Reflex sympathetic dystrophy treated with gabapentin

98

Reflex Sympathetic

Dystrophy

Treated With Gabapentin

Gary A. Mellick, DO, Larry B. Mellick, MD ABSTRACT. Mellick GA, Mellick LB. Reflex sympathetic dystrophy treated with gabapentin. Arch Phys Med Rehabil 1997;78:98-105. The use of the recently released anticonvulsant, gabapentin (Neurontin), in the treatment of severe and refractory reflex sympathetic dystrophy (RSD) pain in six patients ranging in age from 42 to 68 years is reported. Satisfactory pain relief obtained in all six patients suggests that this medication is an effective treatment for RSD pain. In addition to pain control, early evidence of disease reversal in these patients is suggested. Patient 6 is the first documented case of successful treatment and cure of the RSD pain syndrome using gabapentin alone. Specifically, reduced hyperpathia, allodynia, hyperalgesia, and early reversal of skin and soft tissue manifestations were noted. Gabapentin was chosenbecause it has properties similar to other anticonvulsant drugs and because previous studies have shown that it is well tolerated and appears to have a benign efficacyto-toxicity ratio. It was considered an acceptable and compassionate therapeutic choice because previous medical and surgical approaches had been ineffective for these patients, who represent the first case seriesdocumenting the use of gabapentin for pain management. Presently, the mechanism of pain relief in these patients is unknown. In this article, the pathophysiology of RSD is discussed, and a mechanism by which gabapentin provides pain relief is proposed. In view of encouraging results in these and other RSD patients, further scientific investigation is needed to delineate the role of gabapentin in the treatment of reflex sympathetic dystrophy. 0 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation EFLEX SYMPATHETIC dystrophy (RSD) and causalgia R are generic terms that describe signs and symptoms that follow injury to bone, soft tissue, and nerve. It is a syndrome characterized by severe burning pain, hyperpathia, allodynia, vasomotor and sudomotor changes, edema, stiffness, and discoloration; if left untreated, it may progress to fixed trophic changes.’ It typically occurs in an extremity and can result from trauma,‘z3 inflammatory disorders, myocardial infarction, cerebral infarction, osteoarthritis, degenerative joint disease, frostbite, burns, drugs (particularly phenobarbital):’ malignancy, and paraneoplastic syndromes.6No known causeis identified in approximately 35% of the cases of RSD. The burning pain, which may begin within minutes or hours after the injury, is often inordinately intense and completely out of proportion to the original injury. Certain aspects of the From American Pain Specialists, Inc. (Dr. G. Mellick), Elyria, OH, and the Department of Emergency Medicine, Loma Linda University Medical Center (Dr. L. Mellick), Loma Linda, CA. Submitted for publication December 5, 1994. Accepted in revised form April 24, 1995. No commercial party having a direct or indirect interest in the subject matter of this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Gary A. Mellick, DO, President, American Pain Specialists, Inc., 1100 N. Abbe Rd., Suites C C D, Elyria, OH 44035. 0 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/97/7801-3338$3.00/O

Arch

Phys

Med Rehabil

Vol78,

January

1997

syndrome had been previously described,7‘9but the breadth of the syndrome was not recognized until 1864 when Mitchell and colleagueslo described their experiences with Civil War wounded. Since most of the burning pain that Mitchell described resulted from nerve trauma, he labeled the condition “causalgia.“” Characteristically, the initial pain may be confined to a peripheral nerve or specific vascular distribution, but if not treated promptly or effectively, the syndrome will grow beyond the original area of injury, sometimes even spreading to involve the opposite extremity. The syndrome is arbitrarily divided into three stages, but since the condition represents a spectrum, few patients exhibit all the signs and symptoms in the same order.” The acute stage begins at the time of injury, lasts for several weeks to months, and is characterized by aching and burning pain restricted to a vascular, nerve, or root territory. Redness, edema, and decreased range of motion may also be seen in this stage. The dystrophic stage begins approximately 3 months after the injury and is characterized by pain extending outside of the original vascular area or dermatome of injury. There may be increasedjoint thickness, tenderness, and stiffness, and the hyperpathia and swelling is more pronounced. Some muscle wasting and osteoporosis may be seen by the dystrophic stage. The atrophic stage represents end-stage RSD. It usually develops after 6 months of RSD. Pain is usually intense but may be less severe than in the previous stages. The skin is customarily cyanotic, pale, and cool. Conspicuous irreversible trophic changes are evident in the skin and subcutaneous tissues. The skin has a smooth, glossy appearance, with loss of the usual skin folds and wrinkles. The joints are typically thick and tender, and joint motion is more restricted. In this stage, muscle wasting and osteoporosis are more evident. Swelling and hyperpathia may be more pronounced. The mechanism of the development of RSD symptoms is not known, but many of them correlate with sympathetic nervous system activity. In fact, improvement after sympathetic blockade is common. For this reason, most RSD management strategies consist of procedures that will block central or peripheral sympathetic activity to decreasepain intensity and reverse some of the tissue changes of reflex sympathetic dystrophy. Gabapentin (Neurontin”), an anticonvulsant introduced in February 1994, has been recognized as an effective, adjunctive therapy with other antiepileptic drugs for partial seizures with or without secondary generalization. It is approved for patients with epilepsy over the age of 12 years.‘2-14Gabapentin is available in loo-, 300-, or 400-mg capsules and can be taken with or without food. The effective dose range is 300 to 600mg three times per day. The usual starting dose is one 300-mg capsule taken on day 1, two 300-mg capsules on day 2, and three 300mg capsules on day 3.r5 Common side effects of gabapentin include somnolence, dizziness, ataxia, and fatigue. Gabapentin, a structural analogue of y-aminobutyric acid (GABA), was synthesized as a GABAmimetic drug that could crossthe blood-brain barrieri however, its pharmacodynamic mechanism is different from other substances that interact with GABA synapses, such as valproic acid, phenobarbital, benzodiazepines, and vigabatrin. Its binding occurs in the outer layers of the neocortex and the hippocampus, but the receptor and its biochemical function remain undiscovered. Gabapentin does not bind to benzodiazepine,

