Home

Current Issue

Archives

Guidelines for Authors

Classified Ads

Links

Search PubMed

Subscriptions

Subscriber Registration

Guidelines for Website Users

JRheum Update Service

Contact Info

Correspondence

Identification of Radiologic Healing Phenomena in Patients with Rheumatoid Arthritis

To the Editor:

Professor Rau and his colleagues deserve credit for bringing to our attention several cases of rheumatoid arthritis (RA) in which serial radiographic examination has provided evidence that some improvement in erosive bone damage may occur¹. Although not the first to report healing in RA, Dr. Rau has over the last several years reported several such cases. Unfortunately, reproduction of radiographic film images almost always results in some loss of detail and some of the cases illustrated in the literature in the reports of Rau and others are not convincing. However, enough are convincing, so that the possibility of stabilization and some degree of improvement in bony erosions must be taken into account in all followup studies on RA that include serial radiographic study. The question of terminology and whether the term "healing" should be used is controversial and will be discussed in some detail in a forthcoming supplement to The Journal reporting the deliberations of the subcommittee on healing of erosions in RA held at OMERACT 6, Brisbane, Australia, April 2002.

In the article by Rau I would like to call attention to two errors. On page 2609, Rau, et al state "Some authors exclude the possibility of a score reduction expressis verbis ('once an erosion, always an erosion,' Sharp, personal communication)."¹ On the same page they state ". . . in most clinical trials scoring . . . was done knowing the chronological order . . .". To my knowledge I have never made such a statement. From the earliest time when I began the use of radiographic films to evaluate RA joint damage I have pointed out that it is important that even though we did not have evidence at that time that bone damage could improve, we should keep in mind that effective treatment should lead to repair of bony injury. For that reason, after the original report I have always randomized and blinded sequence of serial films in every study I have organized.

JOHN T. SHARP, MD, Affiliate Professor of Medicine, University of Washington School of Medicine, Seattle, Washington, USA.

REFERENCE

1. Rau R, Wassenberg S, Herborn G, Perschel WT, Freitag G. Identification of radiologic healing phenomena in patients with rheumatoid arthritis. J Rheumatol 2001;28:2608-15.

Dr. Rau replies

To the Editor:

We thank Dr. John Sharp for his thoughtful comments to our article¹. Dr. Sharp is completely right that the reproduction of radiographic films almost always results in loss of detail and therefore is less convincing than the original films. Despite intensive efforts to optimize the reproduction using technical advances in processing images, the result is often disappointing.

On the other hand we have published cases followed over many years showing progressive destruction and — later — increasing repair in the same joint, thereby overcoming problems of reproduction and changing projection (Figures 1, 5, 6, published in 1996²). It is not relevant if these changes are called "healing" or "repair." Our study¹ displayed some case reports only for the purposes of illustration. It was a formal study in which 24 cases with and 10 cases without healing were read by 3 observers blinded to sequence. They had to state which of the films was first and which was second and decide if there was healing or not. The agreement regarding both questions was approximately 90%.

The only difference from the study performed by the above mentioned OMERACT subcommittee on healing (perfectly organized by Dr. Sharp) was that the readers had not only digitized images of the one but also the original radiographs of hands, wrists, and feet at 2 time points, making it much easier to find the right sequence of the films. Consequently, the agreement among readers of our study was better than that of the OMERACT trial. Moreover, it might have been also easier to identify healing phenomena.

We have to apologize for offering the impression that Drs. Sharp and Larsen expressis verbis said "once an erosion, always an erosion." This statement was cited from a publication by van der Heijde³. Drs. Sharp and Larsen were mentioned in a second bracket, which should indicate that they were or still are very skeptical to accept the idea of improvement of radiographic findings or healing. This impression is documented again in Dr. Sharp's present letter. We admire the farsightedness of Dr. Sharp to read serial radiographs blinded to sequence in studies he organized. However, the vast majority of trials we are aware of were read with known sequence.

We also appreciate the great job he did by organizing the above mentioned OMERACT subcommittee study, which has demonstrated again that healing or repair is a reproducible phenomenon. Results like that are most convincing if they are reported by authorities who originally were most skeptical and did not expect that result.

ROLF RAU, MD, PhD, Evangelisches Fachkrankenhaus Ratingen, Ratingen, Germany.

