Rheumatic fever was first described by Hippocrates, a Greek physician, between 400-370 B.C.^[1] The term “rheumatic fever” was first used post-Renaissance by Guillaume de Baillou, a French physician, in the early 1600s. T. Duckett Jones, MD was the first person to publish a set of diagnostic criteria in 1944.^[2]

• Between 400-370 B.C., rheumatic fever was first described by Hippocrates, a Greek physician.^[1]
• In the early 1600s, the term “rheumatic fever” was first used post-Renaissance by Guillaume de Baillou, a French physician.
• In the late 1600s, Thomas Sydenham described a case of severe rheumatic fever.
• In 1771, the term “rheumatic fever” first appeared with a description in Encyclopedia Britannica’s first edition.
• In the 1850s, the number of cases of rheumatic fever began to decline worldwide.
• In the 1900s, high-grade cases of rheumatic fever became less common in modernized parts of the world.^[3]
• In World War II, conditions of over-crowding developed in military training sites and rheumatic fever became more common.
• In 1944, U.S. Naval leaders hired T. Duckett Jones, MD to study rheumatic fever’s epidemiology and clinical development. Dr. Jones devised certain a criteria for the diagnosis of rheumatic fever that is still in use today, albeit in a modified form.^[4] The disease was not common in the United States, and usually occurs in isolated outbreaks.
• In the 1980s, the most recent outbreak of rheumatic fever occurred in the United States.
• In the present day, rheumatic fever is more common worldwide, especially in overcrowded areas.

This section covers the diagnostic classification framework for acute rheumatic fever (ARF), centered on the 2015 AHA Revised Jones Criteria — the current standard for ARF diagnosis. It details the population-stratified diagnostic thresholds (low-risk vs. moderate/high-risk), the distinction between initial and recurrent ARF, the concept of “possible” ARF, the role of subclinical carditis, evidence of preceding Group A Streptococcus (GAS) infection, and the broader disease spectrum taxonomy from ARF through chronic rheumatic heart disease (RHD). Clinical descriptions of individual manifestations, echocardiographic technique, laboratory workup, and treatment are covered in their respective microchapters.

The Jones Criteria, originally developed in 1944, underwent their most significant revision in 2015 by the American Heart Association (AHA) to address two key limitations of prior versions: (1) the failure of a single set of criteria to serve populations with vastly different disease burdens, and (2) the exclusion of echocardiography-detected (subclinical) carditis.^[1][2] The 2015 revision introduced population-stratified diagnostic thresholds, with more lenient criteria for moderate/high-risk populations (prioritizing sensitivity) and more stringent criteria for low-risk populations (prioritizing specificity).^[1]

• Low-risk populations: ARF incidence <2 per 100,000 school-aged children (5–14 years) per year, or all-age RHD prevalence ≤1 per 1,000 population per year (e.g., most of North America, Western Europe, Japan).^[1][3]
• Moderate/high-risk populations: ARF incidence ≥2 per 100,000 school-aged children per year, or all-age RHD prevalence >1 per 1,000 population per year (e.g., sub-Saharan Africa, South Asia, Oceania, Indigenous populations of Australia/New Zealand).^[1][3]

When population risk status is uncertain, clinicians should apply the moderate/high-risk criteria to minimize the risk of missed diagnoses.^[1] Of note, Dougherty et al. (JACC 2023) define moderate/high-risk populations using >2 per 100,000 (strictly greater than), whereas Gewitz et al. (AHA 2015) and Hirani et al. (Lancet 2025) use ≥2 per 100,000. This distinction is rarely consequential in practice.^[2][1]

All diagnoses require evidence of preceding GAS infection, with two exceptions: Sydenham chorea and indolent (insidious-onset) carditis may stand alone without documented GAS evidence, because of the long latency period of chorea (months) and the potential for streptococcal markers to have normalized by the time of presentation.^[3][1][4]

• Initial ARF: 2 major manifestations, OR 1 major + 2 minor manifestations.^[3][1]
• Recurrent ARF (in a patient with a prior documented ARF episode): 2 major, OR 1 major + 2 minor, OR 3 minor manifestations.^[3][1]

‡Polyarthralgia should only be considered as a major manifestation in moderate-risk to high-risk populations after exclusion of other causes.^[1][3]

