A cytomorphometric analysis of the oral mucosa in patients with type 2 diabetes mellitus

Materials and Methods: After rinsing the mouth with normal saline, slides were prepared from buccal mucosa and dorsum of tongue and ixed in 95% ethyl alcohol. The slides were stained with Papanicolaou stain and Acridine orange. Fifty clearly deined cells in each slide were visualized under light microscope for cytomorphometric analysis of cells using Image J software and under luorescence microscope for assessment of nuclear alterations like micronuclei, nuclear budding, binucleation, multinucleation and karyorrhexis.

repair processes and causes dysfunction of oral mucosa. 4everal studies suggest a higher prevalence and severity of some pathologies in the oral tissues of patients with diabetes mellitus like gingivitis, periodontitis, dental caries, candidiasis, and other oral manifestations such as alteration of salivary low and oral burning sensation. 4veral studies have examined the deleterious effects of diabetes on the oral mucosa.It was reported by Caldeira EJ et al that diabetes adversely affects the histomorphology of cheek mucosa, which may compromise tissue function to favour the occurrence of oral infections and probably neoplasia. 5though many of the pathological processes affecting the oral mucosa are clinically distinguishable, most lesions require a deinitive diagnosis before commencement of appropriate therapy.The most accepted clinical technique for the diagnosis of lesions in the oral mucosa is incisional or excisional biopsy. 6However, in speciic clinical conditions, such as diabetes mellitus, a great many invasive techniques are relatively contraindicated and necessitate the use of oral exfoliative cytology which may be suitable. 4 exfoliative cytology is considered a moderate and non-invasive technique when compared to conventional anatomopathological examination, it is necessary to make an assessment of the application of oral exfoliative cytological methods in patients with DM patients. 4This will help to better interpret the indings from specimens obtained from the mucosa of such affected persons.The best knowledge of this technique, particularly regarding the morphological and morphometric aspects of the cells may yet enable the implementation of exfoliative cytology in public health programs, comprising one more tool in the screening, diagnosis and follow-up of diabetes mellitus. 4e present study thus attempts to quantitatively and qualitatively assess changes in oral epithelial cells by exfoliative cytology to determine the effects of DM on the oral cavity.

MATERIALS AND METHODS
The present study was conducted in the Department of Oral Pathology and Microbiology, Manipal College of Dental Sciences, Mangalore and Kist Medical College, Imadol, Lalitpur over a period of 24 months from January 2010 to January 2012.The study involved a cytomorphometric analysis of the oral mucosa in patients with type 2 diabetes mellitus using smears obtained from buccal mucosa and dorsum of tongue.

Study Sample:
The study comprised of a total of 60 subjects divided into two groups: Study group: Consisting of thirty individuals above 40 years of age with a known history of type II diabetes mellitus for a minimum period of one year.The subjects with diabetes were selected from among the patients attending the dedicated Diabetic clinic of K. M. C. Hospital, Attavar Mangalore.Patients were included irrespective of whether they were under insulin or oral hypoglycaemic therapy for diabetes.
Control group: Consisted of thirty age and sex-matched healthy individuals without any history of diabetes mellitus (assessed by random blood sugar) 6

Exclusion criteria
Patients/ individuals with the following were excluded from the study:

Methods of collection of Data:
Ethical clearance was obtained from the Institutional Ethics Committee (MCODS, Mangalore) and approval was taken from Department of Internal Medicine, KIST medical college.Subjects of both the study and control groups were informed of the procedure and a written consent was obtained.All the subjects were clinically examined to assess the oral hygiene and to exclude the possibility of any other oral disease or systemic disease with oral manifestations.
For study groups, haematological assessment for blood glucose/glycosylated haemoglobin levels was obtained from patients' medical records.

