Retrospective observation of drug susceptibility of Candida strains in the years 1999, 2004, and 2015

Introduction There is much literature devoted to the problem of drug resistance and decreased susceptibility of fungi to commonly used antifungals. Aim To analyze drug susceptibility of Candida albicans and non-Candida albicans strains isolated from the hands of people without any symptoms of disease over a 16-year period. Materials and Methods The study included a total of 1,274 Candida-type strains isolated from the hands of people without any symptoms of disease, including: in 1999, 432 strains; in 2004, 368; and in 2015, 454 strains. Biological monitoring of hand surface contamination was performed using the Count-TactTM applicator with Count-Tact plates (bioMerieux). Drug susceptibility was evaluated using FUNGITEST®. Results In 1999, the most strains showed resistance to fluconazole (53.2%), in 2004 to itraconazole (52.9%), and in 2015 to fluconazole (85.8%). Resistance to more than one drug was 35.8% in 1999, 64.7% in 2004, and 92% in 2015. Mean resistance to azole antifungals significantly increased from 98 ± 39.7 strains in 1999 to 118.3 ± 29.6 in 2015 (p < 0.001). In 1999, the most strains showed resistance to fluconazole (50.6%), in 2004 to itraconazole (52.9%), and in 2015 to fluconazole (44.9%). Resistance to more than one drug was 52.9% in 1999, 64.3% in 2004, and 88.1% in 2015. Mean resistance to azole antifungals significantly increased from a mean of 76 ± 9.7 strains in 1999, to 95.3 ± 24.2 in 2004, and to 97.3 ± 16.6 in 2015 (p < 0.001). Conclusions We showed increased C. albicans and non-Candida albicans strain resistance to commonly used antifungal chemotherapeutics, mainly imidazole. We found a clear rise in susceptibility of C. albicans and non-Candida albicans strains to several studied antifungals.


INTRODUCTION
A phenomenon constantly emphasized in the literature is the emergence of yeast-like fungi strains resistant to individual drugs or whole groups of antifungals (Pfaller et al., 1998;Tsai et al., 2006;Bailly et al., 2016;Ben-Ami et al., 2016;Zaidi et al., 2016;Sanglard, 2016). There are two types of resistance in fungi: primary and secondary (induced) (Sanguinetti, Posteraro & Lass-Flörl, 2015). Primary resistance pertains to cells with high Minimal Inhibitory Concentration (MIC) values which had no prior contact with a given antimycotic drug. Secondary resistance is associated with strains that were originally susceptible and acquired resistance through induction (or selection of naturally resistant mutants). Cases of primary (about 10-20% strains) and secondary resistance have been widely described in relation to 5-fluorocytosine. Candida glabrata and Candida krusei, for example, show primary resistance to fluconazole (Richardson & Warnock, 1995). The most common strains resistant to amphotericin B are Candida parapsilosis, Candida lusitaniae, Candida quillermondii, Candida tropicalis, and Candida krusei. According to some authors, increased drug resistance could be associated with improper treatment (Vanden Bossche et al., 1998).
Candida strain antifungal resistance is the result of antimycotic drug use for treating fungal infections, especially during preventive and empiric therapy, causing resistant strain selection (Tsimbalari et al., 2015). Most likely to occur during therapy, yeast-like fungi, which are etiologic factors of infections, may be replaced by other species derived from hospital flora, resistant to the drugs used. Candida spp. clinical isolates commonly exhibit a high inherent tolerance level to azole antimycotics. Azole resistance can be acquired through an increased expression of genes encoding ABC transporters (Cdr1, Pdh1, Snq2) or changes in their transcriptional regulatory system (Pdr1, Gal11) (Thakur et al., 2008). Mitochondrial dysfunction and serum utilization via the putative sterol transporter Aus1 also impact the ability of Candida spp. to tolerate high azole levels (Brun et al., 2004;Nagi et al., 2013). In addition, calcineurin signaling has been implicated in azole tolerance in Candida. Therefore, research on the acquisition of genes determining resistance is becoming more common (Macura & Skóra, 2012;Kabir & Ahmad, 2013). There is also a need to observe and monitor changes in antifungal activity.
While antibiotic-resistant bacterial infections are a widely-recognized public health threat, less is known about the effects of antifungal resistance and the burdens caused by drug-resistant fungal infections. These invasive infections cause considerable morbidity and mortality and are common problems in healthcare settings (Magill et al., 2014;Mohamadi et al., 2015). The fungus Candida is the most common cause of healthcare-associated bloodstream infections in the United States (Magill et al., 2014). Each case of Candida infection of the bloodstream is estimated to result in an additional three to 13 days of hospitalization and thus increases healthcare costs significantly (Morgan et al., 2005). Even with current antifungal therapy, mortality associated with candidiasis can be as high as 50% in adults and up to 30% in children (Moran et al., 2009).
It is known that the hands of healthcare workers are responsible for 20-40% of nosocomial infections (Rożkiewicz, 2011). Moreover, the hands of medical students can be a route of transmission of microorganisms.
Healthy people and healthcare workers can also carry Candida on their hands (Yildirim et al., 2007). There have been outbreaks of candidemia linked to healthcare workers' hands (Clark et al., 2004); therefore, hand hygiene in healthy people and healthcare settings is critical for preventing the spread of infections.
Thus, monitoring people's hands for the presence of Candida spp. is important. The aim of this study was to analyze, over the course of sixteen years, drug susceptibility and resistance of yeast-like fungi strains isolated from the hands of people with no disease symptoms.

