Clinical Reports in Rheumatology

(Conditions)

AA AMYLOIDOSIS

By: Patricia Woo FRCP PhD Professor of Paediatric Rheumatology Department of Molecular Pathology University College London Medical School The Windeyer Building 46 Cleveland Street London W1P 6DB

The term ‘amyloid’ means ‘starch-like’ and refers to the atypical staining properties of autopsy specimens with iodine sulphuric acid as described by Virchow in 1854 (1). Subsequently this material was found to stain positive with Congo red, which became green when viewed under polarising microscopy. This is still the histological definition of amyloid. It was not until 1959 that amyloid deposits were found to be composed of fibrils, associated with other proteins and glycosoaminoglycans. The amyloid fibrils have a typical electron microscopic appearance with a diameter of 10-15 nm, consisting of polypeptides arranged in a twisted ß-pleated sheet. Although the precise detail of the electron microscopic appearance of the fibrils have been questioned, nevertheless ß-pleated sheet is the predominant feature. This is the principle component of all amyloid deposits irrespective of the clinical types, genetic or acquired varieties, or experimentally induced as in experimental models of amyloidosis. The polypeptide component varies with different sub-types of amyloidosis - see Table 1.

Table 1: Classification of amyloidosis

Amyloidosis seen in rheumatological practice is usually the type known as ‘reactive amyloidosis’. The amyloid fibril consists of polypeptide chains derived from the acute phase protein found in serum: serum amyloid A (SAA). This form of amyloidosis is mainly associated with long-standing infections or inflammation, less frequently with cancer, mainly renal cell carcinoma or Hodgkin's disease. In countries where the incidence of chronic infection such as osteomyelitis and tuberculosis has declined, reactive amyloidosis is mostly caused by chronic rheumatic diseases, predominantly in adult rheumatoid arthritis, juvenile chronic arthritis and ankylosing spondylitis. The frequency of amyloidosis in these rheumatic diseases varies geographically. For example, there is a marked geographic difference in the prevelance of AA amyloidosis in juvenile chronic arthritis in European countries (5-10%) and in the United States (0.1%) (2). This highlights the variations in referral patterns, complicating factors such as urinary infections, follow-up periods and problems with diagnosis.

Disease severity and chronicity also are factors that could alter the frequency of amyloidosis since the serum level of the acute phase protein SAA tends to be chronically raised in uncontrolled disease. In this context it is interesting that amyloidosis is rare in systemic connective tissue disease such as systemic lupus erythematosus, dermatomyositis, systemic sclerosis and Sjögren's syndrome where there is minimal acute phase protein responses with low to normal levels of SAA. Furthermore, reports of lower frequencies in European countries within the last decade may be due to the fact that patients are treated longer and more aggressively with disease-modifying drugs.

Although it is recognised that severity and chronicity of disease predispose patients to develop amyloidosis, it is hard to predict individual cases at risk. This is unlike the situation of ß₂ microglobulin amyloidosis which occurs as a result of renal dialysis using particular types of filters, where the length of dialysis is proportional to the percentage of patients developing amyloidosis. Not every patient with long-standing inflammatory disease will develop amyloidosis. This may be due to a (genetic?) predisposition which will only become apparent in cases where there is sufficient amount of precursor protein production. Such a mechanism implies that other clinical conditions leading to an acute phase response such as infectious complications or major surgery may enhance the deposition of amyloid in cases of established AA amyloidosis. This should be a major point of attention in the management of patients with AA amyloidosis.

Pathogenesis of reactive amyloidosis

A prerequisite for the development of reactive amyloidosis is a high level of precursor protein production, in this case SAA. The amyloid deposits consist of fibrils with associated proteins, the most abundant being serum amyloid P component (SAP) and glycosoaminoglycans (GAGs) as mentioned before. However, a number of other proteins have also been found in amyloid deposits - see Table 2. A number of these are matrix proteins or proteins that are resistant to proteolytic cleavage. It has been postulated for a number of the proteins that they may have a role in preventing proteolytic cleavage of the amyloid SAA polypeptide chains leading to formation of fibrils. For example, glycosoaminoglycans occur in tissues in close temporal and morphologic relationship to amyloid deposition (3, 4). The large negative charge of the GAGs has an effect on the folding of precursor proteins to form fibrils (3), and may also have a stabilising effect on the fibrils (5). SAP is another protein that is resistant to proteolytic cleavage and is invariably present in amyloid deposits, irrespective of the type of fibril protein or clinical disease. This protein has been shown to protect amyloid fibrils against proteolysis (6). An intriguing factor which still has to be characterised is the ‘amyloid enhancing factor’ produced by reticuloendothelial cells during persistent inflammation (7). This factor precedes the occurrence of amyloid in affected organs in experimental amyloidosis and has been shown to shorten the induction time of experimental amyloidosis, from weeks to between 24 and 48 hours.

Table 2. Other constituents in amyloid

Table 3. Clinical features at presentation of amyloidosis in JCA patients

Clinical features

The presenting symptoms which may cause the physician to suspect amyloidosis are shown in Table 3. Organ involvement may remain undetected clinically for a long time until serious symptoms appear. No routine laboratory test can distinguish between arthritic patients with and without AA amyloidosis. A high index of suspicion should always be present in the clinician caring for patients with rheumatic diseases. Proteinuria or loss of renal function is the most frequent presentation of amyloidosis. In patients with a known associated disease or family history, signs and symptoms such as cardiac failure, hepatomegaly, hypotension, proteinuria or nephrotic syndrome, carpal tunnel syndrome, diarrhoea, haemorrhage and malabsorption should raise suspicion of amyloidosis.

The diagnosis of amyloidosis is histological, ie proof of AA amyloid deposition in biopsy material. Biopsies may be done using a general approach such as rectal biopsy which will pick up 80% of positive amyloid patients, subcutaneous fat aspirate with a 60% positive pick up rate, gastric or duodenal biopsy with more than 90% positive pick up rate, or by a direct approach such as renal biopsy in the presence of proteinuria (8). Renal biopsies have a higher morbidity rate because severe infiltration of blood vessels within the kidneys by amyloid could cause haemorrhage. In general, rectal biopsy is most frequently used in clinical practice. The biopsy must contain sub-mucosal tissue containing blood vessels since amyloid deposits are found around these vessels. Congo red staining method has to be performed strictly according to the method of Puchtler et al 1962 (9), and sections should be at least 5 microns thick to reveal birefringance. A further problem of histological diagnosis is that the biopsy is of necessity blind, and there will always be false negative results obtained, especially in early disease.

Different methods of imaging in order to detect amyloid deposits in a non-invasive manner have been investigated. For example, the accumulation of bone seeking tracers into amyloid deposits was first reported in 1975 (10). Since then numerous descriptions of ^99mTc-labelled methylene diphosphonate and pyrophosphate compounds have been reported. These methods depend on the rich calcium content of amyloid. Therefore, they are non-specific. The sensitivity is also controversial and echocardiography, for example, has been reported to be more sensitive in identifying cardiac amyloid than bone scanning. Another tracer which may localise non-specifically to amyloid is 67gallium citrate (11, 12).

The high affinity of the protein SAP for amyloid fibrils has recently been used successfully in the diagnosis of AA and other systemic amyloidoses. Radio-labelled SAP has been found to localise in amyloid deposits, thus demonstrating accurately the distribution of amyloid in vivo (13). The tracer used is a medium energy short half-life isotope of iodine: ¹²³I. The turnover of this radio-labelled substance gives a measure of the retention of isotope within the body and could be an alternative measure of the amyloid load in the patient. Distribution of amyloid deposits can be visualised using a gamma camera. This has been shown to be useful in cases where there is no proteinuria but suspicion of amyloidosis, as liver and spleen involvement appear to precede renal involvement. Longitudinal studies of amyloid patients that are under treatment also show that amyloid deposits can regress, but the regression rate varies widely from patient to patient. This is the first evidence that degradation of amyloid fibres is an important factor in the pathogenesis of amyloidosis. Among patients with AA amyloid in whom the underlying acute phase response has been supressed, either spontaneously in remission or following treatment, serial scans over a 2-3 year period have shown active progression of amyloid in about one half of cases, and regression in the remainder. Serial SAP isotope studies have also shown that the rates of increase of the amyloid load in the presence of continuing acute phase response also vary. Attempts to make the test more sensitive by using more sophisticated imaging techniques to visualise the isotope within the body have been performed in a limited number of cases. The main problem is that the gamma camera does not show gut, cardiac and cerebral amyloid deposits. It appears that using positron emission tomography (PET) is sensitive enough to visualise cranial amyloid. However, the images are difficult to analyse at present (14). The above technique is not yet available as a routine measure because of the shortage of clinical grade SAP as well as production of the labelled material.

Therapeutic options in AA amyloidosis

Since circulating SAA is the precursor of AA amyloid deposits, reduction of the precursor protein is the most rational approach at present for the management of amyloidosis. Prevention of amyloidosis is preferable to treatment of the established disease. The falling incidence in developed countries of AA amyloidosis as a complication of osteomyelitis and other chronic infections, compared to 40 years ago, clearly indicates the supreme value of prevention by adequately treating an underlying inflammatory disease. Thus aggressive therapy of rheumatic diseases such as RA and JCA is desirable in this context. Such a strategy, however, might result in over-treatment of the majority of patients who will never develop amyloidosis, so this option is only realistic if the patients at risk from the development of amyloidosis could be properly identified. A search for genetic markers has so far not been very fruitful. HLA studies have not revealed any markers in that region identifying the patient at risk. A restriction fragment length polymorphism of the SAP gene seems to be a marker in patients with JCA and amyloidosis, but not in other diseases (15). Further elucidation of candidate genes or using reverse genetics to identify linkage might be worthwhile research to develop this clinically useful indicator.

Early diagnosis would improve the survival rate and SAP scintigraphy or turnover studies could be valuable and may become more accessible as a routine method later.

As far as drug treatment of amyloidosis is concerned, this will have to be tailored to the type of disease. In JCA chlorambucil treatment was initiated in the early 1960’s (16). In a recent retrospective study a clear benefit of this therapy in 79 patients with JCA amyloidosis has been shown. The survival rate was improved in a 15 year follow-up to 68% compared to 0% in the untreated group (17). The increased risk of leukaemia following chlorambucil therapy as well as infertility are major problems. The use of other drugs such as methotrexate to suppress disease is preferable if the acute phase response is abolished. The value of corticosteroid therapy is controversial. There have been some reports suggesting adverse effects of corticosteroids in promoting amyloidosis in some patients with RA (18). Although corticosteroids suppress both CRP and SAA levels in longitudinal studies of patients with RA, this affects CRP more than SAA. This illustrates the importance of monitoring SAA instead of CRP levels. Administration of corticosteroids once a day or on alternate days may induce a rebound phenomenon in acute phase response, including a sharp increase in SAA levels (8).

Dimethylsulphoxide (DMSO) therapy was advocated by Ravid et al in 1982 (19) when they did a pilot study of 7 patients. However, the side effects of DMSO include transient elevation of transaminases, nausea and a major social side effect of an appalling smell produced by the patient consuming DMSO. Thus this has not gained any favour in clinical therapy.

Supportive therapy is needed for advanced cases of amyloidosis, eg for renal function, which includes dialysis and eventually transplant. Recurrence of amyloid in the renal allograft has been documented in several case reports, but the exact frequency is unknown. Colchicine has a preventative effect on the recurrence of amyloid in the renal allograft in patients with familial Mediterranean fever if taken in adequate dosage (20).

Gastrointestinal symptoms are due to hypomotility of the stomach mainly due to dysphagia. Usually therapy with Cisapride is effective. Gastrointestinal ulceration, bowel perforation, bleeding, constipation, diarrhoea and malabsorption have all been documented. Often acute abdomen can occur due to intestinal obstruction. Approach to the management of AA amyloidosis complicating RA is summarised in Table 4.

Table 4. Management of AA amyloidosis in rheumatoid arthritis

Prognosis

The prognosis of patients with AA amyloidosis studied by a number of groups in Europe showed a survival of 50% ranging between 2-4 years. The longest follow-up in JCA was our report on a 15 year follow-up study that showed 68% survival if patients were treated with chlorambucil, and 100% mortality in those not treated with cytotoxic drugs (17). In a study of 63 Dutch patients with systemic AA amyloidosis, 98% of all patients’ deaths were related to amyloidosis. Although the presenting symptoms were renal in 90% of the patients, renal insufficiency accounted for only 35% of the immediate causes of death. Infection, gastrointestinal bleeding, bowel perforation and myocardial infarction were frequent causes of death (21). In JCA 84% of the causes of death in our recent series were due to renal failure (17). Infection is the second most common cause of death, usually due to bacterial septicaemia.

Familial mediterranean fever (fmf) with aa amyloidosis

Amyloidosis associated with familial Mediterranean fever is the only form of systemic amyloidosis known to be inherited as a recessive trait (22). Like other forms of AA amyloidosis nephropathy is the most important clinical feature and a significant cause of death. The amyloidosis tends to occur in patients from certain ethnic groups, eg North African and Sephardi Jews with a family history of familial Mediterranean fever, and Anatolian Turks and Armenians with origins in the Mediterranean area. It appears that FMF and amyloidoses are inherited independently. The gene for familial Mediterranean fever has been located to the short arm of chromosome 16 and not linked to the human SAA gene family on chromosome 11 (23).

The treatment of the amyloidosis of FMF has been studied by Zemer et al in a large prospective study of 1,070 patients over a 4-11 year period (24). Colchicine is the drug of choice in the prevention of amyloidosis in FMF patients. In addition colchicine can prevent additional deterioration of renal function in patients with amyloidosis who have proteinuria but not nephrotic syndrome. It apears that 1.5 mg a day of colchicine is necessary to prevent amyloidosis, irrespective of age, and 2 mg of colchicine a day can cause regression of amyloidosis (Zemer, personal communication). In a more recent Turkish series, only 2% of patients developed amyloidosis whilst on colchicine (25), thus providing corroboration of the efficacy of colchicine.

References

1. Virchow R. Vebereine in Gehirn und Rückenmark des Menschen aufgefundene substanz mit der chemischen Reacktion der Cellulose. Virchows Archives Pathology, Anatomy & Physiology 1854; 6: 135-38.
2. Filipowicz-Sosnowska AM, Restropowicz-Denisfewicz K, Rosenthal CJ, Baum J. The amyloidosis of juvenile rheumatoid arthritis. Comparative studies in Polish and American children. Arth and Rheum 1978; 21: 699-703.
3. Snow AD, Willman J and Kisilevsky R. A close structural relationship between sulfated proteoglycans and AA amyloid fibrils. Lab Invest 1987; 57: 687-98.
4. Stenstad T, Magnus JH and Husby G. Characterisation of proteoglycans in mouse splenic AA amyloidosis. Biochem J 1994; 303: 663-70.
5. Husby G, Stenstad T, Magnus JH, Sletten K, Nordvåg BY and Marhaug G. Interaction between circulating amyloid fibril protein precursors and extracellular tissue matrix components in the pathogenesis of systemic amyloidosis. Clin Imm and Immunopath 1994; 70: 2-9.
6. Tennent GA, Lovat LB, Pepys MB. Serum amyloid P component prevents proteolysis of the amyloid fibrils of Alzheimer’s disease and systemic amyloidosis. Proc Natl Acad Sci USA 1995; 92: 4299-4303.
7. Kisilevsky R and Young ID. Pathogenesis of amyloidosis. In: Reactive Amyloidosis and the Acute Phase Response (ed. G Husby). Bailliere’s Clinical Rheumatology 1994; 8, pp 613-26.
8. Hazenberg BPC and Van Rijswijk MH. Clinical and therapeutic aspects of AA amyloidosis. In: Reactive amyloidosis and the acute phase response (ed. G Husby). Bailliere’s Clinical Rheumatology 1994; 8, pp 661-690.
9. Puchtlen H, Sweat F and Levine M. On the binding of congo red by amyloid. J of Histochemistry and Cytochemistry 1962; 10: 355-363.
10. Van Antwept JD, O'Mara RE, Pitt MJ and Walsh S. Technetium - 99m diphosphonate accumulation in amyloid. J of Nuclear Medicine 1975; 16: 238-240.
11. Lee VW, Skinner M, Cohen A. Renal amyloidosis - Evaluation by gallium imaging. Clin Nucl Med 1986; 11: 642-6.
12. Gertz MA, Brown ML, Nauser MF and Kyle RA. Utility of gallium imaging of the kidneys in diagnosing primary amyloid nephrotic syndrome. J of Nucl Med 1990; 31: 292-295.
13. Hawkins PN, Lavender PJ and Pepys MB. Evaluation of systemic amyloidosis by scintigraphy with ¹²³I-labelled serum amyloid P component. New Eng J of Med 1990; 323: 508-513.
14. Hawkins PN. Diagnosis and monitoring of amyloidosis. In: Reactive amyloidosis and the acute phase response (ed. G Husby). Bailliere's Clinical Rheumatology 1994; 8, pp 635-659.
15. Woo P, O'Brien J, Robson M, Ansell BM. A genetic marker for systemic amyloidosis in juvenile arthritis. Lancet 1987; 767-769.
16. Schnitzer T and Ansell BM. Amyloidosis in juvenile chronic polyarthritis. Arth & Rheum 1977; 20: 245-252.
17. David J, Vouyiouka O, Ansell BM, Hall A, Woo P. Amyloidosis in juvenile chronic arthritis: a morbidity and mortality study. Clin and Exp Rheum 1993; 11: 85-90.
18. Teilum G. Pathogenesis of amyloidosis. Acta Pathologica Microbiologia Immunologia Scandinavica (A) 1964; 61: 21-45.
19. Ravid M, Shapira J, Lang R and Kedar I. Prolonged dimethylenphoxide treatment in 13 patients with systemic amyloidosis. Am Rh Dis 1982; 41: 587-592.
20. Livneh A, Zemen D, Siegal B. Colchicine prevents kidney transplant amyloidosis in familial mediterranean fever. Nephron 1992; 60: 418-422.
21. Janssen S, Van Rijswijk MH, Meijer S. Systemic amyloidosis: a clinical survey of 144 cases. Netherlands J of Med 1986; 29: 376-385.
22. Pras M, Bronshipigel N, Zemer D, Gafin J. Variable incidence of amyloidosis in familial mediterranean fever among different ethnic groups. The John Hopkins Medical Journal 1982; 150: 22-26.
23. Pras E, Aksentijevich I, Gunberg L, Balow JE, Prosent Dean M, Steinberg AD, Pras M and Kastner DL. Mapping of a gene causing familial mediterranean fever to the short arm of chromosome 6. New Eng J of Med 1992; 326: 1509-13.
24. Zemer D, Pras M, Sohan E, Modan M, Cabili S and Gafni J. Colchicine in the prevention and treatment of the amyloidosis of FMF. New Eng J of Med 1986; 314: 100-5.
25. Saatci U, Bakkaloglu A, Orens S and Besbas N. Familial mediterranean fever and amyloidosis in children. Acta Paediatrica 1993; 81: 705-6.

