Case Based Pediatrics For Medical Students and Residents
Department of Pediatrics, University of Hawaii John A. Burns School of Medicine
Chapter VII.4. Rheumatic Fever
David K. Kurahara, MD
November 2002

Return to Table of Contents

An 11 year old Polynesian male presents with fever up to 39 degrees (102 degrees F), joint pain and swelling, along with shortness of breath. The fever comes and goes at random times of the day. The symptoms have been present now for 4 days. Two days ago, his right knee was painful and swollen, but today it has improved. The joints involved today include the right ankle and left knee. They are quite tender, painful and also swollen. The shortness of breath occurs with walking, but he is now unable to walk because of the joint pain. He also has some shortness of breath with lying down flat when he is trying to sleep.

Exam: VS T 38.2, P 160, RR 32, BP 100/60, oxygen saturation 94% in room air. He is tired appearing with tachypnea and tachycardia. HEENT: Enlarged, erythematosus tonsils with exudates. Lungs are clear but with tachypnea. Heart sounds are tachycardic with a holosystolic murmur 3/6 heard at apex with radiation to axilla. No gallops are heard. His PMI is prominent (size of silver dollar) at the 7th intercostal space in the mid-axillary line. His abdomen is soft with normoactive bowel sounds. His liver edge is 6 to 7 cm below the RCM. His left knee is swollen and extremely tender with warmth. He has difficulty with range of motion but can flex his knee 30 degrees passively. His right ankle is very swollen and warm. He has limited subtalar motion. Both his knee and ankle are very tender even to touch. Neuro: No abnormal movements of arms, hands, or tongue are noted. He is unable to walk due to pain.

Clinical course: The child is admitted to the hospital. Initial laboratory work includes a erythrocyte sedimentation rate of 110, a CRP of 9.5, and a chest X-ray with cardiomegaly present. EKG reveals a prolonged PR interval. ASO titer is 754 and streptozyme is 1:600. The diagnosis of acute rheumatic fever (ARF) is made and he is initially started on salicylate therapy at 75 mg/kg/day, and his arthritis improves dramatically. However, the next day an echocardiogram confirms severe mitral insufficiency. Due to the significant cardiac disease with elements of congestive heart failure he is switched to corticosteroids and improves. His heart size decreases over the next 2 weeks, and when it normalizes he is switched back to salicylates for a total treatment duration of 8 weeks. He does have a persistent murmur after this time however. He is started on intramuscular benzathine penicillin, which is given every 4 weeks for streptococcal prophylaxis.

The terms of Acute Rheumatic Fever and Rheumatic Heart Disease are sometimes confused. Proper use of these terms requires some knowledge of the disease entities even though their pathogenesis and relation to streptococcal infection is nearly identical. ARF is usually used to describe the initial or acute onset of the disease. In our case, this being the first initial presentation of the disease, it would be correct to call this ARF. The case fulfills modified Jones criteria as will be discussed below. However, as time goes on it is found that this child has a persistence of the murmur. He also had severe carditis which caused his acute congestive heart failure, as manifestations of ARF, but he subsequently develops chronic heart disease as a sequelae of the ARF carditis and thus it would also be correct to describe him in terms of a more chronic form of the disease, namely Rheumatic Heart disease (RHD). This term implies there has been significant valvulitis, enough to cause valvular scarring. This child is at an increased risk of requiring a valve replacement in the future, especially if he develops another episode of the disease, which puts great emphasis on him receiving long term penicillin prophylaxis, to prevent him from getting streptococcal disease and possible reoccurrence of ARF with worsening RHD.

The study of ARF and RHD parallels the history of modern medicine. At one time in the early 1900s children filled the beds of hospitals dedicated to treat only rheumatic fever. The treatment at that time was simply bed rest, sometimes for up to a year. With improvements in living conditions, reduction of crowding, and industrialization, ARF incidence has steadily decreased in the United States (1). When the link to streptococcal infection was found, the usefulness of using penicillin to prevent future attacks was also established, and ARF incidence decreased further.

However, certain areas of the country and large parts of the under-developed world, including India, Sub-Saharan Africa, Turkey, Australia, New Zealand, and Tonga, still experience many cases of ARF (2). In the United States, there remains a high incidence of ARF in Hawaii and Utah (3-6). In Hawaii, the ethnic groups at greatest risk are those of Polynesian heritage, with Samoan children being at greatest risk (4-6). The Samoan children also appear to be at greater risk of developing carditis (4,5). More than 75% of patients with ARF, in Hawaii, have Polynesian ethnicity within their heritage.