RSD

TREATED

WITH

glutamate, glycine, GABA,, GABA,, and N-methyl aspartate receptors and may have a novel binding site in the nervous system.17 The drug is absorbed orally without interference by food and reaches peak serum concentrations after 2 to 3 hours. Gabapentin is not bound to plasma proteins and is eliminated only by renal excretion in its original form. It is not metabolized by the liver and does not induce hepatic oxidase enzymes. Because of the absence of hepatic metabolism and lack of protein binding, there are almost no interactions between gabapentin and other drugs. Antacids reduce the bioavailability of the drug up to 24%, which may require an appropriate increase in the dose of gabapentin. Although cimetidine induces a slight decrease in renal excretion, it is not expected to be of clinical significance.16

CASE REPORTS Case 1 A 52-year-old woman developed right arm reflex sympathetic dystrophy after she fell down on November 21, 1991. When she grabbed the railing she wrenched her shoulder and right arm. Subsequently, she experienced gradual onset of burning pain and swelling in her right hand, arm, and shoulder. Six months later, after evaluation by 6 different physicians, RSD was diagnosed. In the ensuing months, 5 stellate ganglion blocks were completed, each resulting in an ipsilateral Homer’s syndrome, vasomotor dilatation, and normalization of skin color in the involved extremity. The patient, however, experienced only temporary pain relief. Previous treatments had also included a transcutaneous electrical-nerve stimulation (TENS) unit, physical therapy, peripheral nerve anesthetic blocks, mild to moderate opioid analgesics, nonsteroidal anti-inflammatory drugs (NSAIDs), and occupational therapy that resulted in transient improvement only. When seen in our clinic in January 1993, she complained of burning pain in the right hand and wrist, extending to her shoulder. She also experienced cramping and parestbesias of the right arm and hand associated with numbness and throbbing of the fingers. On a verbal pain scale she rated her pain at 8/10 to lO/lO in severity. The pain was made worse by light touch, rubbing, anxiety, fatigue, strong emotions, raising her arm, or driving her car. Besides RSD, her medical history was significant for multiple medication allergies and a secondarily generalized seizure disorder. Examination revealed a swollen right forearm and hand that was red and exquisitely painful. Her fingers were swollen and finger movement was limited because of flexion deformities. Her skin had a waxy sheen and she withdrew from even the lightest touch by her examiner. The remainder of the examination was significant for findings consistent with rotator cuff tear of the left shoulder. A triple phase bone scan showed symmetric bilateral increased uptake involving the joints of the hands and wrists. The pattern of uptake was consistent with the clinical diagnosis of RSD. RSD developed in her left arm and hand in September 1993 and the patient experienced major depression. She eventually agreed to treatment with gabapentin, and received the initial 300-mg capsule on May 20, 1994, at 5PM. Within 2 hours, she had complete relief of pain. She experienced euphoria, mild disorientation, dizziness, and drowsiness. After her second 300mg gabapentin capsule, she developed a slight headache and a “tight hat band” sensation. She experienced leg cramps and diarrhea after the third dose. (The leg cramps responded to baclofen, but have since resolved.) By day 3 of treatment, the diarrhea was gone and her mood had improved; she awakened with a feeling of increased energy. On day 4, she developed pruritus which was mostly relieved by diphenhydramine HCl.

GABAPENTIN,

99

Mellick

On day 10 of gabapentin therapy, she held her grand nephew for the first time. By day 15, she reported only slight burning in the right upper extremity, increased finger movement, and improved handwriting. Occupational therapy sessions subsequently were less painful, and her therapist (who was unaware of the gabapentin therapy) observed that after a year of treatment, the patient was finally showing improved range of motion. Follow-up examinations showed complete disappearance of allodynia, and hyperesthesia in her right hand. Approximately one month after she began gabapentin therapy, the skin color and texture of her right hand was normal, she had regained use of her middle finger (which previously had been immobilized by the RSD), and she had a strong (5-/5), pain-free grasp. During the first week of therapy, the euphoria and disorientation gradually diminished; nevertheless, the patient’s mood improvement was almost as consequential as her pain relief. She began dressing neatly, wearing make-up and dieting, and she started an exercise program. Family and friends were now able to touch and hug her without inducing pain. Finally, she began a support group for other RSD patients.

Case 2 A 42-year-old woman was referred for evaluation and treatment of RSD of her left arm after she fell and dislocated her elbow on November 14, 1992. The arm was placed in a cast, but she experienced severe pain in her left hand and arm, which was swollen and red when the cast was removed on December 12, 1992. A repeat x-ray showed continued partial elbow dislocation with a small chip fracture. In February 1993, she was told that she had RSD; she was unable to care for her disabled child, or to drive or to curl her hair. Past treatments consisted of ibuprofen 400 mg four times daily, an elbow and hand brace, and a TENS unit. Her medical history was significant for glaucoma and decreased visual acuity. Upon presentation, her left shoulder, arm, and hand pain was described as S-9/10 on a verbal pain scale of severity. The shoulder pain was piercing and burning, and the elbow pain was stabbing in nature. She also described a subjective sensation of cold and burning in her hand and arm. Finger pain was accompanied by tightness and numbness. Rubbing, vibration, light touch, straining, or cold temperatures aggravated the pain. Her left upper extremity was cold to touch and the mottled skin had a bluish hue. The hand and fingers had a white, glossy sheen and there were flexion contractures. Elbow extension was painful and limited to 70”. Her grasp was 3+/5, finger abduction was markedly restricted, and wrist dorsiflexion was 415 but was limited by pain. Allodynia and hyperpathia were present in the left arm and hand. Additionally, it was noted that she had sensory loss to pin prick and temperature in the ring and little finger of the left hand. Initial management of the RSD included stellate ganglion blocks, which provided significant reduction of her pain and temporary reversal of her other symptoms. Peripheral anesthetic nerve blocks gradually reduced her hand pain and physical therapy maintained limited extremity mobility. After appropriate discussion, the patient consented to a trial of gabapentin therapy. After taking only one 300-mg gabapentin capsule in late May 1994, the patient reported that her pain intensity dropped to a tolerable l-2/10 severity on a verbal pain scale. With her second capsule, she noted increased throbbing and enhanced warmth in her left arm and hand. Associated with this warmth was a change in skin color and resolution of the mottled appearance that had been present in her left upper extremity for years. Additionally, she noted a moderate reduction in arm swelling and reported increased finger range of motion

