Association of diabetes mellitus on the development of oral candidiasis

Study setting

A total of 200 subjects were recruited. Confirmed DM patients (n = 100) aged between 20 and 60 years, confirmed endocrine patients (n = 50) (not suffering from DM) aged between 20 and 60, who were registered at the diabetic and endocrinology clinic, Colombo South Teaching Hospital (CSTH), Kalubowila, Sri Lanka were selected as the patient group and the comparative patient group, respectively. Apparently, healthy nursing staff (n = 50) aged between 20 and 60, who were employed at the time of sampling at the same hospital, were selected as the apparently healthy group.

All study participants were randomly selected for the study after obtaining informed consent and they were requested to attend the next clinic with 8–10 h fasting. DM patients and non-diabetic patients who had undergone specific treatment such as anti-neoplastic therapy, those who are diagnosed with HIV/AIDS and other oral cavity infections, patients above 60 years of age or pregnant, and patients undergoing treatments for any chronic disorders (e.g., thyroid disorders, hypertension, dyslipidemia, and hyperlipidemia) were excluded from the study.

Sample collection

All individuals, including patients and healthy groups, were investigated and confirmed by the medical professionals. Oral rinse samples were collected from all after rinsing their mouth thoroughly using 10 mL of sterile phosphate-buffered saline (1M; pH 7.2) for analysis of the presence of fungi species. Samples were collected in to sterile plastic culture containers, after rinsing the mouth thoroughly for 1 minute.^[13] Then, samples were transported to the laboratory within 2 h at 4°C in an iced box. Each oral rinse samples were centrifuged at 5000 rpm for 10 min. The deposit (about 1 microliter amount) was inoculated onto a Sabouraud dextrose agar medium with chloramphenicol and incubated at 35°C for 24 h. Candida growth was morphologically identified as smooth, white or creamy-colored buttery colonies. The individual CFUs were counted manually. The samples which were negative for Candida growth after the first 24 h of incubation were re-incubated at 35°C for a further 24 h. Suspected Candida colonies were further identified using Gram stain and differentiated using germ tube test and the CHROM agar Candida (CHROM agar, France) medium. The light green colonies were identified as C. albicans, dark green colonies as atypical C. albicans, flat pink dry colonies as Candida Krusei, and whitish colonies as other Candida species.

Salivary samples were collected from each of the 200 subjects (100 DM patients, 50 non-diabetic endocrine patients, and 50 apparently healthy individuals) subsequent to the collection of the oral rinse. The salivary flow rate was estimated after pooling saliva in their mouth for 5 min, and the FSG levels in each individual were also measured. FSG levels were determined on the saliva samples (glucose oxidase method, GOD) by a semi-automated analyzer (Chrome Tech B01-CT-8 series double-beam UV/Vis spectrophotometer). The supernatant from centrifuged (centrifugation at 4500 rpm for 5 min) saliva samples were used for the estimation of glucose levels. A sample of 1.5 mL of fasting venous blood was collected from all subjects (100 DM patients, 50 non-diabetic endocrine patients, and 50 apparently healthy individuals) to analyze Fasting Blood Sugar levels (Konelab 30, GOD method). Each individual HbA1c levels were measured using 1.5 mL of whole blood which was collected into ethylenediamine tetraacetic acid (EDTA) tubes (high-performance liquid chromatography, Bio-Rad D10).

Data analysis

Sociodemographic data, family history, and medical and clinical history of the patients were collected using an interviewer-administrated questionnaire. Data were analyzed using the statistical software; Statistical Package for the Social Sciences version 23.0. Continuous variables were analyzed and the analyzed data were classified for further analysis. Sociodemographic data (age, gender, marital status, height, weight, body mass index (BMI), food pattern, patient history, denture wearing, smoking, and alcohol use) and results of laboratory investigations were descriptively analyzed. The correlations between variables were analyzed using bi-variant correlations. The mean values of individual data were compared using independent t-tests. The associations among the groups mentioned above were then obtained using the Chi-square test.

Descriptive statistics

One hundred DM patients, 50 patients from the endocrinology clinic without diabetes, and 50 individuals who appeared to be healthy were analyzed to determine the association between DM and the development of OC in known diabetic mellitus patients. The sociodemographic data and laboratory investigations of the participants were analyzed, as shown in Table 1. All three study groups represented a female predominance and a mean age of ≥45 years.

DM patients were categorized into age groups and a majority (45%) of patients were in the age range 41–60 years. Further, the BMI category of DM patients was analyzed according to the Asian BMI categorization, and it was identified that 62% of the patients were obese, and only 21% had a normal BMI. According to the medical history of the DM patients, 96% of the individuals were suffering from Type 2 DM, and 4% had Type 1 DM. Among the 23 DM patients wearing dentures, 75% have been wearing them for ≤10 years, and 25% have been wearing them for 11–20 years.

The Candida growth of the DM patients was compared between study groups. Fungal growth was given for oral rinse samples grown on sabouraud dextrose agar. According to the Candida growth, 66 (66%) individuals out of 100 DM patients showed positive growth. Among the individuals who had positive Candida isolates, certain individuals had only one species, while other individuals had two species isolated [Table 2]. Importantly, the number of colonies isolated in the oral cavity showed a significant variation among the three study groups [Table 3]. Diabetic patients who gave positive growth in the study group were altogether 66. Greater than 10000 CFU/mL colony count were identified in 41 patients (62.12%), which is significantly higher comparatively to other study groups.

The following comparisons were done using the independent t-test [Table 4]. Candida growth was significantly associated with the food pattern of the participants (P = 0.026). Non-vegetarians had a comparatively higher Candida growth than vegetarians. Wearing of dentures (P = 0.001) and duration of wearing dentures (P = 0.006) also had a significant association with the growth of Candida in the oral cavity. Participants wearing dentures and, more significantly who have been wearing them for more than 10 years had a comparatively higher probability of having Candida growth. Moreover, FSG level had a significant association with Candida growth (P = 0.026), where individuals with an FSG level ≥4.1 mg/dL were identified to be having a greater possibility for Candida growth.

A significant correlation was identified between the Candida growth and the FSG level of participants (P = 0.026, r = 0.239). The log value of the total colony count (per mL) had a significant correlation with the FSG level of the individuals (P = 0.029, r = 0.234). Furthermore, FSG level presented a significant correlation with the fasting blood sugar level of participants (P = 0.005, r = 0.297). Each variable was statistically analyzed against Candida growth, and significant associations were entered into a regression to obtain a model. Accordingly, a model with a higher significance of 0.001 was obtained. Food pattern, wearing of dentures, and FSG were identified to be significant contributing factors affecting the oral growth of Candida.