Background

Neurofibromatosis 1 (NF1, OMIM 162200) is one of the most common genetic diseases and has widely variable expressivity [1]. Oral manifestations are common, [2-6] but there are no studies that investigated functional alterations in salivary glands in NF1. Our aim was to evaluate the salivary flow rate in NF1 individuals, comparing to a control group, and to investigate the possible causes and some consequences of salivary gland alteration.

Methods

This case–control study was approved by the Ethics Committee (#31640/2012) and conducted at Oral Diagnosis Ambulatory and Pathological Anatomy Service of Antônio Pedro University Hospital of Universidade Federal Fluminense, Brazil.

The study group was composed of 49 NF1 individuals, according to clinical criteria [7]. To compare the sialometry values of NF1 individuals, a sex and age-matched control group (n=49) composed by non-NF1 individuals was included. Data about smoking, alcohol use, and intake of hyposalivation-inducing drugs was obtained from both groups.

Sialometry (unstimulated whole saliva flow rate) was performed between 8:00–11:30 AM and the participants did not ingest food or liquids 1 hour before the exam. After discarding all the saliva present in the mouth, saliva was collected for 5 minutes without stimulus. The salivary flow rate (mL/min) was calculated and classified: normal: ≥0.3; low: ≥0.1 and <0.3; very low: <0.1.

Tongue coating was evaluated according to previous studies [8,9], using digital photographs. The tongue dorsum area was divided into nine equal sections and each section was assigned a value: 0 = no; 1 = thin, 2 = thick coating. The sum of the results of each section was calculated, divided by 18 and multiplied by 100 to achieve the tongue coating index (TCI) percentage score.

Oral candidiasis was investigated by clinical and cytopathology exam through scrapping the tongue dorsum with a cytobrush. Glass slides stained with Papanicolaou and Periodic Acid-Schiff were analyzed according to specific criteria [10].

NF1 participants also answered to a questionnaire (Table 1).

Table 1 Questionnaire used in the study group (with NF1)
Full size table

Results

Additional files 1, 2 and 3 show details of the results. Table 2 summarizes the sample characteristics. There was no heterogeneity between the study and control group according to smoking, alcohol use and intake of medications that could cause hyposalivation. Four (8%) NF1 participants had plexiform neurofibromas close to or involving the major salivary glands areas and only one (25%) had hyposalivation.

Table 2 Summary of the sample characteristics
Full size table

In NF1 individuals, 59% (29/49) had hyposalivation; 32% (16/49) had severe hyposalivation. In the control group, 22% (11/49) of individuals had hyposalivation; 8% (4/49) presented severe hyposalivation. There was a statistically significant difference between the sialometry values of the study and control group (P < 0.0001; Wilcoxon rank-sum test). There was a statistically significant association between NF1 individuals and presence of hyposalivation (P = 0.001; McNemar’s χ2 test). The analysis of the adjusted residual showed that the prevalence of hyposalivation in NF1 individuals (46.9%) was 4-fold higher than in controls (10.2%).

Forty-six NF1 participants answered to the questionnaire. The number of positive answers per individual to the oral dryness questions was assessed to investigate them as possible predictor of hyposalivation. The results are shown in Tables 3–4.

Table 3 Hyposalivation and oral dryness questionnaire in study group (with NF1)
Full size table
Table 4 Score of positive answers from oral dryness assessment questionnaire according to hyposalivation in study group (with NF1)
Full size table

Oral candidiasis was present in 22% (11/49) of NF1 individuals. Of these, 54% (6/49) had no clinical signs or symptoms of candidiasis, 36% (3/49) had erythematous and 9% (1/49) pseudomembranous candidiasis.

Fifty-five percent (27/49) of NF1 individuals had more than 50% and 16% had more than 80% of the tongue covered by coating. The mean of TCI was 50% (standard deviation = 26.7, median = 50, first quartile = 25, third quartile = 72).

Any of the investigated variables were associated with hyposalivation in NF1 individuals (Table 5).

Table 5 Correlation between hyposalivation and its causes and consequences in study group (with NF1)
Full size table

Discussion

We showed that NF1 individuals present high prevalence (59%) of hyposalivation comparing to a control group (P < 0.001), with a proportion 4-fold higher than the controls.

Xerostomia often occurs when sialometry value is reduced by about 50% [11]. Due to this lack of correlation between xerostomia and hyposalivation, some authors have investigated whether positive answers to other questions related to the dry mouth symptoms have correlation with sialometry values [11]. In our study, hyposalivation was not associated with xerostomia in NF1 individuals, neither with other questions from the oral dryness questionnaire.

Xerostomia can also occur in individuals with normal sialometry values [11,12]. It can occur in mouth breathers or because of changes in the sialochemistry, viscoelastic properties of saliva or by sensory alterations. In the present study, there was no NF1 individual who complained of xerostomia and did not present hyposalivation.

Saliva exerts many functions in oral cavity, including the control of the composition of the microflora. We found 22% of NF1 participants with cytological evidence of oral candidiasis and 55% with 50% of the tongue covered by coating, but none were associated with hyposalivation.

None of the investigated causes of hyposalivation (medications, low liquid intake, caffeinated or stimulant drink use, alcohol, and smoke) were associated with low salivary flow rate in NF1. Moreover, only 4 (8%) NF1 individuals had plexiform neurofibromas in areas of major salivary glands, which could cause atrophy of the acinar cells by compression or infiltration by the tumor cells, and just one of them had hyposalivation.

Alterations in the salivary glands (acini and ducts) caused by mutations in the NF1 gene may be a possible explanation for the high prevalence of hyposalivation in NF1 individuals. Neurofibromin is a negative regulator of the Ras pathways and there is a complex signaling cross-talks between neurofibromin and other members of the superfamily of small GTPases, including Rho binding domain [13]. Rho family of small GTPases presents a crucial role in lumen morphogenesis of salivary glands in animal models and in acinus formation in human salivary gland cell line [13-16]. Kimura et al. [17] showed that neurofibromin is strongly expressed in ductal cells of the parotids. Other salivary glands were not evaluated in that study.

Moreover, the volume and type of saliva is controlled by autonomic nervous system and the blood supply to the glands influences the salivary secretion. Since neurofibromin is expressed in peripheral and central nervous system, as well as in blood vessel smooth muscle and endothelial cells [18-20], alteration in autonomic nervous system and in blood flux to the salivary glands may also be involved in hyposalivation in NF1 individuals.

Conclusions

Hyposalivation may be a consequence of NF1, as occurs in other genetic diseases. More studies are necessary to understand if there is and what is the relationship between NF1 and hyposalivation.

Availability of supporting data

The data sets supporting the results of this article are included within the paper and its additional files.