35 Arthritis

Heart failure (HF) and arthritis are both common in adults, increasing in prevalence with age. The term arthritis covers a wide spectrum of disease (Box 35.1) which may involve the cartilage, bone, tensile structures (ligaments, tendons, and muscles), or synovial tissue, and pathological processes can impact on any of these structures.

Osteoarthritis is the most common arthritis and results from cartilage fragmentation and loss. The major target for inflammatory arthritides, of which rheumatoid arthritis (RA) is the most common, is the synovium. RA affects 1% of the population, and premature death, predominantly from cardiovascular disease, is associated with an 8–15-year reduction in lifespan compared to age-matched controls.¹ The onset of HF is significant in patients with arthritis as there is evidence that deaths are attributed to HF more often for patients with RA than for the general population. However, this difference in mortality results from an increased incidence of HF rather than a worse prognosis overall.²

The main focus of this chapter is the interaction of HF with the inflammatory arthritides. As in the general population, the most common cause of HF in inflammatory arthritis is atherosclerotic coronary artery disease. Other causes are now less common. There is less extra-articular disease in inflammatory arthritis as a result of more aggressive management and better control of disease. However, other possible causes of HF include:

When HF and arthritis coexist, the pharmacotherapy of either condition can impact on the other, to a beneficial or detrimental effect (e.g. gout triggered by diuretic use).

Much of the literature on arthritis and HF comes from the extensive study of RA, where the clinical presentation and outcome of HF appears different from that in the general population. The extent of cardiac involvement in arthritis has been a source of ongoing interest for decades. In 1943, Bayles described post-mortem findings in the heart of patients with RA.³ As early as 1975 it was reported that there was an excess cardiovascular mortality in patients with RA.⁴ The prevalence of HF in the RA population increases with age and is greater in men than in women (Fig. 35.1).⁵

There is also a higher incidence of HF in patients with RA compared to those without (37.1% vs 27.7%, p 〈 0.001). In addition, RA patients with HF die sooner than patients with HF and no RA. Mortality rates among patients with and without RA were 39.0 and 29.2 per 1000 person-years, respectively (Fig. 35.2).²

As well as being more frequently female, RA patients with HF less frequently suffer from obesity, hypertension, and ischaemic heart disease.⁶ Their symptoms and signs at presentation are often more subtle and consequently they are less likely to have an echocardiogram. However, RA patients are more likely to have preserved systolic function (Fig. 35.3) but they have a higher mortality at 1 year (Fig. 35.4). These findings illustrate the difficulties in the assessment and management of this patient group, and highlight the importance and potential benefit in aggressively screening such patients.

In the general population, ischaemic cardiomyopathy is the most common cause of HF associated with left ventricular systolic dysfunction (LVSD). The presentation may be obvious, with chest pain prompting admission for management of an acute myocardial infarction, or exertional chest pain requiring management of chronic stable angina. However, the presentation may be that of silent ischaemia and insidious development of worsening breathlessness.

It is generally accepted that aggressive management of acute coronary artery occlusion minimizes the impact on systolic function. The optimal reperfusion strategy is currently prompt delivery of primary percutaneous coronary intervention.

As described, there is an excess of cardiovascular disease in many of the arthritides. This often manifests as coronary artery disease. In RA, the risk of coronary artery disease is equally elevated in both men and women and increases with disease severity, disease duration and evidence of extra-articular disease.⁷ The increased risk of developing, and dying from, coronary artery disease is significantly higher than in a non-RA individual.⁸ A recent meta-analysis of 24 observational studies, with a total study population of 111 758 patients, described that the risk of mortality related to coronary artery disease was 59% higher in patients with RA than in the general population.⁹

There is a difference in the presentation of coronary artery disease in patients with RA. Patients are twice as likely to experience unrecognized myocardial infarctions and sudden deaths, and less likely to report angina or to undergo coronary artery bypass grafting.¹⁰ The reduction in the presentation of acute myocardial ischaemia impacts on the therapeutic options available to these patients and the unheralded silent ischaemia persists and reduces systolic function. The risk of coronary artery disease in RA patients often precedes the diagnosis of RA, and the risk cannot be explained by an increased incidence of traditional coronary heart disease risk factors in RA patients.