Key rule: Joint manifestations can be counted in either the major or minor category, but not both simultaneously in the same patient.^[3]

A landmark change in the 2015 revision was the formal inclusion of subclinical carditis (echocardiographic valvulitis without auscultatory findings) as fulfilling the major criterion of carditis.^[1][3] Pathological mitral or aortic regurgitation detected by echocardiography is now diagnostic of carditis regardless of the presence or absence of a murmur.^[3] Isolated pericarditis or myocarditis should almost never be considered rheumatic in origin; valvulitis is the most consistent cardiac feature of ARF.^[3] Detailed Doppler criteria (jet length, velocity, and duration thresholds) are covered in the Echocardiography microchapter.

Subclinical carditis is estimated to occur in approximately 12–21% of ARF cases. A study from Sudan showed that up to 11% of febrile children had subclinical carditis as their only major manifestation of ARF, underscoring the need for echocardiography in all suspected ARF cases.^[3]

One of the following must be documented to satisfy this requirement:^[3][1]

• Positive throat culture or rapid antigen detection test (RADT) for GAS
• Elevated or rising streptococcal antibody titers (ASO, anti-DNase B)
• Recent documented scarlet fever (listed in the AHA statement but rarely the sole means of documentation in practice)

Providing evidence of antecedent GAS infection remains an ongoing challenge, particularly in endemic settings where background seropositivity is high and skin infection (impetigo) may also trigger ARF.^[3]

Patients who do not fully meet the Jones Criteria but have a clinical presentation highly suggestive of ARF may be classified as “possible ARF”, particularly in moderate/high-risk populations. This category is clinically important because it may warrant initiation of secondary prophylaxis pending further evaluation.^[3][1] However, the non-specific nature of some minor features (arthralgia, elevated ESR) within this category carries a risk of misdiagnosis, unnecessary prophylaxis, and patient anxiety.^[3]

Recurrent episodes carry a higher risk of progressive valvular damage and RHD. The 2015 criteria use a lower diagnostic threshold for recurrent ARF: 3 minor manifestations alone (with evidence of GAS infection) can satisfy criteria in a patient with a documented prior ARF episode.^[3][1] This is a key distinction from the initial episode, which always requires at least one major manifestation.

ARF and RHD represent a continuum of disease rather than discrete entities. The following taxonomy, based on the JACC 2023 framework, classifies the spectrum according to detection method (clinical vs. echocardiographic) and timing:^[2]

• ARF with clinical carditis: Valvulitis detected by auscultation during an acute ARF episode
• ARF with subclinical carditis: Valvulitis detected only on echocardiography during an acute ARF episode
• Clinical RHD: Chronic valvular disease detected when symptomatic patients present to care
• Subclinical/latent RHD: Chronic valvular lesions detected by echocardiographic screening in asymptomatic individuals with no known ARF history

The original WHF echocardiographic criteria were published by Reményi et al. (Nat Rev Cardiol, 2012) and are reproduced in Watkins et al. (JACC, 2018).^[5][6] The 2012 WHF guidelines (updated in 2023) introduced a classification for definite RHD with categories A–D based on valve pathology pattern. For individuals aged ≤20 years: (A) pathological MR with ≥2 morphological features of RHD of the MV, (B) MS with mean gradient ≥4 mmHg, (C) pathological AR with ≥2 morphological features of RHD of the AV, and (D) borderline disease of both the AV and MV. For individuals aged >20 years: Categories A and B are the same; (C) pathological AR with ≥2 morphological features of the AV applies only to those aged <35 years; and (D) is redefined as pathological AR with ≥2 morphological features of RHD of the MV. A separate “borderline RHD” category (applicable only to those aged ≤20 years) was also introduced, with 3 subcategories: (A) at least 2 morphological features of RHD of the MV without pathological MR or MS, (B) pathological MR alone, and (C) pathological AR alone.^[6][5]

The 2023 WHF update revised Doppler thresholds for pathological regurgitation, notably introducing weight-based MR jet length criteria (≥1.5 cm for individuals weighing ≤30 kg; ≥2.0 cm for individuals weighing ≥30 kg), replacing the uniform ≥2 cm threshold from 2012.^[3][6] Of note, the inter-rater reliability of the 2012 WHF criteria for specific RHD classification categories (borderline vs. definite, and subcategories) is only fair (κ = 0.51), highlighting the challenge of consistent classification across observers.^[7]

A 2026 review by Zühlke et al. references 2024 WHO recommendations as providing an additional global framework for ARF/RHD management and classification, though the primary WHO guideline document should be consulted directly for specific recommendations.^[8] National ARF guidelines from RHDAustralia (2020)^[9] and the New Zealand Heart Foundation (2014, updated 2019)^[10] provide additional region-specific classification frameworks for high-risk populations and may be consulted for local clinical guidance.

