Epidemiology and Infection Epidemiology, risk factors and outcomes of

cambridge.org/hyg
Candida albicans vs. non-albicans candidaemia
in adult patients in Northeast China
Wei Zhang, Xingpeng Song, Hao Wu and Rui Zheng
Original Paper
Cite this article: Zhang W, Song X, Wu H, Department of Respiratory Medicine, Shengjing Hospital of China Medical University, No. 36 Sanhao Street, Heping
Zheng R (2019). Epidemiology, risk factors and District, Shenyang, Liaoning, 110004, China
outcomes of Candida albicans vs. non-albicans
candidaemia in adult patients in Northeast Abstract
China. Epidemiology and Infection 147, e277,
1–8. https://doi.org/10.1017/ This study aimed to evaluate the clinical characteristics, risk factors and outcomes of adult
S0950268819001638 patients with candidaemia caused by C. albicans vs. non-albicans Candida spp. (NAC). All
adult hospitalised cases of candidaemia (2012–2017) at a tertiary hospital in Shenyang were
Received: 17 March 2019
Revised: 20 August 2019 included in the retrospective study, and a total of 180 episodes were analysed. C. parapsilosis
Accepted: 21 August 2019 was the most frequently isolated species (38.3%), followed by C. albicans (35.6%), C. glabrata
(13.9%), C. tropicalis (10%) and others (2.2%). As initial antifungal therapy, 75.0%, 3.9%, 5.6%
Key words: and 2.2% of patients received fluconazole, caspofungin, micafungin and voriconazole, respect-
Candida albicans; candidaemia; mortality;
non-albicans Candida; risk factor
ively. Multivariate analyses revealed that total parenteral nutrition was associated with an
increased risk of NAC bloodstream infections (BSI) (OR 2.535, 95% CI (1.066–6.026))
Author for correspondence: vs. C. albicans BSI. Additionally, the presence of a urinary catheter was associated with an
Rui Zheng, E-mail: [email protected] increased risk of C. albicans BSI (OR 2.295 (1.129–4.666)) vs. NAC BSI. Moreover, ICU
stay (OR 4.013 (1.476–10.906)), renal failure (OR 3.24 (1.084–9.683)), thrombocytopaenia
(OR 7.171 (2.152–23.892)) and C. albicans (OR 3.629 (1.352–9.743)) were independent risk
factors for candidaemia-related 30-day mortality, while recent cancer surgery was associated
with reduced mortality risk (OR 26.479 (2.550–274.918)). All these factors may provide useful
information to select initial empirical antifungal agents.

Introduction
Candida is an important causative organism of bloodstream infections (BSIs). Over the last
two decades, candidaemia has been reported as the fourth and seventh most common
healthcare-associated BSI in US and European studies, respectively [1, 2]. In a recent multi-
centre point-prevalence survey, Candida species emerged as the most common bloodstream
pathogen and accounted for up to 22% of healthcare-associated BSIs [3]. Additionally, candi-
daemia remains associated with high-mortality rates, prolonged hospital stays and increased
healthcare costs [1, 4–7]. Overall, mortality rates among patients range from 19.6% to 67%
worldwide [1, 6, 8–12] and the major risk factors for candidaemia include receipt of parenteral
nutrition, exposure to broad spectrum antibiotics, presence of central venous catheter (CVC),
prior surgery and ICU stay [13, 14]. Even though C. albicans overall accounts for the majority
of Candida spp. causing candidaemia, the proportion of non-albicans Candida (NAC) spp. is
rising and has even exceeded that of C. albicans in some recent studies from Europe, Asia and
America [8–10, 15–17].
In general, the most frequently isolated NAC spp. from candidaemia cases include C. glab-
rata, C. parapsilosis, C. tropicalis and C. krusei [5, 10, 16, 17]. Reduced in vitro susceptibility to
antifungal agents has been observed among several NAC spp., which may present a therapeutic
challenge. C. glabrata and C. krusei tend to be less susceptible to azole agents than other
Candida spp., while C. parapsilosis shows the highest minimum inhibitory concentrations
(MICs) to the echinocandins than other Candida spp. [18, 19]. Several retrospective studies
suggest that delayed initiation of antifungal therapy after the onset of candidaemia is asso-
© The Author(s) 2019. This is an Open Access ciated with increased mortality, which highlights the importance of early appropriate antifun-
article, distributed under the terms of the
gal therapy [20]. The local epidemiology and variable antifungal susceptibility profiles of
Creative Commons Attribution licence (http://
creativecommons.org/licenses/by/4.0/), which different Candida spp. are critical for the selection of antifungal therapy prior to culture
permits unrestricted re-use, distribution, and and susceptibility data being available.
reproduction in any medium, provided the However, risk factors related to NAC BSI, and the distribution and antifungal susceptibility
original work is properly cited. of Candida spp. from BSI differ geographically, whereas data on risk factors, antifungal sus-
ceptibility and outcomes in C. albicans and NAC BSIs remain scarce in Northeast China.
Therefore, this retrospective study was performed to compare clinical characteristics and out-
comes between C. albicans and NAC candidaemia, analyse prognostic factors and determine
risk factors related to C. albicans or NAC at a tertiary hospital in Shenyang.
2 Wei Zhang et al.

