Mortality‑associated factors of candidemia: a multi‑center prospective cohort in Turkey

Abstract

Candidemia may present as severe and life-threatening infections and is associated with a high mortality rate. This study
aimed to evaluate the risk factors associated with 30-day mortality in patients with candidemia. A multi-center prospective
observational study was conducted in seven university hospitals in six provinces in the western part of Turkey. Patient data
were collected with a structured form between January 2018 and April 2019. In total, 425 episodes of candidemia were
observed during the study period. Two hundred forty-one patients died within 30 days, and the 30-day crude mortality rate
was 56.7%. Multivariable analysis found that SOFA score (OR: 1.28, CI: 1.154–1.420, p < 0.001), parenteral nutrition (OR:
3.9, CI: 1.752–8.810, p = 0.001), previous antibacterial treatment (OR: 9.32, CI: 1.634–53.744, p = 0.012), newly developed
renal failure after candidemia (OR: 2.7, CI: 1.079–6.761, p = 0.034), and newly developed thrombocytopenia after candidemia
(OR: 2.6, CI: 1. 057–6.439, p = 0.038) were significantly associated with 30-day mortality. Central venous catheter
removal was the only factor protective against mortality (OR: 0.34, CI:0.147–0.768, p = 0.010) in multivariable analysis.
Candidemia mortality is high in patients with high SOFA scores, those receiving TPN therapy, and those who previously
received antibacterial therapy. Renal failure and thrombocytopenia developing after candidemia should be followed carefully
in patients. Antifungal therapy and removing the central venous catheter are essential in the management of candidemia.

Full text

Vol.:(0123456789)
1 3
European Journal of Clinical Microbiology & Infectious Diseases
https://doi.org/10.1007/s10096-021-04394-0
ORIGINAL ARTICLE
Mortality‑associated factors ofcandidemia: amulti‑center prospective
cohort inTurkey
MuratKutlu1,2 · SeldaSayın‑Kutlu1· SemaAlp‑Çavuş3· ŞerifeBarçınÖztürk4· MeltemTaşbakan5· BetilÖzhak6·
OnurKaya7· OyaErenKutsoylu3· ŞebnemŞenol‑Akar8· ÖzgeTurhan9· GülşenMermut5· BülentErtuğrul4·
HüsnüPullukcu5· ÇiğdemBanuÇetin8· VildanAvkan‑Oğuz3· NurYapar3· DilekYeşim‑Metin10· ÇağrıErgin11
Received: 12 August 2021 / Accepted: 6 December 2021
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021
Abstract
Candidemia may present as severe and life-threatening infections and is associated with a high mortality rate. This study
aimed to evaluate the risk factors associated with 30-day mortality in patients with candidemia. A multi-center prospective
observational study was conducted in seven university hospitals in six provinces in the western part of Turkey. Patient data
were collected with a structured form between January 2018 and April 2019. In total, 425 episodes of candidemia were
observed during the study period. Two hundred forty-one patients died within 30days, and the 30-day crude mortality rate
was 56.7%. Multivariable analysis found that SOFA score (OR: 1.28, CI: 1.154–1.420, p < 0.001), parenteral nutrition (OR:
3.9, CI: 1.752–8.810, p = 0.001), previous antibacterial treatment (OR: 9.32, CI: 1.634–53.744, p = 0.012), newly developed
renal failure after candidemia (OR: 2.7, CI: 1.079–6.761, p = 0.034), and newly developed thrombocytopenia after candi-
demia (OR: 2.6, CI: 1. 057–6.439, p = 0.038) were significantly associated with 30-day mortality. Central venous catheter
removal was the only factor protective against mortality (OR: 0.34, CI:0.147–0.768, p = 0.010) in multivariable analysis.
Candidemia mortality is high in patients with high SOFA scores, those receiving TPN therapy, and those who previously
received antibacterial therapy. Renal failure and thrombocytopenia developing after candidemia should be followed carefully
in patients. Antifungal therapy and removing the central venous catheter are essential in the management of candidemia.
Keywords Candida· Candida parapsilosis· Catheter removal· Mortality· Risk factors
Introduction
Candida is a significant cause of bloodstream infections
(BSIs). Candida BSIs are mostly nosocomial, or health-
care-associated [1–3]. The incidence of candidemia varies
depending on time, place, and age [2–4]. The incidence of
candidemia in southern Europe, where Turkey is located,
is higher than in northern and western Europe [4]; at a
university hospital in Turkey, Candida ranked as the fifth
most-frequently isolated pathogen among all positive
* Murat Kutlu
muratkutlu72@yahoo.com
1 Infectious Diseases andClinical Microbiology, Pamukkale
University, Denizli, Turkey
2 Infectious Diseases andClinical Microbiology
Department, Pamukkale University, School ofMedicine,
Kınıklı/Pamukkale, 20070Denizli, Turkey
3 Infectious Diseases andClinical Microbiology, Dokuz Eylul
University, İzmir, Turkey
4 Infectious Diseases andClinical Microbiology, Adnan
Menderes University, Aydın, Turkey
5 Infectious Diseases andClinical Microbiology, Ege
University, İzmir, Turkey
6 Medical Microbiology, Akdeniz University, Antalya, Turkey
7 Infectious Diseases andClinical Microbiology, Süleyman
Demirel University, Isparta, Turkey
8 Infectious Diseases andClinical Microbiology, Celal Bayar
University, Manisa, Turkey
9 Infectious Diseases andClinical Microbiology, Akdeniz
University, Antalya, Turkey
10 Medical Microbiology, Ege University, İzmir, Turkey
11 Medical Microbiology, Pamukkale University, Denizli,
Turkey

European Journal of Clinical Microbiology & Infectious Diseases
1 3
blood cultures [5]. The rate of candidemia was found to
be 4.7% in patients with sepsis in a multi-center intensive
care unit (ICU) point prevalence study [6].
Candida albicans is the main causative species among
the causes of candidemia; however, the increasing trend
of non-albicans candida species continues [4, 7]. The
distribution of species that cause candidemia depends on
various factors including comorbidities and geography
[8]. C. glabrata in northern Europe and C. parapsilosis
in southern Europe are detected as the second causative
agents following C. albicans [4, 8]. Recent publications
from southern European countries, such as Italy, Greece,
and Turkey, reported a higher rate of C. parapsilosis when
compared to older studies [9–11].
Candidemia may present as severe and life-threatening
infections and is associated with prolonged hospitaliza-
tions, high costs, and substantial morbidity [1–3]. Patients
with candidemia have a high mortality rate (25–50%),
although the attributable mortality rate may be lower [1–3,
12, 13]. Despite increased awareness, knowledge of the
risk factors, antifungal drug options, guidelines, and infec-
tion control measures, the mortality rate of candidemia
remains high [4, 12, 14].