Arch

Phys

Med Rehabil

Vol78,

January

1997

RSD TREATED

100

WITH

and touch sensation. Initial side effects included diarrhea, euphoria, giddiness, dizziness, mild disorientation, and leg cramps. The diarrhea ceased after 24 hours and the euphoria, giddiness, and disorientation resolved after the first week of therapy. Reduced extremity pain was accompanied by an increase in temperature from 34°C to a maximum of 37°C. This patient twice had dramatic worsening of her pain-once when she missed two doses of gabapentin and once when ciproheptadine, a serotonin antagonist that was used to treat her migraine headaches, resulted in a temporary loss of pain control. Case

3

A 57-year-old woman medical transcriptionist developed severe pain in the right foot in July 1991. To understand what was being said on tapes that she transcribed, she rewound them repeatedly, using her right foot. Her pain syndrome was diagnosed as posterior tibial nerve entrapment and pain was relieved only temporarily by a surgical procedure. X-rays of her foot and magnetic resonance imaging (MRI) of the back obtained in 1993 were negative. Following chiropractic therapy in January 1994 (to correct a leg length discrepancy), the patient also developed severe right back, groin, and hip pain. On presentation, she complained of constant, severe pain (S-10/10) in the right foot. The right knee pain was described as “suffocating.” Her back pain was dull and stabbing; the pain increased with leg movement or ambulation. At initial evaluation she was taking no medications, but past treatments had included NSAIDs. Additionally, she had received 4 anesthetic nerve blocks at a major pain center. The staging questionnaire established that she was in the atrophic stage of RSD. Otherwise, she was in excellent health, without any known medical problems. Examination showed evidence of previous third-degree burns of the right anteromedial leg. The skin on the right foot was smooth and her toenails were coarse. She demonstrated allodynia, hyperesthesia, hyperalgesia, and hyperpathia of the right foot. She could not extend her knee completely and palpation of the medial aspect of the knee and the lumbar spine area caused pain. Straight leg raising was negative. The neurological examination was otherwise within normal limits. A triple phase bone scan showed bilateral diffusely increased uptake of the radiotracer in the joints of the left forefoot and hindfoot as well as along the calcaneal surface consistent with RSD. After consent was obtained, the patient began her first course of gabapentin therapy at 300 mg three times a day in May 1994. She reported pain relief in her right leg, but her foot was most improved. Subsequently, she quantified her pain relief as “good.” Other than feeling groggy, the patient experienced few adverse effects from gabapentin therapy. This side effect has completely resolved. Case 4 A 45-year-old woman was referred for treatment of RSD of both lower extremities. She was injured in June 1991 when she fell while climbing stairs. Burning pain began almost immediately and was accompanied by swelling and dark discoloration of her left leg. In the first three months after the injury, she experienced stiffness, limited mobility of the affected area, and severe burning and aching at the injury site. Later, she noted increased sensitivity to stimuli, accelerated nail growth, localized edema, muscle spasms, and vasospasm. As time passed, she noted muscle wasting, edema, and hair loss. On presentation to our clinic she was experiencing intractable pain of the entire left limb. Her treatment included nortriptyline (Pamelorb) 25 mg at bedtime. By February 1994, despite a series

Arch

Phys

Med Rehabil

Vol78,

January

1997

GABAPENTIN,

Mellick

of intravenous regional sympathetic blocks, she had increased sweating and pain in her right foot. Aerobics and water jogging were prescribed to slow the spread of the RSD, but these activities were routinely too painful to perform. She underwent autonomic testing to determine her potential response to lumbar sympathetic blocks, but she declined this therapy because of apprehension about needle injections. Adaptive behaviors initiated by the patient included tactile desensitization of her foot prior to putting on her sock. Additionally, she wore an oversized shoe and sock to reduce pain caused by their contact with her foot. The pain in both knees, left foot, and ankle was described as burning, sharp, shooting, squeezing, and needlelike in quality and ranged from 5/10 to 9/10 in intensity on a verbal pain scale. Additionally, leg cramps routinely occurred throughout the night and she experienced burning back spasms. A triphasic bone scan obtained on October 14, 1991, showed diffuse uptake of the radiotracer in the joints of the left forefoot and hindfoot and along the posterior calcaneal surface. Neurological examination of the left lower extremity was difficult because the patient’s pain limited motor testing of her leg and foot. She exhibited a withdrawal response when the examiner attempted to touch her left leg. Her gait was antalgic; she had decreased range of motion of the left foot and ankle joints. The skin of her left leg showed dystrophic changes. The patient gave her informed consent to pain therapy with gabapentin at a dose of 300 mg three times a day. The following week, she reported a 75% reduction of her pain intensity. Furthermore, she reported that her pain had dropped to 2-3/10 on a verbal pain scale after taking only 3 gabapentin capsules. Adverse effects of fatigue and drowsiness disappeared by the second day of therapy. Additionally, she had spontaneous resolution of back pain after starting gabapentin, but this was shortlived. A repeat bone scan obtained approximately 1 week after implementing gabapentin therapy showed diffuse increased uptake involving the joints of both feet. The patient continued to have excellent pain control while taking gabapentin therapy. Case 5 This patient is a teacher who developed RSD of the right upper extremity after she fell on February 2, 1993. A comminuted Colles’ fracture of her right wrist resulted in such severe pain that she would not allow the physician to touch her hand or wrist. External fixation of her wrist was completed in March with removal of the last of the pins on May 3, 1993. During the first 3 months, she experienced severe burning pain, aching, and stiffness at the injury site. Subsequently, she also noted increasing sensitivity to stimulus, more diffuse pain, accelerated hair and nail growth, localized edema, muscle spasm, and vasospasm in her right upper extremity. She reported pain in the right side of her neck, shoulder, rib cage, arm, forearm, and wrist. The shoulder verbal pain scale was 500 in severity and the wrist pain was rated at lO/lO. She described a steel bandlike sensation extending from her shoulder to her wrist when she moved her arm. Furthermore, she reported aggravation of her pain with vibration, fatigue, sudden movements, and lack of sleep. Additionally, 4 months before her initial evaluation, she had developed pain in the left hand. Physical findings were consistent with the atrophic stage of RSD. Neurological examination showed weakness and limited motion of the interphalangeal joints of her right hand. She had mild atrophy and weakness of the right triceps muscle. Besides allodynia and hyperpathia, the patient demonstrated decreased pin prick and temperature sensation in the thumb and index finger of the left hand. Musculoskeletal examination of the right shoulder and neck showed trigger point tenderness and palpable

RSD TREATED

WITH

tightness. Her right forearm extensor muscles were tender to palpation. The range of motion of the right wrist was restricted and her little finger was fixed in a flexion deformity. X-rays showed right hand and arm osteoporosis. The patient gave informed consent to a trial of gabapentin starting at 300 mg one by mouth three times a day. She took the first capsule at 8PM but did not note any improvement of her symptoms until she awoke spontaneously at 2:30AM; she then noted that she no longer had right wrist, arm, or shoulder pain. Additionally, she noted complete release of the muscle tightness in her right neck, shoulder, arm, and increased warmth in her right hand. The pain and tightness returned later that morning, but it was relieved when she took a second 300mg gabapentin capsule. She did not experience euphoria or disorientation. Subsequent examinations showed resolution of hyperesthesia or allodynia. She has experienced mild pain (l3110 intensity), which was typically provoked by increased physical activity.

