Asymptomatic Candiduria among Type 1 and 2 Diabetes Mellitus Patients: Risk and Sociodemographic Factors, Prevalence, Virulence Markers and Antifungal Susceptibility

Diabetes mellitus (DM) has been considered as one of the predisposing factors for candiduria and Candida urinary tract infections. The study determined the socio-demographic characteristics, risk factors of DM patients with asymptomatic candiduria and ascertained the prevalence, virulence factors and antifungal susceptibility of Candida isolated. Socio-demographic and risk factors were obtained via questionnaires. Microscopic, macroscopic and chemical analysis of mid-stream urine (MSU) samples were determined by microbiological method and dipsticks. The characterization, virulence factors, antibiotic susceptibility of Candida isolates were determined by conventional, mycological media and disc diffusion techniques, respectively. Of the 51 MSU samples, ≥ 31.4% were amber and clear in colour, contained yeast cells and leukocytes; between 5.9 to 25.5% had hyaline casts, urobilinogen, epithelial cells, red blood cells, pus cells and nitrite, while the specific gravity was ≥ 1.015. The prevalence of candiduria among subjects with respect to age, types and duration of diabetes, gender, tobacco and alcohol consumption were not significant (p ≥ 0.005). Candida dubliniensis and C. parapsilosis prevalence was highest in subjects with random blood sugar (mg/dL) of ≥ 400 and 300-399, respectively. Of the 39 isolates, 64.1% were Fluconazole sensitive, 10.3% were dose dependent susceptible to Ketoconazole, 74.4% exhibited Voriconazole sensitivity, 100% C. dubliniensis were Clotrimazole sensitive, ≤ 28.6% C. tropicalis and C. glabrata were resistant to Amphotericin B and Itraconazole, while between 23.1% and 71.8% isolates produced hydrolytic enzymes and biofilm. This study revealed the socio-demographic characteristics and risk factors among subjects and the necessity to continuously investigate pathogenic Candida against antifungal agents for effective treatments of asymptomatic candiduria in diabetes

the use of prolonged and inappropriate empirical therapy which has further complicated the management of patients (Chakrabarti et al., 1996). Thus, it's important to appropriately identify the causal agents of the infection and administer proper treatment so as to prevent the disease from becoming chronic.
The virulence factors that enhance the pathogenicity of Candida spp are the: phenotypic switching (Soll, 1992); adhesins and invasins on the cell surface (Akinjogunla et al., 2019); yeast hyphal morphogenetic transformation and hydrolytic enzymes e.g. proteases, phospholipases, haemolysins and lipase (Francois et al., 2013). These hydrolytic enzymes aid in adherence, tissue penetration, invasion and destruction of host tissues (Silva et al., 2011;Akinjogunla et al., 2019). Biofilms, communities of microorganisms embedded in an extracellular matrix, similarly contribute to the pathogenicity of Candida spp and confer substantial resistance to antifungal therapy (Donlan and Costerton, 2002). The study determined the socio-demographic characteristics and risk factors of diabetic patients with asymptomatic candiduria and as well ascertained the prevalence, virulence markers / factors (amylase, lipase, phospholipase, caseinase, gelatinase, coagulase, biofilm, haemolysin) and susceptibility profiles of Candida isolated to antifungal drugs (Ketoconazole, Clotrimazole, Voriconazole, Itraconazole, Fluconazole and Amphotericin) using disc diffusion technique.

Collection of Samples
Mid-stream urine (MSU) samples were aseptically collected using sterile wide-necked, leak-proof containers from 51 diabetic patients (aged < 40 yrs and ≥ 41 yrs), with random blood sugar ranging between 200 and ≥ 400 mg/dL, attending some hospitals / clinics in Uyo, Akwa Ibom State. The Ethics Review Committee approval and verbal informed consent of diabetic patients were obtained prior to samples collection. The MSU samples were properly labelled and transported in a cooler box (4 o C) to microbiology laboratory for analysis.

Administration of Questionnaires
Structured questionnaires reflecting the socio-demographic characteristics and associated risk factors were administered to the participants after their verbal informed consent.

Inclusion Criteria
The diabetic patients who agreed and gave verbal consent to participate in the study.

Exclusion Criteria
The non-diabetic patients and diabetic patients who were on antibiotics within one week prior to enrolment and/or those that declined to participate in the study.