Mortality in candidemia was found to be associated
with underlying diseases and comorbidities, APACHE and
Sequential Organ Failure Assessment (SOFA) scores, ICU
admission, multiple invasive interventions, and treatments
showing the severity of illness [2, 11, 15–25]. Mortality was
also associated with certain Candida species and some labo-
ratory parameters [11, 15–19]. The main factors that reduce
mortality are antifungal therapy and source control. A rela-
tionship has also been observed between mortality and some
sub-factors, such as antifungal drug selection, time to start of
treatment, and source control timing [2, 11, 15, 16, 19–22].
This prospective multi-center observational study
aimed to evaluate the risk factors associated with 30-day
mortality in patients in seven university hospitals in west-
ern Turkey.
Methods
Study design andsettings
A prospective observational study was performed among
patients over 18years old with candidemia at 7 university
hospitals between January 2018 and April 2019. These uni-
versity hospitals are in the western part of Turkey, and the
Western Anatolia Fungal Study Association carried out the
study. The Pamukkale University ethics committee approved
this study (date and no: 14.11.2017–15).
Data collection anddefinitions
Patient data was collected through a structured form. The
form was divided into parts: 1—Demographic details; 2—
Pre-hospitalization period (underlying diseases, surgery
within last 3months, etc.); 3—Pre-candidemia period of
hospitalization (mechanical ventilation, central catheters,
parenteral nutrition, use of antibacterial and antifungals
before the candidemia was detected, etc.); 4—When can-
didemia developed and post-candidemia period (SOFA
score, newly developed neutropenia and thrombocytope-
nia, newly developed renal failure, concurrent bacteremia,
etc.); and 5—Candidemia-specific variables (Candida spe-
cies, the origin of candidemia and involvements, treatment
approaches and antifungals, outcome, etc.)
Each center entered the data into the form in Excel and
uploaded it to Google Tables. The data entry was checked
weekly by the study coordinators.
The primary outcome of the study was 30-day mor-
tality. Candidemia was defined as one or more positive
blood cultures for the Candida species in any patient. Only
one episode of candidemia per patient was included in
the study. Hospitalization duration was considered as the
period from admission until the first positive blood culture
was taken. Corticosteroid use was established as 20mg/
day prednisolone for > 7days or 10mg/day prednisolone
for > 2weeks within 4weeks before candidemia diagnosis.
SOFA scoring was conducted with patients hospitalized in
ICUs and patients transferred to ICUs within the first 48h
after candidemia diagnosis.
After candidemia diagnosis, newly developed renal fail-
ure was defined as a 25% decrease in glomerular filtration
rate (GFR) or a 50% increase in basal creatinine value.
After candidemia, newly developed neutropenia and newly
developed thrombocytopenia were defined as decreased
neutrophil count below 500/mm3 and platelet count below
150,000/mm3, respectively. Detection of bacteremia within
72h before and after candidemia diagnosis was defined
as concurrent bacteremia. The time from the diagnosis
of candidemia to the catheter’s removal was recorded
in hours, and removing the catheter in the first 48h was
defined as early removal. The number of days between the
first candida-positive blood culture and initial antifungal
therapy was recorded; if antifungal treatment was started
on or before the day of the first positive blood culture, this
period was considered zero days.
The decision of appropriate antifungal treatment was
made according to the following criteria: (1) ≥ 400mg/day
fluconazole or an echinocandin, or one of the other azoles;
(2) antifungal susceptibility test result showing the sensi-
tivity of the isolated strain to the initiated antifungal agent;
(3) in the case of a lack of a susceptibility test, initiation of

European Journal of Clinical Microbiology & Infectious Diseases
1 3
an antifungal agent that is adequate to the strain in terms
of the natural resistance of that strain; (4) in patients with
focal involvement, an antifungal agent appropriate to the
site of involvement. This study is reported in line with the
Strengthening the Reporting of Observational Studies in
Epidemiology (STROBE) recommendations.
Microbiological studies
Blood samples were cultured with BACTEC 9000 System
(BD Diagnostics, NJ, USA) and BactAlert (bioMérieux,
France). Positive blood cultures were plated on standard
bacteriological media and Sabouraud dextrose agar with
0.4% chloramphenicol media after Gram-staining and micro-
scopic evaluation. After the incubation period, all suspected
colonies were examined microscopically for the presence
of yeasts. Isolates with positive standard germ tubes and
chlamydospore formation on Dalmau plates were presump-
tively identified as Candida albicans. VITEK 2 system and
MALDI-TOF MS were used for acceptable identification of
other isolates.
Antifungal susceptibility testing
Both the broth microdilution method and the gradient strip
test method were performed. The broth microdilution test
was performed according to the CLSI M27-A3 document,
and the gradient strip test was performed according to the
manufacturer’s instructions. Each isolate was tested for
invitro susceptibility to amphotericin B, fluconazole, itra-
conazole, voriconazole, posaconazole, caspofungin, anidu-
lafungin, and micafungin.
Data analysis
Data were analyzed using IBM SPSS 23.0 (IBM Corp.
Released 2015. IBM SPSS Statistics for Windows, Ver-
sion 23.0. Armonk, NY: IBM Corp). The normality of the
variables was examined by one-way ANOVA test. Com-
parisons of continuous variables were made using Student’s
t-test or Mann–Whitney U-test. Categorical variables were
compared using the chi-square test or Fisher’s exact test for
association. Multivariable analysis by logistic regression
was also performed. Included in the model were variables
with p < 0.10: Age; duration of hospitalization time; SOFA
score; female gender; chronic obstructive pulmonary dis-
ease (COPD); malignancy; previous antibacterial treatment;
previous fluconazole treatment; being in the ICU; mechani-
cal ventilation; total parenteral nutrition; central venous
catheter (CVC); concurrent Gram-positive bacteremia;
catheter-related candidemia; newly developed thrombocy-
topenia after candidemia; newly developed renal failure after
candidemia; catheter removal; and echinocandin treatment.
Backward elimination of these variables was performed, and
statistical significance was set at p < 0.05. Patients receiving
and not receiving antifungal therapy were compared using
the Kaplan–Meier survival analysis.