[Back to top]

ANKYLOSING SPONDYLITIS

By: Ashok Rai Senior Registrar Walsgrave Hospitals NHS Trust Coventry West Midlands

G R Struthers Consultant Rheumatologist Walsgrave Hospitals NHS Trust Coventry West Midlands

Summary

Ankylosing spondylitis (AS) is a chronic inflammatory disorder affecting the axial skeleton predominantly, although peripheral joint involvement may also be a significant feature. The disease affects synovial and cartilaginous articulations, and the sites of tendon and ligament attachment to bone (enthesopathy). In the primary care setting it is a clinical diagnosis supported by radiological evidence of sacro-iliitis.

AS and the seronegative spondylarthritides

The seronegative spondylarthritides (see Table 1) form a group of disorders which share a number of characteristic overlapping clinical features (Table 2) and an association with HLA B27. One therefore needs to be aware of these conditions as features of sacro-iliitis or spondylitis may occur later in, for example, patients with Reiter’s syndrome or psoriatic arthritis.

Table 1. Seronegative spondyloarthropathy

Table 2. Seronegative spondyloarthropathy

The association with HLA B27 is greatest with AS (over 95%) and weaker with psoriatic and enteropathic arthritis. Interestingly certain populations have a predilection to develop AS (eg Pima Indians) whilst others develop Reiter’s syndrome. An environmental trigger is also important as suggested by the observation that HLA B27 positive identical twins may be discordant for AS. Although HLA B27 may be important in antigen presentation the precise patho-aetiological role remains to be determined. One school of thought (Arthrogenic Peptide Theory) is that the B27 molecule acts as a receptor for an environmental (eg bacterial) peptide trigger. This leads to the generation of primed cytotoxic T lymphocytes which target sites where mimicking self-peptides are being presented by B27 on the surface of cells. Recent research with HLA B27 transgenic mice offers an animal model for AS. Further advances have been made in subtyping of HLA B27 using PCR (polymerase chain reaction) methods and in determining the three-dimensional structural analysis of the B27 molecule.

Diagnostic features

It is important to assess from the history and examination the features of inflammatory back pain as detailed in Table 3. The significance of distinguishing the causes of ‘low back pain’ is discussed in a previous ARC report (1).

Table 3. Features of inflammatory back pain

AS does affect women and, according to data from Bath, UK, the sex ratio prevalence becomes less marked in males as the age of onset increases from less than 16 years (M:F=6:1) to about 30 years (M:F=2:1).

The typical presentation of AS is a young adult with insidious onset of back-pain and stiffness. A misdiagnosis of mechanical pain with symptoms present for a few years is not unusual. The initial involvement of the sacro-iliac joint often produces pain that radiates to one or both buttocks, sometimes to the back of the knee or the thigh. Fatigue can also be a significant problem as a result of chronic discomfort which frequently disturbs the normal sleep pattern. Lumbar spondylitis is debilitating and progression to the thoracic spine also reduces chest expansion which may impair lung function further if there is associated interstitial lung disease. Enthesopathy may present with pain in the back of the heels (achilles tendinitis), heel pain on weight bearing in the mornings (plantar fasciitis), or ‘pleuritic’ chest pains (intercostal muscle insertions).

In chronic progressive disease there is loss of lumbar lordosis with increased thoracic and cervical kyphosis which becomes fixed as a result of fibrosis and bony ankylosis at which stage pain is generally less prominent.

Examination of the patient should include assessment of:

1) Axial spine involvement.

The following measurements are helpful.

(a) Modified Schober’s test of lumbar flexion (Figure 1).

Figure 1. Flexion of lumbar spine measured as increase in distance between points 5cm below and 10cm above the level of the dimples of Venus (DV).

(b) Finger-floor distance on trying to touch the toes. This test by itself is unreliable as it is more of a test for hip flexion.

(c) Occiput-wall distance or tragus-wall distance (Figure 2).

Figure 2. Wall-tragus (a) and wall-occiput (b) measurements.

(d) Chest expansion.

The serial assessment of these measurements (eg by physiotherapists) is useful to assess progress but other aspects of disability are discussed later.

2) Sacro-iliac (SI) joint examination

(eg springing of pelvis test).

Tenderness over other axial joints eg manubrio-sternal, costovertebral and symphysis pubis may also be present.

3) Peripheral joints, temporo-mandibular joints, shoulders and hips.

Assessment of flexion contractures of the hips and knees at an early stage is particularly important.

4) Enthesopathy

(eg achilles tendinitis, plantar fasciitis).

The intermalleolar straddle (ie distance between medial malleoli with hips abducted wide apart) is reduced in enthesopathy of the pelvic attachments, and in hip joint disease.

5) General examination in relation to the extra-articular features of AS

Self-administered assessments can also be useful, for example the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and the Bath Ankylosing Spondylitis Functional Index (BASFI). Enthesopathy can also be documented formally, eg using the Haywood Hospital Enthesitis Index. It is important to distinguish between disease activity/process measures, eg stiffness, and disease severity/outcome measures, eg assessment of deformity.

Extra-Articular features

General symptoms of fatigue, weight loss and low grade fever are common. Asymptomatic prostatitis is also evident in the majority of male patients. Recurrent acute uveitis (iritis) and conjunctivitis can occur early in the disease, particularly in those with peripheral arthritis, but unrelated to the severity of the spondylitis. Severe uveitis will require ophthalmological assessment and treatment with topical corticosteroids.

Chronic infiltrative and fibrotic changes in the upper lung fields associated with cough, sputum and dyspnoea are recognised in severe cases. Haemoptysis may occur following apical cavitation and infection with aspergillus producing a clinical picture mimicking tuberculosis, but this is rare.

Cardiac involvement may be clinically silent eg picked up on echocardiography, or cause significant problems. Post-inflammatory scarring may lead to aortic/mitral incompetence or conduction defects.

Other rare complications include an association with IgA nephropathy and the cauda equina syndrome secondary to spinal arachnoiditis.

Investigation

Routine blood tests may prove unhelpful. For example, a normal ESR does not exclude active disease. Non specific abnormalities include raised serum IgA levels. HLA B27 tissue typing is an expensive test which is not readily available in district general hospitals and is not recommended as part of the initial screening tests. In fact 8% of the normal population is HLA B27 positive and X-ray of the sacro-iliac joints is of equal diagnostic value. Occasionally (ie in about 20% of cases) patients present with an asymmetrical seronegative peripheral inflammatory arthropathy/enthesopathy - typically they are in the adolescent age group and may also be seen in the combined paediatric/rheumatology clinic. In such circumstances HLA B27 typing can be useful.

The X-ray changes affecting the sacro-iliac joints may develop some years after the onset of symptoms. Changes may be difficult to interpret in the teenage ‘immature’ pelvis or in the older patient with patchy ‘degenerative’ changes. One must not forget to distinguish the characteristic features of osteitis condensans ilii which may be seen incidentally in the multiparous female. Occasionally early changes affecting the SI can be seen on computerised tomography (CT) scanning with equivocal plain X-rays of the joints, but this investigation should not be over-used. In established cases with complications, however, CT scanning is useful in determining spinal fractures, spinal stenosis and thecal diverticulae.

Management

One cannot over-emphasize the importance of a number of general measures which, together with patient/ family education and awareness of the condition, can help to achieve a good prognosis in the majority of cases compared with the non-compliant, uninformed patients, some of whom may be from a poorer socio-economic background and require extra attention.

The objectives in early disease are to relieve pain and stiffness, and to maintain good posture and function. General exercises (eg swimming, extension exercises) are actively encouraged but specific measures are also necessary. Adverse conditions affecting posture at work and at home should be corrected. Occupational therapists can assess the need for any appliances, eg home aids, driving mirrors. Many hospital physiotherapy departments now run ‘spondylitis classes’, often in the evenings, on a regular basis.

Non-steroidal anti-inflammatory drugs provide symptomatic relief and slow release preparations taken in the evenings are helpful. Occasionally there is still a role for the use of phenylbutazone, with monitoring of the full blood count, on a named-patient prescription from a consultant rheumatologist. The use of sulphasalazine is mainly reserved for those with significant active peripheral joint involvement which has progressed despite other local measures including intra-articular joint injection. There is no convincing data or appropriate trials supporting the long term use of various drugs (eg methotrexate, pulse methylprednisolone, other second line agents) in the treatment of active spinal disease.

HLA B27 and genetic counselling

The majority of patients with AS are HLA B27 positive. The risk of a patient transmitting the same HLA B27 antigen to the child is 1 in 2. The risk of an HLA B27 relative developing AS is about 1 in 3. Hence the overall risk for the children to develop AS is around 1 in 6. However, sporadic disease (the majority of cases) tends to be more severe than familial disease. There may be additional genetic factors for disease ‘susceptibility’ or ‘severity’ and other loci under current investigation include HLA B60 and other Class II MHC genes such as TAP (Transporters associated with Antigen Processing) and LMP (Large Multi-functional Protease) genes.

Special considerations

1) Disability and psychological factors

Although much attention can be paid to the medical condition per se, the psychological aspects should not be forgotten. In patients with rheumatoid arthritis anxiety and depression have been shown not to be directly related to objective measurements of arthritis severity. Lack of social support and the experience of social stress are important factors for psychological adjustment.

Skeletal changes (eg kyphosis) make the impact of AS visible to others and may influence perception of self-worth. Problems of ‘body image’ (tending to be focused primarily on people with eating disorders) may also be relevant to psychological adjustment.

Work has shown that patients who perceived a sense of personal control over AS, despite the pain and debilitating effects of the disease, were less psychologically distressed than those who doubted their ability to exert control over various aspects of AS. ‘Doctor’ control belief was shown to play a generally weaker role in the process of psychological adjustment.

Instilling confidence in exercise regimes is also an important factor to ensure compliance.

2) Surgical intervention

About 20% of patients with juvenile onset AS develop significant hip problems requiring total hip replacement, within 20 years of disease onset. However this figure falls sharply with increasing age of onset and the risk of requiring joint replacement is quite small if onset occurs later on in the 30s. The outcome of hip replacement is good and should be considered as a practical option.

Vertebral wedge osteotomy is performed in specialist centres after careful consideration eg to correct marked flexion deformity where forward vision is severely impaired.

Acute painful exacerbation of back symptoms due to spondylodiscitis and/or spinal fracture should be recognised and treatment usually involves temporary restriction of spinal movements by using a corset if necessary.

References

1. Haslock I. The management of low back pain - a personal view. ARC Practical Problems May 1988, No.9, in Reports on Rheumatic Diseases, Series 2.
2. Laurent, R. Are there any anti-rheumatic drugs that modify the course of Ankylosing Spondylitis? Balliere’s Clinical Rheumatology 1990, Vol 4, 2, p387-400.

[Back to top]

BONE AND JOINT TUBERCULOSIS

Introduction

The prevalence of tuberculosis has been rising in the Western world over the last decade, and this is not just confined to specific groups such as the immuno- compromised, the elderly, recent migrants from endemic areas, and the socially disadvantaged. Although these incidence increases have mainly been in pulmonary tuberculosis, there has also been an attendant rise in osteoarticular disease.

Osteoarticular infection accounts for about 1-3% of all tuberculosis. Approximately 50% of osteoarticular cases are associated with concurrent pulmonary disease. A constant awareness of osteoarticular tuberculosis is necessary in order to avoid the diagnosis being overlooked. Timely recognition and management is essential to prevent loss of integrity and function of joints, and to reduce long-term morbidity.

Aetiology

Three species of mycobacteria are recognised as causing tuberculosis in humans - Mycobacterium tuberculosis (the commonest), M. bovis (spread by unpasteurised milk) and M. africanum (rarely found outside Northwest Africa).

Osteoarticular tuberculosis may occur following haematogenous spread or contiguous spread (for example from a lymph node), or as a reactivation of previous infection.

General features of bone and joint tuberculosis

Virtually any bone or joint may be involved. Ninety percent of cases have a monoarticular presentation with an insidious onset and slow chronic progression. Localised symptoms such as pain and swelling often precede radiologic changes and functional impairment. Cold abscess formation is common, especially if pulmonary disease is also present. Systemic features such as weight loss, fever, night sweats, fatigue and cough may be present. Variations in disease severity depend on the immune status of the person as well as the virulence of the organism.

Table 1. Major clinical features of osteoarticular tuberculosis

Diagnosis

A history of residence in an endemic area, or contact with tuberculosis, should raise the clinical suspicion for the diagnosis.

The definitive diagnosis follows the recognition of M. tuberculosis from any skeletal site. Biopsy is therefore the most useful procedure, for obtaining either bone or synovial tissue for histology and culture, and ideally both should be positive. Aspiration of joint fluid, or drainage of a cold abscess, may also provide evidence of infection with positive cultures being found in about 85% of cases where a joint is involved. Although there are no specific radiological features for osteoarticular tuberculosis, suggestive findings may include associated osteopenia, erosion, cyst formation, narrowing of the affected joint space and joint destruction occurring in the late stages (Figure 1).

Figure 1. X-ray features of joint tuberculosis. In this line diagram of a hip joint, the major radiological features are illustrated

In spinal tuberculosis magnetic resonance imaging (MRI) is particularly useful for evaluating soft tissue lesions, abscesses and intramedullary and extradural disease. By comparison, computerised tomographic (CT) scanning allows better evaluation of bone involvement, posterior extension of inflammatory tissue into the spinal cord and additionally allows position guidance for biopsies (Figure 2).

Figure 2. CT scan features of tuberculosis of the thoracic spine: illustrated on a transverse section through the thorax.

A raised ESR or plasma viscosity, an abnormal chest x-ray and evidence of pyrexia on temperature recording may assist in the diagnosis of tuberculosis. Sputum specimens, gastric lavage and early morning urine samples may provide additional sites for culture of the organism.

Malignant disease, pyogenic infections, gout, rheumatoid arthritis, fungal infections, systemic connective tissue diseases including vasculitis and brucellosis may mimic tuberculosis and should be considered in the differential diagnosis.

Specific features of bone and joint tuberculosis

Spondylitis (Pott’s disease)

Pott described tuberculous infection of the spine, and particularly the thoraco-lumbar region. In the author’s experience the sacroiliac joints are also commonly involved. Clinical features may include localised back pain and soft tissue swelling (‘cold abscess’), tenderness on vertebral percussion, reactive muscle spasm and truncal rigidity. Radiology may reveal end-plate sclerosis and erosion of vertebral bodies which may eventually progress to collapse and kyphosis. Nerve root and spinal cord compression may occur in advanced disease.

Appropriate chemotherapy remains the first line of management but with abscesses requiring drainage, especially if causing pressure symptoms.

Table 2. Diagnosis of osteoarticular tuberculosis

Peripheral arthritis

This usually occurs as monoarticular disease and most often affects weight-bearing joints. The ankle is frequently involved. Clinical features include chronic pain with progressive swelling and eventual loss of joint function. Infection may spread to the surrounding tendons and bursa.

It is important to recognise that joints previously affected by inflammatory or degenerative change may secondarily become infected by tuberculosis. Narrowed joint spaces and destructive changes with little reactive bone formation on x-ray are more suggestive of primary tuberculosis.

Osteomyelitis/dactylitis

Tuberculous inflammation may occur in any of the long bones, but most commonly affects the femur, tibia or ulna. Tuberculous dactylitis affecting the metacarpals or metatarsals is mainly seen in children, and usually presents as a localised swelling. A radioisotope bone scan showing increased uptake in the affected areas (‘hot spots’) is usually more helpful than x-rays in this situation.

Tenosynovitis/bursitis

Tuberculous infection of the tendon sheaths or bursa is comparatively rare. Where these structures are involved, however, the trochanteric, olecranon and retrocalcaneal bursae are the most common sites, as are the tendon sheaths of the dominant hand or wrist. Symptoms (mainly pain and swelling) are usually minor and slowly progressive, and hence patients may present only after months or years.

Biopsy is the most useful diagnostic technique, and the presence of small particles of degenerate tuberculous granulation tissue known as ‘rice bodies’ is characteristically seen on macroscopic examination.

Reactive arthritis (Poncet’s disease)

This is a sterile reactive polyarthritis, associated with active non-articular tuberculosis. It is believed to be an autoimmune response to M. tuberculosis. Symptoms usually regress once chemotherapy commences.

Principles of management of osteoaticular tuberculosis

All mycobacterial infections are notifiable conditions. It is essential that a physician with experience of treating tuberculosis, i.e. the thoracic specialist/ infectious disease specialist, is informed to initiate treatment and routine contact tracing. Prompt and appropriate chemotherapy is the cornerstone of effective management.

Table 3. Commonly used anti-tuberculosis drugs

General principles of management include prolonged treatment using at least three drugs to which the infecting organism is sensitive. At least one of the drugs should be bacteriocidal, such as isoniazid, rifampicin, streptomycin or capreomycin. Such drugs rapidly destroy the pool of infecting organisms. Bacteriostatic drugs, such as ethambutol or pyrazinamide, are used synergistically to eradicate the persistent slow-growing organisms and prevent subsequent relapse. Combination therapy also reduces the incidence of drug resistance.

