Blastomycosis with rapid-onset acute respiratory distress syndrome in an urban setting

Abstract

While blastomycosis is endemic to eastern USA and northwestern Ontario, acquisition is an anomaly in urban settings. We present a 54-year-old immunocompetent man from the greater Toronto area with no travel, who presented with a 3-week history of chest pain and dyspnoea. Initial radiographic workup revealed a mass-like opacification in the right apical mediastinum. Extensive investigations including bronchoscopy with bronchoalveolar lavage, mediastinal mass biopsy with fungal and mycobacterial cultures and multiple stains, and CT were unrevealing. The patient progressed to respiratory failure over 4 months. Ultimately, sputum and bone marrow cultures confirmed a diagnosis of disseminated blastomycosis. The patient required prolonged extracorporeal membrane oxygenation and ongoing ventilation postdecannulation. Our case highlights diagnostic challenges with blastomycosis, particularly in immunocompetent individuals with no travel to recreational areas, and emphasises the importance of maintaining a high index of suspicion and sending fungal cultures of appropriate specimens and/or cytopathology in clinically compatible cases.

Background

Blastomyces dermatitidis is one of several medically important dimorphic fungi endemic to North America, and typically causes the clinical manifestations of blastomycosis. Fungal spores known as conidia are found in soil, and can become airborne as a result of excavation or construction. In the absence of prompt phagocytosis, 48–72 hours after inhalation of conidia there is transformation into a yeast phase in the lungs through a temperature-mediated process.¹

In Canada, disease due to B. dermatitidis has been described in four provinces: Saskatchewan, Manitoba, Ontario and Quebec.² The incidence is particularly high in northwestern Ontario where many individuals have recreational properties in ‘cottage country’,^{2 3} and the Kenora region has the highest reported incidence of blastomycosis in the world.⁴ The average incubation rate of blastomycosis is 30–45 days, though has been known to extend up to 106 days.³

Blastomycosis is asymptomatic in at least 50% of cases, and when clinically apparent, the lungs are involved in up to 80% of cases. The condition most commonly presents as isolated pulmonary disease,³ with non-specific presentations that range from acute lobar consolidation to chronic cavitation. Disease severity is also variable, ranging from asymptomatic to acute respiratory distress syndrome (ARDS). Approximately 15% of patients with pulmonary blastomycosis develop ARDS, with an associated fatality rate ranging between 50% and 89%.⁵ Extrapulmonary manifestations may include cutaneous blastomycosis, or dissemination to the skeletal, urogenital or central nervous systems,³ and such findings may increase clinical suspicion of blastomycosis.

Here, we describe a case of pulmonary blastomycosis complicated by disseminated disease and ARDS acquired in an urban region of Ontario with no obvious environmental exposures. This case serves to highlight the potential difficulties with diagnosis particularly in immunocompetent patients from non-endemic regions, the importance of a high index of suspicion in clinically compatible cases and the need to specifically consider fungal cultures of appropriate specimens and/or cytopathology when a diagnosis of blastomycosis may be possible.

Case presentation

A 54-year-old man from Oshawa, Ontario, presented to hospital with a 3-week history of chest pain and dyspnoea. He was previously employed with a printing company and was functionally independent at baseline. He had no recent occupational or recreational exposures to areas with decaying vegetation, and no recent engagement with activities such as hiking, camping or gardening. His only known comorbidity was well-controlled type 2 diabetes mellitus, controlled with oral agents (metformin and gliclazide) and insulin. An initial chest X-ray in the emergency department revealed a mass-like opacification in the right apical mediastinum. This was confirmed on computed tomography (figure 1). He subsequently underwent a bronchoscopy with a broncheoalveolar lavage, which did not demonstrate malignant cells or mycobacteria. Bacterial cultures demonstrated normal respiratory flora. One month later, the patient had a biopsy taken of the mediastinal mass (figure 2), which demonstrated granulomatous inflammation with aggregations of epithelioid and multinucleated giant cells with small foci of necrosis. The differential diagnosis provided by the pathology department was infectious (mycobacterial or fungal) versus non-infectious (sarcoid). There was no evidence of tuberculosis or malignancy, and no fungal or mycobacterial organisms were seen on Grocott’s methenamine silver, periodic acid-Schiff or Ziehl-Neelsen stains. Over the subsequent 4 months, the patient had multiple visits to the emergency department for progressive dyspnoea, chest pain, weight loss, non-productive cough and night sweats. At no point were steroids or other immunosuppression initiated for the patient.

Figure 1.

Mass-like opacification in the right apical mediastinum, confirmed on CT of the thorax on initial presentation.

Figure 2.

CT-guided needle biopsy of the right apical mediastinal mass.

Four months after the patient’s initial presentation, he represented to hospital with significant respiratory distress and was found to be hypoxic and tachypneic. He was admitted and underwent CT of the thorax, which revealed nodular infiltrates with medial bronchiectasis in the right upper lobe. Two days after his admission, the patient had an acute deterioration in his respiratory status and a repeat bronchoscopy was performed. Initial bronchoscopy results did not initially yield a diagnosis and the patient was intubated to facilitate an open lung biopsy. Immediately after intubation, the patient was extremely difficult to ventilate, and required transfer to a quaternary care centre for initiation of venovenous extracorporeal membrane oxygenation (ECMO). Subsequently, the patient’s bronchoscopy culture grew B. dermatitidis and serology returned positive with a 1:4 titre. He was started on liposomal amphotericin B at this time. He also underwent a bone marrow biopsy on admission that demonstrated bone marrow involvement, confirming disseminated blastomycosis.