Patients and methods Statistical analysis

Subjects and study design Continuous variables were expressed as the mean ± standard devi-
ation in normally distributed data or as the median and interquar-
This retrospective observational study was undertaken from
tile range (25th–75th percentiles) in data with non-normal
January 2012 to October 2017 at a tertiary grade A comprehensive
distributions, respectively. Categorical variables were reported as
hospital in Shenyang, China. The setting is a China Medical
number (%). Continuous variables were compared using the
University-affiliated teaching hospital with over 6000 beds cur-
Student’s t-test or Mann–Whitney U-test according to the normal-
rently. All adult hospitalised cases of candidaemia were included
ity of distribution. Categorical variables were compared using the χ 2
in the study. The following data were collected from medical
test or χ 2 continuity correction test. Variables with P-values of
records: age, sex, major underlying diseases and predisposing fac-
⩽0.10 including age, sex, underlying diseases and predisposing fac-
tors, surgical and invasive procedures, complications, blood tests,
tors were entered into the multivariable logistic regression model to
isolated Candida spp., antifungal susceptibility, antifungal ther-
identify factors associated with NAC vs. C. albicans candidaemia.
apy, duration of hospitalisation and final outcome. The study
All variables with P-values of ⩽0.10 in the univariate analysis of fac-
was approved by the Medical Ethics Committee in Shengjing
tors associated with mortality were included in the multivariate
Hospital of China Medical University (reference number
logistic regression model to identify independent factors associated
2018PS506K).
with mortality. Two-sided P-values of <0.05 were considered statis-
tically significant. Statistical analyses were carried out with SPSS
(IBM SPSS Statistics 21.0).
Microbiological methods
Blood samples were drawn under sterile conditions and processed
using a BD BACTEC 9120 or BD BACTEC FX400 (Becton Results
Dickinson Diagnostic Instrument Systems, Sparks, USA) blood During the study period, 188 episodes of candidaemia in 187
culture system. After the identification of Candida isolates utilis- adult patients were initially identified. Two patients were excluded
ing the VITEK 2 YST (bioMérieux, Durham, USA) card, antifun- from further analysis because they had a single positive blood cul-
gal susceptibility testing of fluconazole, itraconazole, voriconazole ture for Candida spp. without signs and symptoms of infection.
and amphotericin B was determined using a ATB FUNGUS 3 One blood culture was considered to be contaminated due to sim-
(BIOMERIEUX) strip following the manufacturer’s instructions. ultaneous isolation of Staphylococcus epidermidis with C. glabrata.
The interpretation of antifungal susceptibility of fluconazole, itra- Five patients were lost to follow-up within 30 days from the onset
conazole, voriconazole and amphotericin B against C. albicans, C. of candidaemia and were therefore excluded. Thus, the remaining
glabrata, C. parapsilosis, C. tropicalis and C. krusei was performed 180 episodes of candidaemia from 179 adult patients were
by applying clinical breakpoints (CBPs) defined by the Clinical included in the study. Over the study period, 74 patients devel-
Laboratory Standards Institute (CLSI) [21] or European oped candidaemia in the first 3-year period (2012–2014) (74/
Committee on Antimicrobial Susceptibility Testing (EUCAST) 2845, 26.0/1000), which did not evidently differ from cases per
[22]. MIC CBPs for echinocandins and 5-fluorocytosine by the hospitalisation-days in the second 3-year period (2015–2017)
CLSI or EUCAST were not available at the time, and therefore (106/3769, 28.1/1000) (P = 0.611).
are not reported in the current study.

Distribution of Candida spp. causing bloodstream infection

Clinical definitions Among the 180 Candida isolates from blood cultures, C. parapsilosis
An episode of candidaemia was defined as the isolation of a was the predominant species (69/180, 38.3%), followed by
Candida spp. recovered from the first blood culture in a patient C. albicans (64/180, 35.6%), C. glabrata (25/180, 13.9%),
with accompanying clinical signs and symptoms. Subsequent C. tropicalis (18/180, 10%), C. krusei (2/180, 1.1%) and C. famata
positive cultures from the same patient were considered as a (2/180, 1.1%). Mixed infections with two or more Candida spp.
new episode if the interval between the two episodes was more were not observed. Figure 1 shows the distribution of the
than 30 days. The onset of candidaemia was taken as the day Candida spp. from candidaemia cases from 2012 to 2017. NAC
the first positive blood culture for Candida spp. was drawn accounted for the majority and there were no substantial changes
from the patient. Except for recent surgery, defined as y within regarding species distribution patterns between C. albicans and
3 months, the predisposing factors occurred within 30 days NAC over the study period. The proportion of C. parapsilosis BSI
prior to the onset of candidaemia. Elderly patients were defined in surgical patients evidently increased compared with non-surgical
as those 65 years of age or older. Laboratory test results were patients (59/138 (42.8%) vs. 10/42 (23.8%), P = 0.027) (Table 1).
obtained at the onset of candidaemia. Renal failure was defined
as a serum creatinine above 104 µmol/l. Anaemia was defined
Patient characteristics
as a haemoglobin level below 110 g/l for women and 130 g/l for
men. Hypoalbuminaemia was defined as albumin <30 g/l and The mean age of all episodes was 61.8 ± 15.9 years and 122/180
hyponatraemia as a serum sodium concentration <135 mmol/l. (67.8%) were male. C. albicans and NAC spp. were responsible
Leukopaenia was defined as a peripheral white blood cell count for 64/180 (35.6%) and 116/180 (64.4%) cases, respectively.
less than 4 × 109/l and thrombocytopaenia as a platelet count Among surgical patients, 138/180 (76.7%) episodes were docu-
<100 × 109/l. Antifungal therapy was defined as empirical if mented. Other common predisposing factors for candidaemia
started before susceptibility test results were obtained and as tar- included prior antibiotics exposure (163/180, 90.6%) and total
geted thereafter. The outcome was registered after 30 days from parenteral nutrition (TPN) (152/180, 84.4%). Nonetheless, there
the onset of candidaemia. were no statistical differences between C. albicans and NAC
Epidemiology and Infection 3

