Complex Regional Pain Syndrome: What's in a Name? - Thblack.com

The Journal of Pain, Vol 12, No 1 (January), 2011: pp 2-12Available online at www.sciencedirect.comFocus ArticleComplex Regional Pain Syndrome: What’s in a Name?Terence J. CoderreDepartment of Anesthesia and Alan Edward Centre for Research on Pain, McGill University, and McGill UniversityHealth Centre Research Institute, Montreal, Quebec.Abstract: Within a 2-year period in the 1940s, 2 Boston physicians published dramatically opposingviews on the underlying nature of a syndrome now known as complex regional pain syndrome(CRPS). Evans suggested, in several papers in 1946–1947, that sympathetic reflexes maintain painand dystrophy in affected limbs. Foisie, in 1947, suggested arterial vasospasms were key in the etiologyof this pain syndrome. Evans’ hypothesis established the nomenclature for this syndrome for 60years, and his term, ‘‘reflex sympathetic dystrophy,’’ guided clinical treatment and research activitiesover the same period. Foisie’s proposed nomenclature was unrecognized, and had virtually no impacton the field. Recent evidence suggests that Evans’ contribution to the field may have in fact ledclinicians and researchers astray all those years. This focus article on CRPS compares recent observationswith these 2 earlier theories and asks the question—what if we had adopted Foisie’s nomenclaturefrom the beginning?Perspective: This article discusses 2 opposing historical views on the etiology of what is nowknown as CRPS, and how they affected nomenclature, research, and clinical therapy in subsequentdecades. This focus article may help researchers and clinicians realize the importance of syndromenames, and how they may inadvertently misdirect research and treatment.ª 2011 by the American Pain SocietyKey words: Complex regional pain syndrome, CRPS, reflex sympathetic dystrophy, arterial vasospasm,nomenclature, etiology, pathophysiology, ischemia.Complex regional pain syndrome (CRPS; see Table 1for characteristics and diagnostic criteria) 35,61 isthe current name for a syndrome that has beenknown over the years by many names, includingcauslagia, minor causalgia, algodystrophy, shoulderhandsyndrome, and Sudeck’s atrophy. However, themost common and recently used name for CRPS was reflexsympathetic dystrophy. 83,86 Although mechanisticdifferences may not exist, the term CRPS-I has beenused to replace reflex sympathetic dystrophy and CRPS-II to replace causalgia, with the former diagnosis in caseswhere no nerve injury is detected, and the latter whennerve injury is confirmed. 61 Recent discouragementwith the sympathetic nervous system as a therapeutic target,13,25,53 and an underlying pathology, 3,66,84,90 had noSupported by grants from CIHR, NSERC, and the Louise and Alan EdwardsFoundation.Address reprint requests to Dr Terence J. Coderre, Anesthesia ResearchUnit, 1203-3655 Promenade Sir William Osler, Montreal, QC, H3G 1Y6,Canada. E-mail: terence.coderre@mcgill.ca1526-5900/$36.00ª 2011 by the American Pain Societydoi:10.1016/j.jpain.2010.06.001doubt encouraged the taxonomic change to CRPS, 83,86and it has been argued that the former syndromename may have misguided research and treatmentsince its inception. 8 In a treatise on the concept that inspiredWilliam Shakespeare to pen the phrase ‘‘Thatwhich we call a rose by any other name would smell assweet,’’ the question should be asked—‘‘What’s ina (syndrome) name?’’ (Romeo and Juliet 2.2,43-44). 81That is, would history have been different if anothername than reflex sympathetic dystrophy had been adoptedfor the predominant nomenclature of CRPS? Toaddress this question, it would be instructive to reviewthe thinking of the time, back in the mid-1940s, whenthe term reflex sympathetic dystrophy (and an alternativename) was coined.In 1946–47, James A. Evans, a physician at the LaheyClinic in Burlington, Massachusetts, reported onpatients’ intense suffering combined with featurescharacteristic of what he called ‘‘stimulation of the sympathetic,. . . namely rubor, pallor, or a mixture of both,sweating and atrophy. . . .’’ 20 In a series of 4 papers thatdescribed at first 32, 19,20 then a total of 57 patients, 21,22he observed these symptoms in association with2

Coderre The Journal of Pain 3Table 1. International Association for the Studyof Pain (IASP) Diagnostic Criteria for CRPS-Iand CRPS-II 61CPRS-I(REFLEX SYMPATHETIC DYSTROPHY)1. The presence of an initiatingnoxious event, or a cause ofimmobilization.2. Continuing pain, allodynia, orhyperalgesia with which thepain is disproportionate toany inciting event.3. Evidence at some time ofedema, changes in skin bloodflow, or abnormal sudomotoractivity in the region of thepain.4. This diagnosis is excluded bythe existence of conditionsthat would otherwiseaccount for the degree ofpain and dysfunction.Note: Criteria 2 to 4 must besatisfied.CRPS-II (CAUSALGIA)1. The presence of continuingpain, allodynia, orhyperalgesia after a nerveinjury, not necessarily limitedto the distribution of theinjured nerve.2. Evidence at some time ofedema, changes in skin bloodflow, or abnormal sudomotoractivity in the region of thepain.3. This diagnosis is excluded bythe existence of conditionsthat would otherwiseaccount for the degree ofpain and dysfunction.Note: All 3 criteria must besatisfied.fractures (21%), sprains (21%), vascular complications(19%), amputations (9%), joint/bone inflammation(5%), and lacerations (2%), or more minor injuries/conditions including bruises (9%) and static defects ofthe foot (7%). Pain was described as deep, boring,diffuse, but with trigger points, and manipulation oruse aggravated patient suffering. Importantly, Evansreported that often pain in these patients was relievedby sympathetic blocks, as the limb warmed. Evansattributed the pain to abnormality in the sympatheticnervous system, and first used the term reflexsympathetic dystrophy. 