| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The tricyclic antidepressants are often used for the treatment of neuropathic pain. In this study, we evaluated one of these drugs on human cutaneous experimental pain. A randomized, double-blinded, placebo-controlled, crossover design methodology was conducted. Subjects participated in 2 14-day study sessions separated by a 7-day washout period. One session was with desipramine and one with placebo. At baseline, Day 7, and Day 15, quantitative sensory testing was performed to thermal and mechanical stimuli. On Day 15 only, intradermal capsaicin was injected on the volar aspect of the forearm followed by an assessment of pain and hyperalgesia. Oral desipramine had no significant effect on acute sensory thresholds, pain, secondary hyperalgesia, or flare response induced by intradermal capsaicin. Mean peak plasma levels of desipramine were within the therapeutic range for the treatment of depression. This study further supports a lack of effect of the tricyclic antidepressants on acute nociception and experimentally-induced secondary hyperalgesia. IMPLICATIONS: Human experimental pain models have recently been developed; however, the efficacy of the tricyclic antidepressants (TCA) in these models has not been systematically studied. This investigation provides further validation of human experimental pain models and demonstrates that the chronic delivery of a TCA has no effect on human experimental pain.
Current thinking emphasizes that tissue injury can lead to persistent small diameter afferent fiber input that yields several psychophysically distinct components, including the pain sensation arising from the stimulus itself, as well as an exaggerated sensitivity to lower threshold stimuli applied to areas of the injury site (primary) and adjacent (secondary) to the injury site. Experimentally, this facilitated pain state can be evaluated in animals by injecting formalin into the hind paw (1), whereas in humans, this can be accomplished by injecting intradermal capsaicin (2,3). Intradermal capsaicin results in the transient (<2030 min) and selective activation of C fibers. This then results in a brief pain state, which, upon disappearance, is replaced by an enlarged area of tactile allodynia and thermal hyperalgesia that persists for an extended interval (47). Thus, human and animal studies show that repetitive, small-diameter afferent activity can induce a central state of facilitation corresponding to a state of hyperalgesia and allodynia that is referred to a skin region far larger than the extent of the original injury. Although the pharmacology of animal models of facilitated pain has been thoroughly described, the pharmacology of the human correlate is less well characterized. Preclinical animal models indicate that specific components of the facilitated pain state may possess a distinct modulatory pharmacology. This has led to important questions about the mechanisms and pharmacology that subserve the human pain state. Routinely, the investigation of the human pain state has used inadvertent or iatrogenic trauma (e.g., tissue injury or postoperative trauma). Such approaches are limited because (a) they represent multifactorial mechanisms (e.g., tissue and nerve injury or inflammation), (b) they involve multidrug therapies (e.g., anesthetics plus several adjuvants), (c) it is typically impossible to do controlled crossover interventions on the same subject, and (d) timing of the study is defined by clinical expediency. For these reasons, it is reasonable to use experimental interventions that permit the ethical study of the response of the human subject to experimental stimulus conditions that activate components of the systems outlined above, which subserve postinjury pain processing. Human cutaneous experimental pain models have been developed, and the pharmacology of the human cutaneous experimental pain state has recently been addressed with different classes of drugs that include the N-methyl-D-aspartate (NMDA) antagonists, opioids, and sodium channel antagonists (811). Systemic tricyclic antidepressants (TCA) have been widely investigated in both acute nociceptive pain and neuropathic pain, making them interesting targets to study in human experimental pain models. Systemic TCA significantly decrease the hyperalgesia in the formalin model of the rat (12,13); however, Eisenach et al. (9) demonstrated no effect for capsaicin-induced hyperalgesia in humans. The study by Eisenach et al. used acute boluses of amitriptyline, and because of the TCAs long half-life (24 h), therapeutic levels may require days to weeks of dosing. This study sought to evaluate the effects of a 2-wk course of oral desipramine (a TCA) on the pain and cutaneous hyperalgesia induced by intradermal capsaicin in healthy volunteers.