were prescribed initially in 7/180 (3.9%), 10/180 (5.6%) and 4/

180 (2.2%) of episodes, respectively. The overall 30-day mortality
rate was 25.6% and for patients receiving either targeted, empir-
ical, or no antifungal therapy, the 30-day mortality rates were
24.3% (9/37), 24.6% (29/118) and 32.0% (8/25), respectively,
with no statistical difference between the therapy groups (P =
0.728). However, the proportion of patients who had not received
antifungal therapy did not significantly differ for those who died
or survived within 30 days.
On univariate analysis (Table 5), a lower proportion of surgical
patients (P = 0.085), and an increased proportion receiving mech-
anical ventilation (P = 0.096) were observed in patients dying
within 30 days, although not statistically significant. By multivari-
ate regression analysis, four independent risk factors for death
were identified: ICU stay (OR 4.013; CI 1.476–10.906; P =
0.006), renal failure (OR 3.24; CI 1.084–9.683; P = 0.035),
Fig. 1. Distribution of Candida species isolates during the study period.
thrombocytopaenia (OR 7.171; CI 2.152–23.892; P = 0.001) and
the isolation of C. albicans (OR 3.629; CI 1.352–9.743; P =
0.011), while recent cancer surgery was associated with decreased
cases for the proportion of surgical patients and those who mortality (OR 0.038; CI 0.004–0.392; P = 0.006) (Table 6).
received prior antibiotics treatment. At the onset of candidaemia,
148/180 (82.2%) patients presented with anaemia. Moreover, the
overall low frequency of fluconazole resistance to Candida BSI Discussion
isolates (12/180, 6.7%) is noteworthy. The distribution of Candida spp. associated with candidaemia
The comparison of demographics and clinical data of patients varies greatly in different regions of the world. In recent years,
with C. albicans vs. NAC candidaemia is summarised in Table 2. some studies [8–10, 15–17] have observed a shift from the more
NAC-infected patients were more likely to have hyponatraemia antifungal susceptible type species C. albicans towards NAC
than those infected with C. albicans (50/116 (43.1%) vs. 18/64 spp., and this is corroborated by the findings of this 6-year
(28.1%), P = 0.047). Likewise, significant differences between study where NAC isolates accounted for the majority (64.4%) of
NAC and C. albicans patients with BSI were evident for the pres- all adult candidaemia episodes. Likewise, we found C. parapsilosis
ence of a urinary catheter (68/116; 58.6% and 47/64; 73.4% P = to be the most prevalent NAC species (38.3%) which accords with
0.048), the in vitro susceptibility of isolates to fluconazole (72/ a report from Mexico of 37.9% in 398 patients of all ages during a
116; 62.1% and 57/64; 89.1% P < 0.001) and the 30-day mortality 3-year surveillance program from five hospitals [23]. Moreover,
(23/116; 19.8% and 23/64; 35.9% P = 0.018). Other statistically there are several other reports of C. parapsilosis as the most fre-
significant elevated differences between the two groups were the quent NAC species (20–25%) from blood cultures in Spain [5,
frequency of current and former smokers (40/116; 34.5%, and 19], Italy [15], Latin America [24] and Shanghai, China [9, 25].
14/64; 21.9% P = 0.077) and TPN (102/116; 88%, and 50/64; The second most common NAC spp. in the current study was
78.1%, and P = 0.082). C. glabrata (13.9%), which has also been commonly recorded in
Multivariate analysis (Table 3) confirmed that TPN was signifi- studies from the USA [10], the UK [13], Greece [26], Beijing,
cantly associated with an increased risk of NAC candidaemia China [17] and Taiwan [27]. In contrast, C. tropicalis ranked as
(odds ratio (OR) 2.535; 95% confidence interval (CI) 1.066–6.026; the third most common in our series (10%) but was the most fre-
P = 0.035), whereas the presence of a urinary catheter was distinctly quent NAC isolate in other studies in China, notably Nanjing
associated with an increased risk of C. albicans infection (OR 2.295; [28], Chongqing [8], Shanghai [29], Shandong [30] and Taiwan
CI 1.129–4.666; P = 0.022). Current and former smokers were more [31]. The reasons for the varied species distribution and frequency
associated with NAC candidaemia (OR 1.824; CI 0.885–3.756; P = remain unclear but patient demographics and underlying medical
0.103), but the trend did not reach statistical significance. conditions are most likely contributory factors [16, 19]. The
higher frequency of C. parapsilosis BSI recorded here in surgical
Antifungal susceptibility testing patients clearly warrants further investigation.
Table 4 shows that of the 180 Candida isolates tested for antifun- Several possible risk factors have been previously identified
gal susceptibility, 12 (6.7%) exhibited resistance to fluconazole, for the development of NAC candidaemia and include prior
including C. albicans (2/64; 3.1%), C. parapsilosis (2/69; 2.9%), antifungal exposure, malignancy, immunosuppressive ther-
C. tropicalis (6/18; 33.3%) and both isolates of C. krusei. apy, abdominal surgery [32], presence of CVCs [26], artificial
Overall, voriconazole resistance was uncommon (10/180; 5.6%) surgical implants [17], head trauma and bacterial sepsis [29].
and except for one isolate of C. glabrata, all other Candida isolates Our finding of an association of an indwelling urinary cath-
were susceptible to amphotericin B. According to the species- eter and a higher risk of BSI caused by C. albicans vs. NAC
specific CBPs, all 25 isolates of C. glabrata showed susceptible- is consistent with other studies [33]. Multivariate analysis
dose dependence (SDD) to fluconazole. also confirmed that TPN was associated with an increased
risk of NAC spp., compared with C. albicans spp., but this is
in contrast with Chow et al. [34] who linked TPN with a
Therapy, outcomes and risk factors associated with mortality
decreased risk of BSI caused by NAC spp., vs. C. albicans. In
The majority of patients (135/180, 75.0%) received fluconazole as the current study, C. parapsilosis accounted for 59.5%, of all
first line treatment; caspofungin, micafungin and voriconazole NAC isolates whereas in Chow’s series [34], C. glabrata
4 Wei Zhang et al.

