Incidence and demographic characteristics of patients
A total of 393 consecutive episodes of Candida BSI were collected, occurred in 370 inpatients, during an 11-year study period with an incidence of 0.39 episodes/1000 admissions, which has increased steadily, as shown in Figure 1. (0.21 episodes per 1,000 admissions in 2008, 0.24 in 2009, 0.26 in 2010, 0.44 in 2011, 0.42 in 2012, 0.37 in 2013, 0.40 in 2014, 0.45 in 2015, 0.47 in 2016, 0.59 in 2017 and 0.33 in 2018).
The demographic characteristics of the patients are summarized in Table 1. The mean age of patients was 57.6±19.0 years and 74.3% were male. Among 393 episodes, 148 episodes (37.7%) were from surgical ward, 167 episodes (42.5%) were from ICU and 78 episodes (19.8%) were from internal medicine ward, as shown in Figure 2.
C. albicans was isolated in 19.3% of cases in internal medical wards while in 41.2% and 38.9% of cases in surgery wards and ICU, respectively. In contrast, a higher proportion of C. tropicalis (34.7%) was found in internal medicine wards compared to that in surgery wards (21.6%) and ICUs (21.6%).
Underlying diseases and clinical features
The majority of patients with Candida BSI had at least one co-morbidity. 118 (30%) patients had solid tumors, 48 (12.2%) patients had hematological malignancies, 77 (19.6%) patients had diabetes mellitus, 124 (31.6%) patients had chronic cardiac disease, 52 (13.2%) patients had chronic pulmonary disease, 42 (10.7%) patients had chronic renal failure, 26 (6.6%) patients’ skin barrier has been considered compromised, 244 (62.1%) patients had prior surgical intervention, 54 (13.7%) patients with corticosteroid use, 88 (22.4%) patients had prior antifungal agents use, 255 (64.9%) patients received antibiotic prior Candida BSI onset. 244 (72%) patients had at least two co-morbidities. No patient had human immunodeficiency virus (HIV) infection. Regarding the severity, 309 (78.6%) patients had fever, 180 (45.8%) patients received parenteral nutrition, 147 (37.4%) patients received mechanical ventilation, 49 (12.5%) patients received renal replacement therapy, and 42 (10.7%) patients had neutropenia. The clinical characteristics of patients, by Candida species are shown in Table 2.
Candida species and antifungal susceptibility testing
393 Candida species were isolated in total. 141 (35.9%) of Candida BSI were C. albicans and 252 (64.1%) were non-albicans Candida. In non-albicans Candida sepsis cases, 87 (22.1%) were caused by C. parapsilosis, 69 (17.6%) by C. tropicalis, 48 (12.2%) by C. glabrata, 20 (5.1%) by C. guilliermondii, 8 (2.0%) by C. sake, 5 (1.3%) by C.krusei, and 15 by other species (4 C. gum, 3 C. lusitaniae, 2 C. intermedia, 2 C. theae, 2 C.lipolytica, 1 C. famataand 1 C. haemulonii).
Among 393 Candida species, there were 378 episodes with antifungal susceptibility testing results. According to CLSI breakpoints 2012 (CBPs), the rates of susceptibility of C. albicans, C. parapsilosis, C. tropicalis and C. parapsilosis to fluconazole and voriconazole were 94%, 89.5%, 52.2%, 0% and 93.3%, 91.9%, 61.2%, 95.8%, respectively. Their susceptibilities to itraconazole were worse (82.1%, 87.2%, 37.3%, 0%). Amphotericin B and 5-flucytosine remained superior to 95% susceptibility against common Candida spp., except for C. krusei and C. guilliermondii, which were only 20% and 47.4% susceptible to flucytosine, respectively (Table 4).
Antifungal therapy and outcome
Antifungal therapy was administered in 299 (76.1%) of the cases. 94 (23.9%) patients did not receive any antifungal treatment. Among the patients who received antifungal therapy, 247 (62.8%) received appropriate antifungal therapy, and 52 (13.2%) received delayed antifungal therapy (5 days after onset of Candida BSI). Fluconazole was most frequently used as empiric therapy, followed by echinocandins and voriconazole. 18 (4.6%) patients with Candida BSI received a combination therapy.
The overall, 30-day mortality rate was 28.5%. Mortality rate was significantly higher in internal medicine wards and in ICUs than in surgical wards (37.2% and 34.7% vs. 16.9%, respectively, P<0.001) (Figure 3a). The mortality for those who received appropriate or delayed antifungal therapy was 25.5% or 23.1% (P=0.012 or P=0.046), as compared to 39.4% for those who hadn’t receive any antifungal therapy. However, there was no significant difference between the mortalities for those who received appropriate antifungal therapy and for those who received delayed antifungal therapy (Figure 3a and 3b).
In univariate analysis, age, solid tumor, diabetes mellitus, chronic cardiac disease, chronic renal failure, skin disease, prior surgical intervention, mechanical ventilation, neutropenia, and antifungal therapy were associated with 30-day mortality. In multivariate Cox regression analysis, advanced age (HR=1.025; 95%CI, 1.013-1.037; P<0.001), chronic renal failure (HR=2.018; 95%CI 1.234-3.299; P=0.005), mechanical ventilation (HR=1.950; 95%CI 1.307-2.912; P=0.001), neutropenia (HR=4.347; 95%CI 2.462-7.675; P<0.001), were independent risk factors for 30-day mortality, while antifungal therapy (HR=0.570; 95%CI 0.382-0.849; P=0.006) was independent protective factor for 30-day mortality (Table 3).