by Isabella Mori
This is a two-part investigation, comprising a) a short review of pharmacologic approaches to RSD as found in the literature and b) a pilot study of medication use among 45 RSD sufferers who subscribe to an RSD-related internet listserv. Among others, the review turned out an endorsement of antidepressants and/or gabapentin, a new anticonvulsant as treatment of first choice, equivocal and generally negative evaluation of opioids, negative evaluation of non-narcotic analgesics (NNAs), and controversial evaluation of systemic steroids. The pilot study found, inter alia, high and successful use of opioids, equivocal information on the effectiveness of gabapentin, the use of muscle relaxants, which were not found in the literature review, a confirmation of the low effectiveness of NNAs, and no use of steroids. Different possible ways of the effect some drugs have on the experience of pain are discussed. This study appears to be the first epidemiological approach to medication use in RSD.
Reflex Sympathetic Dystrophy (RSD) – or Chronic Regional Pain Syndrome Type 1, as it is now referred to by some authors – is typically characterized by severe burning pain, allodynia/hyperaesthesia, sudomotor and vasomotor abnormalities, changes in skin, muscle and bone tissue, as well as mood, sleep and cognitive disturbances. There is also suggestion that RSD is a movement disorder (Dotson, 1993). RSD pain is a type of neuropathic pain; it characteristically occurs in the extremities and follows an injury. However, few actual cases are typical, making diagnosis quite challenging. It is helpful to think of RSD as progressing in stages.
Onset – onset is typically within days or weeks after initial injury; sometimes within months. The injury may be quite innocuous – a fall, a simple fracture. The term “injury” must be used quite broadly here – for example, in some cases it can stem from surgery. Appendix A contains a list of conditions and events that have been seen to be connected to RSD, as precipitating, predisposing for and/or accompanying RSD. The injury can also be so trivial that the patient does not remember it. In some cases, a precipitating incident cannot be established.
Stage 1 (acute) – The patient typically experiences constant burning or aching pain as well as allodynia and hyperpathia. Hyperalgesia and hyperaesthesia may also occur. Some type of inflammation such as localized edema is present. Muscle spasm and tenderness occur almost invariably. Pain is worsened by emotional stress and movement. The skin may start out warm, red and dry; at the end of stage 1, it typically turns cyanotic, cold, and sweaty. This stage may last anywhere from a few weeks to six months. If diagnosis occurs and proper treatment occur in the first six months, the prognosis for full recovery is very good – up to 80% (Hooshmand, 1995; Rogers & Valley, 1994)
Stage 2 (dystrophic) – characterized by continuous burning, aching or throbbing pain. Often, the skin will be cool, appear doughy, and skin colour will be pale grey or cyanotic. Allodynia, hyperalgesia and hyperpathia may be more marked than in stage 1. Additional signs and symptoms: abnormal hair growth, brittle, cracked and heavily grooved nails, thickened joints, and muscle wasting, resulting in limited movement. Osteoporosis may set in. Patients may develop psychological disturbances such as reactive depression. Stage 2 may last 6-12 weeks. (Rogers & Valley, 1994; Kozin, 1992; McNerney, 1991).
Stage 3 (atrophic) – irreversible marked tissue changes are encountered: muscle and diffuse bone atrophy, weakened joints, limited motion or loss of motion; eventually, joints may become ankylosed. Pain, allodynia and hyperpathia can be less severe, although patients may experience spasmodic “breakthrough pain” and, occasionally, proximal spread of pain. The skin is cool, smooth, glossy, pale, or cyanotic. Digits may be thin and pointed. Full recovery at this stage is difficult, although the authors surveyed seem to be in disagreement as to the tractability of RSD in general (e.g., Hareau, 1996, McNerney, 1991). There are anecdotal and indirect indicators that persons with intractable cases of RSD have a higher rate of suicide than the population average (Hooshmand, 1995, Fishbain et al., 1991).
Not all patients will exhibit all the signs and symptoms during the particular stages, and stages often overlap. In addition to the above signs and symptoms, some other techniques are useful diagnostic adjuncts (but rarely establish RSD conclusively). These are, among others, thermography, radiology, bone scintigraphy and/or densitometry, and MRI and PET scans (Kozin, 1992). While some workers in the field (e.g., Kozin, 1992) seem to conceptualize RSD as having very fuzzy boundaries, others have developed relatively clear diagnostic guidelines for RSD.
Examples are the following:
Hooshmand (1995): a) pain, b) muscle spasms and vasomotor abnormalities, c) inflammation and d) insomnia and emotional disturbances. All these must be present for a diagnosis to occur.
Gibbons & Wilson (1992): 1. allodynia or hyperpathia; 2. burning pain; 3. edema; 4. colour or hair growth changes; 5. sweating changes; 6. temperature changes; 7. radiographic changes (demineralization); 8. quantitative measurement of vasomotor/sudomotor disturbance; 9. bone scan consistent with RSD. When a criterion is met unequivocally, 1 point as assigned, 1/2 point if met equivocally. A patient who scores below 3 points is considered not to have RSD, between 3 and 4.5 points possibly to have RSD, and probably to have RSD if 5 points or more are scored.
Stanton-Hicks et al. (1995) refer to RSD as Chronic Regional Pain Syndrome (CRPS) Type 1. This new name resulted from a special workshop held by the International Association for the Study of Pain (IASP) in 1993, designed, among other things, to disentangle the various terms and conditions associated with RSD, such as algoneurodystrophy, Sudeck’s atrophy, causalgia, etc. (see Appendix B for a list).
The criteria for CRPS are as follows: 1) a syndrome that develops after an initiating noxious event; 2) spontaneous pain or allodynia/hyperalgesia occurs, is not limited to the territory of a single peripheral nerve, and is disproportionate to the inciting event; 3) there is or has been evidence of edema, skin blood flow abnormality, or abnormal sudomotor activity in the region of the pain since the inciting event; 4) this diagnosis is excluded by the existence of conditions that would otherwise account for the degree of pain and dysfunction.
Etiology and pathophysiologic mechanisms
The etiology of RSD is as yet unknown. Generally, there is agreement that RSD involves the sympathetic nervous system – but not to what degree or whether all cases of RSD are always sympathetically maintained (Czop et al., 1996, Gragnani, 1994; Stanton et al. 1993, Levine, 1991). In the case of sympathetically maintained RSD, Czop et al. point out that because abnormal receptor-neurotransmitter regulation responds to sympathetic blocks, at least to some degree, RSD could be a receptor disease or aberration.
This seems to go along with Roberts’ (1986) hypothesis of oversensitized wide-dynamic-range neurons in the dorsal horn. Czop et al. suggest that the autonomic dysfunction found in RSD is related to a combination of effects of norepinephrine and other substances. Norepinephrine is a vasoconstrictor controlled by mechanisms that modulate blood flow to digits and produce adrenergic effects which are partially dependent upon microvascular mechanisms. The vasomotor irregularities seen in RSD appear to be, to some degree, a reflection of an imbalance between vasoconstriction (which is influenced by norepinephrine) and vasodilation (resulting from reflex action on local control factors). Levine (1991) suggests that motor inactivity leads to stagnation of lymph, a change in capillary permeability, and localized acidosis.