To accurately diagnosis ARF, one should adhere to the modified Jones criteria (7). These criteria have been modified over the years since it was first developed by T. Duckett Jones. The last modification removed the minor criteria of "a history of ARF", since there are fewer cases seen on the continental United States, and the authors wanted to concentrate on first time cases, rather than recurrent ones (8). These criteria were developed to accurately diagnose ARF. It is very important to use these criteria when making the diagnosis. If the criteria are not used, and the patient is misdiagnosed, you may be subjecting the patient to needless penicillin injections for years. It is sometimes difficult for ARF patients to get life insurance and medical insurance later, due to the implications of the cardiac disease. Therefore, the diagnosis must fulfill the modified Jones criteria.

The modified Jones criteria are categorized into Major and Minor criteria. These criteria are based on how specific the manifestation is to the diagnosis of ARF. In other words, a Major criterion is much more specific to ARF than the Minor criteria. Therefore, if a child that has two Major criteria, they can fulfill Jones criteria for the diagnosis, as long as they have some evidence of streptococcal disease. On the other hand, if there is evidence of only one Major criterion, they need two minor criteria to fulfill the diagnosis, along with evidence of streptococcal infection. Since the minor criteria are less specific for the diagnosis of ARF, you cannot make the diagnosis of ARF with just minor criteria. The symptoms may be dampened by giving aspirin or other non-steroidal anti-inflammatory medications too early, thus not allowing the manifestations to fully develop.

Modified Jones Criteria (two majors or one major + two minors required) (7)
. . . . . Major criteria: carditis, migrating polyarthritis, chorea, erythema marginatum, subcutaneous nodules.
. . . . . Minor criteria: fever, arthralgia, elevated acute phase reactant (CRP or ESR), prolonged PR interval (i.e., first degree AV block). Leukocytosis used to be a minor criterion, but it no longer is.
. . . . . Plus: All must have evidence of streptococcal infection (positive ASO titer, Streptozyme, positive streptococcal throat culture.

The polyarthritis must be migratory. This manifestation is one of the most common of the major criteria in ARF. Usually one joint becomes involved and over a few days resolves, then another joint(s) becomes involved as demonstrated in our case. Occasionally, the first joint does not resolve completely by the time the second joint becomes involved, and this is termed "additive arthritis", and also fulfills a diagnosis of migrating polyarthritis. In ARF, two or more joints are considered polyarthritis. If migrating polyarthritis is present you cannot use the minor criteria of "arthralgias", as virtually all the children with polyarthritis from ARF have a significant amount of pain. The most common joints involved are large joints, usually those that weight bear. Knees and ankles are most often involved, although elbows and wrists can also be involved. Metatarsophalangeal joints can be involved and one can screen for their involvement by squeezing them together, across the foot, and eliciting pain. The joint pain of ARF is typically very severe even if the visual findings are not very impressive. Merely touching the joint often elicits severe pain. Lower extremity joint involvement renders these patients non-ambulatory.

The presence of a new murmur due to cardiac disease (i.e., carditis) is always sought on physical exam of a child that presents like our case. A very careful cardiac examination should include a description of the PMI and its location (for evidence of congestive heart failure). Our case had an enlarged PMI that was displaced to the lateral side indicating cardiomegaly. The enlarged liver size gave further evidence of congestive heart failure. These findings are important to note, especially in a child with possible symptoms of orthopnea. Congestive heart failure is a severe form of carditis in ARF, and is managed more aggressively, often needing corticosteroids, diuretics, digoxin, and occasionally inotropic agents.

More often, the carditis of ARF is not quite this severe, but can be problematic. The most common valve involved is the mitral valve. The second most common valve involved is the aortic valve. Classic mitral insufficiency sounds like a holosystolic murmur heard at the apex which radiates to the axilla. There are very few cardiac lesions that can be heard in the axilla. Besides mitral insufficiency, a ventricular septic defect could be heard in the axilla, but this murmur is usually heard all over the precordium. The murmur of aortic insufficiency is a diastolic murmur (difficult to hear) that is usually heard best at the upper left sternal border. There is often a decrescendo component to this murmur that is sometimes very high pitched. One should also listen for a rub which would indicate pericarditis and a gallop for evidence of congestive heart failure.