19-22 Evans suggested thatexcessive afferent nerve inputs associated with injuryinitiated central nervous system activity that spreadthrough an ‘‘internuncial pool’’ of hyperactive spinalneurons that stimulated sympathetic neurons. Activitygenerated in postganglionic sympathetic efferents wasthen proposed to induce arterial spasms and ischemiathat increased capillary filtration pressure withresultant edema and swelling. Evans’ conclusions wereclearly influenced by Lorento de No, 51 who discussedhow ‘‘prolonged bombardment of pain impulses’’ triggers‘‘pools of constant circling of activity across (spinal)synapses’’ and ‘‘a vicious circle’’ which causes sympatheticneurons to trigger vascular spasms that raise filtrationpressure causing oedema. He was also inspired by Livingston,50 who had applied Lorento de No’s above ideas andMacKenzie’s 52 theory of an ‘‘irritable focus’’ in spinalcord (in relation to referred pain) to the condition of causalgia,and who had adopted the techniques of Leriche 49by performing sympathectomies to break the viciouscircle.One year after Evans coined the term reflex sympatheticdystrophy, Philip S. Foisie, a surgeon at BostonCity Hospital, concluded that a persisting low-grade arterialspasm after soft-tissue injury can lead to a syndromewith severe pain, extreme tenderness or allodynia, tissueswelling, muscle atrophy, bone osteoporosis, jointstiffness, and limited motion. Basically, the constellationof symptoms that Evans would diagnose as reflexsympathetic dystrophy, Foisie termed ‘‘traumatic arterialvasospasm.’’ 23 According to Foisie, subacute, undetectedarterial vasospasms led to decreases in nutritional bloodsupply that resulted in degenerative changes primarily inthe musculoskeletal system. Foisie was careful to pointout that these vasospasms were not in large vesselsthat produce tissue-threatening ischemia, but rather insmall arterioles. As causalities of war, some of the injuriesto Foisie’s patients were severe and may have includednerve as well as soft-tissue injury, but others were less severeand the pain was disproportionate to the injury. Hedescribed these symptoms in a single paper on woundedsoldiers with fractures, soft-tissue injuries, minor gunshotand shell-fragment wounds, but not in extensive‘‘wide-open’’ wounds. 23 Importantly, he observed thatthe syndrome was even more likely in patients with compressioninjuries that resulted in increased tissue tension(ie, where microvascular pressure is reduced because ofincreased intracompartmental pressure). He also notedthat swelling was often present early on, particularly followingcasting and immobilization, and concluded thatedema should be expected as ischemia causes microvascularinjury that increases plasma extravasation withincreased permeability of capillary walls. Additionally,these patients would sometimes respond to a combinationof sympathetic blocks, massage, and physical therapyto encourage the restoration of nutritional bloodsupply back to deep tissues. Foisie described 2 cases inwhich pain relief was obtained following upper dorsal(sympathetic) ganglionectomy or lumbar (sympathetic)ganglionectomy. Foisie was probably influenced by thework of Leriche, 49 who had shown the sympathectomieswere useful in the treatment of posttraumatic contracturesand causalgia. He was likely also influenced byHomans 39 (also from Boston), who discussed the possibilitythat ‘‘gross arterial spasm’’ and ‘‘diffuse peripheralspasm’’ contributed to minor causalgia.The most interesting point to note is that althoughEvans and Foisie were both suggesting that sympatheticblocks could be useful to relieve pain in their patients,they came to different conclusions as to what pathologyproduced the pain. Although, like Foisie, Evans identifiedarterial vasospasms and capillary permeability askey factors, it was their emphasis on sympathetic nervoussystem activity that differed dramatically. For Evans, increasedactivity in sympathetic efferents was crucial tothe maintenance of the vicious circle, and spinalhyperactivity maintained the reflex sympathetic drive.For Foisie, it was the nature of injury itself that initiatedthe vasospasm (particularly in cases where injuries increasedintracompartmental pressure), and sympathetic

4 The Journal of Pain Historical Reflection on Syndrome Names for CRPSblocks were seen as a way to overcome the vasospasm.And as history has shown through the citation and keyword record searches (see Fig 1), Evans’ conclusionswere almost universally adopted, whereas Foisie’s werealmost completely ignored. Indeed, the volume of referencesto the term reflex sympathetic dystrophy clearlyshow that this name dominated medical thinkingthroughout the latter half of the 20 th century. In fact,the emphasis of the sympathetic contribution to reflexsympathetic dystrophy has been a large driver in boththe predominant therapeutic practice and mechanisticresearch activity for the past 6 decades.403530252015105EvansFoisieIf Not Sympathetically Maintained Pain,Then What?Now that the pendulum has swung away from theview that sympathetic nervous system is causal for thepain of CRPS, the question is whether it is about timeto seriously consider Foisie’s hypothesis. From the obviouslack of attention that Foisie’s theory