Predictors and outcomes of peritoneal dialysis-related infections due to filamentous molds (MycoPDICS).

We sought to evaluate the predictors and outcomes of mold peritonitis in patients with peritoneal dialysis (PD). This cohort study included PD patients from the MycoPDICS database who had fungal peritonitis between July 2015-June 2020. Patient outcomes were analyzed by Kaplan Meier curves and the Log-rank test. Multivariable Cox proportional hazards model regression was used to estimating associations between fungal types and patients' outcomes. The study included 304 fungal peritonitis episodes (yeasts n = 129, hyaline molds n = 122, non-hyaline molds n = 44, and mixed fungi n = 9) in 303 patients. Fungal infections were common during the wet season (p <0.001). Mold peritonitis was significantly more frequent in patients with higher hemoglobin levels, presentations with catheter problems, and positive galactomannan (a fungal cell wall component) tests. Patient survival rates were lowest for non-hyaline mold peritonitis. A higher hazard of death was significantly associated with leaving the catheter in-situ (adjusted hazard ratio [HR] = 6.15, 95%confidence interval [CI]: 2.86-13.23) or delaying catheter removal after the diagnosis of fungal peritonitis (HR = 1.56, 95%CI: 1.00-2.44), as well as not receiving antifungal treatment (HR = 2.23, 95%CI: 1.25-4.01) or receiving it for less than 2 weeks (HR = 2.13, 95%CI: 1.33-3.43). Each additional day of antifungal therapy beyond the minimum 14-day duration was associated with a 2% lower risk of death (HR = 0.98, 95%CI: 0.95-0.999). Non-hyaline-mold peritonitis had worse survival. Longer duration and higher daily dosage of antifungal treatment were associated with better survival. Deviations from the 2016 ISPD Peritonitis Guideline recommendations concerning treatment duration and catheter removal timing were independently associated with higher mortality.

Déclaration de conflit d'intérêts

The authors have declared that no competing interests exist.