Aetiology

The common aetiological thread of these disorders is their striking association with HLA-B27, particularly ankylosing spondylitis (AS). HLA type B27 is present in >90% of Caucasians with AS but only 8% of controls. HLA-B27 exhibits a number of unusual characteristics including a high tendency to mis-fold but its aetiological relevance remains unclear. The role of class I HLA antigens in pathogenesis is supported by the fact that HLA-B27 transgenic mice spontaneously develop arthritis, skin, gut and genitourinary lesions.

There are clues that infections play a role, possibly by molecular mimicry, with parts of the organism which are structurally similar to the HLA molecule triggering cross-reactive antibody formation. This is unproven. AIDS is increasing the prevalence of reactive arthritis and spondylitis in sub-Saharan Africa even in the absence of HLA-B27. The explanation for this changing epidemiology is unclear.

The types of arthritis that follow a precipitating infection are called reactive arthritis (p. 529).

The specialized immune systems of the gut and genitourinary mucous membranes may also play a causal role, perhaps reacting to local infections or to antigens which cross the damaged mucosa.

Epidemiology and pathogenesis

Environmental factors may also be involved but although Gram-negative organisms, e.g. Yersinia, Klebsiella, Salmonella, Shigella, can cause a reactive arthropathy, there is no conclusive evidence for their involvement in the pathogenesis of AS.

There is lymphocyte and plasma cell infiltration and local erosion of bone at the attachments of the intervertebral and other ligaments (enthesitis). This heals with new bone (syndesmophyte) formation.

Clinical features

Episodic inflammation of the sacroiliac joints in the late teenage years or early 20s is the first manifestation of AS. Pain in one or both buttocks and low back pain and stiffness are typically worse in the morning and relieved by exercise. Initially the diagnosis is often missed because the patient is asymptomatic between episodes and radiological abnormalities are absent. Retention of the lumbar lordosis during spinal flexion is an early sign. Later, paraspinal muscle wasting develops.

Criteria for classifying inflammatory back pain as ankylosing spondylitis are shown in Box 11.14.

Spinal stiffness can be measured by Schober’s test: a tape measure is placed in the midline 10 cm above the dimples of Venus. Any movement of a marker at 15 cm during flexion is recorded. A reading of <5 cm implies spinal stiffness. Individuals may be able to touch the floor with a stiff back if they have good hip movements but serial measurement of the finger tip to floor distance highlights any change.

Non-spinal complications (uveitis or costochondritis) suggest the diagnosis of spondyloarthritis (Box 11.15). Costochondral junction inflammation causes anterior chest pain. Measurable reduction of chest expansion is due to costovertebral joint involvement.

Peripheral joint involvement is asymmetrical and affects a few, predominantly large joints. Hip involvement leads to fixed flexion deformities of the hips and further deterioration of the posture. Young teenage boys occasionally present with a lower-limb monoarthritis (see p. 546), which later develops into AS.

Acute anterior uveitis is strongly associated with HLA-B27 in AS and related diseases and is occasionally the presenting complaint. Severe eye pain, photophobia and blurred vision are an emergency (see p. 1062).

Overall clinical assessment is based on pain, tenderness, stiffness and fatigue using, e.g. the Bath Ankylosing Spondylitis Disease Activity Index.

Investigations

Figure 11.20 X-ray of ankylosing spondylitis. The sacroiliac joints are eroded and show marginal sclerosis (white arrows). There is bridging syndesmophyte formation at the thoracolumbar junction (black arrows).

Figure 11.21 X-ray of bamboo spine in ankylosing spondylitis. In advanced disease there is calcification of the interspinous ligaments and fusion of the facet joints as well as syndesmophytes at all levels. The sacroiliac joints fuse.

Treatment

Prognosis

With exercise and pain relief, the prognosis is excellent and over 80% of patients are fully employed. Anti-TNF therapies are likely to reduce the morbidity of severe disease, reducing the risk of permanent spinal stiffness and progressive peripheral joint disease.

Patients should be made aware that they risk passing the HLA-B27 gene to 50% of their children. HLA-B27 positive offspring then have a 30% risk of developing AS.

Clinical features

Patterns of psoriatic arthritis include:

Figure 11.22 Hand showing psoriatic arthritis mutilans. All the fingers are shortened and the joints unstable, owing to underlying osteolysis.

Radiologically, psoriatic arthritis is erosive but the erosions are central in the joint, not juxta-articular, and produce a ‘pencil in cup’ appearance (Fig. 11.23). The skin and nail disease can be mild and may develop after the arthritis.

Figure 11.23 X-ray of psoriatic arthritis. There is osteolysis of the metatarsal heads and central erosion of the proximal phalanges to produce the ‘pencil in cup’ appearance (circle). All the lesser toes are subluxed.

Treatment and prognosis

NSAIDs and/or analgesics help the pain but they can occasionally worsen the skin lesions. Local synovitis responds to intra-articular corticosteroid injections.

In milder, polyarticular cases, sulfasalazine or methotrexate slows the development of joint damage.

When the disease is severe, methotrexate or ciclosporin is given because they control both the skin lesions and the arthritis. Anti-TNF-α agents, e.g. etanercept and golimumab (see p. 1210) and ustekinumab, are highly effective and safe for severe skin and joint disease. They should be given when methotrexate has failed. Corticosteroids orally may destabilize the skin disease and are best avoided but are valuable when injected into a single inflamed joint. Rituximab has no role in treating psoriatic arthritis.

The prognosis for the joint involvement is generally better than in RA.

Aetiology

A variety of organisms can be the trigger, including some strains of Salmonella or Shigella spp. in bacillary dysentery. Yersinia enterocolitica causes diarrhoea and a reactive arthritis. In NSU, the organisms are Chlamydia trachomatis or Ureaplasma urealyticum.

People with reactive arthritis are not more susceptible to infection but appear to respond differently. Bacterial antigens or bacterial DNA have been found in the inflamed synovium of affected joints, suggesting that this persistent antigenic material is driving the inflammatory process. The methods by which HLA-B27 increases susceptibility to reactive arthritis may include:

These are not mutually exclusive.

