INTRODUCTION
We describe a case where pulmonary oedema was initially mis-diagnosed and treated as pulmonary tuberculosis for 3 months. The patient was later found to have destructive fungal endocarditis.
CASE PRESENTATION
A 36-year-old, HIV-positive female patient presented to a peripheral hospital 3 months prior to admission to our institution with a 2-week history of respiratory symptoms. She was initially treated for pneumonia and due to her lack of clinical response, she was started on empiric anti- tuberculosis treatment despite having negative sputum for tuberculosis. As she was not on any antiretroviral therapy, she was started on combined antiretroviral therapy with a baseline CD4 count of 45 cells/mm3. A month post-discharge from hospital, she noticed that her legs had begun to swell and found that she was becoming short of breath.
She presented to our institution with signs and symptoms of left heart failure. She had a displaced volume loaded apex with peripheral signs of severe aortic regurgitation. Her auscultatory findings were that of a to-and-fro, machinery murmur at the base of the heart. At the time, she was apyrexial and had no clinical features of acute infective endocarditis (IE). Her inflammatory markers were not elevated and blood cultures (including for fungi) were all negative. Her transthoracic echocardiographic findings showed a thickened rheumatic aortic valve with a large (1.67 × 0.47 cm2) right coronary cusp vegetation resulting in severe aortic regurgitation. The parasternal short axis view revealed a similar sized pulmonary valve vegetation (1.79 × 1.12 cm2) with severe pulmonary regurgitation.
In search of an explanation for endocarditis of both the aortic and pulmonary valves, a trans-oesophageal echocardiograph (TEE) was performed. This confirmed endocarditis of both aortic and pulmonary valves (Figures 1 and 2). TEE also identified a communication between the aortic root and the right ventricle (Figure 3). This was confirmed in the cardiac catheterization laboratory, as contrast injected above the aortic valve filled both left and right ventricles. The diagnosis of chronic IE with a sinus-of-Valsalva fistula extending from the aorta to the right ventricle was made. Due to the refractory nature of the heart failure and severe haemodynamic lesions, she was referred to for surgery where a metallic aortic valve replacement together with a bio-prosthetic pulmonary valve replacement was performed. Post-operatively she was unstable, requiring ionotropic support and she then developed systemic inflammatory response syndrome. Her native valve cultures came back positive for Aspergillus fumigatus. Despite starting her on amphotericin B, our patient never showed satisfactory clinical response and demised 10 days post valve replacement from a newly acquired methicillin resistant Staphylococcus aureus infection.
Trans-oesophageal echocardiographic long axis view showing the large aortic valve vegetation. AO = Aorta; LV = left ventricle; LA = left atrium
DISCUSSION
The precise incidence of IE with HIV infection is difficult to ascertain because case definitions have varied over time between authors and between clinical centres. HIV-infection itself places one at a higher risk of developing IE, with reports in the literature of patients with a CD4 cell count <50 cells/mm3 and high viral loads (>100,000 copies/ml) having a 4-fold increased risk of developing IE.(1) Generally, endocarditis involving both the left and right sides of the heart simultaneously is uncommon and is predominantly described in the congenital heart disease group with an incidence of 2.3%.(2)
The incidence of fungal endocarditis has been reported to comprise 1%–10% of all IE cases. Candida accounts for almost half of all fungal cases and Aspergillus is the second most common (25%) offending organism.(3) Diagnosing fungal endocarditis is a challenge and poses a clinical dilemma as only 50% of blood cultures are positive for Candidiasis and cultures are rarely positive for Aspergillus species.(4) A review of the literature by Kalokhe et al. found that fungal endocarditis predominately affects the left side of the heart (72%) with right-sided endocarditis accounting for 19%. There were no reports of both left and right heart endocarditis in this review.(5)
Mortality of fungal endocarditis is more than 50% and treatment requires antifungal therapy and surgical valve replacement. The recommended antifungal therapy for invasive Aspergillus infections is voriconazole. The superiority of voriconazole to amphotericin B deoxycholate was demonstrated in a large, randomized controlled trial of invasive Aspergillus infections.(6) Compared to amphotericin B, voriconazole was associated with improved survival, and less nephrotoxicity, electrolyte abnormalities and infusion-related events.(6) Suppressive long-term therapy with oral azoles is recommended for life once fungal IE is acquired.
It is remarkable that such destructive endocarditis was silent for so long. It is possible that in our patient the fungemia was kept at bay by the Rifampicin that had been prescribed for her mis-diagnosed tuberculosis. Interestingly, Rifampicin has not only shown to have some antifungal activity but also offers synergistic effects to Amphotericin B, thereby becoming fungicidal rather than only being fungistatic.