Disseminated Histoplasmosis and Aortic Valve Infective Endocarditis in a Patient With Recent Aortobifemoral Bypass

Introduction

Fungal endocarditis constitutes 1% to 3% of all infective endocarditis (IE) cases, and fewer than 10% of fungal endocarditis cases are caused by Histoplasma spp. [1,2]. Histoplasma spp. are dimorphic fungi endemic to the Ohio and Mississippi River Valleys in the United States [3]. Infection occurs through inhalation of fungal spores found in soil contaminated by bird or bat droppings [4]. Although pulmonary involvement is most common, disseminated forms occur and can affect multiple organ systems, and, rarely, manifest as IE. In an analysis of 270 fungal IE cases over more than 30 years, only 16 cases (6%) were attributed to Histoplasma capsulatum, emphasizing the rarity of this condition [5].

The symptoms of Histoplasma endocarditis are nonspecific, which makes diagnosis challenging. Blood cultures frequently yield negative results, and the clinical presentation overlaps with that of bacterial endocarditis, including fever, heart murmurs, and evidence of valvular involvement [6]. Previous valvular disease is a major antecedent risk factor in immunocompetent individuals. Diagnostic evaluation for Histoplasma endocarditis is also challenging; it includes DNA probes targeting damaged endocardial tissue, echocardiography, radiological imaging, various cultures (bone marrow, valves, or tissue), serology, and molecular assays. Microbial cell-free DNA (mcfDNA) testing can detect pathogen-derived DNA fragments in biological specimens even when blood culture findings are negative. The current gold standard for diagnosis is recovery of Histoplasma spp. from a clinical sample [7], although false negatives occur in up to 15% of cases [8]. Standard treatment for Histoplasma endocarditis involves a combination of antifungal therapy and surgical valve replacement [9]. We encountered a highly unusual case that illustrates the complex interplay of multiple comorbidities, including recent vascular surgery and reactivation of asymptomatic histoplasmosis with dissemination to the aortic valve. The following report presents a presumptive case of Histoplasma endocarditis, outlines the diagnostic uncertainties, and discusses implications for clinical practice. This case also highlights the role of advanced molecular diagnostics in identifying H. capsulatum when routine test results are inconclusive.

Case Report

A 59-year-old woman was admitted with the chief complaint of sudden onset of severe nausea and vomiting. Abdominal computed tomography (CT) revealed a suspected 4.4-cm loculated fluid collection in the anterior pelvis, suggestive of an abscess (Figure 1). The patient’s family and social histories were noncontributory. Her medical history included a successful aortobifemoral bypass for aortoiliac occlusive disease, performed 2 months earlier. This procedure was complicated by severe thrombocytopenia, splenomegaly, and recurrent retroperitoneal bleeding, which required multiple exploratory laparotomies and a splenectomy. She had also been diagnosed with sick sinus syndrome around the time of her bypass, leading to implantation of a Micra® leadless pacemaker that same month. At age 39, she experienced pregnancy-related immune thrombocytopenia but delivered a healthy infant at term. She had never received intravenous immunoglobulin or corticosteroid therapy, and she was informed that her platelet count had normalized. Additionally, her medical history included gangrene of the right hallux secondary to atherosclerotic disease, several months before her current presentation. She also had a remote history of untreated intermittent atrial fibrillation with a negative diagnostic workup. She denied any history of recurrent acute pulmonary infections.

On admission, the patient denied shortness of breath, weight loss, night sweats, orthopnea, or chest pain. She reported no joint pain, dermatologic changes, or splinter hemorrhages. She had no seizure activity; neurological examination findings of claudication and weakness were consistent with her regional thrombosis. Her blood pressure was 95/75 mmHg and pulse was 101 beats/min; she was afebrile and had a midline surgical scar with a left upper quadrant Jackson-Pratt drain. Her abdomen was guarded but not rigid. A loud systolic murmur was present on chest auscultation. Blood cultures were drawn, and laboratory studies were performed. The patient did not meet systemic inflammatory response syndrome criteria, although her initial blood work was notable for leukocytosis (white blood cell count 25,200 cells/μL; reference range, 4500–11 000 cells/μL), 1.1 mmol/L lactic acid (reference range, 0.5–2.0 mmol/L), and 52 U/L aspartate aminotransferase (reference range, 5–34 U/L), along with hypotension. Abdominal CT revealed complete occlusion of her recent bypass (Figure 2). Chest radiography showed no granulomas or adenopathy.