The outcome is further capillary permeability and more edema. The leaking edema fluids organize into adhesions; this results in fibrosis. The fibrous tissue inhibits vascular supply and adversely affects somatic and sympathetic nerves. This escalates pain intensity and decreases voluntary motion – the vicious circle is complete.
Some authors suggest a significant proportion of psychological involvement in the predisposition for or the development of RSD (e.g., Egle et al, 1990; Dotson, 1993; Gragnani, 1994; McNerney, 1991; Berry, 1996). Some posit such a (partial) etiology especially in the case of paediatric RSD (e.g. Stanton et al. (1993). However, Lynch (1992), in a review of the psychological aspects of RSD in both adult and paediatric literature, concluded that while the psychological aspects of RSD are indeed important and can be quite detrimental, there is no good evidence pointing towards psychopathological or personality factors as predispositions for RSD.
This observation should in no way detract from the importance of the psychological support needed to deal with the psychological consequences of RSD as well as to build the psychological and social stamina, skills and tools that are crucial for the effective management of this condition.
Treatment – General
Very little agreement exists on the course of treatment. The necessity for early diagnosis, for mobilization (through exercise and/or physical therapy) (Hareau, 1996), for psychological support (esp. stress reduction techniques) and generally for the interruption of sympathetic activity (esp. via sympathetic blocks) seem to exhaust the extent of agreement. Otherwise, states a cynical Gragnani (1994), “the choice of treatment is dictated more by the practitioner’s specialty, personal experience and patient complaints than by any sound medical guidelines or repeatedly useful methods”. Some of the authors surveyed suggest that RSD, even in the later stages, could be cured (Mellick & Mellick, 1997, Gallien et al., 1995, Hooshmand, 1995, Stanton et al., 1993, McNerney, 1991).
Particularly controversial are the number of sympathetic blocks and the use of sympathectomy, spinal cord stimulators, systemic steroids, opioid analgesics, splints and other similar devices, the application of ice, and complementary or “alternative” treatment modalities, e.g. acupuncture, osteopathy or hyperbaric oxygen. Other methods of treatment and management (excluding pharmacologic treatment – oral medication, patches and infusion pumps – which will be discussed in more detail below) include electrical stimulation (NMES and/or TENS (Hareau, 1996) and/or dorsal column stimulation (Rogers & Valley, 1994), desensitization (Gragnani (1994), prolonged intravenous regional blockade (Rogers & Valley, 1994), radiation, hypnosis, topical capsaicin and/or counterirritants during stage 1 (Kozin, 1992), physiotherapy, occupational therapy (van Laere & Claessens, 1992), heat (esp. water), ultrasound, massage, mechanical edema control, abstention from caffeine and alcohol to control vasospasm (McNerney, 1991), biofeedback, muscle reeducation, trigger point injections, tendon release, psychotherapy, diathermy (Levine, 1991), and diet (Hooshmand, 1995).
Many workers in the field acknowledge that proper staging and the use of a multidisciplinary therapeutic team (e.g., general practitioner, specialist, anaesthetist, neurologist, internist, physical therapist, psychologist and social worker) are crucial in the successful treatment of RSD. Stanton et al. (1993) suggest that improvement can occur without medical intervention.
Pharmacologic treatment (excluding blocks)
The major aim of this paper is to investigate analgesic drugs which are regularly used by persons who have RSD. The first part of the remainder of this paper is a summary of this drug use as found in the literature; this is followed by a report on a study of analgesic drug use as found in 45 RSD patients. The first part is categorized by what appear to be their most outstanding effects on RSD. Of course, such categorization is somewhat arbitrary and not always clearcut; for example, there is some overlap between the anxiolytics and the muscle relaxants. Nevertheless, I hope that the categories help in giving the reader as orderly an overview as possible.
This type of drug seems to be especially effective for managing sharp, shooting and lancinating pain. It is possible that anticonvulsants stabilize excitable nerve membranes, limit neuronal hyperexcitability, and inhibit transynaptic neuronal impulses in the CNS. Gabapentin (neurontin), a GABA-mimetic, seems currently (since 1994) to be the anticonvulsant used most widely in North America in the treatment of RSD. The fact that this medication is so enthusiastically prescribed may be due to the possibility that gabapentin may reverse RSD. The mechanism of action (e.g. receptor and biochemical action) is not completely known.
Gabapentin binds in the outer layer of the neocortex and the hippocampus and may be effective via a receptor associated with the L-system amino acid carrier protein; its anticonvulsant effects may be mediated by increasing the promoted release of GABA. It has both analgesic and anxiolytic effects. In the formalin test with mice, gabapentin selectively blocks nociception associated with inflammation, suggesting a central site of action, perhaps by blocking the sensitization of dorsal horn neurones that occurs during inflammation (Singh et al. 1996). Gabapentin appears to differ from the other anticonvulsants in its mechanism of action. Gabapentin is not metabolized and can therefore safely be given with other anticonvulsants. It is well tolerated and has few adverse effects (mostly somnolence, fatigue, ataxia and dizziness); these tend to decrease with continued usage (Mellick & Mellick, 1997; Rosenberg et al., 1997; Singh et al. 1996; Czop et al., 1996). Other anticonvulsants used for RSD are phenytoin (dilantin), carbamezapine (tegretol) and valproic acid (depakene).
Hooshmand (1995) strongly recommends antidepressants such as sertraline, paroxetin, trazodone and fluoxetin as the pharmacologic treatment of choice for RSD, not only for their analgesic effects but also for providing rest and sleep. However, he advocates against other antidepressants, such as amitriptyline (Elavil) or imipramine (Tofranil) because of their side effects. Tricyclic antidepressants (TCAs) generally interfere with amine reuptake and block serotonergic, adrenergic, histaminergic and muscarinic receptors. Different TCAs influence these factors in different ways; more than one attempt may have to be made to establish which TCA works for which patient.
Adverse effects include sedation and hypotension. Interactions with other drugs are common. The analgesic dose of TCAs is generally thought to be lower than that for depression. Amitriptyline (elavil) may be the most frequently prescribed TCA. It is not entirely clear whether its analgesic effects are independent of its mood-elevating effects, or whether and how these two might work together. Amitriptyline also appears to be an NMDA antagonist, which seems to be significant in the inhibition of neuropathic pain (Bryson & Wilde, 1996). McNerney (1991) suggests amitriptyline for its mood elevating and anxiety reducing effects. Nortriptyline (Pamelor) and doxepin (Sinequan) are also used for RSD. (Czop et al., 1996; Rogers & Valley, 1994).
The atypical antidepressant trazodone (Desyrel) preferentially blocks serotonin reuptake. Adverse effects may be significant (e.g. tremor, hypertension, bradycardia).