received, couldit be said that while being preoccupied by the notions ofreflex sympathetic dystrophy and vicious circles, andlater by theories of sympathetically maintained painand sympathetically independent pain (see below),Foisie’s hypothesis was the one that got away? What ifrather than reflex sympathetic dystrophy and sympatheticallymaintained pain, traumatic arterial vasospasms,and ischemia-maintained pain were the catch phrasesfor this syndrome? Would a diagnosis of traumatic arterialvasospasm have changed the way the patient wouldhave been treated, or would a syndrome so named havechanged the course of research history? Given the lack ofprogress in the field during the reflex sympathetic dystrophyfixation, and recent research findings that supportFoisie’s conclusion, it is likely that the answer is yes!So what has been learned over the 60-odd years sinceEvans’ and Foisie’s respectively renowned and ignoredhypotheses? First, it is becoming clear that sympatheticblocks do not always relieve pain in CRPS patients. 13,25,53Clearly, some patients have remarkable but often shortlivedresponses to sympathetic blocks. Other patientsdo not respond at all, and still others respond, as notedby both Evans and Foisie, only following multiple blocks.However, since many of these trials lack control subjects,gradual improvement over time may reflect the naturalhistory of the syndrome, rather than the use of sympatheticnerve blocks. These findings lead to 2 possible conclusions:1) that the pain of CRPS depends on thesympathetic nervous system in some patients but notothers; or 2) that the sympathetic nervous system is notcritical to pain in CRPS, but may influence pain by affectingother crucial factors. The first conclusion was adoptedby proponents of the once-popular notions ofsympathetically maintained 70 and sympathetically independent24 pain (where the pain is proposed to be dependenton, or independent of, activity in sympatheticefferent fibers, respectively), whereas the latter conclusionhas not, until recently, received much attention.012001000800600400200040's 50's 60's 70's 80's 90's 00'sReflex sympatheticdystrophyTraumatic arterialvasospasm50's 60's 70's 80's 90's 00'sFigure 1. Impact of Evans’ and Foisie’s papers on the field. (A)Histogram of the numbers of Science Citation Index citationsfound for the papers of Evans 19-22 and Foisie, 23 respectively bydecade between 1940 and 2009. (B) Histogram of the numbersof Medline citations found for the terms ‘‘reflex sympatheticdystrophy’’ and ‘‘traumatic arterial vasospasm’’ by decadebetween 1950 and 2009.A second important finding is that numerous investigatorshave now reported that sympathetic activity isactually blunted, rather than enhanced, particularly inearly-stage CRPS (when the limb is often hot andswollen). 45,77,95 This findings leads to the importantconclusion that CRPS cannot depend on enhancedactivity in sympathetic efferent fibers, at least not atstages during which the activity is blunted. It alsosupports the theory that late-phase CRPS (when the limbis often cold and cyanotic) may depend on denervationhypersensitivity of sympathetic vasoconstrictors, 27,53,89a phenomenon that would be expected if sympatheticactivity is blunted for prolonged periods.Role of Ischemia-Reperfusion InjuryA third critical finding is that many CRPS patients havesubtle but convincing evidence of persistent tissue ischemiain skin and particularly deep tissue. Thus, muscle tissuefrom an amputated CRPS limb was found to exhibit

Coderre The Journal of Pain 5lipofuscin pigment, atrophic fibers, and severely thickenedbasal membrane layers of the capillaries. 90 Althoughamputated tissue may represent extreme casesof CRPS, this conclusion is also supported by functionalobservations of increased density of perfused vessels,lower capillary filtration capacity, arteriovenous shunting,and ischemia of peripheral subcutaneous tissue inintact CRPS limbs. 57,78 It has also been reported thatCRPS limbs have high arterial flow, but low oxygenconsumption, as well as high lactate flux—indicative oftissue ischemia, despite increased flow in largevessels. 29 There is also an impairment of high-energyphosphate metabolism in muscle tissue of CRPS limbs, 30,37suggestive of mitochondrial oxygen deficiency. All ofthese observations suggest that ischemia in deep tissuescontributes to CRPS. Others have also found evidence ofmicrovascular injury in skin, including hypertrophicmultilayered, and sometimes collapsed, endothelial cellwalls around the arteries and arterials from the skin ofamputated CRPS limbs, 3 and reduced skin nutritive flowand capillary hemoglobin oxygenation in intact CRPSlimbs. 44,45,71So it seems that Evans may have been wrong aboutsympathetic hyperactivity. However, if sympathetic hyperactivitydoes not cause the arterial vasospasms thatboth Foisie and Evans described, what does? Foisie provideda hint about a possible mechanism. The hint wasthat arterial vasospasms and painful symptoms weremore likely (and more severe) after compression injuriesthat increased tissue tension (that is, lowered microvascularpressure because of increased intracompartmentalpressure). Foisie suggested that traumatic arterial vasospasmtakes place when increased tissue tension occursafter increased vascular permeability and accumulationof interstitial fluid. He further suggested that arterialvasospasm caused a ‘‘prolonged arterial inadequately’’or an oxidative ‘‘malnutrition’’ which ‘‘. . . results indegenerative changes of the musculoskeletal systemprimarily.’’ 