Discussion

A high index of suspicion of infective endocarditis should be considered in congenital heart disease children who present with a non-specific febrile illness, irrespective of the duration of fever, fever pattern, or the resolution of fever with antipyretic medications. Our case series demonstrate that a fever of ≥38°C without any associated focal symptoms in paediatric congenital heart disease patients should be evaluated for infective endocarditis with the inclusion of blood culture studies.

Congenital heart disease is the leading risk factor for infective endocarditis in children.^{Reference Knirsch and Nadal1} The current diagnosis of infective endocarditis is based on modified Duke’s criteria that amalgamate clinical features, microbiological tests, and cardiovascular imaging.^{Reference Habib, Lancellotti and Antunes2} The two major criteria in Duke’s classification require specific tests such as serial blood cultures and echocardiography to be undertaken to confirm the diagnosis of infective endocarditis.^{Reference Habib, Lancellotti and Antunes2} Blood culture is the most important initial laboratory test in the work-up of infective endocarditis. Bacteraemia is usually continuous and the majority of infective endocarditis patients have positive blood cultures.^{Reference Murdoch, Corey and Hoen3} Echocardiography is the second cornerstone of diagnostic testing in suspected infective endocarditis cases.^{Reference Habib, Lancellotti and Antunes2} Transthoracic echocardiography is usually sufficient for children weighing less than 60 kg to fully understand cardiovascular findings in definite or presumptive infective endocarditis.^{Reference Barbour, Louise and O’Keefe4} Patients with congenital heart disease may have had palliative surgeries (shunts and conduits) that are difficult to visualise on transthoracic echocardiography.^{Reference Martin, Neches and Wald5} In these circumstances, transesophageal echocardiography may be more valuable to diagnose paravalvular leakage or dehiscence, left ventricular outflow tract complications, including root abscesses, involvement of the sinuses of Valsalva, and prosthetic valve endocarditis.^{Reference Barbour, Louise and O’Keefe4} Vegetations in right ventricular-to-pulmonary artery conduits and prosthetic pulmonary valves, however, are difficult to visualise on both forms of echocardiography and combined studies may give a better yield than individually.^{Reference Miranda, Connolly and Bonnichsen6} In the absence of a well-defined vegetation, a newly identified prosthetic valve or conduit stenosis on echocardiogram in the presence of isolated fever should raise the suspicion for infective endocarditis.^{Reference Miranda, Connolly and Bonnichsen6} Additional tests including combined positron emission spectroscopy-computed axial tomography may facilitate the diagnosis of infective endocarditis affecting right ventricular-to-pulmonary artery conduits in patients with negative echocardiogram studies.^{Reference Saby, Laas and Habib7,Reference Mauritz, Wagenaar, van der Jagt and Bouman8} Other laboratory investigations for fever including elevated C-reactive protein or erythrocyte sedimentation rate, leucocytosis, anaemia, and microscopic haematuria lack specificity and are not included in the current diagnostic criteria of infective endocarditis.^{Reference Habib, Lancellotti and Antunes2} Minor criteria including vascular phenomena such as Janeway lesions, splinter haemorrhages, mycotic aneurysms, major arterial emboli, and immune phenomena such as Roth spots, Osler nodes, rheumatoid factor, and glomerulonephritis are rarely seen in children.^{Reference Martin, Neches and Wald5,Reference Jortveit, Klcovansky, Eskedal, Birkeland, Dohlen and Holmstrom9} Many children with underlying complex congenital heart disease have pre-existing murmurs that make the recognition of a new murmur difficult.^{Reference Martin, Neches and Wald5}

The signs and symptoms of infective endocarditis in congenital heart disease children are insidious and non-specific.^{Reference Martin, Neches and Wald5} Many children present initially with alternative diagnoses and are already receiving antibiotics.^{Reference Martin, Neches and Wald5,Reference Lewena10} Previous studies have failed to reveal specific initial symptoms in children that prompted the primary team to consider infective endocarditis as a possible diagnosis.^{Reference Kelchtermans, Grossar and Eyskens11} A discrepancy of more than 50% in the diagnosis of infective endocarditis on admission has been observed in paediatric populations.^{Reference Martin, Neches and Wald5,Reference Lewena10} Fever, a minor Duke’s criterion, is the only consistent clinical feature reported in retrospective studies of infective endocarditis in congenital heart disease children.^{Reference Martin, Neches and Wald5,Reference Lewena10,Reference Kelchtermans, Grossar and Eyskens11} Fever, however, is also the most common chief complaint for children younger than 15 years of age presenting to the emergency department and accounts for 15% of all visits per year.¹² Fever of short duration lasting less than 7 days in healthy children is usually due to self-limiting viral infections (common cold, gastroenteritis) or uncomplicated bacterial infections (otitis media, pharyngitis).^{Reference Dayal and Agarwal13} Fever without a focus in a paediatric congenital heart disease patient, however, must be approached more diligently than in healthy children as seen in our two patients.