Selective serotonin reuptake inhibitors (SSRIs) are used for RSD; however their effectiveness for chronic pain conditions is not well documented. SSRIs potentiate serotonin neurotransmission by preferential binding to presynaptic serotonin reuptake carriers. Different SSRIs have different degrees of selectivity (Julien, 1995). Adverse effects, which may be quite intense and typically include nausea, nervousness, anorexia and insomnia, tend to decrease after 10-15 days. Examples of SSRIs are fluoxetin (Prozac), paroxetine (Paxil) and sertraline (Zoloft) (Czop et al., 1996).
Some -adrenergic transmitters andof the antihypertensives involve receptors, which figure importantly in the mediation of SMP. A commonly used antihypertensive in RSD is clonidine (Catapres), a selective 2 agonist. Its usefulness may be in the reduction ofpresynaptic allodynia and hyperalgesia. Adverse effects include sedation, dry mouth, dizziness and constipation. Terazosin (Hytrin) falls into the same broad category; it acts through the selective blocking of 1 receptors. Another recommended antihypertensive ispostsynaptic prazocine (Rogers & Valley, 1994).
Phenoxybenzamine (Dibenzyline), a sympatholyticum is also recommended by various authors (Muizelaar et al. 1997; Rogers, 1994; Kozin; 1993). McNerney (1991) recommends it in combination with anxiolytics. Muizelaar et al. report it as second choice treatment after unsuccessful use of nifedipine. This drug has quite severe adverse effects: orthostatic hypotension, dizziness, general malaise, impotence in men, also nausea and diarrhea. These adverse effects may make the drug less useful for ambulatory practice; Muizelaar at el. suggest that since the drug can be so effective it is worthwhile to deal with these problems by very frequent physician-patient contact or, for those with severely debilitating side effects, by hospitalizing the patient during some time of the treatment (8-12 weeks).
McNerney (1991) advocates the use of anxiolytics, esp. diazepam (valium), chlorpromazine, chlorazepate and phenytoin for their psychologic effects; they also aid in muscle relaxation. On the other hand, Hooshmand (1995) contends that benzodiazepines such as diazepam or lorazepam (Ativan) can, in some patients, actually perpetuate chronic pain. There seems to be some evidence that lorazepam effects pain relief via anxiolysis, not via separate analgesic action (Patterson et al., 1997). The beta-blocker popranolol (Inderal) is also suggested (van Laere & Claessens, 1992, Levine, 1991), esp. for the later stages (McNerney, 1991); it may work through vasoconstriction. In chronic TMD/myofascial pain, it was found that clonazepam (Klonopin) provides pain relief similar to that of amitriptyline but has fewer or less intense side effects than the latter and that therefore, medication compliance was much higher for clonazepam (Harkins et al., 1991).
Calcium channel blockers
This type of medication effects vasodilation which appears to occur by decreasing sympathetic tone via blocking adrenergically mediated increase in movement of calcium through the calcium channel. Nifedipine (Procardia) inhibits the entry of calcium into cardiac and smooth-muscle cells. This produces vasodilation, which results in increased peripheral blood flow and in counteracting the effects norepinephrine has on blood vessels. Headaches (usually mild and esp. of the vascular kind and accompanied by photophobia and phonophobia), orthostatic hypotension are adverse effects that might be expected, as well as some dizziness, lightheadedness and nausea. (Muizelaar et al., 1997; Czop et al. 1996)
No mention of the use of muscle relaxants for RSD could be found in the medical/psychological literature. They are, however, mentioned by Durrett (1996) in the drug list published by the Reflex Sympathetic Dystrophy Network, a patient-run organization. In central pain following stroke or spinal cord injury, Taira et al. (1995) found the GABA agonist baclofen (Lioresal) to provide substantial pain relief as well as a diminishing of allodynia and hyperalgesia. The relief of muscle spasm pain (which also occurs sporadically with RSD) appears to be secondary to the decrease in spasticity. Pain relief may involve spinal as well as supraspinal mechanisms, and is not mediated by the endogenous opioid system. Glycine neurons might be involved in the effect of baclofen on allodynia. Examples of other muscle relaxants are cyclopenzaprine, methocarbonol (Robaxin), carisoprodol (Soma) and tizanidine (Zanaflex).
Non-narcotic analgesics (NNA).
These drugs include acetaminophen and the non-steroidal anti-inflammatory drugs (NSAIDs). They inhibit synthesis and release of prostaglandins and thereby block the generation of nociceptive impulses. All of them produce analgesia; the NSAIDs also have an anti-inflammatory effect (Julian, 1995). McNerney (1991) finds that other in stage 1, NSAIDs are of little use. With prolonged use, NSAIDs cause toxicity, including gastrointestinal or CNS problems. For patients with osteoarthritis, Clyman (1996) suggests recurrent switching from one family of NSAIDs to another. He appears to endorse long-term use of acetaminophen (Tylenol) whereas Merskey (1997) does report the same gastrointestinal problems as with NSAIDs. Examples of NSAIDs are ibuprofen (Advil), aspirin, etodolac (Lodine) and nabumetone (Relafen). For more severe pain, acetaminophen or NSAIDs are routinely combined with opioids, e.g. Empracet, a codeine-acetaminophen combination.
Gragnani (1994) recommends avoidance of the use of “addictive pain killers”, stating that patients with sympathetic dysfunctions tend to be at risk for addiction. Arn�r and Meyerson (1988) suggest that clients who do not experience much decrease in pain intensity per se from opioids but rather a decrease in the “suffering” aspect of pain are candidates for addiction. Interestingly, the opioid-induced euphoria that is deemed to be involved in the endorphin-mediated reward system which is said to cause addiction (Julien, 1995) is not reported by any of the writers surveyed as an effect of opioid analgesics.
Rogers (1994) supports the use of opioid analgesics (sustained-release morphine compounds or methadone) prophylactically before physical therapy and as a last resort for those who have not responded to other treatment modalities. Hooshmand (1995) argues for the use of the opioid agonists/antagonists butorphanol (Stadol) and tramadol (Ultram) or the less addicting drugs nalbuphine (Nubain) or talacen (pentazocine with acetaminophen). He maintains that the high doses of opioid agonists found when taken orally are harmful to RSD patients because of their addicting qualities and because they inhibit endogenous endorphin production; however, when given in minimal doses in drip irrigation pumps, agonists such as morphine supposedly circumvent these problems.
In contrast, Becker et al. (1995) report the successful use of high doses (up to 109 mg/day) of intrathecal morphine in 2 patients with intractable RSD. These were given during exacerbations; as soon as these had passed, the pump was returned to much lower doses (54 mg/day in one case, 7 mg/day and decreasing in the other). This type of adjustment raises questions about some common assumptions about opioid tolerance and addiction.
The morphine seemed not only to have lessened the pain but also to have brought under control the swelling, the allodynia and abnormally low temperature. Possibly, successful management of RSD may occur because of the ability of opioid alkaloids to modulate nociceptive transmission at the level of the primary afferent synapse between the nociceptive afferent fibre and the dorsal horn projection neuron.