Table 1. Distribution of Candida species among patients according to prior surgery or history of solid tumours

Species Non-surgical patients (42) Surgical patients (138) P Non-solid tumours (100) Solid tumours (80) P

C. albicans 19 (45.2%) 45 (32.6%) 0.134 36 (36%) 28 (36%) 0.889

C. parapsilosis 10 (23.8%) 59 (42.8%) 0.027 35 (35%) 34 (35%) 0.304
C. glabrata 8 (19.0%) 17 (12.3%) 0.270 15 (15%) 10 (12.5%) 0.630
C. tropicalis 4 (9.5%) 14 (10.1%) 1.0 13 (13%) 5 (6.3%) 0.134

Table 2. Demographics and clinical characteristics of patients with C. albicans and non-albicans candidaemia

Characteristics C. albicans n = 64 (35.6%) Non-albicans n = 116 (64.4%) P Value

Age 61.3 ± 15.1 62.2 ± 16.5 0.726

Male sex 41 (64.1%) 81 (69.8%) 0.428
Underlying diseases
Haematological malignancy 3 (4.7%) 4 (3.4%) 0.993
Solid tumours 28 (43.8%) 52 (44.8%) 0.889
Diabetes mellitus 16 (25%) 25 (21.6%) 0.597
Predisposing factors
Dialysis 2 (3.1%) 6 (5.2%) 0.795
Current and former smokers 14 (21.9%) 40 (34.5%) 0.077
ICU stay 19 (29.7%) 37 (31.9%) 0.759
Surgical patients 45 (70.3%) 93 (80.2%) 0.134
Recent abdominal surgery 35 (54.7%) 73 (62.9%) 0.280
Recent cancer surgery 25 (39.1%) 46 (39.7%) 0.938
Prior antibiotics exposure 60 (93.8%) 103 (88.8%) 0.276
Prior antifungal exposure 8 (12.5%) 18 (15.5%) 0.581
TPN 50 (78.1%) 102 (87.9%) 0.082
Presence of CVC 28 (43.8%) 40 (34.5%) 0.220
Mechanical ventilation 3 (4.7%) 9 (7.8%) 0.632
Presence of urethral catheter 47 (73.4%) 68 (58.6%) 0.048
Presence of gastric tube 42 (65.6%) 80 (69%) 0.646
Laboratory findings
Renal failure 13 (20.3%) 25 (21.6%) 0.845
Anaemia 49 (76.6%) 99 (85.3%) 0.140
Hypoalbuminaemia 26 (40.6%) 44 (37.9%) 0.723
Hyponatraemia 18 (28.1%) 50 (43.1%) 0.047
Leukopaenia 9 (14.1%) 16 (13.8%) 0.960
Thrombocytopaenia 10 (15.6%) 18 (15.5%) 0.985
Fluconazole susceptibility 57 (89.1%) 72 (62.1%) <0.001
Outcome
Crude 30-day mortality 23 (35.9%) 23 (19.8%) 0.018
LOS (days) 28 (21–38.3) 30 (20–48) 0.469
ICU, intensive care unit; TPN, total parenteral nutrition; CVC, central venous catheter; LOS, length of hospital stay.

(49%) was more than twice as frequent as C. tropicalis (19%) negative association with C. tropicalis [35]. Therefore, we
or C. parapsilosis (18%), which may explain the different find- speculate that the possible reason for divergent views may
ings. Frequent usage of TPN has been linked with C. parapsi- be the variable proportion of Candida spp. in the local epi-
losis candidaemia compared with C. albicans, as well as a demiological setting.
Epidemiology and Infection 5

Table 3. Multivariate logistic regression analysis of factors associated with an increased risk of candidaemia due to NAC species vs. C. albicans

Risk factors B S.E. Wald P Value OR (95% CI)

Current and former smokers 0.601 0.369 2.655 0.103 1.824 (0.885–3.756)
TPN 0.93 0.442 4.432 0.035 2.535 (1.066–6.026)
Presence of urinary catheter −0.831 0.362 5.263 0.022 0.436 (0.214–0.886)
Constant 0.205 0.426 0.232 0.63 1.227 (–)
B, coefficient; S.E., standard error; Wald test statistic; OR, odds ratio; CI, confidence interval; TPN, total parenteral nutrition.

