Comparison of Exfoliative Pap Stain and AgNOR Counts of the Tongue in Smokers and Nonsmokers
- First Online:
- Cite this article as:
- Fontes, P.C., Corrêa, G.H.M., Issa, J.S. et al. Head and Neck Pathol (2008) 2: 157. doi:10.1007/s12105-008-0059-0
Objective To compare exfoliative cytology from the oral mucosa of smokers and nonsmokers, with emphasis on proliferative activity. Methods Exfoliative cytology specimens were obtained from clinical normal mucosa from the lateral border of the tongue in 30 nonsmokers and 30 smokers ranging in age from 40 to 70 years of age, who were seen at the Heart Institute’s Patient Center and the Smoking Cessation Program of the University Hospital, University of São Paulo Medical School (InCor-HCFMUSP). The cytologic specimens were evaluated by Papanicolaou staining and AgNOR quantification in order to evaluate the presence of cytological alterations suggestive of inflammation, dysplasia, keratinization, and proliferative activity of epithelial cells. Results Only Papanicolaou Class I and Class II smears were observed. Inflammatory alterations were found in 90% of smokers and in 87% of nonsmokers. The number of AgNORs/nucleus differed significantly between smokers and nonsmokers (3.372 ± 0.375 versus 2.732 ± 0.236). Conclusions Within the limitations of this research, the results indicate higher proliferative activity in smoking patients compared to nonsmoking patients, even in the absence of clinical lesions.
KeywordsCytodiagnosisSmokingNucleolar organizer regionDisease preventionOral mucosa
Smoking is currently the most preventable cause of diseases and death worldwide and is one of the main risk factors for the development of cancer in different organs. Therefore, smoking patients should be carefully monitored in view of the series of alterations that tobacco can cause [1–6].
According to the World Health Organization, cancers of the mouth and oropharynx are the most frequent head and neck tumors, with about 390,000 new cases per year worldwide, corresponding to 4% of all cancer cases . Brazil has one of the highest incidences of oral and oropharyngeal cancer worldwide. According to the Brazilian National Institute of Cancer (INCA) , tumors of the oral cavity is the eighth most common cancer in Brazil and estimates for 2006 predict 13,470 new cases of oral cancer, including 10,060 among men and 3,410 among women.
Oral cancer mainly affects the floor of the mouth, lateral border of the tongue and soft palate, although other areas of the mouth may also be involved, and smoking alone or in combination with frequent alcohol consumption represents the main risk factor [3, 4]. According to Winn, smoking is the leading cause of oral cancer in 91% of men and in 59% of women . Many cases of oral cancer are diagnosed at advanced stage, a fact that results in an unfavorable prognosis and high mortality, in addition to high treatment costs coupled with an increased number of complications. Thus, early diagnosis is of extreme importance in managing this disease.
Considering the possible influence of smoking on the occurrence of oral cancer and of precursor lesions, together with the accessible location of the lesions which permits easy visualization and collection of material for analysis, exfoliative cytology might be a useful tool for the detection and monitoring of initial alterations [9, 10] and for the establishment of adequate treatment in smokers [3, 11–16]. It is a complementary diagnostic method which presents several advantages such as rapid and easy execution, low cost, diagnostic safety, efficacy and noninvasiveness, and can be repeated several times [9, 12, 15–17].
Papanicolaou staining is used as a routine method for the analysis of cytological aspects and permits the identification of basic inflammatory, dysplastic or malignant alterations . Histochemical AgNOR quantification consists of the nucleolar organizer regions (NORs) staining with silver (Ag) salts. NORs are proteins that are associated with the fibrillar centers and dense fibrils of the cell nucleus during interphase and are responsible for the replication of RNA. Thus, the larger the number of NORs, the higher the replication rate of ribosomes and cells. This technique has therefore been used for the quantification of cell proliferation in different tissues and lesions [11, 13, 17].
Silver staining of exfoliative cytology smears was first used by Howell et al. in 1975 . Subsequently, several investigators have applied this method in cytology for the study of different oral lesions. Studies evaluating cell replication in smoking patients by AgNOR quantification in exfoliative cytology specimens have demonstrated an increased cell proliferation when compared to nonsmokers [11, 17, 19].
In the present study, we examined exfoliative cytology from the lateral border of the tongue, which is a site associated with a high incidence of oral cancer, in both smokers and nonsmokers. The objective of this study was to determine the influence of smoking habit (time and number of cigarettes consumed per day) on the proliferative activity, and whether a correlation exists between the results obtained by Papanicolaou staining and by histochemical AgNOR quantification.
