Spinal cord injury pain

Spinal cord injury pain

European Journal of Pain 7 (2003) 335–338 www.EuropeanJournalPain.com Spinal cord injury pain Aleksandar Beric * NYU School of Medicine, Hospital f...

68KB Sizes 6 Downloads 95 Views

European Journal of Pain 7 (2003) 335–338 www.EuropeanJournalPain.com

Spinal cord injury pain Aleksandar Beric


NYU School of Medicine, Hospital for Joint Diseases, 301 East 17 Streeet, New York, NY 10003, USA Received 28 March 2003; accepted 8 April 2003

Abstract Awareness that SCI pain is common emerged during the past decade. However, there are a number of unresolved issues. There is a need for variety of experimental models to reflect diversity of SCI pains. Current classification is not as user-friendly as it should be. More attention should be given to a condition of the spinal cord below and above the SCI lesion. A consensus for what is an optimal SCI functional assessment for patients with sensory complaints and pain should be developed. Further extensive SCI pain research is needed prior to spinal cord regeneration trials in order to be able to cope with a potential for newly developed pains that may appear during incomplete spinal cord regenerative attempts. Ó 2003 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Science Ltd. All rights reserved. Keywords: Spinal cord injury; Pain; Central pain; Spinal cord; Spinal cord regeneration

One wonders why a larger impact on the understanding and treatment of spinal cord injury (SCI) pain has not been made during the last decade. Actually, contributions have been made, but the root cause of pain is still unknown and the efficacy of treatment in the SCI patient is not predictable. This is occurring as the Decade of the Brain and health politiciansÕ promises of curing neurologic diseases such as ParkinsonÕs disease has passed and left us with a need for new promises and big name actions. One thing has definitely happened that may help the SCI pain field move forward into this and the next decade; awareness that SCI pain exists, is common, and may be as devastating as other SCI consequences (Bonica, 1991; Yezierski and Burchiel, 2001). This is a prerequisite for any expansion of the field. Another important issue that has evolved is a need for classification (Beric, 1997; Siddall et al., 1997). In order to further this field types of pain need to be categorized, their frequency and how to interpret the results of pain treatment in light of different types of SCI pain. As an additional consequence of SCI pain awareness there are now a number of experimental models that were


Tel.: +1-212-598-6185; fax: +1-212-598-6009. E-mail address: [email protected]

specifically developed to tackle this problem (Hao et al., 1991; Vierck and Light, 1999; Yezierski et al., 1998). A breakthrough in SCI pain management can happen in either two ways; through serendipity or as a consequence of systematic contributions and incremental understanding of the biological consequences of SCI and the mechanism of pain in general. More specifically, understanding neurogenic and central pain. However, the problem is that not a lot is known about SCI and even less is known regarding central pain. The field of SCI pain is intricately connected to both the SCI field and pain research, but to such an extent that both clinical and experimental researchers need to respect this complexity. SCI leads to a variety of different clinical presentations, from complete, incomplete, with sacral sparing, without sacral sparing, more sensory involvement, more motor involvement, mainly upper motor neuron (UMN) features, combined UMN and lower motor neuron (LMN) features, etc. In addition, cauda equina injury and pain is a separate root/ peripheral nerve problem. Most of these different presentations can be assessed by American Spinal Injury Association (ASIA) standard exam, although clinical researchers studying SCI pain have not fully taken this approach. It is less clear how basic researchers are handling these complexities. To complicate things even

1090-3801/03/$30 Ó 2003 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Science Ltd. All rights reserved. doi:10.1016/S1090-3801(03)00045-4