Table 4. In vitro antifungal susceptibility of Candida species isolated from 180 candidaemia episodes

MIC (μg/ml) No. (%) of isolates by new CBPsa

Species
(n = 180) Antifungal agent Ranges MIC 50 MIC 90 S SDD/I R

C. albicans Fluconazole ⩽0.5–16 ⩽1 1 57 (89.1%) 5 (7.8%) 2 (3.1%)

(n = 64) Voriconazole ⩽0.03–4 ⩽0.06 0.125 57 (89.1%) 3 (4.7%) 4 (6.3%)
Itraconazole 0.062–1 ⩽0.125 0.125 NA NA NA
Amphotericin B ⩽0.25–1 ⩽0.5 0.5 64 (100%) 0 0
C. parapsilosis Fluconazole ⩽0.5–8 ⩽1 ⩽4 61 (88.4%) 6 (8.7%) 2 (2.9%)
(n = 69) Voriconazole ⩽0.03–1 ⩽0.06 0.125 66 (95.7%) 2 (2.9%) 1 (1.4%)
Itraconazole ⩽0.062–0.25 ⩽0.125 0.125 67 (97.1%) 0 2 (2.9%)
Amphotericin B ⩽0.25–1 ⩽0.5 0.5 69 (100%) 0 0
C. glabrata Fluconazole ⩽1–16 4 16 0 25 (100%) 0
(n = 25) Voriconazole 0.06–2 0.125 0.25 NA NA NA
Itraconazole ⩽0.125–1 0.25 1 NA NA NA
Amphotericin B ⩽0.25–2 ⩽0.5 0.5 24 (96%) 0 1 (4%)
C. tropicalis Fluconazole ⩽1–⩾128 2 16 11 (61.1%) 1 (5.6%) 6 (33.3%)
(n = 18) Voriconazole ⩽0.06–8 0.125 4 12 (66.7%) 2 (11.1%) 4 (22.2%)
Itraconazole ⩽0.125–8 0.125 2 9 (50%) 0 9 (50%)
Amphotericin B ⩽0.25–0.5 ⩽0.5 0.5 18 (100%) 0 0
C. kruseib,c Fluconazole 64–128 0 0 2 (100%)
(n = 2) Voriconazole 0.5–>8 1 (50%) 0 1 (50%)
Itraconazole 1–>4 NA NA NA
Amphotericin B 1 2 (100%) 0 0
C. famata b
Fluconazole ⩽1–4 NA NA NA
(n = 2) Voriconazole ⩽0.06–0.125 NA NA NA
Itraconazole ⩽0.125–0.125 NA NA NA
Amphotericin B ⩽0.25–⩽0.5 NA NA NA
a
Except lack of C. famata specific CBPs, CBPs for Candida susceptibility to fluconazole and voriconazole were obtained from CLSI [21], while CBPs for susceptibility of Candida against
itraconazole and amphotericin B were from EUCAST [22].
b
MIC 50 and 90 values were not calculated for antifungal drugs against C. krusei and C. famata due to the small number of C. krusei and C. famata cases.
c
Isolates of C. krusei are considered to be intrinsically resistant to fluconazole. S, susceptible; I, intermediate; SDD, susceptible-dose dependent; R, resistant; NA, non-applicable.

Smoking increases susceptibility to a wide range of bacterial multivariate analysis, current and former smoking was not inde-
and viral infections [36] as well as oral candidiasis in HIV infected pendently associated with the occurrence of NAC BSI, it was likely
adults [37]. Based on pooled data from contemporary cohort to be a predictive factor for NAC candidaemia as according to our
studies in the USA, the risk of death from infections in current data current and former smokers had an increased risk for NAC
smokers is more than twice that of individuals who have never BSI, but no significant relationship was found between smoking
smoked [38]. However, data on invasive candidiasis and candi- and death due to candidaemia. More epidemiological, clinical
daemia in adult smokers remain scarce. Here, we explored the and mechanistic approaches are needed to study further the
role of smoking in NAC and C. albicans BSI. Although, by impact of smoking in such patients.
6 Wei Zhang et al.

Table 5. Univariate analysis of factors associated with 30-day mortality of patients with candidaemia

Characteristics Survivors n = 134 (74.4%) Deaths n = 46 (25.6%) P Value

Age 61.5 ± 15.3 63 ± 17.8 0.570

Male sex 90 (67.2%) 32 (69.6%) 0.764
Underlying diseases
Haematological malignancy 3 (2.2%) 4 (8.7%) 0.130
Solid tumours 67 (50%) 13 (28.3%) 0.010
Diabetes mellitus 30 (22.4%) 11 (23.9%) 0.831
Predisposing factors
Dialysis 7 (5.2%) 1 (2.2%) 0.652
Current and former smokers 42 (31.3%) 12 (26.1%) 0.502
ICU stay 25 (18.7%) 31 (67.4%) <0.001
Surgical patients 107 (79.9%) 31 (67.4%) 0.085
Recent abdominal surgery 88 (65.7%) 20 (43.5%) 0.008
Recent cancer surgery 63 (47%) 8 (17.4%) <0.001
Prior antibiotics exposure 119 (88.8%) 44 (95.7%) 0.281
Prior antifungal exposure 20 (14.9%) 6 (13%) 0.754
TPN 112 (83.6%) 40 (87%) 0.586
Presence of CVC 42 (31.3%) 26 (56.5%) 0.002
Mechanical ventilation 6 (4.5%) 6 (13%) 0.096
Presence of urethral catheter 88 (65.7%) 27 (58.7%) 0.395
Presence of gastric tube 95 (70.9%) 27 (58.7%) 0.127
Laboratory findings
Renal failure 19 (14.2%) 19 (41.3%) <0.001
Anaemia 109 (81.3%) 39 (84.8%) 0.599
Hypoalbuminaemia 42 (31.3%) 28 (60.9%) <0.001
Hyponatraemia 60 (44.8%) 8 (17.4%) 0.001
Leukopaenia 16 (11.9%) 9 (19.6%) 0.197
Thrombocytopaenia 10 (7.5%) 18 (39.1%) <0.001
C. albicans 41 (30.6%) 23 (50%) 0.018
Fluconazole susceptibility 99 (73.9%) 30 (65.2%) 0.261
Empirical antifungal therapy 89 (66.4%) 29 (63%) 0.678
No antifungal therapy 17 (12.7%) 8 (17.4%) 0.426
Outcome
LOS (days) 30 (21–48) 27.5 (16–41.3) 0.191
ICU, intensive care unit; TPN, total parenteral nutrition; CVC, central venous catheter; LOS, length of hospital stay.