Sixty men ranging in age from 40 to 70 years, volunteers of the Outpatient Clinic of the Heart Institute, University Hospital, University of São Paulo Medical School (InCor-HCFMUSP), participated in the study after signing an informed consent form. The following inclusion criteria were applied: not addicted to alcohol, absence of any oral lesion, and with no prior or present history of benign or malignant oral neoplasms . Two groups were analyzed: smokers and non-smokers. Smokers were defined as those individuals that smoked over 20 cigarettes or more per day for at least 30 years . In this research, only smokers of filtered cigarettes were included. Pipe smokers or consumers of tobacco in other ways were not included, because of concentration variation, which may affect the oral mucosal cells with different intensity, and other systemic effects . Non-smokers were defined as people who have never smoked.
Material for exfoliative cytology was collected with a metal spatula from the lateral border of the patient’s tongue, and smears were prepared on properly identified glass smears and fixed in 99.3% ethanol.
The samples were then submitted to AgNOR and Papanicolaou staining. AgNORs were stained according to the method recommended by the International Committee on AgNOR Quantitation, revised by Trerè, with pre-staining fixation for 30 min in Carnoy’s solution (acetic acid: absolute ethanol, 1:3) . NORs were directly counted under a light microscope according to the parameters established by Crocker et al., i.e., well-defined black dots in the nucleus were counted, with aggregations (overlapping or fused black dots) being considered a single structure .
The results were reported as mean ± standard deviation. The results were submitted to statistical analysis, with the level of significance set at 5%, using the MINITAB for Windows program (version 2000, 3.1; Minitab Inc., State College, PA). For the variables of number of cigarettes smoked per day and duration of smoking, significant differences between groups were determined by the Student t-test.
The number of smears adequate for analysis of AgNORs/nucleus was 25 for smokers and nonsmokers each. This smaller number was due to technical artifacts such as an insufficient number of cells, the presence of a great amount of residual staining and overlapping structures.
The mean age of the population was approximately 51 (±6) years for smokers and 58 (±10) years for nonsmokers. The mean duration of smoking was approximately 36 (±7 years) and the mean number of cigarettes smoked per day, as reported by the patients, was approximately 26 (±7).
Comparison of the mean number of AgNORs showed a significant difference between nonsmokers (2.732 ± 0.236 AgNORs/nucleus) and smokers (3.372 ± 0.375 AgNORs/nucleus) (t = 7.22, d.f. = 48, P = 0.0001; 95% CI: 0.46–0.82 AgNORs/nucleus).
Mean (±standard deviation) of AgNORs per nucleus in the group of 25 smokers divided according to the number of cigarettes consumed per day
Patients (n = 25)
Mean ± SD
3.38 ± 0.385
3.31 ± 0.364
3.50 ± 0.459
Mean (±standard deviation) of AgNORs per nucleus in the group smokers divided according to the duration of smoking (years)
Duration of smoking (years)
Patients (n = 25)
3.03 ± 0.679
3.42 ± 0.368
3.46 ± 0.311
3.06 ± 0.218
Number of AgNORs per nucleus in smears from smoking and nonsmoking patients divided according to the cytologic evaluation
3.302 ± 0.234
3.389 ± 0.405
2.651 ± 0.414
2.843 ± 0.312
In the present study, a site showing one of the highest incidences of oral cancer, the lateral border of the tongue, was analyzed by both AgNOR and Papanicolaou staining . Histochemical AgNOR quantification demonstrated a significantly higher proliferative activity in smokers compared to nonsmokers.
Numerous studies have used different markers of cell proliferation as an auxiliary tool in the diagnosis of oral cancer. These cell proliferation markers are important for advancing the understanding of cellular alterations. However, studies analyzing cell proliferation markers by exfoliative cytology are scarce [11, 17, 19, 23–25].
Prior to staining of the smears by the histochemical AgNOR technique, pilot smears were prepared to improve the technique and to prevent the loss of smears. The first results were unsatisfactory since the smears were poorly stained and analysis was impossible. The technique described by Ploton et al. was used during this phase . Many smears had a great amount of residual staining. Tests without celloidin were performed but the results were also unsatisfactory. The technique proposed by the International Committee on AgNOR Quantitation and revised by Trerè yielded the best results and was adopted in the present study . The difficulties of the technique are mainly related to ambient temperature, speed of the silver nitrate reaction in solution, and the sensitivity of silver nitrate to light.
In the present study, the initial sample of 30 smears in each group was reduced to 25 after staining due to technical difficulty. This sample size is similar to those reported in the studies of Sethi and Shah, Gedoz et al., Sampaio et al. and Orellana-Bustos et al. [17, 19, 23, 24]. The smears could not be analyzed with the Image Tools 3.0 software and AgNORs were therefore counted manually. The condition of the sample also did not permit cytomorphometric analysis due to the same problems that led to the exclusion of part of the specimens. Various attempts were made to use software for AgNOR quantification without success.
