Acute Rheumatic Fever

Acute rheumatic fever is a disease that causes inflammation throughout the body, especially in the joints. Now rare in developed countries, it is an important cause of heart disease in developing countries and is most common in children aged five to 15 years.


Rheumatic fever is believed to be an autoimmune disorder (in which the body’s immune system attacks its own tissues) induced by certain strains of streptococcal bacteria. Rheumatic fever almost always follows a throat infection (or rarely a skin infection - see below). The development of rheumatic fever can usually be prevented by treatment with antibiotic drugs.

Symptoms and signs

Rheumatic fever causes fever with pain, inflammation, and swelling of one or more of the larger joints. The heart valves may be scarred, leading to mitral stenosis (narrowing of the mitral valve) or mitral incompetence (leaking of the mitral valve). If the nervous system is involved, Sydenham’s chorea may occur, in this complication there are irregular, uncontrollable, jerky movements and usually some emotional upset.

Diagnosis and treatment

Rheumatic fever may be suspected whenever arthritis (joint inflammation) moves from joint to joint, but may be discovered only after development of heart failure or a heart murmur. Treatment is with penicillin drugs to eradicate streptococci; with aspirin or other salicylate drugs to control joint pain and inflammation and to minimize heart damage; and, in some cases, with corticosteroid drugs. If damage to the heart valves occurs, heart valve surgery may be required.

Read more:

Rheumatic fever - non-technical article

Acute rheumatic fever in detail - technical

Topics covered:

  • Essentials
  • Introduction
  • Epidemiology
  • Pathogenesis
  • Clinical manifestations 
  • Diagnosis
  • Treatment
  • Prognosis and follow-up
  • Recurrences
  • Prevention of acute rheumatic fever
  • Further reading


Acute rheumatic fever is an immunologically mediated multisystem disease induced by recent infection with group A streptococcus. About 5% of people have the potential to develop acute rheumatic fever after infection by a strain of streptococcus with propensity to cause the condition. Most cases (97%) occur in developing countries, particularly in sub-Saharan Africa, Pacific nations, Australasia, and the Indian subcontinent. Children aged 5 to 15 years are most commonly affected, and rheumatic heart disease remains the most common acquired heart disease of childhood in the world.

Presentation—after a latent period (1–5 weeks in most cases, but up to 6 months for presentation with chorea) the disease presents with one or more of the following major criteria: (1) carditis—most typically manifest as an apical pansystolic murmur of mitral regurgitation; (2) polyarthritis—severe, large-joint, and migratory; (3) chorea; (4) subcutaneous nodules; (5) erythema marginatum. Other minor criteria that can support the diagnosis include fever, polyarthralgia, elevated acute phase reactants or neutrophil count, prolongation of the PR interval on the ECG.

Diagnosis—in addition to the criteria described above, evidence of preceding group A streptococcal infection is required: (1) positive throat culture, or (2) elevated or rising anti-Streptolysin O or other streptococcal antibody, or (3) rapid antigen test for group A streptococcus, or (4) recent scarlet fever (contentious).

Prognosis and management—untreated acute rheumatic fever lasts for about 3 months. All patients with acute disease should be given penicillin to eradicate the group A streptococcus that precipitated the attack. Children with arthritis or severe arthralgia should be treated with nonsteroidal anti-inflammatory medication (usually salicylates). For severe carditis, many clinicians use oral prednisone or prednisolone at a dose of 40 to 60 mg/day (1–2 mg/kg per day in children), tapering after 2 or 3 weeks, but benefit is not proven. Important prognostic factors are the severity of the acute carditis and the number of recurrences: 30 to 50% of patients with a first episode of acute rheumatic fever will develop chronic rheumatic heart disease, but more than 70% of those with severe carditis at the first episode, or with recurrent episodes.

Secondary prophylaxis—every patient with acute rheumatic fever should immediately commence intramuscular benzathine penicillin G every 3 or 4 weeks, or twice daily oral penicillin V. In patients without carditis, this should continue for 5 years or until age 18, whichever comes later; with mild or healed carditis, for 10 years or until age 25, whichever is longer; those with more severe valvular disease or after valve surgery should have secondary prophylaxis for life.

Primary prophylaxis—a full course of penicillin treatment commencing within 9 days of the onset of symptomatic group A streptococcal pharyngitis will prevent the subsequent development of acute rheumatic fever in most cases, but this is not widely practised in most developing countries and would not prevent those cases of acute rheumatic fever which do not follow a sore throat.


Acute rheumatic fever is an immunologically mediated multisystem disease induced by recent infection with group A streptococcus. Most medical practitioners in industrialized countries will rarely, if ever, see a case. However, the dramatic decline in incidence of acute rheumatic fever in industrialized countries during the second half of the 20th century was not replicated in many developing countries, or among some indigenous and other populations living in poverty in industrialized countries. Moreover, acute rheumatic fever has recently returned as an important public health problem in some middle-class regions of the United States of America. Rheumatic heart disease remains the most common acquired heart disease of childhood in the world.


It was recently estimated that between 15 and 19 million people are affected by rheumatic heart disease, with approximately 280 000 new cases and 230 000 deaths occurring each year as a result. Ninety-seven per cent of acute rheumatic fever cases and deaths occur in developing countries. Although acute rheumatic fever and rheumatic heart disease are relatively common in all developing countries, they occur at particularly high rates in sub-Saharan Africa, Pacific nations, Australasia, and the Indian subcontinent.

There have been dramatic declines in incidence in recent decades in many Latin American and Asian countries with improving economic and living conditions. In most populations with high incidence, the predisposing conditions are those that promote endemicity and high levels of transmission of group A streptococci: these include overcrowded housing, poor personal and community hygiene, poor access to medical services, and, in some circumstances, widespread skin infection and scabies infestation.

