A Prospective Surveillance Study of Candidaemia: Epidemiology, Risk Factors, Antifungal Treatment and Outcome in Hospitalized Patients

Abstract

This study provide an up-to-date overview of the epidemiology and risk factors for Candida bloodstream infection in Scotland in 2012/2013, and the antifungal susceptibility of isolates from blood cultures from 11 National Health Service boards within Scotland. Candida isolates were identified by chromogenic agar and confirmed by MALDI-TOF methods. Survival and associated risk factors for patients stratified as albicans and non-albicans cases were assessed. Information on the spectrum of antifungals used was collected and summarized. The isolates sensitivity to different antifungals was tested by broth microdilution method and interpreted according to CLSI/EUCAST guidelines. Forty one percent of candidaemia cases were associated with Candida albicans, followed by C. glabrata (35%), C. parapsilosis (11.5%), and remainder with other Candida spp. C. albicans and C. glabrata infections were associated with 20.9 and 16.3% mortality, respectively. Survival of patients with C. albicans was significantly lower compared to non-C. albicans and catheter line removal in C. albicans patients significantly increases the survival days. Predisposing factors such as total parenteral nutrition, and number of days on mechanical ventilation or in intensive care, were significantly associated with C. albicans infections. Fluconazole was used extensively (64.5%) for treating candidaemia cases followed by echinocandins (33.8%). Based on CLSI breakpoints, MIC test found no resistance to any antifungals tested except 5.26% fluconazole resistance among C. glabrata isolates. Moreover, by comparing to EUCAST breakpoints we found 3.95% of C. glabrata isolates were resistant to anidulafungin. We have observed a shift in Candida spp. with an increasing isolation of C. glabrata. Delay and choice of antifungal treatment are associated with poor clinical outcomes.

Keywords: Candida albicans; Candida glabrata; antifungals; candidaemia; drug resistance.

FIGURE 1

Association between patient mortality and the timing of antifungal treatment. The timing of antifungal therapy was determined to be before (BPS) or from the time when the first blood sample for different Candida culture positive was collected to the time when antifungal treatment was first administered to the patient. Number above each bar represents the overall % mortality over different antifungal timings and the black portion represent mortality associated with Candida albicans and white portion for NCAS infections.

FIGURE 2

Survival analysis for candidaemia patients. Patients with bloodstream infections (BSIs) were grouped by C. albicans (n = 52) or non-C. albicans (NCAS) spp. (n = 72) and censored at death, or day 30. Cox-regression plots adjusted for patient age is shown. Comparison between these curves found a statistically significant difference in mortality rate (p < 0.05).

FIGURE 3

Days in the intensive care unit (ICU) or days in the ventilation before the extended prevalence of C. albicans, C. glabrata or other Candida spp. infection. Values in bracket indicate number of patients included in each group. ^∗C. albicans vs NCAS (p < 0.05).

FIGURE 4

Survival analysis for candidaemia patients with and without line removal. Cox-regression plots adjusted for patient age is shown (A) Patients with C. albicans infection (n = 24); p < 0.05, (B) patients with non-C. albicans infections (NCAS [n = 37]). p > 0.05 for C. albicans vs NCAS.