• Chorea and indolent (insidious-onset) carditis are the only two manifestations that do not require evidence of preceding GAS infection for ARF diagnosis.^[3][4]
• PR prolongation cannot be used as a minor criterion if carditis is already counted as a major criterion in the same patient.^[1][3]
• Joint manifestations can only be counted in one category (major OR minor) per patient — never both.^[3]
• In moderate/high-risk populations, monoarthralgia alone can serve as a minor criterion, substantially lowering the diagnostic bar.^[1]
• The addition of subclinical carditis and monoarthritis to the criteria increased sensitivity in high-risk populations. Application of the 2015 modifications (subclinical carditis, monoarthritis, lower fever threshold) to a North Queensland cohort increased diagnostic sensitivity from 71.4% to 91.8% compared with the 1992 criteria.^[1]
• The 2015 AHA statement uses ≥38.5°C/≥38°C and CRP ≥3.0 mg/dL for fever and CRP thresholds, while Hirani et al. (Lancet 2025) uses >38.5°C/>38°C and CRP >3 mg/dL (strictly greater than). The table above follows the AHA formulation. In practice these distinctions are rarely consequential, but clinicians should be aware of the minor discrepancy between sources.^[1][3]

• Applying low-risk population criteria in endemic settings leads to missed diagnoses. Clinicians in moderate/high-risk regions should use the more sensitive thresholds.^[1]
• Diagnosing ARF without documenting GAS infection (except for chorea and indolent carditis) is insufficient; GAS evidence is mandatory.^[3][1]
• Counting arthritis as both a major and minor criterion is incorrect; joint findings can only be used once.^[3]
• Ignoring subclinical carditis by failing to perform echocardiography can miss a substantial proportion of carditis cases, potentially leaving patients without secondary prophylaxis.^[3]
• Confusing “possible ARF” with “no ARF”: Patients with possible ARF require close follow-up and, in many guidelines, initiation of secondary prophylaxis while awaiting further clarification.^[3][1]
• Overlooking the lower diagnostic threshold for recurrent ARF (3 minor criteria alone can suffice).^[3][1]

Acute rheumatic fever (ARF) is an immune-mediated (autoimmune) multisystem sequela of superficial group A Streptococcus (GAS; Streptococcus pyogenes) infection, not a result of direct bacterial invasion of target organs. It arises 2–4 weeks after pharyngitis or skin infection in genetically susceptible hosts whose immune systems have been primed by prior streptococcal exposure.^[1]

The mechanisms that inform diagnosis, prognosis, and secondary prevention include repeated-infection priming, molecular mimicry and neo-antigen generation, the antibody-then-T-cell cascade producing valvulitis, two-phase evolution of carditis, chemokine-driven fibrosis, and host genetic susceptibility. Clinical Jones criteria, echocardiographic findings, and treatment protocols are addressed in their dedicated microchapters and are referenced here only for mechanistic context.

• The initiating event is a superficial GAS infection. Pharyngitis is the classic and best-established trigger; skin infection (impetigo/pyoderma) is increasingly recognized as a trigger or early-life priming exposure, particularly in high-burden Indigenous and Pacific populations.^[1]
• Repeated infections appear necessary to prime the immune system for loss of self-tolerance. This represents an update from the older single-infection model and is supported by epidemiologic and serologic data.^[1]
• Only a subset of untreated children develop ARF, and only a minority of those progress to rheumatic heart disease (RHD), implicating host genetic susceptibility as a necessary co-factor.^[2]

After GAS adheres to and invades pharyngeal or skin epithelium, neutrophils, macrophages, and dendritic cells phagocytose bacteria and present antigen to T cells. Both B cells (producing IgM, IgA, IgG) and CD4+ T cells are activated.^[1]