Outbreaks of acute rheumatic fever occurred in middle-class areas of the United States during the 1980s and 1990s. These outbreaks arose because of the emergence of virulent strains of group A streptococci, particularly belonging to M serotypes 1, 3, and 18. By contrast, outbreaks of acute rheumatic fever have rarely, if ever, been described from developing countries; most cases appear to arise from the ongoing circulation of pathogenic group A streptococcal strains in the population.

Recurrent episodes are almost as common as primary episodes in many populations with high incidence rates of acute rheumatic fever. These may lead to accumulated cardiac valvular damage and are therefore responsible for many cases of rheumatic heart disease, yet they are almost entirely preventable using secondary prophylaxis (see later).

In many developing countries, females are affected more than males, although this gender association is stronger for rheumatic heart disease (especially mitral stenosis) than for acute rheumatic fever; this may reflect a greater tendency to recurrences among females. Any female preponderance may relate to inherited characteristics, to greater exposure to group A streptococci because of the increased involvement of girls and young women in child-rearing in most cultures, or to reduced access by females to primary and secondary prophylaxis.

The maximum incidence of acute rheumatic fever is between the ages of 5 and 15 years in all populations. Approximately 5% of cases occur in children younger than 5 years, but very rarely are children younger than 3 years affected. This age distribution parallels that of group A streptococcal pharyngitis, and supports the hypothesis that all cases of acute rheumatic fever follow this condition. However, it may be that cases do not occur in infants or very young children because of the need for maturity of the immune system (particularly of cellular immunity), or sensitization of the immune response by prior streptococcal infections. New cases occur occasionally up to age 30, but rarely beyond. Hypotheses to explain the reduced incidence in adulthood include development of non-type-specific immunity to primary group A streptococcal infections, further maturation of immune responses, or reduced sensitization by recurrent streptococcal infections.


Despite a century of research, the pathogenesis of acute rheumatic fever remains incompletely understood.

Host factors

Epidemiological evidence suggests that less than 5 to 6% of people have the potential to develop acute rheumatic fever after relevant streptococcal exposure, and that this proportion does not vary substantially between populations. Attack rates of acute rheumatic fever after untreated group A streptococcal pharyngitis vary from less than 1 to 3%. Genetic susceptibility to acute rheumatic fever was first suggested by its familial aggregation and by a greater concordance in monozygotic than in dizygotic twins. The mode of inheritance is uncertain; autosomal recessive or autosomal dominant with partial penetrance have been suggested.

The basis for genetic susceptibility is not known. Recent work suggests an association of rheumatic heart disease with certain HLA class II alleles. A B-cell alloantigen (D8/17) is expressed in a high percentage of B cells from patients with acute rheumatic fever and their family members in many populations. However, D8/17 may not predict susceptibility in all populations; studies in India suggest that different B-cell alloantigens may identify patients with acute rheumatic fever there. However, it is not yet clear whether these putative markers are involved in the pathogenesis of acute rheumatic fever.

Organism factors

The observation that outbreaks of pharyngitis due to certain serotypes of group A streptococcus resulted in high attack rates of acute rheumatic fever, whereas no cases occurred after infection with other serotypes, led to the concept of ‘rheumatogenicity’—that only some strains of group A streptococcus have the potential to cause acute rheumatic fever. M serotypes 1, 3, 5, 6, 14, 18, 19, 24, 27, and 29 were most frequently implicated in studies predominantly from the United States of America. However, recent studies from regions with high endemicity of group A streptococcal infections have not found consistent M serotype, or emm genotype, associations with acute rheumatic fever. There may be substantial genetic diversity among strains belonging to a particular M serotype, and not all strains of ‘rheumatogenic serotypes’ appear to cause acute rheumatic fever. Therefore, rheumatogenicity may be strain specific rather than serotype specific; i.e. any group A streptococcus may acquire the potential to cause acute rheumatic fever.

The pathogenic factor(s) are not known. Parts of the organism have immunological cross-reactivity with human tissue; there is close homology between regions of the M protein and human myosin, tropomyosin, keratin, actin, laminin, vimentin, and N-acetylglucosamine. Other components of group A streptococci, including the hyaluronic acid capsule, the cell-wall associated group-specific carbohydrate, and the cell membrane, cross-react with a variety of human tissues damaged in acute rheumatic fever, including components of heart muscle and valves, joints, and brain. Acute rheumatic fever-associated strains of group A streptococcus also tend to be heavily encapsulated with hyaluronic acid, and not to express opacity factor. Group A streptococci possess components which act as superantigens, selectively stimulating subsets of T cells without the need for antigen presentation. Their role in acute rheumatic fever pathogenesis is not yet clear.

Site of infection

Although it is widely accepted that acute rheumatic fever may result from group A streptococcal infection of the upper respiratory tract, but not of the skin, there is some evidence that this may not always be the case. Upper respiratory tract infection certainly accounts for most, if not all, episodes of acute rheumatic fever in countries with a temperate climate. However, in tropical countries where streptococcal impetigo is highly endemic but group A streptococcal pharyngitis less common, it may be that skin infection accounts for many cases of acute rheumatic fever, either de novo or after subsequent throat infection. Determining whether group A streptococcal skin infection may have a role in pathogenesis of acute rheumatic fever would have enormous public health implications, as it may redirect present approaches to primary prevention (see later).

