Article Text
PDF
Abstract
A 24-year-old fit and well Caucasian man was referred to acute hospital via his General Practitioner with chest pain, palpitations, shortness of breath and an antecedent sore throat. Investigations revealed pericardial and pleural effusions, pericardial thickening on MRI, mild mitral regurgitation on echocardiogram and a raised Antistreptolysin O (ASO) titre.
He was treated as acute rheumatic fever (ARF) with a prolonged course of penicillin, supportive therapy with bisoprolol and colchicine with lansoprazole cover. The patient made a full recovery and subsequent cardiac MRI showed resolution of all changes.
- cardiovascular medicine
- pericardial disease
- infectious diseases
Statistics from Altmetric.com
Background
Acute rheumatic fever (ARF) is a systemic inflammatory autoimmune disease that classically occurs 2–3 weeks following throat infection by group A streptococci (GAS).1–4 ARF can affect multiple systems potentially resulting in carditis, arthritis, chorea, erythema marginatum and subcutaneous nodules.5 While the majority of these symptoms resolve without consequence, carditis can lead to chronic rheumatic heart disease5 in the form of permanent valvular damage, resulting in serious long-term health implications,1 6 including arrhythmias, stroke, complications during pregnancy, heart failure and death.5 The carditis most commonly affects the mitral valve, resulting in regurgitation.7 The arthritis is migratory and asymmetrical affecting mostly the large joints causing pain, swelling and limited movement.2 7 Chorea affects fewer than 15% of people and results in involuntary movements, clumsiness, unsteady gait or fidgeting.2 7 Chorea typically occurs after a longer latent period and may follow a fluctuating course of several months.2 Erythema marginatum is a rare presentation seen in <5% and appears as an annular erythema with raised edges, which is not painful or itchy.7 Finally, subcutaneous nodules are another rare finding, generally seen as small painless nodules on the extensor surfaces.7
Pathogenesis is thought to consist of infection with a rheumatogenic strain of group A streptococcus causing an aberrant host immune response in a genetically susceptible individual.3 Antibody and T-cell responses against bacterial virulence factors cross-react with host tissue proteins in the myocardium, synovial tissue and in the chorea of the basal ganglia, causing an inflammatory response,2–4 leading to the above symptoms. This cross-reactivity has been known for some time8 and this theory is supported by the finding of anticardiac autoantibodies.9 Although classically thought to be due to group A streptococci, there are documented cases following group G infection10 11 and there are studies demonstrating autoimmune carditis in rat models of rheumatic heart disease in response to group G.4 Group G strep is a normal commensal organism often isolated from the throat12 but one hypothesis is that virulence factors could be acquired by horizontal transfer from group A streptococcus.1
Incidence of ARF fell throughout the 20th century and is now a rare condition in high and middle-income countries1 9 13 and as such is a condition that today’s junior doctors in the UK have little experience of. This reduced incidence is felt to be due to improved living conditions, nutrition and access to healthcare,1 13 but ARF continues to cause significant morbidity in low-income countries and is still seen in indigenous communities in Australasia and during outbreaks in some middle and high-income countries. The majority of cases occur in children aged 5–15 years but young adults are occasionally affected.2
This report describes the presentation and subsequent treatment of a man with a working diagnosis of ARF. As a result of the varying manifestations of ARF, the diagnosis is clinical and based on a set of criteria rather than a single diagnostic test. This case highlights the difficulty and ambiguity that doctors can face when making a medical diagnosis, in a wider respect the uncertainty we face on a daily basis and how this affects patients.
Case presentation
A 24-year-old Caucasian man presented to his General Practitioner with an initial report of a 2-week history of left shoulder pain, which had begun while lying on a beach on holiday in Spain. Pain had initially shown some improvement with physiotherapy but then moved into his chest and there was associated shortness of breath. The chest pain was sharp and pleuritic in nature and he reported palpitations. ECG showed a sinus tachycardia and given this marked tachycardia in an otherwise fit and healthy young man, he was referred into hospital for further investigation.
Chest X-ray (see figure 1) showed cardiomegaly and a small left pleural effusion. D-dimer was 2419, but as the patient had no risk factors or clinical evidence of deep vein thrombosis, a focused V-scan was performed (as opposed to CT pulmonary angiogram as pulmonary embolus was not suspected). This showed a 10 mm pericardial effusion but no evidence of tamponade. A formal echocardiogram confirmed this along with moderate left ventricular systolic dysfunction and dilated right atrium with raised pressures (>15 mm Hg).