In order to evaluate whether there is a difference in mor-
tality among antifungal drugs, the use of fluconazole before
candidemia is included in the model; any antifungal use
before candidemia was not included in the multivariate anal-
ysis. For the same reason, echinocandin treatment for can-
didemia was included in the model. Requirement of dialysis
developed after candidemia was not included in the model
because newly developed renal failure after candidemia was
included. Transfer to the ICU after candidemia diagnosis, no
fever response, and lack of negative blood culture were not
included in the multivariable analysis because these have
already been proven as indicative of poor prognosis.
Results
During the study period, a total of 425 patients with candi-
demia were identified. Of the 425 patients, 199 (46.8%) were
female, and 226 (53.2%) were male. The median age was
65 (55–76) years. Of these, 247 (58.1%) were hospitalized
in ICUs, 131 (30.8%) in internal medicine services, and 47
(11.1%) in surgery services when candidemia is detected.
Among all the patients, 241 patients died within 30days,
and the 30-day crude mortality rate was 56.7%. The number
of cases and the mortality rates by study centers are shown
in Fig.1. The demographics, comorbidities, and univariate
analysis of the mortality are presented in Table1.
Candida species
Candida albicans (169; 39.7%) and Candida parapsilosis
(143; 33.6%) were the most common species causing can-
didemia. Other Candida species were Candida tropicalis
(37; 8.7%), Candida glabrata (28; 6.6%), Candida kefyr
(10; 2.4%), Candida krusei (8; 1.9%), Candida dubliniensis
(2; 0.5%), Candida lusitaniae (2; 0.5%), Candida cifferi (1;
0.2%), Candida pelliculosa (1; 0.2%), and Candida guill-
ermondi (1; 0.2%). Nineteen Candidaisolates (4.5%) were
not typed. In 4 patients (0.9%), twoCandida species were
isolated. In terms of mortality, no differences were seen
between species (Table1). Demographic factors and sta-
tistically significant variables according to Candida species
are presented in Table2, excluding the patients with isolated
unidentified and multiple Candida species.
The origin andinvolvementsof candidemia
One hundred fifty-four (36.2%) cases were evaluated as
catheter-related candidemia. Twenty-seven (6.4%) patients

European Journal of Clinical Microbiology & Infectious Diseases
1 3
had intra-abdominal involvement. Cardiac and eye evalua-
tions were performed in 80 (18.8%) and 57 (13.4%) cases,
respectively. Endocarditis was detected in 3 (0.7%) patients,
and endophthalmitis was found in 2 (0.47%) patients.
Treatment approaches andantifungals
Before the candidemia was detected, 63 (14.8%)
patients—23 of whom were prophylactic and 40 of whom
empirical—were using antifungal drugs. In these cases, the
antifungals used were fluconazole (n = 39), echinocandins
(n = 19), and voriconazole or posaconazole (n = 5). The use
of any antifungal drug and the use of fluconazole before can-
didemia diagnosis were found to be associated with higher
mortality in the univariate analysis (p = 0.001 and p = 0.004,
respectively).
Fifty-one (12%) patients did not have any antifungal drug
treatment. The median age of these patients was 65 (56–77),
and 35 patients (69%) were male. The most common comor-
bidities were malignancy (24 patients, 47%), diabetes mel-
litus (13 patients, 26%), and renal failure (9 patients, 18%).
Eighteen (35%) patients had undergone surgery within the
last 3months. When candidemia was detected, 33 (65%)
patients were hospitalized in ICUs, 13 (25%) in internal
medicine services, and 5 (10%) in surgical services. Forty-
eight (94%) of these patients had broad-spectrum antibiotic
use, and 29 (57%) had TPN therapy. CVC and mechanical
ventilation were present in 37 (73%) and 32 (63%) patients,
respectively.
Twelve (23.5%) of these 51 patients did not die within
30days. Eight of these 12 patients had undergone catheter
removal or various source control procedures. Thirty-nine
(76.5%) of 51 patients died within 30days. Of these 39
patients, 34 (87%) died within the first 4days after the posi-
tive blood culture was obtained. Antifungal drugs were used
to treat candidemia in 374 patients. A Kaplan–Meier sur-
vival analysis of patients receiving and not receiving anti-
fungal therapy is shown in Fig.2 (p < 0.001).
The median time from onset of candidemia to antifungal
drug therapy was 2 (1–3) days, with no difference between
patients who died and those who survived (p = 0.123).
Among the patients who died, the median time from the start
of treatment to death was 7 (3–15.5) days. Any antifungal
drug treatment was found to prevent mortality in univari-
ate analysis (p = 0.002). The antifungal drugs used and their
relationships with mortality are included in Table3.
The initial antifungal drug was changed in 80 (21.4%)
patients who used antifungal treatment. No difference was
found between patients who died and patients who sur-
vived in terms of antifungal exchange (43 (17.8%) vs. 37
(20.1%), p = 0.319). In 31 of the patients who underwent a
drug change, the reason for the change was drug resistance,
and there was no difference between those who died and
those who survived (20 (8.3%) vs. 11 (6.0%), p = 0.236).
Inappropriate antifungal therapy overlapped with antifungal
resistance and was detected in 12 patients. In 6 patients, the
reason for the change was due to the pharmacokinetic/phar-
macodynamic properties of the drug. In the remaining 39
patients, the reason for the change was side effects or switch-
ing to oral therapy (no statistical evaluation was made).
