Candida Carriage Rate and Growth Characteristics of Saliva in Diabetes Mellitus Patients: A Case-Control Study

Background and aims. The aim of this study was to establish a relationship between salivary glucose levels and Candida carriage rate in type 2 diabetes mellitus patients and assess the growth characteristics and acid production of Candida in glucose-supplemented saliva. Materials and methods. A total of 90 subjects, 30 with controlled type 2 diabetes, 30 with uncontrolled type 2 diabetes and 30 without diabetes (control subjects), aged 30‒60 years, participated in the study. Unstimulated saliva was collected and investigated for glucose levels (GOD-POD method), colony-forming units (CFU) of Candida and salivary pH, using Indikrom paper strips). Analysis of statistical significance of salivary glucose and PH levels was carried out using post hoc Tukey HSD test. Correlation of Candida carriage rate with salivary glucose and salivary PH in the study groups and control group was made using Pearson’s correlation. Results.Candida CFUs were significantly higher in diabetic subjects, with a significant and positive correlation with salivary glucose levels. There was a negative correlation between salivary PH levels and Candida carriage rate. Conclusion. Increased salivary glucose was associated with increased prevalence of oral Candida in diabetic subjects. The growth of Candida in saliva was accompanied by a rapid decline in PH, which in turn favored their growth.


Introduction
iabetes mellitus is a complex multisystem disorder representing one of the major chronic health problems the world is facing today. The prevalence among adults aged 20-70 years is expected to rise from 285 million in 2010 to 438 million by the year 2030. 1 Oral candidiasis and other opportunistic fungal infections are some of the early, non-specific signs of uncontrolled diabetes. 2 The carriage of Candida species and the density of Candida growth in the oral cavity is frequently claimed to increase in patients with diabetes mellitus contributing to acquiring oral candidiasis in such patients. 3,4 Candida is known to be a normal commensal of the oral cavity. But during hyperglycemic episodes, the environmental alteration in the oral cavity such as immune dysfunction, increased salivary glucose and acid production favor the transition of the harmless commensal to a pathogen. Hence, estimating the critical Candida carriage rate in saliva at which it transforms to produce pathogenesis, especially when it is omnipresent during commensal as well as pathogenic states, can be a valuable aid in identifying patients with increased risk for the disease before development of the occult disease. Institution of prevention strategies at this stage itself can reverse the impending disease as well as its multiple sequelae. Saliva offers an inexpensive and noninvasive screening method as compared to serum in terms of collection, storage and voluminous sampling. Therefore, the present study was undertaken to establish a relationship between salivary glucose levels and Candida carriage rate in type 2 diabetes using saliva as a screening medium. Considering the paucity of available information, this study might lay a foundation for the application of saliva as a tool for predicting candidiasis in the population susceptible to diabetes mellitus.

Study Setting, Design and Subjects
This randomized case control study was conducted on the south Indian population during 2011-2012. The study was approved by the ethics committees of the University. Participants provided written informed consent prior to data collection. The study sample consisted of 90 subjects in the age group of 30 to 60 years and included 31 females and 59 males. They were divided into 3 equal groups of 30 patients-Group I included control patients with Random Non Fasting Plasma Glucose (RNFPG) levels less than 120 mg/dL; Group II included patients with controlled diabetes mellitus (RNFPG levels in the range of 120 mg/dL to 200 mg/dL); Group III consisted of patients with uncontrolled diabetes mellitus (RNFPG levels more than 200 mg/dL). As HbA1c reflects the average blood glucose concentration over an extended period of time and remains unaffected by short-term fluctuations in blood sugar levels, 4 RNFPG level was employed to achieve direct and simultaneous correlations with saliva glucose concentrations. Patients with a history of any systemic or oral mucosal disease, patients who were on medications other than anti-diabetic drugs and patients with habits like smoking, tobacco or betel nut chewing and alcohol consumption were excluded from the study. To avoid bias in the results, confounding factors which might alter the parameters of saliva like oral mucosal or dental disease and subjects with inadequate oral hygiene were excluded from the study.

Sample Collection and Laboratory Techniques
Unstimulated saliva was collected after 12 to 16 hours of fasting in the morning using a "spit technique" which represented whole mouth fluid contributed by secretions from major and minor salivary glands and potentially gingival crevicular fluid. Salivary glucose was estimated by GOD-POD method. Saliva sampling for estimation of colony-forming units (CFUs) of Candida was performed using the oral rinse technique. The rinse was immediately concentrated by centrifuging at 1,700 rpm for 10 minutes. The supernatant was discarded and 0.1 μL inoculating loop was used to spread the sample onto Sabouraud's dextrose agar plates supplemented with chloramphenicol (10 mg/mL). Cream-colored opaque colonies with yeasty odor were identified. CFUs were counted manually, and the number was multiplied by 1,000 and expressed as CFU/mL. The salivary pH of the collected saliva was estimated using Indikrom paper strips and reported accordingly.