A. Beric / European Journal of Pain 7 (2003) 335–338

further, basic researchers lack central pain models. However, peripheral nerve models are well established. Since a majority of SCI pain is either nociceptive or caused by peripheral nerve dysfunction, what is known in the periphery should, therefore, be applied to SCI pain. Only a minority of SCI pain is central. The problem of the dissociation of clinical presentations of SCI and SCI pain research is further exemplified by the active suppression of a syndromatological approach and the devising of models that are marginally relevant to the clinical problem. There are no known naturally occurring surgical, chemical or physical rat or monkey SCIs. The only point for these models would be if they were relevant to human SCI since they have no veterinary value. There appears to be some difference between spinal cord and more central occurring lesions, as allodynia is less common in SCI in the chronic stage compared to post-stroke pain. Cold allodynia is extremely rare in SCI pain below the lesion. On the other hand, allodynia and hyperalgesia of every type is a hallmark of transitionalat level-pains. Is transitional zone pain a proximal root and rootlet injury or is it a genuine spinal cord injury pain? To complicate matters further, are both root and cord involved? Therefore, it is difficult to interpret the relevance of allodynic-hyperalgesic models to SCI pain. There should at least be a disclaimer regarding which clinical situation these models represent. Things are even more complicated with below level pains, especially at the conus-cauda equine level where there is no relevant animal model. This is where a good classification may be helpful. The operative word here is ‘‘good’’. At present, an elaborate classification in absence of basic understanding of SCI and pain in general is overkill. In order to construct a prospective treatment trial, or even a natural history study, more information is always better than less. More information should yield a detailed clinical evaluation with the ASIA exam as a beginning. The quality of the trials and especially their future value in comparing with future trials would be greatly increased with the appropriate neurophysiologic, radiological and urodynamic studies. We would know more about what is happening above, at and below the SCI. Quantitative sensory studies would be invaluable, despite the fact that they occasionally yield conflicting results. The conflict is mainly in the interpretation based on the aim of the study. Therefore, results can be used over-and-over again when comparing similar aspects in various studies when the difference in the aim is removed from the equation. The beauty of this extensive approach is that data is subject to both prospective study hypothesis analyses and post-hoc analyses. In the absence of well-collected original data there is no hope for viable comparative analysis or meaningful meta-analysis. It cannot be stressed enough that there is no SCI pain, there are only SCI pains (Beric, 1997). They are so

different that a proper comparison would be trying to explain migraine and post-puncture headache with one simple mechanism. Trying to reduce the number of SCI pain types artificially in order to improve statistics would most likely lead to erroneous conclusions. Over time it may became even a bigger problem if individual researchers lump together different pain types in various combinations. So is there a solution? The answer is most likely ‘‘yes’’, although it is not simple. One should be reluctant to classify SCI pain patients into any pre-study scheme, as there is currently no consensus as to the minimal and optimal assessment of patientsÕ clinical condition. This should be established at some level before insisting on a classification. It certainly is related, but the assessment will define the extent and depth of classification, not the other way around. The reason for this is that there is no etiologic assessment available, so one cannot start from that end. What may be needed is a classification of a descriptive nature, or neurologic nature, or psychosocial nature, and perhaps not even have a single classification. On the other hand if ASIA and IASP come to an agreement on a minimal and/or optimal assessment criteria, this can be used as a basis for a meaningful classification. It should not be forgotten that pain is only one aspect of SCI, even when the goal is SCI pain classification (Siddall et al., 2001). It is very different when classifying pain in general. SCI pain is, therefore, one group of pains, and does not go beyond that, or in any previous IASP classifications (Merskey and Bogduk, 1994), as it has not arisen from SCI research interest in pain classification. It may sound trivial, but the function of the cord below the level of the lesion is important; it may even be etiologically related to pains below the level of the lesion (Beric et al., 1992). If this is the case, it would be important for the SCI pain researcher to understand the expected outcome of sphincterotomy in quadriplegics, which is not common knowledge in the pain research community. The problem is that what is important in assessment and ancillary studies have never been addressed in the pain community. Perhaps bladder and sphincter function is irrelevant in SCI pain understanding and treatment. However, it may be important, so it should not be singled out as too detailed and irrelevant to the advent of this field. With hard data, this should hopefully be brought up between the SCI and pain fields. Until that time, the syndromatologic approach cannot be dismissed as it helps define some relatively well-recognized SCI and SCI pain presentations. Meanwhile, prospective studies can be carried out with these better defined subgroups in order to determine minimal required assessment to classify all, or a majority of SCI pains. This complex assessment approach to SCI pain research is incompatible with typical telephone pain