REFERENCES

1. Rau R, Wassenberg S, Herborn G, Perschel WT, Freitag G. Identification of radiologic healing phenomena in patients with rheumatoid arthritis. J Rheumatol 2001;28:2608-15.

2. Rau R, Herborn G. Healing phenomena of erosive changes in rheumatoid arthritis patients undergoing disease-modifying antirheumatic drug therapy. Arthritis Rheum 1996;39:162-8

3. van der Heijde DMFM. How to read radiographs according to the Sharp/van der Heijde method. J Rheumatol 1999;26:743-5.

How Many Angels Could Dance on the Head of a Pin?

To the Editor:

Medieval theologians argued how many angels could dance on the head of a pin. Contemporary rheumatologists seem to be repeating this futility by arguing about fibromyalgia (FM). The exchange of letters^1,2 in the Journal in February refers to FM as a disease. It isn't. It is merely a portion of the spectrum of diffuse chronic pain, and its definition and boundaries are so variable that it has lost any claim to consideration as a diagnostic entity. As Werle, et al³ point out, the specificity of antibodies against serotonin and phospholipids is questionable, and I would maintain that neither the clinical features nor the laboratory findings can be used to classify or to ascribe etiology. It is high time we abandoned the misleading diagnosis "FM," which support groups and other interested parties have distorted, recognize that the classification criteria⁴ merely assure that series are comparable but have no diagnostic significance, and help the patients abandon victimhood and get on with their lives.

GEORGE E. EHRLICH, MD, Philadelphia, Pennsylvania, USA.

REFERENCES

1. Klein R, Berg PA. Diagnostic relevance of antibodies to serotonin and phospholipids in fibromyalgia syndrome [letter]. J Rheumatol 2002;29:395-6.

2. Klein R, Berg PA. Diagnostic relevance of antibodies to serotonin and phospholipids in fibromyalgia syndrome [letter]. Eich W, Werle E, Mueller A. [Reply]. J Rheumatol 2002;29:395.

3. Werle E, Fischer HP, Mueller A, et al. Antibodies against serotonin have no diagnostic relevance in patients with fibromyalgia syndrome. J Rheumatol 2001;28:595-600.

4. Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 1990;33:160-72.

Drs. Klein and Berg reply

To the Editor:

We appreciate the sophisticated statement by Dr. Ehrlich concerning the problem of defining fibromyalgia (FM) as a disease. However, the author of this letter has probably not carefully read our reply¹ to the article by Werle, et al². From our arguing it is obvious that, first, we never classified FM as a single disease; second, we referred to other authors who defined FM as belonging to the "functional somatic syndromes"³; and third, we interpreted our serological data about the occurrence of antibodies to serotonin, gangliosides, and phospholipids in patients with FM as an indicator for its "heterogeneity" and outlined that the presence of these antibodies in different clinical manifestations is a further argument that FM may belong to the spectrum of functional somatic syndromes. There are quite a few reports that indicate that features suggesting an alteration of the sympathic adrenergic and sensory nerve system as well as of immunological functions can be found quite frequently in patients with FM or other functional somatic syndromes^4-6. The underlying pathogenic mechanisms in FM may, therefore, also relate to a disturbance in the neuroendocrine immune network⁶.

Although FM has been primarily defined by generalized pain and psychological alterations, we should not worry too much how closely FM criteria are related to the real clinical condition reflecting the manifold above mentioned psychoneuroimmunological disturbances frequently occurring in this disease. Indeed, we should rather listen carefully with an open mind to these patients, hoping to improve our understanding of this still badly defined and multifaceted syndrome.

REINHILD KLEIN, MD; PETER A. BERG, MD, PhD, Department of Internal Medicine II, University of Tübingen, Tübingen, Germany.

REFERENCES

1. Klein R, Berg PA. Diagnostic relevance of antibodies to serotonin and phospholipids in fibromyalgia syndrome [letter]. Eich W, Werle E, Müller A. [Reply]. J Rheumatol 2002;29:395-6.

2. Werle E, Fischer HP, Mueller A, Fiehn W, Eich W. Antibodies against serotonin have no diagnostic relevance in patients with fibromyalgia syndrome. J Rheumatol 2001;28:595-600.