Some workers in the field (e.g., Arnor & Meyerson, 1988) hypothesize that neuropathic pain is intrinsically unresponsive to opioids. However, Portenoy et al. (1990) point out that these hypotheses are flawed because none of them took into account the fact that drug response needs to be looked at from many different perspectives, including patient-related factors (predisposition to side effects, psychological distress or suffering, prior opioid exposure, history of psychological dependence and possibly genetic factors), pain-related factors (temporal characteristics, rapid tempo of nociception and pain mechanisms) and drug-selective effects.
Thus, while it is possible that statistically, responsiveness to opioids in neuropathic pain states is relatively low, a favourable response in any individual case should not be precluded (Portenoy, 1996). Portenoy also details that adverse factors associated with opioid use such as increased tolerance, addiction, rebound and competition with endogenous endorphins can certainly occur but should never be assumed to necessarily develop; that the concepts of physical dependence and addiction must be separated, with the former being quite acceptable under appropriate circumstances (just as insulin dependence is acceptable for diabetics); and that opioid therapy should only be considered once other therapeutic avenues have failed.
Systemtic Corticosteroids (SC)
Czop et al. (1996) state that success rates of up to 82% have been reported with the use of SC in RSD patients. The IASP classification of chronic pain mentions success with high doses of SC in the early phase of RSD (IASP, 1986). Possibly, SC work by inhibiting spontaneous neural discharge (Kozin, 1993). It has not been established whether SC treat symptoms only or underlying causes (McNerney, 1991). Apparently, SC are used more widely in the U.S.A. than in Europe. According to van Laere & Claessens (1992), their effectiveness has not been proven, and their most serious disadvantage is a tendency to provoke osteoporosis. McNerney (1991) advocates against the use of SC past 3 or 4 weeks. Kozin (1993) recommends prednisone or equivalent.
In addition to the above, the following have been proposed as useful for RSD pain: biphosphonates (van Laere & Claessens, 1992); intravenous guanethidine, a false neurotransmitter (used mostly in Europe) which displaces norepinephrine from storage sites and shows good results (Czop et al., 1996); reserpine (not available in U.S.A.) which decreases catecholamine reuptake and also shows good results (Czop et al., 1996; Kozin, 1992); bretylium tosylate, which blocks norepinephrine release but for which more studies are needed (Czop et al., 1996; Kozin, 1992); and calcitonin, which induces vasodilation in skin and vasoconstriction in the bone (van Laere & Claessens, 1992) – McNerney (1991) recommends it for later stages. It is also of interest to mention that Stanton et al. (1993) report that among his patients, pharmacological approaches (e.g. lorazepam, carbamezapine or amitriptyline) were taken but showed only subjective improvement which was not substantiated by independent observation or functional assessment.
Analgesic drug use by 45 clients who have RSD.
What follows is a report on an exploratory study of analgesic drug use by 45 clients who suffer from RSD.
A total of 45 persons suffering from RSD participated in this study: 37 female, 6 male, 2 did not disclose their gender. The preponderance of females over males seems to be typical of RSD. Mean age was 45 (range: 28-76). Mean duration of RSD was 54 months (range: 1.5 – 262); 38 (84%) had RSD longer than 6 months and could thus considered to be in stage 3. The majority (38) had RSD in multiple locations; 4 had it in a single location, 3 did not disclose the location. The most frequent locations were: leg: 23; arm: 21; shoulder: 17; foot: 15; hand: 14. Participants responded to 3 consecutive requests for participation posted in January and February of 1998 at an internet listserv that serves as support for persons who suffer from RSD. A listserv is an internet mailing list to which subscribers can post e-mail and through which they can receive e-mail posted by other subscribers. The listserv has approximately 650 subscribers (the majority – appr. 90% – in the U.S.A.); however, the majority of them are not active participants.
Participation consisted in completing a questionnaire; the questionnaire was prefaced by an explanation of the nature and purpose of the study. Twice, in conspicuous places in the questionnaire, participants were urged to send the completed questionnaire by private e-mail to me; most followed this instruction but a few participants sent their answers to the listserv. When answers seemed unclear, an attempt was made to contact the participant for clarification. A copy of the questionnaire is at Appendix C. It should be noted that participants gave general information (“general information”) on all of the medication taken as well as more detailed information (“detailed information”) on one or more medication. Because some clients provided more than one detailed information, the total of general and detailed information is 58. The majority of the important answers to the questionnaire were in open-ended format; therefore, analysis was done using IZE, a text-based database software package. The results section will report both quantitative and qualitative data. Because of the exploratory nature of the study and the relatively low number of participants, quantitative analysis was restricted to simple descriptive measures.
Medication consumption – general information.
The mean number of medications taken was 4.1 (mode=2; range=1-12). In total, I counted 90 different medications taken. 67 of them would appear to have a direct effect on RSD. However, some of them, especially the antihypertensives, may have been prescribed primarily for other purposes. Appendix D shows a list of all the medications taken. Except for opioids, Catapres and hormone replacement, all drugs were taken orally. 4 participants have opioid infusion pumps and 5 participants used patches (4 of them containing catapres). Table 1 shows types of medication, broken down by number of clients who take it, and providing the name of the most frequently used drug. This was prepared using the general information. Since most clients took more than one drug, the percentages add up to more than 100%. Some participants also took more than one medication of a specific category, e.g., amitriptyline and doxepin, 2 antidepressants); this is not reflected here.
The most frequently taken type of drug were anticonvulsants, unless one were to put both the opioids and the opioid/NNA combinations into one category, in which case, opioids would be the most frequently used category. However, this would raise the question of how to categorize NNAs, since they occur in two different categories as well. This would create more confusion than necessary; I have therefore opted to put the opioid/NNA combinations in a category all by themselves.
The picture in Table 1 is different from what the above literature recommends. Muscle relaxants, which were not at all mentioned, were taken by 27% of the participants. Opioids and opioid/NNA combinations, despite their controversiality, were used by 73%. Only one participants took a calcium channel blocker, although this type of drug seems to be a promising approach. Conversely, 22% take NNAs even though the literature stresses their limited effectiveness. No-one took steroids.
Table 1 – Pain relief (For a complete version of this paper including tables please contact me )
On average, pain relief was between 25% and 50%; this figure was calculated using the detailed information.
Table 2 shows a breakdown according to types of medication.
Table 2 – Pain relief by medication
(For a complete version of this paper including tables please contact me )
Of course, these average numbers and percentages need to be interpreted with caution. Since the e-mail format provided clients with ample room to make comments (and comments were encouraged), some clients remarked on what the percentage numbers meant to them. For example, explanatory comments on 50% pain relief or more were “yes, to a point”, “some”, “sometimes – depends on the pain level”, “yes, indeed”. The highest pain relief seemed to come with the opioid/NNA combinations (directly followed by opioids only); not surprisingly, the effectiveness of NNAs only is far below the average and provided the lowest pain relief overall.
The mean number of adverse effects per medication was 2.5 (range=0-12, mode=1). Antihypertensives were reported to produce the most side effects (mean=4.5), followed by anticonvulsants (3.6), antidepressants (3.1), opioids only (2.7), opioid/NNAs (2.4), muscle relaxants (1.3) and NNAs (0.2). The most frequently mentioned adverse effects were constipation (experienced by 15 clients), followed by somnolence, fatigue and tiredness (10), memory problems (10), drowsiness (9), nausea (8), dizziness (8), pain and other dysaesthesias (8), dry mouth (7), weight gain or loss (7), gastrointestinal upset (6), blurred vision (6) and problems with the urinary tract (5).
