Molecular Identication, And Antifungal Susceptibility Patterns of Candida Species Isolated From Candidemia Patients in Yasuj, Southwestern Iran

Candidemia is the most common systemic infection in hospitalized patients and causing high mortality. Hence, the diagnosis of this infection in the early-stage with appropriate antifungal therapy has been attributed to the lowest mortality. The aims of this study are molecular identication of Candida species isolated from candidemia patients and evaluated the in vitro antifungal susceptibility patterns of these strains to uconazole, amphotericin B, and caspofungin. In the present study, 800 hospitalized patients who suspected candidemia were sampled. Candida species were isolated and identied based on morphological and PCR-sequencing of the ITS1-5.8S-ITS2 region. Antifungal susceptibility tests for uconazole, amphotericin B, and caspofungin was performed according to the Clinical and Laboratory Standards Institute M27-A3. Also, clinical data were recorded from patient's records. Overall, 27 candidemia patients were detected among hospitalized patients. 33.3% of candidemia patients were treated with amphotericin B, however, the mortality rate was 14.8%. The majority of patients (59%) were from the neonatal intensive care unit and premature born was the most underlying condition. C. albicans (n = 18; 66.6%) was the most common species isolated from blood cultures, followed by C.parapsilosis (n = 7; 25.9%), C.pelliculosa ( n = 1;3.7% ) and C.tropicalis (n = 1;3.7%). Only one C. albicans isolate were resistance to uconazole (MIC = 32 µg/mL). Generally, C. albicans has been the most frequent causative agent of candidemia. Resistance to antifungal drugs among candidemia agents was rare. Also, the identication of Candida isolates at the species level with in-vitro antifungal susceptibility tests can manage and decrease the mortality rate among candidemia patients.


Introduction
One of the most prevalent causes of hospital-acquired infections, more speci cally in severe cases, is systematic fungal infections. Candida species are the most common causative agents of these infections. Candidemia is the most common systemic infection in hospitalized patients and causing high mortality (1). Prolonged antibiotic therapy, immune dysfunction, corticosteroids, renal failure, and dialysis are introduced as the predisposing risk factors for candidemia (2). Candida albicans is the most common agent of candidemia. However, the prevalence of candidemia caused by non-albicans species including C. tropicalis, C. glabrata, C. krusei, and C. parapsilosis is increasing (3)(4)(5)(6). The change in the epidemiology of Candida infection has contributed to the development of antifungal resistance so that many Candida species are resistant to uconazole (FLZ) (6, 7). Resistance to antifungal agents can increase the risk of treatment failure, mortality, and prolonged hospitalization (8). Therefore, accurate diagnosis of the fungal infection, identi cation of the species, monitoring incidence, and in vitro antifungal susceptibility pro les are essential to managing systemic Candida infection (9). DNA-based methods such as PCR-sequencing, and restriction fragment length polymorphism (RFLP) are reliable methods for the identi cation of Candida species (10,11).
The present study was assessed to identify Candida species isolated from candidemia patients using molecular methods. Also, the antifungal susceptibility pro les of these species were evaluated.

Clinical samples and Candida isolation
This study focused on patients with candidemia admitted to the medical centers in Yasuj, Shahid Beheshti, Imam Sajjad, and Provincial reference laboratory from November 2016 to November 2019.
During the study period, 800 blood samples were collected from patients and inoculated into a biphasic blood culture medium (Tebo Sadegh, Iran) and incubated at 37 ºC aerobically for more than 7 days.
During this period, one ml of each blood culture was subcultured on CHOROM agar Candida (CHOROM agar, France) and incubated at 37 ºC. After that, Candida species were initially identi ed based on classical methods. Molecular identi cation DNA extraction DNA was extracted and puri ed using the boiling method described by Tay et al,. (12). Brie y, a loop full of a fresh colony of each yeast isolates was added to a 1.5 ml tube containing 300 µl of distilled water.
The tubes were placed in boiling water for 20 min. After that, tubes vortexed for 5 min and centrifuged at 8000 rpm for 1 min. Finally, supernatant (DNA) was separated and stored in -20 ºC and used for PCR ampli cation.