The immune response

Molecular mimicry between group A streptococcal epitopes and human tissue is the basis for the autoimmune response that leads to rheumatic fever. It is thought that epitopes in cardiac myosin, normally sequestered from the immune response, are exposed by normal cardiac cell turnover. This leads to sensitization of T cells, which may then be recalled following subsequent exposure to group A streptococci. However, myosin cross-reactivity with M protein does not explain the valvular damage that is the hallmark of rheumatic carditis. The link may be laminin, another α-helical coiled-coil protein like myosin and M protein, which is found in cardiac endothelium and is recognized by anti-myosin, anti-M protein T cells. Moreover, antibodies to cardiac valve tissue cross-react with the N-acetylglucosamine of group A streptococcal carbohydrate, and there is some evidence that these antibodies may be responsible for valvular damage. Overall, it is not entirely clear if the initial damage in rheumatic fever is primarily due to cellular or humoral immunity, but it does appear that ongoing damage is mainly due to T cell and macrophage infiltration.

Clinical manifestations

There is always a latent period between group A streptococcal infection and the development of acute rheumatic fever. This varies from 1 to 5 weeks in most cases (usually c.3 weeks), but may be shorter in recurrences. Chorea may occur up to 6 months after the precipitating streptococcal infection. The preceding infection is asymptomatic in about two-thirds of cases.

The tissues most commonly affected are the heart, joints, and brain. Although the symptoms due to each can be disabling in the short term, only cardiac damage may be permanent and progressive. Therefore, the focus in controlling or treating acute rheumatic fever is always to prevent the development of rheumatic heart disease.

The frequency with which the various clinical manifestations have occurred in recent descriptions of acute rheumatic fever is listed in Table 1.

Table 1 Frequency of clinical manifestations in acute rheumatic fever
Manifestation Proportion of patients with manifestation (%)
Choreaa absent Choreaa present
Carditis 40–60 20–30
Polyarthritis 50–75 <10
Erythema marginatum 1–10 0–1
Subcutaneous nodules 1–10 0–1
Fever >37.5 °C >90 10–25
Arthralgia <10–20 <5
Elevated acute phase reactants >90 10–25
Prolonged PR interval 30–50 5–10

a Chorea is present in <10% to >30% of patients with acute rheumatic fever, depending on the population.


Although inflammation in acute rheumatic fever may affect the pericardium (causing pericardial rubs and occasionally pleuritic chest pain) or the myocardium (sometimes causing cardiac failure, and evident on biopsy with pathognomonic Aschoff bodies), endocardial inflammation is the most important cause of cardiac damage. If either acute cardiac failure or chronic cardiac disease occurs, it is almost always due to damage to the cardiac valves.

A murmur is the most common evidence of acute valvular disease, usually the apical pansystolic murmur of mitral regurgitation, with or without a low-pitched mid-diastolic (Carey–Coombs) murmur. Occasionally an aortic regurgitant murmur may be heard, mainly in older adolescents or young adults. Murmurs of tricuspid or pulmonary regurgitation are rare and are usually secondary to increased pulmonary venous pressures resulting from mitral regurgitation or stenosis. Sinus tachycardia or gallop rhythms may also be present in acute carditis.

Valves affected by rheumatic carditis may have a characteristic appearance or pattern of regurgitation on Doppler echocardiography (when interpreted by experienced technicians), which may be found even in the absence of a cardiac murmur. This may be useful for diagnosis when other clinical manifestations are not definitive. However, echocardiographic criteria have not yet been standardized, and it is difficult to distinguish acute carditis from previous rheumatic valve damage.

Mitral or aortic stenosis may develop as later complications of severe and/or recurrent acute carditis due to scarring and contraction following the acute inflammatory process. Rarely, mitral stenosis may occur in young children with acute rheumatic fever—so-called ‘juvenile mitral stenosis’—the reasons for the development of this condition are not clear.

Damage to the electrical conduction pathways may result in prolongation of the PR interval on electrocardiography. Although a subset of healthy people may have this finding, the presence of a prolonged PR interval that resolves over the ensuing few days to weeks may be a useful diagnostic feature in cases where the clinical manifestations are not clear. Occasionally, in the acute phase, second- or third-degree heart block or a nodal rhythm may be present.


The characteristic joint manifestation of acute rheumatic fever is severe, large-joint, migratory polyarthritis. The knees, ankles, wrists, and elbows are most commonly involved; only rarely, and usually only when the patient is untreated for several days, are the hips or small joints of the hands or feet inflamed. One joint characteristically becomes exquisitely painful and inflamed as another is waning. Most patients have only one or two joints affected at any one time, and each joint may be involved for just a few hours or up to 1 or 2 days. The arthritis is so responsive to nonsteroidal anti-inflammatory medication (NSAIDs) that its persistence more than 1 or 2 days after commencing high-dose aspirin should lead one to consider alternative diagnoses.

Arthritis of a single large joint is increasingly described in acute rheumatic fever from regions with high rates of disease. This is sometimes, but not always, due to early administration of anti-inflammatory medication, before the typical migratory pattern has emerged. Other causes of mono-arthritis, including septic arthritis, should first be excluded before a diagnosis of acute rheumatic fever is entertained. Arthralgia (joint pain without objective evidence of inflammation) is usually migratory and affects large joints, and like the arthritis of acute rheumatic fever is very responsive to NSAIDs.

Sydenham’s chorea

In 1686 the English physician Thomas Sydenham described rheumatic chorea, initially naming it ‘St Vitus’ dance’. It is the most intriguing manifestation of acute rheumatic fever, particularly as it commonly occurs in the absence of other manifestations, usually follows a prolonged latent period (up to 6 months) after the precipitating group A streptococcal infection, and occurs most commonly in females (and almost never in postpubertal males). The rapid, jerky, involuntary movements affect predominantly the upper limbs and face, may be asymmetrical, and may be sufficiently severe to render the patient unable to eat, drink, walk, or perform other activities of daily living. Mild chorea can sometimes be detected by having the patient join palms above the head to reveal occasional twitches of the arms or the head. Typical signs include the ‘milkmaid’s grip’ (rhythmic squeezing when the patient grasps the examiner’s fingers), spooning of extended hands (caused by flexion of the wrists and extension of the fingers), darting of the protruded tongue, and the ‘pronator sign’ (the arms and palms turn outwards when held above the head). As with other forms of chorea, the disorder usually becomes more evident with anxiety or purposeful movements (such as drinking or writing). Movements may appear semi-purposeful, and symptoms subside during sleep. Sydenham’s chorea is often associated with excessive emotional lability or personality changes, which may precede the abnormal movements.