Case 6 A 68year-old woman underwent right thumb basal joint arthroplasty for degenerative joint disease July 26, 1994. There was pain in the thumb and tingling in all her fingers the next day. The patient complained in subsequent weeks of pain and stiffness in her fingers with increased pain when she tried to flex her fingers. On August 24, 1994 she complained that her cast was too tight, but a replacement cast did not relieve her pain. Examination revealed a red and swollen hand that restricted extension and flexion of her fingers. On September 2, 1994 her hand surgeon diagnosed RSD. Examination showed marked stiffness of the fingers with flexion contractures of the proximal interphalangeal joints of her fingers and thumb. She was unable to completely open her hand. X-rays showed marked osteopenia which was most pronounced around the metacarpalphalangeal joints. Referred to our clinic for stellate ganglion blockade on September 13, 1994, she rated her pain as S/10-10/ 10 in severity on the verbal pain scale. Her previous treatment consisted of amitriptyline 20mg every night, Medrol Dosepak,’ and physical therapy, but had resulted in little pain relief. Motor examination of the right hand was pain limited (4/5) in all planes of testing. Sensory examination of the hand showed allodynia. Light pressure over the palm and dorsal hand surface produced a rapid withdrawal response. The thumb and index finger showed diminished sensation to pain and temperature. Her reflexes were 2/4 and symmetrical in the upper and lower extremities. The remainder of her neurological examination was noncontributory. The skin of the hand was red and a waxy sheen was most noticeable over the thumb. Medical history was significant for hypertension. The patient was referred for a series of stellate ganglion blocks, but she requested a trial of gabapentin prior to sympathetic blocks and denied our request for a triple phase bone scan. She agreed to stellate ganglion blocks only if gabapentin therapy was not effective in relieving her pain. After informed consent was obtained she began gabapentin therapy at 900mgi d. One week later, she reported little pain relief at this dose of gabapentin; when the dose was gradually increased to 600mg four times a day, she noticed a rapid increase in finger mobility. After 5 days, her pain was reduced to 4-5110 in severity. In week 3 of treatment she had little RSD pain and was once again writing script. She had an estimated 70% to 80% decrease in hand redness and swelling. There was more flexibility in finger movement, but her thumb mobility was only 20% to 30% improved. She had increased strength in all her fingers and was able to grasp without hypersensitivity. In addition to pain relief, the patient reported a significant improvement in her mood.

GABAPENTIN,

101

Mellick

Her finger mobility gradually returned, and by October 1994 she could extend and flex her fingers. The hand allodynia and hypersensitivity was completely gone. In December 1994 her pain was 2/10 in severity and she estimated her condition was 90% improved. Stiffness in the joints was greatly reduced, but pain over the radial nerve persisted. She had normal sensation in her hand, and grasp and thumb strength was 5-15. The hand color had almost completely normalized. Gabapentin therapy was discontinued, and 1 month later the patient described only persisting mild aching in the shoulder and thumb. Her persisting shoulder pain has been controlled with NSAID therapy. When this patient began gabapentin therapy 6 weeks after onset of RSD, she appeared to be in the late acute or early dystrophic stage. Her pain and dystrophic symptoms resolved only after prescription of higher doses (2,40Omg/d) of gabapentin.

DISCUSSION The six patients’ histories are summarized in table 1. Sympathetic pain is a phylogenetically more primitive and hyperpathic pain that follows a different route in central nervous system (CNS) (paleospinothalamic tract, medial pain system) than the more sophisticated somatic pain (neospinothalamic tract, lateral pain system). The paleospinothalamic fibers take the path of the spinoreticular and spinomesencephalic tract. They are multisynaptic, diffuse in their projection, and stimulate the reticular formation at the periaqueductal gray, the hypothalamus, nucleus submedium, and medial intralaminar nuclei of the thalamus. Eventually, these fibers stimulate the limbic forebrain system, causing a feeling of hyperpathia and allodynia, unpleasant pain, and exacerbated and augmented pain. This type of pain involves the paleocortex and is quite diffuse with a strong emotional component, in contrast to neospinothalamic type of pain. It has ipsilateral as well as contralateral projections to the frontal lobe. The ipsilateral projections terminate in the mesial frontal lobe regions with resultant anxiety, emotional alarm (hyperpathia and extreme anxiety), hypertension, insomnia, and secondary depression.‘8 Gabapentin was chosen for pain management because it has qualities similar to other anticonvulsant drugs utilized to reduce pain. Its spectrum of activity most closely resembles carbamazepine and phenytoin in that it depresses segmental and descending excitatory mechanisms while facilitating segmental inhibitory mechanisms. It differs in that it also enhances descending inhibitory mechanisms.” Furthermore, gabapentin, like valproic acid, causes an increase in brain GABA synthesis rates. The GABA-increasing effect of valproic acid is limited to the substantia nigra and corpus striatum, whereas gabapentin causes increased GABA synthesis throughout the brain.20 Long-term studies of gabapentin use with animals and humans have not documented permanent or long-lasting toxicity.*’ Most adverse effects resolve after the dose is decreased or the drug is discontinued. Patient complaints are similar to those heard with use of other anticonvulsant drugs: dizziness, diplopia, ataxia, tremor, somnolence, nausea, and vomiting. Other reported problems include a worsening of preexisting absence seizures, weight gain, and headaches. Gabapentin has no active metabolites, no known antiepileptic drug interactions, and has a 5- to 7-hour elimination half-life.2’ It is believed to possess a unique, as yet unidentified, receptor in brain tissue.17 Recent gabapentin-binding studies have linked this receptor with a site resembling the L-system amino acid transporter protein, since neutral L-amino acids are potently inhibited.17 The functional reciprocal relationship between serotonergic and noradrenergic systems in the CNS is well known.22-26 A clearly defined anatomic relationship exists in that 5HT nerve