Currently recommended initial treatment of adult tuberculosis is 300 mg isoniazid, 600 mg rifampicin and 1.5 grams pyrazinamide administered on a daily basis, for a period of 6 months. (The commonly used anti-tuberculosis drugs and their dosages and toxic effects, are summarised in Table 3). Some authorities advocate continuing treatment for osteoarticular tuberculosis for a longer period, and the author has persisted with therapy for 12 months when the clinical response appears to be inadequate or if complications have occurred (e.g. spinal tuberculosis requiring surgical intervention). Directly observed treatment has been shown to improve compliance, reduce relapse and prevent the emergence of resistant strains of the organism.

Immobilisation of affected joints with limited weight-bearing is recommended in some cases, for example if significant joint destruction has already occurred. Most patients are adequately treated as out-patients, and the therapeutic response is usually determined by clinical and radiological assessment. After completion of treatment, 6 monthly follow-up is recommended for two years.

Abscess drainage may be necessary for a number of indications, such as to relieve pressure symptoms. Severely damaged joints or unstable vertebrae may require arthrodesis or surgical fusion. Children should be managed using the same principles with appropriately adjusted dosages. It is recommended that a specialist paediatrician should be involved in the management from an early stage.

Management of HIV-positive cases

Tuberculosis is a common opportunistic infection in HIV-positive individuals. There is a high incidence of osteoarticular disease, and an atypical presentation may predominate. Mortality may be high and early diagnosis and treatment reduces this and diminishes the public health risk of tuberculosis for the general population. A standard 6 months regime is recommended, followed by isoniazid for life to prevent relapses.

Multi-drug resistant (MDR) tuberculosis

Approximately 1-3% of mycobacteria are resistant to isoniazid alone, and multi-drug resistance (MDR) is becoming an increasing global problem, with a poor prognosis and high mortality. The highest rates of MDR strains have been found in Nepal (48%), Gujarat, India (33.8%), New York (30%), Bolivia and Korea.

Both primary and acquired drug resistance may occur. The latter is more common and due to inappropriate treatment regimes and poor compliance. Drug sensitivity tests are strongly recommended for all patients with osteoarticular tuberculosis. Standard therapy should, however, be commenced whilst awaiting sensitivity results.

Treatment of MDR tuberculosis must be individualised, and at least four drugs to which the organism is sensitive must be used. Resistance to both rifampicin and isoniazid presents as a particularly difficult therapeutic problem. Collaboration with physicians with a particular interest in antituberculosis therapy, for example, thoracic physicians, is essential if the management of MDR tuberculosis is to be successful.

Atypical infections

Atypical mycobacteria such as M. kansasii, M. fortuitum, M. marinum, M. avium, M. chelonae and M. xenopi may occasionally lead to osteoarticular disease in man.

These organisms are of low virulence and tend to affect HIV-positive or immunocompromised individuals. The disease pattern is usually insidious with peripheral mono- or oligoarthritis.

The general principles of management are as those for M. tuberculosis infection. However, atypical organisms tend to be more frequently associated with drug resistance.

Table 4. Key points

Conclusion

The recognition of osteoarticular tuberculosis requires a high level of suspicion. Early diagnosis and treatment is essential to reduce morbidity and subsequent disability and mortality. HIV status, multiple drug resistance and atypical organisms should be borne in mind when treating non-typical patients, and collaboration with a specialist thoracic or infectious diseases physician, experienced in the management of tuberculosis, is strongly recommended in all cases.

[Back to top]

HERITABLE PROBLEMS OF THE SOFT CONNECTIVE TISSUES

Paul Wordsworth FRCP Clinical Reader in Rheumatology University of Oxford

Dorothy Halliday MRCP British Heart Foundation Clinical Research Fellow Wellcome Trust Centre for Human Genetics Oxford OX3 7BN

Introduction

Rheumatologists know the connective tissues as the battlefield on which many inflammatory diseases are fought. In contrast, the heritable disorders of connective tissue are much less familiar. However, it is important to have a working knowledge of their clinical features, classification and underlying biochemistry because they frequently present to musculoskeletal clinics. Abnormalities associated with these monogenic disorders include abnormalities of growth, joint laxity, deformity and pain which are symptoms only too familiar to rheumatologists.

These disorders are characterised by symptoms and signs in the soft tissue elements of the musculoskeletal system which can be highly variable. Although there are many well characterised syndromes within this group they frequently overlap not only with each other but also with the normal population. This can present significant diagnostic, and sometimes prognostic, difficulties which may only be clarified after careful clinical evaluation and appropriate specialist investigation. Considerable progress has been made in recent years in our understanding of the genetic and biochemical basis for many of these disorders and it is hoped that in the future this will form the basis of a more rational classification.

Marfan syndrome

The Marfan syndrome is an autosomal dominant disorder. It has an estimated prevalence of about 1 in 5000 with approximately 25% of cases representing new mutations. Life expectancy is reduced by about 40%, premature death nearly always being due to proximal cardiovascular disease. However, in recent years there has been a significant improvement in the prognosis which is largely attributable to greater awareness of the condition and improvements in the surgical treatments of the associated aortic and mitral valve problems. It is important to diagnose the condition early, particularly to try to reduce progressive dilatation of the aorta, aortic reflux and dissection of the ascending aorta. Diagnosis has until recently been based on the clinical criteria established in the Berlin nosology in 1986 (1). New criteria have been proposed to take into account increasing knowledge of the molecular basis and pathogenesis of the condition (2).

Table 1. Major criteria for the diagnosis of the Marfan syndrome

The diagnosis of the Marfan syndrome is clinical, requiring for a single index case characteristic involvement of at least three organ systems, including major manifestations in at least two of these. The mutations giving rise to the phenotype are in FBNI, a gene on chromosome 15 which encodes the extracellular matrix protein fibrillin, a major component of the microfibrils widely distributed in mesenchymal tissues. The defects are found predominantly in those connective tissues containing microfibrils, including the blood vessels of the proximal cardiovascular tree, the suspensory ligament of the lens and the skeleton. The cardinal features of the condition are therefore ectopia lentis, dilatation or dissection of the ascending aorta and a typical body habitus. A further major criterion is lumbosacral dural ectasia which is seen in up to 63% of those with the Marfan syndrome and very rarely in normal controls (3). Minor features include: pectus excavatum of moderate severity; a high arched palate with crowding of the teeth; joint hypermobility; mitral valve prolapse; spontaneous pnuemothorax and apical blebs; and recurrent hernias and striae (See reference 2 for more complete information on the revised criteria).

The revised criteria are also designed to improve diagnostic accuracy with regard to the first-degree relatives of those unequivocally affected. In these individuals in the presence of a relative, or a mutation in FBN1 associated with the phenotype, a major criterion in one organ system is required plus minor involvement of a second system. However, the degree of involvement in these systems varies widely both within and between families, producing a spectrum of disease ranging from almost normal to very clearly affected. This can cause major difficulties in diagnosis, emphasising the need for clear guidelines in the Marfan syndrome.

Fibrillin-1

As more becomes known about the FBN1 gene and the biochemistry of fibrillin-1 we hope to be able to define the limits of the disease phenotype more clearly. FBN1 is a large gene with 65 exons and spanning 110 kb of DNA. The coding sequence shows fibrillin-1 to be a modular protein comprised predominantly of two types of domain; epidermal growth factor (EGF)-like domains most of which are calcium binding, and 8-cysteine repeats (see Figure 1) (4).

Figure 1. Domain organisation of human fibrillin-1

Fibrillin is the main structural component of the microfibrils. Mutations in the gene have been shown to cause changes in microfibril morphology (5). To date about 100 different mutations have been identified in FBN1. As most families studied have a different mutation, and because of the size of the gene, mutation detection is not feasible on a routine basis. The mutations occur throughout the gene and are most commonly mis-sense mutations where, due to a single base change, an alternate amino acid is encoded. For classic Marfan syndrome these mutations frequently substitute a highly conserved amino acid on one of the calcium-binding EGF domains. This impairs the ability to bind calcium, which seems crucial to the structural integrity and function of the protein (6).

There are several conditions which need to be differentiated from Marfan syndrome. These are congenital contractural arachnodactyly (CCA), familial thoracic aortic aneurysm or dissection, familial ectopia lentis and familial Marfan-like habitus. Apart from CCA, which is caused by mutations in the related protein FBN2 on chromosome 5, mutations have been found in FNB1 in all these conditions (7, 8). In contrast, the MASS phenotype (Myopia, Mitral valve prolapse, Aortic root at the upper end of normal, and Skin and Skeletal involvement) shows involvement of two or more systems but no major criteria (2). Many individuals previously borderline for Marfan syndrome will fit accurately into one of these categories. The diagnosis of each of these three familial disorders depends on the presence of dominant criteria from the same organ system in at least two affected individuals in the family.

Treatment

Much work is still needed to understand the pathogenesis of Marfan syndrome. The main practical aim is to identify individuals with the condition early so that those at high risk can receive appropriate treatment. They should be counselled about avoiding contact sports and activities involving anaerobic effort or Valsalva manoeuvres. Regular echocardiographic measurement of the diameter of the aortic root should be carried out and also, where indicated, prophylactic beta-blockade, as there is increasing evidence that early intervention can slow the progression of the aortic dilatation and thus reduce morbidity. Pregnancy is better tolerated in the early twenties and should be monitored by echocardiography every eight weeks. Early cardiological advice is appropriate and, should surgical intervention be necessary because of aortic insufficiency, this is much better undertaken electively than in a crisis. Orthopaedic complications are not rare: some patients develop scoliosis of such severity that spinal surgery may be appropriate; others may require surgical footwear or surgery to correct deformity; occasionally hormonal treatment to accelerate puberty may be appropriate for those who are exceptionally tall; and multiple extractions for dental crowding are common. Refractory errors need careful correction and more than 90% of patients can be corrected to vision better than 20/40 without lens surgery. Careful monitoring for glaucoma, cataract and retinal detachment is advisable.

Finally, it should be remembered that if an accurate diagnosis can be made this may exclude the diagnosis in individuals who had previously been considered to be affected. The importance of this should not be underestimated in a condition with such an adverse prognosis. Many individuals in the general population exhibit some of the features commonly associated with the Marfan syndrome. The use of the term ‘Marfanoid’ to describe them is to be deprecated. All possible attempts should be made to define precisely whether they are affected. If they do not fulfil the appropriate criteria they can be reassured, while, if they are affected, the appropriate monitoring and treatment as necessary can be implemented. The related conditions generally do not carry the same adverse prognosis because of the relatively mild cardiological involvement. This is true, in particular, for the MASS phenotype. It is useful to be aware of the existence of these conditions, which breed true in families, when considering the differential diagnosis of the Marfan syndrome.

Ehlers danlos syndrome

This syndrome comprises a variety of conditions affecting the soft connective tissues including the blood vessels, skin, tendons and ligaments (Table 2). The aggregation of certain clinical features such as excessive skin fragility, joint laxity, bruising and skin hyperelasticity were noted in early descriptions by Ehlers and Danlos at the beginning of the century. Although ten clinical variants are currently recognised, fewer than half of the cases seen in practice can be easily accommodated within these descriptions. Indeed, some of the subtypes are very rare, with some doubt as to whether they all represent discrete entities. There is also considerable overlap with other disorders involving the soft tissues including Marfan syndrome (which is associated with joint hypermobility, striae and mitral valve prolapse) and osteogenesis imperfecta, in which a variable degree of skin fragility and joint hypermobility is not uncommon.

Table 2. Ehlers Danlos syndromes

Fibrillar collagens

The collagens are structural proteins characterised by the presence of triple helical domains with the amino acid sequence glycine-x-y where x and y are often proline or hydroxyproline. The fibrillar collagens have clear stress-bearing functions in both the hard and soft connective tissues; a large number of specific mutations have now been described in these proteins in many heritable connective tissue disorders. Genetic, biochemical and ultrastructural analysis of collagen has incriminated the fibrillar collagens type I, II, III and V in Ehlers Danlos syndrome, while in some variants enzymes involved in the post-translational modification of collagen (essential for the formation of stable cross-links) are incriminated.

Variation in the clinical phenotype

Although there are certain unifying features across the spectrum of Ehlers Danlos syndrome there is significant variation between the major subtypes. Severe bruising, skin fragility and ligamentous laxity may preclude participation in contact sports for the majority of patients affected and premature rupture of the membranes in pregnancy is to be expected in the more severe variants. The most severe disease is caused by the type III collagen deficiency found in EDS IV. Although excessive bruising and thin skin in infancy is relatively uncommon and complications are rare before the age of 20 in this variant, the average age of death is in the early 30s, with 90% of deaths caused by arterial haemorrhage. Less commonly, rupture of the bowel (particularly the sigmoid colon) occurs and uterine rupture is also well recognised. Surgery for any of these complications is difficult, because of the fragility of the tissues.

EDS VI is particularly interesting because it was the first collagen disorder to be characterised at the molecular level. It is uncommon and has quite pronounced clinical features, amongst the most striking of which is ocular fragility. Intracerebral and other haemorrhages are also relatively common in this variant. EDS VII can be associated with marked short stature and substantial disability resulting from dislocation of the hips and other joints. It is most typically caused by mutations in COL1A1 and COL1A2, the structural genes for type I collagen (see below), but very rarely the same condition can result from an enzyme deficiency when it is associated with very pronounced joint laxity, bruising, skin fragility and blue sclerae.

Genetics and biochemistry

A variety of mutations have now been described in the fibrillar collagens (types I, II and V) in several variants of EDS. Interestingly, some individuals with osteogenesis imperfecta, which is linked to type I collagen (COL1A1), have striking features of EDS and there are a few reports of mutations in COL1A2 causing relatively mild EDS-like features. A very specific type of mutation in either COL1A1 or COL1A2 is responsible for the majority of cases of EDS VII. Typically a mutation in one of these genes causes exon skipping with the deletion of exon 6 from the corresponding mRNA. This contains the cleavage site for the N-terminal propeptide, which is consequently retained (mutations in COL1A1 are known as EDS VIIA, in COL1A2 as EDS VIIB). In a very small minority a deficiency of enzyme procollagen N-proteinase is responsible for a somewhat more severe form of the disease (EDS VIIC).

Perhaps the best studied abnormality is that relating to type III collagen in EDS IVix. In this, deficient synthesis of type III collagen can be attributed to a variety of mutations in the COL3A1 gene; these include glycine substitutions in the triple helical domain, exon skipping and other deletions. The defect can be established by finding deficient synthesis of type III collagen in fibroblast culture, which can be used for prenatal diagnosis. More recently, linkage studies have shown that EDS I/II is linked to one of the type V collagen genes (COL5A1) and in some cases specific glycine mutations in type V collagen have been described (5,6).

EDS VI, the earliest-characterised collagen abnormality, is caused by a deficiency of the enzyme lysyl hydroxylase on chromosome 5. Specific mutations within this gene have been described and for prenatal diagnosis lysyl hydroxylase activity can be studied from fibroblast culture. EDS IX is also associated with abnormal lysyl hydroxylase activity and is now known to be caused by mutations in the same gene as Menke’s syndrome. However, in this case the primary abnormality is, in fact, on copper metabolism, since copper is an important co-factor for lysyl hydroxylase and lysyl oxidase activity. The genetic defect lies on the X chromosome in a gene encoding a cation membrane transport ATPase which is involved in copper handling. The other associated clinical features (including a degree of mental retardation) probably reflect the importance of copper in a variety of other enzyme pathways.

Conclusions

The heritable disorders of connective tissue represent for many the ultimate nightmare in nomenclature with a mixture of acronyms conspiring to complicate the field still further. In this review we have concentrated on those conditions which have a propensity to involve the soft connective tissues and skeleton to a significant extent, excluding disorders such as cutis laxa, pseudoxanthoma elasticum and epidemolysis bullosa (caused by mutations in type VII collagen) which mainly involve the skin. In recent years we have begun to see real hope of reclassification on the basis of the underlying genetic and biochemical defect which has yielded insights into the molecular basis of these diseases and has also helped very considerably in their recognition, diagnosis and classification.

References

1. Beighton P, de Paepe A, Danks D et al. International nosology of heritable disorders of connective tissue, Berlin, 1986. American Journal of Medical Genetics 1988; 29: 581-594.
2. de Paepe A, Devereux RB, Dietz H et al. Revised diagnostic criteria for the Marfan syndrome. American Journal of Medical Genetics 1996; 62: 417-426.
3. Pyeritz RE, Fishman EK, Bernhardt BA et al. Dural ectasia is a common feature of the Marfan syndrome. American Journal of Human Genetics 1988; 43: 726-732.
4. Handford PA, Mayhew M, Baron M et al. Key residues involved in calcium binding motifs in EGF-like domains. Nature 1991; 351: 164-167.
5. Kielty CM, Rantamaki T, Child AH et al. Cysteine to arginine point mutation in a "hybrid" eight-cysteine domain of FBN1: consequences for fibrillin aggregation and microfibril assembly. Journal of Cell Science 1995; 108: 1317-1323.
6. Downing AK, Knott V, Werner JM et al. Solution structure of a pair of calcium binding epidermal growth factor like domains: implications for the Marfan syndrome and other genetic disorders. Cell 1996; 85: 597-605.
7. Kainulainen K, Karttunen L, Puhakka L et al. Mutations in the fibrillin gene responsible for dominant ectopia lentis and neonatal Marfan syndrome. Nature Genetics 1994; 6: 64-69.
8. Dietz HC and Pyeritz RE. Mutations in the human gene for fibrillin-1 (FNB1) in the Marfan syndrome and related disorders. Human Molecular Genetics 1995; 4: 1799-1809.
9. Pope FM, Martin GR, Lichtenstein JR et al. Patients with Ehlers Danlos syndrome type IV lack type III collagen. Proceedings of the National Academy of Science 1975; 72: 1314-1316.
10. Loughlin JK, Irvine C, Hardwick L et al. Linkage of type II Ehlers Danlos syndrome to the COL5A1 locus. Human Molecular Genetics 1995; 4: 1649-1651.
11. Burrows NP, Nicholls AC, Yates JRW et al. The gene encoding collagen a1(V) (COL5A1) is linked to mixed Ehlers-Danlos syndrome type I/II. The Journal of Investigative Dermatology 1996; 106: 1273-1276.