Most patients can be reassured that Sydenham’s chorea will resolve completely and leave no long-lasting effects, usually within 6 weeks and almost always within 6 months, but rarely lasting up to 3 years.

Subcutaneous nodules and erythema marginatum

Both of these manifestations are found in less than 2% of patients with acute rheumatic fever, although they were described in up to 10 to 20% of patients in earlier studies from the United States of America and the United Kingdom. Subcutaneous nodules are firm, painless lumps, usually between 0.5 and 2 cm in diameter, commonly found in crops of three or more, and usually appear 2 to 3 weeks after the onset of acute rheumatic fever. They occur mainly over extensor surfaces or bony protuberances, particularly the hands, feet, occiput, and back. The nodules are similar, though often smaller, to those found in rheumatoid arthritis, and are most likely to be associated with severe carditis. Nodules usually last from a few days to 2 or 3 weeks.

The characteristic rash, erythema marginatum, appears as a light pink macule that spreads outwards with a serpiginous, well-demarcated edge, while the central portion clears. It appears, disappears, or moves before the observer’s eyes. Multiple areas are often involved, usually over the trunk, occasionally over the proximal portions of the limbs, but rarely, if ever, the face. It usually appears together with the other initial symptoms of acute rheumatic fever, but may recur intermittently for weeks or even months. This does not indicate ongoing rheumatic inflammation, and patients can be reassured that the rash will eventually disappear without complications.


With the exception of those with pure chorea, 90% of patients will have a temperature at presentation higher than 37.5 °C. Although it has been reported that the temperature usually exceeds 39 °C, others have found only 25% of confirmed cases with fever to that level. Any temperature above 37.5 °C should be considered a minor manifestation. As with arthritis, fever is very sensitive to NSAIDs, usually resolving completely within 1 or 2 days of commencing high-dose salicylates.

Elevated acute phase reactants

Almost all patients, except those with pure chorea, have a dramatically elevated erythrocyte sedimentation rate or serum C-reactive protein. There appears little difference between these measurements in their diagnostic usefulness. The C-reactive protein may return to normal more rapidly than the sedimentation rate when rheumatic activity subsides. Mild to moderate peripheral leucocytosis is common, although this is a less sensitive marker of rheumatic inflammation.

Other features

Severe central abdominal pain is found at presentation in a small proportion of patients. It may be associated with other features of acute rheumatic fever; if not, these features usually appear within 1 or 2 days. The pain responds quickly to NSAIDs. Epistaxis was reported frequently in historical accounts of acute rheumatic fever, but does not feature prominently in recent descriptions. Pulmonary infiltrates may be found in patients with acute carditis; this has been labelled ‘rheumatic pneumonia’ although it is not clear whether the infiltrates represent rheumatic inflammation or another process. There may be microscopic haematuria, pyuria, or proteinuria; also mild elevations of liver transaminases: these are nonspecific and not usually severe.

Associated poststreptococcal syndromes

Poststreptococcal reactive arthritis has been differentiated from rheumatic fever by some authors because it has a shorter incubation period after streptococcal infection, sometimes follows non-group A β-haemolytic streptococcal infection, may have a different pattern of arthritis (including small joint involvement), and is less responsive to NSAIDs. Because of the lack of cardiac involvement, these patients are said not to require secondary prophylaxis. However, descriptions of patients who have subsequently developed carditis have led other authors to question the distinction between poststreptococcal reactive arthritis and rheumatic fever. If poststreptococcal reactive arthritis is diagnosed, secondary prophylaxis should be prescribed for at least 1 year and discontinued if there is no evidence of carditis. In populations with high incidence rates of acute rheumatic fever, it may be prudent to treat all cases of possible poststreptococcal reactive arthritis as acute rheumatic fever.

The frequent finding of emotional lability, motor hyperactivity, and occasional obsessive–compulsive symptoms in patients with Sydenham’s chorea led to the observation that group A streptococcal infections may precipitate or exacerbate other disorders of the basal ganglia. These include tic disorders, Tourette’s syndrome, and obsessive–compulsive disorder, and the term PANDAS (paediatric autoimmune neuropsychiatric disorders associated with streptococcal infections) has been coined. The existence of PANDAS is not universally accepted. Patients with PANDAS are said not to be at risk of developing carditis. There is evidence that these patients, and some children with autism, have high proportions of circulating B cells expressing D8/17 antigen, which is a proposed marker of rheumatic fever susceptibility. It is not yet clear whether these syndromes are linked with acute rheumatic fever.


Because of the diversity of symptoms and signs, and the nonspecific nature of most of them, Dr T Duckett Jones developed a set of criteria to aid in the diagnosis of acute rheumatic fever in 1944. The Jones criteria have subsequently been revised and updated a number of times to improve their positive and negative predictive values, most recently in 1992. These 1992 criteria are to be used only for the diagnosis of the initial episode of acute rheumatic fever. In response to uncertainty about how to use the Jones criteria for the diagnosis of recurrent episodes, a World Health Organization (WHO) expert committee published the 2002–2003 WHO Criteria, which are now the standard for acute rheumatic fever diagnosis (Table 2).

