Case history of a patient with multiple sclerosis and scleroderma.

British Journal of Rheumatology 1997;36:502–509

LETTERS TO THE EDITOR Case History of a Patient with Multiple Sclerosis and Scleroderma S—We describe the case of a 48-yr-old woman who developed multiple sclerosis (MS) in her early twenties and who subsequently presented with features of systemic sclerosis. This rare association is interesting due to overlapping pathogenesis and ensuing severe disability [1]. Miss SH presented in 1972 with a spastic paraparesis and one episode of optic neuritis. By 1980, she had become a wheelchair user and required assistance in many activities of daily living. She was admitted in August 1993 with pressure sores on both heels, knees and sacrum as a result of flexor spasms and contractures at the hips and knees. These sores were treated conservatively and tenotomies were performed on the hip adductors and knee flexors to improve her wheelchair posture. She was also noted to have Raynaud’s disease and developed small hypopigmented patches, suggesting a diagnosis of scleroderma. Further investigation included a barium swallow which showed poorly coordinated peristalsis and reflux oesophagitis. There was a negative rheumatoid factor (latex), a positive ANA titre of 1:320 with a speckled pattern, but anti-double-stranded DNA was negative. Hand X-rays showed soft-tissue swelling without calcification. The ESR was q100, CRP was 64 and anticentromere antibodies and ANCA were absent. A forearm skin

biopsy was consistent with a diagnosis of scleroderma. CT scan and X-ray of the chest were both normal. Visual evoked responses had previously confirmed the diagnosis of MS. A recent MRI scan revealed changes consistent with demyelination. On mobilization, her scleroderma became active and a further sore developed. She underwent a surgical debridement and healed quickly [2]. In the meantime, her skin tightness worsened and she was treated with penicillamine [3]. Three weekly courses of daily prostacyclin infusions were given for her Raynaud’s symptoms, without objective improvement. Despite penicillamine for 25 weeks (500 mg daily), her skin changes persisted, and CRP and ESR remained high. On stopping the drug, her symptoms and upper limb function paradoxically improved, which was difficult to explain. She remains quadriparetic from her MS and severely affected by her scleroderma, but with no further cutaneous progression (Fig. 1). Her sore had not fully resolved, but is stable. She returned home with considerable community care support and, 6 months later, both diseases remain controlled. Miss SH had both MS and systemic sclerosis which have been described together in only two cases in the British literature since 1984 [4]. Attention is drawn to these two important disorders, in which there are pathophysiological similarities with respect to

F. 1.—The hands show tightening of the skin with a shiny appearance.

502

LETTERS TO THE EDITOR

T-lymphocyte abnormalities [5], and their subsequent clinical pictures and profound disability. S. H. J, A. A, A. B. W

North Staffordshire Rehabilitation Centre, The Haywood, North Staffordshire Hospital NHS Trust, High Lane, Burslem, Stoke-on-Trent ST6 7AG Accepted 11 October 1996 Correspondence to: S. H. Jawad. 1. Baker HWG, Balla JI, Burger HG, Ebeling P, Mackay IR. Multiple sclerosis and autoimmune diseases. Aust NZ J Med 1972;3:256–60. 2. Grossman JA, Barrall DT, Dennison A, Lally EV. Successful combined medical and surgical treatment of a lower extremity sclerodermal ulcer. Ann Plast Surg 1988;20:582–5. 3. Black CM, Denton CP. The management of systemic sclerosis. Br J Rheumatol 1995;34:3–7. 4. Trostle DC, Helpich D, Medsger TA Jr. Systemic sclerosis and multiple sclerosis. Arthritis Rheum 1986;29:124–7. 5. Seibold JR. Connective tissue diseases characterised by fibrosis. In: Kelly WN, Harris ED, Ruddy S, Sledge CB, eds. Textbook of rheumatology, Vol. 2, 4th edn. Philadelphia: WB Saunders, 1993:1123–30.