C. albicans blood isolates showed significantly greater suscep- recommendations from the updated guideline for the manage-
tibility to fluconazole (89.1%) than NAC isolates (62.1%) (P < ment of candidiasis that transition to higher-dose fluconazole
0.001) by applying the species-specific new CBPs. In contrast to should be considered for patients with fluconazole-susceptible
the overall high-fluconazole susceptibility rate of C. parapsilosis C. glabrata isolates [14]. We identified only two patients with
(88.4%) and C. tropicalis (61.1%) recorded here, some studies C. krusei, one of which was resistant to voriconazole; each was
from other cities in China have reported relatively reduced flucon- effectively treated with micafungin and caspofungin, respectively.
azole susceptibility in their corresponding Candida spp. For Treatment of candidaemia is increasingly problematic owing
example, a survey from Nanjing reported that only 74.2% of to accumulated resistance of Candida isolates to antifungal agents,
C. albicans, 57.7% of C. parapsilosis, 9.1% of C. glabrata and especially fluconazole worldwide [24, 39]. Considering safety and
31.6% of C. tropicalis were susceptible to fluconazole [28]. efficacy aspects of treatment, echinocandins are now recom-
Based on the CBPs, all 25 isolates of C. glabrata in this study mended especially in the early treatment of candidaemia, by
exhibited SDD to fluconazole, which supports the both USA and European guidelines [14, 39]. In the current
Epidemiology and Infection 7

Table 6. Multivariate logistic regression analysis of risk factors associated with 30-day mortality of patients with candidaemia

Risk factors B S.E. Wald P Value OR (95% CI)

Solid tumours 1.845 0.975 3.581 0.058 6.329 (0.936–42.778)

ICU stay 1.389 0.51 7.417 0.006 4.013 (1.476–10.906)
Surgical patients 0.793 0.658 1.451 0.228 2.209 (0.608–8.02)
Recent abdominal surgery 0.069 0.647 0.011 0.915 1.072 (0.302–3.805)
Recent cancer surgery −3.276 1.194 7.53 0.006 0.038 (0.004–0.392)
Presence of CVC 0.476 0.491 0.939 0.333 1.61 (0.615–4.216)
Mechanical ventilation 0.521 0.98 0.283 0.595 1.683 (0.247–11.48)
Renal failure 1.176 0.559 4.429 0.035 3.24 (1.084–9.683)
Hypoalbuminaemia 0.861 0.474 3.303 0.069 2.366 (0.935–5.991)
Hyponatraemia −0.988 0.538 3.376 0.066 0.372 (0.13–1.068)
Thrombocytopaenia 1.97 0.614 10.292 0.001 7.171 (2.152–23.892)
C. albicans 1.289 0.504 6.543 0.011 3.629 (1.352–9.743)
Constant −3.491 0.729 22.947 <0.001 0.03 (–)
B, coefficient; S.E., standard error; Wald test statistic; OR, odds ratio; CI, confidence interval; ICU, intensive care unit; CVC, central venous catheter.