Table 2 2002–03 World Health Organization criteria for the diagnosis of rheumatic fever and rheumatic heart disease (based on the 1992 revised Jones criteria)
Diagnostic categories Criteria
Primary episode of rheumatic fevera Two major or one major and two minor manifestations plus evidence of preceding group A streptococcal infection
Recurrent attack of rheumatic fever in a patient without established rheumatic heart disease Two major or one major and two minor manifestations plus evidence of preceding group A streptococcal infection
Recurrent attack of rheumatic fever in a patient with established rheumatic heart diseaseb Two minor manifestations plus evidence of preceding group A streptococcal infectionc
  • Rheumatic chorea
  • Insidious onset rheumatic carditisb
Other major manifestations or evidence of group A streptococcal infection not required
Chronic valve lesions of rheumatic heart disease (patients presenting for the first time with pure mitral stenosis or mixed mitral valve disease and/or aortic valve disease)d Do not require any other criteria to be diagnosed as having rheumatic heart disease
Major manifestations Carditis
Erythema marginatum
Subcutaneous nodules
Minor manifestations Clinical: fever, polyarthralgia
Laboratory: elevated ESR or leucocyte counte
Electrocardiogram: prolonged PR interval
Supporting evidence of a preceding streptococcal infection within the last 45 days
  • Elevated or rising anti-Streptolysin O or other streptococcal antibody, or
  • A positive throat culture, or
  • Rapid antigen test for group A streptococcus, or
  • Recent scarlet fevere

a Patients may present with polyarthritis (or with only polyarthralgia or monoarthritis) and with several (three or more) other minor manifestations, together with evidence of recent group A streptococcal infection. Some of these cases may later turn out to be rheumatic fever. It is prudent to consider them as cases of ‘probable rheumatic fever’ (once other diagnoses are excluded) and advise regular secondary prophylaxis. Such patients require close follow up and regular examination of the heart. This cautious approach is particularly suitable for patients in vulnerable age groups in high incidence settings.

b Infective endocarditis should be excluded.

c Some patients with recurrent attacks may not fulfil these criteria.

d Congenital heart disease should be excluded

e 1992 Revised Jones criteria do not include elevated leucocyte count as a laboratory minor manifestation (but do include elevated C-reactive protein), and do not include recent scarlet fever as supporting evidence of a recent streptococcal infection.

WHO Expert Consultation on Rheumatic Fever and Rheumatic Heart Disease, Rheumatic fever and rheumatic heart disease: report of a WHO Expert Consultation (WHO Technical Report Series, 923), p23. Geneva; World Health Organization. 

The manifestations are divided into major, those which are most predictive of acute rheumatic fever, and minor, those which are commonly found in acute rheumatic fever but are less specific. The diagnosis of an initial episode requires the presence of either two major, or one major and two minor criteria, plus the demonstration of a current or recent group A streptococcal infection. Evidence of group A streptococcal infection is not required for chorea, where the onset may be delayed up to 6 months after streptococcal infection, and late-onset carditis, when low-grade inflammation may persist for prolonged periods after the precipitating infection. Recurrences can be diagnosed with less stringent criteria.

Proof of a recent group A streptococcal infection can include demonstrating the organism in the upper respiratory tract, either by culture or rapid antigen techniques. However, most children with acute rheumatic fever no longer have a group A streptococcus detectable by these methods, and up to 15 to 25% of normal children in temperate climate countries may carry the organism in their throats. Serological techniques are therefore most commonly used, particularly the antistreptolysin O, anti-DNase B, or antihyaluronidase titres. One of any two of these tests will be positive in well over 90% of recent streptococcal infections. Their usefulness is increased by performing more than one serological test, or by demonstrating rising titres in paired sera. Serology is of limited value in regions with high prevalence rates of streptococcal impetigo, where children may have positive antistreptococcal titres most of the time. There is therefore a need for a better diagnostic test of recent streptococcal infection, or an objective diagnostic test for acute rheumatic fever itself.

The most common clinical presentation, that of a child with fever and polyarthritis, raises multiple differential diagnoses that will vary by region. Table 3 lists some alternative diagnostic possibilities for the three most common major manifestations.

Table 3 Differential diagnoses of common major presentations of acute rheumatic fever
Polyarthritis and fever Carditis Chorea
Differential diagnoses Septic arthritis (including gonococcal) Innocent murmur SLE
Connective tissue and other autoimmune diseasea Mitral valve prolapse Drug intoxication
Viral arthropathyb Congenital heart disease Wilson’s disease
Reactive arthropathyb Infective endocarditis Tic disorderc
Lyme disease Hypertrophic cardiomyopathy Choreoathetoid cerebral palsy
Sickle cell anaemia Myocarditis—viral or idiopathic Encephalitis
Infective endocarditis Pericarditis—viral or idiopathic Familial chorea (including Huntington’s)
Leukaemia or lymphoma   Intracranial tumour
Gout and pseudogout   Lyme disease

SLE, systemic lupus erythematosus.

a Includes rheumatoid arthritis, juvenile chronic arthritis, inflammatory bowel disease, systemic lupus erythematosus, systemic vasculitis, sarcoidosis, among others.

b Mycoplasma, cytomegalovirus, Epstein–Barr virus, parvovirus, hepatitis, rubella vaccination, and yersinia and other gastrointestinal pathogens.

c Possibly including PANDAS (pediatric autoimmune neuropsychiatric disorder associated with streptococcal infection).

d Includes oral contraceptives, pregnancy (chorea gravidarum), hyperthyroidism, hypoparathyroidism.

Reprinted from The Lancet, Vol. 366, Carapetis JR, McDonald M, Wilson NJ, Acute rheumatic fever, pp155–68.


If untreated, acute rheumatic fever lasts on average for 3 months. Except in the case of life-threatening acute carditis, there is no evidence that presently available treatments alter the outcome. Most treatments are designed to provide symptomatic relief or are based on theoretical (but unproven) approaches to attenuating the long-term damage.