3. Barsky AJ, Borus JF. Functional somatic syndromes. Ann Intern Med 1999;130:910-21.

4. Clauw DJ, Chrousos GP. Chronic pain and fatigue syndromes: overlapping clinical and neuroendocrine features and potential pathogenic mechanisms. Neuroimmunomodulation 1997;4:134-53.

5. Elenkov IJ, Wilder RL, Chrousos GP, Vizi ES. The sympathetic nerve — an integrative interface between two supersystems: the brain and the immune system. Pharmacol Rev 2000;52:595-638.

6. Demitrack MA, Crofford LJ. Evidence for and pathophysiologic implications of hypothalamic-pituitary-adrenal axis dysregulation in fibromyalgia and chronic fatigue syndrome. Ann NY Acad Sci 1998;14:684-97.

Drs. Werle and Eich reply

To the Editor:

How many angels could dance on the head of a pin? The scholastics, medieval philosophers of the Catholic Church, like Thomas Aquinas could have very reasonably focused on this question, for it does concentrate several of their points of dispute, including whether angels have a corporeal (bodily) or merely spiritual existence. These theologians of the Middle Ages had no idea what sort of thing an angel was. What is knowledge? There was a time when everyone knew what a demon was, or that the world is composed of 4 elements. To answer this ridiculous sounding question about dancing angels on the top of a pin one should have enough information about angels, and we must have more data to answer today's scientific topic — fibromyalgia.

Therefore, in our recent article¹, we reevaluated repeatedly published data on laboratory tests for the diagnosis of FM. In his letter, however, Dr. Ehrlich argues against FM as a disease entity and strongly recommends we abandon this misleading diagnosis. His opinion may be the conclusion from his own experience and from more than 100 articles published in 2001 in peer reviewed journals with "fibromyalgia" as a key word in the title. In most of the recent papers the American College of Rheumatology criteria² were used to define the patient groups under investigation. The fact is that chronic pain patients exist and that tender points reflect a decreased pain threshold. The nature and the etiology of these phenomena are discussed controversially. Several authors assume that these patients may suffer, besides a chronic pain disorder, from a depression or an anxiety disorder or a somatization disorder. Others authors think of neuroplasticity.

The aim of our study¹ was to investigate whether or not the clinical symptoms in this patient group may be related to antibodies as reported. In our large cohort of patients recruited for integrated psychological and physical group therapy, we demonstrated that measurement of antibodies against serotonin was not associated with clinical scoring of, for example, pain intensity, depression, or activities of daily living. In addition, antibodies against serotonin showed no diagnostic relevance in these patients at all. Therefore, we only can refuse those unnecessary, even expensive, measurements for routine diagnostics, thus avoiding a fixation on the laboratory data of these patients.

Taking account of these reports we cannot definitely answer the question whether there are angels, or whether FM is a dispensable term, but we are quite sure that serial measurements of the antibodies we investigated have no diagnostic relevance and may, as suggested by Dr. Ehrlich, even have some psychological adverse effects on our patients' lives.

EGON WERLE, MD, PhD; WOLFGANG EICH, MD, PhD, University of Heidelberg, Heidelberg, Germany.

REFERENCES

1. Werle E, Fischer HP, Mueller A, Fiehn W, Eich W. Antibodies against serotonin have no diagnostic relevance in patients with fibromyalgia syndrome. J Rheumatol 2001;28:595-600.

2. Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 1990;33:160-72.

Polymorphism in the Matrix Metalloproteinase-1 Promoter Gene and Severity of Rheumatoid Arthritis

To the Editor:

In a recent report Constantin, et al¹ found no association between collagenase-1 (MMP-1) gene polymorphism and susceptibility to or severity of rheumatoid arthritis (RA). We agree that MMP-1 polymorphism is not involved in RA susceptibility, but have some doubts about the relationship between MMP-1 polymorphism and RA severity.

The search for prognostic markers for early RA is a very important issue. RA is characterized by a variable clinical course, with a poorer prognosis associated with the presence of erosive damage to joints, which is a very early feature of the disease². Therefore, the challenge for the physician is to predict, as early as possible, which patients will have a more disabling course necessitating an aggressive therapy. Identification of RA severity markers is thus urgently required to guide treatment strategies, in particular to avoid overtreating patients who will respond to cheaper and less toxic conventional therapies.

