Clinical and mycological predictors of cryptococcosis-associated immune reconstitution inflammatory syndrome.

Objective: HIV-infected patients with treated cryptococcal meningitis are at risk for further neurological deterioration after commencing combination antiretroviral therapy (cART), mostly because of cryptococcosis-associated immune reconstitution inflammatory syndrome (C-IRIS). Identifying predictors of C-IRIS could enable risk stratification.

Design: Prospective, longitudinal cohort study for 24 weeks.

Setting: Durban, South Africa.

Participants: One hundred and thirty HIV-infected patients with first cryptococcal meningitis episode

Intervention: : Antifungal therapy (amphotericin 1 mg/kg median 14 days, followed by consolidation and maintenance fluconazole) and cART (commenced median of 18 days from cryptococcal meningitis diagnosis).

Main Outcome Measure: Clinical, blood, and cerebrospinal fluid (CSF) markers associated with C-IRIS before and during cART and clinical significance of CSF cryptococcal culture negativity pre-cART commencement.

Results: Of 106 patients commencing cART, 27 (25.5%) developed C-IRIS, 16 (15.1%) neurological deterioration-not C-IRIS, and 63 (59.4%) no neurological deterioration. On multivariable analysis, C-IRIS was associated with persistent CSF cryptococcal growth [hazard ratio (HR) 0.27, P=0.026] and lower CSF protein (HR 0.53, P=0.059) prior to cART and lower CD4 T-cell increases (HR 0.99, P=0.026) but not change in HIV viral load during cART. Using survival analysis, patients with a negative cryptococcal culture pre-cART commencement (n=51; 48.1%) experienced fewer episodes of neurological deterioration, C-IRIS, and cryptococcal relapse/persistence than patients with culture positivity (n=55; 51.9%, HR 0.33, 0.33, and 0.12 and P=0.0003, 0.0042, and 0.0004, respectively).

Conclusion: Persistent CSF cryptococcal growth at cART initiation and poor CD4 T-cell increases on cART are strong predictors of C-IRIS. Approaches aimed at achieving CSF culture negativity prior to cART should be evaluated as a strategy to reduce rates of C-IRIS.