23Importantly, injuries that produce enough edema toincrease intracompartmental pressure can result in microvascularchanges and injury that depend both onthe degree and length of time of the increased intracompartmentalpressure. Minor inflammatory injuries causemicrovascular disturbances that increase capillaryfiltration pressure and vascular permeability due to theformation of gaps in the endothelial cell lining of microvessels.58 However, as interstitial fluid increases, there isa compensatory increase in lymphatic flow, which preventsincreases in intracompartmental pressure. 62 Moresignificant inflammation, after moderate soft tissue injuryor fracture, produces plasma extravasation ata rate that cannot be compensated for by lymphaticflow, resulting in increased intracompartmental pressure.64,82 Intramuscular intracompartmental pressuresthat approach diastolic blood pressure (typicallyaround 70 mmHg) for at least several hours producea phenomenon known as acute compartmentsyndrome (although prolonged intracompartmentalpressures above 40 to 45 mmHg are typically a warningthat an acute compartment syndrome may occur, andindicate that aggressive treatment is warranted). 32,97Acute compartment syndrome results when thepressure from excessive interstitial fluid within a closedanatomical compartment compresses small bloodvessels and blocks oxygen supply to the tissue forseveral hours or more. 32The consequences of a very prolonged acute compartmentsyndrome can be catastrophic or even fatal, as theischemic muscle eventually degenerates, producing toxicmetabolites that lead to systemic inflammatory responsesyndrome and multiple organ dysfunction, characteristicof crush syndrome or tourniquet shock. 32,79 However,a critical study by Vollmar et al 92 has shown that intracompartmentalpressures much lower than those neededto produce acute compartment syndrome also have significanteffects on microvascular function. Thus, intracompartmentalpressures between 25 and 35 mmHgblock arterial flow in smaller arterioles and venules. Furthermore,an even lower intracompartmental pressureof 12 mmHg produces a 50% reduction in the numberof perfused capillaries, reducing nutritive blood flow tothe involved tissue. 92 Importantly, it was also foundthat subsequent reduction of intracompartmental pressureby as little as 4 mmHg restores capillary flow. 92Therefore, for a significant number of capillaries, capillaryflow can be blocked (causing ischemia) and reestablished(causing reperfusion) with very small changes inintracompartmental pressures, at levels that are muchlower than required to cause acute compartment syndrome.If the ischemia associated with the capillaryblockade occurs for a few hours, the tissue within thecapillary bed will be subjected to an ischemiareperfusion(IR) injury, once the pressure is reduced andthe tissue is reperfused.Importantly, it has been shown that after closed softtissueinjury of the rat hind limb, intracompartmentalpressures are elevated between 13.6 and 25.4 mmHg inthe anterior compartment and 7.9 and 16.1 mmHg inthe posterior compartment for up to 72 hours postinjury.Similar to tourniquet-induced IR injury, these increaseslead to significant vascular permeability, edema formation,leukocyte rolling and adherence, and decreases infunctional capillary density, but not acute compartmentsyndrome. 75,88 More severe soft-tissue injury also producesarterial constriction and capillary perfusion failurelasting 5 days, 6 and the microvasculature dysfunction associatedwith severe soft-tissue injury is alleviated withthe free radical scavenger N-acetylcysteine. 74 Soft-tissueinjury in the tibial compartment has also been shownto produce similar microvascular dysfunction in the rattibial periosteum, 76 whereas closed fracture of the rattibia produces microvascular dysfunction in adjacentskeletal muscle. 101 Indeed, patients often exhibit highmuscle intracompartmental pressures (over 40 mmHgfor 24 hours) after tibial fractures. The increased intracompartmentalpressures produce simultaneous reductionsin perfusion pressure (reflecting low capillaryflow), which correlates highly with later muscle deficits.64 Thus, it seems the mechanism on which Foisiewas not able to elaborate in his patients may havebeen microvascular dysfunction following IR injury. IR

6 The Journal of Pain Historical Reflection on Syndrome Names for CRPSinjuries are so named because of the recognition thatmost of the damage that occurs following prolonged ischemiaoccurs after the tissue is reperfused, a discoverymade 3 decades after Foisie’s paper.Importantly, recent evidence indicates that IR injuriesmay cause some of the bewildering symptoms observedin CRPS patients. IR injuries initially produce increasedcapillary and postcapillary venule permeability leadingto extensive plasma extravasation and edema, 58 a verysignificant feature of early-stage CRPS. 48,96 IR injuriesare produced by the generation of oxygen free radicalsand proinflammatory cytokines, 41,87,100 which produceboth microvascular dysfunction and additional IR injuryin neighboring tissues, including muscle, bone, andskin. 10 Recent evidence suggests that both oxygen-freeradicals and proinflammatory cytokines are elevatedin CRPS, 18,38 and CRPS is alleviated by treatments aimedat reducing these mediators. 40,67 CRPS often spreadsbeyond dermatomal borders, and affects both skin andvarious deep tissues. 54 This parallels what happens withIR injuries, where all tissues (including muscle, bone,and skin) within the occluded area suffer the same consequences(although bone and skin are more resistant thanmuscle). 