The initial valvulitis of ARF results in valvular insufficiency. Subsequently as RHD develops, if enough inflammation has occurred on the valve leaflets of the mitral valve, the leaflets may scar and become adherent to each other, resulting in mitral stenosis (usually seen late in the patient's course, sometimes after repeated episodes of ARF). The murmur of mitral stenosis is a diastolic murmur, although it is described as occurring in mid-diastole, rather then later in diastole like aortic insufficiency. Similarly, aortic stenosis may subsequently result from initial aortic insufficiency.

The other major criteria describe manifestations that are less often seen in ARF. These manifestations are usually seen in less than 20% of cases. Chorea is the more common of these three, and is often difficult to diagnosis. It is also known as Sydenham's chorea or St. Vitus dance, and causes purposeless and involuntary movements. The hands and tongue are often involved. Parents may also notice the child having mood swings or just "not acting right". Emotional lability is often seen with this manifestation of chorea. Occasionally, chorea occurs so late in the illness that the laboratory tests including ASO titers, ESR, and CRP titers may all be normal. Thus, chorea is often termed a "subacute" phenomenon of rheumatic fever (as opposed to acute rheumatic fever). Despite this lack of evidence of inflammation these patients can develop cardiac disease. Typically, the chorea is not present while sleeping. The chorea usually resolves with time.

Both subcutaneous nodules and erythema marginatum are less common in ARF, but if they are seen there, there is a greater chance of the patient developing carditis. The nodules are usually small being <0.5 cm in diameter, and are seen in <20% of patients with ARF. They are located over areas that tend to be more prominent and rub against surfaces causing microtrauma. For example, they can be located at the tips of the elbows, around the joints, and the bony prominences of the spinal column. It is worthwhile spending some time looking for the nodules as their presence heralds severe carditis (9).

Erythema marginatum is the most rare of the major signs/criteria seen in ARF. It occurs in 5 to 10% of cases. It is a rash usually present over the trunk, and almost never seen over the face. The erythema is described as an evanescent pink eruption with irregular but well-demarcated borders (9). Individual lesions usually last for hours and then disappear, which is why it seen so infrequently. If this rash is found, careful cardiac exams should be done, as these children are at greater risk to develop carditis.

When evaluating a child with acute onset arthritis, the differential diagnosis can be quite overwhelming. Certain elements of the history and physical can help lead to the correct diagnosis. In ARF, the modified Jones criteria are very helpful, but there are other findings which can also help confirm your suspicion of ARF. For example, you should be able to describe the type of arthritis you are observing. Are the joints swollen and without much tenderness, but very stiff in the morning like is seen in Juvenile Rheumatoid Arthritis? Are the effusions rather bland and non-tender lasting for a few days as they are in Systemic Lupus Erythematosus? Is the joint so tender and swollen it can not be moved even a few degrees as is seen in a septic joint? In ARF the joints are usually somewhere between these extremes of pain/tenderness. They can be very painful, but yet if you do not move them, the child is still fairly comfortable. In a septic joint, the child usually has pain even at rest. The classic ARF joint is very warm, only sometimes erythematosus, and very tender. Even the weight of the bed-sheet can cause pain, and this finding is sometimes called the "bed-sheet sign". The tenderness is almost hyper-esthetic, with light pressure causing pain.

The treatment of ARF and RHD is often confusing for the medical students and housestaff. If it is confusing to such well trained individuals, just think of the frustration parents may feel when trying to understand the treatment regiment. To simplify the treatment we will separate the regimen into acute management of the inflammatory condition of ARF, and prevention of further episodes of ARF or antibiotic prophylaxis. With any good treatment plan a "healthy" amount of translating medical jargon into simple terms for the parents is needed, which will help compliance issues. This is especially important when dealing with a long term treatment like benzathine penicillin injections on a monthly basis.

The acute arthritis of ARF will normally respond very dramatically to high dose salicylate therapy. The aspirin dose is 70-100 mg/kg/day divided into QID dosing with a maximum dose of 975 mg QID. Aspirin tablets come in 81 mg, 325 mg, and 975 mg. Use enteric coated tablets if available, and ask patients to eat prior to taking the aspirin. Monitor salicylate levels and liver function tests while on aspirin. Be very careful with ARF patients who have some elevation in liver function tests prior to being put on aspirin, since a low grade inflammatory hepatitis can be seen in ARF. The aspirin could aggravate this problem. The treatment duration is usually 4 to 6 weeks or until the ESR or CRP returns to normal. If it is stopped too early, the arthritis usually returns.

