Capsaicin evoked pain and allodynia in post-herpetic neuralgia

The hypothesis that the pain and allodynia associated with post-herpetic neuralgia (PHN) is maintained by a combination of input from preserved primary afferent nociceptors and sensitization of central pain transmitting neurons was examined in 17 subjects with PHN. Pain, allodynia, thermal sensory function, cutaneous innervation, and response to controlled application of 0.075% capsaicin were measured. Compared to mirror-image skin, applying capsaicin on a 9 cm(2) area of PHN skin significantly increased overall PHN pain and allodynia in 11 of 17 subjects. These 'capsaicin responders' were characterized by higher average daily pain, higher allodynia ratings, and relatively preserved sensory function at baseline compared to the non-responders. In three of the 'capsaicin responders' the area of allodynia expanded into previously non-allodynic and non-painful skin that had normal sensory function and cutaneous innervation. These observations support the hypothesis that allodynia in some PHN patients is a form of chronic secondary hyperalgesia maintained by input from intact and possibly 'irritable' primary afferent nociceptors to a sensitized CNS.     

Post-herpetic neuralgia: the relation of pain complaint, sensory disturbance, and skin temperature

Twelve otherwise healthy patients with longstanding postherpetic neuralgia (PHN) were prospectively studied using clinical examination, infrared thermography and response to local anesthetic skin infiltration. All had at least 2 of 3 possible components to their PHN pain: continuous, neuralgic, or allodynic. In patients with allodynia, maximal reported pain and the location of maximal allodynia on sensory examination were largely overlapping and were often warm thermographically. Areas of dense sensory loss and skin scarring without allodynia were usually cool thermographically. Local anesthetic skin infiltration produced substantial pain relief in all 9 patients (essentially complete relief in 7) with allodynia: the 3 patients with predominantly continuous pain were not relieved. In 7 of 8 skin infiltration responders, the same dose of lidocaine i.m. in the deltoid muscle also produced significant, though less complete pain relief. These results suggest that PHN patients can be divided into at least 2 clinical groups: those with predominantly continuous pain localized to a region of significant sensory loss and those in whom allodynia is the most prominent sensory disturbance. The latter group has pain localized to areas with relatively preserved sensation. The differences in clinical features and response to lidocaine suggest that there are at least 2 different mechanisms contributing to the pain of PHN.     

Natural history of cutaneous innervation following herpes zoster

As part of a comprehensive study of the natural history of herpes zoster (HZ), 57 of 94 subjects in a cohort at elevated risk for post-herpetic neuralgia (PHN) consented to collection of 3-mm skin punch biopsies from affected, mirror-image, and distant control skin at baseline and followup visits. As cutaneous innervation is reduced in longstanding severe PHN, we tested the hypothesis that development of PHN is correlated with severity of initial neural injury and/or a failure of neural recovery. Quantitative analysis using single-label PGP9.5 immunofluorescence microscopy showed epidermal profiles were reduced in zoster skin by approximately 40% at study entry compared to control and mirror skin. The density of the subepidermal plexus was approximately 15% lower in zoster skin. Mirror skin was not denervated compared to control skin. Although not significant at all visits, correlations between epidermal nerve fiber density in HZ skin and thermal sensation, allodynia, capsaicin response, and average daily pain all associated more severe abnormalities with lower epidermal innervation. There was limited evidence that the initial neural injury was more severe in the 15 eventual PHN subjects. Overall, pain and pain-related disability resolved the fastest. Sensory abnormalities and symptom aggravation by focal capsaicin application showed partial and selective recovery over 6 months. In contrast, cutaneous innervation showed no recovery at all by 6 months, conclusive evidence that resolution of pain and allodynia does not require cutaneous reinnervation. A much longer period of observation is needed to determine if zoster-affected skin is ever reinnervated.     

Naloxone antagonizes the local antihyperalgesic effect of fentanyl in burnt skin of healthy humans.