Re: Three Year Follow-up of a Case of Giant Cell Arteritis Presenting with a Chronic Cough and Upper Limb Ischaemic Symptoms S—The association of giant cell arteritis and pulmonary disease, exemplified by a recent case report [1], has its corollary in the rare association of allergic asthma and cryptogenic fibrosing alveolitis (CFA) [2] in a patient with a previous history of polymyalgia rheumatica (PMR). At the age of 60, this patient had a clinical diagnosis of PMR characterized by weight loss and a lumbosacral distribution of myalgia [3], and concurrent investigations (with reference ranges) showing an erythrocyte sedimentation rate of 95– 124 mm/h (Westergren), positive test for C-reactive protein, serum alkaline phosphatase 160–212 IU/l (25–125), g-glutamyl transferase 106–248 IU/l (0–45), and an elevated serum immunoglobulin G level of 314 IU/ml (80–220). Rheumatoid factor was absent, but the lupus erythematosus test was weakly positive. His symptoms and hepatic biochemical derangements resolved (with the exception of immunoglobulin G, which remained elevated, and immunoglobulins A and M, which later increased, despite the absence of alcohol abuse), and the ESR fell, after treatment with corticosteroids, which were discontinued after 12 months. At the age of 73, he presented with acute asthma, characterized by sputum eosinophilia, and a striking improvement in peak expiratory flow rate (from zero to 380 l/min) after corticosteroids and bronchodilators. Although he had marked finger clubbing, his chest X-ray showed no focal abnormality. Two years later, however, bilateral basal mottling was noted on chest radiography, and computerized axial tomography displayed the typical peripheral crescentic pattern of honeycomb shadowing attributable to CFA. As expected, lung function tests showed an obstructive profile, compatible with obstructive airways disease, and an impairment of pulmonary diffusing capacity consistent with interstitial lung disease. The lupus

503

erythematosus test was again weakly positive, but the relevant tests were negative for antinuclear factor and rheumatoid factor. His histocompatibility status was HLA1 B8. O. M. P. J

Tameside General Hospital, Fountain Street, Ashtonunder-Lyne, Lancashire OL6 9RW Accepted 5 September 1996 1. Rischmueller M, Davies RP, Smith MD. Three year follow-up of a case of giant cell arteritis presenting with a chronic cough and upper limb ischaemic symptoms. Br J Rheumatol 1996;35:800–2. 2. Jolobe OMP. Asthma with CFA could arise from altered immunity. [Letter] Care Elderly 1994;6:346. 3. Hamilton CR, Shelley WM, Tumulty PA. Giant cell arteritis: including temporal arteritis and polymyalgia rheumatica. Medicine (Baltimore) 1971;50:1–27.

Ehlers–Danlos Support Group S—I was pleased to see Dr D. L. Scott’s review of our booklets in British Journal of Rheumatology 1996;35:913–4. In his review, Dr Scott has suggested that a ‘telephone helpline would be useful for patients to ask difficult questions when necessary’. I am sure your members would be interested to know that such a helpline does exist as anyone phoning the support group phone number (01748 823867) will be given all the help they require. V. B

Ehlers–Danlos Support Group, 1 Chandler Close, Richmond, N. Yorks DL10 5QQ Accepted 17 September 1996 Hypersensitivity Vasculitis Related to Aceclofenac S—Hypersensitivity vasculitis (HV), also referred to as leucocytoclastic vasculitis, is thought to be an immunopathogenic disease resulting from the deposition of immunocomplexes, mainly in the small vessels. The precipitating antigen can be endogenous (in the context of connective tissue disorders or other chronic diseases) or exogenous (drugs or infections) [1]. The proportion of idiopathic cases ranges from 20 to 60% [2]. In its most benign form, the skin is the most commonly involved organ, mainly as a palpable purpura, usually in the lower limbs. However, any organ can be affected. Drugs have been implicated as the only possible precipitating factor in 12–18% of two series of non-selected consecutive cases, and as a cofactor in an additional 16–22% [2, 3]. Antibiotics and non-steroidal anti-inflammatory drugs (NSAIDs) are among the drugs most commonly associated with HV [2, 3]. Aceclofenac, a new NSAID marketed in Spain and Portugal since 1992, which has been submitted for approval in most European countries, is a phenylacetic acid derivative. As far as we know, only three reports of HV associated with the use of this new NSAID have been published [4–6]. We report on five additional cases of HV in patients treated

504

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 4

TABLE I Cases of hypersensitivity vasculitis associated with aceclofenac Case no.

Age/ sex

1* 2†

66/F 67/F

3‡

74/F

4

67/F

5 6† 7†

50/F 72/M 68/F

8†

52/M

Indication for use of aceclofenac

Daily dose (mg)

Induction period (days)

Recovery period (days)

Sciatica Polymyalgia rheumatica Osteoarthritis of the knee Osteoarthritis of the knee Low back pain Humeral fracture Minor trauma of the ankle Ankylosing spondylitis

200 200

50 18

3 11

200

15

15

100

12

18

100 200 200

1 1 3

3 28 3

200

5

10

F, female, M, male. *The patient had also taken Nervobion8 (a fixed-drug combination of cyanocobalamin + cocarboxylase + pyridoxal-5phosphate). †Diagnosis confirmed by cutaneous biopsy. ‡Positive rechallenge. Case 6 was published by Go´mez Rodrı´ guez et al. [4]. The patient also presented with microscopic haematuria. Case 7 was published by Epelde and Boada [5]. The patient also presented with haemoptysis. Case 8 was published by Nu´n˜ez et al. [6]. The patient also presented with microscopic haematuria.