[Back to top]

OSTEOARTHRITIS - CLINICAL FEATURES & MANAGEMENT

Dr C W Hutton FRCP Consultant Rheumatologist Mt Gould Hospital Plymouth

The nature of Osteoarthritis

Osteoarthritis is a condition characterised by hyaline cartilage damage, changes in subchondral bone of sclerosis and osteophyte formation. There is considerable controversy about the mechanisms and significance of these changes. More understanding may, in future, enable interventions that reduce the destructive changes, slow or even reverse progression. However presently there are no interventions which are known to alter the long term outcome. There are however many possible approaches to ameliorate the clinical problems the disease causes.

Epidemiological and clinical observation emphasise the diversity of the condition and its natural history. Changes are common in the population but relatively few have symptoms. Even fewer have a painful disabling disease that presents to doctors.

Groups and subgroups

Unlike most joint diseases there is no shortage of known conditions that can cause osteoarthritis. This has led to the concept of osteoarthritis being a final common pathway of many different joint problems. It is seen as the process that a joint undergoes as it responds to injury. Elements of this may be adaptive, others are maladaptive. The balance results in stable non-progressive and often asymptomatic disease or progression to joint destruction.

Table 1. Secondary Osteoarthritis

In many people the disease occurs spontaneously and no previous cause can be identified. In these people there are however associations and patterns in primary, idiopathic osteoarthritis. Its incidence increases with increasing age. Some joints are often involved whereas others are rarely affected. Different patterns suggest there may be subgroups within primary osteoarthritis. The presence of multiple joint involvement suggests some people have a generalised osteoarthritis. The commonest distribution is bilateral involvement of the thumb carpometacarpal joint, the interphalangeal joints, the knee and metatarsophalangeal joint of the great toe. This group in turn has been subdivided depending on the presence of bony swellings on the dorsum of the distal interphalangeal joints: Heberden’s nodes.

The subgroup, generalised nodal osteoarthritis, is particularly common in women. Its onset is in middle age often with painful inflamed Heberden’s nodes. These may settle leaving a bony swelling. The onset around the menopause suggests an endocrine cause but there is no evidence of oestrogen-dependent effects. Whether hand osteoarthritis is associated with an increased risk of developing major joint disease is still unknown. An apparent association may be coincident involvement of common independent sites, and may be better termed a mono arthritis multiplex.

Figure 1. Pathological features of OA.

The concept of generalised disease is helpful clinically as an unusual pattern of joint involvement should suggest a possible secondary cause. Prominent metacarpophalangeal joint disease may be associated with haemochromatosis and pyrophosphate deposition. Shoulder involvement also suggests pyrophosphate deposition. It may also indicate that a second different inflammatory arthritis, such as rheumatoid, is developing in someone with pre-existing osteoarthritis.

Figure 2. Pain around the hip and knee joints.

Relationship with joint use

Confusion remains about how use influences the development of osteoarthritis. The classic study of Hadler suggests that use influences at least the pattern of involvement. Severe trauma with intra-articular fracture produces rapid onset disease. It is unclear what the tolerance of a normal joint to peak or repetitive loading is, what happens if the person is trained to high fitness and what happens if there is already joint damage. Nevertheless the importance of maintaining muscle means activity should be encouraged.

Clinical problems

The two major clinical problems are pain and impaired function.

Pain

Hyaline cartilage has no nerve supply but many of the other structures in the joint are richly innervated although the mechanisms of pain are only partially understood. The pain is often described as worse with joint movement or weight bearing. There may be severe pain on initial movement that eases and then worsens with continuing activity. The pain may be localised or be more diffuse. It may be referred to an unaffected joint. Hip pain may present with knee pain or vice versa.

Initially pain may be variable, with periods that are symptom free. Sudden exacerbations that may be extremely severe may occur. This may be associated with joint swelling and a coincident episode of some other type of arthritis must be considered, such as acute pseudogout. Single joint presentation must initially be investigated as a subacute infection and diagnosed as osteoarthritis by exclusion.

Pain is associated with soft tissue structures like ligament and tendon insertions and bursae. This gives important treatable pain patterns different with each joint. Examples are the lateral pain of the trochanteric bursitis in the hip and the medial focal pain of an anserine bursitis of the knee (Figure 2).

Later in disease pain becomes constant; it may be described as burning and may be associated with night pain. This should be distinguished from pain on movement at night. Both, however, produce sleep disturbance, sleep deprivation and amplify the pain problem.

Dysfunction and handicap

The loss of function produced by an osteoarthritic joint varies with the joint. The handicap that results will reflect a multitude of interacting factors. This means that apparently identical joint pathology may give a wide spectrum of clinical problems. In the leg instability, a sense of instability, and joint locking combine with pain to decrease mobility. In the arm weakened power of grip and problems with dexterous movement may make it difficult to continue to work.

Assessment

The diagnosis is by exclusion of other diseases and is confirmed by radiography. Three clinical questions need to be considered:

(1) Is there a contributing or complicating disease?

(2) In particular is there pyrophosphate or urate crystal deposition?

(3) Is there a resultant or contributing mechanical factor?

Look for instability of the joint, leg length inequality, marked valgus or varus deformity. Simple corrective orthosis may reduce pain. Hallux rigidus may be made less symptomatic by a rocker sole.

Is there associated deformity and muscle wasting?

Much of the disability results from secondary weakness which if partially corrected may markedly reduce handicap. Effective motivation of the patient to improve muscle power is a major challenge to the clinician.

Figure 3. Distribution of joints involved in primary osteoarthritis.

Management

As no treatment can influence the disease progression, treatment might seem pointless. There is a widespread sense of nihilism on one side and misplaced enthusiasm for fashions on the other. Most interventions are untested, reflecting the difficulty of doing long-term studies in a complex variable condition. Clinical management must be based on being logical and not doing harm.

The principles are the same as managing any chronic disease. Educate the patient to understand the problem. Identify realistic expectations for interventions. Simple understanding of the diagnosis and low probability of a crippling outcome will reassure many. It is important for the patient to create a sense of control over his or her disease.

Second is to correct any contributing factors which have been identified – the primary cause, the dysfunction with muscle strengthening exercises and orthotics. A walking stick of the correct height held in the contralateral hand and a non-worn ferrule will allow weight to be deflected on to the good leg. Correct flexion deformities; with exercise or possibly using hydrotherapy; a splinting system such as continuous passive movement that gradually works the joint straight, or flowtron splints that squeeze it straight. Unfortunately contractures recur readily as enthusiasm for maintenance exercise has a short half life! Emphasise the importance of muscle strengthening.

The third objective is to control pain. The doctor or therapist should work with the patient to analyse their pain and work towards a ‘pain strategy’. There is a set of modalities to try. Simple low toxicity analgesia, possibly in combination, should be tried first. Anti-inflammatory drugs are then added even though it is often regarded as a non-inflammatory condition. If the drugs have a poor effect they should be discontinued. Different strategies suit different individuals. The dilemma of giving long-term non-steroidals to patients with a history of peptic ulcer with or without use of gastro-protection remains. Evidence is beginning to support using misoprostol, H2 blockers or proton pump blockers with non-steroidals in the elderly. However this needs continuous review as new trial evidence becomes available.

Joint effusions are common, particularly in the knee. They may cause discomfort, be associated with pain and are associated with reflex quadriceps wasting. This sets up a cycle of weakness and instability. The effusion may be inflammatory but often has a low cell count and probably results from changes in the fluid flows in the synovium. They often respond to intra-articular injection of crystalline steroids such as triamcinolone, but they often recur. Repeated injection in the joint must be done with caution as concern about promoting collapse of the subchondral bone remains. Intra-articular injection is commonly used in joints to reduce pain, with widespread anecdotal evidence of transient effectiveness. Great care must be taken not to damage periarticular structures as the joint may be very disorganised and fibrosed making location of the needle difficult.

Correct other contributing lifestyle factors... encourage exercise and control of weight.

Anticipate future problems particularly in relation to potential major joint surgery. One common difficulty with lower limb surgery is leg ulceration, so care is necessary to prevent ulcers.

Surgery

The outlook for patients with severe hip and knee osteoarthritis has been transformed by arthroplasty. At other large joints surgery is possible but not so well established. Elbow and shoulder replacement are increasingly routine. In the foot and hand arthroplasty is less well established.

The indication is uncontrolled pain. In the hip this is often clear, but knee pain is more difficult. It seems to phase more gradually from tolerable to intolerable. Milder disease is extremely common. With monoarticular joint disease and a considerable expectation of activity after surgery, loosening may occur early. So even though the procedures are technically feasible there is concern that early surgery may increase the risk of difficult later revision. Procedures such as osteotomy may buy time in a younger patient. In elderly patients with multiple pathology the risks of the procedure, particularly perioperative vascular events, need to be weighed carefully against potential benefits. Despite anti-thrombotic prophylaxis young people can die from pulmonary embolism.

Figure 4. Pain amplification in osteoarthritis

 In the smaller joints interventions are more difficult. The various operations available for hallux rigidus underline the varied results. Controlled studies are needed. Various procedures including arthroplasty, tendon reorientation, fusion or scaphoid exision have been proposed for thumb base arthritis, but should be considered only in persistent severe, painful disease not responding to intra-articular injection.

The future

The outlook for osteoarthritic patients is changing. At last the conceptual challenge of influencing the natural history of the process has been accepted and research is gearing up to understand the biology in a way that should lead to therapeutic advances. However, even with more effective intervention, the principles of care will be similar to those we should follow today.

Further reading

Hadler NM, Gillings DB, Imbus HR, Levitin PM, Makuc D, Utsinger PD, Yount WJ, Slusser D, Moskovitz N. Hand structure and function in an industrial setting. Influence of three patterns of stereotyped repetitive usage. Arthritis & Rheumatism 1978; 21(2): 1019-1025.

Bland JH, Cooper SM. Osteoarthritis. A review of the cell biology involved and evidence for reversibility. Management rationally related to known genesis and pathophysiology. Seminars in Arthritis & Rheumatism 1984; 14(2): 106-133.

Peyron JG. Epidemiologic and etiologic approach of osteoarthritis. Seminars in Arthritis & Rheumatism 1979; 8(4): 288-306.

Sokoloff L. The pathology of degenerative joint disease. Chicago University Press, (1969).

[Back to top]

PAEDIATRIC GOUT: ROLE OF ABERRANT PURINE AND PYRIMIDINE METABOLISM IN RHEUMATIC AND ALLIED DISORDERS

H Anne Simmonds Purine Research Laboratories Clinical Science Laboratories, UMDS Guy’s Hospital, London

Summary

Gout and uric acid stones are traditionally associated with the middle-aged male. Not so widely recognised is the fact that such symptoms, together with crystalluria, can occur in children with genetic purine or pyrimidine enzyme deficiencies. Three such genetic defects can present as paediatric gout, sometimes as renal failure, often both, two unusually affecting both sexes. Others cover the full spectrum of human disease, and include immunodeficiency, neurological deficits, or any combination of these and can present at any age. Owing to their relatively recent description the majority are little known. This diagnostic problem is compounded by a) the fact that the symptoms are common to other genetic metabolic disorders, or non-genetic disorders (eg AIDS or rheumatoid arthritis); b) the considerable genetic heterogeneity, with infinite variation in the phenotype and its biochemical expression. Crystals on the nappy, in the urine, or evident by renal ultrasonography, have sometimes provided the first clues to the underlying disorder. Purine and pyrimidine defects can also be the unsuspected basis of intolerance of analogue therapy. Knowledge derived from these disorders may assist the treatment of more common conditions. The importance of these pathways to the immune response stimulated the development of a new generation of purine and pyrimidine analogues targeting the enzyme defective in a particular disorder. Several are now in clinical trial in rheumatoid arthritis and organ transplantation.

Introduction

The first known purine disorder, ‘primary’ gout, a disorder manifest typically in the over-indulgent, overweight, hypertensive middle-aged male, and beloved by 19th century cartoonists, has been known for centuries. It is a disease of plenty, as evidenced by the fact that it was almost unknown during and immediately after the first and second world wars. This type of gout results directly from over-indulgence in purine-rich foods and indirectly from the manner in which the human kidney handles uric acid. Humans have no requirement for dietary purines. The intestinal mucosa rapidly degrades dietary nucleotides, nucleosides and bases to the extremely insoluble metabolic waste, uric acid, rather than the more soluble allantoin of other species (1).

The recognition of genetic defects affecting the pathways of purine and pyrimidine synthesis or degradation is relatively recent (1,2). The biochemical basis for the first purine defect to be identified, xanthinuria (XDH deficiency), was established only in 1959. An unusual form of genetic gout, a dominant disorder affecting young men, women and children, now termed familial juvenile hyperuricaemic nephropathy (FJHN) was also first described in 1962. By 1973 the number of disorders identified was still only six. All owed their recognition, then as now, to the fact that they can present as either crystalluria, or gout, hypeuricaemia, kidney stones, or renal failure; alone or any combination of these (1-3). The other four were partial and complete deficiency of the enzyme hypoxanthine-guanine phosphoribosyltransferase (HPRT), its companion X-linked disorder phosphoribosylpyrophosphate superactivity (PRPS), and the refractory anaemia sometimes associated with immunodeficiency, hereditary oroticaciduria (1-4).

Purine and pyrimidine disorders are more than gout, crystalluria or kidney stones

Since the 1970s there has been an exponential increase in the number of such disorders recognised. They now number 23 - only a few of which have no apparent clinical consequence. Their recognition followed the tremendous technological advances which made possible identification of the specific metabolites, or the absence of normal metabolites, which accumulate in consequence of these defects. These disorders cover the full spectrum of human disease. This underlines the central importance of purine and pyrimidine metabolic pathways for many cellular functions and accords with their early origins (2).

Cellular purines and pyrimidines are derived almost exclusively from endogenous sources and are anchored intracellularly, predominantly as the ribonucleoside triphosphate. They are best known as the energy supply for many cellular reactions (ATP), or the DNA and RNA essential for cell replication and protein synthesis. In addition, they play an important role in signal transduction, translation and membrane biosynthesis. They are built up de novo from simple precursors by energetically expensive, multi-step pathways. Alternatively, and more generally, these ribo- and deoxyribonucleotide pools are sustained by the recycling of the basic ring structure derived from their catabolism during the normal processes of cell turnover - red cell senescence, wound healing, exercise, and so on. The latter, the so-called ‘salvage’ pathway is a single-step, less energetically expensive process, which normally exerts feed-back control on the de novo path. The metabolic pathways of purine metabolism and the defective enzymes identified so far are summarised in Figure 1.

Such recent description means that many of these disorders are little known in the clinic or district hospital laboratory. This problem is compounded by the fact that they can have multiple forms of presentation.The considerable genetic heterogeneity in these disorders means that they can present at birth, leading to early death; or, with milder forms, up to the 80s (2).

Figure 1. Schematic representation depicting the metabolic pathways of purine metabolism which sustain the principal purine pools (ATP, DNA, RNA, GTP) in humans. These include the 10 step de novo synthetic route (synthesis) and the routes of nucleotide interconversion, breakdown and recycling (salvage). Metabolic end products (2,8-hydroxyadenine, xanthine, uric acid) are not recycled, but degraded or excreted, principally in the urine (degradation). The eight different steps in these processes, which have been identified as defective are indicated by a minus sign, or an upward-pointing arrow to denote superactivity, after the appropriate abbreviation. summary of the associated clinical manifestations is given at the top left: ADA, adenosine deaminase; PNP, purine nucleoside phosphorylase; AMPDA, adenylate deaminase; ASA, adenylosuccinate lyase; HPRT, hypoxanthine phosphoribosyltransferase; GPRT, guanine phosphoribosyltransferase; PRPS, phosphoribosylpyrophosphate synthetase; APRT, adenine phosphoribosyltransferase; XDH xanthine dehydrogenase.

Figure 2a. Model of urate transport in the proximal tubule derived from studies in brush-border membrane vesicles from the human kidney: 1) Intraluminal filtered urate is exchanged at the brush-border membrane against an intracellular anion (X) by the urate anion-exchanger. 2) Urate leaves the cell through the basolateral voltage-sensitive pathway. 3) Part of the urate which has entered the tubular cell can return to the lumen by the brush-border voltage-sensitive pathway. 4) This mechanism shows basolateral entry of urate into the cell by a transport system which resembles (but is distinct from) the PAH transporter. This mechanism is as yet undefined. 5) Luminal Na+-anion co-transport is shown, involving anions (X) such as lactate, nicotinate etc.

Figure 2b. Schematic diagram depicting the net reabsorption occurring in the proximal tubule of the human kidney in healthy adults (mean values for men 92%, women 88% and children 82%) which means that only 8%, 12% and 18% respectively of the filtered urate is excreted in the urine.

Role of the kidney in the genesis of gout

The ‘primary’ gout of the middle-aged male needs little further description here. Diagnosis and treatment is well documented, with the exceptions discussed below. However, understanding the underlying problem in ‘primary’ gout is relevant to the other genetic disorders described.

The body pool of uric acid in humans, and hence the plasma urate concentration is the result of a balance between endogenous production, dietary ingestion and excretion. Recent studies have expanded our knowledge of the role of the kidney in the genesis of gout. This relates to the other peculiarity humans display: namely that the renal tubule reabsorbs around 90% of the filtered uric acid (Figure 2b). Net reabsorption is slightly higher in normal males (mean 92%) than in females (88%) and, except in the newborn, is lower in children of either sex (80-85%). This explains why gout is extremely rare in women and children, but means that uric acid is already circulating in the plasma of adult males at concentrations close to its solubility limit (420mmol/l). Net reabsorption is even higher in subjects with ‘primary’ gout (> 94%), hence their greater susceptibility to dietary purine intake (1). Several different models to explain the handling of uric acid in the proximal tubule of the human kidney, involving a combination of reabsorption, secretion and post-secretory reabsorption were proposed by earlier workers (1). Progression from a two- to a four-component model resulted from studies using pyrazinamide and probenecid.