Admitting chest X-ray showing cardiomegaly due to pericardial effusion, and small left-sided effusion.
He was managed as pericarditis plus effusion with non-steroidal anti-inflammatories, colchicine, bisoprolol and an angiotensin-converting enzyme inhibitor and discharged with a plan for urgent outpatient MRI, repeat echo and follow-up in clinic. The MRI, however, showed moderate size increasing pericardial effusion (~17 mm) and a thickened appearance of both visceral and parietal pericardium up to ~6 mm with some septations within the pericardial fluid (see figure 2). The patient was also feeling more unwell and spiked a temperature of 38.8°C and he, therefore, was subsequently readmitted.
Initial cardiac MRI showing a moderate global pericardial effusion and globally thickened and inflamed pericardium (small left pleural effusion present).
On further direct questioning, it transpired that he had had a sore throat prior to admission. This had been mild and resolved spontaneously and as such he had not sought medical opinion. Routine bloods at the time of readmission showed C-reactive protein (CRP) >200, white cell count of 11.0, normal renal and liver function tests, troponin <5. Throat swabs grew Group G streptococcus (sensitive to penicillin), and antistreptolysin O (ASO) titre was moderately raised (200) but antideoxyribonuclease B titre was non-diagnostic at <200. Other microbiological tests for causes of myo-/peri-carditis were negative, as were immunological screening tests such as antinuclear antibodies and extractable nuclear antigens. Repeat ASO titre (3 weeks later) was again moderately raised, suggesting previous infection rather than active infection.
Given increasing shortness of breath and a large pleural effusion clinically, chest X-ray (see figure 3) was repeated, which confirmed marked left-sided pleural effusion (and small right-sided effusion). Pleural fluid aspirate revealed a straw coloured, sterile exudate containing blood, inflammatory cells and reactive mesothelial cells. Serial echocardiograms revealed increasing pericardial effusion, reducing ejection fraction (<35% at one point) and bowing of both mitral leaflets with mild regurgitation.
Repeat chest X-ray 1 week later revealing significant left-sided effusion.
Diagnosis
As a result of the varying manifestations of ARF, the diagnosis is clinical and based on the Jones criteria rather than a single confirmatory investigation or examination finding. The Jones criteria, first published in 1944,14 were created to serve as a guide to diagnosis since they ‘occurred together with a frequency that far exceeded chance’.15 These criteria have since been modified to include different criteria for low-risk and moderate-to-high-risk populations.13 Many patients who present with ARF do not remember having antecedent upper respiratory tract symptoms and it is for this reason that confirmation of GAS infection involves demonstration of an immune response (such as increasing ASO titre).9 15 Over time, with these revisions and the decreasing incidence in high-income countries, these criteria have increased in specificity but decreased in sensitivity.1
Modified Jones criteria for ARF for low-risk populations13:
Evidence of preceding GAS infection and two major criteria or one major plus two minor
Major criteria
carditis
polyarthritis
chorea
erythema marginatum
subcutaneous nodules.
Minor criteria
polyarthralgia
fever
CRP ≥3.0 mg/dL or peak erythrocyte sedimentation rate ≥60 mm
prolonged PR-interval on ECG (unless carditis is a major criterion).