The following quotes from participants also include other disturbances and provide good examples of the qualitative aspects of adverse effects:
“I’m kind of like a zombie” (gabapentin)
“painful swelling of my knees which grossly limits my mobility” (gabapentin)
“one time, it made my peripheral vision on the right go haywire. It got all wavy and out of focus, and it moved over like it was pushing into my forward line of vision” (gabapentin)
“weight gain without noticeable appetite gain” (amitriptyline)
“hypomanic behaviour” (amitriptyline)
“most frequent, most resented side effect is drowsiness” (amitriptyline)
“kept me awake for most of the night” (nortriptyline)
“horrible terrible unbearable mademecry constipation” (oxycodone)
“I feel like I am living behind my body” (Percocet)
“enlarged my colon” (methadone)
“even if I swallow it very quickly, the unpleasant taste comes back. It also affects the sinuses” (carisoprodol)
The comments regarding nortriptyline, when compared to the immediately preceding comment on amitriptyline (a very similar TCA), emphasize that sometimes, adverse effects can go in different directions (e.g., drowsiness and insomnia). A similar pattern was found with weight gain. While these adverse effects were attributed to certain types of medication, they could have been the result of taking other, additional medication, of drug interaction, of RSD itself, and/or of other factors.
Functional gain and loss.
Participants reported, for example, that the medication they take enabled them to “tolerate” RSD, or to “live a normal life”, or to work. Other functional gains were being able to move, to relax, and/or to lead a somewhat normal social life. One important functional gain was sleep – something that except for Mellick & Mellick (1997) and Hooshmand (1995), was not brought up by the authors surveyed. However, the issue turned up repeatedly in participants’ comments.
Following are some typical statements:
“I can sleep through the night … if not, I am awake due to pain” (amitriptyline)
“The primary benefit is sleep, pain relief is secondary, but pain is reduced with real sleep, so it helps all around” (amitriptyline)
“I can get some sleep” (amitriptyline)
“I can sleep for longer periods at night” (paroxetine)
“I think that the medicine assures that I have an excellent night’s sleep … When I finally do wake up in the morning, I feel very well rested and energetic” (doxepin)
“I also get better sleep so I can cope better” (methadone)
Other drugs credited with sleep improvement were Darvocet, guanethidine and cyclobenzaprine. The above comments show some of the different aspects of the interaction between sleep, medication and RSD, as well as the continuum of sleep improvement (e.g. from “some sleep” to “longer periods of sleep” to “excellent sleep”).
The most frequent functional loss (11 out of 28 who reported such loss) as a result of medication was being unable to drive. Another relatively frequently mentioned functional loss was mental hypofunction (e.g. difficulty thinking). All in all, relatively little was reported in terms of functional loss, perhaps partly because this overlaps with adverse effects.
Other pain relief modalities.
Relaxation techniques were the most frequently (17) mentioned types of additional relief modalities. This was followed by warmth (15) (esp. warm water, e.g. whirl pool use), the use of non-physician health professionals (e.g. acupuncture), distraction (10), exercise (9), “attitude” (5), and other assorted techniques. 6 participants reported receiving blocks.
High pain relief.
In exactly 50% (n=29) of all the detailed information, clients reported experiencing 50% or more pain relief. All 4 participants who use opioid infusion pumps are in this group. Table 3 shows distribution by type of medication, the number (percentage) of participants who ascribed pain relief directly to this type of medication, and number (percentage) of participants who take this medication.
That is, a participant may have chosen to provide detailed information on gabapentin and ascribed a high level of pain relief to it; however, it is also important to consider that medications interact, and that pain relief may not stem entirely from the medication it has been ascribed to. Opioids were the type of drug taken by the highest number of participants in this subgroup. The low percentage for NNAs was to be expected. The low percentage for antidepressants is notable; the question could be raised whether they simply do not provide adequate symptom relief for these participants or whether are efficient in other areas, such as sleep improvement.
Table 3 – Drug profile of clients experiencing more than 50% pain relief. (For a complete version of this paper including tables please contact me )
High pain relief and high function.
In 14 cases, participants reported not only high pain relief resulting from a particular medication but also that this medication enabled them to be highly functioning. Function was assessed with the question, “what does this medication enable you to do?” Persons with chronic pain generally seem to have a desire to live a normal life (something their condition has often required them to learn to treasure), and society’s aim is to have independent (i.e. working) citizens. Thus, those who experience both high pain relief and high function would be those who would appear to meet expectations both by the individual and by society.
I have taken all the answers to the above which appear to point to “normal life” and “independence” and interpreted them as high function. In contrast, low function would be an answer such as “I don’t cry all day anymore.”
Examples of high function can be found in the following quotes:
“It gives me a couple of hours of some pain relief to do my normal everyday tasks such as housekeeping chores”
“… enjoy life more (when pump is titrated well – it may take one year!) Go out with friends, not be a hermit … [can] … drive a van (I’m in a power chair) and do crafts”
“… live my life as if everything was normal. Can work, can listen to music, visit with or be visited by friends”
I consider the last quote to be one of the most positive among this subgroup of participants. Nevertheless, for a person who enjoys normal health, the situation described would still be well below optimal.
Table 4 gives detailed information on the 14 participants in question. Notes: a) column “*” denotes to which medication relief in column “% relief” and adverse effects have been attributed; b) # of Drugs refers to number of medications taken. Table 5 is a comparison of drug use between the 14 participants in question and the remainder, using general information.
Table 4 – Details of high relief/high function participants (For a complete version of this paper including tables please contact me )
These 14 participants had multiple locations of RSD, except for two (one in left foot, and one with no answer on that item). Mean age was 45 (the same as the remaining participants), mean duration of RSD was 57 months (range=5-120, sd=36.7) (remaining participants: mean=55, range=1.5-262, sd=75.1). The mean number of adverse effects for the 14 participants was 2.4 (mode=2) (remainder: 2.7, mode=1). On average, the 14 participants used 5.6 medications (remainder: 3.5). 12 of the 14 participants had RSD in multiple locations, 1 in the left foot, and 1 did not specify. 3 of the 4 participants who used opioid infusion pumps are in the high relief/high function group. Considering the question of whether non-pharmacological pain relief approaches may have been involved in the positive results found among these participants, I found that the only approaches which were used more frequently among this group were the use of warmth (e.g. immersion of the affected limb in warm water) and “attitude” (example: “determination not to give in to this disease”).
The other additional approaches were either used in the same proportion as by the remainder (e.g., exercise and the use of other health care professionals) or less than the remainder (e.g., relaxation and distraction). Only one person reported receiving blocks.
Table 5 shows a comparison between drug use by these participants and the remaining participants. Note that the last column does not add up to 100% because other medications not taken by the 14 participants were not accounted for in this table.
Table 5 – Drug use comparison between high relief/high function group and the remainder of participants.