PCR-sequencing
The ITS1-5.8S-ITS2 region of rDNA complex was ampli ed using V9g/LS266 primers (V9g, 5′-TTACGTCCCTGCCCTTTGTA-3′, and LS266, 5′-TCCTCCGCTTATTGATATGC-3′) for all strains. Then, the PCR products were sequenced, manually veri ed, and aligned by the MEGA6 software. All sequence data were compared to reference sequences in the GenBank (NCBI) and CBS database via the nucleotide BLAST ™ algorithm to obtained a de nitive identi cation (similarity values ≥ 99%). Finally, all nucleotide sequences representative were submitted in the GenBank database and accession numbers were obtained.

Antifungal susceptibility testing (AFST)
The antifungal assay was performed according to CLSI protocol (M27-A3) (13). Brie y, overnight cultures of Candida species on Sabouraud Dextrose Agar (SDA) (Merck, Germany) were used for this purpose. Suspensions were adjusted to 0.5 McFarland and then diluted 1:1000 (1:20 and 1:50). The serial dilutions of the antifungal agents were prepared from 0.0625to 32 µg/ml for FLZ, 0.03-16 µg/ml, and 0.015-8 µg/ml for AMB and CAS, respectively (13). 100 µL of yeast suspension and 100 µL of serial dilution of each tested antifungal were added into each well of the microplate. Microplates were incubated at 35ºC for 24-48 h. Then the minimum inhibitory concentration (MIC) range, MIC 50 , MIC 90 , and MIC GM were calculated. De ned CLSI guidelines breakpoints were used for susceptible, dose-dependent, and resistance (14). Also, there is no clinical breakpoint for amphotericin B so, an epidemiological cut-off value (ECV) was used for amphotericin B (15) ( Table 1). Besides, C. parapsilosis ATCC 22019 was used for quality control.
The data were analyzed using the Chi-Square and Kruskal Wallis one-way ANOVA tests by SPSS version 21.0. P-value was considered a default value of 0.05 for signi cance.

Results
A total of 800 patients were suspected of sepsis and candidemia enrolled in this study. Among these patients, 27 patients with candidemia were based on mycological criteria. A total of these patients, 18 (66.7%) and 9 (33.3%) were male and female, respectively (Table 2). Also, the age range of these patients was varied from three days in the neonate to 80 years in adult patients. 51.8% of positive candidemia patients had a history of received antimicrobial drugs such as vancomycin, amikacin, gentamycin, and cipro oxacin. Over 59% of the yeast isolates recovered from the neonatal intensive care unit (NICU) and the rest (33.3%) were obtained from the intensive care unit (ICU). Among the underlying disease, premature borne (16,59.3%) and diabetes (4, 14.8%) were the most frequent conditions in admitted patients. 33.3% of candidemia patients were treated with amphotericin B, followed by uconazole and a combination used with both of these drugs. The mortality rate was obtained at 14.8% and associated with C. albicans.  According to sequence analysis, C. albicans (n = 18; 66.6%) was the most common isolated species followed by C. parapsilosis (n = 7; 25.9%), C. pelliculosa ( n = 1; 3.7% ), and C. tropicalis (n = 1; 3.7%).
The minimum inhibitory concentration (MIC) range of each isolate and MIC range, MIC 50 , MIC 90, and MIC GM for all isolates were calculated and have been shown in Table 3. As shown, exception one C.