The study was approved by the University of California at San Diegos IRB. Twelve healthy volunteers (nine women and three men) were used in the study. Average age was 33 yr (range, 1951 yr), and average weight was 72 kg (range, 53106 kg). Informed consent was obtained after thorough explanation of the study protocol. A randomized, double-blinded, placebo-controlled, crossover design methodology was conducted. Subjects participated in 2 14-day study sessions separated by a 7-day washout period. One session was with desipramine, and one was with placebo. A research pharmacist prepared the desipramine and placebo capsules, which looked identical so as to blind the subject and investigator. Exclusion criteria included patients with peripheral nerve injury or neuropathic pain syndromes, persons currently taking antidepressants, patients with heart, hepatic, or renal insufficiency, allergy or hypersensitivity to desipramine, pregnancy, or presence of psychiatric illness that would interfere with the interpretation of results. An electrocardiogram was performed on all patients before participation and on Day 14 of each study session. Baseline blood pressure, heart rate, respiratory rate, and temperature were measured, then baseline neurosensory testing was performed on the volar aspect of the subjects left forearm. Subjects were instructed to take a single dose of desipramine at bedtime using the following schedule: Day 13, 50 mg; Day 46, 100 mg; Day 710, 200 mg; and Day 1114, 300 mg. Each day of the study, the patient was evaluated for side effects on a scale of 0100 (0 = none and 100 = the worst imaginable side effect). The patient was instructed to decrease to the next smaller dosing schedule if side effects were more than 30 of 100. Plasma levels were drawn on the morning of Days 7 and 15. At baseline, Day 7, and Day 15, quantitative sensory testing was performed to thermal and mechanical stimuli. Three neurosensory tests were performed: (a) warm and cool sensation, (b) hot and cold pain, and (c) touch. These tests were performed on the volar aspect of the left forearm. The same order of the stimuli was used in all subjects: touch, cool, warm, cold pain, and hot pain. This order was chosen because it goes from the lowest stimulus (touch) to the highest stimulus (hot pain). Warm and cool sensations were measured using a Thermal Sensory Analyzer (Medoc Advanced Medical Systems, Minneapolis, MN). This device consists of a thermode measuring 46 x 29 mm. The temperature of the thermode can either increase or decrease (at a rate of 1.0°C/s) depending on the direction of current flow through the device. The patient holds a switch that is pressed at the first sensation of warmth or cold; pressing the switch reverses the temperature change, returning to a neutral temperature of 32°C. Warm and cold pain measurements also use the Thermal Sensory Analyzer, but the end-point is pain instead of temperature change sensation. It uses a temperature change rate of 1.5°C/s. Touch was measured using von Frey hairs. Calibrated von Frey hairs are filaments of varying size. The filament was selected at random, and three successive stimuli were applied for 2 s at 5-s intervals per filament applied in an ascending pattern. The patient was instructed to report if the stimulus was felt. Thresholds are expressed in grams and measured as positive if the patient felt any of the three successive stimuli. At the stimulus intensity evoking a report of discomfort, the next stimulus is one unit lower. This stimulus reversal is repeated twice, and the average reversal intensity is defined as the threshold. This method is a modification of the widely used method of Dixon in animal and human psychophysical testing (see for example Wallace et al. (11)). On Day 15 only, intradermal capsaicin (8-methyl-N-vanillyl-6-nonamide) 100 µg/10 µl of a 20% cyclodextran vehicle was injected on the volar aspect of the left forearm. At 0, 5, 10, and 15 min after the capsaicin injection, spontaneous pain scores and evoked pain scores to von Frey hair, stroking, and heat (40°C) were measured using a visual analog scale (VAS). At 15 min after the injection, secondary hyperalgesia and flare response was measured. Secondary hyperalgesia was evaluated with a 5.18 von Frey hair (touch), a foam brush gently stroked on the skin (stroking), and a 2 x 2-cm probe heated to 40°C (heat). These stimuli started away from the injection site in a nonpainful area and moved progressively closer in radius until the subject reported pain or tenderness. That site was marked on the skin, and a total of eight determinations of the borders of secondary hyperalgesia were outlined on the skin. The area of secondary hyperalgesia and flare response was outlined onto a transparency for area determination (cm2). The postcapsaicin neurosensory thresholds were tested at a distance halfway between the edge of the area of hyperalgesia and the injection site. Pain scores were measured using a VAS. The subject places a mark along the line that corresponds with their pain. This line is 100-mm long with no pain at 0 mm and the worst imaginable pain at 100 mm. The distance (in millimeters) gives the measurement of pain. Using previously acquired data (11), sample size determinations for percentage change in pain scores were performed. Setting type I error rate at 0.05 and the type II error rate at 0.20 (i.e., power = 0.80), a sample size of 12 was determined to be adequate to detect a clinically relevant change in pain score of 30% or more (14). Data are expressed as the mean ± SD. Spontaneous and elicited (von Frey, stroking, and heat) pain scores postcapsaicin were analyzed by two-factor repeated-measures analysis of variance with both drug treatment and time postcapsaicin (0, 5, 10, and 15 min) as within-subjects factors. Allodynic areas (postcapsaicin) were analyzed by a paired t-test. Data for each sensory threshold measure and pain score collected at baseline, Day 7, and Day 15 of treatment were compared using a two-factor repeated-measures analysis of variance with both drug treatment (desipramine versus placebo) and day of test (baseline, Day 7, and Day 15) as within-subjects factors. Simple linear regression analysis was used to test for significant correlations between desipramine plasma levels and each component of the pain test battery and flare response on Day 15.