Statistical Analysis
Data was analyzed with SPSS 18. Data obtained was analyzed using one-way ANOVA and chi-squared test for comparison between the groups. Analysis of statistical significance of salivary glucose and pH levels was made using post hoc Tukey test. Correlation of Candida carriage rate with salivary glucose and salivary pH in the study groups and control group was made using Pearson's correlation.

Results
The mean age in group I was 48.20±7.131 years whereas the mean ages in groups II and III were 48.33±8.46247 and 47.833±7.022 years, respectively. Overall a male predominance was observed in all the 3 groups (Table 1). The mean salivary glucose level in group I was calculated at 1.18±0.675 gm/dL, which was lower than that in group II, i.e. 4.95±6.61 gm/dL. The mean salivary glucose level in group III was 13.35±6.61 gm/dL which was the highest. Statistically significant differences were obtained when salivary glucose levels were compared between group I, II and III ( Table 2). The statistical comparison of the salivary pH levels of group I and group II showed no statistically significant differences (P=0.990). Comparison of salivary pH levels between groups II and III showed a significant difference (P=0.002). Similarly, comparison of the salivary pH levels of groups I and III revealed significant differences (P=0.001) ( Table 3).
In group I, 46.7% of patients had no Candida growth while 40.0% had a Candida count of 10 CFU/mL. No patients were observed in the colony forming range of 10 3 to 10 4 CFU/mL. In group II 43.3% of patients had colony forming units of 10 2 CFU/mL whereas no patients with Candida growth of more than 10 3 CFU/mL were observed. The percentage distribution of Candida counts in group III showed that Candida growth ranged from 10 to 10 4 CFU/mL with Candida growth predominantly in 10 3 CFU/mL range. None of the patients in Group III were observed with negative Candida cultures.
Comparison between descriptive statistics of groups I, II and III showed that Candida count was highest in group III, followed by groups II and I (Table 4).
Excellent correlation was observed between salivary glucose levels and Candida carriage rates in the three groups, with statistically significant p-value of <0.001 and r-values of 0.844, 0.925, 0.901, respectively. The correlation was seen in 71.2336% of patients in group I, 85.5625% in group II and 81.1801% in group III (Figure 1).
A negative correlation was noted between salivary pH and Candida carriage rates in patients with uncontrolled diabetes with 39.5641% of cases showing the correlation with an r-value of -0.629 while the correlation observed in group II was poor with only 3.385% of cases showing correlation with an r-value of -0.184 (Figure 2).

Discussion
The present microbiological study comprehensively investigated oral yeast colonization; candidal CFUs were significantly higher in diabetic subjects (group III > group II) when compared with non-diabetic    4 However, the carriage rate of Candida in the oral cavity was different in various studies, which might be attributed to the different sampling methods (such as swabs, water and buffer etc) used by different workers. The oral rinse technique with phosphate-buffered saline was used for sampling in the present study since this is known to be a sensitive technique for estimating the oral candidal carriage. 9 Moreover, as C. albicans is not uniformly distributed in the oral cavities of healthy or diabetic patients, the oral rinse technique can be used where quantitative and overall carriage of yeasts are to be assessed (as in the present study) whereas swabs can be used for examination of specific sites for carriage. 10 In the present study there was a significant and positive correlation between salivary glucose and CFU of Candida, which was in accordance with previous studies where increased Candida reflects increased salivary glucose levels. 4,11 Normal glucose levels in the saliva do not significantly affect oral health or support the growth of microorganisms. However, high levels of salivary glucose increase candidal adherence to buccal epithelial cells. 12 Salivary glucose forms chemically reversible glycosylation products with proteins in tissues dur-