10 While IR injuries clearly cannot account formotor abnormalities or altered sympathetic or sudomotoractivity often seen in CRPS limbs, or the occasionalspread of symptoms to the contralateral limb, it is possiblethat afferent input from injured muscle may beinstrumental in producing central sensitization 93 thatcould contribute to these phenomena.It could be argued that many CRPS patients do nothave the type of tissue damage that leads to IR injury.However, CRPS commonly follows factures, sprains, contusionsand crush injuries, athroscopic surgery, overlytight casting, and other edematous soft-tissue injuries.4,26,73 A common feature of all these conditions isan early inflammatory response in soft tissue that hasthe potential to produce microvascular and ischemicchanges in various tissues. While others have notedthat CRPS depends on an exaggerated inflammatoryresponse, 30,37 most CRPS patients are seen by painspecialists many weeks, months, or years after theiroriginal injuries, and initial inflammatory symptomsmay be long forgotten. If the soft-tissue injury andinflammation is severe enough to cause a temporary(hours long) increase in intracompartmental pressurethat occludes capillary blood flow, a microvascular IR injurywill occur. These microvascular injuries would thenbe expected to produce sustained hypoperfusion and reducednutritive blood flow. Importantly, since this is mostlikely to occur in deep tissues, the ensuing ischemicchanges may go undetected.Role of Microvascular DysfunctionIR injuries produce significant arterial vasospasms orenhanced vasoconstriction due to microvascular dysfunctionassociated with increased responsiveness of smoothmuscleadrenoceptors to the sympathetic transmitternorepinephrine, 72 and a reduced production of the endogenousvasodilator nitric oxide. 47,94 Both Foisie andEvans described arterial vasospasms, and recentevidence indicates that CRPS limbs exhibit bothreduced nitric-oxide 33 and enhanced vasoconstrictor responsivenessto norepinephrine. 2,7,16,89 Importantly, themain cause of arterial vasospasm or enhancedvasoconstriction following IR injury is the dramatic lossof nitric oxide (much of which is scavenged byincreased oxygen-free radicals to produce nitrogen-freeradicals). 10,94 Thus, arterial vasospasms can bemaintained after IR injury, even if sympathetic activityis blunted in CRPS patients, which itself would bea natural homeostatic response to the vasospasm.A very significant later consequence of IR injury is damageto capillaries. Free radicals and cytokines generatedby IR injury trigger neutrophil recruitment and adhesion,capillary plugging, endothelial cell swelling, and theeventual collapse of the capillary lumen, leading toa phenomenon called capillary no-reflow. 36,60 Capillaryno-reflow results in arterial venous shunting, 42 and a significantreduction of nutritive blood flow within theaffected capillary. Importantly, arterial venous shuntingand collapsed capillaries have been observed in the limbsof patients with CRPS. 57,90 Furthermore, IR injuriesinitially produce enhanced vascular permeability thatcontributes to the damage causing vasospasms, but asvasospasms progress to no-reflow, plasma extravasationis diminished, as there is reduced blood flow within themicrovasculature. 59 This shift from initial high plasma extravasationto later capillary no-reflow, could explainwhy many CRPS patients start out with significantedema, but then progress to a state with less edemaand more ischemia. 9,11,45 Later capillary no-reflow mayalso explain why CRPS is less likely to respond to antisympathetictreatment with greater time following syndromeonset, 1 as antisympathetic treatment likelyrelieves vasospasms, but not capillary no-reflow.Recent evidence further links the fields of CRPS and IRinjury by showing that after a hind paw IR injury, rodentsdevelop vascular and painful symptoms similar to that exhibitedby CRPS patients. 14,46 Allodynia that developsfollowing rat hind paw IR injury also depends on thegeneration of oxygen-free radicals and proinflammatorycytokines. 46 Importantly, after hind paw IR injury, ratswith vasoconstrictive hypersensitivity to norepinephrine(vasospasms) are more likely to develop allodynia, as wellas painful responses to intraplantar norepinephrine. 98Similar to CRPS patients, 17 sympathetic blockade usingguanethidine is more effective at relieving allodynia inrats when performed early after hind paw IR injury. 99 Aswell as vasospasms, rats with hind paw IR injuries developno-reflow over a time course that parallels allodynia, andexhibit muscle ischemia and increased muscle lactate (asdo CRPS patients 29 ) that is directly correlated with the allodynia.46 Like CRPS patients, 91 rats with hind paw IR injuriesalso display painful responses to exercise, and parallelincreases in muscle lactate and allodynia after exercise. 46Novel Therapeutics and ConclusionsThere is growing evidence that targeting pathologiesassociated with microvascular dysfunction and ischemia