A. Beric / European Journal of Pain 7 (2003) 335–338

research. There is already reasonable information from studies about pain presence in SCI population. It is time to learn what these pains are and how to predictably help these patients. This would require patient examination and follow-up, as well as documentation of changes in pain and clinical-neurologic presentation, either as a natural progression or as a result of intervention if involved in therapeutic trials. It cannot be emphasized enough, but SCI pains change with time. There is also a change in the underlying neurologic condition, more frequently occurring early after the injury, but not exclusively, making examination an absolute necessity. For example, I have seen patients treated for pain who sustained SCI and never fully undressed during initial examination. Were these patients atrophic below the level of the lesion with signs of LMN, or not? Furthermore, was their bladder showing residual urine because of DSD or hyporeflexia? Was their sensory deficit present above the level of the original SCI? If a traditional clinical shot-gun approach to pain treatment is used, and clinicians go through their arsenal, it may not be important to understand the spinal cord condition of every patient. Hopefully, their problems with spasticity, bladder, bowels and dysreflexia would be taken into account when dispensing treatment. However, this approach is useless for the purpose of furthering the field of SCI pain, as there is no idea of the patientsÕ real spinal cord condition. Pain at the level of the lesion can be caused by root lesion, cord lesion, syringomyelia, muscle overuse, spine instability and so on. Therefore, neurologic and physical therapy exam, ASIA score, Xrays, MRI, sensory quantification, reflex assessment and bladder assessment is needed to include these patients into a potential database of treated and untreated SCI pain patients for further analysis. This approach would also facilitate multicenter studies even across continents, as the number of these well-studied patients may not be large enough. Another problem in SCI pain research is an almost theatrical negativistic approach which uses a single case or a small number of unrelated cases to describe the lack of consistency in SCI pains. The danger is that these anecdotal reports may get picked up and repeated in reviews, possibly becoming mainstream despite hundreds of other patients that show some pattern of abnormality or respond to some treatment. Special difficulties are treatment-resistant patients who underwent a series of different management approaches. Usually, the very last approach is used to prove something, ignoring the previous history of events leading to this final intervention. Although these patients may have a special value, they are not the bread and butter of SCI pain patients, and should not be mixed with patients naıve to invasive treatments who should be a mainstay of prospective trials. Only in conjunction with prospective trials of well-defined SCI patient groups can


these exceptional single cases be taken into proper perspective. Most clinicians have encountered these types of patients, but our goal is to understand the condition and find a common denominator and not to use a journalistic approach in science and become fatalistic. Despite the fact that the uninitiated usually consider SCI pain as central, actual presumed central pains in SCI are the minority (Beric, 1997). Therefore, we should at least exert an equal effort in investigating central versus other SCI pains if our intention is to help patients with SCI pain. The funding agencies should be aware of that too, so there should eventually be a balanced approach to funding clinical and experimental studies that would lead to the understanding and control of all pains in SCI. This, in turn, suggests a need for more experimental models and making every existing experimental model even more valuable. This may be in contradiction with the aforementioned critique of experimental models. Actually, it is not a contradiction, as little is known regarding different underlying causes of SCI pains, and evidence of clinical relevance is not fact-mechanism based, but symptom based. If the results of experimental research are carefully extrapolated to the clinical turf without speculative correlation, clinicians would be more receptive, and the combined effort may yield some fruits, sooner rather than later. There is great interest in spinal cord regeneration for the potential improvement, and even cure, of SCI. There are different approaches, but the essence of this procedure is bridging the injury with viable neurons that are supported by trophic factors. The role of stem cell research cannot be over-emphasized in this context. There are a number of issues to be addressed, including how to modify the lesion site so as not to interfere with nerve regeneration; how to deliver cells and trophic factors; how to guide trophic factors; and especially how to control regeneration. Given the assumption that these efforts lead to a complete and functional restoration of spinal cord function, concerns regarding implication of pain in this regeneratory effort should be minor. However, if the assumption is for different degrees of regeneration outcome, and that patients with different degrees of spinal cord lesion and dysfunction would be enrolled in clinical trials, it is inescapable that the trials will result in incomplete SCI lesions. SCI pain is frequently present in incomplete SCI patients; perhaps even more prevalent than in complete SCI patients. Development of pain and dysesthesia in these patients could be expected to be either temporary during regeneration, or continuous if the end result is an incomplete SCI profile, usually leading to SCI pain (i.e., anterior cord syndrome-like situation). The overall extent of SCI regeneration, and especially the degree of functional connections in different spinal cord pathways, will be difficult to predict. A recent example of this comes to mind regarding a rush into