The differential diagnosis of chest pain in patients with systemic inflammatory athritidies includes:

It is now well established that RA patients have an increased risk of cardiovascular disease. As a result, classic cardiovascular risk factors (smoking, hypertension, hyperlipidaemia) are increasingly targeted and annual screening for these risk factors is now advocated.¹¹

There is a well-described dyslipidaemia in RA, with an inverse association between inflammatory markers (C-reactive protein or ESR) and HDL cholesterol.¹² A reduction in inflammation with RA-modifying drugs correlates with increase in HDL and total cholesterol. Interestingly, the Trial of Atorvastatin in Rheumatoid Arthritis (TARA) study demonstrated that hydroxymethylglutaryl-coenzyme A reductase inhibitor (statin) therapy conferred an improved lipid profile and achieved clinically apparent anti-inflammatory effects with reduction in RA disease activity score.¹³ The pleiotropic effects of statins are thought to include anti-inflammatory, antifibrotic, and antioxidant effects;¹⁴ prevention of left ventricular hypertrophy;¹⁵ reduction of endothelial dysfunction;^16,17 inhibition of neurohormonal activation; and prevention of cardiac arrhythmias.¹⁸ These are all potentially beneficial in the management of HF.

Psoriasis is a systemic inflammatory condition that primarily manifests as a skin disorder in 1–3% of the population. Of those affected, 6–11% have psoriatic arthropathy. In the United States, the prevalence of all heart diseases in patients with psoriasis has been estimated at 14.3% compared with 11.3% for the entire US population.¹⁹ There is an increased prevalence of diabetes, hypertension, hyperlipidaemia, smoking, and increased body mass index (BMI) in patients with psoriasis. These key cardiovascular risk factors that comprise the metabolic syndrome are more strongly associated with severe rather than mild psoriasis.²⁰

The literature continues to debate whether psoriasis is an independent risk factor for myocardial infarction. Gelfand et al.²¹ demonstrated that the risk of myocardial infarction associated with psoriasis is greatest in young patients with severe psoriasis, is attenuated with age, and remains increased even after controlling traditional cardiovascular risk factors (Fig. 35.5). In a recent study from the United States, the 10-year risks for coronary heart disease were 28% greater for patients with psoriasis than in the general population.²² However, a very large Dutch cohort did not find psoriasis to be a clinically relevant risk factor for ischaemic heart disease hospitalizations.²³ It remains to be seen whether aggressive control of the traditional risk factors and of systemic inflammation will impact on the evolution of coronary artery disease in patients with psoriasis and more specifically psoriatic arthropathy. In RA classic risk factors alone do not explain the excess vascular disease. Del Rincon et al.²⁴ reported a fourfold higher incidence of cardiovascular events relative to a community-dwelling cohort. Adjusting for conventional risk factors, this risk ratio was only minimally attenuated, suggesting that classic risk factors do not fully explain the accelerated atherogenesis in RA.

The role of autoimmune processes is well-described in relation to several arthritic conditions including RA and SLE. A causal association between autoimmune disease and primary heart disease continues to be an area of research. There is increasing evidence that active autoimmune arthritis is associated with heart disease including HF. Sattar et al.²⁵ hypothesized that it is the severity and chronicity of the systemic inflammation that is particularly damaging. Even when the arthritis is clinically quiescent, cytokine production remains high and continues to promote vascular risk.

Inflammatory mediators can be identified in coronary heart disease, but require high-sensitivity assays. Parallels can be drawn between coronary heart disease, with inflammatory molecules and immune cells in the cap region of unstable atheromatous plaques, and the inflammatory process in synovitis. Cytokines, particularly tumour necrosis factor (TNF) α and interleukins IL-1β and IL-6, can be identified in both HF and the inflammatory arthritides. TNFα has been implicated in the pathogenesis of HF and cardiac cachexia.²⁶ In a follow-up study of the SOLVD (Studies Of Left Ventricular Dysfunction) population levels of TNFα were significantly higher in patients with HF compared with the normal controls.²⁷ Furthermore, higher cytokine concentrations are associated with a worse prognosis in patients with advanced HF (Fig. 35.7).²⁸