If the carditis is mild and the child is asymptomatic from a cardiovascular standpoint, then salicylate therapy is usually given. However, if there is evidence of severe carditis, then corticosteroids are indicated. Severe carditis is manifested by evidence of congestive heart failure (e.g., gallop rhythm, cardiomegaly, etc.) or severe myocardial disease (e.g., two valve disease or a new or a worsening arrhythmia). Close follow-up and evaluation by the cardiology service is warranted. Repeat echocardiograms will be needed. Corticosteroids are indicated for severe carditis under the direction of a cardiologist. Prednisone is usually given for 2 to 3 weeks followed by aspirin while the corticosteroids are tapered.

Some RHD patients will develop an indolent flare-up of their cardiac disease which is far removed in time from their first episode of ARF. These patients are extremely challenging. During this indolent flare-up, they develop no fever or arthritis, but just present with worsening cardiac disease. Sometimes, this is found on repeat echocardiograms, or by symptomatic CHF returning without other warning signs of a reoccurrence of rheumatic fever. Often an increase in the ESR, CRP, and ASO titer is also seen, indicating a sub-clinical case of streptococcal infection leading to the recurrence of the immune reaction in ARF. These patients may respond to another course of corticosteroids. This underscores the importance of close follow up by the cardiology service.

Antibiotic prophylaxis against streptococcal infections is utilized to prevent a recurrence of ARF, and thus prevent further damage to the valves. Long term prophylaxis needs to be carefully described to the parent and child. Many of the families do not understand why the child needs penicillin injections when he or she feels fine, following the episode of ARF. Many mistakenly think the injections are for the arthritis and therefore do not comply with this regiment once the arthritis has resolved.

There is currently some debate about whether the penicillin injections should be given every 3 or 4 weeks, as well as, the length of treatment (10), but these arguments are beyond the scope of this article. Suffice it to say, these children require the prophylaxis as long as they are at greatest risk of contracting streptococcal infection, which means at least until adulthood, and some require it for their lifetime. We have recommended that our patients receive it every 4 weeks, partly due to compliance concerns. With every 3 weeks it is difficult for families to remember when to get their injection, and this has an increased negative effect on compliance.

Oral antibiotics can also be used but have higher recurrence rates of ARF, than the intramuscular injections. If the child forgets one or two days of oral antibiotics, they are at risk of contracting streptococcal infection, and this is the reason for the higher recurrence rates with oral antibiotics. In penicillin allergic patients, the only option is to utilize oral antibiotics.

It is important to counsel families on the importance of preventing subacute bacterial endocarditis (SBE) from occurring in RHD patients. Like children with other cardiac malformations, once a child is diagnosed with RHD, they are at similar risk of developing SBE. Antibiotics for prophylaxis against alpha-hemolytic viridans streptococci valvular infection is important prior to and following any dental or gastrointestinal procedure. These recommendations can be found elsewhere in this textbook.

The development of persistent cardiac disease is dependent on the amount of inflammation suffered by the cardiac structures during the acute period of disease and by the number of recurrences. Each recurrence will cause increased damage to valvular components and an increased likelihood of mitral stenosis, and the need for valve replacement. The mortality from ARF and RHD probably lies somewhere between 1 to 5 %, although most of the prognostic studies were done decades ago. A classic study demonstrated that with increased carditis severity, there is an increased risk of subsequent cardiac disease (see below).

Estimated occurrence of Rheumatic Heart Disease 5 years after ARF (11,12)
Initial clinical status (%risk of subsequent cardiac disease)
No carditis (4%)
Soft apical murmur (18%)
Loud apical murmur (32%)
Diastolic murmur (47%)
CHF or pericarditis (100%)

Few prognostic studies have been done in the recent past, and should probably be repeated to understand the current risk to children with ARF developing chronic cardiac disease and RHD.

The diagnosis of ARF can be challenging and difficult to make. However, the modified Jones criteria can be extremely helpful in assisting the clinician in this process. It is important to verify the development of the major criteria before starting treatment, because treating too early may stop migration of the arthritis and make fulfilling Jones criteria more difficult. Without fulfilling Jones criteria it is difficult to justify long term penicillin prophylaxis, which may last decades, to patients and their families.