CVC removal and other source controls (removing other
foreign bodies, drainage, and surgery) were performed in
Fig. 1 The number of cases and
mortality rates of each center
(n; %)

European Journal of Clinical Microbiology & Infectious Diseases
1 3
Table 1 Univariate analysis of risk factors for 30-day mortality in patients with candidemia
Variable Patients who died (%)
n = 241 (56.7)
Patients who survived (%)
n = 184 (43.3) p value
Median age (median, IQR), years 66 (58–76) 62 (52–76) 0.014
Female 120 (49.8) 79 (42.9) 0.096
Comorbidities
Diabetes mellitus 54 (22.4) 45 (24.5) 0.351
Renal failure 49 (20.3) 31 (16.8) 0.217
COPD 47 (19.5) 25 (13.6) 0.069
HIV infection 1 (0.4) 1 (0.5) 0.679
Steroid use 24 (10) 15 (8.2) 0.321
Malignancy (any) 116 (48.1) 76 (41.3) 0.096
Hematological malignancy 40 (16.6) 23 (12.5) 0.149
Solid organ malignancy 76 (31.5) 53 (28.8) 0.309
Otolog HSCT 3 (1.2) 2 (1.1) 1
Allogeneic HSCT 0 2 (1.1) 0.187
Solid organ transplantation 3 (1.2) 1 (0.5) 0.637
Surgery within last 3months 73 (30.3) 50 (27.2) 0.277
GI surgery within last 3months 26 (10.8) 17 (9.2) 0.361
Chemotherapy within last month 43 (17.8) 34 (18.5) 0.482
Before candidemia
Broad-spectrum antibiotic use 236 (97.9) 169 (91.8) 0.003
Antifungal use 47 (19.5) 16 (8.7) 0.001
Echinocandin 13 (6.3) 6 (3.4) 0.152
Fluconazole 30 (13.4) 9 (5.1) 0.004
Voriconazole or posaconazole 4 (2.0) 1 (0.6) 0.379
Surgery within candidemia hospitalization 90 (37.3) 59 (32.1) 0.152
Gastro-intestinal surgery within candidemia hospitalization 35 (18.8) 29 (18.8) 0.553
When candidemia is detected
The time from admission to candidemia diagnosis (median, IQR), days 25 (14–38) 17 (9–40) 0.025
Intensive care units 153 (63.5) 94 (51.1) 0.007
Internal medical wards 68 (28.2) 63 (34.2) 0.110
Surgical wards 20 (8.3) 27 (14.7) 0.028
Urinary catheter 216 (89.6) 138 (75.0) < 0.001
Central venous catheterization 184 (76.3) 116 (63.0) 0.002
Mechanical ventilation 147 (61.0) 74 (40.2) < 0.001
Total parenteral nutrition 161 (66.8) 84 (45.7) < 0.001
Fever or hypothermia 202 (83.8) 159 (86.4) 0.274
SOFA score (median, IQR) 10 (7–13) 6 (4–9) < 0.001
Bacteraemia 103 (42.7) 62 (33.7) 0.036
Gram-negative bacteraemia 51 (21.2) 38 (20.7) 0.498
Gram-positive bacteraemia 68 (28.2) 37 (20.1) 0.035
After candidemia
Neutropenia 15 (6.2) 7 (3.8) 0.186
Thrombocytopenia 100 (41.5) 30 (16.3) < 0.001
Renal failure 83 (34.4) 14 (7.6) < 0.001
Hemodialysis requirement 21 (8.7) 5 (2.7) 0.013
Transfer to the intensive care unit 48 (19.9) 14 (7.6) < 0.001
The origin of candidemia or involvement
Catheter-related candidemia 93 (38.5) 61 (33.1) 0.091
Endophthalmitis (n = 57) 2 (7.4) 0 (0) 0.220
Endocarditis (n = 80) 1 (3.6) 2 (3.8) 1

European Journal of Clinical Microbiology & Infectious Diseases
1 3
119 and 36 patients, respectively. CVC removal was pro-
tective against mortality in univariate analysis, while non-
catheter source control did not affect mortality (p = 0.007
and p = 0.491, respectively).
Mortality predictors
In multivariate analysis, high SOFA score (OR: 1.28, CI:
1.154–1.420, p < 0.001), total parenteral nutrition (TPN)
Table 1 (continued)
Variable Patients who died (%)
n = 241 (56.7)
Patients who survived (%)
n = 184 (43.3) p value
Intraabdominal focus (n = 425) 17 (7.1) 10 (5.4) 0.319
Candida species
C. albicans 95 (43.0) 74 (40.9) 0.373
C. parapsilosis 77 (34.8) 66 (36.5) 0.407
C. tropicalis 20 (9.0) 17 (9.4) 0.520
C. glabrata 13 (5.9) 15 (8.3) 0.227
C. kefyr 7 (3.2) 3 (1.7) 0.522
C. krusei 6 (2.7) 2 (1.1) 0.304
Other Candida spp.* 3 (1.4) 4 (2.2) 0.706
Polymicrobial candidemia** 2 (0.8) 2 (1.1) 1.0
Control blood culture had taken 142 (58.9) 152 (82.6) < 0.001
No negative result of the blood culture 44 (31.0) 10 (6.6) < 0.001
Breakthrough candidemia 20 (8.3) 22 (12.0) 0.138
The time between the onset of candidemia and death (median, IQR),days 8 (3–15.5) - -
*C.dubliniensis (2), C.lusitaniae (2), C.cifferi (1), C.pelliculosa (1), C.guillermondi (1).
**C.albicans and C.parapsilosis (2), C.albicans and C.kefyr (1), C.parapsilosis and C.glabrata (1).
Table 2 Comparison of variables according to Candida species in 402 candidemia cases
*C.kefyr (10), C.krusei (8), C.dubliniensis (2), C.lusitaniae (2), C.cifferi (1), C.pelliculosa (1), C.guillermondi (1).
1 p = 0,039, 2p = 0.022, 3p = 0.002, 4p = 0.026, 5p = 0.016, 6p = 0.017, 7p = 0.039, 8p = 0.040, 9p = 0.037, 10p < 0.001, 11p = 0.011, 12p = 0.036,
13p < 0.001, 14p = 0.018 15p = 0.014, 16p = 0.004, 17p = 0.018.
Variable C. albicans
(n = 169, 42%) C. parapsilosis
(n = 143, 36%) C. tropicalis
(n = 37, 9%) C. glabrata
(n = 28, 7%)
Other identified Can-
dida spp.* (n = 25,
6%)
Median age (median, IQR), years 65 (55–76) 68 (56–78) 64 (55–75) 60 (44–72) 59 (48–66)
Female 83 (49) 67 (47) 12 (32.4)113 (46.4) 16 (64)
Comorbidities
Diabetes mellitus 48 (51.6)230 (32.3) 5 (5.4) 5 (5.4) 5 (5.4)
Chemotherapy within last month 26 (36) 25 (34) 7 (10) 4 (6) 11 (15)3
Malignancy 74 (41) 55 (30)420 (11) 16 (9) 17 (9.3)5
Surgery within last 3months 55 (47) 32 (27)616 (14)710 (9) 4 (3)
Before candidemia
Antifungal use 18 (33) 23 (42) 3 (6) 4 (7) 7 (13)8
When candidemia is detected
Intensive care units 87 (38)9101 (44)10 14 (6)11 15(7) 12 (5)
Surgical wards 26 (55)12 6 (13)13 9 (19)14 4 (9) 2 (4)
Central venous catheterization 113 (41) 109 (39)15 23 (8) 20 (7) 13 (5)
Mortality 95 (43.0) 77 (34.8) 20 (9.0) 13 (5.9) 16 (7.2)
Antifungal drug change 27 (34) 39 (50)16 5 (6) 1 (1) 7 (9)
Antifungal drug change due to resistance 9 (29) 17 (55)17 1 (3) 1 (3) 3 (10)

European Journal of Clinical Microbiology & Infectious Diseases
1 3
(OR: 3.9, CI: 1.752–8.810, p = 0.001), CVC removal
(OR: 0.34, CI:0.147–0.768, p = 0.010), previous antibac-
terial treatment (OR: 9.32, CI: 1.634–53.744, p = 0.012),
newly developed renal failure after candidemia (OR: 2.7,
CI: 1.079–6.761, p = 0.034), and newly developed throm-
bocytopenia after candidemia (OR: 2.6, CI: 1. 057–6.439,
Fig. 2 Kaplan–Meier sur-
vival analysis for candidemia.