TNFα may mediate cardiac myocyte hypertrophy and influence other pathways involved in cardiac remodelling culminating in cardiac dysfunction. TNFα may have time-dependent opposing effects and other cytokines (e.g. IL-1β, IL-10) may also carry independent pathophysiologic importance in HF.²⁹

The metabolic effects of cytokines (particularly TNFα, IL-1β, and IL-6) include lipid alterations and peripheral insulin resistance. These effects can be seen as beneficial in the acute-phase response to acute inflammation, but chronic cytokine activity and related metabolic effects are proatherogenic.

The end result of the inflammatory process on cardiac structure and function has recently been described with cardiac magnetic resonance (CMR) imaging.³⁰ Compared with non-RA patients, RA patients demonstrate a markedly lower left ventricular mass, and a modest decrease in ejection fraction and stroke volume. Disease severity, suggested by higher titres of antibodies to cyclic citrullinated peptide (anti-CCP) and the use of biologic agents, probably correlates with lower left ventricular mass, left ventricular stroke, and end-diastolic volumes.

Myocarditis is an uncommon presentation of inflammatory arthritis. In patients with SLE it is often asymptomatic and can affect between 8–25% of patients,^31,32 particularly in African-American patients. Myocarditis should be considered if there is a persistent tachycardia with nonspecific ST and T wave abnormalities. Echocardiography is indicated and may show systolic or diastolic dysfunction. Establishing the diagnosis of myocarditis can be challenging and may require endomyocardial biopsy. Histology demonstrates mononuclear cell infiltration of the myocardium. In established cases with a more prolonged myocarditic process the inflammation may lead to fibrosis, which may be present clinically as dilated cardiomyopathy, or cause conduction abnormalities. These may present as bradycardia or tachyarrhythmias, both of which may further impact on HF. Myocarditis and HF is also described in other non-ANCA-associated vasculilides, e.g. Takayasu’s arteritis.

Churg–Strauss syndrome (CSS) is the ANCA-associated vasculitis most likely to involve the heart, with 14% of patients developing myocardial involvement and HF. Chronic HF is a significant cause of death in CSS.³³

Pericarditis is described in a number of inflammatory arthritides including RA, ankylosing spondylitis, SLE, scleroderma, and vasculitis. In RA clinical symptoms of pericarditis are infrequent (3%); however, echo studies have reported up to 50% of patients affected and at autopsy up to 30% of patients are found to have had pericardial disease. If effusions are present they are usually small and do not require treatment.³⁴

Valvular disease—in particular mitral and aortic insufficiency—is documented in RA. In ankylosing spondylitis aortic regurgitation is more common. Patients with SLE can develop valve thickening or Libman–Sacks endocarditis. Valvular heart disease due to scleroderma is rare. In Takayasu’s vasculitis aortic dilatation may involve the aortic valve and in Wegener’s granulomatosis both aortic and mitral regurgitation are documented.

Diastolic dysfunction is more common with increasing age. Impaired ventricular filling and other measures of diastolic dysfunction, including transmitral blood flow velocities and valve annulus velocities, have been assessed in patients with arthritis. These echocardiographic findings are more common in patients with RA than in the general population,^35,36 and a correlation with RA disease duration, perhaps due to subclinical myocardial involvement, has been demonstrated.

Diastolic dysfunction may predate clinically apparent HF. In a study of patients with SLE without clinically evident cardiovascular disease, no significant difference in systolic function was identified, but there was evidence of diastolic dysfunction.³⁷

Reactive or secondary (AA) amyloidosis refers to predominantly extracellular tissue deposition of fibrils composed of fragments of serum amyloid A protein, an acute-phase reactant. In poorly controlled chronic inflammatory arthritis, mainly seropositive RA and ankylosing spondylitis, cardiac AA amyloidosis can occur. In the context of RA, it has been reported that up to 40% of patients have cardiac involvement,³⁸ although significant deposition of AA amyloid in the heart is uncommon and is therefore rarely the cause of death.