A. Beric / European Journal of Pain 7 (2003) 335–338

clinical trials. Fetal cell transplant in ParkinsonÕs disease led to frequent, uncontrollable dyskinesias in the absence of medication (S. Fahn, unpublished, W. Olanow, unpublished). After successful transplants resulted in uncontrollable delivery of dopamine, the only solution was to perform a pallidotomy to abolish severe and resistant dyskinesias that developed in some patients. A promising neuroaugmentative procedure was followed by destructive brain surgery in order to control the sideeffects of the first surgery. If more were known about dyskinesias and their management and about ways to control dopamine release from fetal grafts, perhaps these trials may have had a greater chance to succeed. Hypothetically, a patient that developed pain and dysesthesias in future SCI regeneration trials would require cordotomy in order to control newly developed sensory symptoms. This would certainly lead to the permanent closure of otherwise promising trials. Therefore, it would be prudent to make a great effort in the understanding of pain and dysesthesia before stem cell and other trials are undertaken in order to control pain if it appears, allowing these patients to comfortably make it through regenerative processes. In summary, it would be helpful to develop, in parallel, both central and peripheral animal models pertinent to SCI. There should be a consensus for what is an optimal SCI functional assessment for patients with sensory complaints and pain. Classification should remain descriptive and detailed and include both pain and clinical findings at, above and below SCI. Prospective studies of the natural progression and treatment outcome of SCI pain should be developed in a well-defined

SCI patient population to find a common denominator. As a caveat one should be aware of exceptional case studies which are currently inexplicable. A final point is that a better understanding of pain and dysesthesias in incomplete SCI patients needs to be obtained before undertaking large-scale trials of regeneration. References Beric A, Dimitrijevic MR, Light JK. Pain in spinal cord injury with occult caudal lesions. Eur J Pain 1992;13:1–7. Beric A. Post-spinal cord injury pain states. Amesthesiol Clin North America 1997;15:445–63. Bonica JJ. Introduction:semantic, epidemiologic, and educational issues. In: Casey KL, editor. Pain and central nervous system disease: the central pain syndromes. New York: Raven Press; 1991. p. 13–29. Hao J-X, Xu X-J, Aldskogius H, Seiger A, Wiesenfeldt_Hallin Z. Allodynia-like effect in rat after ischemic spinal cord injury photochemically induced by laser irradiation. Pain 1991;45:175–85. Merskey H, Bogduk N, editors. Classification of chronic pain: descriptions of chronic pain syndromes and definitions of pain terms. Seattle: IASP Press; 1994. Siddall PJ, Taylor DA, Cousins MJ. Classification of pain following traumatic spinal cord injury. Spinal Cord 1997;35:69–75. Siddall PJ, Yezierski RP, Loeser JD. Taxonomy and epidemiology of spinal cord injury pain. In: Yezierski RP, Burchiel K, editors. Spinal cord injury pain: assessment, mechanisms, management. Seattle: IASP Press; 2001. p. 9–24. Vierck CJ, Light AR. Effects of combined hemotoxic and anterolateral spinal lesions on nociceptive sensitivity. Pain 1999;83:447–57. Yezierski RP, Liu S, Ruenes GL, Kajander KJ, Brewer KL. Excitotoxic spinal cord injury: behavioral and morphological characteristics of a central pain model. Pain 1998;75:141–55. Yezierski RP, Burchiel K, editors. Spinal Cord Injury Pain: Assessment, Mechanisms, Management. Seattle: IASP Press; 2001.