The aim of this study was to investigate local opioid effects in the inflamed skin of healthy human volunteers. To induce inflammation, the circular tip of a 10-mm-diameter probe was heated to 48 degrees C and applied for 120 seconds to a site on each forearm of 24 healthy participants. Thirty minutes later, 0.2 mL of normal saline was injected subcutaneously into 1 inflamed site, and the opioid antagonist naloxone hydrochloride (80 microg in 0.2 mL) was injected subcutaneously into the other inflamed site. Participants completed tests of pain sensitivity (heat pain thresholds, heat pain ratings, and mechanical pain ratings) before and after the injections. Fentanyl citrate (10 microg in 0.2 mL) was then injected into the pretreated sites, and pain sensitivity was measured again. The thermal injuries produced thermal and mechanical hyperalgesia that did not differ between the saline and naloxone sites. After the fentanyl injections, decreases in thermal and mechanical hyperalgesia were greater at the saline site than the naloxone site. These findings demonstrate that pretreatment with naloxone blocks local opioid effects produced by the subcutaneous injection of a low dose of fentanyl in the inflamed skin of healthy humans. Thus, peripheral opioid receptors could be a therapeutic target for painful cutaneous disorders. PERSPECTIVE: This article demonstrates that activation of opioid receptors in the skin inhibits sensitivity to painful mechanical and thermal stimuli. Thus, local application of low-dose opioid medications could relieve painful skin disorders.     

Relief of post-herpetic neuralgia by surgical removal of painful skin

We present a case of longstanding PHN treated by skin excision of the area of greatest pain (11.3 x 26.0 cm(2)). The operation reduced pain, eliminated tactile allodynia, and facilitated greatly reduced medication use over a 1-year follow-up period. Fourteen punch biopsies and 10 strips of skin (each 10 mm long) from the excised painful PHN skin were qualitatively assessed by double-label immunofluorescence using antibodies against protein-gene-product 9.5 (PGP9.5), 200 kDa neurofilament protein (NF), calcitonin gene-related peptide (CGRP) and vanilloid receptor-1 (VR-1). Compared with a punch biopsy from mirror image skin, the pattern of cutaneous innervation in PHN skin was consistently and substantially different. The results may explain the anatomical basis of the capsaicin-response test and have implications for our understanding of clinical mechanisms underlying PHN pain.     

Relief of post-herpetic neuralgia by surgical removal of painful skin: 5 years later

Surgical removal of painful skin was first attempted as a treatment for chronic intractable post-herpetic neuralgia (PHN) more than a century ago, but long-term follow-up has rarely been reported. A patient who underwent surgical excision of 294cm(2) of thoracic skin comprising the entire area of pain and allodynia in October 2000 has been followed for 5.5years post-surgery. Our initial report presented evidence of benefit in the form of reduced pain, elimination of allodynia, and reduced medication consumption during the first post-operative year. Unfortunately, pain steadily increased and now exceeds pre-surgery levels despite increased medication use. Pain topography and characteristics are different from pre-surgery and may relate to the pathophysiology of PHN. Skin resection cannot be recommended as a treatment for PHN.     

Central metabotropic glutamate receptors differentially participate in interleukin-1beta-induced mechanical allodynia in the orofacial area of conscious rats.

The present study investigated the role of central metabotropic glutamate receptors (mGluRs) in interleukin-1beta (IL-1beta)-induced mechanical allodynia and mirror-image mechanical allodynia in the orofacial area. Experiments were carried out on male Sprague-Dawley rats weighing 230 to 280 g. After administration of 0.01, 0.1, 1, or 10 pg of IL-1beta into a subcutaneous area of the vibrissa pad, we examined the withdrawal behavioral responses produced by 10 successive trials of an air-puff ramp pressure applied ipsilaterally or contralaterally to the IL-1beta injection site. Subcutaneous injection of IL-1beta produced mechanical allodynia and mirror-image mechanical allodynia in the orofacial area. Intracisternal administration of CPCCOEt, a mGluR1 antagonist, or MPEP, a mGluR5 antagonist, reduced IL-1beta-induced mechanical allodynia and mirror-image mechanical allodynia. Intracisternal administration of APDC, a group II mGluR agonist, or L-AP4, a group III mGluR agonist, reduced both IL-1beta-induced mechanical allodynia and mirror-image mechanical allodynia. The antiallodynic effect, induced by APDC or L-AP4, was blocked by intracisternal pretreatment with LY341495, a group II mGluR antagonist, or CPPG, a group III mGluR antagonist. These results suggest that groups I, II, and III mGluRs differentially modulated IL-1beta-induced mechanical allodynia, as well as mirror-image mechanical allodynia, in the orofacial area. PERSPECTIVE: Central group I mGluR antagonists and groups II and III mGluR agonists modulate IL-1beta-induced mechanical allodynia and mirror-image mechanical allodynia in the orofacial area. Therefore, the central application of group I mGluR antagonists or groups II and III mGluR agonists might be of therapeutic value in treating pain disorder.     