Coderre The Journal of Pain 7Figure 2. Consensus views of therapy for CRPS at the end of the 20 th century. Therapeutic algorithm (‘‘Diagnosis CRPS CareContinuum’’) with emphasis on therapeutic options in response to patient’s clinical progress in the rehabilitation pathway. Basedon the Interdisciplinary Clinical Pathway for CRPS prepared by an Expert Panel in Cardiff, Wales in 2001 (reprinted with permissionfrom Stanton-Hicks et al, An updated interdisciplinary clinical pathway for CRPS: Report of an expert panel, Pain Practice, John Wileyand Sons 85 ).are effective against CRPS. Thus, CRPS is alleviated withfree radical scavengers, 31,67,104 and antitumor necrosisfactor-a antibodies, 40,56,80 which would mitigate theinflammatory and damaging effects of oxygen-free radicalsand proinflammatory cytokines. CRPS has also beenshown to be alleviated by nitric-oxide donors and othervaosodilators, 34,55,63 or a-adrenergic antagonists, 63,68and topical or regional clonidine, 15,28,43,69 which act byinhibition of the effects of sympathetic vasoconstrictorson vasculature smooth-muscle cells. Antioxidant vitaminC administration has also recently been shown to reducethe incidence of CRPS when administrated as a prophylactictreatment after fracture. 5,102,103The very recent shift in thinking about the treatmentof CRPS is reflected in the differences between a therapeuticalgorithm for CRPS outlined by an expert panelthat met in Cardiff in 2001 85 (Fig 2) versus theevidence-based guidelines for CRPS published by theNetherlands Society of Rehabilitation Specialists andthe Netherlands Society of Anaesthesiologists in 2007 65(Fig 3). The 2001 algorithm lists sympathetic nerve blocksas a first interventional pain-management procedure

8 The Journal of Pain Historical Reflection on Syndrome Names for CRPSFigure 3. Very recent views on therapy for CRPS. Summary chart of the 2007 evidence-based guidelines for complex regional painsyndrome reported by the Netherlands Society of Rehabilitation Specialists and the Netherlands Society of Anesthesiologists. 65Reprinted with permission of the Dutch Institute for Health Care Improvement, Evidence-Based Guidelines Development programme(EBGD) of the Order of Medical Specialists.and lists surgical sympathectomy as a more invasive procedureto be performed in case of inadequate or partialresponse. The 2007 Dutch evidence-based guidelines listfree radical scavengers, 50% DMSO, and n-acetylcysteineunder drug treatments, as well as vasodilators for ‘‘coldCRPS’’ and vitamin C as primary prevention for wrist

Coderre The Journal of Pain 9fractures. Importantly, these later guidelines do not listsympathetic blockade as a first-line treatment, and recommend‘‘extreme caution’’ when considering surgicalsympathectomy. Importantly, sympathetic blocks maystill be useful in alleviating CRPS pain associated with microvasculardysfunction and deep tissue ischemia, butthey should not imply that sympathetic abnormalitiesare causative.While recent evidence points to the strong possibilitythat following IR injury, arterial vasospasms, no-reflow,and deep-tissue ischemia may be key underlying mechanismsof CRPS, one may ask the question: Why were thesemechanisms overlooked for so long? Both Evans andFoisie described arterial vasospasms and ischemia as keyfeatures present in their patients. Perhaps the sympatheticreflex hypothesis derived by Evans was just toocompelling to allow us to entertain other possibilities.And by concentrating on the sympathetic nervous system,everyone lost sight of the key role of vasospasmsand ischemia. Evans’ hypothesis was based on the earliertheories of Livingston, 50 a clinician-scientist who hada large influence on his contemporaries and futurepain researchers. The ideas of Evans and Livingstonclearly had an impression on John Bonica, key founderand second president of the International Associationfor the Study of Pain. Bonica 11,12 was a strongproponent of the idea that reflex sympatheticdystrophy depended on a vicious circle and sympatheticefferent hyperactivity, and he was instrumental in theadoption of the nomenclature proposed by Evans. Butmore than people, it is ideas that shape researchtrends, and it seems that, at the time, Evans’ idea wasmore attractive than Foisie’s. This may have stemmedfrom a neuron-centric view of pain processing, with anoveremphasis of the importance of nerve cells to thegeneration, as well as transmission, of pain signals.Hopefully, recent research findings outlined here will inspirepain researchers and clinicians to consider endothelialcells and the microvasculature as important potentialgenerators of the noxious inputs critical for CRPS, andput to rest not only the name, reflex sympathetic dystrophy,but the misguided theories, research, and limitedtreatment it spawned. Perhaps more than 6 decadeslater, Foisie and his little-known paper will finally receivetheir due recognition, and patients with CRPS will benefitfrom the multitude of additional potential therapiesnew thinking will generate.AcknowledgmentsThe author wishes to thank Dr. Gary J. Bennett forcomments on the manuscript.References1. AbuRahma AF, Robinson PA, Powell M, Bastug D,Boland JP: Sympathectomy for reflex sympathetic dystrophy:Factors affecting outcome. Ann Vasc Surg 8:372-379,19942. Ackerman WE, Munir MA, Zhang J-M: Assessment of laserDoppler imaging for the diagnosis of complex regional painsyndrome I. J Neuropathic Pain Symp Pal 1:13-20, 20053. Albrecht PJ, Hines S, Eisenberg E, Pud D, Finlay DR,Connolly MK, Paré M, Davar G, Rice FL: Pathologic alterationsof cutaneous innervations and vasculature in affectedlimbs from patients with complex regional pain syndrome.Pain 120:244-266, 20064. Allen G, Galer BS, Schwartz L: Epidemiology of complexregional pain syndrome: A retrospective chart review of134 patients. Pain 80:539-544, 19995. Amadio PC: Vitamin C reduced the incidence of reflexsympathetic dystrophy after wrist fracture. J Bone JointSurg Am 82:873, 20006. Amon M, Laschke MW, Harder Y, Vollmar B, Menger MD:Impact of severity of local soft-tissue trauma on long-termmanifestation of microcirculatory