Arch

Phys

Med Rehabil

Vol78,

January

1997

102

RSD TREATED

Table

1: Description

Site of RSD

Right greater than left upper extremity Left upper extremity Right lower extremity (and neuropathic Left greater than right lower extremity Right upper extremity Right upper extremity

pain)

of RSD Patients

WITH

and Their

Phys

Med Rehabil

Vol78,

January

1997

Pain Reduction

Mellick

Using

Age

Gender

onset

52 42 57 45 64 68

Female Female Female Female Female Female

1 l/21/91 1 l/41 4192 0719 1 6/13/91 212193 l/27/94

terminals in the locus coeruleus whose cells contain catecholamines (nearly all of which are norepinephrinez7) respond to inhibitory outflow from the serotonin-secreting cells of the raphae nuclei. The projections of the locus coeruleus have a wide distribution throughout the neuraxis, including descending noradrenergic fibers that supply preganglionic sympathetic neurons in the intermediolateral cell column at thoracic and upper lumbar levels.28 Furthermore, sympathetic preganglionic neurons can be inhibited by stimulation of the raphe nuclei.29 Most of the raphe nuclei midline cell groups located in the medulla, pons, and midbrain contain neurons that synthesize serotonin, which is released at the terminals of widely distributed raphe. Basbaum et a13’ identified bilateral spinal projections of the nucleus raphe magnus that descend in the dorsolateral funiculus and terminate mostly in Laminae I and II and other serotonergic fibers that descend in the anterior and lateral funiculi ending in the anterior and lateral gray horns. Furthermore, they found that stimulation of the nucleus raphae magnus produced an analgesic effect by an inhibitory action on sensory neurons.3’ CNS networks that modulate nociceptive transmission are well defined, and brain stem control of nociception in the spinal cord dorsal horn has been thoroughly studied. Pain modulation via the afferent projection from the midbrain to rostra1 ventromedial medulla (RVM) and from the RVM to the dorsal horn has been verified by research. 32,33The RVM contains the nucleus raphe magnus and adjoining ventral reticular formation. The major descending projections of the RVM are to the spinal and trigeminal dorsal horns.34-36 The axons of the RVM neurons project to the spinal dorsal horns via the spinal dorsolateral funiculus to terminate in laminae I, II, V, VI, and VII.34,35,37-39 These same laminae are known to contain the terminals of small-diameter nociceptive primary afferents,40-42 as well as neurons that respond to noxious stimuli and project to the brain stem and thalamus.43344 Several RVM neurons that project to the spinal cord contain 5HT.45-48 Moreover, the RVM is the major source of dorsal horn 5HT.49,50 It is thought that 5HT may participate in the local intercommunications among off-cells because micro-injection of serotonin into the RVM has an antinociceptive effect51.52 and drugs that release 5HT or block its reuptake have similar antinociceptive effects.53 Injection of the serotonin antagonist methysergide into the RVM lowers the pain threshold in rats and increases the pain (tail-flick response).54 This suggests that local 5HT release may excite off-cells, resulting in reduced nociception. Norepinephrine (NE) has nociceptive modulatory neurons and binding sites in the RVM,55*56 and a,- and o12-adrenergic binding sites have been described there.57,58 Iontophoresis of NE into the RVM caused an al-mediated increase in on-cell activity consistent with an enhancement of nociceptive transmission.59*60 Clonidine produces antinociception resulting in long-lasting suppression of on-cell firing, presumably mediated by an action at an c;u2-adrenergic receptor. These experiments indicate a permissive or facilitating effect of on-cells on nociceptive transmission.60

Arch

GABAPENTIN,

Gabapentin

Therapy D0?.e

400mg 300mg 300mg 300mg 300mg 800mg

tid tid qid tid tid tid

Reduction of Pain (Verbal Pain Scale)

95% 85% 60% 90% 90% 100%

Primary afferent nociceptors have their neurons in the sensory dorsal root ganglion. The peripheral nociceptors are subject to modulation at both the central and peripheral nerve terminals and undergo activity-dependent long-term changes. Repeated noxious stimulation increases excitability (sensitization) and a lower threshold for activation with increased and prolonged firing (hyperalgesia).6’ Other authors have described peripheral-central mechanisms to explain sympathetically maintained pain.62-64 We propose that RSD syndrome is the result of a central mechanism stemming from a disparity between the CNS monoaminergic transmitters and their pathways working in conjunction with provocating peripheral events. A gabapentin-induced elevation of CNS 5HT has been demonstrated in both humans65 and animals.66 Furthermore, 5HT plays an important role in the inhibition of pain via the raphespinal descending control system. This system carries signals from the raphe magnus to inhibit nociception at the substantia gelatinosa of the spinal cord. The role of descending control systems in pain modulation was emphasized in Melzack and Wall’s original gate control theory.67 This region contains a high density of projections from the raphe magnus and contains substance P terminals, opiate receptors, and serotonin terminals.68 5HT has a prominent role in this type of analgesia, whereas NE depresses the analgesic effect. Further proof of a 5HT role in pain modulation was provided by Sicuteri et al,‘j9 who found that after administration of pChlorophenylalanine, 4 of 16 patients developed a painful syndrome of hyperalgesia, hyperpathia, and spontaneous pain. PChlorophenylalanine can deplete total brain 5HT to 10% or less of normal levels in the rat by blockade of 5HT biosynthesis at the tryptophan hydroxylation step.70 Sicuteri’s group hypothesized that painful impulses from the periphery are blocked by a physiological concentration of 5HT in the brain. Similar pain responses were identified in rats that demonstrated a reduced morphine induced analgesia when receiving p-chorophenylalanine.70 We believe that the primary mechanism of RSD pain relief is the result of an increase in CNS 5HT levels that modulate monoaminergic pathways. We postulate that the successful control of RSD symptoms results from gabapentin’s capacity to cross the blood-brain barrier where it provides an inhibitory action at its receptor sites. We hypothesize that a gabapentininduced increase in 5HT or a serotonergic-like activity of gabapentin at a novel gabapentin-specific receptor site in the CNS and via the raphe-spinal descending control system inhibits pain sensation and causes an early reversal of at least some of the soft tissue and skin changes characteristic of RSD. A gabapentininduced increase in 5HT bioavailability or gabapentin activity at its receptor sites presumably acts to reciprocally block catecholamine outflow, resulting in a gradual reduction of the noradrenergic-induced hyperalgesia. This could explain the reduced allodynia and hyperalgesia witnessed in our patients. A reversal of depressed affect was seen in some of our patients. This apparent mood reversal happened quickly and occurred with pain remission. We do not know whether the mood