Cardiac involvement may be suspected by low-voltage complexes in the limb leads or a pseudo-infarct pattern on ECG. Amyloid infiltration of the heart results in increased echogenicity, and therefore echocardiography is the noninvasive test of choice. Early features include left ventricular wall thickening and evidence of diastolic dysfunction. Right ventricular diastolic dysfunction can also occur. If AA amyloidosis is clinically suspected and suggested on noninvasive tests then it can be diagnosed by rectal, subcutaneous abdominal fat, skin, or, if necessary, cardiac biopsies.

However, in the current era of arthritis care, with more aggressive disease control, it is very unusual for patients to develop HF related to AA amyloidosis. A recent study of the natural history and outcome in systemic AA amyloidosis found that cardiac failure attributable to AA amyloidosis was present in only 1 of the 374 patients assessed (60% of whom had chronic inflammatory arthritis as the cause of AA amyloidosis); 224 patients underwent echocardiography and only 2 of them had findings consistent with cardiac infiltration.³⁹

The management of arthritis and HF involves polypharmacy. Many of the agents used to treat arthritis can negatively impact on HF, and in some cases may be the cause of HF.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are at the front line of disease control. NSAIDs offer both analgesic and anti-inflammatory properties, but do not slow disease progression. NSAIDs have several potential adverse cardiovascular effects, including interference with the antiplatelet action of aspirin, an increase in cardiovascular events including myocardial infarction, and exacerbation of established HF.

The attenuation of the antiplatelet effect of aspirin is well described with both ibuprofen and naproxen.^40,41 The presumed mechanism is competitive binding at the cyclooxygenase (COX)-1 receptor. The impact of this interaction is the potential for increased thrombotic coronary events, and subsequent HF. This has yet to be proved.

Studies have not demonstrated an increase in first episodes of HF with NSAID use.⁴² However, in established HF NSAIDs are associated with relapse of HF symptoms and increased mortality. The proposed mechanism is an increase in afterload resulting from NSAID-induced systemic vasoconstriction, which can lead to a further reduction in cardiac contractility and cardiac output. This effect is exacerbated by hyponatraemia which is a marker of advanced HF. The adjusted risk of rehospitalization for HF is significantly increased in patients on diclofenac or ibuprofen.⁴³ There was a dose-dependent increase in risk of death, which was highest with diclofenac (adjusted hazard ratio 2.08). Higher doses of ibuprofen (〉1200 mg/day) and naproxen (〉500 mg/day), but not lower doses, were also associated with an increased risk of death.

In an attempt to overcome the gastrointestinal adverse effects of nonselective NSAIDS, COX-2 selective inhibitors were developed. Unfortunately, because of significant cardiovascular toxicity many of the agents have been withdrawn. There is an increase in ischaemic coronary events that are presumed to be due to the reduced prostacyclin production by vascular endothelium without inhibition of production of the prothrombotic platelet thromboxane A₂.

In a cohort study to assess the impact of COX-2 selective inhibitors on the incidence of HF, crude rates of hospitalization for HF per 100 patient-years of exposure were 0.9 for the controls, 2.4 for the patients treated with rofecoxib, and 1.3 for the patients treated with celecoxib.⁴⁴ Adjusting for potential confounding risk factors, the risk of hospitalization with HF was significantly higher in patients treated with rofecoxib (compared to controls), but not celecoxib. The rate of death was considered in another study that identified that there was a dose-dependent increase in risk of death that was highest with rofecoxib, celecoxib, and diclofenac.⁴³ A recent meta-analysis looking at both conventional NSAIDs and COX-2 selective inhibitors concluded that the risk of cardiac failure, albeit small, was similar with both types of NSAIDs and that pre-existing cardiac failure increased the risk.⁴⁵

The current evidence therefore suggests that both selective and nonselective NSAIDs should be used with caution in patients with established HF. If new HF develops, the potential cardiovascular adverse effects of these drugs, including myocardial infarction, should be considered as a potential mechanism.

