Dear Editor,

I was intrigued by the observations of Drs. Marasini, De Monti and Ghilardi on the risk factors for accelerated atherosclerosis in patients with systemic lupus erythematosus. Most interesting was the high percentage of ANA negative lupus patients they evaluated. Given the sensitivity of the Hep2 substrate, I would expect no more than one such patient in a total population of 48.

The fact that these patients were older and more likely to have comorbidities (hypertension), I suspect other less specific measures were used to make the diagnosis of SLE (proteiuria, cytopenias, arthritis, rash...). In my institution a negative ANA (using Hep2 substrate and Immunoflorecence Microscopy) is considered to make the diagnosis of SLE highly unlikely.

Possibly this study makes the case that the ARA criteria should be updated to require a positive ANA for the diagnosis of SLE.

Dear Editor,

We read with interest the case report by Denschlag et al. [1]. In addition, von Lilienfield-Toal et al. also recently described the presentation of adult onset Still’s disease associated with occult breast malignancy [2]. These case reports remind us that rheumatic symptoms may precede the diagnosis of an underlying neoplasm. We recently observed an atypical presentation of scleroderma associated with pseudomyxoma peritonei, a rare tumour of appendiceal or ovarian origin.

A 50-year old female presented with a two-year history of intermittent right iliac fossa pain and weight loss. Recent laparoscopy and extensive radiological investigations had identified no explanation. In addition she described new onset skin photosensitivity, Raynauds phenomenon and acute dyspnoea on exertion which was diagnosed as pleuro-pericardial serositis on high resolution computed tomography. Antinuclear antibodies were positive, titre 1:640 (speckled pattern). IgM anti-cardiolipin antibodies were positive (46 MPLU/ml) on one occasion only but IgG antibodies were negative as was lupus anticoagulant (she had a history of six first trimester miscarriages). Systemic lupus erythematosus was diagnosed and appropriate therapy commenced.

Five months later she felt skin tightness over her hands and complained of pruritus, indigestion and arthralgia. Clinical examination revealed scleroderma extending to the elbows and involving the neck and face, with digital ulcers and a nail bed vasculitic lesion. Skin biopsy was consistent with scleroderma. Just two months later, the patient developed a recurrence of the right iliac fossa pain now associated with bleeding per vagina. Ultrasound revealed a 16 weeks’ size pelvic mass of ovarian origin and she underwent surgery. Histological examination was consistent with pseudomyxoma peritonei. She is currently undergoing appropriate treatment for this tumour.

This tumour could have caused the initial presentation with weight loss and abdominal pain, and the close temporal relationship between the appearance of the tumour and the features initially of a lupus like disease and latterly of scleroderma suggest that they were paraneoplastic manifestations. It is now ten months since tumour resection and the cutaneous scleroderma is gradually improving.

Scleroderma has been previously reported as a paraneoplastic phenomenon coexisting most frequently with adenocarcinomas of lung, breast and stomach [3,4]. These patients were three times more likely to be female: all had cutaneous manifestations of scleroderma but less than half had systemic features [4]. We found no previous reports of scleroderma in association with pseudomyxoma peritonei although dermatomyositis has been reported preceding the diagnosis of this tumour by six months [5].

These collected case reports highlight that occult malignancies can present with musculoskeletal symptoms and the clinician should always be vigilant for these paraneoplastic manifestations.

References

1. Denschlag D, Reiner E, Vaith P, Tempfer C, Keck C. Palmar fasciitis and polyarthritis as a paraneoplastic syndrome associated with tubal carcinoma: a case report. Ann Rheum Dis 2004; 63: 1177-1178.

2. von Lilienfeld-Toal M, Merkelbach-Bruse S, Dumoulin FL. An unusual presentation of a common disease. Ann Rheum Dis;63:887-888.

3. Fam AG. Paraneoplastic rheumatic syndromes. Ballieres Clin Rheumatol 2000; 14: 515-33.

4. Caldwell DS, McCallum RM. Rheumatological manifestations of Cancer. Med Clin N Am 1986;70: 385-417.

5. Maekawa Y, Nakamura K, Nogami R. A Case of Dermatomyositis Associated with Pseudomyxoma Peritonei Originating from Mucinous Adenocarcinoma of the Appendix. J Dermatol 1992; 19: 420-3.

Dear Editor,

We read with interest the article by Kauppi, Barcelos and da Silva which discussed the importance of the specific evauation of the cervical spine in the patients with rheumatoid arthritis (RA) [1]. Magnetic resonace imaging (MRI) is becoming surgical golden standard for the confirmation of compressive cervical myelopathy but is expensive and time consuming. Functional MRI is the most reliable examination but almost impossible to perform in majority of patients with atlantoaxial instability and long lasting RA. Forward atlantoaxial subluxation is the most frequent dislocation and easy to detect on standard and dynamic radiograms of the cervical spine. Steel's rule of third is plausible screening method for assessment of sagittal spinal canal in C1 level (in sagittal diameter of atlas one third of the space occupies dens, one third spinal cord and one third is free space that allows secure movements in this region) [2]. Theoretically the maximal and still safe atlantodental distance could be one third of the atlas saggittal diameter [2, 3]. This is partial explanation for low incidence of neural complications in patients with RA of craniovertebral junction. The aim of clinical evaluation is to detect early signs of neural involvement. Somatosensory evoked potentials (SSEPs) represent a neuroelectrophysilogical method that estimate the function of the posterior column of the spinal cord [4, 5, 6, 7, 8]. It allows early detection of disorders of cervical neural pathways before the clinical symptoms develop. SSEPs could be perforemed in neutral and dynamic fashion what gives information is there any compressive moment in different positions [5].