Discussion
Candidemia is one of the nosocomial systemic infections with a high mortality rate, especially in neonates and pediatric patients (16). The prevalence of candidemia in published reports was varied (17)(18)(19). In our study, the incidence of candidemia was 3.4% among hospitalized patients with positive blood cultures. Agree with our study, this rate was reported to be 3.5% in Motta et al,from Brazil (20). However, the incidence of candidemia in published reports from the US and Europe was lower than from our study (18,19). Amphotericin B was the most widely used antifungal drug. However, 40.7% of our patients were not treated with antifungal drugs. This result disagrees with other studies that reported uconazole was the most antifungal drug received in candidemia patients (21,22). The low mortality rate (14.8%) were obtained in our study was mostly lower than other reported in Iran (28-47%) (21,22). It seems that this difference may due to different conditions of candidemia patients such as underlying disease, antifungal therapy, the type of Candida species, and the length of hospitalization. In the present study, C. albicans was the only cause of mortality in candidemia patients. However, C. tropicalis was reported for the high mortality rates in another study (23,24).
This study showed that C. albicans was the most frequent species causing Candidemia (%78.5). These results are in agreement with the other studies conducted in Iran, such as Razzaghi et al. and Sadrossadati et al,. who reported that C. albicans was the most common (25,26). The present study is also compatible with the other studies reported from different regions of the world (27,28). However, in recent years, the emergence of infections caused by non-albicans species increased as the common etiology of candidemia (29,30). In studies such as those performed by Chander et al. in India (31), and Ghahri et al. in Iran (29), C. tropicalis (40.8%) and C. parapsilosis (34.4%) were found to be the most frequent cause of candidemia, respectively. The reason for the emergence of non-albicans species can be associated with some underlying conditions which were different for each species (25). In our study, the second most frequent Candida species were C. parapsilosis. Similar to our results Sadrossadati et al. (26) and Kooshki et al. (32) revealed that C. parapsilosis was the second most prevalent isolated species. However, Arastefar et al, showed that C. glabrata was the second agent of candidemia in Shiraz (22). This contrast may be due to the difference age groups of patients.
In our study, the MIC range for FLZ was evaluated between 0.0625 and 32 g/ml for the 27 Candida strains. Besides, resistance to uconazole was detected in 5.6% of C. albicans strains, however, all nonalbicans strains were sensitive to this drug. Similarly, Arastehfar et al., were reported that 4.42% of Candida strains isolated to blood culture resistant to uconazole (22). Although initially, it seems that triazoles are highly e cient, but overusing them is associated with azole resistance in Candida species (33 Even though amphotericin B is one of the most toxic antifungal drugs in clinical use, it is still considered a standard and inexpensive treatment (25,36). In this study, 100% C. albicans strains were wild-type to amphotericin B. This result similar to Arastehfar et al,. shows that all candidemia isolates were wild-type against amphotericin B (22). In international studies such as Motta et al, also, amphotericin B resistance among Candida bloodstream isolates was rare (20). Therefore, it seems that amphotericin B can be a choice of therapeutic agent for patients with candidemia.
Echinocandins are generally effective against candidemia and it has been shown that their therapeutic use is in line with low mortality (1,37). Among echinocandin antifungal agents, caspofungin is more used in Iran (38). This drug has been demonstrated as the most effective against resistance Candida species to other antifungals (25). In this study, antifungal susceptibility tests indicate that caspofungin was the most e cient drug against Candida species (MIC ≤ 1 g/mL). In the other studies carried out in Iran, different rates of resistance have been reported. The susceptibility pro le of caspofungin in our study was similar to that reported in Iran and other countries (20,25). In some studies, resistance to this drug has been observed in non-albicans species, such as C. parapsilosis, and C. krusei (39,40).

Conclusion
In conclusion, our study demonstrated that C. albicans has been the most frequent agent of candidemia in the southwest of Iran. Resistance to antifungal drugs among candidemia agents was rare. Also, the identi cation of Candida isolates at the species level with in-vitro antifungal susceptibility tests can manage and decrease the mortality rate among candidemia patients.

Declarations Acknowledgment
This study was the result of a thesis nancially supported by Yasuj University of Medical Sciences, Yasuj, Iran.
Compliance with ethical standards