(For a complete version of this paper including tables please contact me )
As can be seen, the biggest differences were in the use of opioids – the 14 subgroup participants took considerably more of these than the remainder. The situation was similar for muscle relaxants. Many more of them used antidepressants than the remainder; on the other hand, no-one attributed high pain relief to this drug category. Conversely, the 14 participants take considerably less anticonvulsants than the remainder and even fewer attributed high pain relief to this type of drug.
Gabapentin was the drug most commonly used (by 23 or 51% of the participants) and also the one drug on which participants the most gave detailed information. Because of this and because gabapentin seems to be currently the drug of choice for RSD, it is useful to investigate more closely the responses of those who take gabapentin (see Table 6).
Table 6 – Detailed information on gabapentin use (n=11) 1 = no relief; 2 = 25% or less relief; 3 = 25% or more relief; 4 = 50% or more relief; 5 = 75% or more relief(For a complete version of this paper including tables please contact me )
9 of these participants were female, 2 were men. Mean age was 40, which is younger as the remainder (47). They also had RSD for a shorter period of time (40 months, compared to 59 months). The latter two differences may be due to the fact that gabapentin has only been available since 1994. Mean pain relief for this group was 3.3, the same mean as that for the remaining participants. Mean number of side effects is 3.7, which compares unfavourably to the remainder of the participants (2.2).
However, this comparison must be made cautiously, as another 12 participants among the remainder also use gabapentin. Dizziness, blurred vision and memory problems are the most frequent adverse effects. The dosages lay more or less within the parameters advocated by Mellick & Mellick (900-1800 mg/day) and Czop et al. (900-2400 mg/day). The effect of the medication was reported to set in between 1-2 hours; the effect was reported to last for 4-6.5 hours. For 2 participants, pain relief from gabapentin was reported to be constant.
Co-existing medical conditions
The interaction between co-existing conditions, RSD and medication use was not investigated at this point. Except for sinus problems, which were reported twice, no one single condition was reported more than once.
The most consistent and conspicuous finding in this exploratory study was the high and successful use of opioids. Overall, the highest pain relief was provided by opioid/NNA combinations, closely followed by opioids alone. Considering the subgroups of participants who experienced 50% or more pain relief as well as the subgroup of those who in addition also reported high functional ability, one can see that the drug category most used, and to which high pain relief was most often attributed, were also opioids. 3 of the 4 patients who used opioid infusion pumps were among the 14 high relief/high function clients; perhaps this treatment modality should be considered more often. These findings are somewhat surprising in the light of little support for the use of opioids in the literature, especially regarding the notion that opioids are of little benefit for neuropathic pain. However, Portenoy et al. (1996) point out that neuropathic pain is not a homogenous entity. It is possible that RSD belongs to a subgroup of neuropathic pain that responds well to opioids.
Alternatively, if RSD itself is not homogenous either (which is most likely the case), it is possible that certain manifestations of RSD respond better than others to opioids, and that those who responded well to opioids in the present sample constitute a sample of such a subgroup. The high use of opioids also raises the question of whether the reluctance – for whatever reason – to prescribe this type of drug on the part of physicians is really as widespread as generally thought. The question of tolerance or addiction was not addressed in this study. The above findings indicate a great need for a more in-depth and, if possible, controlled and prospective study of opioid use in RSD, at which point these issues could be addressed.
The effectiveness of anticonvulsants, and specifically gabapentin (Neurontin), as a pain reliever was somewhat uncertain. Anticonvulsants generally, and gabapentin specifically, provided as much pain relief as the average of all medications surveyed. The high relief/high function group used gabapentin considerably less than the remainder. Adverse effects, on average, appeared considerably higher with gabapentin than with the remainder of the drugs. However, since gabapentin was by far the most prescribed medication, the question is: what special benefits is this medication is expected to have?
It does appear that gabapentin “takes the edge off” the burning pain – it is possible that even though the intensity of the pain does not decrease very much, gabapentin favourably changes pain quality. Furthermore, Mellick & Mellick (1997) tentatively indicate that gabapentin might have curative powers.
Nevertheless, the findings in this study suggest that gabapentin may not always be as useful for RSD as apparently thought by some of the prescribing physicians. The effectiveness of antidepressants is somewhat puzzling. These drugs were among the very frequently prescribed drugs, and they were used the most by the high relief/high function group, which would give some indication of their effectiveness. However, only 1 of the 7 participants who gave detailed information on an antidepressant reported it to provide more than 50% pain relief. A possible interpretation of this picture is that, while these drugs may not directly provide high pain relief, they may provide alleviation of symptoms indirectly, perhaps by providing better sleep, mood elevation or the reduction of anxiety.
However, this interpretation would call into question the generally accepted idea that antidepressants help in chronic pain situations specifically and directly via analgesia.
Equally puzzling is the matter of muscle relaxants. They are not discussed in the literature on RSD; however, in this study, they were prescribed to over 1/4 of the participants. In the high relief/high function group, they were taken by proportionately more than twice as many participants than the remainder. However, just as with antidepressants, this study did not provide any evidence for particularly outstanding strength in terms of pain relief.
The finding that was most consistent with the literature was the low effectiveness of non-narcotic analgesic drugs. The pain relief provided by them was less than half that of most other drugs. A positive aspect of these drugs was that the reported adverse effects were also extremely low. Most likely, however, these do not include the long-term toxicity often found with NNAs. One explanation for the continued use of this type of medication could simply be the fact that it is the pain medication that typically comes to mind and that it is easily available.
Three observations made of the high relief/high function group deserve repeating. One is that this group, on average, uses more drugs than the remainder. Could this mean that their medication is successfully addressing all the different aspects of RSD? The second is that they experience a lower number of adverse effects. Is this a result of the type of medication they are using, of personal dispositions that make them less prone to adverse effects, or of other factors? Third, this was the group which had by far the highest number of persons report that “attitude” helped them in managing RSD. It is quite possible that this contributed to a large degree to their high functioning; it would be interesting to investigate whether and how a positive attitude actually potentiates medication use, or whether it has a rather independent effect on managing RSD.
One may, of course, ask whether high pain relief in connection with any specific medication is indeed as important a measure as I seem to imply. Obviously, the most important goal is to provide high pain relief for the client as a whole; but when, for example, high pain relief is already provided by an opioid, the role of other medications may be to deal with additional pain relief (which would necessarily be lower), with the relief of pain of a different quality, or with other symptoms such as excessive sweating, mood disturbances, etc. It is possible that antidepressants and muscle relaxants fulfil those adjuvant roles quite well. Anticonvulsants and antihypertensives may fulfil similar roles; however, in this study, they show a higher average number of adverse effects.
This study has a number of limitations. Some are due to the fact that this was a pilot study, some are due to the small number of participants, some are due to the design in general, and some are due to the difficulties inherent in investigating the subject of medication use in general and from an epidemiological perspective specifically. Examples of the first type are that participants were not given a specific rationale for giving detailed information on a specific drug (e.g., they could have been asked to give information on the drug that gave them the most pain relief; this would have removed most of the problems regarding attribution of pain relief as found above); they were not asked how long they had taken the medication; they were not asked about their use of alcohol, tobacco and recreational drugs; and the inclusion of a scale regarding functional gain would have been helpful. The small number of participants, together with the fact that this was a pilot study, makes any generalization very tentative. This is the reason why I provided only descriptive statistics. Even the use of percentages might sometimes be misleading, as in those situations where a certain percentage corresponded to only one or to participants.