In our department we are following the next steps in the approach to the patient with RA and cervical spine involvement: 1. radiological assessment (standard and lateral flexion-extension views of cervical spine and open mouth views of dens) of all patients with RA and cervicogenic symptoms regardless of the duration of disease and all patients with RA lasting 5 years and more; 2. SSEPs for median nerve in standard and functional position of patient' head when there is radiologically significant atlantoaxial dislocaton (> 4mm or more in maximum flexion) or any other changes related to RA activity; 3. MRI is last step depending of the previous findings; 4. consideration of more agressive pharmacological treatment, cervical orthosis or surgery. Neuroradiological analyses visualise bone displacment and erosions, proliferative rheumatoid tissue and compression of spinal cord but do not express functional integrity of neural pathways. SSEPs is an accepted method for evaluating sensory functions of the central nervous system with the possibility to detect subclinical disorders. SSEPs should be registered in neutral position, flexion and rotations of the cervical spine.

We investigated 56 female hospital patients with RA who were admitted to the Department of Rheumatology, University Hospital Rebro, Zagreb, Croatia during one year [5]. Twenty five (44%) had forward atlantoaxial subluxation ranging from 4 to 17 mm. A total of 12 patients had pathological SSEPs latencies for median nerve and among them 11 had neck symptoms, 6 had abnormal neurological findings and 6 had forward atlantoaxial subluxation. The pathological SSEPs findings were provoked by movements of the neck in 9 out of 12 patients. In conclusion, we would like to add radiological analysis of cervical spine according Steel's rule of third and SSEPs functional examination as valuable diagnostic tool for evaluation of rheumatoid cervical spine. SSEPs results help to identify patients with cervical cord dysfunction before detectable clinical signs develops and thus contribute to the therapeutic decision making.

Ð Babiæ-Nagliæ¹, B Æurkoviæ ¹

Department of Rheumatology University Hospital Rebro Zagreb Croatia
Correspondence to: Prof. Ðurða Babiæ-Nagliæ, MD Kišpatiæeva 12 Zagreb 10000 Croatia dnaglic@kbc-zagreb.hr

References

1. Kauppi MJ, Barcelos A, da Silva JAP. Cervical complications of rheumatoid arthritis. Ann Rheum Dis 2005; 64: 355-358.

2. Steel HH. Anatomical and mechanical consideartions of the atlantoaxial articulations. J Bone Surg 1968; 50: 1481-1482.

3. Babiæ-Nagliæ Ð, Potoèki K, Æurkoviæ B. Clinical and radiological features of atlantoaxial joints in rheumatoid arthritis. Z Rheumatol 1999; 58: 196-200.

4. Giraud P, Mauguiere F. Somatosensory evoked potentials in rheumatoid polyarthritis with radiologic involvement of the cervical spine. Neurophysiol Clin 1997; 27: 33-50.

5. Babiæ-Nagliæ Ð, Nesek-Maðariæ V, Potoèki K, Lelas-Bahun N, Æurkoviæ B. Early diagnosis of rheumatoid cervical myelopathy. Scand J Rheumatol 1997; 26: 247-52.

6. Tsiptsios I, Fotiou F, Sitzoglou K, Fountoulakis KN. Neurophysilogical investigation of cervical spondylosis. Electromyogr Clin Neurophysiol 2001; 41: 305-13.

7. Hayashida T, Ogura T, Hase H, Osawa T, Hirasawa Y. Estimation of cervical cord disfunction by somatosensoray evoked potentials. Muscle Nerve 2000; 23: 1589-1593.

8. Dvorak J, Sutter M, Herdmann J. Cervical myelopathy: clinical and neurophysiological evaluation. Eur Spine J 2003; 12 (suppl 2): S181-187.

Dear Editor,

With the increasing number of patients now being treated with tumour necrosis factor alpha (TNF-alpha) blocking agents, the publication of an updated consensus statement in the Advances in Targeted Therapies VI supplement of Annals of the Rheumatic Diseases was both timely and relevant. However we notice that conflicting advice on the management of patients with Hepatitis B Virus (HBV) infection is given in this supplement in 2 separate articles. Furst et al[1] comment on the paucity of data on this situation which leads them to conclude that "TNF blockers should not be used in patients with HBV infection (category C evidence)". In a subsequent article, Calabrese et al[2] review the evidence obtained from 4 cases and conclude that when there is no alternative to TNF blockade the potential exists for safe administration of TNF blocking agents through careful monitoring of viral status, prior or concurrent prescription of lamuvidine and/or temporary discontinuation of anti-TNF therapy should HBV re-activation occur. Given that HBV infection and inflammatory arthritis are common conditions, this is a not uncommon situation that most rheumatology centres will encounter at some point and we wish to clarify the correct position on the management of patients with HBV infection and TNF-alpha blocking agents.