and microlymphaticdysfunctions. J Trauma 61:924-932, 20067. Arnold JMO, Teasell RW, Macleod AP, Brown JE,Carruthers SG: Increased venous alpha-adrenoceptors responsivenessin patients with reflex sympathetic dystrophy.Ann Intern Med 83:185-192, 19938. Bennett GJ: Epilogue, in Harden RN, Baron R, Janig W(eds): Complex Regional Pain Syndrome, Progress in Pain Researchand Management, Vol 22. Seattle, WA, IASP Press,2001, pp 323-3259. Birklein F, Riedl B, Claus D, Neundorfer B: Pattern of autonomicdysfunction in time course of complex regional painsyndrome. Clin Auton Res 8:79-85, 199810. Blaisdell FW: The pathophysiology of skeletal muscle ischemiaand the reperfusion syndrome: A review. CardiovascSurg 10:620-630, 200211. Bonica JJ: Causalgia and other reflex sympatheticdystrophies, in Bonica JJ (ed): The Management of Pain,2nd ed. Philadelphia, PA, Lea & Febiger, 1990, pp 220-24312. Bonica JJ: The Management of Pain, 1st ed. Philadelphia,PA, Lea & Febiger, 195313. Cepeda MS, Lau J, Carr DB: Defining the therapeutic roleof local anesthetic sympathetic blockade in complex regionalpain syndrome: A narrative and systematic review.Clin J Pain 18:216-233, 200214. Coderre TJ, Xanthos DN, Francis L, Bennett GJ: Chronicpost-ischemia pain (CPIP): A novel animal model of complexregional pain syndrome-type I (CRPS-I;reflex sympatheticdystrophy) produced by prolonged hindpaw ischemia andreperfusion in the rat. Pain 112:94-105, 200415. Davis KD, Treede RD, Raja SN, Meyer RA, Cambell JN: Topicalapplication of clonidine relieves hyperalgesia in patientswith sympathetically maintained pain. Pain 47:309-317, 199116. Dayan L, Salman S, Norman D, Vatine JJ, Calif E, Jacob G:Exaggerated vasoconstriction in Complex Regional PainSyndrome-1 is associated with impaired resistance artery endothelialfunction and local vascular reflexes. J Rheumatol35:1339-1345, 200817. Driessen JJ, van der Werken C, Nicolai JP, Crul JF: Clinicaleffects of regional intravenous guanethidine (Ismelin) inreflex sympathetic dystrophy. Acta Anaesthesiol Scand 27:505-509, 1983

10 The Journal of Pain Historical Reflection on Syndrome Names for CRPS18. Eisenberg E, Shtahl S, Geller R, Reznick AZ, Sharf O,Ravbinovich M, Erenreich A, Nagler RM: Serum and salivaryoxidative analysis in Complex Regional Pain Syndrome. Pain138:226-232, 200819. Evans JA: Reflex sympathetic dystrophy. Surg Clin NorthAmerica 26:780-790, 194620. Evans JA: Reflex sympathetic dystrophy. Surg GynecolObstet 82:36-44, 194621. Evans JA: Reflex sympathetic dystrophy: report of 57cases. Ann Intern Med 26:417-426, 194722. Evans JA: Sympathectomy for reflex sympathetic dystrophy:A case report of twenty-nine cases. JAMA 132:620-623,194623. Foisie PS: Traumatic arterial vasospasm. N Engl J Med237:295-302, 194724. Frost SA, Raja SN, Campbell JN, Meyer RA, Khan AA:Does hyperalgesia to cooling stimuli characterize patientswith sympathetically maintained pain (reflex sympatheticdystrophy)? In Dubner R, Gebhart GF, Bond MR (eds):Proceedings of the 5th World Congress on Pain, PainResearch and Clinical Management, Vol. 3. Amsterdam,The Netherlands, Elsevier, 1988, pp 151-15625. Furlan AD, Lui PW, Mailis A: Chemical sympathectomyfor neuropathic pain: Does it work? Case report and systematicliterature review. Clin J Pain 17:327-336, 200126. Galer BS, Henderson J, Perander J, Jensen MP: Course ofsymptoms and quality of life measurement in ComplexRegional Pain Syndrome: A pilot survey. J Pain SymptomManage 20:286-292, 200027. Gibbs GF, Drummond PD, Finch PM, Phillips JK: Unravellingthe pathophysiology of complex regional painsyndrome: Focus on sympathetically maintained pain. ClinExp Pharmacol Physiol 35:717-724, 200828. Gintautas J, Housny W, Kraynack BJ: Successful treatmentof reflex sympathetic dystrophy by bier block with lidocaineand clonidine. Proc West Pharmacol Soc 42:101, 199929. Goris RJ: Conditions associated with impaired oxygenextraction, 1st ed, in Guitierrez G, Vincent JL (eds): Tissue OxygenUtilization. Berlin, DE, Springer, 1991, pp 350-36930. Goris RJ: Reflex sympathetic dystrophy: Model of a severeregional inflammatory response syndrome. World J Surg 22:197-202, 199831. Goris RJ, Dongen LM, Winters HA: Are toxic oxygenradicals involved in the pathogenesis of reflex sympatheticdystrophy? Free Radic Res Commun 3:13-18, 198732. Gourgiotis S, Villias C, Germanos S, Foukas A,Ridolfini MP: Acute limb compartment syndrome: A review.J Surg Educ 64:178-186, 200733. Groeneweg JG, Huygen FJ, Heijmans-Antonissen C,Niehof S, Zijlstra FJ: Increased endothelin-1 and diminishednitric oxide levels in blister fluids of patients with intermediatecold type complex regional pain syndrome type 1. BMCMusculoskelet Disord 7:91, 200634. Groeneweg JG, Niehof S, Wesseldijk F, Huygen FJ,Zijlstra FJ: Vasodilative effect of isosorbide dinitrate ointmentin complex regional pain syndrome type 1. Clin JPain 24:89-92, 200835. Harden RN, Bruehl S, Stanton-Hicks M, Wilson PR:Proposed new diagnostic criteria for complex regionalpain syndrome. Pain Med 8:326-331, 200736. Harris AG, Steinbauer M, Leiderer R, Messmer K: Role ofleukocyte plugging and edema in skeletal muscle ischemiareperfusioninjury. Am J Physiol 273:H989-H996, 199737. Heerschap A, den Hollander JA, Reynen H, Goris RJA:Metabolic changes in reflex sympathetic dystrophy: A31P NMR spectroscopy study. Muscle Nerve 16:367-373,199338. Heijmans-Antonissen C, Wesseldijk F, Munnikes RJ,Huygen FJ, van der Meijden P, Hop WC, Hooijkaas H,Zijlstra FJ: Multiplex bead array assay for detection of 25 solublecytokines in blister fluid of patients with complex