SNI rats is surprising. The dose of imipramine used in the current study is in the dose range shown to be effective in other pain models and rat depression models; thus, the dose is not likely to be an issue (23). Others, however, have also reported a lack of efficacy of TCAs in nerve injury models (24,25). Decosterd et al. (24) tested (repeated doses of) amitriptyline in rats between two and four weeks after SNI and showed no efficacy against mechanical allodynia. The data seem to be the combination of results at two to four weeks, showing an apparent lack of effect. Rodrigues-Filho et al. (25) may have missed imipramine’s efficacy (testing 20–40 days after nerve injury) for a similar reason. In the current study, a small reversal of hypersensitivity was observed with imipramine four weeks after injury. Given the current data, one may conclude that the SNI model is insensitive to TCAs at particular times after nerve injury. Thus, the reported lack of effect of TCAs may be due to previous evaluation at time points when these drugs have little or no effect on mechanical hypersensitivity. Determining why this is the case may uncover a time-dependent variable that underlies the effect of TCAs in human neuropathic pain.

There are numerous variables that hinder defining drug efficacy in human neuropathic pain, one of which is duration of time after injury. With animal models, however, it is possible to control for this particular variable to critically evaluate drug effects long after the injury and onset of neuropathic pain symptoms. The increasing body of evidence suggests that the maintenance of neuropathic pain is a dynamic process, involving anatomical and functional changes over time. Thus, mechanistic studies and drug discovery efforts may need to examine a chronic pain state over various time points rather than at a single arbitrary time point. A better understanding of the evolving process in chronic pain states may lead to significantly more effective treatments.

Back to Top | Article Outline


1. McQuay HJ, Tramer M, Nye BA, et al. A systematic review of antidepressants in neuropathic pain. Pain 1996;68:217–27.
2. Bonica JJ. Management of pain. 2nd ed. Philadelphia: Lippincott Willams & Wilkins, 1990.
3. Jensen TS. Anticonvulsants in neuropathic pain: rationale and clinical evidence. Eur J Pain 2002;6:A61–8.
4. Mogil JS. The genetic mediation of individual differences in sensitivity to pain and its inhibition. Proc Natl Acad Sci USA 1999;96:7744–51.
5. Amanzio M, Pollo A, Maggi G, Benedetti F. Response variability to analgesics: a role for non-specific activation of endogenous opioids. Pain 2001;90:205–15.
6. Benedetti F. How the doctor’s words affect the patient’s brain. Eval Health Prof 2002;25:369–86.
7. Aubrun F, Langeron O, Quesnel C, et al. Relationships between measurement of pain using visual analog score and morphine requirements during postoperative intravenous morphine titration. Anesthesiology 2003;98:1415–21.
8. Wiffen P, Collins S, McQuay H, et al. Anticonvulsant drugs for acute and chronic pain. Cochrane Database Syst Rev 2000:CD001133.
9. Ozaki S, Narita M, Narita M, et al. Suppression of the morphine-induced rewarding effect in the rat with neuropathic pain: implication of the reduction in mu-opioid receptor functions in the ventral tegmental area. J Neurochem 2002;82:1192–8.
10. Truong W, Cheng C, Xu QG, et al. Mu opioid receptors and analgesia at the site of a peripheral nerve injury. Ann Neurol 2003;53:366–75.
11. Wang H, Sun H, Della Penna K, et al. Chronic neuropathic pain is accompanied by global changes in gene expression and shares pathobiology with neurodegenerative diseases. Neuroscience 2002;114:529–46.
12. Goff JR, Burkey AR, Goff DJ, Jasmin L. Reorganization of the spinal dorsal horn in models of chronic pain: correlation with behavior. Neuroscience 1998;82:559–72.
13. Cameron AA, Cliffer KD, Dougherty PM, et al. Time course of degenerative and regenerative changes in the dorsal horn in a rat model of peripheral neuropathy. J Comp Neurol 1997;379:428–42.
14. Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain: a review and meta-analysis (2000). Neurophysiol Clin 2000;30:263–88.
15. Bruggemann J, Galhardo V, Apkarian AV. Immediate reorganization of the rat somatosensory thalamus after partial ligation of sciatic nerve. J Pain 2001;2:220–8.
16. Erichsen HK, Blackburn-Munro G. Pharmacological characterisation of the spared nerve injury model of neuropathic pain. Pain 2002;98:151–61.
17. Decosterd I, Woolf CJ. Spared nerve injury: an animal model of persistent peripheral neuropathic pain. Pain 2000;87:149–58.
18. Lee BH, Won R, Baik EJ, et al. An animal model of neuropathic pain employing injury to the sciatic nerve branches. Neuroreport 2000;11:657–61.
19. Kim SH, Chung JM. An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 1992;50:355–63.
20. Tallarida RJ, Murray RB. Manual of pharmacologic calculations with computer programs. New York: Springer-Verlage, 1983.
21. Luo ZD, Calcutt NA, Higuera ES, et al. Injury type-specific calcium channel alpha 2 delta-1 subunit up-regulation in rat neuropathic pain models correlates with antiallodynic effects of gabapentin. J Pharmacol Exp Ther 2002;303:1199–205.
22. De Vry J, Kuhl E, Franken-Kunkel P, Eckel G. Pharmacological characterization of the chronic constriction injury model of neuropathic pain. Eur J Pharmacol 2004;491:137–48.
23. Korzeniewska-Rybicka I, Plaznik A. Analgesic effect of antidepressant drugs. Pharmacol Biochem Behav 1998;59:331–8.
24. Decosterd I, Allchorne A, Woolf CJ. Differential analgesic sensitivity of two distinct neuropathic pain models. Anesth Analg 2004;99:457–63.
25. Rodrigues-Filho R, Campos MM, Ferreira J, et al. Pharmacological characterisation of the rat brachial plexus avulsion model of neuropathic pain. Brain Res 2004;1018:159–70.
© 2005 International Anesthesia Research Society