Cytomorphometric Assessment:
Sample collection procedure: After detailed clinical examination, the subjects were requested to rinse the mouth with normal saline.Smears were obtained from the buccal mucosa (BM) and the dorsum of tongue of each subject using a wooden spatula moistened in distilled water.Two smears from each site were obtained.The smears were transferred onto grease-free glass slides and ixed with 95% ethyl alcohol.
Two smears each from both the sites (buccal mucosa and tongue) were stained with Papanicolaou stain to visualize under compound light microscope for cytomorphometric analysis of cells (for NuA, CA, CyA and N/C ratio) using Image J software.
The other two smears from each site were stained with Acridine orange to visualize under luorescence microscope for assessment of nuclear alterations like micronuclei, nuclear budding, binucleation, multinucleation and karyorrhexis.Cytomorphometric assessment was done in 50 cells of each PAP-stained smear.Nuclear alterations were also evaluated in 50 cells of each smear stained with Acridine orange and evaluated using the criteria set by Countryman and Heddle et al. 9

Statistical Analysis:
The data was analysed using

RESULTS
Table 1 enumerates the various parameters showed that the total nuclear area was higher in cases (53.96%) than in controls (44.13%) in both recorded sites, i.e., tongue and buccal mucosa.The same trend of higher values was also seen in the total cellular area in cases (1965.90)compared to controls (1662.52).Likewise, total cytoplasmic area in cases (1922.52)was greater than in controls (1616.07)and total N/C ratio in cases (0.0332) was also higher than in controls (0.0293) both in tongue as well as buccal mucosa..031 Glucose level = 28.359+ 0.075 (total cytoplasmic area) Lamichhane RS et al.
36.22 and 13.08 ± 1.56 while the mean RBS of controls was 99.06 ± 13.54.Similarly, the mean hemoglobin of cases was 13.08 ± 1.56 and that of controls was 13.09 ± 1.74.
Correlation of glucose level with various parameters using linear regression (Table 4) showed slight to mild correlation.The total nuclear area showed slight correlation (R=0.217), the total cellular area (R=0.336) and total cytoplasmic area (R=0.345)showed mild correlation while the total N/C ratio (R=0.083)showed no correlation with glucose levels.
Although, the glucose level values when compared with various parameters (Table 4) showed statistically signiicant correlation, only 11.3% of the total cellular area (p=0.024) and 11.9% of the total cytoplasmic area (p=0.020)showed variation with alteration in glucose levels.By looking at R value and R2 value, the Cellular area explains only 11.3% of variation in glucose level (R= 0.336, p=0.024).Also by looking at R value and R2 value, the Cytoplasmic area explains only 11.9% of variation in glucose level (R= 0.345, p=0.02).
Forward stepwise linear regression was carried out for total CA and total CyA to conirm the cytomorphometric parameter (deemed signiicant by Pearson's correlation) correlating with glucose level changes.The regression equation derived for CA from the coeficient was Glucose level = 33.910+ 0.070(total cellular area).The regression equation indicated that for every one unit increase in cellular area, the glucose level increased by 0.07 units.(Table 6) The regression equation derived for CyA from the coeficient was Glucose level = 28.359+ 0.075(total cytoplasmic area).The regression equation indicated that for every one unit increase in cytoplasmic area, the glucose level increased by 0.075 units.(Table7) Comparison of hemoglobin level with various parameters using Pearson's correlation showed only slight to mild correlation.The total nuclear area (R=0.149),total cellular area (R=0.192) and total cytoplasmic area (R=0.149)showed slight correlation while the total N/C ratio (R=0.251)showed mild correlation.The test further showed that only 37% of the total cellular area, 22% of the total cytoplasmic area, 2.2% of the total nuclear area, 6.3% of the total N/C ratio showed variation correlating with hemoglobin level and these were not statistically signiicant.(Table 5) Also the epithelial cells from the diabetic group showed features such as nuclear budding, micronuclei, binucleation, karyorrhexis and perinuclear halo as seen in igure (1-6).