RSD

TREATED

WITH

change was solely the result of chronic pain relief or if there was also a biochemical etiology for their improved mood. A salutary effect on the limbic system induced by gabapentin in conjunction with its RSD pain relieving mechanism may have occurred in these patients. The coexistence of chronic pain and depression is well documented. Some studies7’ have demonstrated that complaints of chronic pain occur in 30% to 100% of patients presenting with depression; conversely, other studies7’ have identified clinical depression in 22% to 78% of patients with chronic pain. According to Sternbach,73,74 the shared mechanism behind chronic pain and the vegetative signs of depression may be a depletion of 5HT and possibly endorphins. Additionally, Sternbach believes that depletion of central 5HT activity causes impaired endogenous serotonergic pain suppressing mechanisms resulting in diminished pain tolerance, depression, and sleep disruption in these patients7’ Tertiary amine tricyclics are believed to increase pain tolerance by raising pain tolerance levels.75 Several studies reviewed by Terenius76 showed that patients with neuropathic or “organic pain” have low endorphin and 5HT levels, but these levels are normal or elevated in patients with psychiatric disorders or “psychogenic pain.” Our series of patients reported improved sleep quality and sleep consolidation with far fewer nocturnal awakenings. Sleep enhancement may be solely a result of pain reduction. Some of our patients who had sleeping problems and restless legs syndrome77,78 have reported similar sleep improvement. Serotonergic raphe neurons are involved in sleep induction. Historically, tryptophan, an essential amino acid and precursor to serotonin, has reduced alertness, shortened sleep latency, and reduced pain perception. Rao et a165 identified both an increase in whole blood 5HT and stage 3 and 4 sleep in healthy young men who were given regular doses of gabapentin. In that study, gabapentin administration did not influence REM sleep. Jouvet’~~~ study showed increased brain 5HT bioavailability and also demonstrated augmented sleep stages 3 and 4 in experimental animals. Whatever the psychological, physiological, or biochemical mechanism, our patients have reported an enhanced sleep quality and sleep consolidation.

CONCLUSION Prior to gabapentin therapy, our patients had been treated for RSD by at least one other specialist in pain management. The patients had experienced years of severe, intractable pain and had undergone multiple treatments that variously included stellate ganglion sympathetic blocks, peripheral nerve anesthetic blocks, and drug therapy. We believe this is the first reported series of RSD patients treated with gabapentin (Neurontin). The effective control of pain and reversal of most RSD symptoms after administration of gabapentin was unanticipated. Patient 6 is the first patient to receive gabapentin therapy with complete and sustained resolution of RSD subsequent to discontinuation of gabapentin therapy. A spontaneous resolution of the RSD in this patient is unlikely because pain control occurred rapidly only after taking higher doses of gabapentin. The patient’s trophic manifestations resolved except for mild tlexion/extension restrictions in her fingers. This suggests that other patients may achieve complete reversal of their RSD by taking gabapentin. Our 6 patients represent the second description of the use of gabapentin for the treatment of RSD.79 Our initial clinical experience with gabapentin has convinced us that it possesses significant pain-relieving potential for RSD patients. We have yet to encounter gabapentin treatment failures in the management of RSD. Some patients may require higher gabapentin doses (2,400 to 3,600mg/24hr). In selected RSD patients, stellate ganglion blocks, lumbar sympathetic blocks,

GABAPENTIN,

103

Mellick

and peripheral nerve blocks are likely to continue to be therapeutic, especially in RSD’s first two stages. Some patients with accompanying neuropathic pain or orthopedic pain syndromes may experience more modest pain relief. These conditions should be differentiated from RSD to provide appropriate treatment. Control of the sympathetically maintained pain followed by treatment directed at accompanying pain conditions will often result in further significant pain reduction. We believe that gabapentin may become the drug of choice for RSD treatment; we recognize, however, the need to subject this new drug therapy to randomized blinded prospective studies.

5.

6.

7. 8.

9. 10 11

12.

13. 14.

15. 16. 17.

18. 19.

20.

21.

22.

23.

References Waldman SD, Waldman KA. Reflex sympathetic dystrophy. Int Med Mag 1990; 11:62-8. Fields HL. Pain. New York: McGraw-Hill, 1988. Schwartzman RJ, McLelan RL. Reflex sympathetic dystrophy: a review. Arch Nemo1 1987;44:555-61. Horton P, Gerster JC. Reflex sympathetic dystrophy syndrome and barbiturates. A study of 25 cases treated with barbiturates compared with 124 cases treated without barbiturates. Clin Rheumatol 1984; 3:493-500. Taylor L, Payne R, Young D, Stem R, Posner J. Phenobarbital rheumatism in primary brain tumor patients. Ann Neurol 1987;22: 117-8. Michaels RM, Sorber JA. Reflex sympathetic dystrophy as a probable paraneoplastic syndrome: case report and literature review. Arthritis Rheum 1984;27:1183-5. Paie Q, Les oeuvres d’Ambroise Pare (histoire de defunct), Roy Charles IX. Tenth Book. Paris: Gabriel Buon. 1598:41:401. Denmark A. An example of symptoms resembling ‘tic douleureus by a wound in the radial nerve. Royal Med Chir Tram 1produced 1813;4:48-52. Hamilton J. On some effects resulting from wounds of nerves. Dublin J Med SC 1838; 13:38-55. Mitchell SW, Morehouse GR, Keen WW. Gunshot wounds and othe injuries of nerves. Philadelphia: JB Lippincott, 1864. Rizzi R, Viscentin M, Maxetti G. Reflex sympathetic dystrophy. In: Benedetti C, Chapman CR, Monica G, editors. Advances in pain research, vol. 7. New York: Raven Press, 1984:451-64. Goa KL, Sorkin EM. Gabapentin: a review of its pharmacological properties and clinical potential in epilepsy. Drugs 1993;46:40927. Andrew J, Chadwick D, Bates D, Cartlidge N, Boddie G, Bole1 P; et al. Gabapentin in partial epilepsy. Lancet 1990;335:1114-7. Ojemann LM, Wilensky AJ, Temkin NR, Chmelir T, Ricker BA, Wallace J. Long-term treatment for partial epilepsy. Epilepsy Res 1992; 13:159-65. Neurontin@ (Gabapentin Capsules). Product monograph. Morris Plains (NJ): Parke-Davis, 1994. Foot M, Wallace J. Gabapentin. Epilepsy Res 1991;3 Suppl:10914. Hill DR, Suman-Chauhan N, Woodruff GN. Localization of [3H]gabapentin to a novel site in rat brain: autoradiographic studies. Eur J Pharmacol 1993;244:303-9. Hooshmand H. Chronic pain: reflex sympathetic dystrophy prevention and management. Boca Raton (FL): CRC Press, 1993. Kondo T, Fromm GH, Schmidt B. Comparison of gabapentin with other antiepileptic and GABAergic drugs. Epilepsy Res 1991;s: 226-3 1. Loscher W, Honeck D, Taylor CR. Gabapentin increases aminooxyacetic acid-induced GABA accumulation in several regions in rat brain. Neurosci Lett 1991; 128:150-4. Bartoszyk GD, Meyerson N, Reimann W, Satzinger G, von Hodenberg A. Gebapentin. In: Meldrum BS, Porter RJ, editors. New anticonvulant drugs. London: John Libby & Co., 1986. Lewis BD, Renaud B, Buda M, Pujol JF. Time-course variation in tyrosine hydroxylase activity in the rat locus coeruleus after electrolytic destruction of the nucleus raphe dorsalis or raphe centralis. Brain Res 1976; 108:339-44. Crespi F, Buda M, McRae-Dequeurce A, Pujol JF. Alteration of

