Determination of the palatal masticatory mucosa thickness by dental MRI: a prospective study analysing age and gender effects

Objectives:

The aim of this prospective study was to evaluate age and gender effects on the palatal masticatory mucosa thickness by using non-invasive and non-ionizing MRI.

Methods:

40 periodontally healthy participants of five gender-balanced age groups (20–29, 30–39, 40–49, 50–59 and 60–69 years, respectively) underwent dental MRI at 3 Tesla using a contrast-enhanced, high-resolution 3D-sequence. The palatal masticatory mucosa was measured at 40 standard measurement points by two independent observers. Statistical analysis was performed by using intraclass correlation coefficient (ICC), Shapiro-Wilk test, two-way analysis of variance and post-hoc Tukey HSD test.

Results:

Measurements of the palatal masticatory mucosa thickness were highly reliable with a mean intraobserver ICC of 0.989 and a mean interobserver ICC of 0.987. Mean palatal masticatory mucosa thickness increased with the distance from the gingival margin in all tooth regions. Molars showed a considerably lower average palatal masticatory mucosa thickness at intermediate heights in comparison to canines and premolars. Average palatal masticatory mucosa thickness continuously increased with age and significantly differed between age groups (p < 0.01). A significant increase was observed between the age groups of 30–39 years and 40–49 years (p = 0.04). Gender had no significant effect on average palatal mucosa thickness (p = 0.19). However, there was a tendency towards thicker mucosa in males (mean ± SD, 3.36 mm ± 0.47) compared to females (3.23 mm ± 0.44).

Conclusions:

In the present study, dental MRI allowed for a highly reliable determination of the palatal masticatory mucosa thickness. Considerable intra- and interindividual variations in palatal masticatory mucosa thickness were observed. Average palatal masticatory mucosa thickness was dependent on age but not on gender.

Introduction

The palatal masticatory mucosa is the main donor site for soft tissue grafts in periodontal plastic surgery. Tissue grafts from this area are used in various surgical procedures such as root coverage,^1,2 augmentation around implants,^3,4 augmentation of alveolar ridge thickness⁵ or vestibuloplasty.⁶ Due to a high degree of intraindividual and interindividual variation^7,8 the thickness of the palatal masticatory mucosa should be assessed before surgery. Various methods for the determination of the palatal masticatory mucosa thickness have been suggested, including bone sounding,^9–11 ultrasound,^7,12 conventional computed tomography (CT)¹³ and cone-beam computed tomography (CBCT).¹⁴

MRI is the modality of choice in head and neck imaging and allows for an excellent visualization of soft tissues. A major benefit of this non-invasive and widely available imaging technique is that it does not involve ionizing radiation. Despite these advantages, MRI has not yet been established in routine dental imaging. In the past, the development and implementation of dental MRI was limited due to the lack of three-dimensional (3D) sequences, comparatively low resolution and long examination time. However, recent technical improvements such high field strength systems,¹⁵ parallel imaging methods,¹⁶ 3D-techniques¹⁷ and dedicated coil systems^18–20 have made decisive contributions to overcome these limitations. These technical milestones open a wide range of possible applications of dental MRI.

We developed a 3D dental MRI technique for high-contrast visualization of hard and soft tissues. Using this non-invasive and non-ionizing technique, a prospective in vivo study was conducted to evaluate whether dental MRI allows to assess the palatal masticatory mucosa thickness at established measurement points. We were particularly interested in analysing effects of age and gender on the palatal masticatory mucosa thickness.

Methods and materials

Patients

40 periodontally healthy Caucasian participants were prospectively enrolled in this study (20 females, 20 males). The study cohort consisted of five gender-balanced age groups of 20–29, 30–39, 40–49, 50–59 and 60–69 years, respectively. Mean age ± SD was 44.2 years ± 14.7, median age was 45.0 years, range was 23–69 years. All participants presented with a full dentition in the upper jaw from the canines to the second molars on both sides and received a complete dental examination including the assessment of periodontal conditions. Exclusion criteria were: destructive periodontal disease with a possible exception of localized gingival recession, history of soft tissue removal or other surgical interventions in the palate, history or present diagnosis of any disease in the palate, asymmetries of the maxilla requiring treatment, use of any medication possibly affecting gingival thickness (e.g. cyclosporine A, calcium channel blockers or phenytoin) and contraindications to contrast enhanced MRI.