Microscopic and Macroscopic Examination of Mid-stream Urine Samples
Two loopful of each uncentrifuged MSU sample was aseptically placed on a clean grease-free slide and covered with a cover slip. The presence of epithelial cell, pus cell, yeast cell, granular cast, crystals, hyaline cast and red blood cells were observed using 10 x and 40 x objectives with condenser iris sufficiently closed. Macroscopic examination was done by observing colour of the MSU samples as previously described by Akinjogunla and Divine-Anthony (2013).

Chemical (Dipstick) Analysis of Mid-stream Urine samples
The chemical analysis of MSU samples was carried out using dipsticks (Medi-Test Combi 10 SGL, Macherey-Nagel, Germany). The parameters evaluated were: urobilinogen, leukocytes, protein, ketone, bilirubin, nitrites, specific gravity and blood. The dipstick was dipped into each fresh uncentrifuged MSU sample and removed after 5-10 secs. A colour change of the dipstick within 2 mins was observed and recorded according to the manufacturer's instructions.

Mycological Analysis of Mid-stream Urine samples
A loopful of each MSU sample was aseptically inoculated onto each plate of Sabouraud Dextrose Agar (SDA, Oxoid UK) supplemented with chloramphenicol and aerobically incubated at 35°C for 48 hrs. After incubation, the plates were observed for yeast growth and were subcultured onto fresh plates of SDA and incubated at 35°C ± 2 for 48 hr. After incubation, the pure isolates were subcultured onto plates of CHROMagar Candida (Difco BBL., USA), aerobically incubated at 35°C ± 2 for 48 hr and their colonial morphology and pigmentation on CHROMagar Candida were used for identification. The Candida species were further identified using Gram staining, temperature tolerance, chlamydospore and germ tube formation, sugar fermentation and assimilation tests. The Candida isolates were maintained on SDA slant at 4°C.

Detection of Amylase producing Candida isolates
Amylase producing Candida isolates was detected using starch agar (SDA, 2% starch). The Candida isolates were streaked onto starch agar plates and aerobically incubated for 48-72 hr at 35°C ± 2. After incubation, 3 drops of 10 % Lugol iodine was put on the culture plates and allowed to react for 10 min. Clear zone around the colony indicated the production of amylase (Akinjogunla et al., 2107).

Detection of Biofilm Producing Candida isolates
A loopful of each Candida isolate from overnight agar plate was aseptically inoculated onto each tube of Glucose-Sabouraud Dextrose Broth (10 ml SDB, 2% glucose). Each tube was incubated aerobically for 48 hrs at 35°C ± 2. After incubation, the contents of each tube were poured out, each tube washed with phosphate buffer saline (PBS, pH 7.2), dried and stained with safranin (1%). Excess stain in each tube was rinsed with PBS and tubes dried in an inverted position.

Detection of Coagulase Producing Candida isolates
The production of coagulase by Candida isolates was determined using EDTA-human plasma. Ten microlitre of each Candida isolate from overnight agar plate was aseptically inoculated onto each tube of EDTA-human plasma. Each tube was incubated at 35°C ± 2 and observed for clot formation after 2, 4, 6 and 24 hr. The presence of clot that could not be resuspended by gentle shaking indicated positive for coagulase production (Rodrigues et al., 2003).

Statistical Analysis
The Statistical Package for Social Sciences (IBM SPSS Version 22.0. Armonk, NY: IBM Corp.) was used for data analysis. The significant difference in the socio-demographic characteristics and associated risk factors among the diabetic patients at p ≤0.05 were determined using chisquare (χ2) test. Descriptive data were presented as charts and percentages.