There are other organisms that also trigger reactive arthritis but have a different genetic basis; see post-streptococcal arthritis (p. 533), gonococcal arthritis (p. 546) and brucellosis (p. 533). In these, the borderline between reactive arthritis and septic arthritis is more indistinct and they can cause both.

Clinical features (Fig. 11.24)

The arthritis is typically an acute, asymmetrical, lower-limb arthritis, occurring a few days to a couple of weeks after the infection. The arthritis may be the presenting complaint if the infection is mild or asymptomatic. Enthesitis is common, causing plantar fasciitis or Achilles tendon enthesitis (see p. 509). Seventy per cent recover fully within 6 months but many have a relapse.

Figure 11.24 Clinical features of reactive arthritis.

In susceptible individuals with reactive arthritis, sacroiliitis and spondylitis may also develop. Sterile conjunctivitis occurs in 30%. Acute anterior uveitis complicates more severe or relapsing disease but is not synchronous with the arthritis.

The skin lesions resemble psoriasis:

Treatment

Treating persisting infection with antibiotics alters the course of the arthritis, once it has developed. Cultures should be taken and any infection treated. Sexual partners must be screened.

Pain responds well to NSAIDs and locally injected or oral corticosteroids. The majority of individuals with reactive arthritis have a single attack which settles, but a few develop a disabling relapsing and remitting arthritis. Relapsing cases are sometimes treated with sulfasalazine or methotrexate (Table 11.16). TNF-α blocking agents remain the drugs of next choice in severe and persistent disease but are rarely necessary.

Treatment

The inflammatory bowel disease should be treated (see p. 272). In all cases of enteropathic arthritis, the symptoms are helped by NSAIDs, although they may make diarrhoea worse. A monoarthritis is best treated by intra-articular corticosteroids. Sulfasalazine is more frequently prescribed than mesalazine as it may help both bowel and joint disease. The TNF-α blocking drug infliximab is used in inflammatory bowel disease (p. 272) and can help the arthritis.

Aetiology

Two main types of crystal account for the majority of crystal-induced arthritis. They are sodium urate and calcium pyrophosphate and are distinguished by their different shapes and refringence properties under polarized light with a red filter. Rarely, crystals of calcium apatite (see p. 515) or cholesterol cause acute synovitis.

Crystal arthritis. (a) Needle-shaped urate crystals. (b) A small intracellular pyrophosphate crystal. Both viewed under polarized light with a red filter.

Epidemiology

The prevalence of gout is increasing mainly in developed countries, due to changing diets – purine rich foods, high saturated fats, fructose containing drinks and alcohol. The prevalence is 1.4% in the UK (increasing with age to 3% in women and 7% in men) and 2.7% in the USA. Asian populations are more at risk as their diet becomes more Western. Gout develops in men more than women (10 : 1) and rarely occurs before young adulthood (when it suggests a specific genetic defect), and seldom in premenopausal females. Some 85–90% of cases are idiopathic. The prevalence in older females is increasing with increased diuretic use. Hyperuricaemia is common in certain ethnic groups (e.g. Maoris).

The last two steps of purine metabolism in humans are the conversion of hypoxanthine to xanthine and of xanthine to uric acid, catalysed by the enzyme xanthine oxidase. Primates lost the gene for uricase during their evolution about 10–20 million years ago. Hyperuricaemia possibly offered an evolutionary advantage.

Uric acid levels are higher in men than in women. There is a normal distribution of serum uric acid in the population with a skewed distribution at the upper end of the range. Hyperuricaemia is defined as a serum uric acid level greater than two standard deviations from the mean (420 µmol/L in males; 360 µmol/L in females). This is close to the limit of solubility – 360 µmol/L at 35°C and 300 µmol/L at 30°C.

Most people with hyperuricaemia are asymptomatic. Osteoarthritic joints are more prone to gouty attacks. The range for gouty individuals is higher than for normals, but the curves overlap (Fig. 11.25). Serum uric acid levels increase with age, obesity, a ‘Western’ diet (see above) and combined hyperlipidaemia, diabetes mellitus, ischaemic heart disease and hypertension (metabolic syndrome, p. 218). There is often a family history of gout.

Figure 11.25 Serum uric acid levels in controls and in those with gout. 7.9 mg/dL is equivalent to 474 µmol/L, 5.1 mg/dL is equivalent to 306 µmol/L.

(From Snaith ML, Scott JT. Uric acid clearance in patients with gout and normal subjects. Annals of Rheumatic Disease 1971; 30:285–289, with permission from the BMJ Publishing Group.)

Pathogenesis of hyperuricaemia and gout

Uric acid is the final product of endogenous and dietary purine metabolism in humans and levels in the blood depend on the balance between purine synthesis and the ingestion of dietary purines, and the elimination of urate by the kidney (66%) and intestine (33%).

Some 90% of people with gout have impaired excretion of uric acid (10% have increased production due to high cell turnover and <1% due to an inborn error of metabolism). Renal excretion is coordinated by a group of renal tubular urate transport molecules and a complex process of glomerular filtration, proximal tubule reabsorption via the urate transporter-1 (URAT-1) and active resecretion (Fig. 11.26). GLUT9 transports uric acid along with glucose and fructose into the cell from the tubule and back into the circulation. Both the entry of uric acid via the URAT-1 mechanism and the exit into circulation by GLUT9 can be blocked by uricosuric drugs such as probenecid. The body pool is about 1000 mg and 60% is turned over daily. Low-dose aspirin blocks uric acid secretion. Insulin resistance enhances uric acid resorption. Causes of hyperuricaemia are shown in Table 11.18.

Figure 11.26 Urate renal transport. The net result is that about 5–10% of the glomerular load is excreted in the urine under normal circumstances.

Table 11.18 Causes of hyperuricaemia

Clinical features

Hyperuricaemia may be asymptomatic. It also causes:

Acute gout presents typically in a middle-aged male with sudden onset of agonizing pain, swelling and redness of the first MTP joint. The attack occurs at any time, but may be precipitated by too much food or alcohol, by dehydration or by starting a diuretic. Untreated attacks last about 7 days. Recovery is typically associated with desquamation of the overlying skin. In 25% of attacks, a joint other than the great toe is affected.