Although a polygenic component in susceptibility and severity of RA is very likely, the bulk of the genetic component is unknown, except for sex and HLA-DRB1 genes that confer susceptibility^1,3.

A strong candidate gene for RA is the MMP-1, since a variant (2G) has been identified in its promoter region, associated with increased transcription and hence with a more aggressive matrix degradation⁴, and higher circulating MMP-1 levels have been associated with rapidly progressive erosive RA⁵.

We studied polymorphism in the MMP-1 gene promoter in 56 patients with RA (41 women, 15 men, mean ± SD ages 49 ± 16 yrs), all fulfilling the American College of Rheumatology criteria⁶. Patients had symptoms for not more than 6 months at the time of the study and did not take any medication except nonsteroidal antiinflammatory drugs. One hundred sixty-four sex and age matched subjects with no rheumatic complaints served as controls. Patients and controls gave written informed consent. Patients' clinical assessment included counts of tender/swollen joints, measurement of erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), serum rheumatoid factor (RF), and radiographs of the hands. A single observer blinded to clinical features evaluated the presence/number of erosions. Joint erosions were selected as an accepted measure of joint damage⁷.

In accord with Constantin, et al¹, MMP-1 promoter gene did not contribute to the RA susceptibility (Table 1); but in contrast with their findings, MMP-1 promoter gene polymorphism was significantly associated with erosive RA. Erosive disease was observed in 7 patients, all of whom carried the 2G allele (Table 2). The number of involved joints was significantly higher in the 2G/2G (3 to 21 joints) and 1G/2G (1 to 18 joints) genotypes than in the 1G/1G genotype (1 to 10 joints; p = 0.015 and p = 0.027, respectively). The 2G allele mutation was observed in 12 of the 14 total alleles of patients with erosive RA (86%) and in 45 of the 98 total alleles of those with nonerosive RA (46%) (chi-square 6.252, p = 0.013, OR 7.067, 95% CI 1.382-48.243). No relationship was observed between MMP-1 polymorphism, ESR, CRP, or the presence or titer of RF.

Table 1. Frequency distribution of MMP-1 genotypes in patients and controls.

Table 2. Frequency distribution of MMP-1 genotypes in patients with erosive and nonerosive RA.

The reason for the discrepancy between our results and Constantin's data is not clear. Since the control groups had similar frequency distribution of MMP-1 genotypes, ethnic differences should be ruled out. However, if treatment with disease modifying drugs might have altered the progression of the disease in Constantin's patients, the 6 month period for symptoms might be insufficient for conclusive results. More studies are needed to evaluate whether MMP-1 polymorphism can be considered a reliable prognostic marker of erosive RA.

MARCO MASSAROTTI, MD, Department of Medicine, Surgery and Dentistry, S. Paolo Hospital; ANTONIO MARCHESONI, MD, Orthopedic Institute G. Pini; MARIA LUISA BIONDI, MD, Clinical Chemistry Laboratory; BIANCA MARASINI, MD, Department of Medicine, Surgery and Dentistry, S. Paolo Hospital, University of Milan, via Di Rudini' 8, Milan 20142, Italy.

REFERENCES

1. Constantin A, Lauwers-Cancès V, Navaux F, et al. Collagenase-1 (MMP-1) and HLA-DRB1 gene polymorphisms in rheumatoid arthritis: a prospective longitudinal study. J Rheumatol 2002; 29:15-20.

2. Drossaers-Bakker KW, deBuck M, van Zeben D, Zwinderman AH, Breedveld FC, Hazes JM. Long-term course and outcome of functional capacity in rheumatoid arthritis: the effects of disease activity and radiologic damage over time. Arthritis Rheum 1999;42:1854-60.

3. Wordsworth BP, Bell J. Polygenic susceptibility in rheumatoid arthritis. Ann Rheum Dis 1991;50:343-6.

4. Rutter JL, Mitchell TI, Buttice G, et al. A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription. Cancer Res 1998;58:5321-5.

5. Cunnane G, FitzGerald O, Beeton C, Cawston TE, Bresnihan B. Early joint erosions and serum levels of matrix metalloproteinase 1, matrix metalloproteinase 3, and tissue inhibitor of metalloproteinases 1 in rheumatoid arthritis. Arthritis Rheum 2001;44:2263-74.

6. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.

7. Van der Heijde D, Boers M, Lassere M. Methodological issues in radiographic scoring methods in rheumatoid arthritis. J Rheumatol 1999;26:726-30.

Dr. Constantin, et al reply

To the Editor:

Dr. Massarotti and colleagues report an association between a MMP-1 gene promoter polymorphism and severity of rheumatoid arthritis (RA), whereas we found no such association in our study¹. Massarotti used a qualitative approach of RA severity by classifying patients with early RA in erosive or nonerosive disease groups (assessed on radiograph of the hands only), whereas we used a quantitative approach by quantifying RA severity on the basis of a validated radiographic damage score (calculated from radiographs of hands and feet according to the Sharp/van der Heijde method²).

Since this methodological difference may account for the discrepancy between these 2 studies, we performed a complementary analysis of our data using the same approach as Massarotti, et al. Patients were classified in the erosive disease group if the joint erosion score was ³ 1 and in the nonerosive group if the joint erosion score equalled 0. Using this qualitative approach, we found no association between MMP-1 gene polymorphism and severity of RA, neither at inclusion of patients in our prospective longitudinal study (Table 1) nor after 4 years of followup (Table 2).

Table 1. MMP1 genotypes in patients with erosive (n = 47) and nonerosive (n = 55) RA at inclusion.

Table 2. MMP1 genotypes in patients with erosive (n = 74) and nonerosive (n = 21) RA after 4 years of followup.

Thus, a methodological difference in the assessment of RA severity could not account for the discrepancy between the report of Massarotti and colleagues and our own. Further studies are needed on the value of MMP-1 gene polymorphism as a marker of RA severity.

ARNAUD CONSTANTIN, MD, Department of Rheumatology, CHU Rangueil, 1 ave Jean Poulhès, 31403 Toulouse Cedex 4; VALÉRIE LAUWERS-CANCÈS, MD; ANNE CAMBON-THOMSEN, MD, INSERM, Unité 558; ALAIN CANTAGREL, MD, INSERM, Unité 395, Toulouse, France.

REFERENCES

1. Constantin A, Lauwers-Cancès V, Navaux F, et al. Collagenase-1 (MMP-1) and HLA-DRB1 gene polymorphisms in rheumatoid arthritis: a prospective longitudinal study. J Rheumatol 2002; 29:15-20.

2. van der Heijde DMFM. How to read radiographs according to the Sharp/van der Heijde method. J Rheumatol 1999;26:743-5.

Dermatomyositis with Normal Creatine Kinase and Elevated Aldolase Levels

To the Editor:

I read with interest the report by Carter, et al¹, and I would like to describe a similar case seen in our hospital.

A 24-year-old man presented in November 2001 with a 3 month history of arthralgias, Raynaud's phenomenon, and rash on his face. On examination, he had a heliotrope rash with associated edema of the upper eyelids, Gottron's papules on metacarpophalangeal, proximal phalangeal, and elbow joints, and digital ulcers. Manual testing of muscle strength revealed minimal proximal muscle weakness of upper and lower extremities. Laboratory evaluation showed an elevation of aspartate aminotransferase (AST) of 203 IU/l, and creatine kinase (CK) was normal (57 IU/l). Anti-dsDNA, human immunodeficiency virus, and renal function tests were negative or normal. Chest radiograph was normal. A muscle biopsy specimen from the right deltoid showed perivascular inflammation, without necrobiotic degeneration or atrophy of the muscle fibers.

He was treated with oral prednisone and vasodilators. On December 10, 2001, AST was 87 IU/l, alanine aminotransferase (ALT) was 69 IU/l, and CK was 59 IU/l. On January 14, 2002, AST was 60 IU/l, ALT 44 IU/l, CK 33 IU/l, and aldolase level was 9.9 IU/l (normal 0-8.1 IU/l). On January 23, 2002, CK was 42 IU/l and aldolase was 11.5 IU/l. A few days later, he was admitted to our hospital because of a possible seizure. On examination, no focal neurologic deficit was found. A lumbar puncture was normal. Electrocardiogram and echocardiogram results were normal. He denied drug abuse. He was discharged 48 h later to continue the prednisone therapy.