with aceclofenac which have been reported to the Catalan Centre of the Spanish System of Pharmacovigilance. Between 1 and 50 days after starting treatment with 100–200 mg/day of aceclofenac, palpable purpura developed in all patients (Table I). All patients recovered completely 3–18 days after withdrawing aceclofenac. Except for one patient (patient no. 1), who had also taken a fixed-dose combination of cyanocobalamin + cocarboxylase + pyridoxal-5-phosphate 1 month before purpura developed, other known causes of HV had been ruled out. Patient no. 2 had polymyalgia rheumatica and although this syndrome is not a connective disease related to HV, its possible predisposing role cannot be completely ruled out; this patient had biopsy-proven leucocytoclastic vasculitis. No patient had systemic involvement and, except for case no. 2, diagnoses were made by clinical inspection of the cutaneous lesions. The positive rechallenge experienced by patient no. 3 strengthens the causal association between aceclofenac and HV. Although HV is commonly a benign disease, in some instances its outcome may be harsh. Alclofenac and fenclofenac, two NSAIDs structurally similar to aceclofenac, were withdrawn in the late 1970s and 1980s because of severe cutaneous adverse reactions, including several cases of generalized vasculitis associated with alclofenac [7]. Unfortunately, our data do not allow estimation of the risk of this previously undescribed adverse effect of aceclofenac. We thank the reporting physicians. This work was supported by Servei Catala` de la Salut.

R. M, A. F, D. C`, J.-R. L

Institut Catala` de Farmacologia, Universitat Auto`noma de Barcelona, Servei de Farmacologia Clı´ nica, CSU Vall d’Hebron, 08035 -Barcelona, Spain Accepted 27 September 1996 Correspondence to: J.-R. Laporte, Institut Catala` de Farmacologia, Servei de Farmacologia Clı´ nica, CSU Vall d’Hebron, 08035-Barcelona, Spain. 1. Fauci AS, Haynes BF, Katz P. The spectrum of vasculitis. Clinical, pathologic, immunologic, and therapeutic considerations. Ann Intern Med 1978;89:660–76. 2. San Jose´ A, Bosch JA, Knobel H et al. Vasculitis por hipersensibilidad. Estudio de 106 casos. Rev Clin Esp 1986; 178:368–72. 3. Lozano F, Martı´ n E, Luaces C, Cardellach F. Vasculitis leucocitocla´sticas: ana´lisis de 62 casos. Med Clı´ n (Barc) 1987;88:437–41. 4. Go´mez Rodrı´ guez N, Moratiel Valde´s D, Formigo Rodrı´ guez E, Ferreiro Seoane JL. Vasculitis por hipersensibilidad y aceclofenaco. Med Clı´ n (Barc) 1993;101:239. 5. Epelde F, Boada L. Leukocytoclastic vasculitis and hemoptysis after treatment with aceclofenac. Ann Pharmacother 1995;29: 1168. 6. Nu´n˜ez M, Miralles ES, Harto A, Ledo A. Hypersensitivity vasculitis associated with aceclofenac. J Dermatol Treat 1995; 6:54. 7. Mann RD. Withdrawal of alclofenac. Br Med J 1979;2:133.

Re: Synovitis Associated with an Electrical Injury S—Electrical injuries may be due to several mechanisms: (1) direct contact; (2) arcing of electricity; (3) exposure to the intense heat of an arc flash; (4) fires ignited by the heat; (5) mechanical injuries associated with the electrical accident. Musculoskeletal complications are frequently encountered and are most commonly fractures due either to falls or sustained tetanic contractions [1]. Acute synovitis is not a recognized complication of electrical injury. I wish to report a case of acute synovitis of the wrist and hand following a domestic electrical injury. A 72-yr-old Asian man developed acute painful swelling of the right wrist associated with stiffness and erythema. The symptoms came on 3 h after touching a live domestic electrical wire (240 V, AC) with the right hand. Three weeks later, he developed increasing pain and stiffness of the second and third digits of the right hand. Tenoxicam 20 mg/day was commenced with slight improvement of wrist swelling and pain. Six weeks after the electrical injury, his right hand showed diffuse swelling of the dorsum of the hand as well as firm swelling of the second and third fingers. Mild, tender synovial swelling was evident of the second and third PIP and DIP joints, all MCP joints and the wrist. Only 5° flexion and extension were possible in the wrist. A blue–brown discolouration of the skin was present, overlying the dorsal and ventral surfaces of the wrist with desquamation of the wrist, fingers and thumb. The left hand and wrist were normal. An X-ray of the right hand demonstrated carpal osteopenia and soft-tissue swelling about the wrist and second and third fingers. There was no chondrocalcinosis. The ESR, 43 mm/h 2 weeks earlier,