Because her presentation was consistent with severe infection, the patient was administered empiric antibiotics: piperacillin–tazobactam 4.5 g in 0.9% NaCl 100 mL intravenously (IV) piggyback and vancomycin 1000 mg in 0.9% NaCl 250 mL IV piggyback. She subsequently underwent interventional radiology–guided aspiration of the abdominal fluid collection, but culture results were negative. Transthoracic echocardiography revealed a large aortic valve vegetation, raising concern for IE (Figure 3). At this point, the patient was monitored off antibiotics in anticipation of an endocarditis workup. Blood cultures remained unrevealing. mcfDNA testing later returned a high positive molecules-per-microliter result for H. capsulatum and low-level Streptococcus mitis/oralis. Repeat mcfDNA testing 4 days later yielded the same findings, increasing concern for concomitant H. capsulatum and S. mitis/oralis IE. Histoplasma antibody complement fixation and immunodiffusion tests demonstrated negative results. A Histoplasma/Blastomyces antigen assay showed positive findings; antinuclear antibody test results were negative. Due to evidence of coinfection, the patient was administered ceftriaxone 2 g IV in sterile water for 4 weeks, along with liposomal amphotericin B 5 mg/kg for 14 days. To evaluate potential disseminated histoplasmosis, a bone marrow biopsy was performed, and samples were sent for polymerase chain reaction analysis; results were unrevealing.

Panorex imaging revealed dental disease, which may explain the source of S. mitis/oralis. The patient underwent extraction of root tips 3, 14, and 20, as well as alveoplasty. The source of H. capsulatum remains uncertain. The patient is a native Floridian and denied exposure to exotic pets, stating that she owns only a cat. Notably, she reported remote travel to Indiana (i.e., within the Ohio River Valley region where Histoplasma spp. are endemic) in 1999 and 2000, during which she visited family and stayed for at least 6 months on each trip.

Ultimately, the patient underwent aortic valve replacement with root enlargement, placement of a 21-mm Inspiris® valve, and left atrial appendage ligation using a 45-mm Atriclip®. She was extensively counseled on the advantages and disadvantages of biological versus mechanical valves; she selected a bioprosthetic valve because it did not require use of warfarin. Her height was 1.57 m and weight was 61 kg (body surface area, 1.6 m2). Intraoperatively, a very small aorta and sinotubular junction were identified. The barrel end of a 19-mm Inspiris® valve sizer would not pass through the sinotubular junction. Therefore, root enlargement was performed to facilitate placement of a 21-mm Inspiris® valve, avoiding patient–prosthesis mismatch and preserving future transcatheter options. One month after the operation, transthoracic echocardiography showed a normal mean gradient of 9 mmHg across the bioprosthetic aortic valve. Vascular surgery subsequently corrected the complete occlusion of her bilateral femoral arterial tree.

Pathology of the aortic valve tested positive according to Grocott methenamine silver staining, revealing fungal elements morphologically compatible with Histoplasma. Gram staining and acid-fast bacilli staining of the valve showed negative results, excluding bacterial and mycobacterial organisms. Fungal, acid-fast bacilli, aerobic, and anaerobic cultures – and their respective stains – displayed negative findings (Figure 4).

Two weeks after valve replacement, the patient was transitioned from liposomal amphotericin B to itraconazole. The itraconazole regimen included a loading dose of 200 mg 3 times daily for 3 days, followed by a maintenance dose of 200 mg every 12 h for 6 months. She also received 4 weeks of ceftriaxone 2 g in sterile water via IV syringe. Since discharge, the patient has continued outpatient follow-up with our clinic and remains on itraconazole therapy.