Molecular mimicry is the prevailing hypothesis: streptococcal antigens structurally resemble human proteins, generating cross-reactive antibodies and T cells. The two most-implicated streptococcal antigens are the surface M protein and the group A carbohydrate epitope N-acetyl-β-D-glucosamine (GlcNAc).^[3][4]

• M protein and cardiac myosin are both α-helical coiled-coil molecules; cross-reactive T cells recognize both. Because myosin is absent from valves, cross-reactivity with valvular laminin and collagen is invoked to explain valvulitis.^[5][4]
• Autoantibody reactivity is broad and evolves via epitope spreading. IgG3 is elevated in ARF, and the anti–M-protein response is IgG3-polarized.^[6]

A complementary mechanism posits that GAS penetrates the epithelial basement membrane, where an octapeptide motif on M protein (PARF) binds the CB3 region of type IV collagen, rendering it immunogenic and inducing anti–type IV collagen antibodies and systemic inflammation. These antibodies may cross-react with type I collagen in the valve.^[3]

The molecular-mimicry and neo-antigen theories are not mutually exclusive and may act sequentially, consistent with the presence of both anti-myosin/laminin and anti-collagen antibodies in rheumatic carditis.^[3]

The mechanistic sequence that produces valve damage proceeds as follows:

1. Antibody deposition and endothelial activation — Cross-reactive (antiendothelial) antibodies bind valve endothelium and up-regulate vascular cell adhesion molecule-1 (VCAM-1).^[5]
2. T-cell infiltration — VCAM-1 permits CD4+ T cells (expressing integrin α4β1/VLA-4) to adhere and extravasate into the valve. CD4+ cells predominate over CD8+ in the inflamed valve.^[5]
3. Cytokine-mediated damage — Infiltrating T cells and macrophages generate a Th1 (and Th17) response with IFN-γ, TNF-α, and IL-17, driving valve breakdown.^[4]
4. Epitope spreading — Tissue breakdown liberates endogenous antigens (collagen, laminin, myosin, tropomyosin) that recruit further autoreactive T cells, amplifying and perpetuating injury across successive episodes.^[7]

Carditis occurs in more than 50% of first ARF episodes and is typically a pancarditis that almost always includes valvulitis (inflammation of the valvular endocardium).^[5] Subclinical carditis—echocardiographic valvulitis without auscultatory findings—has a weighted pooled prevalence of approximately 17% (95% CI 11.9–21.6%; range across studies 0–53%) and provides the mechanistic rationale for echocardiographic screening.^[8][9]

• CXCL9 and CXCL10 are elevated in ARF/RHD and recruit CXCR3+ T cells to the heart; circulating CXCR3+ cells fall as they home to valve tissue, providing direct evidence of T-cell trafficking to the valve.^[1][6]
• TGF-β drives fibrotic remodeling and neovascularization; neovascularization grants immune cells greater valve access in later episodes. Tenascin C links inflammation to fibrosis.^[1][7]
• Damage concentrates in the left-sided (mitral > aortic) valves, which endure the highest pressure gradients and have limited repair capacity, explaining the anatomic distribution of clinical disease.^[1]

• Exudative phase (first 2–3 weeks): interstitial edema, cellular infiltration, collagen fragmentation, and fibrinoid deposition.^[10]
• Proliferative phase (months to years): formation of Aschoff bodies—the pathognomonic lesion, consisting of central fibrinoid necrosis surrounded by T lymphocytes, plasma cells, and activated macrophages (Anitschkow cells), found chiefly in the myocardium.^[10][1]
• The conduction system is rarely structurally involved; the common PR prolongation is functional/inflammatory rather than a fixed lesion.^[10]
• Chronic RHD is the cumulative product of one or more episodes of carditis: repeated inflammation causes fibrosis, leaflet thickening/calcification, commissural fusion, and chordal shortening/fusion, producing mitral regurgitation (acute/early) and, over years, mitral stenosis.^[5][11]
• Progression to RHD depends on severity of initial carditis, number of recurrences, and adherence to secondary penicillin prophylaxis—the biologic basis for long-term secondary prevention.^[10][11]