Glucocorticoids are frequently used to achieve inflammatory control in the acute flare of arthritis. However, their use has been associated with increased rates of HF, myocardial infarction, stroke, and all-cause mortality in a dose-dependent manner.⁴⁶ The risk of HF increases with the daily dose of glucocorticoids, with a relative risk of 3.72 for prednisolone doses of 7.5 mg/day or more. Ongoing steroid use was associated with a higher risk, than intermittent courses.

A number of disease-modifying antirheumatic drugs (DMARDs) are used to treat RA, SLE, and other inflammatory arthritides. Cardiac toxicity secondary to DMARDs is uncommon. Methotrexate is the most commonly prescribed DMARD.

Sulphasalazine is also frequently used particularly in the UK. Potential cardiac side-effects listed for these agents are pericarditis (Methotrexate and Sulphasalazine) and myocarditis (Sulphasalazine). In practice this is very uncommon. Treatment would include withdrawal of the drug and steroids if required.

Chloroquine and hydroxychloroquine have a beneficial effect on the lipid profile in SLE. However, toxicity from either agent can cause a generalized myopathy, conduction abnormalities, and rarely a cardiomyopathy. In the cardiomyopathy, the ECG shows nonspecific T-wave changes, and echocardiography or invasive catheterization demonstrates a restrictive physiology.⁴⁷ Endomyocaridal biopsy may show myocyte degeneration.⁴⁸ Importantly, withdrawal of the agents achieves reversal of the HF syndrome.

Leflunomide and ciclosporin used in the management of RA and psoriatic arthritis can cause hypertension, and regular blood pressure monitoring is done for both these agents along with blood monitoring.

TNFα is a key agent in systemic inflammation coordinating the stimulation and release of the inflammatory cytokines (including IL-1β, IL-6, IL-8); up-regulation of endothelial adhesion molecules and chemokines; and the migration of leucocytes to targeted organs. The failing heart produces TNFα, but the normal heart does not. Clinical trials of TNFα blockers in advanced HF have been negative.^49,50

There is ongoing debate as to whether anti-TNFα therapy, which is being used increasingly in the treatment of systemic inflammatory diseases including arthritis, affects the risk of HF in these patients.⁵¹ The current evidence base to assess the risk include clinical trials and registry data. A German register of RA patients reported no increased risk of HF related to TNFα inhibition.⁵² On the other hand, Setoguchi et al.⁵³ found an adjusted hazard ratio of 1.70 for HF hospitalization in TNFα-treated patients versus those on methotrexate. All RA patients in this cohort were aged 65 or over and therefore the risk of HF in the elderly RA patient treated with TNFα antagonist may be greater. On the other hand, TNFα antagonist therapies in RA may ameliorate the deleterious cardiac effects of circulating TNFα. A recent study demonstrated that blocking TNFα in RA patients without evident HF decreases NT-proBNP levels by around 18%, suggesting no treatment-induced deterioration in cardiac function, and a potential cardiovascular risk benefit.⁵⁴ Despite this the current recommendation is to avoid use of anti-TNFα agents in patients with HF, especially in those with NYHA classes III or IV.⁵⁵

TNFα antagonists offer a significant advance in the management of inflammatory arthritis and so it is important that such a therapy be available to as many patients as appropriate. In patients with NYHA class I or II, TNFα antagonists can be considered but a baseline echo should be performed and careful clinical and echocardiographic follow-up should be employed. High doses of the agents should be avoided. If clinical HF symptoms develop or deteriorate then the TNFα antagonists should be stopped.

Gout is a clinical syndrome that results from the deposition of urate crystals in the joints. Granulocytes phagocytize the crystals

and then secrete inflammatory mediators that modulate an intense inflammatory reaction, which can result in an intensely painful acute arthritis and lead to joint destruction. The inflammatory process becomes self-perpetuating with increased lactic acid production reducing the synovial fluid pH, which favours further deposition of urate crystals.⁵⁶

The diagnosis is made by aspirating urate crystals from the joint. Serum urate levels are less helpful. Patients with acute gout can have a normal serum urate and patients with an elevated serum urate can be asymptomatic. Serum urate levels vary with age and sex, and also blood pressure, renal function, diet, and alcohol intake. However, the incidence of gouty arthritis among men with urate levels exceeding 540 µmol/L (9.0 mg/dL) is less than 5% per year.⁵⁷