Patients treated with antifungal
drugs and patients not treated
with antifungal drugs were
included (log rank p ≤ 0.001)
Table 3 Univariate analysis of treatment choices and approaches for 30-day mortality in patients with candidemia
Variable Patients who died (%)
n = 241 (56.7)
Patients who survived (%)
n = 184 (43.3) p value
Antifungal (any) 202 (83.8) 172 (93.5) 0.002
Echinocandin 141 (69.8) 133 (77.3) 0.064
Fluconazole 46 (22.8) 35 (20.3) 0.330
Voriconazole/posaconazole 5 (2.5) 1 (0.6) 0.224
Liposomal amphotericin-B 10 (5.0) 3 (1.7) 0.154
The time interval from candidemia onset to the start of antifungal treatment
(SD), days
2.17 (2.3) 2.28 (1.8) 0.610
The time between the start of treatment and death (median, IQR), days 7 (3–15.5) - -
Catheter removal 56 (24.8) 63 (36.8) 0.007
Early (first 48h) catheter removal 38 (15.8) 30 (16.3) 0.492
Other source control procedures 21 (8.7) 15 (8.2) 0.491
Removing other foreign bodies 3 (1.2) 6 (3.3) 0.184
Drainage 14 (5.8) 6 (3.3) 0.159
Tissue or organ debridement 7 (2.9) 6 (3.3) 0.524
Correction of anatomical disorder 6 (2.5) 3 (1.6) 0.738
Antifungal drug changing 43 (17.8) 37 (20.1) 0.319
Drug resistance 20 (8.3) 11 (6.0) 0.236
Inappropriate antifungal drug 12 (5.9) 10 (5.8) 0.569
Antifungal susceptibility 136 (56.4) 116 (63.0) 0.101

European Journal of Clinical Microbiology & Infectious Diseases
1 3
p = 0.038) were found to be significantly associated with
30-day mortality (Table4).
Discussion
In this study, the 30-day mortality rate from candidemia
was 56.7%. The 30-day mortality rate in patients with can-
didemia has been reported as being between 40 and 83%
in several studies in Turkey [26–31]. Doğan etal. [11]
reported a 10-day mortality rate of 32.2% in a recent multi-
center study from Turkey. Koehler etal. [4] showed that
the incidence and mortality rates of candidemia in Europe
have increased over time and that mortality was higher in
university and teaching hospitals than in general hospitals,
as found in this study. Patients who did not receive antifun-
gal therapy were also included in the mortality analysis in
our study. A high SOFA score (OR: 1.28, CI: 1.154–1.420,
p < 0.001), TPN (OR: 3.9, CI: 1.752–8.810, p = 0.001), pre-
vious antibacterial treatment (OR: 9.32, CI: 1.634–53.744,
p = 0.012), newly developed renal failure after candidemia
(OR: 2.7, CI: 1.079–6.761, p = 0.034), and newly devel-
oped thrombocytopenia after candidemia (OR: 2.6, CI: 1.
057–6.439, p = 0.038) were found to be independent risk
factors of 30-day mortality from candidemia in this multi-
center prospective observational study.
CVC removal was the only factor associated with sur-
vival/decreased mortality (OR: 0.34, CI: 0.147–0.768,
p = 0.010). However, the impact of CVC removal on the
survival of patients with candidemia is disputed [15, 24,
32]. A Cochrane analysis revealed that a randomized con-
trolled trial was not available on this subject, and observa-
tional studies were insufficient to make conclusions [33].
On the other hand, the current IDSA guidelines recommend
that CVCs should be removed as early as possible when the
infection source is presumed to be the CVC [32].
Andes etal. [15] found that CVC removal was associ-
ated with decreased mortality in a patient-level quantita-
tive review of randomized trials. Garnacho-Montero etal.
[21] reported that CVC removal within the first 48h was
a determinant of survival in patients with catheter-related
candidemia; however, CVC removal did not influence the
outcome of non-catheter-related candidemia in that study.
Delayed CVC removal was associated with increased mor-
tality in patients with a low Charlson Comorbidity Index
(CCI) score, but not in patients with a high CCI score in
another study [16]. Puig-Assensio etal. [2] reported that
CVC removal and antifungal treatment prevented 7-day
mortality, but not 30-day mortality. Comorbidity factors
are associated with an increased rate of 30-day mortality.
In a study comparing early (first 48h) catheter removal and
catheter removal at any time, the catheter removal had a
mortality-preventing effect, but this effect was not detected
in early removal [19]. Unfortunately, the literature about the
effect of early CVC removal on survival is conflicting since
different studies have distinct results. In the current study,
early catheter removal was not associated with death, even
in univariate analysis (Table3). For this reason, when we
comment on the mortality preventing effect of CVC removal
at any time, it should be kept in mind that the severely ill
patients had no chance of early CVC removal, which affected
the study findings in those patients.
More than one-third of candidemia cases (36.2%) in this
study were related to a CVC. Candida parapsilosis is the
most common species associated with catheter-related can-
didemia [2, 34]. This species is considered to cause less
severe disease, and it has also been associated with lower
mortality in some studies [1, 16, 19, 35]. In this study, no
difference was found among the species in terms of mortal-
ity; still, in the Kaplan Meier survival analysis, mortality
was lower with C. parapsilosis between the 10th and 20th
days (data not shown, log rank p = 0.134, Breslow p = 0.05,
and Tarone-Ware p = 0.07). C. parapsilosis was the second-
most common cause of candidemia (33.6%) in this study.
Higher C. parapsilosis rates (~ 25%) have been reported in
some studies, including pediatric patients, in Turkey [29,
36]. However, in an earlier study, the rate of C. parapsilo-
sis among species that caused candidemia was only 14.5%
[37]. In addition to CVCs, C. parapsilosis is associated with
nosocomial factors such as TPN, ICU admission, and broad-
spectrum antibiotic use, and it is less susceptible to echino-
candins than other Candida species [1, 4, 35]. Consequently,
the increase of this species is remarkable and alarming.