CK measurement in serum has remained the best marker for detection and monitoring of inflammatory disease of skeletal muscle. However, of the 3 widely used muscle enzyme measurements (AST, CK, and aldolase), any one may be normal even in an active disease state. Therefore, we recommend that all 3 enzyme tests be performed during evaluation of a patient with myositis.

ULISES MERCADO, MD, MS, FACR; Hospital General Mexicali and Universidad Autonoma de Baja California, Mexicali, México.

REFERENCE

1. Carter JD, Kanik KS, Vasey FB, Valeriano-Marcet J. Dermatomyositis with normal creatine kinase and elevated aldolase levels. J Rheumatol 2001;28:2366-7.

Dr. Carter, et al reply

To the Editor:

The literature maintains that as many as 10% of patients with active myositis can present with normal creatine kinase (CK) levels^1,2. Dr. Mercado describes another patient who meets this clinical scenario, i.e., a patient with biopsy proven dermatomyositis who was found to have normal CK concentrations with elevated serum aspartate aminotransferase (AST) and aldolase. He correctly concludes that we must rely on all of the available muscle enzyme results in those patients who have normal CK levels in spite of active myositis. Aldolase is the most specific muscle enzyme after CK.

Clinicians are quick to disregard an abnormal laboratory value if it does not fit the clinical scenario. This reaction is completely justified. A perfect example of this is a patient who presents with mechanical low back pain and a weakly positive antinuclear antibody. A diagnosis of systemic lupus erythematosus is inappropriate in this situation. The clinical picture guides the diagnosis, not a single laboratory test. The converse should also hold true. If your clinical suspicion of a certain disease is very high, a single incongruent laboratory test should not dissuade your opinion. In this instance, it is myositis with a normal serum CK. We must search for other markers of disease activity, i.e., aldolase, AST, lactate dehydrogenase. It is imperative that we use diligence when our laboratory testing does not match our clinical suspicion. If, after reevaluation, our hypothesis remains the same, it is our duty to find other ways to confirm our suspicion.

JOHN D. CARTER, MD; JOANNE VALERIANO, MD; FRANK B. VASEY, MD, Division of Rheumatology, University of South Florida, 12901 Bruce B. Downs Blvd., MDC 81, Tampa, Florida 33612, USA.

REFERENCES

1. Bohan A, Peter JB, Bowman RL, Pearson CM. A computer-assisted analysis of 153 patients with polymyositis and dermatomyositis. Medicine (Baltimore) 1977;56:255-86.

2. Vignos PJ, Goldwyn J. Evaluation of laboratory tests in diagnosis and management of polymyositis. Am J Med Sci 1972; 263:291-308.

Chronic Nonmalignant Pain

To the Editor:

In a recent research report, Mailis, et al¹ investigated nondermatomal somatosensory deficits (NDSD) (nonanatomical sensory deficits) in patients with chronic nonmalignant pain seen for an independent medical examination. They found that 25.3% of the patients had NDSD. In addition, most interestingly, Mailis, et al reported that these patients were more likely to have their pain lateralized to one side of the body or worse on one side of the body.

Our group has performed some research that resonates with the data in the report from Mailis, et al. First, we reported that 37.8% of patients with chronic pain demonstrated NDSD². Second, in a followup study we reported in abstract form³ on the relationship between pain location and location of NDSD. We found that there was a strong statistical relationship between location of NDSD and pain location (study in press)⁴. Third, NDSD are important because their presence to some physicians indicates the possible presence of either malingering or conversion disorder⁵. We recently performed a structured evidence based review⁵ of studies addressing this issue. It was concluded that nonorganic findings, including NDSD, do not discriminate between organic and nonorganic problems and are not associated with secondary gain, which is a prerequisite for malingering. Overall, this group of studies, including that of Dr. Mailis and colleagues, indicate that NDSD are not psychological phenomena.

DAVID A. FISHBAIN, MD, FAPA, Professor of Psychiatry and Neurological Surgery and Anesthesiology, University of Miami School of Medicine, University of Miami Comprehensive Pain and Rehabilitation Center, 600 Alton Road, Miami, Florida 33139, USA.