had fallen to 22 mm/h and CRP was Q10 mg/ml. He was seronegative for rheumatoid factor and ANA was negative. A technetium bone scan demonstrated increased uptake of tracer in the right wrist, carpus, MCP and PIP consistent with synovitis. The right wrist joint was injected with 40 mg of triamcinolone acetonide. No synovial fluid could be aspirated from the joint. There was rapid resolution of the symptoms in the right wrist over 48 h. The range of movement of the fingers improved slightly over the first 48 h and with continued active physical therapy the right hand and wrist were normal after 12 weeks. This case describes synovitis of the wrist and small joints of a hand following a low-voltage electrical injury. The synovitis was evident clinically and supported by an elevated ESR, the presence of osteopenia on radiographs and bone scintigraphy. The synovitis was associated with cutaneous changes consistent with a resolving thermal injury. The cutaneous features may also be compatible with resolving cellulitis; however, there was no evidence of infection and the patient did not receive antibiotic therapy. Current markings were not present. The synovitis was minimally responsive to non-steroidal anti-inflammatory drug therapy. Intra-articular corticosteroid therapy resulted in rapid clinical improvement. Within 3 months, the synovitis and cutaneous changes had completely resolved. The cause of the synovitis in this case is unclear. In a review of 182 cases of electrical injury, five cases of acute gout were recorded [1]. It is possible that the synovitis in the case I have described was crystal induced; however, the patient had had no previous episodes of acute synovitis, there were no features of a chronic arthropathy, the synovitis involved the majority of the joints of the hand and wrist, and there were no gouty erosions or chondrocalcinosis on X-ray. Reflex sympathetic dystrophy has been described following low-voltage injuries; however, in this case there were no clinical features of reflex sympathetic dystrophy [2]. Direct electrical injury to the joints is unlikely to have occurred in a low-voltage injury. The low resistance of neural tissue makes it prone to electrical injury. It is possible that neural electrical damage resulted in the release of substance P and subsequent joint inflammation as a result of prostaglandin E2 and collagenase release, and synoviocyte proliferation. Trauma to a joint, usually in the form of a direct blow or as a result of forced inappropriate motion, may be followed by an acute synovitis [3]. It is possible that forceful muscle contractions at the time of the electrical insult resulted in direct joint trauma. However, the development of finger involvement was delayed by 3 weeks and suggests that this was not responsible. Trauma has also been reported to trigger the development of seronegative spondyloarthropathies and psoriatic arthritis, and the time course of small hand joint involvement is consistent with this [3]. However, only the hand affected by the electrical injury was involved and the synovitis was

505

self-limiting. A ‘deep’ Koebner phenomenon has been suggested as an explanation for the development of psoriatic arthritis after trauma, as has the release of self-antigens from injured joints in the genesis of reactive arthritis [4]. I believe that this is the first case to be described of an acute inflammatory arthropathy occurring as a direct result of an electrical injury. Awareness of this possible complication of electrical injuries may result in its more frequent detection. A. L. T

Department of Rheumatic Diseases, Royal Perth Rehabilitation Hospital, 6 Selby Street, Shenton Park 6008 , Western Australia, Australia Accepted 6 September 1996 1. Butler ED, Grant TD. Electrical injuries, with special reference to the upper extremities: a review of 182 cases. Am J Surg 1977;134:95–101. 2. Demun EM, Redd JL, Buchanan KA et al. Reflex sympathetic dystrophy after minor electric shock. J Emerg Med 1993;11:393– 6. 3. Foley-Nolan D, Wright V. The relationship between trauma and arthritis. In: Klippel JH, Dieppe PA, eds. Rheumatology. London: Mosby, 1994:2.9.1–8. 4. Langevitz N, Buskila D, Gladman D. Psoriatic arthritis precipitated by physical trauma. J Rheumatol 1990;17:695–7.

Has Hepatitis C Virus a Specific Tropism for the Synovial Membrane? S—Arthritis is not included among the extrahepatic manifestations of hepatitis C virus (HCV) infection [1]. However, three anecdotal reports of polyarthritis associated with HCV infection have recently been reported [2, 3]. Two out of these patients had chronic HCV infection and a rheumatoid arthritis (RA)-like disease; HCV RNA was detected not only in the serum, but also in the synovial fluid (SF). The prevalence of antibodies to HCV is increased in patients with RA [4], a finding that may be related to the infection risk inherent in the number of medical procedures performed on them. However, in the absence of population-based studies, the possibility of a link between HCV infection and RA cannot be discarded. This hypothesis would be supported by a specific tropism of HCV for the synovial membrane (SM). As a first step, we have evaluated HCV RNA in both serum and SF of patients with RA and concomitant HCV infection. Four RA patients [5] with concomitant HCV infection were studied (Table I). All the patients had antibodies to HCV by ELISA and Western blotting in the serum, but none of them had elevation of alanine or aspartate transferases. Sera and SF effusions were tested on a mean of three different occasions (range 1–5) 1 month apart for HCV nucleic acids. Qualitative analysis of HCV RNA was performed with a nested polymerase chain reaction (PCR) with four primers complementary to the highly conserved 5' non-coding region of the HCV genome. Quantitative analysis of HCV RNA was performed using the branched-DNA signal amplification assay (Quantiplex, Chiron Corp.,