Candida spp. usually cause infection in patients with
comorbid conditions. Candidemia is also an opportun-
istic infection; it affects patients who are already in a
Table 4 Multivariable analysis
of risk factors for 30-day
mortality in patients with
candidemia
Odds ratio 95% confidence interval p value
SOFA score 1.28 1.154–1.420 < 0.001
Total parenteral nutrition 3.9 1.752–8.810 0.001
Central venous catheter removal 0.34 0.147–0.768 0.010
Broad-spectrum antibiotic use before candidemia 9.3 1.634–53.744 0.012
Newly developed thrombocytopenia after candidemia 2.6 1. 057–6.439 0.038
Newly developed renal failure after candidemia 2.7 1.079–6.761 0.034

European Journal of Clinical Microbiology & Infectious Diseases
1 3
compromised and impaired physiological situation. In this
study, no relationship was found between mortality and
comorbid factors; however, at the time of the candidemia
diagnosis, 58.1% of the patients were in the ICU. Addition-
ally, sixty-two patients in other clinics were transferred to
ICUs after the diagnosis. The severity of the illness on the
day of detection has a powerful effect on survival [24]. The
APACHE II score, including age and comorbid conditions, is
a predictor of candidemia mortality [15, 19–21, 24]. Recent
studies reported that the SOFA score on the day of detection
was associated with survival [25, 38]. Although no severity
scoring system depending on the predictors was performed,
this study showed a strong relationship between SOFA score
and mortality (OR: 1.28, CI: 1.154–1.420, p < 0.001).
Platelet count and renal function, which are among the
SOFA parameters, also had independent associations with
mortality (OR: 2.6, CI: 1. 057–6.439, p = 0.038; OR: 2.7,
CI: 1.079–6.761, p = 0.034, respectively). Renal involve-
ment in candidemia has been shown in autopsy studies [39,
40]. The kidneys are affected by the intense inflammatory
response developed against candidemia, and increased BUN,
creatinine levels, and kidney failure may develop [41, 42].
Clinical studies have shown that renal failure is associated
with mortality in candidemia [11, 17]. This study found that
the development of renal failure after candidemia—not as
a comorbid condition—was a predictive factor for mortal-
ity. Previous studies reported that the number of platelets is
lower in candidemia patients who died [17, 18]. Although
the cause-and-effect relationship is disputed, new studies
point out that platelets play a role in defense against Candida
[43, 44]. Eberl etal. [45] evaluated the interaction between
pathogen and platelets as a nuance in Candida infections.
The authors suggested that platelet-targeted immunotherapy
may be intriguing in invasive Candida infections. Similarly,
preventing acute renal injury in patients with candidemia
might be a strategy for survival.
Broad-spectrum antibiotic use was a risk factor for mor-
tality in this study (OR: 9.32, CI: 1.634–53.744, p = 0.012).
Concurrent Gram-negative bacteremia was not associated
with mortality in patients with candidemia, while Gram-
positive bacteremia was associated with mortality in univari-
ate analysis but not in multivariate analysis. TPN is a risk
factor for candidemia development rather than mortality [1].
Similar to our study (OR: 3.9, CI: 1.752–8.810, p = 0.001),
TPN was reported as a risk factor for mortality by Doğan
etal. [11].
The effects of echinocandins and fluconazole treatments
on candidemia mortality have been reported in previous pub-
lications [15, 22, 46, 47]. A recent prospective study from
Turkey noted that researchers used echinocandins as initial
therapy to reduce mortality [11]. IDSA guidelines recom-
mend starting treatment with an echinocandin in clinically
unstable patients [32]. In this study, no difference in terms of
mortality was found between echinocandins and fluconazole
therapies. Guidelines may guide initial treatment choice,
but the median SOFA score was higher in those using flu-
conazole in the current study (data not shown). It has been
reported that the positive effect of echinocandins on mortal-
ity compared to fluconazole disappeared in those with higher
APACHE scores [15].
Similarly, it has been reported that the positive effect of
echinocandins on mortality in patients without septic shock
was not detected in patients with septic shock [48]. The
authors suggested that echinocandins’ pharmacokinetics
can vary due to old age, comorbid conditions, and septic
shock; therefore, less exposure may play a role in this result.
In addition to these explanations, the high rate of C. parap-
silosis may play a role, with no difference between echino-
candins and fluconazole treatments in the current cohort.
Although this species has higher echinocandin MIC values,
it has been shown that treatment with echinocandins does
not have any negative effects on infections caused by this
species [49–51]. Andes etal. [15] reported that the anti-
mortality effect of echinocandins in C. albicans and non-
albicans Candida infections is not present in C. parapsilosis
sub-group analysis. On the other hand, a recent study from
Turkey reported that a higher fluconazole resistance was
observed in C. parapsilosis mostly isolated from pediatric
patients [52]. In our study, C. parapsilosis was also the lead-
ing cause of the antifungal treatment change due to antifun-
gal drug resistance (p = 0.018, Table2). Fluconazole treat-
ment was replaced with an echinocandin or amphotericin B
in 7 of 17 patients with C. parapsilosis candidemia.
The time to start antifungal therapy impacts mortality in
patients with candidemia [2, 53–55]. Bassetti etal. [20] and
Ala-Houhala etal. [56] did not show any effect of early anti-
fungal treatment on mortality. In the current study, the time
of initiation of the antifungal therapy did not differ between
patients who died and patients who survived.
The study has some limitations. First, each center made
its own Candida species determination, although the meth-
ods used for species determination were similar. Second,
some centers used the broth microdilution method in this
study, while other centers performed gradient strip tests
on isolated strains for antifungal susceptibility. Third, an
antifungal susceptibility test was performed on 59.3% of
all strains; therefore, sensitivity results are not presented or
discussed in this study. Fourth, SOFA scores were available
for ICU patients and for patients who transferred to the ICU
within the first 48h after the diagnosis of candidemia. Fifth,
this study was conducted at university hospitals in western
Turkey, so it may not represent Turkey in general or the
country’s other regions.
This study’s strength is its inclusion of 425 consecutive
candidemia cases from 7 university hospitals with simi-
lar candidemia approaches in 6 provinces in 15months.

European Journal of Clinical Microbiology & Infectious Diseases
1 3
Prospective data collection and making a distinction between
pre- and post-candidemia in data collection in this study
enabled some parameters to be evaluated as prognostic or
predictive factors of candidemia.
In conclusion, in the present study, the mortality rate
was 57%, and the C. parapsilosis rate (33.6%) among the
cases of candidemia was high. SOFA score, TPN, previous
antibacterial treatment, newly developed renal failure after
candidemia, and newly developed thrombocytopenia after
candidemia were independent risk factors of 30-day mortal-
ity in this study. The removal of the central venous catheter
was essential to the management of candidemia.