RESULTS
The microscopic analysis of MSU samples based on age and gender of subjects are shown on Table 1. Of the 51 samples collected, 13.7%, 25.5%, 35.3%, 9.8%, 13.7%, 5.9% and 11.8% had epithelial cells, pus cells, yeast cells, granular cast, crystals, hyaline casts and red blood cells (RBCs), respectively. The occurrence of epithelial cells, pus cells, yeast cells, crystals, hyaline casts and RBCs was higher in female subjects and those aged ≥ 40 yrs than male subjects and those aged < 40 yrs. Only 5.9% MSU samples of subjects aged ≥ 40 yrs contained hyaline casts, while none of the samples of subjects within age group < 40 yrs had hyaline casts ( Table 1).
The results showed that 64.1% Candida isolates were Fluconazole (FLU) sensitive, while 24.5% were FLU resistant (    (Table 5). Table 6 shows the distribution of virulence factors in Candida isolates from MSU samples of subjects. The percentage occurrence of each virulence factor was as follows: 38.5% amylase, 30.8% lipase, 25.6% coagulase and 23.1% biofilm. A higher prevalence, > 50%, was observed for haemolysin, phospholipase, gelatinase and caseinase (71.8%, 59.0%, 61.5% and 64.1%, respectively). Among candida isolates showing biofilm formation were C. albicans (n=5), C. glabrata (n=2), C. parapsilosis (n=1) and C. tropicalis (n=1) while none of the C. dubliniensis and C. krusei were amylase, coagulase and biofilm producers ( Table 6).     (Matulewicz and Meeks, 2016). In our study, 25.5% DM patients had pus cells in their MSU and this value was < 35.5% reported by Abdulla et al. (2015). The pus cells are white blood cells that have succumbed in defense of the body against pathogens that invade it, therefore their presence indicates microbial infections. The MSU of DM patients in this study had specific gravity of ≤ 1.015 and also contained leukocytes, proteins, ketone, urobilinogen, blood, nitrites and bilirubin. The percentage of MSU with protein (19.6%), ketone (31.4%) and leukocytes (35.3%) in our study was higher than 11.3%, 4.5% and 24.9% for protein, ketone and leukocytes, respectively obtained by Abebe et al. (2019) among DM patients at University of Gondar Hospital, Ethiopia. The presence of nitrite has a predictive value for UTIs (Akinjogunla and Divine-Anthony, 2013) and the value obtained for nitrite (11.8%) in this study was lower than 21.5 % MSU with nitrite reported by Fernandes et al. (2018).
The prevalence of candiduria was more in subjects aged ≥ 40 than in subjects < 40 yrs and this substantiated the earlier report of Hassaneen et al. (2014) and Abebe et al. (2019) that candiduria was more frequent in elderly patients. Several reports have showed frequency of candiduria to be more in diabetic women than in diabetic men (Janifer et al., 2009;Abdulla et al., 2015) and this study similarly confirmed these reports. The hypertensive subjects had higher prevalence of candiduria than non-hypertensive subjects in our study and this was in conformity with a study conducted by Muller et al. (2007) in which diabetic patients were more at risk for candiduria / UTI than hypertensive patients without diabetes. Other predisposing factors such as duration of diabetes, family history of diabetes, alcohol in-take and tobacco consumption were observed to be associated with occurrence of candiduria in this study. The prevalence of candiduria was more in subjects with duration of diabetes ≥ 5 yrs than those subjects with duration of diabetes < 5 yrs and this was similar to the findings of Janifer et al. ( (Sanglard and Odds, 2002), over-expression of efflux proteins which act by pumping the drug out of the cell at a rate faster rate than how the drug enters the cell (Akinjogunla and Eghafona, 2012) and modifications of target enzymes have been attributed to the resistance of Candida spp to azole group of antifungal agents (Silva et al., 2011). The transition of Candida isolate from a harmless commensal to potent pathogen is determined by several host predisposing factors and virulence attributes of organisms. The pathogenicity of Candida isolates is attributable to secretion of extracellular hydrolytic enzymes that act synergistically under favourable conditions (Silva et al., 2011). The occurrence of haemolysin, lipase and phospholipase producing Candida spp in this study corroborated the reports of Saha (2018) and Akinjogunla et al. (2019). Phospholipase facilitates Candida invasion by damaging host cellular contents and cleaves phospholipids of host cell membrane (Deorukhkar et al., 2014). Haemolysin aids in lysis of host erythrocytes and strips iron from haemoglobin molecules (Manns et al. 1994). Candida albicans and NAC exhibited biofilm formation and the occurrence of biofilm producing Candida spp agreed with Seneviratne et al. (2008). In our study, 14.3% coagulase producing C. tropicalis obtained was similar to 14.6 % by Deorukhkar et al. (2014), but < 82.6% reported by Rodrigues et al. (2003). Coagulase binds plasma fibrinogen and activates a cascade of reactions that induce clotting of plasma (Rodrigues et al., 2003).

COnCLUSIOn
This study has shown the sociodemographic characteristics and risk factors in diabetic mellitus patients with or without candiduria; revealed the high sensitivity of Candida isolates to azole drugs and varied percentages of haemolysin, coagulase, biofilm, phospholipase and lipase producing Candida isolates in MSU samples and as well showed the necessity to continuously investigate pathogenic candida isolates against appropriate antifungal agents for effective treatments of asymptomatic candiduria in type 1 and 2 diabetes mellitus patients.