Diabetes Mellitus and Cytomorphometry:
Diabetes Mellitus (DM) is a syndrome characterized by abnormal carbohydrate, fat and protein metabolism that results in acute or chronic complications due to absolute or relative lack of insulin. 9Type 2 diabetes mellitus is the ifth most common chronic condition and the sixth leading cause of mortality among the elderly worldwide. 10ntists come across various oral manifestations of this disease.Proper understanding of DM enables early diagnosis that is helpful in control of blood sugar level at an early stage to prevent various complications.Several studies have examined the deleterious effects of DM on oral mucosa with reports stating its adverse effects on the morphology of oral mucosa, which in turn may compromise tissue function to favour the occurrence of oral infections and oral neoplasia.In diabetes, there is a loss of oxidation equilibrium whereby the activities of the antioxidant scavengers and enzymes are depressed by elevated glucose concentration, excessive formation of free radicals and protein glycation.These noxious processes can cause serious damage to the biological structures at a molecular level which can be appreciated by oral exfoliative cytology which has been improved immensely by incorporation of techniques like computer-aided morphometry, making it more reliable, objective and reproducible. 5,11,12e present study evaluated the morphometric and cytologic changes in the exfoliated cells of apparently normal buccal and tongue mucosa in type 2 diabetic patients.There was a signiicant increase in Nuclear area (NuA) obtained from both the sites i.e., buccal mucosa and tongue in the study group.(Graph 1)This inding concurs with the studies done by Alberti et al 4 , Jajarm et al 13 , Shareef et al 14 , Tozoglu U and Bilge OM 15 and Prasad H et al 16 where the mean NuA was signiicantly higher among the diabetic group.(Table8) The increase in NuA among the study group could be explained by delay in keratinisation of oral epithelium, effects of ageing, dehydration/atrophy and inlammatory Delay in the keratinisation is attributed to glycation changes.Sustained hyperglycaemia causes greater accumulation of advanced glycation end products by abnormal glycation of proteins, lipids and nucleic acids in the walls of large blood vessels as well as in the basement membrane of the microvasculature.The progressive narrowing of the vessel lumen leads to decreased perfusion of the affected tissue and consequently decreases cell turnover thereby explaining the delay in the keratinisation process of the epithelium.This delay in the process of epithelial differentiation leads to an increase in the number of mature cells which show a large nucleus as a primary characteristic. 4,13pe 2 DM is a disease of the elderly and generally occurs in patients over 40 years of age.Cellular ageing produces various morphologic alterations and genotoxic damages in cells in the form of pleomorphism, bilobed nuclei, micronucleus, nuclear budding, karyorrhexis, perinuclear halo which were also observed in the smears from diabetic patients in the present study. 16(igure 1-7)      Diabetic patients also suffer from dehydration due to the decreased salivary low rates that may lead to mucosal atrophy.So, when smears from atrophic oral mucosa are made, the larger basal/spinous cells are inadvertently included in the sample.Thus, the primary pattern encompasses non-keratinized cells of parabasal layers which are smaller in cell size but have relatively larger nuclei; thus giving an impression of nuclear enlargement, with or without pleomorphism. 4,16lammation has been attributed to nuclear changes observed in various lesions.Reactive nuclear changes due to inlammation may erroneously mimic pleomorphism / atypia or may even suggest malignancy. 17Inlammation clinically manifests as supericial erosions or ulcerations of the oral mucosa seen as diffuse stomatitis and gingivitis.Cytomorphometric changes in the buccal mucosal cells of cigarette smokers similar to those noted in diabetics have been reported by Ogden et al. 18 In the present study, there was a signiicant increase in Cellular area (CA) of epithelial cells obtained from both the sites, i.e., buccal mucosa and tongue among the study group.(Graph2) Cytoplasmic area (CyA) from tongue also showed an increase in area which was statistically signiicant (p=0.00268).(Graph 3) This is similar to the indings of Prasad H et al16 who found an increase in cellular diameter (CD) among poorly controlled diabetics (PCD) and a similar study by Jajarm et al 13 who also found a signiicant increase in CyA in the diabetic group.However, Shareef et al 14  This increase in CA and CyA could be because of the altered cell membrane integrity leading to excessive accumulation of lipid droplets in the cytoplasm thus creating an increased intercellular space as reported by Caldeira EJ et al 5 in an experimental animal study where they examined the alterations in the histology and ultrastructure of oral epithelium of diabetic mice.In another experimental study in mice with diabetic retinopathy, Pannicke et al20 indicated that altered properties of K+ channels, arachidonic acid metabolites and acute oxidative stress could be the cause for the development of cell swelling (cytotoxic edema) in the glial cells of retina under ischemic and inlammatory conditions.A similar pathogenesis could be projected for increase in CA and CyA of oral exfoliated mucosal cells observed in our study.
In the present study, an increase in Nucleus: cytoplasm (N/C) ratio was evident in patients with type 2 DM, (Graph 4) a inding similar to Alberti et al, 4 Jajarm et al, 13 Shareef et al, 14 and Prasad H et al 16 though it was not statistically signiicant.(Table 8) The relatively greater increase in nuclear size compared to cytoplasmic volume may explain this phenomenon.
Thus, cytomorphometry may be an eficient tool to understand the extent of cellular changes that occur in oral epithelial cells in diabetics that are secondary to microscopic changes like vascular occlusion and molecular changes such as oxidative stress.Therefore, cytomorphometric analysis of oral mucosal cells could be implicated as a non-invasive technique for screening and monitoring of the disease status in diabetic patients.