Arch

Phys Med

Rehabil

Vol78,

January

1997

104

24.

25.

26

21.

28

29.

30.

31.

32.

33. 34.

35.

36.

31.

38.

39.

40.

41.

42. 43. 44. 45.

46.

41.

Arch

RSD

TREATED

WITH

tyrosine hydroxylase activity in the locus coeruleus after administration of p-chlorophenylalanine. Brain Res 1980; 191:501-g. Reader TA, Jasper HH. Interaction between monoamines and other transmiters in cerebral cortex. In: Descarries L, Reader TR, Jasper HH, editors. Monoamine innervation of cerebral cortex. New York: Liss, 1984:195-225. McRae-Deeueurce A. Dennis T, Leger L, Scatton B. Regulation of neuronal acyivity in the rat locus coeruleus by serotonergic afferents. Physiol Psycho1 1985;344:158-61. Rappaport A, Sturtz F, Guicheney P. Regulation of central alpha adrenoceptors by serotonergic denervation. Brain Res 1985;344: 158-61. Leger L, Descatries L. Serotonin nerve terminals in the locus coeruleus of adult rat. A radiolautographic study. Brain Res 1978; 145: l-13. Nygren LG, Olson L. A new major projection from the locus coeruleus: the main source of noradrenergic nerve terminals in the ventral and dorsal columns of the spinal cord. Brain Res 1977; 132:85-93. Carpenter MB, Medulla. In: Tracy MT, Vaughn VM, editors. Core text of neuroanatomy. 3rd ed. Baltimore: Williams and Wilkins, 1985:102-32. Basbaum AI, Clanton CH, Fields HL. Three bulbospinal pathways from the rostra1 medulla of the cat: an autoradiographic - _ study of pain modulating systems. J Comp Nemo1 1978; 178:209-24. Basbaum AI. Clanton CH. Fields HL. Ooiate and stimulus-uroduced analgesia: functional anatomy of meduhospinal pathways. Proc Nat1 Acad Sci U S A 1976;73:4685-8. Basbaum AI, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984;7:309-38. Besson JM, Chaoch A. Peripheral and spinal mechanisms of nociception. Physiol Rev 1987;67:67-186. Basbaum AI, Clanton CH, Fields HL. Three bulbospinal pathways from the rostra1 medulla of the cat: an autoradiographic study of pain modulating systems. J Comp Neurol 1978; 178:209-24. Holstege G, Kuypers HGJM. The anatomy of brainstem pathways to the spinal cord in cat. A labeled amino acid tracing study. Prog Brain Res 1982;57:45-75. Mason P, Fields HL. Axonal trajectories and terminations of onand off-cells in the lower brainstem of the cat. J Comp Neurol 1989;288:185-207. Basbaum AI, Ralston HJ, III. Bulbospinal projections in the primate: a light and electron microscopic study of a pain modulating system. J Comp Neurol 1986;250:311-23. Basbaum AI, Fields HL. The origin of descending pathways in the dorsolateral funiculus of the spinal cord of the cat and rat: further studies on the anatomy of pain modulation. J Comp Neurol 1979; 187:513-32. Ruda MA, Allen B, Gobel S. Ultrastructural analysis of medial brainstem afferents to the superficial dorsal horn. Brain Res 1981; 205:175-80. Light AR, Per1 ER. Reexamination of dorsal root projection to the spinal dorsal horn including observations on the differential termination of coarse and fine fibers. J Comp Neurol 1979; 186: 117-32. Light AR, Per1 ER. Spinal termination of functionally identified primary afferent neurons with slowly conducting myelinated fibers. J Comp Neurol 1979; 186:133-50. Cervero F, Iggo A. The substantia gelatinosa of the spinal cord. A critical review. Brain 1980; 103:717-72. Fields HL, Basbaum AI. Brainstem control of spinal pain-transmission neurons. Annu Res Physiol 1978;40:217-48. Dubner R, Bennett GJ. Spinal and trigeminal mechanisms of nociception. Annu Res Neurosci 1983;6:381-418. Bowker RM, Westlund KN, Coulter JD. Origins of serotonergic projectionsy to the spinal cord in rat: an immunocytochemicalretrograde transport study. Brain Res 1981;226:187-99. Bowker RM, Dilts RP. Distribution of mu opioid receptors in the nucleus raphe magnus and nucleus gigantocellularis: a quantitative autoradiographic study. Neurosci Lett 1988;88:247-52. Lovick TA, Robinson JP. Substance P-immunoreactive and serotonin-containing neurones in the ventral brainstem of the cat. Neurosci Lett 1983;36:223-8.

Phys

Med Rehabil

Vol78,

January

1997

GABAPENTIN,

48.

49.

50.

51.

52.

53.

54.

55. 56.

57.

58.

59.