• Twin studies estimate ARF heritability near 60%; children of parents with RHD have approximately 2.9-fold higher risk even when raised apart.^[7]
• Genome-wide association studies implicate HLA class II alleles (varying by population) and immune-gene loci including CTLA4, FCGR2A, FCN1–3, IL10, IL1RN, MBL2, TLR2, TNF, and TGFB1. The FCN (ficolin) and MBL2 loci converge on lectin-pathway complement activation. The RHDGen study replicated an IGH locus association and identified a novel chromosome-11 locus in Black African individuals.^[4]
• Sex predilection: RHD is predominantly a disease of women (approximately 80% of cases). Prothymosin-α (ProTα), linked to estrogen-receptor-α–associated CD8+ cytotoxicity, has been proposed as a contributor.^[4]

• Molecular mimicry remains a hypothesis rather than proven causation; the pathophysiologic link between GAS and ARF is under active investigation.^[1]
• The role of group A carbohydrate has been questioned because of the quality of antigens used in older studies.^[1]
• Animal models are imperfect: the Lewis rat model most often produces myocarditis rather than valvulitis, raising validity concerns.^[1]
• Superantigen involvement is unresolved: TCR Vβ-skewing has been inconsistent across studies, and direct evidence for a superantigen role in ARF/RHD is currently lacking.^[12]
• Emerging models frame ARF/RHD as mechanistically distinct stages—early valvulitis driven by systemic inflammation and high-titer autoantibodies versus chronic valve-intrinsic pathology sustained by tissue remodeling and local T-cell dysregulation—with therapeutic implications.^[13]

• ARF is a delayed autoimmune sequela (2–4 weeks post-infection), not an active bacterial infection of the heart or joints; antibiotics eradicate residual GAS but do not treat the ongoing immune injury.
• No disease-specific biomarker or diagnostic test exists; diagnosis remains clinical (Jones criteria). No immunomodulatory therapy has been proven to halt ongoing immune injury or prevent progression to RHD; current management is limited to symptom control and GAS eradication, and corticosteroid/IVIG data remain non-definitive.^[1][9][6]
• Subclinical carditis is common enough that a normal cardiac examination does not exclude valvulitis; echocardiography is required for detection.
• Aschoff bodies are pathognomonic but represent a myocardial/proliferative-phase finding and are not a real-time diagnostic tool.
• Recurrence drives cumulative valve damage and progression to RHD. The priming/recurrence biology is the rationale for long-term secondary antibiotic prophylaxis.

• Attributing ARF to direct streptococcal infection of the heart or joints (it is immune-mediated).
• Assuming a single sore throat is sufficient to cause ARF—repeated priming infections are implicated.
• Overlooking skin infection as a trigger or priming exposure, especially in high-burden populations.
• Equating a normal auscultatory examination with absence of carditis (thereby missing subclinical valvulitis).
• Interpreting the histologic presence of Aschoff bodies as evidence of active, clinically detectable disease at a specific point in time.

Rheumatic fever is usually caused by an infection with Group A beta-hemolytic Streptococcus pyogenes.

Common causes of rheumatic fever include:^[1][2][3]

• Group A beta-hemolytic Streptococcus pyogenes pharyngitis
• Mistreated or untreated scarlet fever
• Antibody type II hypersensitivity reaction

Rheumatic fever must be differentiated from other diseases that cause fever, skin rash, nausea and fatigue, such as typhoid fever, malaria, lassa fever, ebola, and scarlet fever.

Mitral stenosis must be differentiated from the following:^{[1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]}

The incidence of rheumatic fever among developed countries such as the USA and Canada was approximately 20-40 per 100,000 individuals between 1970-1990 with rheumatic heart disease occurring sporadically. Over past two decades, the prevalence of rheumatic fever has decreased to 5-20 per 100,000 individuals in Canada and <5 per 100,000 individuals in the USA with no new cases of rheumatic heart disease.

Between 1990 and the present, the prevalence of rheumatic fever has decreased to 5-20 per 100,000 individuals in Canada and <5 per 100,000 individuals in the USA with no new cases of rheumatic heart disease.^[1] More information regarding prevalence of rheumatic fever can be found here.

Between 1970-1990, the incidence of rheumatic fever among developed countries such as the USA and Canada was approximately 20-40 per 100,000 individuals between 1970-1990 with rheumatic heart disease occurring sporadically. More information regarding incidence of rheumatic fever can be found here.