A new picture has emerged from the recent recognition of a urate anion-exchanger in brush border membrane (BBM) vesicles of the human kidney (Figure 2a). This anion-exchanger also has a high affinity for lactate, ketones and related organic acids (5) and is absent in species which secrete urate, such as the pig and rabbit. Secretion in these species, and to some extent in humans, is via the voltage-sensitive pathway which transports urate from the cell to the tubular lumen. Secretion is favoured because of the electro-negativity of the cell relative to the tubular lumen. Dietary compounds or drugs with an affinity for the anion-exchanger also enhance urate reabsorption and thereby reduce the renal clearance of uric acid. Iatrogenic gout due to treatment with hypotensive agents, including diuretics over long periods, currently accounts for over 50% of new patients attending gout clinics. The number of elderly females in this group is unusually high and the clinical presentation frequently atypical (6).

The only way to establish the exact contribution of diet is to place the subject on a purine-free diet for 7 days and measure the urinary excretion of urate over 24 hours. Excretion will then equal endogenous production (1). In this way fewer than 5% (some estimates are as low as 1%) of patients with gout have been found to excrete abnormally large amounts of urate (>3 mmol/day). A more acceptable yardstick is to persuade the patient to adhere to a low purine diet (one non-organ meat meal per day) and then measure the 24h uric acid excretion relative to creatinine. A ratio in excess of 0.35 in adults is suggestive of an underlying defect of purine metabolism associated with genetic uric acid over-production. This requires confirmation and work-up in a specialised centre. It is important for clinicians to be aware of this, especially when presented with a young patient with gout, or an older male with a history dating back to adolescence, particularly if siblings are involved. Treatment will differ.

Gout in childhood or adolescence is usually genetic and should be investigated

The message here is that gout in the young should always be investigated. These rarer forms of genetic gout, as mentioned above, can be the manifestation of three disorders. Even more unusual is the fact that two of them can affect males and young females equally (2-4,7-9). Unrecognised and untreated, or not treated appropriately, they can progress to dialysis, transplantation or even death - tragic for the families and costly for local health services. This type of gout presents typically in adolescents, children and even infants, sometimes secondary to acute renal failure in the first weeks of life. There is infinite variation not only in the phenotype, but in its biochemical expression. This is clearly evident in the X-linked disorder HPRT deficiency, characterised by the severe Lesch-Nyhan syndrome (LNS; complete HPRT deficiency) at one end of the spectrum (3). Adolescent gout or kidney stones only are found at the other (Kelley Seegmiller syndrome or partial HPRT deficiency), with varying degrees of neurological involvement in between (1-3). LNS is frequently diagnosed only after presentation with gout or acute renal failure in an infant (Figure 3), or in adolescents institutionalised for ‘cerebral palsy’ of unknown cause (2). The latter have been missed because of a normal plasma uric acid found at screening. This is due to the high renal clearance characteristic of children, a factor frequently overlooked (7). The same heterogeneity is evident in the de novo synthetic defect, PRPS. This is characterised in childhood by a variety of neurological deficits, including the inability to walk or talk, abnormal facies, sometimes inherited nerve deafness (1,2,6). Presentation only with gout or stones can occur in early adulthood. PRPS, unlike HPRT deficiency, may also manifest in the carrier female. Thus PRPS should be suspected in any seemingly X-linked defect where HPRT is normal and the mother presents with gout or hyperuricaemia and may be deaf (6).

Figure 3. Acute gout in the first, second and third fingers of the left hand of a one month old boy presenting in acute renal failure of unknown cause. This eventually drew attention to the possibility of a purine metabolic disorder as the underlying basis of the symptoms. The baby was found to have no detectable HPRT activity in intact or lysed erythrocytes and subsequently to have the full Lesch-Nyhan syndrome.

The third condition is FJHN (familial juvenile hyperuricaemic nephropathy), which typically presents in childhood or adolescence, affecting females and males equally (2,7-9). FJHN differs from the two defects above in that it is a dominant disorder and is not associated with uric acid overproduction, but gross underexcretion. Hitherto, unrecognised and treated only as ‘a familial renal disease’ rapid progression to dialysis, transplantation, or death in the 30’s was frequent (2,7). Awareness is important in the recognition of FJHN and explains why more patients have been identified in the UK than have been reported collectively worldwide. The genetic basis is unknown, but FJHN must be excluded in young men, women and children presenting with gout, hyperuricaemia, or renal disease. The hallmark, and only diagnostic marker at present, is hyperuricaemia associated with a grossly reduced fractional uric acid clearance (FEur). We have identified this hallmark by biochemical screening in 50% of seemingly healthy family members of all ages and sexes in FJHN kindreds (2,7-9). This in turn has enabled us to confirm the dominant nature of FJHN and establish that the equally striking finding in affected children is the magnitude of the reduced FEur: 5.0 ± 0.5% compared with healthy UK children (FEur 18.4 ± 5.1%). Even more unusual is the fact that this FEur is virtually identical with that which we reported previously for young adult UK women and men with FJHN (mean 5.1 ± 1.5% and 5.1 ± 1.6% respectively) (9). FJHN thus lacks the normal child/adult difference in FEur and the normal post-pubertal sex difference as well.

Treatment of asymptomatic hyperuricaemia is important in FJHN

Another important observation in FJHN is our finding of this extremely reduced FEur in a greater proportion of children with as-yet normal renal function at diagnosis (42%), than adults (only 16%) (Figure 4). This, coupled with the fact that we have found hyperuricaemia without renal dysfunction but never renal dysfunction without hyperuricaemia (7-9), supports the contention that the hyperuricaemia due to the reduced FEur is the primary event and precedes a reduction in GFR. Such enhanced reabsorption of uric acid could produce unacceptably high urate concentrations in the tubular cell or renal vasculature, eventually leading to permanent renal damage. This hypothesis is supported by the fact that in patients who are good compliers and diagnosed prior to the onset of severe renal disease (creatinine clearance> 50ml/min) allopurinol has ameliorated the hitherto rapid progression to dialysis and transplantation seen in earlier generations (2,7-9). These studies underline the importance of a) biochemical screening of all family members in FJHN kindreds; b) early presymptomatic detection followed by sustained treatment with allopurinol of all asymptomatic as well as symptomatic subjects with FJHN.

Other clinical presentations of genetic metabolic purine and pyrimidine disorders

Two other disorders, adenine phosphoribosyltransferase (APRT) and XDH deficiency (1,2) can present in acute renal failure in the first year of life due to the over-excretion of insoluble purine bases other than urate. Uraemic coma has been reported, with renal ultrasound giving the first hint of the underlying cause (1,2). These conditions must thus be distinguished from the uric acid overproduction disorders by biochemical screening. Symptoms and signs, such as colic or haematuria occurring from birth, have often been overlooked and renal damage has developed, leading to costly dialysis and transplantation, which can be avoided in APRT deficiency. The stones of 2,8-hydroxyadenine formed by XDH from adenine accumulating in APRT deficiency may be misdiagnosed as uric acid stones unless more sophisticated screening methods are used (1,2). The same applies to xanthine stones in XDH deficiency.

In addition to HPRT and superactive PRPS, seven other disorders are associated with neurological deficits of varying severity: they include purine nucleoside phosphorylase (PNP) deficiency, combined xanthine dehydrogenase/ sulphite oxidase (XDH/SO) deficiency, myoadenylate deaminase (AMPDA) deficiency or adenylosuccinase (ASA) deficiency. Symptoms range from exercise intolerance to psychomotor delay, autistic behaviour, epilepsy, neonatal fitting, sensorineural deafness, self-mutilation, choreoathetosis and the inability to walk or talk (1-4,10). Hyper (or hypo) tonicity, microcephaly, mental retardation, dysmorphic features and early death are frequent.

Genetic purine and pyrimidine disorders presenting as t-cell immunodeficiency or autoimmune disease

Failure to thrive and recurrent life-threatening infections, especially if there is a family history of consanguinity, should evoke suspicion of adenosine deaminase (ADA) and PNP deficiency, both recessive genetic immunodeficiency disorders (1,2,11-13). ADA deficiency generally presents soon after birth, associated with projectile vomiting, diarrhoea, generalised candidiasis, severe combined immunodeficiency (SCID), and early death if untreated. The most successful treatment is bone marrow transplantation. The defect was established recently in two sisters in their thirties presenting with deteriorating respiratory function, CD4+ T-lymphocytopenia, generalised warts and oral and vaginal candidiasis, but negative HIV-1 status (11). Both had a history of recurrent chest infection dating from their teens. This demonstrates that, as in the other disorders, ADA and PNP deficiency should be suspected at any age. Disparate phenotypes in siblings also occur in ADA deficiency and have been related to variations in splicing efficiency (12) and further emphasise the importance of screening all family members in kindreds with these disorders.

In the companion disorder, PNP deficiency, the immunological problems are milder initially and PNP generally presents first with neurological abnormalities, commonly spastic diplegia, similar to those in HPRT deficiency. Although HPRT is not defective, the enzyme cannot function without the substrates normally provided by PNP (1,2,13). Consequently, although there is a history of recurrent infections, otitis media, thrombocytopenia, autoimmune disease and T-cell immunodeficiency - ultimately the cause of death at around five years - the underlying defect may not be suspected originally. Again, less severe cases present later, with mild to absent neurological abnormalities. Low to undetectable plasma uric acid is an early clue, but here too tissue variation in enzyme expression may mask the diagnosis, as in a child presenting first with developmental delay and spastic diplegia at 13 months. The child, admitted subsequently into intensive care aged 2 with severe chicken pox, was initially not considered to have PNP deficiency because of a normal plasma uric acid (2). Consequently, clinical as well as biochemical parameters must be taken into consideration in the diagnosis of these disorders.

Immunodeficiency has also been identified in some children presenting with the rare disorder hereditary oroticaciduria (UMP synthase deficiency), a defect of de novo pyrimidine synthesis, although refractory megaloblastic anaemia is the more common form of presentation (1,2).

Purine and pyrimidine disorders and analogue therapy: role in the development of improved antimetabolites and treatment of rheumatoid arthritis (RA)

The application of knowledge derived from these rare disorders to treating the more common killers of mankind is a rapidly developing field, particularly in immunosuppression, organ transplantation, malignancies and viral infections. The association between the inherited disorders ADA, PNP and UMPS deficiency and immunodeficiency syndromes highlighted the importance of purine and pyrimidine metabolism to the immune response. This in turn stimulated the development of novel analogues, such as the antimetabolite 2-chloro-deoxyadenosine, to simulate ADA deficiency. New purine and pyrimidine analogues, have been developed and introduced into clinical trial either in RA, or to prevent rejection after organ transplantation (14-16). They include leflunomide (LFM), brequinar (BQR) and mycophenolic acid-mofetil (MPA). In addition, methotrexate (MTX) is effective at low doses in RA and is considered (based on studies of its known effect in malignancy) to inhibit folate-dependent reactions in purine synthesis, as well as synthesis of the pyrimidine thymidylic acid (Figure 4). Whether it is immunosuppressive or anti-inflammatory in RA is controversial (17). Both BQR, a known inhibitor of de novo pyrimidine synthesis at the level of the enzyme dihydroorotate dehydrogenase (DHODH), and MPA, an inhibitor of the first committed step of guanine ribonucleotide synthesis catalysed by IMP dehydrogenase (Figure 4), have been in clinical trial as immunosuppressants to overcome rejection after organ transplantation (14). Use of MPA in RA has also been proposed (16) and LFM is in phase III clinical trial in RA (14). Structural analogues of LFM developed for use in organ transplantation are effective in prolonging graft survival in different animal model systems (15). However, the mechanism of action of LFM is controversial. It is considered to inhibit de novo pyrimidine synthesis in a manner similar to BQR (Figure 4). Alternative, or additional hypotheses involve cytokine networks.

Figure 4. Diagram illustrating the importance of de novo purine and pyrimidine pathways for the synthesis of DNA, RNA, the purine and pyrimidine sugars and lipids involved in membrane synthesis, glycosylation of proteins and the nucleotides involved in signal transduction and translation. The enzymes targeted by the different immunosuppressive analogues currently used (or in trial) in RA and organ transplantation are indicated. BQR, brequinar; LFM, leflunomide; MPA, mycophenolic acid-mofetil; MTX, methotrexate.

Purine and pyrimidine disorders and adverse reactions to analogue therapy

It is equally important for clinicians to be aware that these genetic disorders can also be the basis of adverse reactions during treatment with purine and pyrimidine analogues. The use of therapy including azathioprine (AZA) for transplant immunosuppression in patients with thiopurine methyltransferase (TPMT) deficiency, and in an undiagnosed case of XDH deficiency, has had potentially catastrophic consequences (1,2,18). Both enzymes normally metabolise AZA by alternative routes, thereby reducing the effective therapeutic dose. The same applies to the pyrimidine analogue 5-fluorouracil used principally in colon and breast cancer. Severe neurotoxicity has developed in patients with the pyrimidine disorder dihydropyrimidine dehydrogenase (DHPD) deficiency, a disorder now being recognised with increasing frequency in the general population (2).

How can we treat these disorders?

Apart from ‘primary’ gout and APRT deficiency, both of which generally respond to allopurinol, the majority of these disorders are as yet untreatable. Allopurinol has been used to treat ‘primary’ gout for the last 3 decades. It has been particularly useful in severe tophaceous gout, especially in the presence of renal failure, but allergic reactions do occur and may be life-threatening. Successful lowering of plasma urate and resolution of tophi has been achieved in such cases using benzbromarone, a potent uricosuric drug, even in renal failure (19). Benzbromarone is more potent than probenecid and is used widely throughout Europe, but interest in its undoubted efficacy in such problem cases is very recent in the UK. It is available for prescription on a named patient basis.

Figure 5. Renal ultrasound of the kidney of a child with long-standing Lesch-Nyhan syndrome on allopurinol therapy. Image kindly supplied by Dr I J Kenney.

Allopurinol can also reduce the risk of gout or kidney stones in LNS, as well as young adults with partial HPRT deficiency, or PRPS, but purine metabolism in such patients is exquisitely sensitive to the drug. The dose must be carefully controlled (never more than 300 mg/day) to prevent replacing uric acid (the solubility of which can be improved ten-fold by urine alkalinisation) with the extremely insoluble xanthine (insoluble at any pH), leading to xanthine nephropathy, xanthine stones and chronic renal failure (7). The development of progressive nephropathy (Figure 5) in LNS patients treated long-term with allopurinol is probably caused by build-up of xanthine. This scenario is exemplified by two brothers presenting with gout in their 20’s (which should have suggested a genetic disease) prescribed allopurinol up to 900 mg/day. Their partial HPRT deficiency remained undiagnosed until they were in their 50s when uric acid had been replaced in the urine almost completely by xanthine and the GFR was 30ml/min (2,7). Allopurinol will also inhibit the formation of the nephrotoxic 2,8 hydroxyadenine in APRT deficiency and thereby prevent the urolithiasis and progression from acute renal failure to dialysis and transplantation, as has happened in the past (1,2). As mentioned earlier, allopurinol also ameliorates the progression of the renal lesion in FJHN and early diagnosis and treatment of all these disorders is thus vital (1,2,7).

Diagnosis and associated problems

The clinical diagnosis of gout is well defined. However, for the X-linked disorders associated with paediatric gout, crystals on the nappy, or the tip of the penis and in the urine may be an early warning sign noted by the parent or local doctor. This also applies to the other disorders manifest by the overexcretion of insoluble purines and pyrimidines. Renal ultrasonography has sometimes provided the first important clue to the underlying disorder in children presenting in acute renal failure (2).

The ability to confirm the defective gene product(s) involved in purine or pyrimidine disorders is based on demonstrating the specific metabolite(s) which accumulate in consequence of the defect (Figure 1) and/or direct assay of the defective enzyme in erythrocytes or leucocytes, all by HPLC (20). However, specific metabolite identification may be hampered by interference from dietary components, drugs, or sometimes blood transfusion (2,20).

In rare instances - for example where the enzyme is expressed only in the liver, or where no abnormal metabolites accumulate - altered renal urate clearance or abnormal erythrocyte nucleotides may provide the only guide (2,21). Accurate measurement of uric acid requires the use of specific enzymic methods (9). In patients with FJHN first presenting when renal dysfunction is severe, diagnosis from the grossly reduced fractional clearance will be difficult because of the well known increment in renal uric acid clearance occurring with progressive renal dysfunction (7). However, the increment is usually less than in renal disease of comparable degree from other causes. Another problem is how to distinguish early onset primary gout due to life style in young males, from FJHN. Both will have a reduced FEur. An important clue, until we have a genetic marker, is diet, weight and family history. FJHN patients are generally slender, normotensive and rarely given to overindulgence in purines. Moreover, there is generally a strong family history of gout and/or renal disease over several generations. The young male with ‘primary’ gout is frequently overweight, a heavy beer drinker, with no family history of renal disease. A combination of tests and careful clinical evaluation is thus invaluable to identify all these disorders.

Key words: genetic purine and pyrimidine disorders, paediatric gout, Lesch Nyhan syndrome, FJHN, PRPS superactivity, renal failure, neurological deficits, immunodeficiency, AIDS, rheumatoid arthritis.