Tactile allodynia in patients with postherpetic neuralgia: lack of change in skin blood flow upon dynamic stimulation

Tactile allodynia is a common, troublesome feature of neuropathic pain. Allodynia has been proposed to involve abnormal Abeta-afferent coupling in the dorsal horn resulting in C-fibre activation and increased skin blood flow (SBF). Thus, changes in SBF could provide an objective measure of allodynia. We searched for this mechanism in patients with postherpetic neuralgia (PHN) with varying degrees of cutaneous sensory loss. We mapped the allodynic area in PHN patients using cotton buds and von Frey hairs. Quantitative thermal testing was performed to assess small fibre function in the affected and mirror-image areas. At a subsequent visit the area of allodynia was remapped. Then the SBF in the affected and control areas was quantified before and after allodynic stimulation using laser Doppler imaging and subsequent single point continuous monitoring to detect rapid changes. We enrolled 10 PHN patients (medians: age 77 yrs, duration 20 months, ongoing pain 5). The allodynic area (range 11-546 cm2) was stable across the sessions. Thermal testing showed similar (n=5) or reduced (n=5) warmth and pain sensation in the affected versus control area. Following allodynic stimulation (median evoked pain-5) we saw no changes in SBF using either imaging (repeated measures ANOVA, P=0.73) or single point monitoring. This was the case for all patients regardless of the degree of sensory impairment in the affected dermatome. In conclusion, in a representative population of PHN patients we found no evidence of changes in SBF in response to allodynic stimulation. Hence, SBF measurements are not suitable for assessing allodynia.     

Topical lidocaine reduces pain in post-herpetic neuralgia

We report the results of a single session, non-blinded, trial of topical application of 10% lidocaine in gel form to the painful skin of 11 patients with well established post-herpetic neuralgia (PHN). Pain decreased as measured by 100 mm VAS pain scale and 100 mm VAS pain relief scale in both trigeminal and thoracic PHN patients.     

Dynamic mechanical allodynia following finger amputation: Unexpected skin hyperinnervation

The development of chronic pain after amputations is not an uncommon event. In some cases the most disabling problem is represented by the symptom called dynamic mechanical allodynia, characterized by the painful sensation evoked by gently stroking the skin. Despite the growing interest in understanding pain mechanisms, little is known about the mechanism sustaining this peculiar type of pain. We present here the case of a 53-year-old female patient who complained of severe tactile allodynia in the hand after amputation of her left second finger, resistant to several medical and surgical treatments. In order to gain information about the pain mechanism, two neurodiagnostic skin biopsies were obtained from the area of tactile allodynia and from the contralateral, normal skin area. Skin biopsies showed an unexpected increased innervation of the allodynic skin compared to the contralateral, normal skin area (+ 80.1%). Hyperinnervation has been proposed as a mechanism of pain following nerve lesions, but the increased innervation described here could be also attributed to neuronal plasticity occurring in chronic inflammatory conditions. Independently from the uncertain cause of the epidermal hyperinnervation, in this patient we tried to reduce the elevated number of epidermal nerve fibres by treating the skin with topical capsaicin (0.075%) three times a day, and obtained a persistent pain relief. In conclusion, neurodiagnostic skin biopsy might represent an useful tool for detecting derangements of epidermal innervation in patients with dynamic mechanical allodynia and can help to select an individually tailored therapeutic strategy in such difficult clinical conditions. Further studies are needed to clarify this issue and try to gain better understanding of chronic pain mechanisms in patients who underwent finger amputation.     

Psychophysical observations on patients with neuropathic pain relieved by a sympathetic block