In severe attacks, overlying crystal cellulitis makes gout difficult to distinguish clinically from infective cellulitis. A family or personal history of gout and the finding of a raised serum urate suggest the diagnosis but, if in doubt, blood and joint fluid cultures should be taken.

Chronic tophaceous gout (see below).

Investigations

The clinical picture is often diagnostic, as is the rapid response to NSAIDs or colchicine.

Treatment

The use of NSAIDs or coxibs in high doses rapidly reduces the pain and swelling. The first dose should be taken at the first indication of an attack:

After 24–48 hours, reduced doses are given for a further week. Caution: NSAIDs may cause renal impairment. In individuals with renal impairment or a history of peptic ulceration, alternative treatments include:

Diagnosis

The diagnosis is made by detecting rhomboidal, weakly positively birefringent crystals in joint fluid, or deduced from the presence of chondrocalcinosis on X-ray. The joint fluid looks purulent. Septic arthritis must be excluded and joint fluid should be sent for culture. The attacks may be associated with fever and a raised white blood cell count.

Treatment

Aspiration of the joint reduces the pain dramatically but it is usually necessary to use an NSAID or colchicine, as for gout. If infection can be excluded, an intra-articular injection of a corticosteroid helps.

Clinical features

Suspected septic arthritis is a medical emergency. In young and previously fit people, the joint is hot, red, swollen and agonizingly painful and held immobile by muscle spasm. In the elderly and immunosuppressed and in RA the clinical picture is less dramatic, so a high index of suspicion is needed to avoid missing treatable but potentially severely destructive and occasionally fatal septic arthritis.

In 20% of patients, the sepsis affects more than one joint. Chronic destructive arthritis due to tuberculosis is rare.

Investigations

Treatment

This should be started immediately on diagnosis because joint destruction may occur within weeks. The joint should be immobilized initially and then physiotherapy started early to prevent stiffness and muscle wasting. Intravenous antibiotics should be given for 1–2 weeks. It is usual to give two antibiotics to which the organism is sensitive for 6 weeks, then one for a further 6 weeks, orally. Monitor clinically and with the ESR and CRP.

Epidemiology

SLE occurs worldwide and is about nine times as common in women as in men, with a peak age of onset between 20 and 40 years. The prevalence varies between ethnic groups, being highest (at 1 : 250) in African/Caribbean women. In other populations, the prevalence varies between 1 : 1000 and 1 : 10 000.

Aetiology

The cause is unknown but there are several predisposing factors.

Pathogenesis

When cells die by apoptosis, the cellular remnants appear on the cell surface as small blebs which carry self-antigens. These antigens include nuclear constituents (e.g. DNA and histones), which are normally hidden from the immune system. In people with SLE, removal of these blebs by phagocytes is inefficient so that they are transferred to lymphoid tissues where they can be taken up by antigen-presenting cells. The self-antigens from these blebs can then be presented to T cells which in turn stimulate B cells to produce autoantibodies directed against these antigens (p. 69). It has been shown that in some patients the autoantibodies are present in stored blood samples that were taken years before the patient developed clinical features of SLE. The combination of availability of self-antigens and failure of the immune system to inactivate B cells and T cells which recognize these self-antigens (i.e. a breakdown of tolerance, see Ch. 3, p. 69) leads to the following immunological consequences.

Pathology

SLE of the skin and kidneys is characterized by deposition of complement and IgG antibodies and influx of neutrophils and lymphocytes. Biopsies of other tissues are carried out less frequently but can show vasculitis affecting capillaries, arterioles and venules. The synovium of joints can be oedematous and may contain immune complexes. Haematoxylin bodies (rounded blue homogeneous haematoxylin-stained deposits) are seen in inflammatory infiltrates and are thought to result from the interaction of antinuclear antibodies and cell nuclei.

The pathology of lesions in other organs is described in the appropriate chapters.

Clinical features

The manifestations of SLE vary greatly between patients. Most patients suffer fatigue, arthralgia and/or skin problems. Involvement of major organs is less common but more serious (Fig. 11.28).

Figure 11.28 Clinical features of systemic lupus erythematosus (SLE).

Investigations

Management

Course and prognosis

An episodic course is characteristic, with exacerbations and complete remissions that may last for long periods. However, SLE can also be a chronic persistent condition. The mortality rate in SLE has fallen dramatically over the last 50 years; the 10-year survival rate is about 90%, but this is lower if major organ-based complications are present. Deaths early in the course of disease are mainly due to renal or cerebral disease or infection. Later coronary artery disease and stroke become more prevalent. Chronic progressive destruction of joints as seen in RA and OA occurs rarely, but a few patients develop deformities such as ulnar deviation. People with SLE have an increased long-term risk of developing some cancers, especially lymphoma.

Pathogenesis

Negatively charged phospholipids and β₂-glycoprotein I are present on the outer surface of apoptotic blebs and so aPLs are believed to arise by a similar mechanism to the lupus autoantibodies described above. Pathogenic aPLs bind to the N-terminal domain of β₂-glycoprotein I and this interaction is facilitated when the protein is bound to phospholipid on the surface of cells such as endothelial cells, platelets, monocytes and trophoblasts. This alters the functioning of those cells leading to thrombosis and/or miscarriage.

Clinical features

Since APS is defined by the presence of thrombosis and/or pregnancy loss it is not surprising that these are the most common features. Ischaemic strokes occur in about 20% of patients and deep vein thrombosis in about 40%. Unlike most causes of thrombophilia, APS can cause either arterial or venous thrombosis (though rarely both in the same patient). Of women who have had two or more spontaneous miscarriages, 27% have APS.

However, large studies show that people with APS can also have many other features including:

Occasionally, APS is catastrophic. Catastrophic APS is a rare variant (about 1% of cases) in which multiple infarcts in different organs of the body cause failure of multiple organs simultaneously. There is a high mortality from catastrophic APS.

Treatment

In people with APS who have had one or more thrombosis, the recommended treatment to prevent further thrombosis is long-term anticoagulation with warfarin. The optimal target INR is unclear and many patients are managed with lower target INR. Pregnant women with APS are given oral aspirin and subcutaneous heparin from early in gestation. This reduces the chance of a miscarriage but pre-eclampsia and poor fetal growth remain common. There are no definite guidelines for managing people with aPL who have never had thrombosis. Aspirin or clopidogrel are sometimes given prophylactically, especially in those with high IgG aPL. Warfarin is given much more rarely in these circumstances.

