Aspergillus Test Profiles and Mortality in Critically Ill COVID-19 Patients.

The literature regarding COVID-19-associated pulmonary aspergillosis (CAPA) has shown conflicting observations, including survival of CAPA patients not receiving antifungal therapy and discrepancy between CAPA diagnosis and autopsy findings. To gain insight into the pathophysiology of CAPA, we performed a case-control study in which we compared Aspergillus test profiles in CAPA patients and controls in relation to intensive care unit (ICU) mortality. This was a multinational case-control study in which Aspergillus test results, use of antifungal therapy, and mortality were collected from critically ill COVID-19 patients. Patients were classified using the 2020 European Confederation for Medical Mycology and the International Society for Human and Animal Mycology (ECMM/ISHAM) consensus case definitions. We analyzed 219 critically ill COVID-19 cases, including 1 proven, 38 probable, 19 possible CAPA cases, 21 Aspergillus-colonized patients, 7 patients only positive for serum (1,3)-β-d-glucan (BDG), and 133 cases with no evidence of CAPA. Mortality was 53.8% in CAPA patients compared to 24.1% in patients without CAPA ( = 0.001). Positive serum galactomannan (GM) and BDG were associated with increased mortality compared to serum biomarker-negative CAPA patients (87.5% versus 41.7%,  = 0.046; 90.0% versus 42.1%,  = 0.029, respectively). For each point increase in GM or 10-point BDG serum concentration, the odds of death increased (GM, odds ratio [OR] 10.208, 95% confidence interval [CI], 1.621 to 64.291,  = 0.013; BDG, OR, 1.247, 95% CI, 1.029 to 1.511,  = 0.024). CAPA is a complex disease, probably involving a continuum of respiratory colonization, tissue invasion, and angioinvasion. Serum biomarkers are useful for staging CAPA disease progression and, if positive, indicate angioinvasion and a high probability of mortality. There is need for a biomarker that distinguishes between respiratory tract colonization and tissue-invasive CAPA disease.