Reply to Fekkar et al.: Fungal Infection during COVID-19: Does Aspergillus Mean Secondary Invasive Aspergillosis?

From the Authors:

We thank Fekkar and colleagues for their thoughtful comments on our case series of patients with coronavirus disease (COVID-19)–associated pulmonary aspergillosis (CAPA) (1). The main points that are raised include the distinction between Aspergillus colonization and invasive infection and the subsequent classification. The presented cases are all classified as possible or probable CAPA, and none are histologically proven invasive aspergillosis, implying colonization without infection is a possibility.

First, we acknowledge the association between chronic obstructive pulmonary disease and Aspergillus spp. colonization. In our letter, two patients with chronic obstructive pulmonary disease are presented; here, we cannot rule out prior colonization. In addition, one of these patients received systemic corticosteroids for 2 days before admission, and one other patient was on a weaning scheme of oral corticosteroids preadmission for rheumatoid arthritis. A fourth patient with underlying bronchial asthma was treated with inhalation fluticasone for 1 month before admission. The use of corticosteroids is a known risk factor for colonization and invasive aspergillosis. However, the patients described in our series received a low daily dosage or a short duration of corticosteroids.

Cohort studies in patients with influenza-associated pulmonary aspergillosis (IAPA) in the ICU demonstrated that any indication of Aspergillus through positive culture or galactomannan (GM) detection is highly indicative of invasive aspergillosis (2). In this specific setting, a single positive test, such as serum GM or BAL GM, is considered sufficient to classify probable IAPA according to an expert panel. Both influenza and treatment with corticosteroids are considered risk factors for IAPA (3).

The direct microscopy findings of respiratory samples for fungal hyphae have no additional diagnostic value to the presented workup according to the latest criteria for IAPA (3). Nor can the criteria presented by Schauwvlieghe and colleagues rule out invasive pulmonary aspergillosis in the presented cases (2).

Furthermore, we would like to state that the start of antifungal therapy was always in multidisciplinary consultation on the basis of clinical deterioration and after reasonably excluding other causes. Indeed, a sole positive culture for Aspergillus might simply indicate colonization. However, rapid clinical deterioration with positive mycological evidence could not be ignored after the first cases of presumed CAPA with high mortality. We emphasize that starting antifungal therapy should always be considered in the context of the clinical status and in consultation with the attending ICU physicians. The overuse of antifungal treatment should be limited, because of adverse events such as liver and renal damage as well as the financial costs and the selection of resistant isolates.

The pathogenesis of COVID-19 is different from that of influenza, both regarding the tropism of the virus as well as the effect of the virus on (fungal) host defenses (4). As a consequence, the risk of invasive aspergillosis in patients with COVID-19 infection may be lower than in patients with influenza. Reports of presumed CAPA cases that survive without antifungal therapy, such as those presented by Fekkar and Alanio, are very informative and suggest that in patients with COVID-19, Aspergillus colonization is more common compared with in patients with influenza (5). In patients with COVID-19, additional factors, such as corticosteroid therapy, might contribute to an increased risk for developing invasive aspergillosis.

We agree with Fekkar and colleagues that a more stringent classification may be required for CAPA cases compared with existing classifications. Ultimate proof of CAPA can only be obtained through showing invasive hyphal growth in tissue samples. A recent case series included four proven CAPA cases, all of which were BAL culture and GM positive (6). However, all four cases were serum GM negative, underscoring the need for a better understanding of the pathophysiology of CAPA and the performance of diagnostic tests. Facing this uncertainty, in critically ill patients with COVID-19, the risk of further diagnostic testing, including bronchoscopy and/or lung biopsy, should be carefully weighed against delaying the initiation of antifungal therapy.

In conclusion, one mycological argument on a respiratory sample does not prove invasive aspergillosis. However, clinical deterioration in critically ill patients with COVID-19 that is not due to other causes, such as thromboembolic complications, inflammatory diseases, or secondary bacterial or viral infection, may indicate aspergillosis. However, radiological presentation can be atypical for invasive fungal disease in COVID-19 pneumonia, resembling influenza. The quantity of mycological arguments or the variety of assays is not decisive, although the quality of clinical specimens is conclusive for proving invasive aspergillosis.