Thirteen volunteers were initially involved in the study. One participant withdrew because of intolerable side effects. Of the 12 patients that completed the trial, five reached the maximum dose of 300 mg/d. The mean dose on Days 7 and 15 was 100 mg (range, all patients reached 100 mg) and 225 mg (range, 100300 mg), respectively. Mean plasma levels on Days 7 and 15 were 89.4 ± 51.9 ng/mL and 204.2 ± 269.6 ng/mL, respectively. Intradermal capsaicin produced an immediate spontaneous pain sensation scored as 50 or higher on the VAS in 83% of patients. In addition, intradermal capsaicin evoked a secondary hyperalgesia to von Frey, stroking, and heat in 92%, 50%, and 42%, respectively. There was a measurable flare response in 92% of subjects. Using spontaneous and evoked pain scores after capsaicin as our primary efficacy measures, it was found that oral desipramine had no significant effect on spontaneous pain, nor was there a significant effect on the pain scores reported from thermal or mechanical pain stimuli after intradermal capsaicin (Fig. 1). Similarly, oral desipramine was without significant effect on the secondary hyperalgesia and flare response induced by capsaicin (Fig. 2). Moreover, there was no significant correlation between plasma levels on Day 15 and any pain measurement on that day (one subject who had a plasma level of 821 µg/mL more than two SD removed from the mean was excluded from this analysis).
Finally, neither placebo nor oral desipramine produced any significant changes from baseline upon any thermal, mechanical, or pain thresholds measured after 7 or 15 days of treatment (Table 1).
All but one subject reported side effects with desipramine. The two most common side effects were dry mouth and drowsiness, which occurred in eight and five subjects, respectively. Other side effects reported included sweating (three subjects), headache (two subjects), shaking (two subjects), insomnia (two subjects), nausea (two subjects), constipation (two subjects), vomiting (one subject), and dizziness. Figure 3 shows the mean peak side effect scores for the two most common side effects.
Although there are many well controlled studies that have demonstrated the efficacy of the TCA in neuropathic pain states, there are few on the effects of these drugs on acute nociception and only one on the effect on models of cutaneous hyperalgesia (9). In animal models of inflammatory pain, both amitriptyline and desipramine were effective (15,16). In placebo-controlled, blinded studies in postoperative pain, desipramine, but not amitriptyline, resulted in an opioid-sparing effect (1719). In animal models of neuropathic pain, amitriptyline and desipramine relieved the thermal hyperalgesia, but not the mechanical allodynia, of the Chung model (13,20). In contrast, Abdi et al. (21) demonstrated a positive effect of amitriptyline on the mechanical allodynia of the Chung model. However, Esser and Sawynok (20) used much larger doses of amitriptyline (10 mg/kg versus 1.5 mg/kg) and demonstrated a hyperesthesia of the contralateral paw, suggesting hyperalgesia at large doses. In the Bennet model, venlafaxine, a new antidepressant that inhibits both norepinephrine and serotonin reuptake, has been shown to relieve the thermal hyperalgesia (22). In addition, the acute and chronic administration of amitriptyline and desipramine was effective in the Bennet model (23). In human neuropathic pain, the TCA were effective in diabetic neuropathy (24,25) and postherpetic neuralgia (26,27).
The analgesic effects of the TCA in neuropathic pain states make it an interesting target for studies in human experimental pain. The antidepressants are complicated drugs that serve to block the reuptake of catecholamines, thereby enhancing adrenergic transmission (28,29). They also serve as NMDA receptor antagonists (30). It is appreciated that bulbospinal noradrenergic pathways can regulate dorsal horn nociceptive processing by the release of noradrenaline and the subsequent activation of
Previous studies on experimentally-induced hyperalgesia have demonstrated different effects of the opioids and non-opioids. The acute systemic delivery of opiates and NMDA antagonists and the spinal delivery of
Intradermal capsaicin results in the transient (<2030 minutes) and selective activation of C fibers. In addition, intradermal capsaicin results in a rapid onset of a flare response, which peaks at approximately three to five minutes (2). This response represents antidromic invasion of the axon collaterals and the subsequent release of neuropeptides (34). Previous studies have shown that the systemic delivery of sodium channel antagonists significantly decreases the flare response secondary to intradermal capsaicin, suggesting a peripheral stabilization of the unmyelinated nerve terminals (11). There are several lines of evidence that NMDA and One criticism of this study could be the disproportionate number of men and women (nine women and three men). Studies have shown that there are sex differences in both pain response and analgesic response (39). This could be one explanation for the lack of effect of desipramine on capsaicin-induced hyperalgesia. Desipramine is a TCA that is more selective for norepinephrine than serotonin reuptake inhibition. It has a small anticholinergic effect thus fewer side effects (40). This small side effect profile makes it an attractive drug and is the rationale for choosing the drug for this study. Despite this small side effect profile, all but five subjects had dose-limiting side effects. However, the mean plasma level of 204.2 ng/mL was well within the antidepressant therapeutic range. This study further supports a lack of effect of the TCA on acute nociception and experimentally-induced secondary hyperalgesia. It also suggests that the TCA have a weak, if any, NMDA antagonist property, because other studies using NMDA antagonists have demonstrated significant effects on experimental secondary hyperalgesia. Whether the TCA have a better therapeutic index when administered intraspinally will be difficult to determine because animal studies have demonstrated a possible spinal toxicity (41). In addition, this study further demonstrates the utility of human experimental pain in evaluating analgesic efficacy.
Supported, in part, by NIH K08 NS01909-01A2 and the Pharmaceutical Research and Manufacturers of America Grant.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|