The index of suspicion for ARF in healthcare providers may be low due to decreasing incidence of the disease and a clinical presentation that can mimic other conditions. However, the need for awareness of this condition and having a high index of suspicion even when patients do not meet the Jones criteria has been highlighted by a case in Alaska in which a 6-year-old girl reportedly had a negative streptococcal screen but ultimately required urgent mitral valve replacement surgery for congestive heart failure approximately 6 weeks after presenting with cough and fever.6
This patient, therefore, presented with one major criterion (carditis as evidenced on the echocardiogram) and two minor criteria (fever and raised CRP), leading to the working diagnosis of ARF. The left shoulder pain was felt to be referred pain through diaphragmatic irritation rather than an arthralgia. In addition, the classical presentation of arthritis in ARF is a migratory asymmetrical polyarthritis of the large joints2 and it is only this which is included in the Jones criteria in low-risk populations.13 There was no relevant family history to suggest a genetic susceptibility
However, the diagnosis of ARF in this patient is debatable since his ASO titre was raised not rising (suggesting carriage rather than active infection) and his antideoxyribonuclease B titre was also non-diagnostic. The American Heart Association guidelines recognise the difficulty in reaching a clear diagnosis with some cases and therefore recommend that in cases where there is uncertainty, it is not unreasonable to offer a prolonged course of prophylactic antibiotics with careful re-examination.13 The working diagnosis here was ARF, possibly group G streptococcus mediated, as while streptococcus G is a normal commensal organism,12 pathogenesis has been previously postulated in aboriginal Australian and Pacific Islanders.16 Of note are the differing diagnostic criteria in various countries. For example, in New Zealand, this case could be classified as ‘possible ARF’ because of the clinical suspicion and criteria met but without definitive evidence of group A streptococcus infection.17
Treatment
The treatment of ARF consists of antibiotics, generally penicillins (erythromycin for those with a pencillin allergy), to eradicate the streptococcus infection, and supportive therapy for sequalae such as arthritis, carditis or chorea.1 2 Following discharge from acute hospital, this patient was treated with a 6-month course of oral amoxicillin, bisoprolol and colchicine with lansoprazole cover.
Outcome and follow-up
Patients with ARF are at risk of recurrences and, therefore, cumulative damage to cardiac valvular tissue and are, therefore, often offered secondary antibiotic prophylaxis, the duration of which is a matter of clinical judgement.5 9 15 Serial repeat cardiac MRI (see figures 4 and 5) showed complete resolution of both pleural and pericardial effusions, normal ventricular function, no pericardial thickening and, importantly, normal valvular function. This case was deemed relatively low risk because of the low incidence of rheumatic fever in the general population, the complete resolution of cardiac changes seen on MRI and the fact that recurrences are extremely uncommon after the age of 21 years.5
Follow-up cardiac MRI approximately 4 weeks later showing significant reduction in size of the pericardial effusion. Percardium remains thickened. Constriction sequences were positive. left ventricular function improved, EF=47%. Left-sided pleural effusion remains.
Further follow-up cardiac MRI—6 weeks after the previous showing complete resolution of pericardial and pleural effusions. Pericardium is no longer thickened measuring 0.18 cm. Normal LVF, EF=58%. Normal myocardial pattern on gadolinium enhancement.
He has subsequently been discharged from regular cardiology follow-up. He is fit and well with no on-going issues, no recurrence of symptoms and now on no regular medication. Of note, however, it is that during his final cardiology follow-up appointment, he reported a sore throat and a repeat throat swab isolated Streptococcus anginosus, the significance of which remains unclear.
‘Medicine is a science of uncertainty and an art of probability’18 is a famous quote from Osler, which holds true a century later in the age of evidence-based medicine. This uncertainty is something that all doctors are familiar with and deal with on a daily basis. However, this uncertainty is not something that patients are familiar or necessarily comfortable with. The statement above from the patient highlights the importance of key communication with patients regarding diagnosis and management plans especially when the diagnosis is unclear. As a result of this experience, he visited the patient advice and liaison services after discharge to raise concerns regarding the rotation of doctors and resultant lack of clarity.
Patient’s perspective
The uncertainty in medicine plus a different weekly consultant covering the cardiology ward ‘led to contrasting views on the best way to proceed with different doctors steering the direction of treatment in different directions daily. The end result was that it felt like no one actually took hold of and had a grasp of the full case starting from my first admission to hospital. Diagnosis wise I am still not 100% sure what is officially in my notes, if it is rheumatic fever, I can not understand why as I am no treatment for that and have had no repeat episodes’.
Learning points
Acute rheumatic fever is an increasingly rare condition in high and middle-income countries, but one that can be treated appropriately if detected early but lead to significant life long complications if missed.
The fact that this patient was initially discharged with a diagnosis of pericarditis raises, the question about whether this condition is more common than suspected and simply not picked up in mild cases.
This case highlights the uncertainty that doctors face with diagnosis or management plans, and the importance of clear communication with patients about this.
Ethics statements
Patient consent for publication
Footnotes
Contributors This case report was a joint piece of work from selection of case, drafting the work through to final approval. The two authors involved were ZMW and KC and both were equally involved with each part of the case report. Further editions from now on will be done by ZMW.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.