506


TABLE I Age, sex and laboratory results in four patients with RA and concomitant HCV infection HCV RNA Patient 1 2 3 4

Age

Sex

IgM RF

Serum

Synovial fluid

65 67 57 33

M M F F

+ + − −

negative positive negative negative

negative negative negative positive (?)

Emeryville, CA, USA). Informed consent was obtained from all patients. In three patients, HCV infection was community acquired and in the fourth it occurred after blood transfusions. Two patients had consistently negative results both on serum and SF. One patient had consistently positive results in the serum on four occasions (26.09 to 61.35 × 105 copies), whereas all three SF specimens were negative. In the fourth patient, HCV RNA was not detected in three different tests in the serum, but was present at the low concentration of 5.92 × 105 copies in 1/3 SF specimens. HCV RNA was searched for in an arthroscopic biopsy of the SM by PCR with a negative result. Her disease was classified as Still’s disease, although the criteria for RA were also fulfilled. The patient had complained of arthritis since age 4 after smallpox vaccination. She had a remission at age 12, but later arthritis recurred with high spiking fever, severe anaemia, leucocytosis, pericarditis, and hepato- and splenomegaly. IgM rheumatoid factor (RF), ANA, anti-DNA and anti-ENA antibodies were negative. Interestingly, arthritis recurred a few years after the blood transfusions that most probably caused HCV infection. In our experience, HCV did not show a tropism for the SM, despite the inflammation-related increase in permeability and vascularization, since even the patient with consistently positive tests in the serum had no detectable amounts of HCV RNA in the simultaneously aspirated SF samples. Moreover, the only positive test in the SF was not confirmed when the SM was evaluated. The diagnosis of HCV infection was made on the basis of positive ELISA and Western blot tests, and should not have been influenced by the possible interference of IgM RF [6] or by high concentrations of IgG [7]. Our data indicate the need for further studies on the role of the SM as a site for HCV persistence, and on SF analysis for viral RNA as a tool for differentiating RA and concomitant HCV infection from HCV-associated arthritis. M. A. C, A. P,* N. S,* R. B,* S. A

Rheumatology and *Gastroenterology Units, Dipartimento di Medicina Interna, Universita` di Genova, Viale Benedetto XV, 6 , 16132 Genova, Italy Accepted 22 September 1996 1. Gumber SC, Chopra S. Hepatitis C: a multifaceted disease. Review of extrahepatic manifestations. Ann Intern Med 1995; 123:615–20.

2. Ueno Y, Kinoshita R, Kishimoto I, Okamoto S. Polyarthritis associated with hepatitis C virus infection. Br J Rheumatol 1994;33:289–91. 3. Ueno Y, Kinoshita R, Tsujinoue H, Kato M. A case of hepatitis C virus (HCV)-associated arthritis. Quantitative analysis of HCV RNA of the synovial fluid and the serum. Br J Rheumatol 1995;34:691–2. 4. Baffoni L, Frisoni M, Miniero R, Righetti F, Sprovieri G, Ferri S. True positive anti HCV tests in rheumatoid arthritis. Br J Rheumatol 1993;32:349–50. 5. 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. 6. Theilmann L, Blazek M, Goeser T, Gmelin K, Kommerel B, Fiehn W. False positive anti-HCV test in rheumatoid arthritis. Lancet 1990;ii:1346. 7. Borque L, Elena A, Maside C, Rus A, Del Cura J. Rheumatoid arthritis and hepatitis C virus antibodies. Clin Exp Rheumatol 1991;9:617–9.