Availability of data and material Data archiving is not mandated but
data will be made available on reasonable request.
Code availability Not applicable.
Declarations
Ethics approval This study was performed in line with the principles
of the Declaration of Helsinki. Approval was granted by the Ethics
Committee of University of Pamukkale ethics committee (date and
no: 14.11.2017–15).
Consent to participate Informed consent was obtained from all indi-
vidual participants included in the study.
Consent for publication Not applicable.
Conflict of interest The authors declare no competing interests.
References
1. Pfaller MA, Diekema DJ (2007) Epidemiology of invasive can-
didiasis: a persistent public health problem. Clin Microbiol Rev
20:133–163
2. Puig-Asensio M, Padilla B, Garnacho-Montero J etal (2014) Epi-
demiology and predictive factors for early and late mortality in
Candida bloodstream infections: a population-based surveillance
in Spain. Clin Microbiol Infect 20:O245-254
3. Toda M, Williams SR, Berkow EL etal (2019) Population-based
active surveillance for culture-confirmed candidemia - four sites,
United States, 2012–2016. MMWR Surveill Summ 68:1–15
4. Koehler P, Stecher M, Cornely OA etal (2019) Morbidity and
mortality of candidaemia in Europe: an epidemiologic meta-
analysis. Clin Microbiol Infect 25:1200–1212
5. Alp S, Arikan-Akdagli S, Gulmez D etal (2015) Epidemiology of
candidaemia in a tertiary care university hospital: 10-year expe-
rience with 381 candidaemia episodes between 2001 and 2010.
Mycoses 58:498–505
6. Baykara N, Akalin H, Arslantas MK etal (2018) Epidemiology
of sepsis in intensive care units in Turkey: a multicenter, point-
prevalence study. Crit Care 22:93
7. Pfaller MA, Diekema DJ, Turnidge JD etal (2021) Twenty years
of the SENTRY Antifungal Surveillance Program: results for Can-
dida species from 1997–2016. Open Forum Infect Dis 6(Suppl
1):S79–S94
8. Guinea J (2014) Global trends in the distribution of Candida
species causing candidemia. Clin Microbiol Infect 20(Suppl
6):5–10
9. Martini C, Torelli R, de Groot T etal (2020) Prevalence and clonal
distribution of azole-resistant Candida parapsilosis isolates caus-
ing bloodstream infections in a Large Italian Hospital. Front Cell
Infect Microbiol 10:232
10. Siopi M, Tarpatzi A, Kalogeropoulou E etal (2020) Epidemiologi-
cal trends of fungemia in Greece with a focus on candidemia dur-
ing the recent financial crisis: a 10-year survey in a tertiary care
academic hospital and review of literature. Antimicrob Agents
Chemother 64:e01516-1519
11. Dogan O, Yesilkaya A, Menekse S etal (2020) Effect of initial
antifungal therapy on mortality among patients with blood-
stream infections with different Candida species and resistance
to antifungal agents: a multicentre observational study by the
Turkish Fungal Infections Study Group. Int J Antimicrob Agents
56:105992
12. Bassetti M, Merelli M, Righi E etal (2013) Epidemiology, species
distribution, antifungal susceptibility, and outcome of candidemia
across five sites in Italy and Spain. J Clin Microbiol 51:4167–4172
13. Seyedmousavi S, Ilkit M, Durdu M etal (2015) Candida and can-
didosis: updates on epidemiology, diagnosis, treatment, antifungal
drug resistance, and host genetic susceptibility. Turk Mikrobiyol
Cemiy Derg 45:1–11
14. Bassetti M, Righi E, Montravers P etal (2018) What has changed
in the treatment of invasive candidiasis? A look at the past 10
years and ahead. J Antimicrob Chemother 73(suppl_1):i14–i25
15. Andes DR, Safdar N, Baddley JW etal (2012) Impact of treat-
ment strategy on outcomes in patients with candidemia and other
forms of invasive candidiasis: a patient-level quantitative review
of randomized trials. Clin Infect Dis 54:1110–1122
16. Lee YM, Kim DY, Kim YJ etal (2019) Clinical impacts of
delayed central venous catheter removal according to the sever-
ity of comorbidities in patients with candidaemia. J Hosp Infect
103:420–427
17. Zhang W, Song X, Wu H etal (2019) Epidemiology, risk factors
and outcomes of Candida albicans vs. non-albicans candidaemia
in adult patients in Northeast. China Epidemiol Infect 147:e277
18. Jia X, Li C, Cao J etal (2018) Clinical characteristics and predic-
tors of mortality in patients with candidemia: a six-year retrospec-
tive study. Eur J Clin Microbiol Infect Dis 37:1717–1724
19. Nucci M, Braga PR, Nouer SA etal (2018) Time of cath-
eter removal in candidemia and mortality. Braz J Infect Dis
22:455–461
20. Bassetti M, Righi E, Ansaldi F etal (2014) A multicenter study
of septic shock due to candidemia: outcomes and predictors of
mortality. Intensive Care Med 40:839–845
21. Garnacho-Montero J, Diaz-Martin A, Garcia-Cabrera E etal
(2013) Impact on hospital mortality of catheter removal and ade-
quate antifungal therapy in Candida spp. bloodstream infections.
J Antimicrob Chemother 68:206–213
22. Colombo AL, Guimarães T, Sukienik T etal (2014) Prognostic
factors and historical trends in the epidemiology of candidemia
in critically ill patients: an analysis of five multicenter studies
sequentially conducted over a 9-year period. Intensive Care Med
40:1489–1498
23. Kollef M, Micek S, Hampton N etal (2012) Septic shock attrib-
uted to Candida infection: importance of empiric therapy and
source control. Clin Infect Dis 54:1739–1746
24. Nucci M, Anaissie E, Betts RF etal (2010) Early removal of cen-
tral venous catheter in patients with candidemia does not improve
outcome: analysis of 842 patients from 2 randomized clinical tri-
als. Clin Infect Dis 51:295–303
25. Jung IY, Jeong SJ, Kim YK etal (2020) A multicenter retrospec-
tive analysis of the antifungal susceptibility patterns of Candida

European Journal of Clinical Microbiology & Infectious Diseases
1 3
species and the predictive factors of mortality in South Korean
patients with candidemia. Medicine (Baltimore) 99:e19494
26. Dizbay M, Fidan I, Kalkanci A etal (2010) High incidence of
Candida parapsilosis candidaemia in non-neutropenic critically
ill patients: epidemiology and antifungal susceptibility. Scand J
Infect Dis 42:114–120
27. Yapar N, Akan M, Avkan-Oğuz V etal (2014) Risk factors, inci-
dence and outcome of candidemia in a Turkish intensive care unit:
a five-year retrospective cohort study. Anaesth Pain Intensive Care
18:265–271
28. Tukenmez Tigen E, Bilgin H, Perk Gurun H etal (2017) Risk
factors, characteristics, and outcomes of candidemia in an adult
intensive care unit in Turkey. Am J Infect Control 45:e61–e63
29. Ulu Kilic A, Alp E, Cevahir F etal (2017) Epidemiology and cost
implications of candidemia, a 6-year analysis from a developing
country. Mycoses 60:198–203
30. Yesilkaya A, Azap O, Aydin M etal (2017) Epidemiology, species
distribution, clinical characteristics and mortality of candidae-
mia in a tertiary care university hospital in Turkey, 2007–2014.