Figure 2. Correlation between salivary pH and candidal carriage in study groups and control group.
ing hyperglycemic episodes, and this leads to accumulation of glycosylation products on buccal epithelial cells, which in turn may increase the number of available receptors for Candida. 4 This enhances their colonization on teeth and oral mucous membranes. Glucose also serves as a nutrient for Candida microorganisms and suppresses the killing capacity of neutrophils, which further accentuates colonization and likely consequences can be proposed as a result of these elevated salivary glucose levels in diabetes. 13 Although the present study confirmed that Candida is more prevalent in the oral cavity of diabetics than non-diabetics, some studies have failed to show a correlation between candidal carriage rates and diabetic status. 14,15 Based on a hypothesis diabetes mellitus is not in itself usually responsible for a significant increase in candidal populations, but only when it interacts with a local factor, it can be a possible explanation for these findings. 16 Khaled Abu-Elteen et al 17 in their study also observed that the frequency of Candida isolation was significantly higher in smokers than in the non-smokers in both diabetics and controls.
Candida becomes more easily established in the oral cavities of diabetics than in healthy subjects, but in the dentate diabetic patients, the number of yeasts on the mucosa remains within the normal range as it is limited by masticatory movements, salivary flow and epithelial desquamation. 18 In the present study, patients with oral mucosal diseases and smoking habits were excluded but dentate and edentulous patients were included in the study indiscriminately, which could be a possible cause for an increase in candidal carriage in these subjects.
In the present study, 53.3% of non-diabetic patients were positive for candidal culture, suggesting candidal carriage even in healthy individuals. But a higher carriage rate was found in diabetic patients as compared to normal controls. The present study also confirmed that the patients with uncontrolled diabetes carried Candida more frequently and in higher numbers than those with moderate or good control diabetics. This observation was consistent with previous studies in which a correlation between diabetic controls at any one time and the extent of yeast colonization of oral mucosa sites commonly inhabited by yeasts was established. 14 The degree of colonization of the mouth by yeasts might be quickly and directly altered by blood glucose levels. However, in a study by LM Tapper Jones et al, 18 no differences in candidal status could be detected according to the degree of control of diabetes, which is contrary to our study. These contradictory findings may be due to differ-ences in sampling techniques or in dental status and smoking habits between patients.
The total number of yeasts present is important to the development of infection. In order to change from a commensal to a pathogenic mode of life, yeasts need to aggregate in large numbers to accumulate sufficient enzymes for penetration of mucous membranes and epidermis. 14 Colony forming unit (CFU) is usually recorded to obtain the clinical data, which is essential to establish a clinical diagnosis of oral candidiasis. In the present study, the majority of the patients with controlled diabetes had candidal carriage of 0 to 1000 CFU/mL of saliva while patients with uncontrolled diabetes demonstrated a carriage ranging from 10 to 10,000 CFU/mL of saliva. Negative cultures were obtained in 46.7% of control patients. These observations were consistent with a study by Renner et al, 16 in which C. albicans never exceeded 100 organisms per square centimeter in the control population whereas the concentration always exceeded 10,000 organisms per square centimeter in patients with candidiasis.
However, Epstein et al 8 demonstrated that carriers and patients with oral candidiasis can be reliably distinguished on the basis of quantitative cultures. Patients with candidiasis had >400 colony-forming units per mL of saliva, whereas carriers of C. albicans had <400 colony-forming units per mL. The possible explanation for higher values of CFU/mL in the present study could be differences in colony counting methods used in the two studies. In the present study, CFUs were counted manually, and the number was multiplied by 1,000 and expressed as CFU/mL whereas automated analysis methods were used in the study by Epstein et al. 8 Thus this observation in the present study requires more comprehensive evaluation with emphasis on laying down cut-off limits for CFU between healthy and diabetics, which might serve as a useful clinical indicator and aid in the diagnosis of oral candidiasis.
A statistically significant difference was seen in salivary pH levels between normal patients and uncontrolled DM patients but statistical comparison of the normal group and group with controlled DM showed no significant differences. The fall in pH in diabetic patients in the present study can be attributed to the decrease in salivary flow rate in diabetic patients and acid production of oral isolates of Candida species in the presence of glucose.
In the present study there was a negative correlation between salivary pH levels and Candida carriage, consistent with studies carried out by Sama-ranayake et al, 5 where the growth of C. albicans in human saliva supplemented with glucose was accompanied by acid production. Their study provided the first, detailed qualitative and quantitative data on the short-chain carboxylic acids produced by Candida species cultured in human whole saliva. Pyruvates and acetates are, by far, the predominant ionic species that bring about a rapid decrease in pH with both species of Candida. The acidic pH conditions prevailing intraorally could, in turn, activate the extracellular phospholipases and acid proteases of Candida, 5 further promoting yeast adhesion to oral mucosal surfaces. 12 MacFarlane et al 19 also supported the possible mechanism whereby low pH conditions may promote oral colonization due to the superior ability of yeasts to adhere to epithelia and denture acrylic surfaces at a low pH of approximately 2 to 4.14. Another mechanism may be the aciduric and acidophilic nature of Candida species, which allows them to thrive in a low-pH milieu.
Therefore, the present findings taken together would strongly indicate that production of acids in Candida species could play a contributory role in the pathogenesis of oral candidiasis.

Conclusion
In conclusion, the present findings show that the increase in salivary glucose levels in diabetic subjects is likely to contribute to their increased candidal carriage and the potential for increased susceptibility to oral candidiasis. Thus saliva can be used as a reliable noninvasive tool to monitor candidal carriage in diabetic subjects.