60. 61. 62. 63. 64. 65.

66. 67. 68.

69.

70.

71.

Mellick

Skagerberg G, Bjorklund A. Topographic principles in the spinal projections of serotonergic and non-serotonergic brainstem neurons in the rat. Neuroscience 1985; 15:445-80. Dahlstrom A, Fuxe K. Evidence for the existence of monamine neurons in the central nervous system. II. Experimental demonstration of monamines in the cell bodies of brain stem neurons. Acta Physiol Stand 1964;62 Suppl 232:1-55. Oliveras JL, Bourgoin S, Hery F, Besson JM, Hamon M. The topographical distribution of serotonergic terminals in the spinal cord of the cat: biochemical mapping by the combined use of microdissection and microassay procedures. Brain Res 1977; 138:393-406. Llewelyn MB, Azami J, Roberts MHT. Effects of 5-hydroxytryptamine applied into nucleus raphe mangus on nociceptive thresholds and neuronal firing rate. Brain Res 1983;258:59-68. Llewelyn MB, Azami J, Roberts MHT. The effect of modification of 5-hydroxytryptamine function in nucleus raphe magnus on nociceptive threshold. Brain Res 1984;306:165-70. Inase M, Nakahama H, Otsjik T, Fang J. Analgesic effects of serotonin microinjection into nucleus raphe magnus and nucleus raphe dorsalis evaluated by the monosodium urate (MSU) tonic pain model in the rat. Brain Res 1987;426:205-11. Aimone LD, Gebhart GF. Stimulation-produced spinal inhibition from the midbrain in the rat is mediated by an excitatory amino acid neurotransmitter in the medial medulla. J Neurosci 1986;6: 1803-13. Fuxe K. Evidence for the existence of monoamine neurons in the central nervous system. Acta Physiol Stand Suppl 1965;247:37-84. Takagi H, Yamamoto K, Shiosaka S, Senba E, Takatsuki K, Inagaki S, et al. Morphological study of noradrenaline innervation in the caudal raphe nuclei with special reference to fine structure. J Comp Neurol 1981;203:15-22. Young WS, III, Kuhar MJ. Noradrenergic alpha, and alpha2 receptors: light microscopic and autoradiographic localization. Proc Nat1 Acad Sci U S A 1980;77:1696-700. Unnerstall JR, Kopaftic TA, Kuhar MJ. Distribution of alpha2 agonist binding sites in the rat and human central nervous system: analysis of some functional, anatomic correlates of the pharmacologic effects of clonidine and related adrenergic agents. Brain Res Rev 1984;7:69-101. Heinricher MM, Haws CM, Fields HL. Opposing actions of norepinephrine and clonidine on single pain modulating neurons in rostra1 ventromedial medulla. In: Dubner R, Gebhart GF, Bond MR, editors. Proceedings of the 5th World Congress on Pain. Amsterdam: Elsevier, 1988; 590-4. Fields HL, Heinricher MM, Mason P. Neurotransmitters in nociceptive modulatory circuits. Annu Rev Neurosci 1991; 14:219-45. Levine JD, Fields HL, Basbaum AI. Peptides and primary afferent nociceptor. J Neurosci 1993; 13:2273-86. Melzack R. Phantom limb pain: implications for treatment of pathological pain. Anesthesiology 1971; 35:409. Livingston WK. Pain mechanisms: a physiologic interpretation of causalgia and its related states. New York: Macmillan, 1943. Ochoa J. The newly recognized painful ABC Syndrome! Thermographic aspects. Thermology 1986;2:65. Rao ML, Clarenbach P, Vahlensieck M, Kratzschmar S. Gabapentin augments whole blood serotonin in healthy young men. J Neural Transm 1988;73:129-34. Jouvet M. Biogenic amines and the states of sleep. Science 1969; 163:32-41. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965; 150:971-9. Besson JMR. Supraspinal modulation of the supraspinal transmission of pain. In: Kosterlitz HW, Terenius LY, editors. Pain and society. Weinheim, Germany: Verlag Chemie, 1980:161-82. Sicuteri F, Anselmi B, Del Bianco P. 5-Hydroxytryptamine supersensitivity as a new theory of headache and central pain: a clinical pharmacological approach with p-chlorophenylalanine. Psychopar macologia (Berl) 1973;29:347-56. Tenen SS. The effects of p-chlorophenylalanine, a serotonin depletar, on avoidance, acquisition, pain sensitivity and related behaviour in the rat. Psychopharmacologia (Berl) 1967; 10:204-19. Stembach RA, editor. Psychology of pain. New York: Raven Press, 1986.

RSD

TREATED

WITH

72. Mersky H. Development of a universal language of pain syndromes. Plenary session address, Third World Congress, International Association for the Study of Pain. In: Bonica JJ, et al, editors. Advances in pain research and therapy, vol 5. New York: Raven Press, 1983: 37-52. 73. Sternbach RA. Pain patients: traits and treatment. New York: Academic Press, 1974. 74. Stembach RA. Acute versus chronic pain, In: Wall PD, Melzack R, editors. Textbook of pain. London: Churchill Livingstone, 1984: 173-7. 75. Sternbach RA, Janowsky DS: Huey LY, Segal DS. Effects of altering brain serotonin activity on human chronic pain. In: Bonica JJ, Albe-Fesard DL, editors. Advances in pain research and therapy, vol 1. New York: Raven Press, 1976:601-6. 76. Terenius LY. Biochemical assessment of chronic pain. In: Kosterlitz HW, Terenius LY, editors. Basel: Verlag Chemie, 1980;355-64.

GABAPENTIN,

105

Mellick

77. Mellick G, Mellick L. Successful treatment of restless legs syndrome with gabapentin (Neurontin) [abstract]. Neurology 1995;45: 285-6. 78. Mellick G, Mellick L. Management of restless legs syndrome with gabapentin (Neurontin). Sleep 1996;19:224-6. 79. Mellick GA, Seng M. The use of gabapentin in the treatment of reflex sympathetic dystrophy and a phobic disorder. AJPM 1995; 5:7-9. Suppliers a. Parke-Davis, Division of Warner-Lambert Company, 201 Tabor Road, Morris Plains, NJ 07950. b. Sandoz Pharmaceuticals Corporation, Route 10, East Hanover, NJ 07936. c. The Upjohn Company, 7000 Portage Road, Kalamazoo, MI 49001.

Arch

Phys Med

Rehabil

Vol78,

January

1997