Does Application of Radiographic Contrast Medium in Radiation Synovectomy Influence the Stability of Yttrium-90 Colloid? S—Radiation synovectomy is a well-established method of non-surgical synovectomy for chronic inflammatory arthritis [1–5]. It is carried out by an intra-articular (i.a.) injection of a radiopharmaceutical. It is necessary to ensure that the needle is positioned correctly in the joint space before injecting the radiopharmaceutical. This is to prevent the spread of radioactive material in the adjacent joint structures. Radiographic contrast medium can help to confirm the correct position of the needle in the joint, and can provide information about the spread of the radionuclide in the joint space and adjacent bursae. For that aim, 1 ml of radiographic contrast is sufficient. Radiographic contrast medium is especially used when there is only a small amount of effusion, as we often saw in pigmented villonodular synovitis [2] and in haemophilic haemarthrosis [6]. Some radiographic contrast media contain chelating compounds e.g. ethylenediaminetetra-acetate (EDTA). In the information sheet of the European Association of Nuclear Medicine (EANM), entitled ‘Radionuclide therapy: from palliation to cure’, it is reported that yttrium-90 (Y-90) colloid has an affinity to some chelates (e.g. EDTA, which is used in iodine contrast medium). This may lead to a separation of the radionuclide from the colloid and hence, because of altered biodistribution after intracavitary administration, add to the bone marrow radiation dose [7]. So, the Y-90–EDTA complex can migrate into the blood circulation and be cleared by the kidneys. It may give undesirable radiation exposure of the whole body and possibly a decreased therapeutic effect in the inflamed joint. We have not found studies in the literature on this phenomenon. We have therefore examined the influence of radiographic contrast medium on the biodistribution of Y-90 colloid injected in the knee joint. A controlled comparative trial was carried out comparing Y-90 colloid and radiographic contrast medium Omnipaque (containing EDTA), Y-90 colloid and radiographic contrast medium Urografine (without EDTA), and

507


Y-90 colloid with saline. The injections in the knee joint were followed by collection of the urine for 48 h and the radioactivity in the urine was measured. Fifteen patients with an active chronic synovitis of the knee joint, which persisted after adequate treatment with anti-inflammatory drugs and i.a. corticosteroid injections, were included in this study. Patients who were treated with Y-90 colloid i.a. in both knees simultaneously were excluded for this study. All patients had given informed consent. The design of the work conforms to standards currently applied in the Netherlands. The patients were randomly divided into three groups of five patients: one group of five patients received an i.a. injection of 1 ml Omnipaque (Nycomed) containing 0.1 mg EDTA, a second group of five patients received i.a. 1 ml Urografine 60% (Scheding) without EDTA and a control group of five patients received i.a. 1 ml saline. Thereafter, each patient received an i.a. injection of 185 MBq (5 mCi) Y-90 citrate colloid (CIS bio international). Preceding the Y-90 colloid treatment, the following inflammatory activity variables of the knee joint were scored: tenderness, swelling, effusion and warmth on a four-point scale (0 = absent, 1 = slight, 2 = moderate, 3 = severe). Furthermore, the ESR, the CRP, serum creatinine and an X-ray of the knee joint were performed. The X-rays were scored according to Larsen et al. [8]. The method for radiation synovectomy has been described before [2, 3]. Briefly, 185 MBq Y-90 colloid were diluted in the vial up to 3 ml with saline and transferred into a 5 ml syringe. With the patient in a supine position and using an aseptic technique, the injection site was cleaned. With a standard medial or lateral approach, a needle was inserted into the synovial cavity and the knee was drained as completely as possible. Using a three-way tap, 1 ml of one of the contrast media or 1 ml of saline was injected and followed by 185 MBq Y-90 colloid, then the needle was flushed with 1 ml of saline. Next, the needle was withdrawn, pressure applied to the injection site and the knee flexed twice before being immobilized in full extension with a bandage and/or a splint. The patient was then confined to bed for 3 days. The urine of each patient was collected for 48 h. The total volume of each urine sample was determined and 4 ml of each urine sample were used for counting the radioactivity. Five microlitres of the solution of the injected doses were used as a standard and mixed with 4 ml of blank urine. Also 4 ml of blank urine were counted. To each urine sample of 4 ml, 16 ml of scintillation solvent were added (Hionic Fluor Packard). The b-radiation of the Y-90 was counted in a liquid scintillation counter for 10 min. Calculation of the percentage Y-90 in the urine was performed as follows: % of the dose in the urine = [(A/4) × (C − B)]/[(3000/5) × (S − B)] × 100, where A is the volume of a urine sample, C is the count of the urine sample in c.p.m., B is the count of the blank urine sample in c.p.m. and S is the count of the standard in c.p.m.

TABLE I Nature and dose of radiographic contrast medium and urine radioactivity after Y-90 colloid measured as a percentage of the dose administered in the knee joint No. of patients 5 5 5

Radiographic contrast medium

Radioactivity in urine (mean 2 ..)