study, fluconazole was the most frequently used antifungal agent C. krusei and C. albicans (20 ± 40%), the lowest being C. parapsi-
for primary therapy, followed by echinocandins and voriconazole, losis [40]. Accordingly, the predominance of C. parapsilosis in our
although amphotericin B exhibited excellent in vitro activity over- study may have partly contributed to the observed lower mortality
all against Candida spp. The plausible explanations for the preva- in patients with NAC.
lent use of fluconazole are the species distribution pattern of Apart from C. albicans, the other variables strongly associated
isolates, their high rates of susceptibility of C. albicans and C. with mortality were ICU stay, renal failure and thrombocyto-
parapsilosis to the agents, and its safety and tolerability compared paenia, which suggest that the more severely ill patients had a
with amphotericin B, as well as a significantly lower cost. Due to worse prognosis. Additionally, almost all clearance of candidae-
the lack of echinocandin susceptibility testing in the present study, mia in survivors who did not receive antifungal treatment
these agents were used as initial empirical therapy for critically ill occurred in surgical patients following the removal of an indwel-
patients or those considered likely to have a fluconazole-resistant ling CVC or drainage catheter. Our results indicated that recent
Candida spp. based on the recommendations of the USA and cancer surgery was associated with a higher probability of survival
European guidelines. [14, 39] but an explanation for this remains unclear. Cancer patients who
Several studies have reported discrepant results between out- were judged reasonably able to tolerate surgery, often lacked dis-
comes of candidaemia and C. albicans spp. in comparison with tant metastases and had less disease severity and fewer complica-
NAC spp. Differences in mortality were not statistically significant tions. Such host factors may therefore be contributory to a
in patient cohorts in the USA (58% vs. 57%) [34], Beijing (34.7% reduced mortality rate for recent cancer surgery, and may aid clin-
vs. 38.6%) [17] and Shanghai (37.3% vs. 27.9%) [29]. Here, we icians better to judge the prognosis of these patients with
found that C. albicans BSI was an independent risk factor asso- candidaemia.
ciated with the mortality rate compared with NAC BSI (OR Our study has several main limitations. Firstly, owing to its
3.629; CI 1.352–9.743; P = 0.011), which mirrors data from retrospective cohort design, factors such as the management of
Shandong and Nanjing in China [8, 30]. By contrast, in Greece, CVC, the use of appropriate antifungal treatments, immunosup-
Dimopoulos et al., [26] reported NAC candidaemia to be pressive, cancer and glucocorticoid therapy, and underlying
associated with higher mortality than C. albicans in non- chronic lung and heart disease, were not evaluated owing to
immunosuppressed, non-neutropenic patients after ICU ad- incomplete medical data acquisition. Secondly, in vitro echino-
mission (OR 6.7; 95% CI 1.2–37.7; P = 0.03), which could be candin susceptibility testing of Candida isolates was not per-
attributed to inappropriate or delayed therapy owing to the slower formed, and thirdly, the relatively small sample size may impact
growth of NAC isolates on primary culture; patients with NAC on the confidence intervals and analysis of risk factors. Lastly,
candidaemia may also have been more critically ill than those the study was limited to a single centre and thus, the results
with C. albicans. may not be applicable to other settings.
Mortality rates of candidaemia have been attributed to the In summary, over the last 6 years, NAC was predominant
relative virulence of different Candida spp., a failure of among Candida isolates from adult candidaemia at Shengjing
host-defence mechanisms, patient’s underlying diseases and com- Hospital in Shenyang. TPN was associated with an increased
plications, inappropriate or delayed treatment and other factors risk of developing NAC candidaemia compared with C. albicans
[5, 20, 40]. In animal models, C. parapsilosis and C. krusei exhibit and patients with a urinary catheter were clearly at an increased
less virulence than C. albicans, C. tropicalis and C. glabrata [40]. risk of BSI due to C. albicans. Four independent risk factors for
A meta review of dozens of studies concluded in terms of both candidaemia-related death were identified, namely, ICU stay,
overall and attributable mortality that C. tropicalis and C. glabrata thrombocytopaenia, the isolation of C. albicans as a significant
are associated with the highest mortality (40 ± 70%), followed by predictor of survival and recent cancer surgery.
8 Wei Zhang et al.