References

1. Various authors. Purines and Pyrimidines. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. McGraw Hill, New York, 1995; 7th edition, Vol II, 1655-1799.
2. Simmonds HA. Purine and Pyrimidine Disorders. In Holton JB, ed. The Inherited Metabolic Diseases. Churchill Livingstone, Edinburgh,1994; 297-351.
3. Watts RWE. Defects of tetrahydrobiopterin synthesis and their possible relationship to a disorder of purine metabolism (the Lesch-Nyhan syndrome). Adv Enz Reg 1985; 23: 25-58.
4. Becker MA, Puig JG, Mateos FA, Jiminez ML, Kim M, Simmonds HA. Inherited superactivity of phosphoribosylpyrophosphate synthetase: association of uric acid overproduction and sensorineural deafness. Am J Med 1988; 85: 383-390.
5. Roch-Ramel F, Werner D, Guisan B. Urate transport in the brush-border membrane of the human kidney. Am J Physiol 1994; 266: 767-805.
6. Platt PN, Dick WC. Diuretic-induced gout: the beginnings of an epidemic. The Practitioner 1985; 229: 281-4.
7. Cameron JS, Moro F, Simmonds HA. Gout, uric acid and purine metabolism in paediatric nephrology: Nephrology Review. Pediatr Nephrol 1993; 7: 105-118.
8. Moro F, Simmonds HA, Cameron JS, Ogg CS, Williams GD, McBride MB, Davies PM. Does allopurinol affect the progression in familial juvenile gouty nephropathy? Adv Exp Med Biol 1991b; 309A: 199-202.
9. Simmonds HA, McBride MB, Hatfield PJ, Graham R, McCaskey J, Jackson M. Polynesian women are also at risk for hyperuricaemia and gout because of a defect in renal urate handling. Br J Rheumatol 1994; 733: 932-937
10. Stone RL, Aimi J, Barshop BA, Jaeken J, Van den Berghe G, Zalkin H, Dixon JE. A mutation in adenylosuccinate lyase associated with mental retardation and autistic features. Genetics 1992; 1: 59-63.
11. Shovlin CL, Simmonds HA, Fairbanks LD, Deacock S, Hughes JMB, Lechler RI, Webster ADB, Sun Xio-Ming, Webb JC, Soutar AK. Adult onset immunodeficiency due to adenosine deaminase deficiency. J Immunol 1994; 153: 2331-2339.
12. Arrendondo-Viga FX, Santisteban I, Kelley S, Schlossman CM, Umetsu DT, Hershfield MS. Correct splicing despite mutation at the invariant first nucleotide of a 5’ splice site: a possible basis for disparate clinical phenotypes in siblings with adenosine deaminase deficiency. Am J Hum Genet 1994; 54: 820-830.
13. Markert ML. Purine nucleoside phosphorylase deficiency. Immunodeficiency Rev 1991; 3: 45-81.
14. Davis JP, Cain GA, Pitts WJ, Magolda RL, Copeland RA. The immunosuppressive metabolite of leflunomide is a potent inhibitor of human dihydroorotate, dehydrogenase. Biochemistry 1996; 35: 170-173.
15. Williamson RA et al (18 co-authors) Dihydroorotate dehydrogenase is a high affinity binding protein for A77 1726 and mediator of a range of biological effects of the immunomodulatory compound. J Biol Chem 1995; 270: 22467-72
16. Allison AC, Kowalski WJ, Muller CJ, Waters RV, Eugui EM. Mycophenolic acid and brequinar, inhibitors of purine and pyrimidine synthesis block the glycosylation of adhesion molecules. Transplant Proc 1993; 25: 67-70.
17. Kremer J. Possible mechanisms of action of methotrexate in patients with rheumatoid arthritis. Br J Rheumatol 1995; 34: 26-29.
18. Chocair P, Duley JA, Simmonds HA, Cameron JS. Is thiopurine methyltransferase (TPMT) activity important for the use of azathioprine in transplant recipients? Transplantation 1992; 53: 1051-1056.
19. Grahame R, Simmonds HA, McBride MB, Marsh FP. How should we treat tophaceous gout in patients with allopurinol hypersensitivity? Clin Biochem 1997 (in press).
20. Simmonds HA, Duley JA, Davies PM. Analysis of purines and pyrimidines in blood, urine and other physiological fluids. In Hommes F, ed. Techniques in diagnostic human biochemical genetics: A Laboratory manual. Wiley-Liss, New York, 1991; 397-425.
21. Simmonds HA, Fairbanks LD, Morris GS, Webster DR, Harley EH. Altered erythrocyte nucleotide patterns are characteristic of inherited disorders of purine and pyrimidine metabolism. Clin Chim Acta 1988; 171: 197-210.
22. Kenny U. Renal sonography in long-standing Lesch-Nyhan syndrome. Clin Radiol 1991; 43: 39-41.

[Back to top]

SEPTIC ARTHRITIS

RA Hughes MRCP MD Rheumatology Department St Peter’s Hospital Guildford Road Chertsey Surrey KT16 0PZ

Definition

Septic arthritis is defined as joint inflammation caused by the presence of live intra-articular micro-organisms. Septic arthritis may be associated with osteomyelitis, infection of an adjacent bone, especially in childhood. Septic arthritis arises as a result of infection with bacteria, viruses, fungi and, more rarely, other esoteric micro-organisms such as protozoa. Septic arthritis is uncommon but diagnosis must be made early and effective anti-microbial treatment started without delay. Joint sepsis must be distinguished from reactive arthritis in which synovitis is triggered by a primary infection at a site distant from the joint. Although bacterial antigens may be found in the joint in reactive arthritis, joint material is sterile on microbial culture and antibiotic treatment is of unproven efficacy. In this report the main focus will be on the investigation and treatment of bacterial septic arthritis.

Clinical presentation

Septic arthritis most commonly presents as an acute hot joint or joints together with the cardinal signs of acute inflammation: swelling and joint effusion, redness, pain and loss of function. The presence of an acutely painful or swollen joint, especially in association with a joint prosthesis, underlying inflammatory joint disease or in a patient with diabetes mellitus, should prompt the suspicion of joint infection. However the presentation of septic arthritis may be atypical in infants, when commonly the patient presents with systemic malaise and fever, with few specific signs or symptoms of joint disease. In the young child, sepsis may occur in the hip which is held immobile in flexion and abduction. In the elderly, and in individuals with coexistent joint disease, immunosuppression and in those on steroid treatment, the clinical signs of an infected joint may be atypical with a minimum of evidence of inflammation. The appearance of sinuses or fistula and tissue necrosis surrounding prosthetic joints should suggest joint infection.

Extra-articular manifestations of infection may be more prominent than joint signs in some cases of septic arthritis. Gonococcal septic arthritis may be associated with urethritis, erythematous skin lesions with subsequent necrosis and sometimes with prominent tenosynovitis. Joint infection can occur by haematogenous spread in patients with bacterial endocarditis in whom heart murmurs and splinter haemorrhages may be more prominent. Infection with Staphylococcus aureus - a common joint pathogen - may present in children with Scalded Skin syndrome and in adults with the clinical features of Toxic Shock syndrome or erythema multiforme.

Differential diagnosis

The differential diagnosis of septic arthritis is that of the acute hot joint. Damaged, inflamed and prosthetic joints are more susceptible to infection; this must not be forgotten as infection occurs more frequently in abnormal joints and the two pathologies can co-exist. Crystal arthritis, acute inflammatory arthritis, post-traumatic synovitis, extra-articular inflammation (for example olecranon bursitis), and haemarthrosis can all mimic septic arthritis. Septic arthritis is the most important of the differential diagnoses of the acute hot joint since a misdiagnosis and failure to treat with antibiotics can have devastating consequences for the joint and for the patient.

The route of infection

Infection of the joint in septic arthritis can occur by one of five routes (Figure 1). Most commonly, spread of the micro-organism is haematogenous, such as seen following septicaemia; frequently septic arthritis arises from infective foci in the skin such as wound infections and abscesses, sepsis in the mouth and teeth or after dental procedures or in association with infection of the respiratory or urogenital tract. Direct penetrating trauma to the joint with sharp objects such as thorns or needles or during major traumatic injury can lead to joint infection. Diagnostic and therapeutic techniques of joint aspiration and injection and surgical procedures such as joint replacement can result in joint infection. Osteomyelitis can spread to involve the joint especially in young children. Finally, infection of the soft tissues adjacent to the joint, such as inflamed bursae or tendon sheaths, can spread to involve the joint space.

Spread of infection by the haematogenous route is still the most frequent cause of joint sepsis. The incidence of septic complications following joint replacement surgery has fallen as more elaborate peri-operative precautions are taken to avoid infection, together with the widespread use of prophylactic antibiotics and advances in surgical technique.

Figure 1. Routes of access to the joint in bacterial septic arthritis

The pathogenesis of joint destruction in septic arthritis

The presence of live micro-organisms in the joint stimulates the activation of immunological defence mechanisms. Neutrophils aggregate in the joint space and synovium in order to phagocytose invading micro-organisms whilst recruitment of lymphocytes and activation of the complement cascade result in accentuation of the acute inflammatory response. Increase in vascular permeability causes joint effusion and intra-articular pressure increases, which, in itself, can compromise the integrity of cartilage. The release of cytokines and degradative enzymes such as elastase and collagenase, and the generation of free radicals in the joint result in rapid and irreversible destruction of joint tissue and adjacent bone and atrophy of surrounding muscles. Sepsis can also stimulate pannus formation which persists long after micro-organisms have been eliminated. Rapid treatment of joint sepsis is paramount if joint damage is to be limited.

Micro-organisms in septic arthritis

The likely microbial aetiology of joint sepsis will vary with the age of the patient, the coexistence of underlying disease, the route of spread of infection and the distribution of affected joints (Table 1). It is important to know the likely pathogens in particular clinical situations as antibiotic treatment is usually started before results of culture and sensitivity are fully available. In all ages and clinical situations the most common micro-organism to cause joint sepsis is Staphylococcus aureus. In neonatal septic arthritis, Escherichia coli and Haemophilus influenzae must be considered as candidate pathogens, whilst in children from age 1 month to 5 years Haemophilus influenzae is the most common cause of haematogenous joint sepsis. Gram-negative intestinal bacterial are more common in patients who are elderly and incontinent and in patients with diabetes mellitus or prosthetic joints. In cases of penetrating injury, and in intravenous drug abusers, infection with Pseudomonas aeroginosum or Staphylococcus epidermidis may be found. Septic arthritis is uncommon in fit young adults and should prompt a search for Neisseria gonorrhoeae or meningococcal infection. Chronic low grade septic arthritis, especially in the spine, can be the result of infection with micro-organisms such as Mycobacteria or Brucella abortus and biopsy may be necessary to establish diagnosis. As HIV infection becomes more widespread so the range of joint pathogens becomes more diverse.

Immunosuppression and low CD4 T lymphocyte counts have been associated with an increase in cases of atypical mycobacterial and fungal joint infections.

Table 1. Micro-organisms causing bacterial septic arthritis

Investigation and diagnosis

A history should be obtained to include evidence of prior or current infection elsewhere in the body, especially the genito-urinary tract in the adult. Any risk of HIV infection should be identified and counselling for HIV testing considered. A history of prior exposure to antibiotics should be obtained as recent oral treatment can mask ongoing joint infection and the joint may appear to be sterile.

Where joint sepsis is suspected joint aspiration must be performed and referral to hospital for in-patient investigation considered. Joint aspiration should be carried out under aseptic conditions and any joint aspirate despatched in a sterile container for immediate Gram stain, special stain for acid-fast bacilli and subsequent culture. Where immunosuppression is suspected, the microbiologist should be alerted to culture for unusual micro-organisms. Synovial fluid should be sent for a cell count as high neutrophil and total white cell counts raise the probability of infection. Synovial fluid should also be examined under polarised light to search for crystals. Blood should be taken into blood culture bottles and the ears and throat should be examined and swabbed especially in children. If gonococcal infection is suspected in the adult, genital, throat and anal swabs should be taken into appropriate media and sent immediately to the laboratory. In suspected cases where no microbiological proof of infection can be established synovial biopsy may be necessary.

Treatment of joint sepsis

All patients with proven septic arthritis should be admitted to hospital for treatment with intravenous antibiotics and rest of the affected joint. The duration of treatment will depend on severity and clinical response.

Treatment should only be started after samples have been taken for microbiological identification of the relevant micro-organism. An initial regimen of antibiotic treatment is suggested for use prior to the availability of culture and sensitivity results (Table 2). The prevalence of antibiotic resistance is changing. An increasing number of Haemophilus strains are resistant to penicillin (approximately 15% of isolates at the present time) prompting the choice of a cephalosporin in the young. As more strains of multiple-resistant Staphylococcus aureus (MRSA) emerge so the choice of antibiotics may become more difficult. Treatment should always include adequate analgesia by an appropriate route as this condition can be very painful. Anti-pyretics and anti-inflammatories can be useful.

Table 2. Recommended initial regimen for antibiotic therapy in septic arthritis

Following the availability of results of culture and sensitivity, and dependent on clinical response, the regimen can be altered subsequently. Intravenous antibiotics are normally continued for up to 14 days followed by oral antibiotics for up to four weeks depending on clinical response. In cases of prosthetic joint infection antibiotics may need to be continued long-term. There is no need for intra-articular injections of antibiotics in the treatment of septic arthritis.

The role of repeated aspiration and of surgery in the treatment of septic arthritis is still uncertain. No well-designed randomised clinical trials have been carried out that address this issue. In joints that are difficult to aspirate due to loculation of fluid and presence of fibrin, an initial joint arthroscopy should be considered. Repeated needle aspiration of infected joints is sometimes performed but there is little evidence to support this practice. Orthopaedic treatment of joint sepsis will usually involve invasive joint washout and subsequent irrigation of the joint although, in uncomplicated cases, there is no clinical evidence that this is better than intravenous antibiotic treatment alone. In a young child with an infected hip, joint washout is often performed and arthroscopy appears to be as beneficial as open arthrotomy.

Caution should be exercised when any invasive joint manoeuvre is considered as new pathogens may be introduced into the joint. Invasive procedures should be avoided where possible. Immobilisation of the infected joint in the early days of treatment can be aided by the use of a plaster back slab or splint. Surgery, with removal of the joint prosthesis and all associated foreign material, is often necessary when joint replacements become infected. Enthusiastic rehabilitation must follow antibiotic treatment with involvement of physiotherapy for muscle building and mobilisation.

Guidelines from the RCP and BSR working parties

In an attempt to ensure the early identification and correct treatment of septic arthritis a working party from the British Society for Rheumatology and the Royal College of Physicians produced guidelines in 1992 for the management of the acute hot joint (Table 3).

Table 3. BSR and RCP guidelines for the management of the acute hot joint (see references)

An audit of the adherence to these guidelines has shown them to be practical in both hospital and general practice. The audit revealed that casualty department staff relied too heavily on x-rays in the assessment of the acute hot joint and that over-confidence in clinical diagnosis of septic arthritis in hospital meant that aspiration was not always performed before treatment was started. The suggested outcome measures were unhelpful and require further modification. The greatest delay before diagnosis of septic arthritis was made and treatment started was the result of initial hesitation by patients themselves before seeking advice.

References for further reading

1. Hedstrom SA, Lidgren L. Practical Rheumatology Eds Klippel JH, Dieppe PA. 1995 Mosby pps 277-288.
2. Currey HLF. Acute Monarthritis - Differential Diagnosis, and Management. Practical Problems, ARC Reports on Rheumatic Diseases (Series 2) September 1988 No 10.
3. Frederiksen B, Christiansen P, Knudsen FU. Acute osteomyelitis and septic arthritis in the neonate, risk factors and outcome. European J Pediatrics 1993; 152: 577-580.
4. Lambert HP, O’Grady FW. Antibiotic and Chemotherapy 1992, Churchill Livingstone, London.
5. Report of a joint working group of the Brititsh Society for Rheumatology and the Research Unit of the Royal College of Physicians. Guidelines and a proposed audit protocol for the initial management of an acute hot joint. J R Coll Physicians Lond 1992; 26: 83-85.
6. Walker DJ, Young I, Hassey GA, Smith AMM. The acute hot joint in medical  practice. J R Coll Physicians Lond 1995; 29: 101-104.

[Back to top]

SYSTEMIC SCLEROSIS

Professor Carol Black Royal Free Academic Rheumatology and Connective Tissue Diseases Unit Royal Free Hospital and School of Medicine Pond Street London NW3 2QG

Introduction

The prevalence and impact of systemic sclerosis (scleroderma, SSc) make it the most important of the ‘scleroderma-spectrum’ disorders. Understanding and managing the disease are made possible by a combination of accurate subsetting, including recognition of the population at risk (Table 1), staging the subsets and monitoring the internal organs by relevant investigations at appropriate intervals. Therapy may then be matched as closely as possible to the pathogenic process (1).

Until a better understanding of the pathogenesis is achieved, the aim of therapy must be to halt disease progression and prevent further extension. This report will briefly outline: 1) pathogenesis; 2) the clinical task in subsetting and staging, emphasising the chronic and often subclinical but persistent nature of the disease, and the need for expert management; 3) an attempt to match therapy to pathology.

Table1. Classification: systemic sclerosis (SSc) subsets

Pathology

The major pathological processes occurring in SSc are increased deposition of extra-cellular matrix (ECM) in the skin and internal organs, apparently the result of disruption of the normal steady state of connective tissue turnover and regulated repair, vascular change and immune cell activation. It is thought that these processes are closely related. Currently the most favoured explanation for them implicates widespread intimal vascular damage and/or activation resulting in increased vascular permeability and leucocyte-endothelial adhesions, with subsequent leucocyte migration into the interstitium. Either concomitantly with or following on from these inflammatory events, mediators are released and a subset of fibroblasts develops a fibrogenic phenotype (2). Whether this fibroblast abnormality is wholly acquired or whether it depends upon an inherited predisposition is unknown, but the final result is an excess deposition of normal matrix with subsequent organ dysfunction (Figure 1).

Figure 1. Interactions between endothelial cells, leucocytes and fibroblasts in scleroderma pathogenesis. Reproduced from The Oxford Textbook of Rheumatology, 2nd edition (eds. Maddison PJ, Isenberg DA, Woo P, Glass DN) (in press) by kind permission of Oxford University Press.