Omnipaque 1 ml Urografine 1 ml Saline 1 ml

3.0 2 1.8 3.0 2 3.7 3.1 2 3.6

Quality control of three batches of Y-90 citrate colloid was performed by paper chromatography (Whatman 3 in acetone). About 3% of the radioactivity administered into the joint was found in the 48 h urine sample (Table I). No differences in radioactivity counts were found between the three patient groups. There was no correlation between the radioactivity in the urine and the inflammatory activity variables of the knee joints, such as tenderness, swelling, effusion, warmth, ESR and CRP. Serum creatinine was normal. No relationship was found between the radiological score [8] and the amount of radioactivity in the urine. Our radiochemical quality control showed a purity of Y-90 colloid of q99%. In this study, no significant differences in radioactivity excreted in the urine were found between three groups of patients treated with radiosynovectomy for a knee joint synovitis, whether or not a contrast medium, with or without EDTA, was added. Although on theoretical grounds one could expect some differences between these patient groups, no such differences could be determined. There was also no relationship between the inflammatory activity parameters or the radiological score [8] of the treated knee joint and the radioactivity excreted in the urine. In this small group of patients, we cannot confirm the suggestion in the EANM information sheet that for radiation synovectomy simultaneous administration of Y-90 colloid i.a. and iodine contrast media containing EDTA should be avoided. M. J. A. M. F, E. B. K,* A. M. T. B,† W. C. A. M. B,* J. A. M. L,‡ L. B. A. V  P†

Department of Rheumatology, Sint Maartenskliniek, Nijmegen and Departments of *Nuclear Medicine, †Rheumatology and ‡Radiology, Academic Hospital, Nijmegen, The Netherlands Accepted 7 October 1996 Correspondence to: Agnes M. Th. Boerbooms, Department of Rheumatology, Academic Hospital Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands. 1. Deutsch E, Brodock JW, Deutsch KF. Radiosynovectomy revisited. Eur J Nucl Med 1993;20:1113–27. 2. Franssen MJAM, Boerbooms AMTh, Karthaus RP, Buijs WCAM, Van de Putte LBA. Treatment of pigmented villonodular synovitis of the knee with Y-90 silicate; prospective evaluation by arthroscopy, histology and 99m Tc-pertechnetate uptake measurements. Ann Rheum Dis 1989;48:1007–13.

508


3. Boerbooms AMTh, Buijs WCAM, Danen M, Van de Putte LBA, Vandenbroucke JP. Radiosynovectomy in chronic synovitis of the kneejoint in patients with rheumatoid arthritis. Eur J Nucl Med 1985;10:446–9. 4. Edmonds J, Smart R, Laurent R et al. A comparative study of the safety and efficacy of Dysprosium-165 hydroxide macroaggregate and Y-90-silicate colloid in radiation synovectomy: a multicentric double blind clinical trial. Br J Rheumatol 1994;33:947–53. 5. Clunie G, Ell PJ. A survey of radiation synovectomy in Europe, 1991–1993. Eur J Nucl Med 1995;22:970–6. 6. Van Kasteren MEE, Nova´kova´ IRO, Lemmens JAM, Boerbooms AMTh. Long-term follow-up of radiosynovectomy with yttrium-90 silicate in haemophilic hemarthrosis. Ann Rheum Dis 1993;52:548–50. 7. EANM. Monograph of the Task Group Radionuclide Therapy from the European Association of Nuclear Medicine, 1993. Radionuclide therapy: from palliation to cure. 8. Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn Stockholm 1977; 18:481–91.

Re: Rheumatology in Israel S—We have read with interest the International Letter ‘Rheumatology in Israel’ by Ehrenfeld et al. (British Journal of Rheumatology 1996;35:778–80). Your readers may be interested to learn of the important part played by British rheumatology in the development of rheumatology in Israel. The Rheumatology Department of the Rambam Medical Centre in Haifa, which opened its doors in 1974, was firmly established in the best traditions of Taplow, Guy’s and the Kennedy where Professors Yehuda Scharf and Menahem Nahir (neither mentioned in the Ehrenfeld article) received their specialist training in the 1970s. The Rambam unit remains, to this day, the only designated university rheumatology in-patient unit in Israel, having 12 beds (not 18 as reported in the article) dedicated for the treatment of rheumatic diseases. It was from there that the first president of the fledgling Israel Society of Rheumatology was elected in 1984, that the successful struggle leading to the heavily resisted recognition of the speciality was waged, that the curriculum of the Specialty Board in Rheumatology was developed and where, until 1994, all qualifying examinations for the boards were held. Over the years, an impressive, modern and popular service has been built up spanning rheumatology in the widest sense, and matched by a commensurate programme of wide-ranging clinical and laboratory research (of which, contrary to what is stated in the report, AIDS research plays no part). Of particular importance to patient care in this unit has been, firstly, the introduction of the rheumatology nurse practitioner (based on the Leeds model) and, secondly, a uniquely close collaboration with their two British-trained rheumatological surgical colleagues, Professors Daniel Reis and Michael Besser. The most recent British influence came when Ms Helen