Pathology and pathogenesis

Clinical features (Fig. 11.29)

Raynaud’s phenomenon

Raynaud’s phenomenon is seen in almost 100% of cases and can precede the onset of the full-blown disease by many years.

Figure 11.29 Clinical features of systemic sclerosis.

Limited cutaneous scleroderma (LcSSc): 70% of cases

This usually starts with Raynaud’s phenomenon many years (up to 15) before any skin changes. The skin involvement is limited to the hands, face, feet and forearms. The skin is tight over the fingers and often produces flexion deformities of the fingers. Involvement of the skin of the face produces a characteristic ‘beak’-like nose and a small mouth (microstomia). Painful digital ulcers and telangiectasia with dilated nail-fold capillary loops are seen. Digital ischaemia may lead to gangrene. Gastrointestinal tract involvement is common. Pulmonary hypertension (PHT) develops in 21% of people with LcSSc and pulmonary interstitial disease also occurs.

Diffuse cutaneous scleroderma (DcSSc): 30% of cases

Initially oedematous in onset, skin sclerosis rapidly follows. Raynaud’s phenomenon usually starts just before or concomitant with the oedema.

Diffuse swelling and stiffness of the fingers is rapidly followed by more extensive skin thickening, which can involve most of the body in the severest cases. Later the skin becomes atrophic. Early involvement of other organs occurs with general symptoms of lethargy, anorexia and weight loss.

Heartburn, reflux or dysphagia due to oesophageal involvement is almost invariable and anal incontinence occurs in many patients. Malabsorption from bacterial overgrowth due to dilatation and atony of the small bowel is not infrequent, and more rarely dilatation and atony of the colon occurs. Pseudo-obstruction is a known complication.

Renal involvement is acute or chronic. Acute hypertensive renal crisis used to be the most common cause of death in systemic sclerosis. ACE inhibitors and better care along with dialysis and renal transplantation have changed this.

Lung disease, both fibrosis (in 41% of cases) and pulmonary hypertension (17% of cases), contributes significantly to mortality in SSc. PHT can be isolated or secondary to fibrosis, and high plasma levels of endothelin-1 are seen.

Myocardial fibrosis leads to arrhythmias and conduction defects. Pericarditis is found occasionally.

Sometimes, these systemic features occur without skin involvement (SSc sine scleroderma).

Investigations

Management

Treatment should be organ-based in order to try to control the disease. Currently, there is no cure. In contrast to many other ARDs, corticosteroids and immunosuppressants are rarely used in SSc, with the exception of SSc-related pulmonary fibrosis.

Prognosis

In limited cutaneous scleroderma the disease is often milder, with much less severe internal organ involvement and a 70% 10-year survival. Pulmonary hypertension is a significant later cause of death. Lung fibrosis and severe gut involvement also determine mortality. In diffuse disease, where organ involvement is often severe at an earlier stage, many patients die of pulmonary, cardiac or renal involvement. Overall, pulmonary involvement (vascular or interstitial) accounts for around 50% of scleroderma-related deaths.

Localized forms of scleroderma occur either in patches (morphea, p. 1218) or linear forms. These are more commonly seen in children and adolescents and do not convert into systemic forms, although ANA may occur in localized scleroderma and very occasionally there is co-existence of localized and systemic forms.

Clinical features

Adult polymyositis

Women are affected three times more commonly than men.

The onset can be insidious, over months, or acute. General malaise, weight loss and fever can develop during the acute phase, but the cardinal symptom is proximal muscle weakness. The shoulder and pelvic girdle muscles become wasted but are not usually tender. Face and distal limb muscles are not usually affected. Movements such as squatting and climbing stairs become difficult. As the disease progresses, involvement of pharyngeal, laryngeal and respiratory muscles can lead to dysphonia and respiratory failure. These severe complications are rare if the disease is treated early.

Adult dermatomyositis

This is also more common in women. Apart from muscle weakness these patients often suffer from myalgia, polyarthritis and Raynaud’s phenomenon but DM is primarily distinguished from PM by the characteristic rash. This typically affects the eyelids, where heliotrope (purple) discoloration is accompanied by periorbital oedema, and the fingers where one sees purple-red raised vasculitic patches. These patches occur over the knuckles (Gottron’s papules) in 70% of patients, and this appearance is highly specific for DM. Ulcerative vasculitis and calcinosis of the subcutaneous tissue occurs in 25% of cases. In the long term, muscle fibrosis and contractures of joints occur.

Purplish Gottron’s papules over the knuckles (dermatomyositis).

(Courtesy of David Paige, London.)

Other organ involvement (antisynthetase syndrome)

Some 20–30% of people with PM or DM have antibodies to tRNA synthetase enzymes. These people are more likely to develop pulmonary interstitial fibrosis, Raynaud’s phenomenon, arthritis and hardening and fissuring of skin over the pulp surface of the fingers (mechanic’s hands). This variant of PM/DM is sometimes called antisynthetase syndrome and often has a poor outcome. Respiratory muscles are affected in PM/DM and this compounds the effects of interstitial fibrosis. Dysphagia is seen in about 50% of patients owing to oesophageal muscle involvement.

Association with other ARD

There is an association with other ARD (e.g. SLE, RA and SSc) with their associated clinical features such as deforming arthritis, malar rash and skin sclerosis.

Association with malignancies

The relative risk of cancer is 2.4 for male and 3.4 for female patients, and a wide variety of cancers have been reported. The onset and clinical picture does not differ from that of typical DM/PM. The associated cancer may not become apparent for 2–3 years, and recurrent, refractory or ANA-negative DM should prompt a search for occult malignancy.

Malignancy (e.g. lung, ovary, breast, stomach) can also predate the onset of myositis, particularly in males with DM.

Childhood dermatomyositis

This most commonly affects children between the ages of 4 and 10 years. The typical rash of DM is usually accompanied by muscle weakness. Muscle atrophy, subcutaneous calcification and contractures may be widespread and severe. Ulcerative skin vasculitis is common and recurrent abdominal pain due to vasculitis is also a feature.