Rheumatic fever tends to affect a younger age group, generally between the ages of 5-15.^[2] Group A beta-hemolytic streptococcus pharyngitis is uncommon in children less than 3 years of age, and rheumatic fever is extremely rare. Rheumatic fever among adults is less frequent and accounts for approximately 20% of cases.

No predilection towards either gender exists. However, certain manifestations of rheumatic fever such as mitral valve prolapse, carditis, and Sydenham chorea are commonly often observed among females than males.^[3]

The incidence of rheumatic fever among developed countries such as the USA and Canada was approximately 20-40 per 100,000 individuals during the period 1970-1990 with rheumatic heart disease occurring sporadically. Over past two decades, the prevalence of rheumatic fever has decreased to 5-20 per 100,000 individuals in Canada and <5/100,000 individuals in the USA with no new cases of rheumatic heart disease.^[1] The incidence of rheumatic fever among developed nations is low, likely due to improved hygienic standards and routine use of antibiotics for acute pharyngitis.^[4]

The rates of rheumatic fever and rheumatic heart diseases among developing nations have mixed trends. Recently, the incidence rate of rheumatic fever has decreased in India, China, and African countries, which has been mainly attributed to improved access to medical treatment. This has led to improved survival rates even among individuals with rheumatic heart disease. As a result, an increase in prevalence rate has been observed.^[1] Newer studies relying on echocardiography in the diagnosis of rheumatic heart disease have demonstrated that rates of subclinical carditis are up to 10x higher than that diagnosed by clinical examination.^[5][6][7][8] India, Pakistan, Russia, and African countries have higher prevalence rates of rheumatic heart disease. It is estimated that approximately there are 62-78 million rheumatic heart disease patients worldwide, which could potentially result in 1.4 million deaths per year.^[9][10] Occurrence of rheumatic fever is associated with low socioeconomic and over crowded conditions.

Common risk factors in the development of rheumatic fever include poor sanitation, low socioeconomic status, and family history.

Common risk factors in the development of rheumatic fever include:^[1][2][3]

• Poor sanitation
• Low socioeconomic status
• Positive family history
• Inadequate healthcare
• Dampness at home
• Overcrowding
• Malnutrition
• Traveling to endemic regions, such as developing countries

Screening of rheumatic fever is important as many cases of rheumatic heart disease are subclinical. Echocardiography among inhabitants of high risk regions is recommended. If any abnormality is detected on echocardiography, further cardiac evaluation is performed followed by antimicrobial therapy.

• According to the World Health Organization, screening for rheumatic fever by echocardiogram is recommended among patients with from high risk communities with abnormal auscultatory findings. Early detection of subclinical disease process helps in early treatment and improves prognosis by preventing recurrence.^[1] Screening is performed to prevent significant valvular disease.^[2]
• Screening of rheumatic heart disease involves:
  • Eliciting history of rheumatic fever
  • Cardiac auscultation for murmurs
  • Echocardiography

If left untreated, patients with rheumatic fever may progress to develop arrhythmias, systemic emboli, and endocarditis, which may lead to cardiac failure. Common complications of rheumatic fever include mitral/aortic valve stenosis, carditis, and stroke. Prognosis is generally poor if left untreated.

If left untreated, rheumatic fever may cause valvular diseases including stenosis, regurgitation of mitral/aortic valves and myocarditis. This may lead to decreased cardiac output, pulmonary edema, and ultimately cardiac failure. If an individual develops rheumatic fever, they will develop an increased sensitization to Streptococcus pyogenes autoantigens. Future infection will likely cause an elevated, autoimmunological response and a more severe case of rheumatic fever will develop.

It is estimated that the recurrence rate of rheumatic fever is decreased by about 85% by providing prophylactic penicillin therapy. Recurrence rate of 0.2/patient/year follow-up was noted among those not receiving regular treatment.^[1]

Complications to rheumatic fever include:^[1][2][3]

• Mitral stenosis
• Aortic stenosis
• Endocarditis
• Myocarditis
• Pericarditis
• Cardiac failure
• Arrhythmias
• Systemic emboli
• Stroke
• Recurrence of rheumatic fever

For cases without carditis, the prognosis is excellent, demonstrating no residual heart disease. In cases with preexisting or developing heart disease, the prognosis is poor, leading to mortality.^[4]