If practical, all patients with acute rheumatic fever should be admitted to hospital to confirm the diagnosis, perform baseline investigations to ascertain the status of the heart, provide adequate treatment for the acute phase, commence secondary prophylaxis, allow communication of details to personnel responsible for long-term follow-up of the patient, and begin education of the patient and family. The mainstays of treatment are bed rest, penicillin, and salicylates.

Bed rest

Previous recommendations that children with acute rheumatic fever be rested in bed until all signs of active inflammation abated were probably more extreme than is necessary. Once symptoms of arthritis have subsided and any cardiac failure is controlled, the child may begin gentle mobilization, which may be increased as tolerated. There is no evidence that bed rest beyond the period where mobilization leads to exacerbation of pain or cardiac failure has any long-term benefit.


All patients with acute rheumatic fever should be given penicillin to eradicate the group A streptococcus that precipitated the attack. This is based on an early finding that, in some cases, prolonged group A streptococcal infection led to more severe acute rheumatic fever. Although in most cases the precipitating organism cannot be cultured, a treatment course of penicillin is prudent in case the strain remains present in low numbers, and to prevent its transmission to other contacts. As the aim is eradication of group A streptococcal infection, penicillin may be administered either as a single intramuscular injection of benzathine penicillin G at a dose of 1.2 million units (600 000 U for patients <30 kg) into the gluteal or quadriceps muscles, or as a 10-day course of oral phenoxymethyl penicillin (V) at a dose of 500 mg (adolescents and adults) or 250 mg (children) given either two or three times daily. In the case of penicillin allergy, the present recommendation is to use oral erythromycin at 20 to 40 mg/kg per day given two to four times daily for 10 days, although in some regions levels of erythromycin resistance among group A streptococci are increasing.


Children with arthritis or severe arthralgia should be treated with NSAIDs; salicylates have been most widely used. Aspirin at a dose of 80 to 100 mg/kg per day (4–8 g/day in adults) usually results in defervescence and resolution of arthritis and arthralgia within 1 to 2 days. Sometimes these doses lead to nausea or vomiting, which can be minimized by increasing from lower starting doses. After a few days or up to 2 weeks, when the initial symptoms are abating, the dose can be reduced to 60 to 70 mg/kg per day for the remaining 2 to 4 weeks. Arthritis or arthralgia may return up to 2 to 3 weeks after discontinuation of therapy; this is usually a brief and mild recrudescence, often associated with increased erythrocyte sedimentation rate or C-reactive protein, and can be managed either with rest and reassurance or a short course of lower-dose NSAIDs.

When the diagnosis is uncertain, salicylates should be withheld for a day or two to look for the development of characteristic migratory polyarthritis. In such cases, paracetamol or codeine can be used to control pain until the diagnosis is confirmed. There is no evidence that salicylates reduce the severity of acute carditis or the risk of chronic cardiac valve damage.


For many years, corticosteroids have been used in acute rheumatic fever, particularly for patients with severe carditis. Two meta-analyses have found no evidence that they reduce the risk of long-term valve damage. However, the studies included in these meta-analyses were all conducted more than 40 years ago and used corticosteroid medications not in common usage today. Many clinicians continue to use oral prednisone or prednisolone at a dose of 40 to 60 mg/day (1–2 mg/kg per day in children), tapering after 2 or 3 weeks, in the belief that this might reduce the severity of acute carditis.

Treatment of cardiac failure

There is no doubting the need to treat cardiac failure. Diuretics, angiotensin-converting enzyme (ACE) inhibitors (especially in aortic regurgitation), and fluid restriction are most commonly employed. Digoxin is usually restricted to cases where atrial fibrillation coexists with cardiac failure, often found in older patients with established mitral stenosis.

If medical therapy fails, cardiac surgery should be considered, even during the acute phase. In populations where fulminant acute carditis is relatively common (e.g. South Africa), mitral valve repair or replacement can be life saving and surgeons have developed techniques for undertaking these procedures despite friable, acutely inflamed valvular and perivalvular tissues. In recent years, there has been a greater tendency to undertake valve repair rather than replacement, or to use homografts or xenografts rather than mechanical prostheses. This is to avoid high rates of thromboembolic complications associated with mechanical prostheses, particularly in populations where compliance with anticoagulation chemotherapy is suboptimal and there are difficulties in monitoring coagulation indices.

Treatment of chorea

Sydenham’s chorea always resolves, and in most cases there is no need for medical treatment. However, medications may reduce abnormal movements in moderate or severe chorea. Carbamazepine or sodium valproate are recommended as first-line treatment, haloperidol less commonly because of its side-effect profile. Other medications sometimes employed include pimozide, chlorpromazine, or benzodiazepines. All of these medications should be used sparingly and only for limited periods, and the tendency to try multiple medications should be avoided.

Salicylates and steroids have no role in treatment of chorea. Psychotherapeutic interventions have little role in the short to medium term, and may increase the stigma of this self-limited organic disease. However, behavioural therapy should be considered if longer-term behavioural abnormalities persist (e.g. emotional lability, obsessive–compulsive traits).

Newer therapies

Because of the autoimmune nature of acute rheumatic fever, immunomodulatory therapies have been tried. Intravenous immune globulin (IVIG) has been given in some small trials. One study showed no apparent benefit on rate of improvement of clinical, laboratory, or echocardiographic parameters of acute carditis, but another suggested that it may accelerate recovery from chorea. Other therapies have yet to be formally assessed.

Prognosis and follow-up

The most important prognostic factors are the severity of the acute carditis and the number of recurrences. Overall, approximately 30 to 50% of patients with a first episode of acute rheumatic fever will develop chronic rheumatic heart disease. This increases to more than 70% in patients with severe carditis at the first episode, or in those who have had at least one recurrence.