Investigations

Treatment

Bed rest may be helpful but must be combined with an exercise programme. Prednisolone is the mainstay of treatment; 0.5–1.0 mg/kg body weight as initial therapy continued until at least 1 month after myositis has become clinically and enzymatically inactive. Tapering of steroids must be slow. Early intervention with steroid-sparing agents such as methotrexate, azathioprine, ciclosporin, cyclophosphamide and mycophenolate mofetil is common, especially where there is clinical relapse or rise in CK as the dose of steroids is reduced. Intravenous immunoglobulin therapy (IVIG) is helpful in some recalcitrant cases. Treatment of childhood DM tends to be more intensive with earlier use of immunosuppressive agents. Use of biological agents such as rituximab has been described but they are not commonly used.

Pathology and investigations

Biopsies of the salivary gland or of the lip show a focal infiltration of lymphocytes and plasma cells.

Management

Symptomatic treatment is with artificial tears and saliva-replacement solutions. Hydroxychloroquine may help fatigue and arthralgia. Corticosteroids are rarely needed but are used to treat persistent salivary gland swelling or neuropathy.

Investigation of PMR

Investigation of GCA

The histological features of GCA are:

Clinical features

These include fever, malaise, weight loss and myalgia. These initial symptoms are followed by dramatic acute features that are due to organ infarction.

Investigations and treatment

Treatment is with corticosteroids, usually in combination with immunosuppressive drugs such as azathioprine.

Clinical features and treatment

The clinical features are:

The persistent fever plus at least four of the other five features should be present to make the diagnosis, or fewer than four if coronary aneurysms can be seen on two-dimensional echocardiography, MRI or angiography.

Cardiovascular changes in the acute stage include pancarditis and coronary arteritis leading to aneurysms or dilatation. Other features include diarrhoea, albuminuria, aseptic meningitis and arthralgia and, in most, there is a leucocytosis, thrombocytosis and a raised CRP. Anti-endothelial cell autoantibodies are often detectable.

Treatment is with a single dose of high-dose intravenous immunoglobulin (2 g/kg), which prevents the coronary artery disease, followed after the acute phase by aspirin 200–300 mg daily. There is no evidence that steroid treatment improves the outcome.

Treatment of small cell vasculitis

The treatment depends on the organs involved. Vasculitis confined to the skin may not require systemic treatment whereas involvement of major organs (e.g. lungs or kidneys in Wegener’s granulomatosis) requires high-dose corticosteroids, immunosuppression and sometimes plasma exchange. Two recent clinical trials have shown that depletion of B cells with rituximab is as effective as cyclophosphamide in treating ANCA-associated vasculitis and it is likely that this will become a common form of treatment in the near future.

Clinical features

The cardinal clinical feature is recurrent oral ulceration. The international criteria for diagnosis require oral ulceration and any two of the following: genital ulcers, defined eye lesions, defined skin lesions, or a positive skin pathergy test (see below). Oral ulcers can be aphthous or herpetiform. The eye lesions include an anterior or posterior uveitis or retinal vascular lesions. Cutaneous lesions consist of erythema nodosum, pseudofolliculitis and papulopustular lesions.

Other manifestations include a self-limiting peripheral mono- or oligoarthritis affecting knees, ankles, wrists and elbows; gastrointestinal symptoms of diarrhoea, abdominal pain and anorexia; pulmonary and renal lesions; thrombophlebitis (especially in the legs); vasculitis; a brainstem syndrome, organic confusional states and a meningoencephalitis. All the common manifestations are self-limiting except for the ocular attacks. Repeated attacks of uveitis can cause blindness.

The pathergy reaction is highly specific to Behçet’s disease. Skin injury, by a needle prick for example, leads to papule or pustule formation within 24–48 hours. Blood tests usually show raised ESR and CRP but not autoantibodies.

Treatment

Corticosteroids, immunosuppressive agents and ciclosporin are used for chronic uveitis and the rare neurological complications. Colchicine helps erythema nodosum and joint pain. Thalidomide may be useful in some cases although side-effects of drowsiness and peripheral neuropathy are common. It should not be used in pregnant women because of phocomelia (limb abnormalities). Anti-TNF agents can be used to control severe uveitis and serious manifestations such as neurological and gastrointestinal Behçet’s disease.

Treatment of JIA

Early recognition and aggressive treatment prevents joint damage and allows normal growth and development. There is no cure but clinical remission is an achievable goal. JIA should always be referred to a specialist paediatric rheumatology unit with facilities to assess and design treatment plans which aim to prevent long-term disability. These units also need facilities for rehabilitation, education and surgical intervention. NSAIDs reduce pain and stiffness but disease-modifying agents such as methotrexate are used to control moderate and severe disease. Corticosteroids are often required in systemic disease: intravenous pulsed methylprednisolone is used, followed by methotrexate (10–15 mg/m²) weekly to control disease and prevent growth suppression.

Cytokine modulators (see Table 11.16) are used if methotrexate fails, and are highly effective in all types except systemic-onset JIA where the results are variable. Etanercept and adalimumab are the commonest drugs used but anakinra, tocilizumab and abatacept are being used in systemic-onset JIA. Anakinra (p. 526), an IL-1β receptor antagonist, helps in methotrexate-resistant systemic onset disease. Sulfasalazine is used only in enthesitis-related JIA. Aspirin may be a cause of Reye’s syndrome and should not be used under the age of 12 years.

Prognosis

Before cytokine modulators, up to 50% of children developed long-term disability; 25% continued to have active arthritis into adult years. Death was due to infection or systemic disease with pericarditis or amyloidosis. The prognosis has much improved but long-term studies, particularly on safety, are awaited.

Bone structure

Bone is comprised of cells and a matrix of organic protein and inorganic mineral. Long bones (femur, tibia, humerus) and flat bones (skull, scapula) have different embryological templates, with varying proportions of cortical and trabecular bone.

Figure 11.32 Diagram of a longitudinal section of a growing long bone.

Bone cells

Bone growth and remodelling

Longitudinal growth occurs at the epiphyseal growth plate, a cartilage structure between the epiphysis and metaphysis (Fig. 11.32). Cartilage production is tightly regulated, with subsequent mineralization and growth finally arrested at between 18 and 21 years when the epiphysis and metaphysis fuse.