REFERENCES

1. Mailis A, Papagapiou M, Umana M, et al. Unexplainable nondermatomal somatosensory deficits in patients with chronic nonmalignant pain in the context of litigation/compensation: a role for involvement of central factors? J Rheumatol 2001;28:1385-93.

2. Fishbain DA, Goldberg M, Meagher RB, et al. Male and female chronic pain patients categorized by DSM-III psychiatric diagnostic criteria. Pain 1986;26:181-97.

3. Fishbain DA, Goldberg M, Ferretti T, et al. The non-dermatomal sensory abnormality (NDSA) and pain perception [abstract]. Pain 1990;S3:A637,S332.

4. Fishbain DA, Cutler B, Rosomoff HL, et al. Is the location of nondermatomal sensory abnormalities (NDSAs) related to pain location? Pain Medicine 2002; (in press).

5. Fishbain DA, Cutler B, Rosomoff HL, et al. A structured evidence based review on the meaning of non-organic physical signs: Waddell signs. Pain Medicine 2002; (in press).

Drs. Mailis and Nicholson reply

To the Editor:

We were interested in Dr. Fishbain's response to our recent report of nondermatomal somatosensory deficits (NDSD) in patients with chronic nonmalignant pain¹. Several groups have now reported on this phenomenon, which appears to be both prevalent and important. We view the presence of NDSD as a poor prognostic factor associated with a number of other problems. Dr. Fishbain makes reference to an article in press in which a structured based review of the literature concluded that "nonorganic findings, including NDSD, do not discriminate between organic and nonorganic problems...." Unfortunately, we cannot discern what this statement might mean and await publication of the report. He also stated that NDSD are not associated with malingering and, primarily upon this finding, it is concluded that NDSD are "not psychological phenomena."

Although a somewhat crude analogy, this sort of statement (which appears not uncommonly in the literature) is akin to stating that because a person has arms and legs, there is no reason to believe that there are any biomedical factors involved in presentation of some problem such as stroke or cancer. In other words, one negative instance (for example, the absence of malingering) is used to rule out an entire class of events (i.e., all possible psychological, psychosocial, or personality factors). We would here caution about mind-body dualism and consider that psychological phenomena of any description also have an organic basis. We strongly suspect that psychological factors are involved in the presentation of NDSD in the context of chronic pain (as well as possible conversion disorders involving sensory deficits independent from chronic pain).

There are several lines of support for this. We have found that certain demographic variables distinguish between chronic pain patients with or without NDSD, i.e., non-Canadian born patients were more likely to have sensory deficits. This may be interpreted as indicating that culture (i.e., the way we view the world) influences the appearance of these deficits. Further, the NDSD subgroup had significantly abnormal pain behaviors, discrepancy between supine and sitting straight leg raising (i.e., behaviors under confrontation and distraction), and much higher levels of disability, as they were all virtually unemployed, as compared with 31% of the non-NDSD group who were employed. Further, Fishbain, et al² demonstrated that in their study of 247 chronic pain patients, all diagnosed with myofascial pain, 77% of the NDSD subgroup had a workers' compensation claim, while in the group without NDSD, only 40% had such a claim. Could these data be interpreted as showing that the stress associated with a workers' compensation claim or perhaps unusual and therefore contested disability contribute to the generation of NDSD? In the same study 27% of the NDSD group had dependent personality disorder, as compared to 12.9% of the non-NDSD group (p < 0.01).

Other groups, as well, have presented evidence that NDSD may be associated with psychological factors. Verdugo and Ochoa reported³ on a group of patients who responded to administration of placebo with either complete or near complete resolution of pain and sensory deficits. This was interpreted as the result of a psychogenic phenomenon.

We have also seen many patients responding to placebo interventions with resolution of their sensory deficits and pain, at times permanently⁴. To study these phenomena we instituted since 1994 in our unit placebo controlled infusions of sodium amytal in a standardized protocol on patients with chronic pain. Response to inert placebo with "shrinking" or disappearance of the NDSD borders has been seen in many patients, and clearly indicates that the deficits are not structural or anatomical (as in the case of neurectomy or other structural nervous system damage). On some occasions, however, NDSD can be superimposed on definite structural deficits⁵.

Our protocols have allowed us to collect a very large database, which is currently under analysis. Preliminary data indicate that personality and psychological factors are indeed associated with the onset, maintenance, severity, or exacerbation of chronic pain and with the presence of NDSD.