Any patient with acute rheumatic fever requires long-term follow-up. Follow-up assessments should focus on cardiac status, adherence to secondary prophylaxis, early treatment of group A streptococcal pharyngitis, and prevention of streptococcal pyoderma (including hygiene and treatment or prevention of scabies infestation). Patients with evidence of cardiac valve damage should be assessed regularly by specialist physicians and considered for cardiac surgery before substantial left ventricular dysfunction occurs. Vasoactive drugs, particularly ACE inhibitors, may delay the need for operation in asymptomatic patients with chronic aortic regurgitation. Regular echocardiography may be useful to follow the progress of rheumatic heart disease, especially in populations where follow-up may be irregular or in whom communication or cultural differences make clinical assessment difficult.


About 75% of all recurrences occur within 2 years of an episode of acute rheumatic fever. The reasons for this are not known, but are thought to relate to a time-dependent sensitization of the immune response. The clinical features of recurrences tend to mimic those present at the initial episode, particularly in the case of chorea. However, this rule is not absolute, and the risk of developing other manifestations, particularly carditis, increases with each recurrence. The practical implication of this is that the absence of carditis at the first episode does not help to identify patients who may not need secondary prophylaxis.

Prevention of acute rheumatic fever

Secondary prophylaxis

Every patient with acute rheumatic fever should immediately commence secondary prophylaxis: long-term, regular antibiotics to prevent primary group A streptococcal infections. This strategy is proven to reduce the incidence of recurrences and the risk of developing chronic rheumatic heart disease.

The optimal regimen is 1.2 million units of intramuscular benzathine penicillin G every 3 or 4 weeks, and this is commonly given in populations with high incidences of acute rheumatic fever and programmes in place to support the regimen. Higher doses (1.8 or 2.4 million U) given every 4 weeks may have similar effect, but further evidence is needed before such regimens can be recommended routinely. An alternative strategy is to use oral penicillin V at a dose of 250 mg twice daily; this is almost as effective as using benzathine penicillin G, but adherence is usually less reliable.

For patients proven to be allergic to penicillin, the present recommendation is to use oral erythromycin at a dose of 250 mg twice daily. Recent trials have shown newer oral cephalosporins to be effective at eliminating upper respiratory tract carriage of group A streptococci. However, none of these antibiotics have been evaluated for their ability to prevent acute rheumatic fever.

The duration of secondary prophylaxis is dictated by the reducing risk of recurrence with increasing age, with time since the last episode, and the possible consequences of recurrences. In patients without carditis, secondary prophylaxis should continue for 5 years following the most recent episode or until age 18 years, whichever comes last. In patients with mild or healed carditis, prophylaxis should be continued for 10 years following the most recent episode or until age 25 years, whichever is longer. Patients with more severe valvular disease or those who have undergone valve surgery should have secondary prophylaxis for life.

Primary prophylaxis

A full course of penicillin treatment commencing within 9 days of the onset of symptomatic group A streptococcal pharyngitis will prevent the subsequent development of acute rheumatic fever in most cases. After the diagnosis has been confirmed by a throat culture or rapid antigen diagnostic test, the treatment of choice is penicillin, administered either as a single intramuscular injection of benzathine penicillin G (600 000 U for children who weigh <30 kg, or 1.2 million U for larger children and adults) or as a full 10 days of oral (phenoxymethyl) penicillin V (250 mg for children or 500 mg for adults given two to three times daily). The importance of completion of the 10-day course, even if symptoms abate quickly, should be stressed to patients and parents. Shorter courses of oral penicillin treatment are associated with higher risks of acute rheumatic fever. There has never been a clinical isolate of group A streptococcus that is resistant to penicillin; therefore, the use of other antibiotics for primary prophylaxis should be restricted to patients who are allergic to penicillin.

In the case of penicillin allergy, a 10-day course of an oral macrolide such as erythromycin is recommended. First-generation oral cephalosporins may also be considered. However, these agents have not been evaluated in populations with high incidences of acute rheumatic fever. Shorter courses (e.g. 5 days) of some later-generation oral cephalosporins and azolides appear to be effective in eradicating carriage, but because of their expense and broader spectrum of antimicrobial activity they should be considered as second-line agents.

It is not possible to predict which episodes of group A streptococcal pharyngitis will precipitate acute rheumatic fever, so this treatment must be offered in all cases to be effective. Unlike prevention of recurrent episodes, which is virtually complete using secondary prophylaxis, penicillin treatment of streptococcal pharyngitis will at best prevent only the one-third or so of cases of acute rheumatic fever that follow a sore throat. However, this important intervention may arrest the spread of pathogenic group A streptococci in the community. Penicillin treatment of group A streptococcal pharyngitis should begin as early as possible in patients with a history of acute rheumatic fever, should they not be taking secondary prophylaxis, but even then may not prevent a recurrence, hence the need for secondary prophylaxis.

In recent years the use of primary prophylaxis has been questioned in some industrialized countries where acute rheumatic fever is now rare. It is argued that the strategy prevents few cases of acute rheumatic fever but contributes to overuse of antibiotics. Similar arguments were raised in the United States of America during the 1970s, but faded somewhat with the resurgence of acute rheumatic fever in that country during the 1980s. Any country considering abandoning primary prophylaxis should first have in place effective surveillance to detect changes in the epidemiology of primary group A streptococcal infections and the appearance of cases of acute rheumatic fever.

Primary prophylaxis is unsuccessful in many developing countries. It requires trained health workers, microbiology laboratories, transportation and communication infrastructure, the availability of penicillin, and a population likely to seek and adhere to treatment for sore throats. In some high-risk populations, all patients with sore throats receive intramuscular benzathine penicillin G without further attempts at diagnosis; the cost-effectiveness of this strategy has not been fully determined. Clinical algorithms to identify patients with group A streptococcal pharyngitis without resorting to laboratory tests have not been validated sufficiently for them to be recommended universally. Even if primary prophylaxis were to be instituted effectively in developing countries, acute rheumatic fever would not disappear, as most cases do not follow a sore throat.