CONCLUSION
In recent years important advances in various invasive techniques have been made for screening and diagnosing diabetes mellitus along with new strategies for its effective treatment.However, in diabetes, use of an invasive technique is a relative contraindication due to changes in blood glucose concentrations and the disease itself.Exfoliative cytology used to evaluate cytomorphometric changes in the oral mucosa in diabetics along with oral candidal colonization may be a promising tool to gauge the molecular/serological alterations in diabetes mellitus.
Two Papanicolaou stained smears each from buccal mucosa and tongue were visualized under compound light microscope for measurement of Nuclear area (NuA), Cellular area (CA) using Image J software and further calculation of Cytoplasmic area (CyA) and Nucleus: Cytoplasm ratio (N/C ratio).The smears were also assessed using Acridine orange under luorescence microscope for genotoxic nuclear alterations like micronuclei, nuclear budding, binucleation, multinucleation and karyorrhexis.The oral rinse was used for assessment of candidal colonization.
In the light of the present study, following conclusions have been derived from cytomorphometric analysis of exfoliated oral epithelial cells: 1. Signiicant increase in Nuclear Area among diabetics when compared to non-diabetics

Signiicant increase in Cellular Area and Cytoplasmic
Area among diabetics when compared to non-diabetics 3. Non-signiicant increase in Nucleus: Cytoplasm ratio among diabetics when compared to non-diabetics The increase in NuA among the study group could be explained by delay in keratinisation of oral epithelium, ageing effect, dehydration /atrophy and inlammatory process.The increase in CA and CyA could be attributed to the alteration of the properties of K+ channels i.e. decrease conductance of K+ ions in response to arachidonic acidinduced intracellular Na+ overload and acute oxidative stress leading to cell swelling.The increase in N/C ratio could be due to relatively greater increase in nuclear size compared to cytoplasmic volume.The objective demonstration of cytomorphometric and nuclear alterations by the oral exfoliated cells indicate the presence of cytological changes in the oral mucosa of diabetic patients despite the apparently normal clinical appearance.Hence, cytomorphometric analysis would aid the health professional as an additional non-invasive tool for the screening and monitoring of DM.While these alterations cannot be considered predictive or diagnostic for diabetes as they are not unique to this disease.

Graph 1 : 3 : 4 :
Indicating the distribution of Nuclear area (NuA) among cases and controlsGraph Indicating the distribution of Cytoplasmic area (CyA) among cases and controlsGraph 2: Indicating the distribution of Cellular area among cases and controlsGraph Indicating the distribution of Nucleus: cytoplasm ratio (N/C) among cases and controls

Figure 7 :
Figure 7: : Flow chart depicting mechanism for nuclear size alteration in DM; LV-Large vessel; BM-Basement membrane

Table 2 : Glucose and hemoglobin level for the study and control group
CA-Cellular area, CyA-Cytoplasmic area, N/C-Nucleus: cytoplasm, BM -Buccal mucosa,

Table 7 : Forward stepwise Linear Regression coeficient for total cytoplasmic area (Deemed signiicant by Pear- son's correlation) in relation to glucose level
in a similar study found a statistically signiicant decrease in CyA which could be due to the cell shrinkage caused by dehydration.(Table8)This theory is supported by the indings of Ogden et al 19 who reported a similar decrease in CyA in patients with alcoholism.