Background

Filamentous basidiomycetes are mainly considered to be respiratory tract colonizers but increasingly these molds are being documented in invasive infections [1]. Hence, the clinical significance of their isolation in a specimen is debatable. Hormographiella aspergillata is a filamentous basidiomycete growing on horse dung. It was found in numerous environmental substrates and first reported as a human pathogen in 1971 [2,3,4]. Since, a few infections were reported all over the world with various clinical outcomes, essentially pulmonary but also disseminated or located to the eye or the skin [2, 5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]. Thus, data are sparse for the diagnosis and management of such infections. Here, we report a new case of human infection involving H. aspergillata and two cases of colonization. We then review all previously published cases and discuss diagnostic strategy and clinical management.

Case presentation

The first case (HA1) was an 70-year-old man admitted to the hematology department for prolonged febrile neutropenia and anorexia. He had a history of acute myeloid leukemia (AML) and hematopoietic stem cell transplantation (HSCT). His C-reactive protein (CRP, positivity threshold value: 3 mg/L) was 135 mg/L and empirical antibiotic therapy (ceftriaxone) was started at day 210 (D210, 7th month) post-HSCT. Chest computed tomography (CT) scan showed right lower lobe opacification (Fig. 1a) that had increased 1 week later (Fig. 1b). Invasive fungal infection (IFI) was suspected, and liposomal amphotericin B (lAmB 5 mg/kg/day) was started on D232 (7th month). Microscopic examination of a bronchoalveolar lavage (BAL) sampled at D237 (7th month) showed septate hyphae (Fig. 2) but cultures on Sabouraud media incubated at 25 °C and 35 °C were sterile after 7 days. H. aspergillata was identified by sequencing the internal transcribed spacer (ITS) region of fungi directly from the BAL. Interestingly, serum galactomannan monitoring was negative (< 0.1 on repeated samples; Platelia® Aspergillus assay, Bio-Rad; positivity threshold index: > 0.5) and β-D-glucan (Fungitell®, Cape Cod; positivity threshold value: 80 pg/mL) was weakly positive on D237 (7th month; 98 pg/mL) but negative on D248 (8th month; 46 pg/mL). In accordance with the 2008 European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) criteria, the patient was classified as having probable IFI [23]. His condition worsened following pulmonary Stenotrophomonas maltophilia infection and so it was decided to initiate palliative care. lAmB was stopped on D253 (8th month), 3 weeks after its introduction. The patient died on D298 (9th month).

Fig. 1
figure 1

Chest computed tomography scan of HA1 patient showing a right lower lobe opacification and b increase in the lesion size 1 week later

Full size image
Fig. 2
figure 2

Microscopy examination of bronchoalveolar lavage from patient HA1 by Gomori-Grocott staining showing the presence of septate hyphae. Scale-bar: 10 μm

Full size image

The second patient (HA2) was a 49-year-old man admitted to the intensive care unit for pneumopathy with acute respiratory failure. He had a history of psychiatric disorders, diabetes mellitus, asthma, smoking and middle cerebral artery stroke with persistent sequelae. CRP was negative on admission. The following day, it was positive at 108.0 mg/L but procalcitonin remained negative. Mechanical ventilation and empirical antibiotic therapy (ceftazidime) were initiated. A mucous plug containing purulent secretions in the left lung was removed by fibroscopy and transmitted to Bacteriology and Mycology Laboratories. Microscopy examinations of samples were negative but cultures identified oropharyngeal microbiota associated with a white mold on Sabouraud media at 25 °C and 35 °C after 7 days. Subcultures of mold grew with white to slightly cream-colored velvety colonies (Fig. 3a and b) on potato dextrose agar media. Microscopy examination of cultures showed hyaline septate hyphae with conidiophores producing cylindrical arthroconidia (Fig. 3c and d). H. aspergillata identification was confirmed by sequencing the ITS region. In vitro antifungal susceptibility testing was performed via broth microdilution technique according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines [24]. Minimum inhibitory concentrations (MICs) are given in Table 1. The chest CT scan was unremarkable and there was no risk factor for IFI and so no antifungal therapy was initiated. The inflammatory syndrome decreased rapidly 3 days later, and the patient’s condition improved. A putative diagnosis of bacterial aspiration pneumonia with fungal colonization was established.

Fig. 3
figure 3

Macroscopic and microscopic morphology of Hormographiella aspergillata on potato dextrose agar (PDA) subculture after 3 days of incubation at 25 °C. a White to cream colored velvety colonies with irregular margin on the recto side. b Verso side of the colonies showing light yellow color. c, d Slide culture of Hormographiella aspergillata showing hyaline septate hyphae with conidiophores and cylindrical arthroconidia without clamp connection, scale-bar: 200 μm (c) and 50 μm (d)

Full size image
Table 1 Antifungal susceptibility testing of Hormographiella aspergillata from the literature and our cases
Full size table

The third patient (HA3) was a 28-year-old woman admitted for investigation of an inflammatory disease affecting the central nervous system treated by methylprednisolone for 3 days (1 g/day). Bronchial fibroscopy was performed along with other investigations. Initial microscopy examination of the sample was negative but H. aspergillata grew after 3 weeks on Lowenstein-Jensen medium at 35 °C because of mycobacterial suspicion (identification confirmed by ITS sequencing). Antifungal susceptibility testing was performed as described above (Table 1). The patient was asymptomatic and her chest CT scan normal, suggesting colonization, and so no antifungal treatment was initiated.