Patients with sympathetically maintained pain (SMP) were tested with noxious heat pulses, innocuous mechanical stimuli, and transcutaneous electrical nerve stimulation before and during local anesthetic sympathetic blocks that relieved their pain. The perceived intensity of the pain evoked by these stimuli was measured by the patients' responses on a visual analog scale and compared to the responses obtained when the same stimuli were applied to contralateral normal skin. In 5 of 7 patients tested, graded noxious heat stimuli (43-51 degrees C) applied to painful skin resulted in heat-pain intensity ratings that were essentially identical to the responses obtained when the same stimuli were applied to the normal side. Of the remaining two patients, one was clearly hypoalgesic for heat-pain and the other was probably hyperalgesic. The normal and subnormal heat-evoked responses obtained from abnormal skin were unchanged during completely successful sympathetic blocks. Trains of noxious heat pulses (52 degrees C) evoked summation of the second pain sensation in each of the 4 patients tested. This summation effect was normal and unaffected by a sympathetic block. Four of the patients had allodynia evoked by mechanical stimulation. In each of the 3 allodynia cases tested, transcutaneous nerve stimulation at an intensity that was at threshold for detection evoked burning pain and a coexistent sensation of tingle, indicating that both sensations were due to the activation of A beta axons. Patients without touch-evoked pain reported that electrical stimuli at threshold for detection produced only the sensation of tingle. The pains evoked by touch and by threshold-strength nerve stimulation were eliminated during sympathetic block. In patients with allodynia, trains of gentle mechanical stimuli and trains of threshold-strength electrical nerve stimuli produced summation of the intensity of the burning pain sensation when the stimuli were presented at 0.3 Hz. These results add to a growing body of evidence indicating that the touch-evoked pain of some patients is due to abnormal central activity evoked by input from A beta low-threshold mechanoreceptors. The coexistence of A beta-evoked pain with normal heat-evoked pain and normal heat-pain summation suggests that the central abnormality cannot be a simple hypersensitivity of wide-dynamic-range neurons. The effect of sympathetic blockade on A beta-evoked pain and its summation suggests that the crucial sympathetic interaction may take place centrally. The results show that there is considerable heterogeneity of sensory abnormalities among patients with SMP.(ABSTRACT TRUNCATED AT 400 WORDS)     

Natural history of sensory function after herpes zoster

The natural history of sensory function in the first 6 months after herpes zoster (HZ) was determined in a cohort of 94 subjects at elevated risk for developing post-herpetic neuralgia (PHN). All four visits included ratings of pain and sensory symptoms, mapping areas of altered sensation and allodynia, and quantitative thermal and mechanical sensory testing. The last three visits included the capsaicin response test. Sensory thresholds in distant control skin were stable. Mirror-image skin was persistently hyperesthetic to warming and mechanical stimuli and hyperalgesic to heat compared to distant control skin. HZ skin showed deficits in all thermal modalities. Sensory recovery was limited and selective. Allodynia area and severity, hyperalgesia to von Frey hair, and cold detection threshold improved, but deficits to warmth and heat pain did not. Capsaicin on HZ skin significantly aggravated pain and allodynia in the majority of subjects at 6–8 weeks after HZ onset. At study entry, eventual PHN subjects had significantly more impairment in detecting warmth and cold, a larger area of altered sensation, a larger area of allodynia, and more severe allodynia. The results support the study hypothesis that severity of initial injury predicts PHN, especially impaired cold sensation in HZ skin. The hypothesis that PHN develops because of a failure to recover normal neural function was not supported. Sensory recovery proceeded at the same rate in eventual pain-free and eventual PHN subjects and is not a requirement for pain resolution. Early interventions that reduce neural injury or enhance recovery should be of benefit.     

Heat, but not mechanical hyperalgesia, following adrenergic injections in normal human skin

The development of adrenergic sensitivity in nociceptors has been suggested as a mechanism of neuropathic pain. We sought to determine if nociceptors in the skin of normal subjects exhibit adrenergic sensitivity. We investigated the effects of intradermal administration of norepinephrine, phenylephrine, and brimonidine on heat pain sensitivity. Norepinephrine and phenylephrine (in concentrations ranging from 0.1 to 10 microM by factors of 10), brimonidine (at 0.01-1 microM), and saline were injected (30 microl volume) in a random, double-blind manner to different sites on the volar surface of the forearm in ten subjects. Before and after the injections, heat testing was performed with a non-contact laser thermal stimulator. Heat pain threshold was measured by means of a 'Marstock' technique in which subjects pressed a reaction time key when they perceived that a slowly increasing heat stimulus (1 degrees C/s ramp from a 36 degrees C base) was painful. In addition, the subjects used magnitude estimation techniques to rate the intensity of pain to a suprathreshold heat stimulus (47 degrees C, 2 s). Mechanical testing was done using 200-microm diameter probes attached to calibrated weights that provided forces over the range of 16-512 mN. The intradermal injections of norepinephrine, phenylephrine and brimonidine produced little evoked pain. However, a dose-dependant decrease in heat pain threshold, but not mechanical pain threshold, was observed. At the highest drug dose injected, all three adrenergic compounds produced a significant decrease in heat pain threshold compared to the saline injection. A significant increase in response to the suprathreshold heat stimulus was also found. One possible explanation for this apparent heat hyperalgesia is that the decrease in perfusion due to the localized vasoconstriction may alter the heat response. However, in control studies we found that the non-adrenergic vasoconstrictors, angiotensin II and vasopressin did not produce heat hyperalgesia at doses that produced comparable decreases in blood flow. In addition, occlusion of blood flow with a blood pressure cuff did not lead to heat hyperalgesia. Thus, the heat hyperalgesia observed with the adrenergic agonists is not due to a decrease in perfusion associated with the injection. These results indicate that alpha(1)- and alpha(2)-adrenoceptor-mediated mechanisms may play a role in sensitization of nociceptors to heat stimuli in normal skin.     