Whitelock, from the RNHRD in Bath, acted as external adviser on a scheme to develop a much-needed hydrotherapy pool, which opened last year. The Rambam Rheumatology Unit has a distinguished international reputation. The list of rheumatological world leaders who have contributed to its academic programme reads like an international ‘Who’s Who’! R. G, L. E. G

Department of Rheumatology, Guy’s & St Thomas’ Hospital Trust, St Thomas Street, London SE1 9RT Accepted 18 September 1996 Erythrocyte Sedimentation Rate (ESR) at Presentation is a Prognostic Indicator for Duration of Treatment in Polymyalgia Rheumatica (PMR) S—Dolan et al. [1] have compared ESRs in PMR patients still on prednisolone after 24 months with ESRs in those off prednisolone at 24 months. They describe a higher pre-treatment ESR and higher ESRs at 12 and 24 months in the group who were still on prednisolone after 24 months. However, they state that the prednisolone dose was reduced ‘according to disease activity and ESR’. This would create a ‘self-fulfilling prophecy’ whereby patients with higher ESRs on treatment would be continued on higher doses of prednisolone for longer. This, in turn, raises the possibility that the apparent association of pre-treatment ESR and duration of prednisolone treatment is due not to true prediction of disease activity, but purely to prediction of later ESR levels regardless of clinical disease activity. In other words, are individuals who tend to run higher ESRs in health likely to have the higher ESRs at presentation of inflammatory disease such as PMR? We are in the unusual position of having a cohort of PMR patients recruited without the requirement for raised ESR at presentation, and in whom relapse was diagnosed clinically without reference to ESR. This cohort is, therefore, a valid group in which to examine the usefulness of ESR at presentation as a prognostic indicator for the duration of treatment in PMR. This cohort has been described previously [2], and of those 44 patients 40 have now been followed either until successfully off prednisolone for 6 months or still requiring prednisolone at 48 months. Of the 40 patients, 29 were female and the median age was 71.5 (range 51–87) yr; 25 had pure PMR, six pure giant cell arteritis (GCA) and nine had both. Six had biopsy-proven GCA. Median ESR at presentation was 46 (range 7–112). Twenty-six patients suffered at least one clinical relapse and the median duration of treatment was 34.5 months (range 12 to q48 months). Both ESR at first relapse and ESR during disease suppressed on treatment did show weak correlations with ESR before treatment (rs = 0.229 and 0.297, respectively), but this did not reach statistical significance (P = 0.261 and 0.149) in this group of 26. This does not convincingly rule out a spurious association of pre-treatment ESR and duration of


treatment wherever ESR is used to govern the duration of treatment. In this cohort, the ESR has not been used in this way, therefore this possible confounding effect is removed. In fact, the pre-treatment ESR did show a significant correlation with duration of treatment: rs = 0.339 (P = 0.032). This confirms the prognostic value of pre-treatment ESR in PMR suggested by Dolan et al. [1]. G. P, B. H*

Hinchingbrooke Hospital, Huntingdon, Cambridgeshire PE18 8NT and *Addenbrooke’s Hospital, Cambridge Accepted 30 September 1996 1. Dolan AL, Dasgupta B, Fernandes L, Panayi GS, Moniz C. ESR at presentation is a prognostic indicator for bone loss and duration of treatment in polymyalgia rheumatica (PMR). Br J Rheumatol 1996;35(suppl. 1): 109. 2. Pountain GD, Calvin J, Hazleman BL. Alpha 1-antichymotrypsin, C-reactive protein and erythrocyte sedimentation rate in polymyalgia rheumatica and giant cell arteritis. Br J Rheumatol 1994;33:550–4.

Reply S—We were interested to read that Pountain and Hazleman were able to confirm our findings that

509

pre-treatment ESR in PMR can give prognostic information about length of treatment. We would agree that subsequently it is difficult to separate the inflammatory process from steroid dose, as corticosteroids are titrated to match disease activity. The important aspect of our work [1] is not just length of treatment, but that those patients who were still on treatment at 2 yr were a subgroup consisting of patients with more severe disease, characterized by: (i) reduced spine bone mineral density at the start of treatment and subsequently; (ii) higher baseline ESR (P = 0.005); and (iii) increased bone resorption markers at 12 months (P = 0.01). We are, therefore, able to suggest that those patients with high initial ESR are those who merit osteoporosis prophylaxis. A. L. D, G. S. P*

Greenwich District Hospital, Vanburgh Hill, London SE10 9HE and *Department of Rheumatology, Guys Hospital, London SE1 1. Dolan AL, Dasgupta B, Fernandes L, Panayi GS, Moniz C. ESR at presentation is a prognostic indicator for bone loss and duration of treatment in polymyalgia rheumatica ( PMR) . Br J Rheumatol 1996;35(suppl. 1): 109.