Questions (authored by Neal Rojas, MD-UCSF Residency Program)

1. What is the main difference between Rheumatic Heart Disease (RHD) and Acute Rheumatic Fever (ARF)?
. . . . . a. In ARF there is an elevated ESR
. . . . . b. In RHD there is a prolonged P-R interval
. . . . . c. In ARF there is a history of arthralgias
. . . . . d. In RHD there is evidence of chronic heart disease
. . . . . e. In ARF there is evidence of erythema marginatum

2. All of the following are included in the revised Jones Major criteria EXCEPT:
. . . . . a. New murmur (carditis)
. . . . . b. Migrating polyarthritis
. . . . . c. Chorea
. . . . . d. Maculopapular rash
. . . . . e. Subcutaneous nodules

3. A 7 year old girl presents with a tender and swollen right knee as well as a more recently appearing swollen left wrist. She also has a fever. This patient fulfills which of the following modified Jones criteria?
. . . . . a. 1 Major 1 minor
. . . . . b. 1 Major 2 minors
. . . . . c. 2 Majors
. . . . . d. 2 Minors
. . . . . e. 1 Major only

4. Which of the following symptom lists of ARF are in the correct order of most common' least common?
. . . . . a. Erythema marginatum, subcutaneous nodules, carditis, fever
. . . . . b. Arthritis, carditis, chorea, erythema marginatum
. . . . . c. Chorea, erythema marginatum, subcutaneous nodules, carditis, fever
. . . . . d. Arthritis, chorea, fever, carditis, subcutaneous nodules
. . . . . e. Fever, chorea, carditis, erythema marginatum

5. Salicylates are directed primarily at what symptom in ARF?
. . . . . a. Rash
. . . . . b. Fever
. . . . . c. Arthritis
. . . . . d. Chorea
. . . . . e. Carditis

6. Corticosteroids are directed primarily at what symptom in ARF?
. . . . . a. Rash
. . . . . b. Fever
. . . . . c. Arthritis
. . . . . d. Chorea
. . . . . e. Severe Carditis


1. Kaplan EL. Global assessment of rheumatic fever and rheumatic heart disease at the close of the century. Circulation 1993;88(4 Pt 1):1964-1972.

2. Steer AC, Carapetis JR, Nolan TM, Shann F. Systematic review of rheumatic heart disease prevalence in children in developing countries: the role of environmental factors. J Paediatr Child Health 2002;38:229-234.

3. Veasy LG, Wiedmeir SE, Orsmond GS, et al. Resurgence of acute rheumatic fever in the intermountain area of the united states. New Engl J Med 1987;316:421-427.

4. Chun LT, Reddy DV, Yamamoto LG. Rheumatic fever in children and adolescents in Hawaii. Pediatrics 1987;79(4):549-552.

5. Chun LT, Reddy DV, Yim GK, Yamamoto LG. Acute rheumatic fever in Hawaii: 1966 to1988. Hawaii Med J 1992;51(8):206-211.

6. Kurahara DK, Tokuda A, Grandinetti A, et al. Ethnic differences in risk for pediatric rheumatic illness in a culturally diverse population. J Rheum 2002;29(2):379-383.

7. 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. Guidelines for the diagnosis of rheumatic fever. JAMA 1992;268:2069-2073.

8. Forster J. Rheumatic fever: keeping up with the Jones criteria. Contemp Pediatr 1993;3:51-60.

9. Amigo MC, Martinez-Lavin M, Reyes PA. Acute rheumatic fever. Rheum Dis Clin 1993;19(2):333-347.

10. Berrios X, Campo E, Guzman B, Bisno AL. Discontinuing rheumatic fever prophylaxis in selected adolescents and young adults. A prospective study. Ann Int Med 1993;118(6):401-406.

11. Cassidy JT, Petty RE. Chapter 12-Arthritis related to infection. In: Cassidy JT, Petty RE (eds). Textbook of Pediatric Rheumatology, third edition. 1995, Philadelphia: W.B. Saunders, pp. 519-520.

12. Collaborative Writing Group United Kingdom and United States Joint report on rheumatic heart disease: the evolution of rheumatic heart disease in children. Five-year report of a cooperative clinical trial of ACTH, cortisone and aspirin. Circulation 1960;22:503.

Answers to questions

1.d, 2.d, 3.a, 4.b, 5.c, 6.e

Questions authored by Neal Rojas, MD-UCSF Residency Progarm

Return to Table of Contents

University of Hawaii Department of Pediatrics Home Page