Other methods of primary prevention are clearly needed in developing countries. Improved living standards and access to primary health care seem to be years or decades away in many places. Although streptococcal skin infections may be linked to acute rheumatic fever pathogenesis, there are no trials of impetigo control programmes to prevent acute rheumatic fever. There is a current focus on attempts to develop a group A streptococcal vaccine. Clinical trials of one prospective vaccine are under way, and others are imminent, but the process will take many years, and recent experience suggests that new vaccines are often beyond the financial reach of most developing countries. For the foreseeable future at least, acute rheumatic fever prevention in many developing countries will depend on improving adherence to secondary prophylaxis and developing new strategies for primary prophylaxis.

Further reading


Anonymous (1995). Strategy for controlling rheumatic fever/rheumatic heart disease, with emphasis on primary prevention: memorandum from a joint WHO/ISFC meeting. Bull World Health Organ, 73, 583–7.
Bach JF, et al. (1996). 10-year educational programme aimed at rheumatic fever in two French Caribbean islands. Lancet, 347, 644–8.
Batzloff MR, et al. (2003). Protection against group A streptococcus by immunization with J8-diptheria toxoid: contribution of J8-and diptheria toxoid-specific antibodies to protection. J Infect Dis, 187, 1598–1608.
Bisno AL (1991). Group A streptococcal infections and acute rheumatic fever. N Engl J Med, 325, 783–93.
Bisno AL, et al. (2005). Prospects for a group A streptococcal vaccine: rationale, feasibility, and obstacles—report of a National Institute of Allergy and Infectious Diseases workshop. Clin Infect Dis, 41, 1150–6.
Carapetis JR, McDonald M, Wilson NJ (2005). Acute rheumatic fever. Lancet, 366, 155–68.
Carapetis JR, et al. (2005). The global burden of group A streptococcal diseases. Lancet Infect Dis, 5, 685–94.
Cilliers AM (2006). Rheumatic fever and its management. BMJ, 333, 1153–6.
Cilliers AM, Manyemba J, Saloojee H (2003). Anti-inflammatory treatment for carditis in acute rheumatic fever. Cochrane Database Syst Rev, CD003176.
Cunningham MW (2000). Pathogenesis of group A streptococcal infections. Clin Microbiol Rev, 13, 470–511.
Cunningham MW (2004). T cell mimicry in inflammatory heart disease. Mol Immunol, 40, 1121–7.
Hu MC, et al. (2002). Immunogenicity of a 26-valent group A streptococcal vaccine. Infect Immun, 70, 2171–7.
Kaplan EL (1993). T. Duckett Jones Memorial Lecture. Global assessment of rheumatic fever and rheumatic heart disease at the close of the century. Influences and dynamics of populations and pathogens: a failure to realize prevention? Circulation, 88, 1964–72.
Lennon D (2000). Rheumatic fever: a preventable disease? The New Zealand experience. In: Martin DR, Tagg JR (eds) Streptococci and streptococcal diseases: entering the new millennium, pp. 503–512. Securacopy, Auckland.
Lennon D (2004). Acute rheumatic fever in children: recognition and treatment. Paediatr Drugs, 6, 363–73.
Martin DR, et al. (1994). Acute rheumatic fever in Auckland, New Zealand: spectrum of associated group A streptococci different from expected. Pediatr Infect Dis J, 13, 264–9.
McDonald M, Currie BJ, Carapetis JR (2004). Acute rheumatic fever: a chink in the chain that links the heart to the throat? Lancet Infect Dis, 4, 240–5.
McDonald M, et al. (2005). Preventing recurrent rheumatic fever: the role of register-based programs. Heart, 91, 1131–3.
National Heart Foundation of Australia (RF/RHD guideline development working group) and the Cardiac Society of Australia and New Zealand (2006). Diagnosis and management of acute rheumatic fever and rheumatic heart disease in Australia—an evidence-based review. National Heart Foundation of Australia, Melbourne.
Quinn RW (1989). Comprehensive review of morbidity and mortality trends for rheumatic fever, streptococcal disease, and scarlet fever: the decline of rheumatic fever. Rev Infect Dis, 11, 928–53. 
Robertson KA, Volmink JA, Mayosi BM (2005). Antibiotics for the primary prevention of rheumatic fever: a meta-analysis. BMC Cardiovasc Disord, 5, 11.
Special Writing Group of the Committee on Rheumatic Fever Endocarditis and Kawasaki Disease of the Council on Cardiovascular Disease in the Young of the American Heart Association (1992). Guidelines for the diagnosis of rheumatic fever. Jones Criteria, 1992 update. JAMA, 268, 2069–73.
Steer AC, et al. (2002). Systematic review of rheumatic heart disease prevalence in children in developing countries: the role of environmental factors. J Paediatr Child Health, 38, 229–34.
Stollerman GH (2001). Rheumatic fever in the 21st century. Clin Infect Dis, 33, 806–14.
Tubridy-Clark M, Carapetis JR (2007). Subclinical carditis in rheumatic fever: A systematic review. Int J Cardiol, 119, 54–8.
Veasy LG, Tani LY, Hill HR (1994). Persistence of acute rheumatic fever in the intermountain area of the United States. J Pediatr, 124, 9–16.
Wannamaker LW (1973). The chain that links the heart to the throat. Circulation, 48, 9–18.
WHO Expert Consultation on Rheumatic Fever and Rheumatic Heart Disease (2004). Rheumatic fever and rheumatic heart disease: report of a WHO Expert Consultation, Geneva, 29 October–1 November 2001. WHO Technical Report Series 923, World Health Organization, Geneva.