MATERIALS
Participation in the study was voluntary. A total of 667 students of the Medical University of Białystok, Poland took part in the study. Respondents' ages ranged from 19 to 25 (22 ± 2.3). The dominant hands of the participants without clinical symptoms were sampled in the morning.
The study included a total of 1,274 Candida-type strains isolated from the hands of people without any symptoms of disease, including the following. In 1999, 432 strains from 229 students; in 2004, 368 from 198 students; and in 2015, 454 strains from 240 students were isolated. To facilitate the analysis of statistical dependence, the studied strains were divided into two groups: I-C. albicans and II-non-Candida albicans.

Identification yeast-like fungi
Biological monitoring of hand surface contamination was performed using the Count-Tact TM applicator with Count-Tact plates (bioMerieux, Marcy l'Etoile, France) containing a medium complying with the Draft European Standard CEN/TC 243/WG2 requirements. CandiSelect (Bio-Rad, Hemel Hempstead, UK) was used to identify yeast-like fungi. It is a selective chromogenic medium designed for the isolation of yeasts, the direct identification of C. albicans and the presumptive identification of C. tropicalis, C. glabrata and C. krusei.
After sampling, Count-Tact plates were incubated at 37C up to 48 h. Next, the fungal culture was inoculated into CandiSelect. The plates were incubated for 72 h to allow for identification. The color and intensity of colonies was assessed every 24 h. C. albicans produced pink to purple colored colonies; C. glabrata produced turquoise colonies that were shiny and flat with a regular outline; C. tropicalis produced intense turquoise colored colonies that were spherical with a regular outline; C. krusei produced large turquoise colonies with a dry appearance and an irregular outline.

Drug concentration
Drug susceptibility was assessed using FUNGITEST R (Sanofi Diagnostics Pasteur, Paris, France in the years, 1999, 2004 since 2015; Bio-Rad, Marnes-la-Coquette, France) (Witthuhn et al., 1999) to analyze yeast-like fungal growth in the presence of two concentrations of six drugs: 5-fluorocytosine, amphotericin B, miconazole, ketoconazole, itraconazole and fluconazole, in modified RPMI 1640 medium, in the presence of a redox indicator. Growth assessment is based on the reduction of the colored indicator which turns the medium from blue to pink. When growth is inhibited by the antifungal agent, the medium remains blue. This test, presented in the form of a 16-well microplate, consists of two growth control wells; 12 wells containing the dehydrated antifungal agents (six antifungal agents at two different concentrations); 5-fluorocytosine (2-32 µg/ml), amphotericin B (2-8 µg/ml), miconazole (0.5-8 µg/ml) kétoconazole (0.5-4 µg/ml), itraconazole (0.5-4 µg/ml), fluconazole (8-64 µg/ml); two negative control wells. The breakpoints have been chosen following the study of the distribution of the antifungal agents M.I.C's obtained with prototype microplates used with the same procedure as Fungitest. Results were interpreted in accordance with the manufacturer's instructions, always by the same person, and always with reference to the color of two wells containing the same drug: a blue color in both wells indicated an in vitro susceptible strain; a pink color at lower concentrations and a blue color at higher concentrations indicated an in vitro strain with low susceptibility; and a pink color in both wells indicated an in vitro resistant strain. The study has been accepted by ethic committee of the Medical University of Białystok, Poland, approval numbers: R-I-003/64/99; R-I-003/222/2004, and RI-002/489/2010.