I should emphasize that I used percentages only for comparison purposes, and not to make any quick generalizations. The design was a mixed qualitative/quantitative design. This is not the place to discuss the relative merits of qualitative versus quantitative approaches. Suffice it to say that perhaps the most significant problem with this is that in psychology, there are no well-established methodologies for this approach and that it is therefore quite possible that my analysis, had I had access to better established methodology (including appropriate software), might have been more sophisticated. The difficulties generally acknowledged to be associated with self-report studies apply here as well; however, it should be remembered that ultimately, pain is a subjective experience. Seen from that perspective, self-report is the most reliable source of information.
The difficulties inherent in studying medication use include interaction with other medication, diet, psychological state and many other factors concerning bioavailability, pharmacodynamics, pharmacokinetics, etc.
This study appears to be the first epidemiological approach to medication use in RSD; in fact, there are very few studies of this sort on chronic pain in general. The mostly open-ended format made it possible for some information to surface which might otherwise have been neglected – examples are the use of muscle relaxants (which were not discussed at all in the literature) and the use of NNAs (which, because of their acknowledged low effectiveness, might otherwise not be studied at all). Furthermore, it appears that other empirical studies on RSD medication (and again, these are also not many) have been done with clients in pain and rehabilitation clinics. It has often been mentioned that clients using these services constitute a very small and specific subgroup and are most likely not very representative of chronic pain patients in general (Gamsa, 1994). Keeping in mind the caveat regarding generalizations, above, we could still ask how representative the present sample is. At the very least, that is, if this sample is just as unrepresentative as pain clinic patients, it nevertheless provides a different perspective.
Finally, a question that needs to be asked is how much benefit in general is provided by pharmacologic approaches such as seen in this study. For example, even though many workers in the field praise the value of exercise, relatively few of the persons studied engaged in it. Would their situation improve with more exercise? Furthermore, none of the persons surveyed mentioned the use of psychological support, even though it is clear that RSD is exacerbated by the emotional and cognitive changes it brings with it. Lastly, possibly some more intense short-term pharmacologic approaches, such as done by Muizelaar et al. (1997) might be useful.
Conditions that have been observed as precipitating or proposed as predisposing for RSD (Gallien et al (1995), Rogers et al. (1994), Stanton et al. (1993), Kozin (1992). Note that according to the CRPS Type 1 definition, many of these predecessors (e.g. diabetes) would preclude a diagnosis of CRPS.
herniated disc disease
infection (e.h. poliomyelitis)
medications (isoniazid, barbiturates)
nerve root avulsion
radiculopathy (e.g., cervical, lumbar)
repetitive strain injury
spinal cord injury
thoracic outlet syndrome
Alternative names for RSD and RSD related conditions
acute bone atrophy
chronic regional pain syndrome
chronic traumatic edema
peripheral acute trophoneuritis
posttraumatic pain syndrome
posttraumatic painful arthrosis
posttraumatic painful osteoporosis
posttraumatic sympathetic dystrophy
major traumatic dystrophy
minor traumatic dystrophy
reflex neurovascular atrophy
reflex neurovascular dystrophy
reflex sympathetic dystrophy
reflex sympathetic dystrophy syndrome
My name is Isabella Mori. I study psychology at Simon Fraser University in Vancouver, B.C., Canada. I have been interested in chronic pain for a number of years, and have been involved with the Canadian Reflex Sympathetic Dystrophy Network since 1996. Currently I am interested in clients’ experience with and reaction to the medications they take for RSD pain, and I was hoping I could get your co-operation. I would like to know what medication(s) you take, whether and how much it reduces your pain, what the side effects are (that includes physical side effects – e.g., blurred vision, and other side effects – e.g., decreased memory). I would also like to know what taking the medication enables you to do (e.g. “these meds enable me to go to work”), or what it prevents you from doing – say, driving a car.
Finally, I am interested in what else you do to relieve your pain (e.g. relaxation exercises), and in any additional comments you may have. You will note the questions are open-ended; I am not only interested in raw information but also your overall experience – similar, perhaps, to the way you tell it here on this list.
If you wish to share your experience, please provide your answers in the space provided below. I will hold all the information you decide to give IN COMPLETE ANONYMITY. The report on this study will be sent to this list some time in April or May of this year. If you have any questions, please email me. If you wish to learn more about some of the work I have done in connection with chronic pain, please visit my web site at […]
** IMPORTANT ** IMPORTANT **
Please send your response to my e-mail address […] otherwise your response will land on the listserv and therefore compromise your anonymity and clog the list. Thanks!
Year of birth
Please list ALL the medication you are currently taking (for RSD pain, for RSD symptoms, and for other purposes)
Name of drug Dosage Form and frequency (e.g. pill, patch, etc.) (e.g. 1000mg/day)
How long have you had RSD?
What part of your body is affected by RSD?
Do you have any other medical conditions you are aware of?
PLEASE PROVIDE ANSWERS FOR ONLY ONE MEDICATION ON THIS FORM! IF YOU WISHTO PROVIDE ANSWERS FOR MORE THAN ONE MEDICATION (BECAUSE YOU USE MORE THAN ONE OR BECAUSE YOU HAVE USED ANOTHER ONE IN THE PAST), PLEASE RE-USE THIS FORM.
1. Medication taken for RSD pain (not for other symptoms) Name
Dose (e.g. 100 mg)
Frequency (e.g. every morning, every 3 hours, when needed, etc.)
Form (e.g. pill, patch, pump, etc.)
2. Does this medication reduce your pain?
3. How much does it reduce your pain? Please tick off one of the answers below.
( ) 100%
( ) about 75% or more
( ) about 50% or more
( ) about 25% or more
( ) less than 25%
4. When does pain relief set in?
5. How long does it last?
6. What are the side effects? (Please list all of them; if you think they might be due to other factors, too, please note that, too.)
7. What does taking the medication enable you to do?
8. What does taking the medication prevent you from doing?
9. What else, beside medication, do you do to relieve your pain?
10. Do you have any further comments?
** IMPORTANT ** IMPORTANT **
Please send your response to my e-mail address […] otherwise your response will land on the listserv and therefore compromise your anonymity and clog the list. Thanks!
THANK YOU VERY MUCH FOR YOU HELP!