MRI examinations

All MRI examinations were performed at a 3T MRI system (MAGNETOM TIM Trio, Siemens Healthineers GmbH, Erlangen, Germany) with a 16-channel multipurpose coil (Variety, Noras MRI products GmbH, Hoechberg, Germany). A T  ₁ weighted, isotropic VIBE (Volumetric Interpolated Breath-hold Examination) sequence with Dixon-based fat suppression^21,22 was applied after intravenous administration of 0.1 mmol kg⁻¹ gadoterate (Gd) meglumine contrast (Dotarem®, Guerbet, France). Sequence parameters were: 3D sequence, time of echo: 2.45 ms, time of repetition: 15.6 ms, bandwidth: 601 Hz/pixel, number of averages: 1, echo train length: 1, flip angle: 14.9°, FOV: 153 × 223 mm², acquisition matrix: 220 × 320, slice thickness: 0.7 mm, number of sections: 88, time of acquisition: 6:50 min. The Dixon technique acquires two sets of images simultaneously, one with fat and water signals in-phase and one with fat and water signals out of phase, resulting in 4 different sequences: (1) acquired in-phase (water + fat), (2) acquired opposed-phase (water − fat), (3) reconstructed fat only (in-phase - opposed-phase) and (4) reconstructed water only (in-phase + opposed phase). Measurements were performed using the water only, fat suppressed images. A representative image from primary MRI datasets is shown in Figure 1.

Analysis of MRI datasets

All measurements were performed using DICOM Imaging Software (Osirix v. 7.0.3, Geneva, Switzerland). As illustrated in Figure 2, a total of 40 predetermined measurements were taken for each study subject. Multiplanar reconstructions (MPR) were obtained separately for canines, premolars and molars on both sides according to the straight lines in Figure 2. Subsequently, for each tooth region measurements of the palatal masticatory mucosa thickness were performed at distances of 2, 5, 8 and 12 mm from the gingival margin (Figures 2 and 3). AH (a radiologist with 4 years of experience in dental imaging) performed measurements of all subjects. For the assessment of intraobserver agreement, AH repeated all measurements in 10 subjects (one female and one male subject per age group) with an interval of 4 weeks to exclude learning effects. TH (a radiologist with 4 years of experience in dental imaging) performed all measurements of the same 10 subjects as well to assess interobserver agreement.

Statistical analysis

Statistical analysis was performed by using software (SPSS v. 24, IBM Corporation, Armonk, NY). Intra- and interobserver agreement was analysed by using intraclass correlation coefficient (ICC) including the 95% confidence interval (CI). Shapiro-Wilk test was conducted to check for normal distribution of the palatal masticatory mucosa thickness. Next, a two-way analysis of variance (ANOVA) was performed to test for the influence of predefined age groups and gender on the palatal masticatory mucosa thickness. Predefined age groups were 20–29, 30–39, 40–49, 50–59 and 60–69 years, respectively. Post-hoc Tukey HSD test was used to test for differences among the age groups. All p-values less than 0.05 were considered to indicate a significant difference.

Results

The acquired MRI datasets allowed a clear depiction of the gingival margins, the surface of the palatal masticatory mucosa and the bony structures (Figures 1 and 3). Intraobserver as well as interobserver agreement was excellent for all measurement points. According to tooth region, intraobserver ICC (95% CI) was 0.989 (0.983–0.993) for canines, 0.991 (0.986–0.995) for first premolars, 0.988 (0.982–0.993) for second premolars, 0.990 (0.984–0.993) for first molars and 0.986 (0.978–0.991) for second molars, respectively. Interobserver ICC (95% CI) was 0.986 (0.979–0.991) for canines, 0.986 (0.978–0.991) for first premolars, 0.985 (0.977–0.990) for second premolars, 0.991 (0.987–0.995) for first molars and 0.987 (0.979–0.992) for second molars, respectively. For all measurement points, intraobserver ICC (95% CI) was 0.989 (0.987–0.991) and interobserver ICC was 0.987 (0.985–0.990).

Table 1 summarizes the mean values and standard deviations of all measurement points for all study subjects, the five age groups as well as females and males. Figure 4 illustrates the mean values for all measurement points showing that measurements were very similar on the right and the left side. Data from Table 1 and Figure 4 reveal that the mucosa thickness increased with the distance from the gingival margin in all tooth regions (2 < 5 < 8 < 12 mm). Furthermore, mucosa thickness of first and second molars was considerably lower at intermediate heights (5 and 8 mm) in comparison to premolars and canines.

As shown in Table 1, average palatal masticatory mucosa thickness consistently increased by age, mean values ± SD were 2.84 mm ± 0.21 (20–29 y), 2.95 mm ± 0.19 (30–39 y), 3.44 mm ± 0.20 (40–49 y), 3.56 mm ± 0.37 (50–59 y) and 3.68 mm ± 0.48 (60–69 y), respectively. As an example, equivalent measurements of two study participants in the region of the left second premolars are presented in Figure 3: A 33-year-old male with comparably thin mucosa (Figure 3a) in comparison to a 59-year-old male with considerably thicker mucosa (Figure 3b). Normal distribution of the palatal masticatory mucosa thickness was confirmed by Shapiro-Wilk test and a significant difference between the five age groups was observed by using 2-way ANOVA (p < 0.01). Post-hoc comparisons indicated a significant difference between the age groups of 30–39 years and 40–49 years (Tukey HSD: p = 0.04), which is also illustrated by the box plots of the five age groups in Figure 5a.