Dynamic mechanical allodynia in humans is not mediated by a central presynaptic interaction of A beta-mechanoreceptive and nociceptive C-afferents

Recently, Cervero and Laird (NeuroReport, 7 (1996) 526-528; Pain, 68 (1996) 13-23) proposed a new pathophysiological mechanism of dynamic mechanical allodynia in skin. Using the capsaicin pain model in humans, they showed that light mechanical stimulation within an area of secondary mechanical allodynia induces vasodilatation measured by laser-Doppler flowmetry. They suggested that the low-threshold A beta-mechanoreceptive fibres depolarize the central terminals of nociceptive primary afferent neurons via interneurons. Consequently, the vasodilatation is produced by impulses conducted antidromically in nociceptive C-axons. The allodynia was proposed to result from depolarization of central terminals of primary afferent neurons with C-fibres with activation of nociceptive dorsal horn neurons. In order to extend these findings, we used the same experimental approach but additionally stimulated the A beta-fibres electrically to evoke secondary allodynia during simultaneous monitoring skin blood flow. Twenty microlitres of a 0.5% capsaicin solution was injected intradermally into the dorsal forearm. Skin sites that demonstrated dynamic mechanical allodynia but were not located within the area of primary hyperalgesia and flare were investigated. Ten mm away from a laser-Doppler probe, dynamic mechanical allodynia was induced for 1 min (1) by moving a cotton swab and (2) by electrically stimulating the afferent nerve endings transdermally. Increasing stimulus intensities were applied (0.3-4 mA, 40 Hz, pulse duration 0.2 ms). After intracutaneous injection of capsaicin, light mechanical stimulation elicited a burning painful sensation (numeric analogue scale (NAS) 1.5-3) and concomitant movement artefacts at the laser signal. Antidromic vasodilatation was never observed. In this area of dynamic allodynia, electrical stimulation at stimulus intensities that were not painful before capsaicin injection (A beta-stimulation) was now able to elicit a burning painful sensation (NAS 1.5-3). No change in blood flow was detected. When the stimulus intensities were increased reaching levels that were also painful before capsaicin treatment (C-fibre stimulation), an increase in blood flow could be induced showing the time course of an axon reflex vasodilatation. In conclusion, electrical stimulation of A beta-fibres in allodynic skin does not induce antidromic vasodilatation. Consequently, interaction of A beta-mechanoreceptive fibres and nociceptive C-fibres at a presynaptic level is unlikely to produce antidromically conducted impulses and therefore cannot explain the pathophysiology of mechanical allodynia. Alternatively, it is much more likely that under pathophysiological conditions, activity in A beta-fibres may activate nociceptive second-order neurons, i.e. in the spinal cord.     

Pain and allodynia/hyperalgesia induced by intramuscular injection of serotonin in patients with fibromyalgia and healthy individuals

The aim of this study was to investigate the effect of injection of serotonin (5-HT) into the masseter muscle on pain and allodynia/hyperalgesia. Twelve female patients with fibromyalgia (FM) and 12 age-matched female healthy individuals (HI) participated in the study. The current pain intensity (CPI) and the pressure pain threshold (PPT) of the superficial masseter muscles were assessed bilaterally. 5-HT in one of three randomized concentrations (10(-3), 10(-5), 10(-7) M) or isotonic saline was then injected into either of the two masseter muscles in a double-blind manner. After the injections the CPI and PPT were recorded ten times during 30 min. The injections were repeated twice with the other concentrations of 5-HT after 1 and 2 weeks, respectively. In the FM-group there was a non-significant increase of CPI after injection that lasted during the entire 30-min period irrespective of whether 5-HT or saline was injected. Neither did the PPT change significantly. In the HI-group pain developed significantly after injection irrespective of whether 5-HT or saline was injected, but significantly more so after 5-HT at 10(-3) M than saline injection. CPI decreased quickly and then remained on a very low level for most of the experiment. 5-HT at both 10(-5) M and 10(-3) M caused a significantly greater decrease of PPT than saline. In conclusion, our results show that 5-HT injected into the masseter muscle of healthy female subjects elicits pain and allodynia/hyperalgesia, while no such responses occur in patients with fibromyalgia.     