In adults, bone is regularly remodelled to ensure repair of microdamage and turnover of calcium and phosphate for homeostasis. Signals regulating initiation of remodelling include changes in osteocytes (apoptosis or altered signalling of sclerostin, prostaglandins and other molecules), resulting in altered balance of RANKL and OPG expression by adjacent osteoblasts. Regulation of bone formation involves reciprocal effects of wnt versus dickkopf (Dkk) and sclerostin on the LRP5/6-β-catenin pathway.

Remodelling is carried out by the basic multicellular unit (BMU) (Fig. 11.33). Retraction of bone lining cells precedes binding of multinucleate osteoclasts to the bone surface, resulting in bone resorption. Unknown factors limit the amount of bone resorbed, after which osteoblasts fill in the resorption cavity. Bone remodelling is said to be coupled as formation normally follows resorption. New bone formation without resorption may, however, occur in the adult skeleton in response to anabolic therapy such as parathyroid hormone peptides. Additional influences include systemic hormones of which oestrogen (in both sexes) is particularly involved, promoting survival of osteocytes and inhibiting osteoclastogenesis.

Figure 11.33 Bone is remodelled in response to alterations in two reciprocal systems. Quiescent bone (centre), may experience reduced load (left); in response, the osteocyte increases expression of sclerostin (SOST+), inhibiting the response to wnt via the LRP5/frizzled co-receptor complex (orange). Microdamage (left) causes osteocyte apoptosis, with direct osteoclast-generating effects via RANKL, and loss of the inhibitory effect of SOST on later bone formation (dashed line, SOST−). The osteoblastic lining cells retract from an area of bone, forming a bone remodelling unit, while increased expression of RANKL and reduced osteoprotegerin (OPG) activates formation of a bone-resorbing multinucleate osteoclast from circulating precursors; the resorbed area is then replaced by new osteoid formed by cuboidal osteoblasts. As new osteocytes are formed, SOST levels rise and the osteoblasts cease formation, the majority undergoing apoptosis. If bone experiences loading without damage (right), sclerostin expression is reduced (SOST−), increasing signal response to wnt, activation of osteoblast bone formation, and increased OPG expression. Anabolic therapy with PTH is likely to act in a similar fashion. Corticosteroids may increase osteocyte SOST expression, and stimulate expression of Dkk, another inhibitor of the wnt/LRP5/frizzled axis. Myeloma cells have dual lytic effects, with enhanced expression of RANKL and expression of Dkk. In rheumatoid arthritis, similarly, RANKL and Dkk are increased, whereas in spondyloarthritis, characterized by new bone formation alongside erosion, Dkk is inhibited, with the increased wnt activity also increasing OPG relative to RANKL.

Parathyroid hormone (PTH)

PTH, an 84 amino-acid hormone, is secreted from the chief cells of the parathyroid gland, which bear calcium-sensing and vitamin D receptors. PTH increases renal phosphate excretion and increases plasma calcium by:

Hypomagnesaemia can suppress the normal PTH response to hypocalcaemia.

Biochemical markers of bone formation and resorption

While these are available in many laboratories, their use is limited by large biovariability and measurement variance. Serial measurements at the same time of day in individual patients are useful in assessing response to treatment of metabolic bone diseases. In addition, measurements of bone turnover markers may have a role in the assessment of fracture risk.

Diagnostic imaging

Figure 11.36 Computerized images of bone biopsies in osteoporosis demonstrate changes in bone architecture over time. Upper pair without treatment, the bone cortex becomes thinner, and numerous trabecular rods and plates are lost. Centre pair in response to bisphosphonate treatment, resorption is reduced, with preservation of bone mineral and structure. Lower pair in response to anabolic therapy, bone is increased at both trabecular and cortical sites.

(From Jiang Y, Zhao JJ, Mitlak BH et al. Recombinant human parathyroid hormone (1–34) improves both cortical and cancellous bone structure. Journal of Bone and Mineral Research 2003; 18:1932–1941, with permission of the American Society for Bone and Mineral Research.)

Definition and incidence

Osteoporosis is defined as ‘a disease characterized by low bone mass and micro-architectural deterioration of bone tissue, leading to enhanced bone fragility and an increase in fracture risk’.

The World Health Organization (WHO) defines osteoporosis as a bone density of 2.5 standard deviations (SDs) below the young healthy adult mean value (T-score ≤−2.5) or lower. Values between −1 and −2.5 SDs below the young adult mean are termed ‘osteopenia’. The rationale for this definition is the inverse relationship between bone mineral density and fracture risk in postmenopausal women and also older men. However, this definition should not be applied to younger women, men or children.

Fractures due to osteoporosis are a major cause of morbidity and mortality in elderly populations, with osteoporotic fractures of the spine causing acute pain or deformity and postural back pain. One in two women and one in five men aged 50 years will have an osteoporotic fracture during their remaining lifetime. Caucasian and Asian races are particularly at risk. As the risk of fracture increases exponentially with age, changing population demographics will increase the burden of disease.

Pathogenesis

Osteoporosis results from increased bone breakdown by osteoclasts and decreased bone formation by osteoblasts leading to loss of bone mass.

Bone mass decreases with age (Fig. 11.37) but will depend on the ‘peak’ mass attained in adult life and on the rate of loss in later life. Genetic factors are the single most significant influence on peak bone mass, but multiple genes are involved, including collagen type 1A1 (p. 494), vitamin D receptor and oestrogen receptor genes. Nutritional factors, sex hormone status and physical activity also affect peak mass.

Figure 11.37 Lifetime changes in bone mineral density (BMD). Peak bone mass is achieved between 20 and 30 years of age (gain), and consolidated up to around the age of 40 years. Then, age-related bone loss occurs in both men and women, with an accelerated loss in women starting around the time of menopause which lasts between 5 and 10 years.

Clinical features

Fracture is the only cause of symptoms in osteoporosis. Sudden onset of severe pain in the spine, often radiating around to the front, suggests vertebral crush fracture. However, only about one in three vertebral fractures is symptomatic. Pain from mechanical derangement, increasing kyphosis, height loss and abdominal protuberance follow crushed vertebrae. Colles’ fractures typically follow a fall on an outstretched arm. Fractures of the proximal femur usually occur in older individuals falling on their side or back. Other causes of low-trauma fractures must not be overlooked, including metastatic disease and myeloma.