In the present study, AgNORs were visualized as well-defined black dots inside the nucleoli [11, 12, 24, 25]. AgNOR quantification revealed a higher proliferative activity in smoking patients. These results demonstrate the importance of exfoliative cytology and of this type of analysis. The technical precision of staining of the smears by the AgNOR technique should be improved to permit its routine application.
A significant difference in the mean number of AgNORs per nucleus was observed between smokers and nonsmokers (3.372 ± 0.375 versus 2.732 ± 0.236). These values are close to those reported by Sethi and Shah and Orellana-Bustos et al., but differ from those obtained by Cançado et al. who observed a mean number of 1.94 ± 0.13 in smoking patients [11, 17, 24]. This difference might be due to the size and characteristics of the sample since the latter authors defined smokers as subjects who had consumed 10–20 cigarettes per day over a period of 10–50 years, in contrast to the inclusion criterion used in the present study. The authors also did not report the mean duration of smoking or mean number of cigarettes consumed per day.
Analysis of the mean number of AgNORs per nucleus in subgroups of patients according to the number of cigarettes consumed per day showed no significant difference, in agreement with Cançado et al. . The mean number of AgNORs per nucleus was highest in group C (3.50) which consumed the largest number of cigarettes (40 cigarettes per day). A larger number of AgNORs per nucleus was observed in group A compared to group B (Table 1). Although there was a trend for increase in AgNORs with increasing cigarette consumption, our data did not reach statistical significance.
Comparison of the mean number of AgNORs per nucleus according to the duration of smoking revealed no significant difference, a result also observed by Cançado et al. .
The mean percentage of cells with more than three AgNORs per nucleus was significantly higher in smokers (43.4 ± 15.2%) when compared to nonsmokers (16.6 ± 8.0%), again demonstrating a higher proliferative activity in the group of smoking patients .
When architectural disturbance is accompanied by cytological atypia, the term dysplasia applies . According to the WHO suggestion of 2005, the traditional schema of grading dysplasia as mild dysplasia, moderate dysplasia, severe dysplasia and carcinoma in situ continues to be used. There are three classification schemes (oral epithelial dysplasia scoring system, squamous intraepithelial neoplasia and Ljubljana classification) recommended by the 2005 WHO Classification for epithelial dysplasia grading for routine use [27, 28]. It was not possible to use the criteria for grading epithelial dysplasia as it is firstly based on architectural features, and then on cytology . It is important to consider that our patient inclusion criteria included non-alcohol use and absence of any clinically evident oral lesion, with no previous or present history of benign or malignant neoplasms.
Pavanello et al. reported a larger number of inflammatory alterations and an increased rate of cell maturation in smokers . In this study, inflammatory cells, as well as characteristics of inflammation such as cell lysis, presence of a perinuclear halo and vacuolization, were observed in smears of both groups. These alterations, even when there are no clinical manifestations, may indicate epithelial modifications in response to a physiochemical environment caused by the substances present in cigarettes.
Although the Papanicolaou classification alone was not significant, it gained meaning when correlated with the results of AgNOR quantification. Comparison of the mean number of AgNORs per nucleus in inflammatory smears showed a significant difference between smokers and nonsmokers.
The effect of smoking was demonstrated by a significantly larger number of AgNORs per nucleus in mucosal cells of the lateral border of the tongue, a site characterized by a high incidence of oral cancer, indicating a higher proliferative activity of these cells . Sampaio et al., using the histochemical AgNOR method, suggested that smoking influences the proliferative activity of cells. Orellana-Bustos et al. reported that the number of AgNORs is not only related to cell proliferation, but also to the metabolic activity of cells [23, 24].
The present results suggest that cigarette smoking produces alterations in the mechanisms of cell growth control. Tobacco has been considered to be an initiating factor in the process of oral carcinogenesis, which is frequently associated with alcohol as a promoting factor . In addition, the results suggest that the oral mucosa is susceptible to the effects caused by cigarette smoking, responding with an increase in cell proliferation [11, 17, 19, 23–25].
The present study demonstrated a larger number of AgNOR proteins in exfoliative cytology specimens obtained from the mucosa of the lateral tongue border of smokers who presented with no clinically visible alterations when compared to nonsmokers.
Within the limitations of this research, the results indicate higher proliferative activity in smoking patients compared to nonsmoking patients, even in the absence of clinical lesions.
The authors would like to thank Prof. Ivan Balducci for the statistical analysis. Research supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP, grant 04/10520-2) and Fundação para o Desenvolvimento da UNESP (FUNDUNESP, grant 00279/04-DFP). The publication of this article was supported by the Program for the Internationalization of Research of UNESP.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.