Patients with chronic HF frequently present with gout (Fig 35.7). There are multiple factors that contribute to the hyperuricaemia seen in HF. These include hypertension and hyperlipidemia, which may be key in the aetiology of the HF; chronic renal failure, which may reflect a more extensive vascular disease or be consequent to the pharmacotherapy of HF; and diuretics, which are the cornerstone in the management of fluid status of HF patients. All diuretics can cause hyperuricaemia, but loop diuretics are the most likely, then thiazide diuretics;⁵⁸ spironolactone is the least likely to cause elevated uric acid levels.⁵⁹ Diuretic-induced hyperuricaemia can be minimized by the concurrent prescription of angiotensin converting enzyme (ACE) inhibitors or an angiotensin II receptor blocker (ARB),^60,61 possibly by inhibition of the proximal sodium and urate reabsorption induced by angiotensin II.

Interestingly, serum urate concentrations correlate with maximal oxygen uptake and NYHA functional class,⁶² as well as circulating markers of inflammation in patients with chronic HF.⁶³

The management of gout in the context of HF can be challenging. Usually, acute gout is managed with NSAIDs. However, as discussed above, NSAIDs are detrimental in HF because of sodium and volume retention, hyperkalaemia, and renal failure. Therefore, a modified approach has been described using a combination of NSAIDs and colchicine, which facilitates the early withdrawal of the NSAID (Fig. 35.8). In the context of acute gout with HF, naproxen is probably the NSAID of choice at as low a dose as is achievable.

The use of colchicine was established as a therapy for acute gout almost 250 years ago. If given soon after the onset of symptoms, it almost invariably achieves resolution of symptoms. Colchicine is an antimitotic agent that decreases the functional activity of locally infiltrating granulocytes and inhibits the release of pro-inflammatory mediators.⁶⁴ However, the serum urate levels are unchanged. It is best used for short (3–4 day) courses to a maximum dose of 6 mg per course. It can be very effective and has the advantages that unlike NSAIDs it does not induce fluid retention or interact with warfarin. If used for longer, however, it has the almost universal side effects of diarrhoea and vomiting.

Glucocorticoids can be used in acute gout both as anti-inflammatory agents and to reduce the serum urate concentration. Oral glucocorticoids are problematic in HF because of the mineralocorticoid effect of fluid retention. However, local joint aspiration and then injection with glucocorticoid targets the site of inflammation and reduces the systemic dose, as well as relieving the tense hot swollen joint by removing some of the synovial fluid. This is a particularly useful adjuvant in patients who are struggling with painful acute gout despite NSAID and colchicine.

In the longer term the serum urate level should be addressed. Patient education about diet modification, reducing the amount of meat, pulses, and alcohol, is important but usually fails to achieve adequate urate reduction. Therefore, pharmacological modification should be considered. This should be cautiously introduced 2 weeks after the acute episode and then up-titrated over the following weeks to reduce the likelihood of a further acute attack. Two classes of drugs are available in the management of hyperuricaemia: uricosuric agents (e.g. probenicid) and xanthine oxidase inhibitors (allopurinol). Allopurinol is the agent of choice and inhibits the synthesis of uric acid. The dose of allopurinol is titrated up from a starting dose of 100 mg/day by 100 mg increments to achieve a serum urate level below 0.3 mmol/L. The dose of allopurinol needs to be modified if renal failure coexists or develops. An additional potential benefit of allopurinol is its effect in reducing oxidative stress by inhibiting xanthine oxidase in the vascular endothelium, and thus preventing the formation of superoxide free radicals. In patients with HF, allopurinol has been shown to reduce markers of oxidative stress and improve endothelium-dependent vasodilatation.⁶⁵ There are also studies demonstrating that allopurinol improves left ventricular efficiency⁶⁶ and reverses ventricular remodelling.⁶⁷

The term ‘pseudogout’ describes an acute synovitis precipitated by calcium pyrophosphate dihydrate (CPPD) crystal deposition in joints. Clinically such an episode resembles gout caused by urate crystal deposition. The diagnosis of pseudogout is achieved by aspiration of the affected joint, and demonstration of positively birefringent CPPD crystals by polarized light microscopy.