Since Calabrese et al’s[2] article, Esteve et al[3] have reported 3 cases of Crohn’s disease patients with chronic HBV who received infliximab. From 80 patients tested prospectively for HBV prior to infliximab therapy for Crohn’s disease, 3 were identified with chronic HBV infection and 2 of the 3 patients developed severe reactivation of HBV at 2 and 3 months after the last infliximab infusion. One patient made a complete recovery after 3 months while the second died from advanced hepatic failure. A third patient was treated with lamivudine therapy prior to treatment with infliximab therapy and has had no clinical or biochemical worsening of liver disease either during or after therapy.

We also have experience of safely and successfully using TNF blockade in 2 patients with RA and chronic HBV infection who had failed standard DMARD therapy and continued to have severely active RA. A 50 year old woman with RA and chronic HBV infection who had failed treatment with 5 DMARDs – including cyclophosphamide – was commenced on etanercept 25mg twice weekly s.c. At present she has completed 12 months of treatment with a reduction in DAS28 from 7.22 to 5.56. She did not receive any antiviral therapy prior to or during etanercept treatment and has had no evidence of HBV reactivation - HBV surface antigen is negative and HBV DNA undetectable. Etanercept was recently stopped due to persistent diarrhoea and she continues on treatment with adalimumab which has been temporarily discontinued during surgery for atlanto-axial subluxation. In a second case, a 62 year old woman with RA and chronic HBV infection (HBsAg negative) who had failed treatment with 3 DMARDs received etanercept 25mg twice weekly in combination with 15mg methotrexate s.c. After 3 months DAS28 reduced from 8.16 to 2.63. No antiviral therapy was given prior to treatment with etanercept, and we have seen no evidence of HBV reactivation to date – HBV DNA undetectable. She has now had 6 months of treatment with etanercept with no complications.

Clearly the limited number of reports of anti-TNF therapy in HBV infected patients raise a number of important issues. We believe that the currently available data suggests that TNF blockade is a therapeutic option in patients with RA and chronic HBV infection, though the risk/benefit ratio must be carefully assessed in each patient. Furthermore, the European Association for the Study of the Liver International Consensus on Hepatitis B has produced guidelines on the safe treatment of patients with HBV who require immunosuppressive treatment. These guidelines led Calabrese et al[2] to advise that TNF blockade may be prescribed in patients with HBV infection. This should include assessment by a hepatologist and close monitoring of liver enzymes and HBV viral load during TNF blockade. Anti-viral therapy is effective for HBV re-activation but it is not yet clear whether anti-viral therapy should be administered prior to or during anti-TNF therapy or solely when HBV re-activation is detected. We believe that since HBV screening has not been mandatory in the pre-treatment assessment of patients receiving TNF blockade there are likely to already be a small but considerable number of patients with HBV infection who are already receiving TNF blockade. Further reporting of these cases would help clarify the risks of HBV re-activation, whether it is more frequent with infliximab, etanercept, or adalimumab and provide information on the outcome of anti-viral treatment.

Graham Raftery 1, Bridget Griffiths 2, Lesley Kay 3, David Kane 4.

Freeman Hospital, High Heaton, Newcastle upon Tyne, NE7 7DN, UK.

Correspondence to: Dr. Graham Raftery, Department of Rheumatology, Freeman Hospital, High Heaton, Newcastle-upon-Tyne, NE7 7DN, UK.

Tel: 44-191-233-6161, Fax: 44-191-223-1159, e-mail: grahamraftery@doctors.org.uk

References

1 Furst D E, Breedveld F C, Kalden J R, Smolen JS, Burmester G R, Bijlsma, J W, et al. Updated consensus statement on biologic agents, specifically tumour necrosis factor alpha; (TNFalpha) blocking agents and interleukin-1 receptor antagonist (IL-1ra), for the treatment of rheumatic diseases, 2004. Ann Rheum Dis 2004;64(Suppl 2):ii2-ii12.

2 Calabrese L H, Zein N, Vassilopoulos. Safety of antitumour necrosis factor (anti-TNF) therapy in patients with chronic viral infections: hepatitis C, hepatitis B and HIV infection. Ann Rheum Dis 2004;63(Suppl 2)ii18-ii24.

3 Esteve M, Saro C, Gonzalea-Huix, Suarez F, Forne M, Viver JM. Chronic hepatitis B reactivation following infliximab therapy for Crohn’s disease patients: need for primary prophylaxis. Gut 2004 Sep;53(9):1363-5.