Case report: septic embolism through patent foramen ovale in mitral valve endocarditis

Abstract

Background

Infective endocarditis with septic emboli is a well-recognized sequala. However, emboli almost invariably are found ‘downstream’ to the vegetation. In the absence of congenital heart disease, the embolization from the left heart to the pulmonary circulation has never been described.

Case summary

A 4-year-old boy presents with a background history of pharyngitis treated with antibiotics. Upon ambulatory review, he is noted to have a new murmur. Transthoracic echocardiogram demonstrated mitral valve vegetation with severe mitral regurgitation; in addition, there was a patent foramen ovale (PFO), and there were no congenital heart defects. The patient was treated for infective endocarditis on high clinical suspicion. He subsequently developed septic pulmonary emboli in the absence of right-sided vegetation. Subsequent mitral valve vegetectomy, resection of infected native anterior mitral valve leaflet, mitral valve repair, and valvuloplasty. The patient made an excellent recovery following the completion of antibiotic therapy.

Discussion

Although not possible to confirm with certainty, this case demonstrates the most plausible explanation for this child's presentation being septic pulmonary emboli originating from left-sided heart vegetation migrating through a PFO.

Introduction

Septic emboli in infective endocarditis are a well-recognized phenomenon.^1,2 However, embolus from a left-sided source to a right-sided destination is incredibly rare in a structurally normal heart. We present a case of probable left-to-right embolism of septic emboli across a patent foramen ovale (PFO).

Summary figure

Case presentation

A 4-year-old boy presented to his local hospital with persistent high fever following 3 weeks of pharyngitis, initially treated with 7 days of erythromycin. Physical examination demonstrated inflamed tonsils, and blood tests demonstrated elevated C-reactive protein. Symptoms resolved following 24 h of co-amoxiclav; however, 3 days later at ambulatory review, a new pan-systolic murmur, located at the apex of the child’s heart with no radiation, was noted. Transthoracic echocardiogram (TTE) showed a large vegetation (11 × 14 mm) on the anterior leaflet of the mitral valve with suspected perforation and severe mitral regurgitation (MR) (Supplementary material online, Videos S1 and S3), with several jets directed both posterolaterally and anterolaterally thought to be from failure of coaptation and secondary to destruction of the anterior leaflet of the mitral valve. The heart was otherwise structurally normal with a small PFO with a left-to-right shunt. Based on modified Duke criteria for infective endocarditis, the patient did not fulfil the criteria for infective endocarditis; however, there was a strong clinical suspicion based on evidence of endocardial involvement and the presence of fever and the subsequent development of vascular phenomenon in septic pulmonary emboli. He was transferred to our institution after intravenous antibiotics (flucloxacillin, amoxicillin, and gentamicin) were empirically started. Upon admission and after microbiology consultation, antibiotics were converted to vancomycin (15 mg/kg TDS), rifampicin (10 mg/kg BD), and gentamicin (1 mg/kg BD), as blood cultures remained negative. Formal examination of his dentition was unremarkable. Multi-disciplinary meeting concluded that surgical management was required over conservative medical management given the high risk of embolization; however, further cross-sectional imaging would be required to evaluate any embolization.

On Day 12, the patient deteriorated clinically with increased work of breathing, tachypnoea, and recurrent temperature spikes. Chest X-ray demonstrated new focal lesions. Thoracic computed tomography (CT) revealed small peripheral foci of consolidation suggestive of septic pulmonary emboli and no other possible sources of left-to-right shunts (Figure 1). Subsequently, it was believed that the source of the septic emboli was from the vegetation on the mitral valve with left-to-right embolus across the PFO. The patient underwent urgent open-heart surgery on Day 13 of presentation and Day 9 of triple therapy antibiotics. Intraoperative transoesophageal echocardiogram (TOE) demonstrated the large vegetation of the mitral valve, the absence of right-sided vegetation, and the presence of a PFO (Figure 2, Supplementary material online, Videos S1–S3) On surgical inspection, there was destruction of the anterior mitral valve leaflet with gross vegetation and superimposition of thrombotic material on the vegetation. Mitral valve vegetectomy, resection of infected native anterior mitral valve leaflet, mitral valve repair and valvuloplasty replacing the body of the anterior mitral valve leaflet with mitral valve homograft, and re-implantation of the chordae to the head of the posterior mitral valve papillary muscle with PFO closure were performed. Although tissue culture from the resected valve was negative, 16 s rDNA sequencing PCR detected a sequence of Streptococcus pneumoniae, sensitive to penicillin.

The patient’s post-operative course was uneventful, and antibiotics were completed 6 weeks after initial presentation. At 12 months follow-up, his clinical examination was normal with only mild residual MR on TTE.

Discussion

Septic pulmonary emboli from left-heart endocarditis through a PFO have not previously been described. An individual case report outlining a case of mitral valve endocarditis migrating through a secundum atrial septal defect has been reported in an adult patient; however, this patient also had septic emboli to multiple systemic organs.³ A small series of patients experiencing septic pulmonary emboli has also been reported in the adult population; however, most patients in this series had an underlying cardiac condition such as ventricular septic defects or were in an immunocompromised state. Those that had no underlying medical conditions presented with right-sided cardiac foci with no emboli between the systemic and pulmonary circulations.⁴ Although it is not possible to say with certainty that the origin of the septic pulmonary emboli was from the mitral valve vegetation, it remains the most likely explanation. Alternative explanations would be that the bacterial burden was sufficiently high giving rise to a secondary infective source disseminating to the lungs such as tricuspid or pulmonary valve infective endocarditis that was then partially treated. However, the patient did not demonstrate any other localizing symptoms or signs prior to their deterioration. A further possibility may be that the septic pulmonary emboli were derived from the original pharyngitis when the bacterial burden was at its highest, although this does not correlate to the onset of respiratory deterioration on Day 12 of presentation. It is our opinion that in this case, septic emboli were likely facilitated by a high left atrial pressure in the setting of severe MR across the atrial septum. Interestingly, the direction of the MR jets was not directed towards the atrial septum. As PFO exists in 25% of the population, similar types of right-sided septic embolism may occur in a context of patients with mitral valve endocarditis and raised left atrial pressures. The reason why this is not a more common phenomena can only be speculated. However, the constellation of a large, mobile vegetation in relatively close proximity to the atrial septum, raised left atrial pressures, and several severe jets of MR reaching the extremities of the atrium and thereby possibly causing currents towards the atrial septum could explain the probable infective embolism.

Lead author biography

Dr Mohammad Abumehdi is a Congenital Interventional Cardiologist. Having trained in Glasgow, Birmingham and London, he is currently working in The Royal Brompton Hospital, London. His interventional interests are pulmonary artery and vein rehabilitation.

Supplementary material

Supplementary material is available at European Heart Journal – Case Reports online.

Acknowledgements

The authors thank Dr Rodney Franklin and Mrs Sheatha Abumehdi.

Consent: Consent was obtained on the premise of anonymized data; however, complete anonymity cannot be guaranteed due to the rarity of the case. Consent complies with the Committee on Publication Ethics (COPE) guidance.

Funding: None declared.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References

Author notes