Management, Outcomes, and Predictors of Mortality of Cryptococcus Infection in Patients Without HIV: A Multicenter Study in 46 Hospitals in Australia and New Zealand.

Abstract

Background: Limited data exist regarding outcomes of cryptococcosis in patients without human immunodeficiency virus (HIV), and few studies have compared outcomes of Cryptococcus gattii versus Cryptococcus neoformans infection.

Methods: We conducted a retrospective study in 46 Australian and New Zealand hospitals to determine the outcomes of cryptococcosis in patients without HIV diagnosed between 2015 and 2019 and compared outcomes of C. gattii versus C. neoformans infections. Multivariable analysis identified predictors of mortality within 1 year.

Results: Of 426 patients, 1-year all-cause mortality was 21%. Cryptococcus gattii infection was associated with lower mortality than C. neoformans (adjusted odds ratio [OR], 0.47; 95% confidence interval [CI], .23-.95), while severe neurological symptoms at presentation were the strongest predictor of death (adjusted OR, 8.46; 95% CI, 2.99-23.98). Almost all (99.5%) patients with central nervous system (CNS) infection received induction antifungal therapy versus 27.7% with isolated pulmonary cryptococcosis. The most common regimen in CNS disease was liposomal amphotericin B with flucytosine (93.8%; mean duration, 31 ± 13 days). Among patients with CNS cryptococcosis, C. gattii infection was associated with higher risk of immune reconstitution inflammatory response (C-IRIS) than C. neoformans (21% versus 3%, P < .001). Nineteen patients received amphotericin B-based re-induction therapy for suspected relapse, but none had microbiological relapse. Serum cryptococcal antigen positivity and lung imaging abnormalities resolved slowly (resolution at 1 year in 25% and 34% of patients, respectively).

Conclusions: Compared with C. neoformans, C. gattii infection demonstrated lower mortality but higher C-IRIS risk in CNS infection. Severe neurological symptoms were the strongest predictor of mortality.