Outcome and follow-up

The patient remained on ECMO for a total of 101 days (CT thorax imaging from admission shown in figure 3). His course was complicated by methicillin-sensitive Staphylococcus aureus bacteraemia, for which he was treated with cloxacillin for 4 weeks and improved. He slowly improved from a respiratory perspective and was decannulated after 101 days on ECMO. The patient remained ventilated after decannulation from ECMO and was transferred to a peripheral hospital for ongoing care and treatment with liposomal amphotericin B.

Figure 3.

CT of the chest on day of transfer to quaternary hospital for venovenous extracorporeal membrane oxygenation initiation.

Discussion

The burden of blastomycosis in Ontario is largely in northwestern regions, where 0.6% of Ontario’s population accounts for 42% of documented blastomycosis hospitalisations, with an average hospitalisation rate of 35.0 per 100 000 per year. Significant geographical variation has been reported across regions, with hospitalisation rates ranging from 1.7 in the Red Lake region to 57.9 per 100 000 per year in the Kenora region. Evidence has suggested that blastomycosis exposure in the spring and summer seasons, followed by a long incubation period, is the most likely predisposing aetiology for the majority of hospitalisations seen in late fall.⁴

The incidence of blastomycosis in Ontario has risen over several decades. Between 1981 and 1989, there were two reported cases per year; in comparison, between 2004 and 2014, the incidence increased to 6.4 cases per year. Case fatality has also been shown to be increasing, varying from 6.3% between 1988 and 1999, to 20.3% between 2004 and 2014 across Manitoba and northwestern Ontario.^{4 6 7} This may relate to increased awareness, detection and reporting of blastomycosis, and expanding ecological niches of Blastomyces in Canada. With climate change, it is projected that the summers in North America will become drier while the winters will be characterised by increased precipitation. These predicted changes are expected to produce optimal conditions for Blastomyces spores to be dispersed.⁴

Our case is unique, first and foremost, in terms of the absence of typical risk factors or characteristic geographic exposures for blastomycosis acquisition. Blastomyces spp classically inhabit a specific ecological niche consisting of soil with an acidic pH, typically due to decaying vegetation or organic material.² Our patient had no apparent occupational or recreational exposures to such environments. Specifically, he did not engage in activities such as camping, hiking or gardening. He resided in Oshawa, Ontario, an urban city on Lake Ontario, 60 km east of Toronto. Despite this, he was admitted and required a prolonged course of ECMO for ARDS, with evidence of disseminated disease. Therefore, clinicians should be aware that Blastomyces infection may cause disease within individuals without traditional travel histories or environmental exposures; when other clinical correlates are suggestive of blastomycosis as a possible aetiology, appropriate fungal cultures and/or cytopathology should be included early in the diagnostic workup.

It is also important to note that our patient developed disseminated blastomycosis in the absence of clear immunocompromise. While he had a history of diabetes, his disease was very well controlled with oral antihyperglycaemics and insulin. Review of his medical history and medications did not reveal any other immunosuppressive comorbidities, nor did the patient receive any immunosuppressive medications. As such, this case adds to existing literature supporting that blastomycosis can cause severe disseminated disease in relatively immunocompetent people, and should be included in the differential diagnosis if other clinical correlates are suggestive.

Perhaps most importantly, this individual was subject to multiple medical assessments and extensive clinical workup prior to the diagnosis of disseminated blastomycosis, close to 5 months after symptom onset. This highlights the considerable difficulty in making a diagnosis of blastomycosis unless the condition is specifically included in the initial differential diagnosis, which is often not the case in a non-endemic area due to its non-specific presentation. However, if considered the diagnosis may be quite straightforward, with a yield as high as 75% from single sputum samples when sent for fungal culture, and of over 90% for specimens obtained during bronchoscopy.⁸ Although direct visualisation of Blastomyces in fluid or tissues sample is uncommon, when seen this may lead to a more rapid presumptive diagnosis of blastomycosis. H&E stains often demonstrate an inflammatory, pyogranulomatous response in blastomycosis, but fungal elements are difficult to visualise⁵ and the diagnostic yield of histopathology is reportedly 36%.⁸ In our patient’s case, pyogranulomatous tissue was demonstrated on the biopsy of the mediastinal mass. While this did raise the question of a possible fungal infection, histopathology of this initial biopsy, which included multiple fungal stains, was non-diagnostic. Other testing modalities, such as serological testing and antigen detection, have poor specificity and sensitivity.⁵ Nucleic acid testing techniques are currently being developed for clinical use, and PCR appears to have high sensitivity and specificity when applied to culture isolates and/or direct clinical specimens.⁵

Learning points.

• Blastomycosis is a frequent cause of hospitalisation in northwestern Ontario, but may also be seen in urban settings not classically endemic for the fungus.

• Blastomycosis may be associated with severe respiratory decompensation and disseminated disease, even in relatively immunocompetent patients.

• Where suspicious clinical features such as pulmonary granulomatous disease are present, clinicians should consider blastomycosis on their differential diagnosis, even in the absence of clear environmental exposure risk.