Mycoses 60:433–439
31. Özdemir K, Kocaoğlu C, Sayın Kutlu S etal (2020) Epidemiology,
risk factors and mortality of candidemia: case-control study. Flora
25:206–212
32. Pappas PG, Kauffman CA, Andes DR etal (2016) Clinical prac-
tice guideline for the management of candidiasis: 2016 update
by the Infectious Diseases Society of America. Clin Infect Dis
62:e1-50
33. Janum S, Afshari A (2016) Central venous catheter (CVC)
removal for patients of all ages with candidaemia. Cochrane Data-
base Syst Rev 7:CD011195
34. Yapar N (2014) Epidemiology and risk factors for invasive can-
didiasis. Ther Clin Risk Manag 10:95–105
35. Pfaller M, Neofytos D, Diekema D etal (2012) Epidemiology and
outcomes of candidemia in 3648 patients: data from the Prospec-
tive Antifungal Therapy (PATH Alliance®) registry, 2004–2008.
Diagn Microbiol Infect Dis 74:323–331
36. Kazak E, Akin H, Ener B etal (2014) An investigation of Candida
species isolated from blood cultures during 17 years in a univer-
sity hospital. Mycoses 57:623–629
37. Yapar N, Pullukcu H, Avkan-Oguz V etal (2011) Evaluation of
species distribution and risk factors of candidemia: a multicenter
case-control study. Med Mycol 49:26–31
38. Paiva JA, Pereira JM, Tabah A etal (2016) Characteristics and
risk factors for 28-day mortality of hospital acquired fungemias
in ICUs: data from the EUROBACT study. Crit Care 20:53
39. Groll AH, Shah PM, Mentzel C etal (1996) Trends in the post-
mortem epidemiology of invasive fungal infections at a university
hospital. J Infect 33:23–32
40. Lehrnbecher T, Frank C, Engels K etal (2010) Trends in the post-
mortem epidemiology of invasive fungal infections at a university
hospital. J Infect 61:259–265
41. Jae-Chen S, Young-Joo J, Seon-Min P etal (2015) Mechanism
underlying renal failure caused by pathogenic Candida albicans
infection. Biomed Rep 3:179–182
42. Duggan S, Leonhardt I, Hunniger K etal (2015) Host response
to Candida albicans bloodstream infection and sepsis. Virulence
6:316–326
43. Schultz CM, Goel A, Dunn A etal (2020) Stepping up to the
plate(let) against Candida albicans. Infect Immun 88:e00784-e819
44. Dewi IMW, Aleva FE, Kullaya VI etal (2020) Platelets modu-
late IFN-gamma production against Candida albicans in periph-
eral blood mononuclear cells via prostaglandins. J Immunol
204:122–127
45. Eberl C, Speth C, Jacobsen ID etal (2019) Candida: platelet inter-
action and platelet activity invitro. J Innate Immun 11:52–62
46. López-Cortés LE, Almirante B, Cuenca-Estrella M etal (2016)
Empirical and targeted therapy of candidemia with fluconazole
versus echinocandins: a propensity score-derived analysis of a
population-based, multicentre prospective cohort. Clin Microbiol
Infect 22(733):e1-8
47. Garnacho-Montero J, Díaz-Martín A, Cantón-Bulnes L etal
(2018) Initial antifungal strategy reduces mortality in critically
ill patients with candidemia: a propensity score-adjusted analysis
of a multicenter study. Crit Care Med 46:384–393
48. Falcone M, Tiseo G, Gutiérrez-Gutiérrez B etal (2019) Impact of
initial antifungal therapy on the outcome of patients with candi-
demia and septic shock admitted to medical wards: a propensity
score-adjusted analysis. Open Forum Infect Dis 6:ofz251
49. Lyon GM, Karatela S, Sunay S etal (2010) Antifungal suscepti-
bility testing of Candida isolates from the Candida surveillance
study. J Clin Microbiol 48:1270–1275
50. Fernández-Ruiz M, Aguado JM, Almirante B etal (2014) Initial
use of echinocandins does not negatively influence outcome in
Candida parapsilosis bloodstream infection: a propensity score
analysis. Clin Infect Dis 58:1413–1421
51. Chiotos K, Vendetti N, Zaoutis TE etal (2016) Comparative
effectiveness of echinocandins versus fluconazole therapy for the
treatment of adult candidaemia due to Candida parapsilosis: a
retrospective observational cohort study of the Mycoses Study
Group (MSG-12). J Antimicrob Chemother 71:3536–3539
52. Arastehfar A, Hilmioğlu-Polat S, Daneshnia F etal (2021) Clonal
candidemia outbreak by Candida parapsilosis carrying Y132F
in Turkey: evolution of a persisting challenge. Front Cell Infect
Microbiol 11:676177
53. Parkins MD, Sabuda DM, Elsayed S etal (2007) Adequacy of
empirical antifungal therapy and effect on outcome among
patients with invasive Candida species infections. J Antimicrob
Chemother 60:613–618
54. Taur Y, Cohen N, Dubnow S etal (2010) Effect of antifungal
therapy timing on mortality in cancer patients with candidemia.
Antimicrob Agents Chemother 54:184–190
55. Garey KW, Rege M, Pai MP etal (2006) Time to initiation of flu-
conazole therapy impacts mortality in patients with candidemia:
a multi-institutional study. Clin Infect Dis 43:25–31
56. Ala-Houhala M, Valkonen M, Kolho E etal (2019) Clinical and
microbiological factors associated with mortality in candidemia
in adult patients 2007–2016. Infect Dis (Lond) 51:824–830
Publisher's note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.