Dissertation committee:
κ .Νικόλαος Κατσιλάμπρος, Ομ. Καθηγητής, κ. Γεώργιος Πετρίκκος, Καθηγητής, (Επιβλέπων), κ. Γεώργιος Δαΐκος, Αναπληρωτής Καθηγητής
Summary:
SUMMARY
There has been a marked increase of invasive Candida infections in recent
years and these fungi are nowadays important hospital pathogens in
immunocompromised or critically ill patients. Species resistant to fluconazole
as well as C. parapsilosis which are less sensitive to echinocandins are also on
the rise. Moreover, there are very few data about the risk of developing
candidemia after candiduria. The aims of the current study were:
1. The identification of all Candida isolates responsible for candidaemia
cases diagnosed in a large tertiary hospital and the correlation between
species distribution and underlying disease or various risk factors.
2. The assessment of the sensitivity of the isolated Candida strains to
antifungal drugs.
3. The assessment of the outcome of patients with candidemia and its
association with the different Candida species.
4. The genotypic analysis of the Candida strains isolated from patients with
candiduria and concurrent candidemia, in order to determine the
frequency of this clinical condition and assess the associated risk
factors.
This dissertation is based on a prospective non-invasive study which took
place in Laikon General Hospital of Athens, between 01-01-2006 and 31-12-
2010. According to our protocol, the patients enrolled in our study were adults
who had a positive blood culture for Candida spp. In addition, it was
investigated whether they had at least one positive for Candida spp. urine
culture two weeks previous to or one week following the positive Candida
blood culture. Recorded data included demographics, the date of admission,
the hospital unit of admission and care, the main and underlying disease, any
risk factors, the cause of admission, any surgical interventions, the immunity
status, the presence of a central venous catheter and site of insertion, the
presence of urine catheter, the administration of parenteral nutrition, the
presence of other Candida infection sites, the presence of other co-infections,
the use of any anti-microbial agents, the former use of antifungal agents, as
well as the treatment and outcome.
Species were identified using API 32C (bioMerieux). The microdilution
colorimetric method Sensititre (Trek Diagnostic Systems Inc. Cleveland, OH)
was employed in order to assess the sensitivity of the isolates to antifungal
agents. The antifungal agents included in this study were the following:
amphotericin B, fluorocytosine, fluconazole, voriconazole, itraconazole,
posaconazole, caspofungin, anidulafungin και micafungin.
For the assessment of the epidemiological relatedness of the strains isolated
from blood and urine of the same patient, a modified, improved version of
pulse-field gel electrophoresis (PFGE) was utilised, after optimization for each
Candida species.
During our study, 141 candidaemia cases were identified, in a total of 134
patients. Twelve different species were isolated with the following frequencies:
51 C. albicans (36.2%), 43 C. parapsilosis (30.5%), 22 C. glabrata (15.6%) and
from the more rare species: 6 C. krusei (4.3%), 6 C. tropicalis (4.3%), 3 C.
sake (2.1%), 2 C. intermedia (1.4%), 2 C. dublinensis (1.4%), 2 Pichia
etchell/carsonii (1.4%), 1 (0.7%) C. famata, 1 C. lusitaniae και 1 C.
guillermondii.
The rare Candida species were more frequently isolated from patients with
haematologic malignancies (41.2% versus 14.5% in non-haematological, p=
0.013). In surgical units, the most frequent species isolated was C.
parapsilosis, a result that was found to be statistical significant (44.4%
compared to 21.8 % in all other units, p= 0,008). The PFGE genotypic analysis
of these isolates revealed that most of them were related, clustered in two
main genotypes, repeatedly appearing during the study period.
Twelve cases of candidemia with concurrent candiduria were observed in
eleven patients (8%). In 6 out of these cases, the fungi isolated from blood
and urine belonged to different Candida species. In two cases the Candida
strains belonged to the same species but they were not PFGE related. Only in
four patients the same Candida species was identified in both blood and urine
samples, as revealed by the identical PFGE patterns.
The observed resistance to antifungal agents was very low. The crude
mortality rate was 32.8% and variation amongst the clinical groups was
between 16.7% and 50%, with most deaths observed in haematological
patients. Regardinge hospitalisation unit, most deaths occurred in the ΙCU.
In summary the conclusions of this study are the following:
1. Non-albicans Candida species were the most frequently identified ones
in our hospital during the study period. However, their sensitivity to
commonly used antifungal agents remained at acceptable levels.
2. Patients with haematologic malignancies have increased possibility of
developing infections caused by rare species of Candida.
3. The risk of developing candidemia after candiduria appears to be
extremely low.
4. Patients with diseases of the gastrointestinal tract, especially those who
have undergone abdominal surgery and are hospitalised in surgical
units have increased risk of developing candidaemia caused by C.
parapsilosis, mainly nosocomial strains.
The last observation highlights the importance of strict compliance with
hygiene rules. It can be hypothesized that under ideal conditions, a substantial
number of these infections could be prevented.