Investigations

Plain radiographs usually show a fracture and may reveal previously asymptomatic vertebral deformities. Such clinically silent fractures may also be detected during the DXA scan with an additional analysis (called lateral vertebral assessment, Fig. 11.38) carried out with a much lower radiation dose than conventional imaging.

Figure 11.38 Vertebral fracture as seen on lateral vertebral assessment (LVA) and plain radiograph. (a) The LVA image shows a vertebra with reduced height, at T12. (b) The radiograph shows a typical wedge compression fracture.

Bone density

Dual energy X-ray absorptiometry (DXA) measures areal bone density (mineral per surface area rather than a true volumetric density), usually of the lumbar spine and proximal femur. It is precise, accurate, uses low doses of radiation and is the ‘gold standard’ in osteoporosis diagnosis (Fig. 11.39). Because of osteophytes, spinal deformity and vertebral fractures, spinal values may be artefactually elevated and should be interpreted with caution in the elderly.

Quantitative ultrasound of the calcaneum. This does not require ionizing radiation and is cheaper than other methods. It cannot be used for diagnostic purposes but is useful as a screening procedure prior to DXA assessment. The WHO T-score definition of osteoporosis should only be applied to DXA carried out at the hip and spine, and not to ultrasound, where a term such as ‘low bone mass’ is preferable.

Quantitative CT scanning allows true volumetric assessment, and distinction between trabecular and cortical bone. However, it is more expensive, requires higher radiation than other techniques, and to date offers no clinical advantage.

Figure 11.39 Dual energy X-ray absorptiometry of the lumbar spine. The AP image of the lumbar spine is shown on the left and on the right, the patient’s value is expressed in relation to the reference range.

Associated disease and risk factors

Investigations to exclude other diseases or identify contributory factors associated with osteoporosis should be performed and are particularly necessary in men, in whom secondary causes are more common (Table 11.23).

Selection of individuals for treatment: risk assessment

The purpose of treatment is to reduce the risk of fractures (Box 11.18). Thus, assessment of absolute fracture risk should be made in every case. Although bone mineral density measurements in the spine and proximal femur provide useful information about fracture risk, they have a relatively low sensitivity and the majority of fragility fractures occur in women with a T-score ≥−2.5. Prediction of fracture risk can be improved by the addition of risk factors that are at least partially independent of bone mineral density (Table 11.23). This forms the basis of a ‘fracture risk tool’ for clinical practice that has been developed under the auspices of the WHO, which provides an algorithm for calculation of 10-year fracture probabilities, based on independent clinical risk factors with and without bone mineral density values. The intervention threshold can then be determined by the cost-effectiveness of treatment and by clinical judgement.

 Box 11.18 Management of osteoporosis

summary

Treatment is guided by risk of fracture, not BMD alone.

If intermediate risk from clinical factors, request DXA scan (see: www.shef.ac.uk/FRAX or other risk calculator until familiar with assessments).

Do not underestimate the risk from steroids or previous fracture.

Many guidelines (e.g. NICE) recommend bisphosphonate as first-line drugs in most cases.Other options include:

strontium ranelate or denosumab:

– in young (to defer bisphosphonate use)

– new fracture on a bisphosphonate or a fall in BMD or bisphosphonate use for 5–10 years

teriparatide if multiple vertebral fractures or high risk

i.v. zoledronate after hip fracture

BMD monitoring is required in:

selected high-risk cases

low-risk cases not treated

All postmenopausal women and older men with a history of fragility fracture should be reviewed for treatment. In those aged >75 years, DXA is often not necessary prior to treatment but in those <75 years, DXA is useful in guiding treatment decisions (Table 11.24). However, other risk factors should also be taken into account when deciding whether to treat, particularly age and a history of previous fracture. Thus individuals with higher BMD values but with other clinical risk factors have a greater fracture probability than those with low BMD values in the absence of risk factors (Fig. 11.40).

Table 11.24 Indications for DXA scanning

Radiographic osteopenia Previous fragility fracture (in those aged <75 years) Glucocorticoid therapy (in those aged <65 years) Body mass index below 19 (kg/m2) Maternal history of hip fracture BMD-dependent risk factors in Table 11.23

In patients presenting with height loss and/or kyphosis, lateral thoracic spine X-ray should be the initial investigation.

Figure 11.40 A graph illustrating the combined effects of age (x-axis) and reduced bone mass (expressed as T-scores on the z-axis) on the 10-year probability of fracture (y-axis) in a population of women. Reduced bone mass osteopenia, T-scores between –1 and –2.5. Osteoporotic bone (T-scores <−2.5) is shown in pale green. Note that the risk of fracture may be greater in an older woman with osteopenia than in a young woman with osteoporosis.

(Data from Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk. Lancet 359:1929–1936.)

Prevention and treatment

Pharmacological intervention (Fig. 11.36)

Most interventions used act by inhibiting bone resorption (anti-resorptives) although parathyroid hormone peptides stimulate bone formation. The mechanism of action of strontium ranelate remains incompletely defined.

The evidence base for anti-fracture efficacy of interventions varies. Adequately powered randomized controlled trials, with fracture as the primary endpoint, exist for alendronate, risedronate, ibandronate, zoledronate, raloxifene, hormone replacement therapy, strontium ranelate, denosumab, teriparatide (recombinant human PTH peptide 1–34), human recombinant parathyroid hormone 1–84 and combined calcium/vitamin D (the latter in frail older individuals only).

Some interventions have been shown to reduce fracture at vertebral and non-vertebral sites, including the hip, whereas others have not been demonstrated to be effective at all sites (Table 11.25). Since a fracture at one site increases the risk of subsequent fracture at any site, treatments with efficacy at all major fracture sites (particularly spine and hip) are preferable. Hence, the bisphosphonates and strontium ranelate are generally regarded as first-line options in the majority of postmenopausal women with osteoporosis.