The management of an acute episode of pseudogout is similar to gout, with NSAID and colchicine being the first-line agents and then local glucocorticoid if required. Long-term prophylaxis is different from that of gout. There are several reports of patients suffering recurrent attacks of pseudogout successfully treated using colchicine 0.6 mg twice daily as oral prophylaxis.⁶⁹

The importance of pseudogout in the context of HF is that it may reveal the underlying cause of HF. The majority of cases of CPPD crystal deposition are idiopathic, and it is more common in elderly people (mean age at symptomatic presentation is 72 years). However, CPPD crystal deposition is associated with several metabolic and endocrine conditions that may cause symptoms from the crystal deposition at a younger age, the best-characterized of which is haemachromatosis. The other accepted association of relevance is hypothyroidism. Both of these conditions may offer a reversible cause of HF.⁷⁰

In the assessment of patients with HF and arthritis there are several possible presentations:

The onset of arthritis in patients with HF may be the evolution of an incidental arthritis, or a complication of the aetiology of, or pharmacotherapy for, HF.

At present, there is no evidence to support screening the asymptomatic patient for evidence of cardiac dysfunction. However, there is a role for risk assessment with the established risk models. The recent European League Against Rheumatism (EULAR) recommendations state:

Annual cardiovascular risk assessment using national guidelines is recommended for all patients with RA and should be considered for all patients with ankylosing spondylitis and psoriatic arthropathy. Any cardiovascular risk factors identified should be managed according to local guidelines. If no local guidelines are available, cardiovascular risk management should be carried out according to the SCORE function. In addition to appropriate cardiovascular risk management, aggressive suppression of the inflammatory process is recommended to further lower the cardiovascular risk.¹¹

New-onset HF may be the first manifestation of incidental heart disease, or it may be either a consequence of the arthritic process, or the result of pharmacotherapy. The cause of the HF may be myocardial, pericardial, valvular, rhythm, or coronary artery related. Irrespective of the presentation, the assessment is similar. As usual, detailed history taking and clinical examination may provide many of the clues. Investigations may then include:

In some cases further assessment is appropriate, such as CMR or invasive investigation. For example, the distinction between constrictive pericarditis and restrictive cardiomyopathy often requires cardiac catheterization with simultaneous right and left heart pressures. Cardiac biopsy can provide tissue to identify potentially treatable causes such as antimalarial-induced cardiomyopathy.

The management of HF in patients with arthritis is similar to that in the general population. There is some evidence that inhibition of angiotensin II with ACE inhibitors or ARBs is associated with benefit in reducing the systemic inflammation.⁷¹ Caution should be used in the prescription of diuretics in patients with HF, as acute gout may be precipitated. Statin therapy is indicated by the risk of coronary heart disease, not by absolute cholesterol levels. Risk is estimated using national risk tables (multiplied by 1.7 to account for the increased vascular risk associated with RA and SLE, and probably ankylosing spondylitis).^8,11

The management of arthritis in the presence of established HF can be challenging. As discussed previously, TNF inhibition, glucocorticoids, and both selective and nonselective NSAIDs are associated with deterioration of HF status. These agents should therefore be avoided whenever possible. If they are introduced, patients should be advised of symptoms to recognize and report. If symptoms develop, the agent should be withdrawn: in the case of chronic glucocorticoid therapy this will require down-titration and gradual withdrawal. Each patient requires individualized care. If arthritis control is poor despite conventional therapies, TNF inhibition may be considered and cautiously introduced. It may even improve cardiac status.

The interactions between HF and arthritis are complex. With an ageing population, it is becoming more common to have patients whose lives are limited by both conditions. Teasing apart the cause and effect of the disease processes and the impact of the various therapies can be challenging. However, as the trend towards earlier disease recognition and more aggressive management of both HF and arthritis persists, hopefully the negative impact of each condition on the other will be attenuated. In both HF and arthritis there are also significant physical and psychological benefits achievable by maintaining the patient’s physical activity and independence.