STATISTICAL ANALYSIS
Statistical analysis of the results was done using the chi 2 test and the Kruskal-Wallis test on Statistica 10.0 software.
In 1999, the C. albicans were most resistant to fluconazole (53.2%), and the least resistant to amphotericin B (1.1%). In 2004, we found the C. albicans were most resistant to itraconazole (52.9%), and least resistant to amphotericin B (3.5%). In 2015, the most C. albicans showed resistance to fluconazole (85.8%), and least resistance to amphotericin B (17%). Results are presented in Table 2. We observed statistically significant differences between the analyzed years pertaining to the studied C. albicans strains: • in susceptibility to amphotericin B (p = 0.011), ketoconazole (p = 0.016), itraconazole (p < 0.001), and fluconazole (p < 0.001) • in decreased susceptibility to all drugs • in resistance to all drugs.
Results are presented in Table 2.
In 1999, the non-Candida albicans was most resistant to ketoconazole (72.9%), and least resistant to amphotericin B (10.6%). In 2004, we detected that non-Candida albicans were most resistant to itraconazole (58.2%), and least resistant to 5-fluorocytosine (5.1%). In 2015, the non-Candida albicans showed most resistance to fluconazole (44.9%), and the least resistant to amphotericin B (1.4%). Results are presented in Table 3.
Results are presented in Table 3. Generally, 60.2% of C. albicans strains showed resistance to one or more of the studied antifungals, and most often to three drugs (24%). Resistance to more than one drug was 35.8% in 1999, 64.7% in 2004, and 92% in 2015. Results are presented in Table 4.
Generally, 71.8% of non-Candida albicans strains showed resistance to one or more of the studied antifungals, and most often to one (26%) or two drugs (25.7%). Resistance to   Table 5. We found significant differences between the analyzed years and resistance of the studied C. albicans and non-Candida albicans strains to one or more antifungal drugs (p < 0001). Generally, 295 ± 71.3 C. albicans strains were resistant to all studied antifungals. The mean resistance of C. albicans strains to antifungals significantly increased: 98 ± 39.7 in 1999, 118.3 ± 29.6 in 2015 (p < 0.001). Details are shown in File S1.
Generally, 259 ± 50.2 of non-Candida albicans strains were resistant to the studied antifungals. The mean resistance of non-Candida albicans to the studied antifungals  1999, 58.2% in 2004, and decreased to 33.5% in 2015 (p = 0.07) and Fluconazole 45.9% of resistant strains in 2004, 52% in 2015, and decreased to 44.9% in 2015 (p = 0.999). We found significant (p < 0.001) differences in resistance to the studied antifungals between C. albicans and non-Candida albicans strains.
In our study, C. albicans strain resistance to amphotericin B significantly increased. This strain resistance pertained to 6.2% strains; however, over the course of 16 years, it increased from 1.1% in 1999, to 3.5% in 2004, followed by 17% in 2015. In the case of non-Candida albicans, resistance to this drug significantly decreased. This resistance pertained to 5.9% strains; over the course of 16 years, it decreased from 10. 6% in 1999, through 8.2% in 2004, to 1.4% in 2015. According to Vanden Bossche et al. (1998), the resistance of Candida spp. strains to 5-fluorocytosine develops during the course of monotherapy using this drug. The authors observed a simultaneous increase in resistance to 5-fluorocytosine and itraconazole among strains. In the study by Cybulski et al. (2003), resistance to 5-fluorocytosine was determined in 8.3% C. glabrata strain isolates from the hospital patients. Oberoi et al. (2012) emphasized that the incidence of candidemia caused by non-albicans infections has increased significantly and is correlated with the increased use of fluconazole. The authors observed a cross-resistance or reduced susceptibility to fluconazole and voriconazole in 11.3% of isolates. Macura & Skóra (2012) assessed the susceptibility of fungi isolated from the vagina to six antifungals (5-fluorocytosine, amphotericin B, miconazole, ketoconazole, itraconazole, and fluconazole) using the Fungitest. The authors found that C. krusei had the highest resistance to antifungal drugs, including fluconazole. Of 23 C. krusei strains isolated from patients with the suspicion of vaginal mycosis, 4.3% showed susceptibility and 87% moderate susceptibility to 5-fluorocytosine; 8.7% susceptibility and 69.6% moderate susceptibility to fluconazole; while 17.4% showed resistance to amphotericin B.
The observations by Witthuhn et al. (1999) on the high resistance to fluconazole are interesting. The authors found 77% of strains were susceptible to fluconazole and 84% to itraconazole in HIV-infected patients.
In the present study, we found a rise in C. albicans strains resistant to fluconazole from 53. 2% in 1999, through 41.9% in 2004, to 67.1% in 2015; and to itraconazole from 43.6% in 1999, through 52.9% in 2004, to 81.8% in 2015. In the case of non-Candia albicans strains, resistance to fluconazole pertained to 45.9% strains in 1999, 52% in 2004, and 44.9% in 2015. In their analysis of susceptibility of C. albicans isolated in the years 1984-1993and 1984-1997, Boschman et al. (1998 found that, up until 1993, all isolates showed susceptibility to fluconazole, ketoconazole, and miconazole. Since 1994, new drug resistant strains have gradually emerged.
We have also observed a significant rise in the percentage of resistance to all imidazole agents. In the case of C. albicans strains, an average of 295 ± 71.3 strains showed resistance to imidazoles, including mean 98 ± 39.7 in 1999, 78.8 ± 10.1 in 2004, and 118.3 ± 29.6 in 2015. In the case of non-Candida albicans strains, an average of 249 ± 39.6 strains showed resistance to imidazoles, including mean 76 ± 9.7 in 1999, 95.3 ± 24.2 in 2004, and 97.3 ± 16.6 in 2015.
It is alarming that 60.2% C. albicans and 78.1% non-Candida albicans strains showed resistance to one or more of the studied antifungals. C. albicans strains were most often resistant to three drugs, and resistance to more than one drug increased from 35.8% in 1999 to 92% in 2015. Non-Candida albicans strains were most often resistant to one drug (26%), and resistance to more than one drug rose from 52.9% in 1999 to 88.1% in 2015.