Table 11.25 Medications to reduce fracture risk in postmenopausal women

Bisphosphonates, synthetic analogues of bone pyrophosphate, adhere to hydroxyapatite and inhibit osteoclasts. Alendronate and risedronate are given most commonly as once weekly doses (70 mg and 35 mg, respectively) and zoledronate is given as a once yearly infusion of 5 mg. Ibandronate is available as a once monthly oral therapy (150 mg per month) or as a three-monthly intravenous injection (3 mg per 3 months). However, anti-fracture efficacy at non-vertebral sites has only been shown in high-risk subgroups with this latter choice.

– Oral bisphosphonates should be taken in the fasting state with a large drink of water, while the patient is standing or sitting upright. The patient should subsequently remain upright and avoid food and drink for at least 30 min.

– Bisphosphonates are generally well tolerated but may be associated with upper gastrointestinal side-effects such as oesophagitis, particularly if the dosing instructions are not closely followed. Bisphosphonates should be used with caution in patients who have chronic kidney disease (stage 4 or 5). Osteonecrosis of the jaw is rarely seen following high dose i.v. nitrogen-containing bisphosphonate in patients who have malignant disease. It is associated with poor dental hygiene. An association with bisphosphonate use in osteoporosis has not been clearly established.

– The optimal duration of bisphosphonate therapy is unknown; prolonged suppression of bone turnover may have adverse effects, including atypical femoral fractures, and it is currently advised to reassess treatment after 5–10 years, although benefits will frequently exceed the very low risk of these uncommon fractures.

Strontium ranelate consists of two stable strontium atoms linked to an organic acid, ranelic acid. Its mechanism of action remains uncertain although it has weak anti-resorptive activity whilst maintaining bone formation. It reduces the risk of vertebral, hip and other non-vertebral fractures in postmenopausal women with osteoporosis. The dose is 2 g daily, given as granules dissolved in water at night. Nausea, diarrhoea and headaches are infrequent side-effects, and there is a small increase in the risk of venous thromboembolism.

Denosumab is a fully human monoclonal antibody to RANKL. It is administered as a single subcutaneous injection every 6 months. Denosumab is an anti-resorptive agent which increases bone mineral density and reduces fractures at the spine, hip and other non-vertebral sites. Fracture risk reduction is equivalent to bisphosphonates. In addition to promoting osteoclastogenesis, RANKL has a role in the immune system and denosumab has been associated with exacerbations of eczema and a small increase in severe cases of cellulitis.

Raloxifene 60 mg daily is a selective oestrogen-receptor modulator (SERM). It has no stimulatory effect on the endometrium but activates oestrogen receptors in bone. It prevents BMD loss at the spine and hip in postmenopausal women, though fracture rates were only reduced in the spine. It also reduces the incidence of oestrogen-receptor-positive breast carcinoma in women treated for up to 4 years. Leg cramps and flushing may occur and the risk of thromboembolic complications is also increased to a degree similar to that seen with HRT. Its use is associated with a small increase in the risk of stroke. Lasofoxifene is a new SERM undergoing clinical trials.

Recombinant human parathyroid hormone peptide 1–34 (teriparatide) and recombinant human parathyroid hormone 1–84 are anabolic agents that stimulate bone formation. Teriparatide reduces vertebral and non-vertebral fractures in postmenopausal women with established osteoporosis, although data on hip fracture are not available. It is given by daily subcutaneous injection in a dose of 20 µg for 18–24 months. Recombinant human parathyroid hormone 1–84 is also administered by once daily subcutaneous injection of 100 µg and has been only shown to reduce vertebral, but not non-vertebral fractures. An anti-resorptive drug should be given after parathyroid hormone peptide therapy to maintain the increase in bone mineral density. Non-osteoporotic bone diseases such as osteomalacia should be excluded prior to treatment. Parathyroid hormone peptide therapy is mainly indicated in severe cases of vertebral osteoporosis or in women who fail to respond to other therapies. Teriparatide causes only mild transient hypercalcaemia and routine monitoring is not required. Nausea and headache may occur. Recombinant human parathyroid hormone 1–84 is associated with a higher incidence of hypercalcaemia and hypercalciuria and routine monitoring is advised. Neither agent should be used in people with skeletal metastases or osteosarcoma.

Hormone replacement therapy (HRT). Because of adverse effects on breast cancer and cardiovascular disease risk, HRT is a second-line option for osteoporosis except in early postmenopausal women at high fracture risk who also have perimenopausal symptoms.

Calcitriol (1,25-(OH)₂D₃) and calcitonin may reduce vertebral fracture rate, although the data are inconsistent.

Combination therapies, either with two anti-resorptive agents or an anti-resorptive and an anabolic agent, often produce larger increases in BMD than monotherapy but have not been shown to result in greater fracture reduction than with monotherapy and combination therapy is not advised.

FURTHER READING

Cummings SR, Martin J, McClung MR et al. Denosumab for prevention of fractures in postmenopausal women with osteoporosis. N Engl J Med 2009; 361:756–757.

Favus MJ. Bisphosphonates for osteoporosis. N Engl J Med 2010; 363:2027–2035.

National Osteoporosis Foundation Clinical Guidelines; http://www.nof.org/professionals/clinical-guidelines.

Rachner TD, Khosla S, Hofbauer LC. Osteoporosis: now and the future. Lancet 2011; 377:1276–1287.

Glucocorticoid-induced osteoporosis

Individuals requiring continuous oral glucocorticoid therapy for 3 months or more (at any dose) should be assessed for co-existing risk factors (age, previous fracture, hormone status). Postmenopausal women, men aged over 50 years and any individuals who have sustained a fragility fracture should receive treatment without waiting for DXA scanning. DXA results and fracture risk assessment guide treatment for other patients (Table 11.24). For these individuals, bisphosphonates and teriparatide are the approved agents. Calcium and vitamin D supplementation should also be given.

Osteoporosis in men

Alendronate and risedronate increase BMD and reduce vertebral fractures in men with osteoporosis. Teriparatide is also approved for use in this setting. In men with osteoporosis who have clinical and biochemical evidence of hypogonadism, testosterone replacement is also used.

Collagen defects

Collagen is responsible for many of the structural, tensile and load-bearing properties in the various tissues where it is found. (The structure of collagen is discussed on page 494.) Thirty or more dispersed genes encode for more than 19 different types of collagen (Table 11.27).

Table 11.27 The major types of collagen

Miscellaneous defects