Literature review

We reviewed the literature since 1971 to date using the terms “Hormographiella aspergillata” or “Coprinus cinereus” and “infection” in MEDLINE database (Tables 1 and 2). For each strain, antifungals MIC with the method used were reported in Table 1 when available. According to the 2008 EORTC/MSG criteria, all probable or proven IFI due to H. aspergillata were reported in Table 2 with significant clinical details.

Table 2 Literature review of Hormographiella aspergillata infections in humans published since 1971
Full size table

Discussion and conclusions

Hormographiella aspergillata is an environmental filamentous basidiomycete found in numerous substrates including soils, leaves, pressmud compost and in the air [3, 4]. It is the anamorph form of Coprinopsis cinerea (formerly Coprinus cinereus), which commonly grows on horse dung. It can be an opportunistic pathogen and is the second filamentous basidiomycete responsible for human infection after Schizophyllum commune [25]. To date, 22 invasive infections involving H. aspergillata have been reported (Table 2), mostly identified by sequencing of the 28S rDNA or ITS regions [2, 5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]. Most cases were diagnosed in Europe, but some were documented in the United States, Japan and India, in both rural and urban areas [2, 5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]. Infection cases occurred mainly in neutropenic patients. Although H. aspergillata is primarily responsible for pulmonary infections it can occasionally cause primary cutaneous lesions [10, 14, 19]. H. aspergillata is able to grow in blood cultures [11] and a few cases of disseminated infections have been reported, affecting the small intestine, the eye and the brain [11, 16, 17, 22]. Interestingly, three cases of IFI have also been reported in immunocompetent patients following cardiac or ophthalmic surgery [2, 9, 21]. The most contributive samples were biopsies, but some cases were diagnosed with BAL. [8, 15] H. aspergillata grows well on different fungal media without cycloheximide at 25 °C or 35 °C. However, diagnosis can be challenging in patients with negative cultures, as for the HA1 patient, whose strain was probably inhibited by the concomitant antifungal treatment. To date, there are insufficient data to draw any conclusions about biomarkers since in all documented reports galactomannan assays were negative and only two observations reported strongly positive β-D-glucan antigens greater than 500 pg/mL [18, 22]. We attempted to evaluate the production of galactomannan, β-D-glucan and glucuronoxylomannan antigens on in vitro cultures. Glucuronoxylomannan is a capsular antigen of Cryptococcus neoformans widely used to diagnose cryptococcosis. Some cross-reactions have already been described with other basidiomycete pathogens such as Trichosporon sp. or even Coprinopsis cinerea [26]. Interestingly, culture supernatants from strains HA2 and HA3 showed that H. aspergillata can produce galactomannan and β-D-glucan but not glucuronoxylomannan (Table 3). Although, as for HA1, results in sera are variable, biomarker assays could provide supplementary evidence in patients with suspected IFI.

Table 3 Galactomannan (GM), β-D-glucan and glucuronoxylomannan antigen assays on culture supernatant. For each strain, 5 to 10 colonies incubated at 35 °C for 4 days on Sabouraud media were suspended in 1 ml distilled water. After vigorous agitation, the suspensions were centrifuged for 5 min at 10,000 g. 1, 1:10 and 1:100 dilutions of the supernatants were then tested with Platelia® Aspergillus assay (Bio-Rad, France), Fungitell® assay (Associates of Cape Cod Inc., USA) and Biosynex® CryptoPS assay (Biosynex, France) according to the manufacturer’s recommendations
Full size table

H. aspergillata can also be a colonizer of the respiratory tract, as illustrated in our three patients, all of whom had an underlying respiratory condition. The weak clinical significance of the isolation of basidiomycetes in healthy subjects, in contrast with their life-threatening potential in immunocompromised patients, has already been described with Schizophyllum commune or Ceriporia lacerata, for example [27, 28]. These fungi are widely present in the environment, and their spores are easily inhaled and can grow in pulmonary alveoli in cases of local or systemic impaired function of alveolar macrophages.

As yet there are no EUCAST nor Clinical and Laboratory Standards Institute (CLSI) breakpoints to interpret the antifungal MICs for H. aspergillata. However, previous articles have reported in vitro resistance to echinocandins, fluconazole along with high MIC for flucytosine (Table 1). We found higher MICs for isavuconazole (4 and 16 mg/L) than what is usually observed for basidiomycetes [28, 29]. In the light of our findings and data from the literature, lAmB and voriconazole have the lowest MICs. However, H. aspergillata infections have a poor prognosis even when surgical debridement is performed.

In conclusion, on isolation of H. aspergillata, its pathogenic potential in clinical samples should be interpreted together with the patient’s history. Formal identification of the fungus can be tricky and usually requires molecular tools in addition to culture. Basidiomycetes can also be contaminants or colonizers and so microscopy examination of samples and/or histology in combination with biomarkers are crucial for diagnosis. Respiratory tract colonization is probably not uncommon given that the fungus is widespread in the environment but seems to be restricted to patients with underlying respiratory diseases. lAmB and voriconazole seem to be the antifungals of choice.