Acknowledgements. We are grateful to Dr Zhijie Zhang and Dr Di Wang at 20. Garey KW et al. (2006) Time to initiation of fluconazole therapy impacts
the Department of Clinical Laboratory, Shengjing Hospital, for their valuable mortality in patients with candidemia: a multi-institutional study. Clinical
advice on microbiological methods to us and to Dr Liqiang Zheng at the Infectious Diseases 43, 25–31.
Department of Clinical Epidemiology, Shengjing Hospital, for his critical 21. Clinical Laboratory Standards Institute (2017) Performance Standards
assistance with statistical analysis. for Antifungal Susceptibility Testing of Yeasts, 1st Edn. CLSI document
M60. Wayne, PA: Clinical and Laboratory Standards Institute. Available at
Financial support. This research did not receive any specific grant from https://clsi.org/standards/products/microbiology/documents/m60/ (Accessed
funding agencies in the public, commercial, or not-for-profit sectors. 12 June 2019).
22. European Committee on Antimicrobial Susceptibility Testing.
Conflict of interest. The authors declare no financial or other competing
Antifungal Agents Breakpoint tables for interpretation of MICs, Version
interests. The authors alone are responsible for the content and the writing 9.0, valid from 2018-02-12. Available at http://www.eucast.org/clinical_-
of the paper. breakpoints/ (Accessed 12 June 2019).
23. Gonzalez GM, Elizondo M and Ayala J (2008) Trends in species distribu-
tion and susceptibility of bloodstream isolates of Candida collected in
References
Monterrey, Mexico, to seven antifungal agents: results of a 3-year (2004 to
1. Wisplinghoff H et al. (2004) Nosocomial bloodstream infections in US 2007) surveillance study. Journal of Clinical Microbiology 46, 2902–2905.
hospitals: analysis of 24,179 cases from a prospective nationwide surveil- 24. Pfaller MA et al. (2011) Geographic variations in species distribution and
lance study. Clinical Infectious Diseases 39, 309–317. echinocandin and azole antifungal resistance rates among Candida blood-
2. Marchetti O et al. (2004) Epidemiology of candidemia in Swiss tertiary stream infection isolates: report from the SENTRY Antimicrobial
care hospitals: secular trends, 1991–2000. Clinical Infectious Diseases 38, Surveillance Program (2008 to 2009). Journal of Clinical Microbiology
311–320. 49, 396–399.
3. Magill SS et al. (2014) Multistate point-prevalence survey of health 25. Wu Z et al. (2014) Candidemia: incidence rates, type of species, and risk
care-associated infections. New England Journal of Medicine 370, 1198– factors at a tertiary care academic hospital in China. International Journal
1208. of Infectious Diseases 22, 4–8.
4. Zaoutis TE et al. (2005) The epidemiology and attributable outcomes of 26. Dimopoulos G et al. (2008) Candida albicans versus non-albicans inten-
candidemia in adults and children hospitalized in the United States: a pro- sive care unit-acquired bloodstream infections: differences in risk factors
pensity analysis. Clinical Infectious Diseases 41, 1232–1239. and outcome. Anesthesia & Analgesia 106, 523–529.
5. Puig-Asensio M et al. (2014) Epidemiology and predictive factors for early 27. Chi HW et al. (2011) Candida albicans versus non-albicans bloodstream
and late mortality in Candida bloodstream infections: a population-based infections: the comparison of risk factors and outcome. Journal of
surveillance in Spain. Clinical Microbiology and Infection 20, 245–254. Microbiology, Immunology and Infection 44, 369–375.
6. Chakrabarti A et al. (2015) Incidence, characteristics and outcome of 28. Ma CF et al. (2013) Surveillance study of species distribution, antifungal
ICU-acquired candidemia in India. Intensive Care Medicine 41, 285–295. susceptibility and mortality of nosocomial candidemia in a tertiary care
7. Hassan I et al. (2009) Excess mortality, length of stay and cost attributable hospital in China. BMC Infectious Diseases 13, 337.
to candidaemia. Journal of Infection 59, 360–365. 29. Wu JQ et al. (2011) Epidemiology and risk factors for non-Candida albi-
8. Zhang XB et al. (2012) Retrospective analysis of epidemiology and prog- cans candidemia in non-neutropenic patients at a Chinese teaching hos-
nostic factors for candidemia at a hospital in China, 2000–2009. Japanese pital. Medical Mycology 49, 552–555.
Journal of Infectious Diseases 65, 510–515. 30. Wang H et al. (2014) Antibiotics exposure, risk factors, and outcomes
9. Yang ZT et al. (2014) Epidemiology, species distribution and outcome of with Candida albicans and non-Candida albicans candidemia. Results
nosocomial Candida spp. bloodstream infection in Shanghai. BMC from a multi-center study. Saudi Medical Journal 35, 153–158.
Infectious Diseases 14, 241. 31. Hii IM et al. (2015) Changing epidemiology of candidemia in a medical
10. Diekema D et al. (2012) The changing epidemiology of healthcare- center in middle Taiwan. Journal of Microbiology, Immunology and
associated candidemia over three decades. Diagnostic Microbiology and Infection 48, 306–315.
Infectious Disease 73, 45–48. 32. Al-Rawahi GN and Roscoe DL (2013) Ten-year review of candidemia in a
11. Gudlaugsson O et al. (2003) Attributable mortality of nosocomial candi- Canadian tertiary care centre: Predominance of non-albicans Candida
demia, revisited. Clinical Infectious Diseases 37, 1172–1177. species. Canadian Journal of Infectious Diseases and Medical
12. Sbrana F et al. (2018) Independent risk factors for mortality in critically Microbiology 24, e65–e68.
ill patients with candidemia on Italian Internal Medicine Wards. Internal 33. Alp S et al. (2015) Epidemiology of candidaemia in a tertiary care univer-
and Emergency Medicine 13, 199–204. sity hospital: 10-year experience with 381 candidaemia episodes between
13. Das I et al. (2011) Epidemiology, clinical characteristics, and outcome of 2001 and 2010. Mycoses 58, 498–505.
candidemia: experience in a tertiary referral center in the UK. 34. Chow JK et al. (2008) Factors associated with candidemia caused by non-
International Journal of Infectious Diseases 15, e759–e763. albicans Candida species versus Candida albicans in the intensive care
14. Pappas PG et al. (2016) Clinical practice guideline for the management of unit. Clinical Infectious Diseases 46, 1206–1213.
candidiasis: 2016 update by the Infectious Diseases Society of America. 35. Chen LY et al. (2013) Associated clinical characteristics of patients with
Clinical Infectious Diseases 62, e1–50. candidemia among different Candida species. Journal of Microbiology,
15. Barchiesi F et al. (2016) Epidemiology, clinical characteristics, and out- Immunology and Infection 46, 463–468.
come of candidemia in a tertiary referral center in Italy from 2010 to 36. Bagaitkar J, Demuth DR and Scott DA (2008) Tobacco use increases
2014. Infection 44, 205–213. susceptibility to bacterial infection. Tobacco Induced Diseases 4, 12.
16. Guinea J (2014) Global trends in the distribution of Candida species caus- 37. Chattopadhyay A and Patton LL (2013) Smoking as a risk factor for oral
ing candidemia. Clinical Microbiology and Infection 20(suppl. 6), 5–10. candidiasis in HIV-infected adults. Journal of Oral Pathology and
17. Ding X et al. (2015) Epidemiology and risk factors for nosocomial Medicine 42, 302–308.
Non-Candida albicans candidemia in adult patients at a tertiary care hos- 38. Carter BD et al. (2015) Smoking and mortality – beyond established
pital in North China. Medical Mycology 53, 684–690. causes. New England Journal of Medicine 372, 631–640.
18. Wang H, Xu YC and Hsueh PR (2016) Epidemiology of candidemia and 39. Cornely OA et al. (2012) ESCMID* guideline for the diagnosis and man-
antifungal susceptibility in invasive Candida species in the Asia-Pacific agement of Candida diseases 2012: non-neutropenic adult patients.
region. Future Microbiology 11, 1461–1477. Clinical Microbiology and Infection 18(suppl. 7), 19–37.
19. Bassetti M et al. (2013) Epidemiology, species distribution, antifungal 40. Krcmery V and Barnes AJ (2002) Non-albicans Candida spp. causing
susceptibility, and outcome of candidemia across five sites in Italy and fungaemia: pathogenicity and antifungal resistance. Journal of hospital
Spain. Journal of Clinical Microbiology 51, 4167–4172. infection 50, 243–260.