regionalpain syndrome type 1. Mediators Inflamm 1:28398, 200639. Homans J: Minor causalgia following injuries andwounds. Ann Surg 113:932-941, 194140. Huygen FJ, Niehof S, Zijlstra FJ, van Hagen PM, vanDaele PL: Successful treatment of CRPS 1 with anti-TNF.J Pain Sympt Manage 27:101-103, 200441. Inauen W, Suzuki M, Granger DN: Mechanisms of cellularinjury: Potential sources of oxygen free radicals in ischemia/reperfusion.Microcirc Endothelium Lymphatics 5:143-155, 198942. Kennedy TJ, Miller SH, Nellis SH, Buck D, Flaim SF,Graham WP 3rd, Davis TS: Effects of transient ischemia onnutrient flow and arteriovenous shunting in canine hindlimb.Ann Surg 193:255-263, 198143. Kirkpatrick AF, Derasari M: Transdermal clonidine: Treatingreflex sympathetic dystrophy. Reg Anesth 18:140-141, 199344. Koban M, Leis S, Schultze-Mosgau S, Birklein F: Tissuehypoxia in complex regional pain syndrome. Pain 104:149-157, 200345. Kurvers HA, Jacobs MJ, Beuk RJ, Van den Wildenberg FA,Kitslaar PJ, Slaaf DW, Reneman RS: Reflex sympathetic dystrophy:Evolution of microcirculatory disturbances in time.Pain 60:333-340, 199546. Laferrière A, Millecamps M, Xanthos DN, Xiao WH, Siau C,de Mos M, Sachot C, Ragavendran JV, Huygen FJ, Bennett GJ,Coderre TJ: Cutaneous tactile allodynia associated with microvasculardysfunction in muscle. Mol Pain 4:49, 200847. Lefer AM, Tsao PS, Lefer DJ, Ma XL: Role of endothelialdysfunction in the pathogenesis of reperfusion injury aftermyocardial ischemia. FASEB J 5:2029-2034, 199148. Leis S, Weber M, Isselmann A, Schmelz M, Birklein F:Substance-P-induced protein extravasation is bilaterallyincreased in complex regional pain syndrome. Exp Neurol183:197-204, 200349. Leriche R: La chirurgie de la douleur, 1st ed. Paris, FR,Masson et Cie, 193750. Livingston WK: Pain Mechanisms, 1st ed. New York, NY,Macmillan, 194351. Lorento de Nó R: Analysis of the activity of chains ofinternuncial neurons. J Neurophysiol 1:207-244, 193852. MacKenzie J: Some points bearing on the associationof sensory disorders and visceral disease. Brain16;321-354,189353. Mailis A, Furlan A: Sympathectomy for neuropathicpain. Cochrane Database Syst Rev CD002918, 200354. Maleki J, LeBel AA, Bennett GJ, Schwartzman RJ: Patternsof spread in Complex Regional Pain Syndrome, Type I(Reflex Sympathetic Dystrophy). Pain 88:259-266, 2000

12 The Journal of Pain Historical Reflection on Syndrome Names for CRPSdirect mechanical trauma applied to leg: In vivo study in ananimal model. J Trauma 51:508-517, 200189. Teasell RW, Arnold JM: Alpha-1 adrenoceptor hyperresponsivenessin three neuropathic pain states: Complex regionalpain syndrome 1, diabetic peripheral neuropathicpain and central pain states following spinal cord injury.Pain Res Manage 9:89-97, 200497. Whitesides TE, Haney TC, Morimoto K, Harada H: Tissuepressure measurements as a determinant for the need of fasciotomy.Clin Orthop Relat Res 113:43-51, 197598. Xanthos DN, Bennett GJ, Coderre TJ: Norepinephrineinducednociception and vasoconstrictor hypersensitivityin rats with chronic post-ischemia pain. Pain 137:640-651,200890. van der Laan L, ter Laak HJ, Gabreels-Festen A,Gabreels F, Goris RJA: Complex regional pain syndrometype I (RSD): Pathology of skeletal muscle and peripheralnerve. Neurology 51:20-25, 199891. Veldman PH, Reyen HM, Arnt IE, Goris RJ: Signs andsymptoms of reflex sympathetic dystrophy: Prospectivestudy of 829 patients. Lancet 342:1012-1016, 199392. Vollmar B, Westermann S, Menger MD: Microvascularresponse to compartment syndrome-like external pressureelevation: An in vivo fluorescence microscopic study in thehamster striated muscle. J Trauma 46:91-96, 199993. Wall PD, Woolf CJ: Muscle but not cutaneous C-afferentinput produces prolonged increases in the excitability of theflexion reflex in the rat. J Physiol (Lond) 356:443-458, 198494. Wang WZ, Anderson G, Fleming JT, Peter FW,Franken RJ, Acland RD, Barker J: Lack of nitric oxide contributesto vasospasm during ischemia/reperfusion injury. PlastReconstr Surg 99:1099-1108, 199795. Wasner G, Heckmann K, Maier C, Baron R: Vascular abnormalitiesin acute reflex sympathetic dystrophy (CRPS I):Complete inhibition of sympathetic nerve activity withrecovery. Arch Neurol 56:613-620, 199996. Weber M, Birklein F, Neundorfer B, Schmelz M: Facilitatedneurogenic inflammation in complex regional painsyndrome. Pain 91:251-257, 200199. Xanthos DN, Coderre TJ: Sympathetic vasoconstrictorantagonism and vasodilatation relieve mechanical allodyniain rats with chronic postischemia pain. J Pain 9:423-433, 2008100. Yan SF, Tritto I, Pinsky D, Liao H, Huang J, Fuller G,Brett J, May L, Stern D: Induction of interleukin 6 (IL-6)by hypoxia in vascular cells. Central role of the bindingsite for nuclear factor-IL-6. J Biol Chem 270:11463-11471, 1995101. Zhang L, Bail H, Mittlmeier T, Haas NP, Schaser KD: Immediatemicrocirculatory derangements in skeletal muscleand periosteum after closed tibial fracture. J Trauma 54:979-985, 2003102. Zollinger PE, Tuinebreijer WE, Breederveld RS,Kreis RW: Can vitamin C prevent complex regional pain syndromein patients with wrist fractures? A randomized, controlled,multicenter dose-response study. J Bone Joint SurgAm 89:1424-1431, 2007103. Zollinger PE, Tuinebreijer WE, Kreis RW,Breederveld RS: Effect of vitamin C on frequency of reflexsympathetic dystrophy in wrist fractures: A randomised trial.Lancet 354:2025-2028, 1999104. Zuurmond WW, Langendijk PN, Bezemer PD, Brink HE,de Lange JJ, van Loenen AC: Treatment of acute reflex sympatheticdystrophy with DMSO 50% in a fatty cream. ActaAnaesthesiol Scand 40:364-367, 1996