The fibrotic response

The first hallmark of the disease is fibrosis. Its existence was demonstrated histopathologically many years ago, then confirmed by physical and biochemical means. Subsequent techniques for culturing fibroblasts, including three dimensional collagen matrix culture, have provided valuable insight into the mechanisms involved in the synthesis of ECM components. Skin fibroblasts from scleroderma patients synthesize increased quantities of collagens types I and III and to a lesser degree IV and VI and also fibronectin and proteoglycan core proteins. Total protein synthesis and growth production are normal. The negative feedback provided by propeptides appears to be normal, and these cells are not transformed or immortalised and have similar doubling times and life-span to normal fibroblasts in conventional monolayer culture. Scleroderma fibroblasts, however, unlike those from normal skin, display persistent proliferation in serum-free medium, and are without doubt deregulated with respect to ECM (3).

Work in the 1970s by Le Roy’s group (4) demonstrated that fibroblasts grown from areas of dermal fibrosis continue to synthesize increased amounts of collagen for several passages in vitro. Later it was demonstrated that the amounts of mRNA for these matrix proteins are increased and localised predominantly to areas surrounding dermal blood vessels. Nucleic acid hybridization techniques have since confirmed that not all fibroblasts are activated to produce more normal collagen but rather that a subset of high collagen producers is responsible. The increased collagen RNA levels could arise through increased transcription rate or by a reduction in mRNA breakdown, and there is evidence that both mechanisms may operate (5).

The transcriptional rate of genes encoding pro-alpha(I) collagen (Col 1A2) is increased in SSc fibroblasts suggesting a change in regulatory transcription factors in these cells.

The mechanism for transcription regulation of collagen synthesis is still unproved but De Crombrugghe et al (6) have demonstrated a pathway which could have a direct bearing on the transcription defect in SSc. Following stimulation by transforming growth factor beta (TGFbeta), fibroblasts containing the promoter enhancer regions of the alpha2(I) collagen gene linked to a chloramphenicol acetyl transferase (CAT) reporter show a marked enhancement of collagen gene expression. This appears to be through the interaction of the specific promoter of the collagen gene and a transacting DNA binding protein, nuclear factor 1. Such elegant experiments provide a prototype for the study of the activation of a pleomorphic genetic response such as fibrosis which is not, of course, restricted to SSc. The cytokine stimulus is also likely to be more complex than TGFbeta alone. The activation and perpetuation of SSc is a complex affair and other mediators such as platelet-derived growth factor (PDGF), interleukins IL1, 4, 6, 8, insulin-like growth factors (IGFs), fibroblast growth factors (FGFs) and interferons alpha and gamma may be involved. The cytokines possibly act via interactive loops or as a cascade.

Immune events

That the immune system and its cellular mediators are important stimulants to pathological change in SSc is without doubt. There is a growing body of evidence in SSc to show that lymphocytes and their mediator products interact with endothelial cells, fibroblasts and the ECM. Lymphocytes normally reside within blood vessels and their permanent exit from the circulation into the extracellular space requires the appropriate expression and engagement of particular receptors and counter-receptors on lymphocytes, endothelial cells and fibroblasts, probably facilitated by locally acting chemotactic factors (7).

Following injury to and activations of the vascular endothelium (see below) adhesion molecules such as E-selectin, VCAM-1 and ICAM-1 are upregulated in response to cytokines and other factors. These endothelial adhesion molecules bind to specific ligands on T and B lymphocytes, platelets, neutrophils, monocytes and natural killer cells, facilitating their adhesion to vascular endothelium and subsequent migration through what have now become in SSc ‘leaky vessels’ into the ECM. Many of the migrating lymphocytes are IL-2 producing cells of the alpha-beta subset expressing typical surface antigens including CD3, CD4, CD45, HLADR and LSA-1. Once in the tissues, lymphocyte localisation and attachment to fibroblasts is dependent upon activation of fibroblasts and expression of functional ICAM-1 to mediate lymphocyte attachment (8). Following cellular attachment, there may be reciprocal activation of both immune cells and fibroblasts by cell contact or by mediators released from these cells, thus providing the potential for autocrine loops. The particular importance of the cellular infiltrate and any mediator may be dependent upon the state of the disease, the disease subset or particular organ involved. These key cells, cytokines and growth factors may provide early targets for therapeutic attack.

The vascular lesions

Damage to the small blood vessel is critical to the pathogenesis of SSc and may be the primary event. The damage in SSc is widespread and can be recognised as:

1.      Vasomotor instability or Raynaud’s phenomenon with repeated transient interruption of tissue perfusion in the digits and internal organs (systemic Raynaud’s) which is often an early event in disease development;

2.      Microvascular abnormalities with structural changes characterised by proliferative intimal arterial lesions and obliteration of the vessels leading to chronic ischaemia. The damage is present in the small blood vessels of virtually all the viscera, muscle, subcutaneous tissue and skin;

3.      Decreased red cell deformability, increased platelet activity and enhanced thrombus formation.

Many factors may be important in the vascular damage but it is the endothelial cell that is thought to have a pivotal role. The endothelium is now known to produce numerous molecules and to regulate many aspects of vascular stability including control of vascular tone, permeability, thrombotic potential and leucocyte trafficking (9).

The growing evidence for endothelial cell (EC) dysfunction in scleroderma is summarised below:

1.      Direct observation of abnormal nailfold capillaries (dilatation and/or drop out).

2.      Increased capillary permeability to tracer molecules with a slowing of flow and increased period of stasis.

3.      Presence of a circulating cytotoxic factor identified as granzyme-1, a serine protease present in granules of activated T-cells.

4.      Changes in circulating levels of endothelial cell products such as von Willebrand factor, endothelin-1, plasminogen activator, angiotensin-converting enzyme, and prostacylin/thromboxane metabolites.

5.      Increased endothelial cell surface expression in vivo and elevated circulating levels of the adhesion molecules, intracellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) and E-selectin.

6.      The presence of autoantibodies that bind to endothelial cells.

The mechanism for vasospasm in SSc is likely to be complex. Interactions between EC products such as endothelin-1 and prostacyclins, nitric oxide, platelet-released products (serotonin and B-thromboglobulin) and neuropeptides (calcitonin gene-related peptide and vasoactive intestinal polypeptide) may all contribute to the abnormal vascular tone (10).

The mechanisms for the development of EC injury are unknown but in addition to the immune (cell-mediated and humoral) cytotoxicity mentioned above non-immune cytoxicity has been implicated and recently several reports have shown that EC damage can occur prior to an inflammatory infiltrate. Prescott et al (11) studied, in a sequential fashion, the pathologic changes in the perivascular spaces in the skin of scleroderma patients and normal controls. Subendothelial oedema was recognised as the first defect, followed by platelet aggregation and lymphocyte migration of both CD4+ and CD8+ T-cells. Tissue fibrosis occurred only after the inflammation had subsided. Supporting evidence for this finding comes from the work of Harrison et al (12), in which damage to endothelial/epithelial surface is shown to be the first ultrastructural change to occur in lung biopsies from SSc patients. Lung sections appearing as normal under light microscopy with no evidence of infiltrating inflammatory cells were, on electron microscopy, shown to contain widespread endothelial/epithelial damage. The nature of the primary vascular trigger to immune stimulation is critical to our understanding and treatment of the disease. Intense vasospasm, which occurs in SSc patients in both the extremities and the internal organs, could lead by reperfusion injury and free radical damage to structural and functional changes in the endothelium and subsequent immune activation (13). An endothelial-seeking virus or environmental agents are also possible candidates.

Early detection, and stabilisation of endothelial changes in SSc is an important goal in management.

The clinical task

The recognition of a pre-scleroderma state is critical and can be done with some confidence. The best predictive markers of the subsequent development of scleroderma are abnormal nailfold capilliaries and disease-specific antinuclear antibodies, and it has been suggested that the presence of both will detect over 90% of those destined to develop SSc (14,15). Other circulating markers now being investigated, and which may aid diagnostic accuracy when trying to distinguish between secondary Raynaud’s and the primary variety, are circulating levels of endothelin-1 and soluble adhesion molecules, E-selectin, VCAM-1 and ICAM-1, although increased levels of these molecules may eventually prove to be most useful as markers of vascular and fibrotic changes in established systemic sclerosis and in identifying evolving organ damage.

The classification of SSc into diffuse or limited cutaneous systemic sclerosis (dcSSc, lcSSc) is based on the extent of skin involvement and is a useful diagnostic tool (16). It is a simple classification and may be changed as knowledge expands. For example other antibody systems and RNA polymerases I, II and III are now being firmly associated with specific organ involvement and subtype and thus may increase the need for a more extensive classification.

Disease staging within a subset is equally important as it helps us approach the patient’s course, prognosis and treatment requirements in a logical manner. This can be achieved by a knowledge of the natural history of the subsets (Figure 2) and by using an investigational plan adapted to the subset and the individual patient’s needs. This is outlined below.

Figure 2. Typical natural history for development of visceral complications in dcSSc and lcSSc. Modified (with permission) from T Medsger.

Investigation plan

Pre-scleroderma: investigation of primary Raynaud’s phenomenon

·         Detailed history and examination to look for evidence of asymptomatic autoimmune rheumatic disease.

·         Plain chest and thoracic inlet x-ray can exclude simple structural lesions (eg cervical rib).

·         Baseline nailfold capillaroscopy and auto-antibody profile (ANA).

If capillaroscopy and autoantibodies are normal/negative and the Raynaud’s is of less than 2 years’ duration the patient is followed annually for another two years and then discharged. If the capillaroscopy or auto-antibody studies are positive then the patient is followed up at 6 months and then yearly for 5 years with repeat capillaroscopy and autoimmune profile, but discharged from follow-up at 5 years if no other disease features have developed.

DcSSc

At presentation

History and general examination generally establish the diagnosis at presentation. Assess the visceral disease by investigations listed in Table 2. Tissue typing (HLADR52a) and autoantibody profile may be useful in diffuse disease to identify those patients destined to develop interstitial lung disease (17). This subset requires the earliest possible diagnosis and aggressive therapy if the disease is to be halted.

Follow-up

This is especially important during the first 5 years from disease onset. Vigilant monitoring for renal involvement should include regular blood pressure checks, six-monthly urine collection for protein excretion and creatinine clearance. Six-monthly lung function tests and ECG with yearly oesophageal motility and echocardiographic studies should also be performed.

Table 2. Suggested investigations of SSc patient

LcSSc

At presentation

By the time of presentation, the history is usually of some years’ duration and the physical signs even if minimal are generally diagnostic. Diligent attention to assessment of the internal organs is necessary with lengthening of the disease duration - the major risk factors being pulmonary hypertension, which occurs in 10-15% of patients, and small and large bowel disease. The disease is assessed at presentation in the same way as the more diffuse variety but follow-up is different.

Follow-up

Annual follow-up is generally undertaken with assessment of visceral disease, including renal function, by creatinine clearance. If pulmonary hypertension is diagnosed, Echodoppler assessments will need to be more frequent. The mid- and lower bowel may also require additional investigations. A few patients with lcSSc carry an anti-topo antibody and are particularly susceptible to lung fibrosis. In these patients lung function parameters should be carefully monitored.

Symptomatic and supportive therapy

Systemic sclerosis is a chronic disabling disease for many patients and it can threaten not only their functional ability and physical appearance but also their professional, social and emotional lives. Its management requires a team approach with input and advice from physiotherapists, occupational therapists, nurse educators, social workers, counsellors and community services. Simple measures such as a warm comfortable environment with additional heating aids can make an enormous difference.

Maintaining adequate lubrication of the skin is difficult and patients should use lubricating creams. Telangiectasiae can be a cosmetic problem. These lesions can be covered with special make-up or, if particularly large, treated with laser therapy.

Early and continuous physical therapy is important in limiting contractures. Both active and passive exercises are necessary (with analgesia if required) to a daily regime. The constant encouragement of the therapist and physician is also critical if function is not to be lost. Of paramount importance to the long-term management of systemic sclerosis is family support and education which is often a much neglected area.

Treatment

Although there is currently no cure for scleroderma, there are therapies available which can offer partial relief, control end organ damage and improve quality of life for the scleroderma patient (18). The choice and evaluation of any treatment regime is not easy. This is because 1) the disorder is heterogeneous and its extent, severity and rate of progression are highly variable - therapy must therefore be closely tailored to the individual patient systems involved; 2) there is a tendency towards spontaneous stabilisation and/or regression after a few years, particularly within the more benign and numerically larger lcSSc subset; 3) the disease is complex and the relationship between immune dysfunction, vascular damage and fibrosis speculative and 4) there is a paucity of both clinical and laboratory features for ascertaining improvement (or deterioration) in the disease especially with respect to visceral change.Therapeutic trials of disease-modifying drugs are essential but the trial design is critical in this disease. There is growing acceptance that any trial of disease-modifying drugs must be controlled, (preferably placebo-controlled), that patients should have early diffuse disease (less than three years) and that studies must be of sufficient duration (one year minimum).

The ideal group to target and treat would be the ‘at risk’ patients, those in the pre-sclerotic state. Many such patients can now be identified by circulating antibodies, cytokine production and nailfold changes. Unfortunately, adequate preventative therapy is still wanting. An extension of this idea and, one which can for example be applied to the lung, is the earliest possible diagnosis of internal organ involvement so that containment therapy may be attempted.

Immune modulating therapies

As the immune system may be involved early in the disease, immuno-modulatory and immuno-suppressive agents have been employed, particularly for early dcSSc. Tables 3 and 4 summarise their use and evaluation. Both anti-metabolites and alkylating agents have been used. Both chlorambucil and 5-fluoruracil failed in placebo-controlled trials. Cyclophosphamide still has an undetermined place (19). As a single agent its efficacy is unknown. In combination with steroids or plasma exchange it may also have a role - unfortunately there are as yet no controlled data. The use of the anti-metabolites 6-thioguanine and azathioprine have been reported but again all the data are anecdotal. Methotrexate is now being approached in a more rational manner. The results of pilot studies have been encouraging and controlled trials are being undertaken. Attempts to target the immune system by lymphoplasmapheresis and total lymphoid irradiation again failed and currently under investigation are the use of rabbit antithymocyte-globulin, and photopheresis. Monoclonal antibody therapy may hold hope for the future.

Table 3. Immuno-modulatory therapies for systemic sclerosis.

Table 4. Anti-fibrotic therapies for systemic sclerosis.

Cyclosporin has been used in a few patients with beneficial effect but high doses of the drug were followed by reports of hypertension and renal failure, both of which might be attributed to either drug or SSc (20). Physicians must be mindful of the fact that trough blood levels are multiplied 2-4 fold in patients taking calcium channel blockers. Therefore a lower dose (5mg/kg/ body weight/day) has been tried with encouraging results, but awaits confirmation in a controlled study. None of these drugs should be used for late-stage diffuse or limited cutaneous disease.

Long-term high-dose steroids have no place in the management of SSc, indeed they are potentially toxic (21) and may be implicated in ‘normotensive renal crisis’. Steroid usage has been restricted to patients with myositis, symptomatic serositis, the early oedematous phase of the skin disease and refractory arthritis and tenosynovitis. The lowest possible therapeutically effective dose should be sought.

Some of the established, and newer, putative antifibrotic agents which have been used in scleroderma are summarised in Table 5. D-penicillamine and colchicine have been used in the treatment of scleroderma for many years. Neither drug stops the fibrotic process, but consensus opinion is that D-penicillamine is the more useful. To derive maximum benefit, the drug must be used correctly. The most suitable subjects for treatment are those with diffuse active skin disease. Long-term treatment in a dose between 500-750mg is needed. D-penicillamine should be used throughout the active phase into the stable period and the dose maintained until there is no further improvement in skin thickening. The drug may then be reduced, but low-dose treatment should be maintained for many years - some recommend 10 years.

The interferons alpha and gamma are currently being investigated in controlled trials after initial pilot studies. Recombinant interferon gamma is a potent inhibitor of collagen production by normal and scleroderma fibroblasts in vitro and has immunoregulatory activities. Interferon alpha, although theoretically less potent than interferon gamma as an inhibitor of collagen synthesis, does not activate the class II cell surface antigens - this may be a distinct advantage in a disease which is HLA-linked.

Such is the frustration of the condition that acupuncture and numerous vitamins, hormones, alternative medicine products, and surgical procedures have been used. Most have been heralded with great enthusiasm only to be abandoned once critically assessed.

Vascular therapy

Raynaud’s is a prominent feature of SSc and occurs in over 97% of cases. It is an aspect of the disease which can be relieved, though the response is variable. This idiosyncratic response may reflect the stage of the disease. If structural damage is present and severe, the patient will respond rather poorly to vasodilators alone. In uncomplicated cases simple measures may suffice. As the attacks become more frequent, prolonged oral drug therapy, possibly on an intermittent basis, may be needed. Intravenous therapy and limited surgery is restricted to the most severe cases (22 - Table 5).

Because of the widespread vascular damage, which not only involves many of the internal organs but also is present from the very beginning of the disease, there has been a search to find drugs to protect injured endothelial cells and to prevent platelet aggregation and subsequent release of platelet-derived mediators. This search has been disappointing to date. Ketanserin, a serotonin antagonist, although useful in Raynaud’s phenomenon, does not improve structural vascular disease. Dipyridamole and aspirin, although reducing the circulating plasma concentrations of beta thromboglobulin or circulating platelet aggregates, were not clinically effective in a randomised double blind trial. Captopril, the angiotensin converting enzyme inhibitor which has been so successful in the treatment of renal crisis, has been considered for the primary and possibly prophylactic treatment of vascular disease.

The search for newer agents continues and of current interest is L-Arginine to reduce the production of nitric oxide, antagonists to endothelin-1 receptors and free-radical scavengers: as yet none of these have been exposed to vigorous clinical trial.

Table 5. Treatment options for Raynaud’s phenomenon

Conclusion

Currently therefore, there is imperfect treatment for scleroderma. However careful consideration of the subset and stage of disease of the individual patient can maximise the use of the drugs currently available. It is hoped, more importantly, that the level and increasing degree of activity of research into the cause and pathogenesis of the condition may eventually result in early, rational, effective treatment.