Discussion

This case is notable for 2 reasons. First, it describes an extremely complex disease process characterized by Histoplasma endocarditis in a patient with severe atherosclerotic disease requiring vascular repair. Generally, these conditions do not occur concurrently unless there are elements of immune compromise [10]. However, 3 prior case reports suggest that progressive disseminated histoplasmosis can be an antecedent risk factor for atherosclerotic peripheral vascular disease [11]. The reverse association has not been documented; it remains uncertain whether atherosclerotic peripheral vascular disease is a risk factor for symptomatic histoplasmosis. An additional complicating factor in the present case was the evidence of 2 infective microorganisms identified as potential IE-causing agents. Only 20% of IE cases are polymicrobial; when 1 of those pathogens is a fungus, that proportion is even lower [12]. Histoplasma IE is more common in individuals with prosthetic valves and in those who are immunocompromised [13]. Our patient had neither of these conditions. Nonetheless, the detection of mcfDNA identifying S. mitis/oralis does not confirm Streptococcus spp. as a coinfecting agent on her native aortic valve. Depending on various factors, a positive Gram stain result for the valve has a high true-positive rate [14]; in the present case, Gram staining of the resected aortic valve revealed negative findings. In contrast, fungal staining results were positive. The microbial landscape of IE is evolving. A recent retrospective study of 222 patients with IE showed a substantial increase in Staphylococcus spp. and a decrease in Streptococcus spp.; Staphylococcus spp. were associated with severe intracardiac involvement [15]. Coinfection with fungal and bacterial organisms is extremely rare in IE, particularly among patients without predisposing conditions such as prosthetic valves or immunocompromise. Diagnosis of fungal–bacterial coinfection is challenging because presentations may be atypical, with overlapping symptoms and differing roles of opportunistic versus endemic pathogens. A 65-year-old patient with combined Enterococcus and Aspergillus species IE was described in 2018, demonstrating the complexity and severity of such coinfections, which require multifaceted treatment approaches [16]. A second notable aspect of the present case is the timeline of disease progression from initial inoculation to end-stage aortic valve destruction. The patient was likely exposed 25 years earlier and remained asymptomatic, unless her atherosclerotic peripheral vascular disease represented a complication of histoplasmosis. However, recent inoculation followed by acute disseminated histoplasmosis leading to rapid aortic valve destruction is also plausible. Histoplasma spp. likely constitute the most common invasive fungal pathogens in humans [10]. Clinical manifestations range from asymptomatic infection to acute histoplasmosis, which typically manifests as pneumonia. The disease also may manifest as chronic pulmonary histoplasmosis or disseminated infection that involves other organ systems, including the endovascular system, central nervous system, gastrointestinal tract, and adrenal glands [17]. The clinical presentation depends on fungal virulence, inoculum size, and host immune status [18]. Our patient had no antecedent congenital or acquired valvular disease and was immunocompetent.

Disease in an asymptomatic individual may remain undetected for decades; depending on immune status, reactivation can occur at any time [19]. The timeline of progression from reactivation to dissemination to cardiac valve involvement is not well established. One feature of Histoplasma endocarditis is known: destruction of the aortic valve can occur rapidly, often within weeks if untreated [20].

The diagnosis of histoplasmosis is challenging despite the availability of multiple diagnostic modalities. The organism is fastidious and slow-growing, and the acquisition of a positive culture – the gold standard – may require weeks [4]. Diagnosing isolated Histoplasma aortic valve vegetation in an immunocompetent individual with dental decay and a concomitant vascular emergency is even more difficult for 2 reasons. First, Histoplasma spp. are not typical pathogens in IE. Second, much of the immediate clinical attention is directed toward the thrombotic catastrophe. Transesophageal echocardiography confirmed aortic valve vegetation, and the mcfDNA findings prompted initiation of antifungal therapy and surgical planning. Ultimately, histopathologic examination of the aortic valve revealed granulomatous inflammation and Grocott methenamine silver–stained fungal elements morphologically compatible with Histoplasma. Nevertheless, the limitations of mcfDNA testing in real-world clinical practice, the patient’s negative histoplasmosis cultures, and the low-level mcfDNA detection of S. mitis/oralis indicate possible mixed IE, rather than isolated histoplasmosis.

Treatment of Histoplasma IE includes surgical replacement of the malfunctioning valve and prolonged antifungal therapy. The patient now has an artificial valve and therefore exhibits increased risk for subsequent infections with various microorganisms. Post-treatment reactivation of histoplasmosis can occur years later, typically due to altered immune status or inadequate initial therapy [21]. Artificial heart valves are a known risk factor for Histoplasma IE, and our patient now has a prosthetic aortic valve. It is unknown whether this prosthetic valve increases her risk for recurrent Histoplasma IE in a manner that warrants modification of her long-term treatment plan. Lifelong antifungal therapy has been described in susceptible patients [21]. At present, no reliable objective screening methods exist to predict which post-infection patients will develop metastatic fungal implantation on a heart valve. In the present case, such screening would not have altered management because the patient’s initial presentation was Histoplasma IE.

Conclusions

Histoplasmosis is an insidious infectious disorder that can arise suddenly and unexpectedly, resulting in substantial morbidity. The present case most likely represents reactivated disease with dissemination to the aortic valve, an occurrence that is extremely rare in immunocompetent individuals with native heart valves. An unusual feature in this case was the absence of fever or signs of heart failure on presentation. Using currently available diagnostic technology (e.g., transesophageal echocardiography and mcfDNA testing), a presymptomatic presumptive diagnosis was established, and a treatment plan was formulated. Diagnostic ambiguity remained, and the possibility of a polymicrobial etiology persisted due to the positive mcfDNA result for S. mitis/oralis.

Further studies addressing reactivation of asymptomatic histoplasmosis in the setting of concurrent serious health events are needed, and stronger pathological confirmation should be obtained in future cases.