CSS for moritherapy.com
List of medications taken
Ambien – zolpidem – anxiolytic
Aspirin – ASA – NSAID
Atenolol – antihypertensive
Ativan – lorazepam – anxiolytic
Butalbital – barbiturate
Citrucel – stool softener
Clindex – chlordiazepoxide hydrochloride & clodinium bromide – antiulcer, anxiolytic
Clonidine – catapres – antihypertensive
Cyclobenzaprine – muscle relaxant
Cytomel – liothyronine – antidepressant potentiator (thyroid med)
Darvocet – propoxyphene & acetaminophen – opioid/NNA
Deconamine – guaifenesin, hydrocodone & pseudoephedrine – opioid
Demerol – meperidine – opioid
Depakote – divalproex sodium – anticonvulsant
Dyazide – triamterene & hydrochlorothiazide – antihyptertensive
Arnér, S. & Meyerson, B.A. (1988) Lack of analgesic effect of opioids on neuropathic and idiopathic forms of pain. Pain, 33, 11-23
Becker, W.J., Ablett, D.P., Harris, C.J. & Dold, O.N. (1995) Long term treatment of intractable reflex sympathetic dystrophy with intrathecal morphine. Canadian journal of neurological sciences, 22, 153-159
Berry, H. (1996) Re: Long term treatment of intractable reflexsympathetic dystrophy with intrathecal morphine. Canadian journal of neurological sciences, 23,(2) 156-157
Bryson, H.M. & Wilde, M. (1996) Amitriptyline. A review of itspharmacologcal properties and therapeutic use in chronic pain states. Drugs & Aging, 8(6). 459-476
Clyman, B.B. (1996) Osteoarthritis: new roles for drug therapy and surgery. Geriatrics, 51(9), 33-36
Czop, C., Smith, T., Rauck, R. & Koman, L.A. (1996) The pharmacologic approach to the painful hand. Hand clinics, 12(4), 633-642
Dotson, R.M. (1993) Causalgia – reflex sympathetic dystrophy -sympathetically maintained pain: myth and reality. Muscle & nerve, 16, 1049-1055
Durrett, R. (1996) Drugs commonly used to treat RSD. Published by The Reflex Sympathetic Dystrophy Network at http://www.rsdnet.org/rframes/RSD-druglist.html, access date: April 5, 1997
Egle, U.T. & Hoffmann, S.O. (1990) Psychosomatic aspects of reflex sympathetic dystrophy. In: Stanton-Hicks, M., Jaenig, W. & Boas, R.A.(eds.): Reflex sympathetic dystrophy. Boston: Kluwer.
Fishbain, D.A., Goldberg, M., Rosomoff, R.S. & Rosomoff, H. (1991) Completed suicide in chronic pain. Clinical journal of pain, 7(1), 29-36.
Gallien, P., Nicolas, B., Robineau, S., Lebot, M.-P. & Brissot, R. (1995) The reflex sympathetic dystrophy syndrome in patients who have had a spinal cord injury. Paraplegia, 33, 715-720
Gamsa, A. (1994): The role of psychological factors in chronic pain, I & II, Pain 57(1), 5-29
Gibbons, J.J. & Wilson, P.R. (1992) RSD score: a criteria for thediagnosis of reflex sympathetic dystrophy and causalgia. Clinicaljournal of pain, 8, 260-270
Gragnani, J. (1994) A review of reflex sympathetic dystrophy and related syndromes. Missouri medicine, 91(11) 680-683
Hareau, J. (1996) What makes treatment of reflex sympathetic dystrophy successful? Journal of hand therapy, 9, 367-370
Harkins, S., Linford, J., Cohen, J., Kramer, T. & Cueva, L. (1991) Adminisitration of clonazepam in the treatment of TMD and associated myofascial pain: a double-blind pilot study. Journal of craniomandibular disorders: Facial and oral pain, 5, 179-186
Hooshmand, H. (1995) RSD puzzles. Vero Beach: Neurological Associates
International Association for the Study of Pain (1986) Classification of chronic pain. Descriptions of chronic pain syndromes and definitions of pain terms. Prepared by the Subcommittee on Taxonomy. Pain, 33, suppl. 3
Julien, R.M. (1995) A primer of drug action. New York: Freeman and Company
Kozin, F. (1992) Reflex sympathetic dystrophy syndrome: a review.Clinical and experimental rheumatology, 10, 401-409
Levine, D.Z. (1991) Burning pain in an extremity. Postgraduate medicine, 90(2), 175-185
Lynch, M.E. (1992) Psychological aspects of reflex sympathetic dystrohy: a review of the adult and paediatric literature. Pain, 49, 337-347
McNerney, J.E. (1991) Reflex sympatehtic dystrophy. Traumatic andpostoperative presentation and management in the lower extremity.Clinics in podiatric medicine and surgery, 8(2), 287-307
Mellick, G.A. & L.B. (1997) Reflex sympathetic dystrophy treated with gabapentin. Archives of physical and medical rehabilitation, 78, 98-105
Merskey, H. (1997) Pharmacological approaches other than opioids in chronic non-cancer pain management. Acta anaesthesiologica scandinavica, 41, 187-190
Muizelaar, J.P., Kleyer, M., Hertogs, I.A.M. & DeLange, D.C. (1997) Complex regional pain syndrome (reflex sympathetic dystrophy andcausalgia): management with the calcium channel blocker nifedipine and/or the a -sympathetic blocker phenoxybenzamine in 59 patients. Clinical neurology and neurosurgery, 99, 26-30
Patterson, D.R., Ptacek, J.T., Carrougher, G.J. & Sharar , S.R. (1997) Lorazepam as an adjunct to opioid analgesics in the treatment of burn pain. Pain, 72, 367-374
Portenoy, R.K., Foley, K.M. & Inturrisi, C.E. (1990) The nature of opioid responsiveness and its implications for neuropathic pain: new hypotheses derived from studies of opioid infusions. Pain, 43, 73-286
Portenoy, R.K. (1996) Opioid therapy for chronic nonmalignant pain: clinician’s perspective. Journal of law, medicine & ethics, 24, 296-309
Roberts, W.J. (1986) A hypothesis on the physiological basis forcausalgia and related pains. Pain, 24, 297-311
Rogers, James N., Valley, Marc A. (1994) Reflex sympathetic dystrophy. Clinics in podiatric medicine and surgery, 11(1), 73-83
Rosenberg, J.M., Harrell, C.H., Ristic, H., Werner, R.A. & de Rosayro, A.M. (1997) The effect of gabapentin on neuropathic pain. The clinical journal of pain, 13, 251-255
Singh, L., Field, M.J., Ferris, P., Hunter, J.C., Oles, R.J., Williams, R.G. & Woodruff, G.N. (1996) The antiepileptic agent gabapentin(neurontin) possesses anxiolytic-like and antinociceptive actions that are reversed by D-serine. Psychopharmacology, 127, 1-9
Stanton, R.P., Malcolm, R., Wesdock, K.A. & Singsen, B.H. (1993) Reflex sympathetic dystrophy in children: An orthopedic perspective.Orthopedics, 16(7), 773-779
Stanton-Hicks, M., Jaenig, W., Hassenbusch, S., Haddox, J.D., Boas, R. & Wilson, P. (1995) Reflex sympethetic dystrophy: changing concepts and taxonomy. Pain, 63, 127-133
van Laere, M. & Claessens, M. (1992) The treatment of reflex sympathetic dystrophy syndrome: Current concepts. Acta orthopedica belgica, 58(suppl.1), 259-261