References

1. Black CM, Stephens CO. Systemic sclerosis (scleroderma) and related disorders. In, Maddison PJ, Isenberg DA, Woo P, Glass DN eds. Oxford Textbook of Rheumatology 1993; 2: 771-89.
2. Black CM. Thick skin thin hypothesis. The Aetiopathogenesis of systemic sclerosis. J Royal College of Physicians of London 1995; 1: 20 No. 2.
3. Kahari VM. Activation of dermal connective tissue in scleroderma. Ann Med 1993; 25: 511-8.
4. Campbell PM and LeRoy EC. Pathogenesis of systemic sclerosis. A vascular hypothesis. Seminars in Arth & Rheum 1975; 4: 351-68.
5. Mauch C, Eckles B, Hunzleman N, Oono T, Kozlowska E, Krieg T. Control of fibrosis in systemic scleroderma. J Invest Dermatol 1993; 100: 92S-96S.
6. De Crombrugghe B, Vuorio T, and Karsenty G. Control of Type I collagen genes in scleroderma and normal fibroblasts. Rheum Dis Clinics of N America 1990; 16: 109-23.
7. Piela-Smith TH and Korn JH. Lymphocyte Modulation of Fibroblast Function in Systemic Sclerosis. Clinics in Derm 1994; 12: 369-377.
8. Abraham D, Lupoli S, McWhirter A, Plater-Zyberk C et al. Expression and function of surface antigens on scleroderma fibroblasts. Arth Rheum 1991; 34: 1134-73.
9. Pearson JB. The Endothelium: Its role in systemic sclerosis. Annals Rheum Dis 1990; 50: 866-71.
10. Kahaleh B and Matucci-Cerinic M. Raynaud’s Phenomenon and Scleroderma. Disregulation neuroendothelial control of vascular tone. Arth & Rheum 1995; 38: 1-4.
11. Prescott RJ, Freemont AJ, Jones CJ, Hoyland J and Fielding P. Sequential dermal microvascular and perivascular changes in the development of scleroderma. J Path 1992; 166: 255-63.
12. Harrison NK, Myers AR, Corrin B, Sooray G et al. Structural features of interstitial lung disease in systemic sclerosis. Am Rev Resp Dis 1991; 144: 706-13.
13. Blann AD, Illingworth K and Jayson MIV. Mechanisms of endothelial cell damage in systemic sclerosis and Raynaud’s Phenomenon. J Rheumatol
    1993; 20: 1325-30.
14. Kallenburg CGM, Wouda AA, Hoet MH, van Venrooij WJ. Development of connnective tissue disease in patients presenting with Raynaud’s Phenomenon: A six-year follow-up with emphasis on the predictive value of antinuclear antibodies as detected by immunoblotting. Ann Rheum Dis 1988; 47: 634-41.
15. Zuffery P, Depairon M, Chamot AM, Monti M. Prognostic significance of nailfold capillary microscopy in patients with Raynaud’s Phenomenon and scleroderma- pattern abnormalities: A six-year follow-up study. Clin Rhematol 1992; 11: 536-41.
16. LeRoy EC, Black CM, Fleischmajer R, Jablonska S et al. Scleroderma (systemic sclerosis); classification, subsets and pathogenesis. J Rheumatol 1988; 15: 202-05.
17. Briggs DC, Vaughan RW, Welsh KI, Myers A et al. Immuno-genetic prediction of pulmonary fibrosis in systemic sclerosis. Lancet 1991; 338: 661-62.
18. Pope J. Treatment of systemic sclerosis. Current Opinion in Rheum 1993; 5: 792-801.
19. Silver RM, Miller KS, Kinsella MB, Smith EA and Schabel SI. Evaluation and management of sclero-derma lung disease using bronchoaveolar lavage. Am J Med 1990; 88: 470-6.
20. Denton CP, Abdullah A, Sweny P and Black CM. Acute renal failure occuring in scleroderma treated with cyclosporin A - a report of three cases. B J Rheum 1994; 33: 90-92.
21. Steen VD, Conte C, Medsger TA Jr. Outcome of renal crisis in systemic sclerosis; relation to the availability of converting enzyme inhibitors (ACE). Annals Int Med 1990; 113: 352-7.
22. Black CM. Raynaud’s Phenomenon. Prescribers’ Journal 1994; 34: 125-33.

[Back to top]

WORK-RELATED UPPER LIMB DISORDERS

Malcolm I V Jayson Rheumatic Diseases Centre University of Manchester, Hope Hospital Salford M6 8HD

Introduction

Upper limb pain in association with work is common. Rheumatic problems are aggravated by forceful activities. Frequently it is difficult to determine whether the problem was caused by the work, or whether the person has difficulties in performing a job because of some rheumatic problem with symptoms which were aggravated by forceful activities. Frequently the whole issue is compounded by psycho-social factors. Compensation issues complicate many cases and play their part in aggravating the clinical problem.

Many epidemiological studies have been undertaken. Commonly they show evidence of increases in upper limb problems in relation to forceful and repetitive work. However the quality of most studies is poor with inadequate clinical descriptions and definition of the upper limb problem and frequently they lack adequate control data.

Upper limb problems can be divided into those for which a specific clinical diagnosis can be made, such as lateral epicondylitis, tenosynovitis or carpal tunnel syndrome and those in which there are diffuse non-specific symptoms and signs and which may be labelled as repetitive strain disorder (RSI) or cumulative trauma disorder (CTD). Particularly in the latter group psycho-social problems and job dissatisfaction appear to play major roles in the pathogenesis of symptoms.

Specific upper limb problems

These are best considered on a regional basis as different types of mechanical pathological entities may occur at any one site.

Neck pain

The prevalence of neck-related discomfort has been estimated as 9 per cent for adult males and 12 per cent for adult females. It may arise due to cervical spondylosis although commonly there is only a poor correlation between symptoms and radiological evidence of degenerative disease in the cervical spine.

Patients complain of aching and stiffness in the back of the neck which may radiate to the occiput and down into the shoulder and the upper limb. Commonly symptoms ease with rest and are aggravated by twisting and turning.

Prolonged static postures particularly with the neck flexed as when working at a computer with a poorly positioned VDU screen (Figure 1) can aggravate these problems. At the other extreme, repeated forceful twisting of the neck in association with work may also aggravate neck problems. There is no evidence that static postures or, at the other extreme, repeated twisting movements experienced during normal working activity cause the development of cervical spondylosis. However they can aggravate the symptoms so that the patient experiences more severe symptoms than would otherwise be the case.

Figure 1. A poorly positioned VDU screen causes prolonged periods of working with the neck flexed.

Frequently anxiety, tension and job dissatisfaction seem to play important roles. They may be associated with spasm of the paravertebral muscles. In many patients there is a vicious circle of a minor neck problem aggravated by a poor posture and job dissatisfaction leading to more pain, increased spasm of the paravertebral muscles and so on.

Careful attention should be paid to a worker’s posture. In particular, care should be taken to ensure that they do not spend long periods of time with the neck flexed. This may be by simple adjustment of the position of the computer screen or other working conditions. Likewise repeated forceful twisting of the neck should be avoided. Patients should be encouraged to get up and stretch at frequent intervals. If necessary physiotherapy treatment may be helpful.

Shoulder pain

The shoulder is a very complicated structure and pain can develop due to problems in the various joints, ligaments and tendons or alternatively be referred from the cervical spine or diaphragm. Shoulder pain is common and experienced by 7 per cent of the general population. Pain-producing pathologies include rotator cuff lesions, subacromial bursitis, bicipital tendinitis, adhesive capsulitis (frozen shoulder), acromio-clavicular joint arthritis and other problems. Careful examination will distinguish these entities but quite often patients will show features combining more than one of these disorders.

Any of these shoulder problems will be aggravated by use of the upper limb. It is commonly difficult to determine whether forceful use of the shoulder led to development of the problem or if it was there anyway in a minor form but the patient experienced difficulties as a result of doing the work.

Working in a poor ergonomic posture (Figure 2) and in particular prolonged periods of working with arms elevated above shoulder height are associated with the development of shoulder pain. Jobs should be redesigned to avoid working in this posture. Prolonged elevation of the shoulder particularly when undertaking repeated or forceful movements (Figure 3) may be associated with a wide variety of shoulder problems.

Figure 2. Working in a poor posture with repeated stresses on the shoulder.	Figure 3. Forceful movements with the arms elevated.

Elbow problems

Arthritis of the elbow is rare. The most common work-related conditions are lateral epicondylitis (tennis elbow) and medial epicondylitis (golfer’s elbow).

In lateral epicondylitis pain is experienced over the outer aspect of the elbow radiating into the dorsal surface of the forearm and sometimes into the back of the wrist. The symptoms are aggravated by forceful use of the forearm and hand. There is localised tenderness at the insertion of the extensor tendons of the forearm into the lateral epicondyle. The pain is reproduced by resisted extension of the wrist and by making a fist. An increased prevalence has been identified in relation to jobs involving repeated forceful use of the forearm such as in butchers. There is no clear evidence that the work was directly responsible for causing lateral epicondylitis but repeated forceful activities using the forearm may well aggravate the problem.

Lateral epicondylitis is easily confused with the radial tunnel syndrome which arises due to compression of the deep branch of the radial nerve as it passes through the supinator muscle. This syndrome seems to arise in relation to repeated flexion, extension, pronation and supination movements of the forearm and may be occupationally related. Patients develop pain often over the lateral aspect of the elbow spreading into the forearm and into the back of the hand. There may be sensory loss in a radial nerve distribution in the dorsal surface of the hand and in particular in the web space between the thumb and index finger. There may be some weakness of extension of the wrist and fingers. The diagnosis may be confirmed by electro-diagnostic studies. If conservative treatment fails to relieve the problem decompression may be helpful.

Medial epicondylitis is similar to lateral epicondylitis but with pain over the medial side of the joint and tenderness at insertion of the forearm flexor muscles into the medial epicondyle. The symptoms are reproduced by resisted flexion of the wrist. Although symptoms may be occupationally related there is no definite evidence that work-related factors caused the development as against the symptoms being aggravated by forceful use of the forearm.

The cubital tunnel syndrome is due to entrapment of the ulnar nerve either behind the medial epicondyle or in the forearm. It may produce pain similar to medial epicondylitis but spreading into the flexor and ulnar side of the forearm, and sometimes into the ulnar border of the hand and into the little finger. There may be sensory loss in an ulnar nerve distribution. Repeated flexion and extension to the elbow have been suggested as the cause of this disorder although no clear evidence exists. However the symptoms are aggravated by forceful use of the forearm.

The treatment of lateral and medial epicondylitis involves reducing the stresses on the forearm. Some patients benefit from local physiotherapy treatment and an epicondylitis clasp. If the problem does not clear up rapidly then a local injection of a slow-release non-fluorinated steroid (eg methylprednisolone) and anaesthetic is usually effective. Very occasionally patients require surgery.

Radial and cubital tunnel syndromes may respond to reduction of excessive stresses on the elbow. A small proportion may require surgical decompression.

Wrist and hand

Common problems in which pain may be experienced in relationship to work include tenosynovitis, carpal tunnel syndrome and osteoarthritis.

The tendons around the wrist are lined by synovial sheaths which allow the normal movements of the wrist and hand. Inflammation of the tendon sheaths gives rise to pain and tenderness along the lines of the tendons aggravated by forceful movements. On palpation there is thickening of the tendon sheath and crepitus may be palpated during tendon movements. Most commonly this condition affects the extensor tendons and in particular the extensor and abductor tendons of the thumb (de Quervain’s tenosynovitis). Finkelstein’s test which is reproduction of pain by grasping the flexed thumb with the fingers and then forcefully abducting the wrist (in an ulnar direction) is a helpful diagnostic aid. This condition is described as developing in association with repetitive forceful movements of the wrists and hands and is a prescribed industrial disease: PD A8 (Figure 4). Treatment is by avoiding excessive forceful repetitive movements: local injection of a non-fluorinated steroid and occasionally surgical decompression may be required.

Figure 4. Work placing repeated stresses on the abductor tendons at the wrist.

The carpal tunnel syndrome arises due to compression of the median nerve deep to the flexor retinaculum. It may develop as a consequence of inflammatory arthritis such as rheumatoid arthritis, in hypothyroidism, obesity or as an idiopathic disorder. It does however appear to be more common than expected in people undertaking repetitive and forceful movements of the wrist and hand and in particular following prolonged exposure to vibration (Figure 5).

Figure 5. Repeated forceful movements combined with vibration exposure may be associated with development of the carpal tunnel syndrome.

Patients with the carpal tunnel syndrome experience pain, paraesthesia and numbness in the fingers spreading up the forearm and sometimes extending to the upper arm and the side of the neck. Although most marked in the median nerve distribution, many patients have difficulties in identifying the precise site of the symptoms. It is characteristic that the problem wakes them at night with numbness and tingling and is relieved by shaking the wrist and flexing and straightening the fingers. Examination may show weakness of abduction of the thumb, sensory loss in a median nerve distribution and a positive Tinel sign (reproduction of symptoms by percussion over the flexor retinaculum) and Phalen test (reproduction of symptoms on prolonged wrist flexion). In advanced cases thenar muscle wasting may be present.

Osteoarthritis commonly affects the small joints of the hands. Most often it affects the carpometacarpal joints of the thumbs, the distal interphalangeal joints (Heberden’s nodes) or the proximal interphalangeal joints (Bouchard’s nodes). They develop bony swelling, limitation of movements and tenderness on palpation and use of the hands becomes painful. As a result patients may experience pain in association with working activities. Osteoarthritis is generally regarded to be a constitutional disorder. Primary generalised osteoarthritis commonly affects these joints and may be associated with osteoarthritis elsewhere. This condition most frequently develops in women and starts around the age of the menopause. Forceful activities aggravate the problem and can lead to employment difficulties. There is limited evidence to suggest that forceful work can cause the development and accelerate the progress of osteoarthritis.

Osteoarthritis in the hands will usually respond to reduction of stresses on the painful joint, physiotherapy, provision of a splint for the thumb base, local steroid injections and very occasionally surgery.

Non-specific upper limb problems

Some patients develop more generalised upper limb symptoms and in particular forearm and hand pain but examination does not show specific evidence of the conditions described. There is diffuse pain and tenderness combined with loss of function. Frequently they are mislabelled as having tenosynovitis despite the lack of specific clinical features of that condition.

The problem may develop in people undertaking repetitive work and has variously been called repetitive strain injury (RSI), and cumulative trauma disorder (CTD). There is no evidence that repetition plays any specific causative role, that there has been strain of the tissues nor that any injury has resulted. The symptom complex is most frequently reported in vulnerable people undertaking boring repetitive jobs particularly if working in stressful conditions.

The problem is particularly difficult as some forms have been designated as a prescribed industrial disease, PD A4, which is cramp of the hand or forearm due to repetitive movements. In particular it may occur in workers involved in prolonged periods of writing, typing or other repetitive movements of the finger or arm. It was first described by Ramazzini in 1713. He described progressive difficulties in writing in scribes and notaries so that eventually writing became impossible. It is of interest that Ramazzini identified intense and incessant application of the mind as being central to this condition.

In the middle of the 1980s there was an epidemic of compensation claims in Australia for repetitive strain injury in workers using computer keyboards. Very large numbers of subjects experienced problems. It spread like an epidemic within individual offices and factories and in certain cities whereas others were relatively unscathed. The whole problem was exacerbated by a compensation system which led to huge numbers of claims. With the requirement that compensation could only be justified when a specific pathological basis for work-related symptoms be identified, the numbers of cases dropped dramatically.

It appears that patients develop pain and disability in association with repetitive activities. In particular it arises in people undertaking prolonged repetitive work with inadequate breaks, working in a poor ergonomic posture, meeting numerous tight deadlines or performance targets, with poor job satisfaction and poor relations with supervising officers. Many will be vulnerable people who have suffered from other stress-related problems such as migraine or irritable bowel syndrome. As a result of this problem they may suffer long continued pain, distress and loss of limb function.

Management primarily lies in improving the working environment. People should enjoy their work and feel fulfilled. Ergonomic analysis helps avoid excessive stresses on the upper limbs. Workers undertaking repetitive tasks should have regular breaks. They should not remain at their work station all day but be encouraged to move about and exercise generally. Variation in work tasks and job rotation should be considered.

When a patient develops forearm or hand pain of this type early intervention is important. Full job analysis should be undertaken and the working conditions carefully monitored. A break from work may be necessary to enable the problem to resolve and then the patient gradually returns to the job but with minimisation of precipitating factors.

The cause of these persistent symptoms is a matter for much debate. Symptoms may be genuinely experienced but there is an absence of specific pathology in the upper limbs. It seems likely that there is an imprinting of pain perception within the central nervous system. Such mechanisms have been described both in the spinal cord and the brain and can lead the patient to experience symptoms in the upper limbs despite the absence of a specific problem in the peripheral areas.

Conclusion

Increasing numbers of patients complain of pain in association with their work. Proper assessment demands a careful clinical examination and occupational history. In some cases the job may be the cause of the problem but frequently it is aggravating the symptoms in a person with an underlying musculoskeletal disorder. A distinction must be drawn between those patients with specific pathologies in the upper limbs and those with diffuse non-specific symptoms often called RSI or CTD and in whom a vulnerable personality, adverse working conditions and psycho-social stress seem important.

Further reading

1. Barton NJ, Hooper G, Noble J, Steel WN. Occupational causes of disorders in the upper limb. British Medical Journal 1992; 304: 309-311.
2. Hadler MM. The roles of work and of working in disorders of the upper extremity. Bailliere’s Clinical Rheumatology 1989; 3: 121-141.
3. Silverstein BA, Fine LJ and Armstrong TA. Hand wrist cumulative trauma disorders in industry. British Journal of Industrial Medicine 1986; 43: 779-784.

[Back to top]