Effects on muscle pain by intramuscular injection of granisetron in patients with fibromyalgia

We have previously reported that the level of 5-HT in the masseter muscle is increased in patients with fibromyalgia as compared with healthy subjects and that high intramuscular level of 5-HT is associated with muscle pain. We have also reported that injection of the 5-HT(3) receptor antagonist granisetron (GRA) into the masseter muscle of healthy subjects reduced pain induced by 5-HT and abolished allodynia/hyperalgesia. The aim of this study was to investigate whether GRA can influence pain and allodynia/hyperalgesia of the masseter muscle in patients with fibromyalgia. Eighteen female patients who met the criteria of fibromyalgia according to the American College of Rheumatology participated in the study. They were examined regarding pain intensity and pressure pain threshold (PPT) over the masseter muscle. One milliliter of GRA (1 mg/ml) was injected into the masseter muscle on one side and 1 ml of isotonic saline on the other side in a randomized and double-blind manner. After the injections, the pain intensity and PPT were recorded during 30 min. The pain intensity increased after injection of saline and to a lower degree after injection of GRA. The PPT increased after injection of GRA, while no such change was observed after saline. The difference between GRA and saline was, however, not significant. Eight of the patients responded to the GRA injection by an increase of PPT during the experimental period that differed from saline. They also showed a tendency to a lower increase of pain intensity after injection of GRA when compared to saline. In conclusion, the results of this study do not prove that injection of the 5-HT(3)-antagonist GRA into the masseter muscle influences local pain and allodynia/hyperalgesia in patients with fibromyalgia.     

Postherpetic neuralgia: clinical experience with a conservative treatment

Ninety-seven consecutive cases of postherpetic neuralgia (PHN) were retrospectively reviewed. Patients comprised 49 women and 48 men with a mean age of 71.6 years. The most common painful locations were the chest and upper back (34%), abdomen and lower back (25.2%), and face (20.2%). Burning pain was the most common type of pain (61.3%). Lancinating pain was reported by 40% and throbbing pain by 22.6%. Treatments included drugs (mainly tricyclic antidepressant, anticonvulsant, and neuroleptic drugs), transcutaneous electrical nerve stimulation (TENS), and dry needling of muscles in the affected dermatomes. Positive response to treatment occurred in 18.5% of the patients after one visit. In 9.3% of the patients, the pain still could not be controlled after 10 visits of 2-week intervals. TENS proved to be effective in patients whose skin sensation was preserved. It was concluded that in most PHN cases, pain can be effectively controlled by conservative noninvasive therapy.     

MULTIDISCIPLINARY PAIN ABSTRACTS: 16

The present double-blind, placebo-controlled study was designed to test quantitatively the effect and of nerve growth factor (NGF) injected into the masseter muscle. Pressure pain thresholds (PPT) and pressure tolerance thresholds (PTOL) were used as indices of mechanical allodynia and hyperalgesia in the jaw-closing muscles. In addition, perceived pain intensity was assessed by the subjects on a 0 to 10 numerical rating scale (NRS) with the jaw at rest and in relation to various oral functions (chewing, yawning, talking, swallowing, drinking and smiling). Repeated measures analysis of variance (ANOVA) was used to test for significant effects. Injection of NGF into the masseter muscle was associated with significantly reduced PPT for 7 days and PTOL for 1 day. Buffered isotonic saline injections into the masseter muscle also significantly lowered the PPT after 1 day and isotonic saline had no significant effect on PTOL. In contrast, assessment of PPT and PTOL in the noninjected temporalis muscles demonstrated a significant increase after 14 to 28 days, which may have reflected an adaptation to the test procedure. NRS scores of chewing and yawning were significantly increased for 7 days following NGF injection. Systemic adverse effects were noted in one subject who reported fever and slight discomfort about 8 hours after the NGF injection. In conclusion, this is the first study to show that injection of NGF into the human masseter muscle causes local signs of mechanical allodynia and hyperalgesia that persist for at least 7 days as well as pain during strenuous jaw movement. The present pain model is safe and may be used to gain further insight into the neurobiological mechanisms of muscle pain and sensitization.