We also believe that there is an "organic" basis, or better said, a "psychobiological" or "psychophysiological" basis for these NDSD phenomena, as our current magnetic resonance imaging data unequivocally confirm⁶. Beyond our recently published report¹, our group's collective experience shows that NDSD phenomena occur in chronic pain patients (1) with significant psychoemotional factors in the absence of detectable pathology; and (2) superimposed on structural deficits. In the vast majority of cases in both groups, psychoemotional factors do seem to be associated with both chronic pain and the appearance of NDSD.

We would urge researchers not to adopt an either/or approach (i.e., either "organic" or "psychological") to such phenomena, but accept them as bridging the "mind-body interface." Ignoring psychological factors carries the risk of overmedicalizing treatment, while dismissing the phenomena as purely psychological, particularly in the presence of specific detectable peripheral pathology, ignores potential nociceptive or neuropathic sources that are treatable as well.

One contentious point involves patients with diffuse myofascial pain syndromes, often classified as fibromyalgia (FM). These patients do not have detectable peripheral pathology in the form of muscle inflammation, etc. Their proven excessive sensitivity to pressure is more likely the product of sensitization at supraspinal rather than peripheral levels, hence some consider the entity as a manifestation of "hypervigilance" due to attentional switches⁷. Kaziyama, et al⁸ reported a 38.2% prevalence of hemisensory deficits to pinprick in 76 women fulfilling the American College of Rheumatology criteria for FM⁹. Overall, their patients may not be different than those reported by Fishbain, et al² (with myofascial pain syndromes) or by Mailis, et al^1,10 (most with diffuse pains and the diagnosis of FM).

ANGELA MAILIS, MD, MSc, FRCPC(PhysMed), Director, Comprehensive Pain Program, Senior Investigator, Toronto Western Hospital Research Institute and Krembil Neuroscience Centre, 399 Bathurst Street, Toronto, Ontario N5T 2S8, Canada; KEITH NICHOLSON, PhD (Psychol), Toronto Western Hospital, University Health Network, Toronto, Ontario.

REFERENCES

1. Mailis A, Papagapiou M, Umana M, Cohodarevic T, Nowak J, Nicholson K. Unexplainable nondermatomal somatosensory deficits in patients with chronic nonmalignant pain in the context of litigation/compensation: a role for involvement of central factors? J Rheumatol 2001;28:1385-93.

2. Fishbain DA, Goldberg M, Steele Rosomoff R, et al. Chronic pain patients and the nonorganic physical signs of nondermatomal sensory abnormalities (NDSA). Psychosomatics 1991;32:294-303.

3. Verdugo JR, Ochoa JL. Reversal of hypoaesthesia by nerve block or placebo: a psychologically mediated sign in chronic pseudoneuropathic pain patients. J Neurol Neurosurg Psychiatry 1998;65:196-203.

4. Mailis A, Nicholson K. Effect of normal saline controlled intravenous administration of sodium amytal in patients with pain and unexplainable widespread non-anatomical sensory deficits: A preliminary report [abstract]. 16th Annual American Pain Society Meeting, New Orleans, LA, USA, Oct. 1997. American Pain Society 1997;709:138.

5. Cohodarevic T, Mailis A, Montanera W. Syringomyelia: pain, somatosensory abnormalities and neuroimaging. J Pain 2000; 1:54-66.

6. Mailis A, Downar J, Kwan C, et al. FMRI in unexplainable widespread somatosensory deficits (WSDs) in patients with chronic pain [abstract]. 19th Annual American Pain Society meeting, Atlanta, GA, USA, Nov. 2000. American Pain Society 2000;760:158.

7. McDermid AJ, Rollman GB, McCain GA. Generalized hypervigilance in fibromyalgia: Evidence of perceptual amplification. Pain 1996;66:133-44.

8. Kaziyama HHS, Texeira MJ, Lin TY, et al. Fibromyalgia and hemisensitive syndromes [abstract]. IASP 9th World Congress, 1999:550.

9. Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 1990;33:160-72.

10. Mailis A, Furlong W, Taylor A. Chronic pain in a family of 6 in the context of litigation. J Rheumatol 2000;27:1315-7.