Mean values ± SD of palatal masticatory mucosa thickness were 3.23 mm ± 0.44 in females compared to 3.36 mm ± 0.47 in males (Table 1). Cumulative data from female and male participants are shown in the box plots in Figure 5b. Even though there was a tendency towards thicker palatal masticatory mucosa in males, gender effects did not reach statistical significance (2-way ANOVA: p = 0.19).

All age groups revealed a substantial variance of palatal masticatory mucosa thickness. Importantly, the widest levels of variance were found in the older age groups. Mean standard deviations were 0.48 and 0.37 mm for the age Groups 60–69 and 50–59 years, respectively. In comparison, mean standard deviations were 0.20, 0.19 and 0.21 mm for the age Groups 40–49, 30–39 and 20–29 years, respectively (Table 1). The degree of dispersion of mean palatal masticatory mucosa thickness in the different age groups is also indicated by the boxplots in Figure 5a. Both genders showed a comparable level of variability in palatal masticatory mucosa thickness (Table 1 and Figure 5b).

Discussion

The present study provides the proof of concept that the thickness of the palatal masticatory mucosa can be reliably determined in vivo by using 3D high-resolution dental MRI. This provides the opportunity for a non-invasive assessment of the palatal masticatory mucosa thickness on both sides of the maxilla in absence radiation exposure.

Our measurements of the palatal masticatory mucosa thickness in different tooth regions showed an increasing thickness from lower to higher measurement points and a distinct decrease at intermediate measurement heights in molars in comparison to premolars and canines. These results correspond very well to earlier studies determining the palatal masticatory mucosa thickness by different measurement techniques. Barriviera et al assessed the palatal masticatory mucosa thickness by using CBCT in a group of 31 patients (20 females) applying exactly the same 40 predetermined measurement points that we used. Importantly, they found an almost identical pattern of mucosa thickness in equivalent tooth regions and measurement heights. Their mean values were expectedly smaller due to a substantially lower mean age of 32 years compared to 44 years in our study.¹⁴ Moreover, the typical pattern of region-dependent thickness of the palatal masticatory mucosa was reported using conventional CT,¹³ ultrasonic devices⁸ and bone sounding probes.^10,11 As our results and multiple previous studies consistently demonstrated a comparatively thin palatal masticatory mucosa at the prominence of the roots of first and second molars, the canine and premolar area appear to be the most appropriate donor site for soft tissue augmentation procedures.

It should be outlined, that the here presented study allowed a clear differentiation between age groups and gender. To the best of our knowledge, for the first time a clear age and gender balanced cohort of periodontally healthy subjects covering young, middle-aged and older adults was examined prospectively to analyse the palatal masticatory mucosa thickness. In comparison to our data, former studies evaluating age and gender effects were performed with unbalanced cohorts which did not present the preferable division in gender-balanced age groups by decades. Therefore, those studies had relevant bias effects and were less conclusive. However, considering this lack of direct comparability, there are many parallels between our results and earlier studies using various methods to determine the palatal masticatory mucosa thickness.

We could show a continuous increase of palatal masticatory mucosa thickness with advancing age. Our data also suggest that the biggest increase occurs in the 40 s, as average palatal masticatory mucosa thickness was significantly higher in the age group of 40–49 years in comparison with the age group of 30–39 years. Wara-aswapati et al used the bone sounding method to measure the palatal masticatory mucosa thickness in 62 healthy Asians (31 females; age range: 14–59 years). They found a significant influence of age by dividing the cohort into a young age group (14–21 years) and an older age group (30–59 years) with thicker mucosa in the older group. Thus, their results correspond very well with ours.¹¹ Song et al selected CT-scans of 100 subjects (42 females, age range 17–71 years) retrospectively. Within the limitations of unbalanced age groups, they found an increase of the average palatal masticatory mucosa thickness with age. Compared to our study, their measurement heights were a little more distant from the gingival margin. According to this, the average mucosa thickness was slightly higher in similar age groups.¹³ Within the limitation of a relatively young cohort not being ideal to test for age effects, Barriviera et al found a thicker mucosa in patients older than 40 years compared to patients younger than 40 years.¹⁴

In the present study, gender had no statistically significant effect on the palatal masticatory mucosa thickness. However, there was a tendency towards thicker mucosa in male subjects. This finding corresponds very well with the data from Wara-aswapati et al who also found thinner palatal masticatory in females compared to males using the bone sounding method, but the differences were not significant either.¹¹ Similarly, Studer et al found no significant effect of gender by performing bone sounding¹⁰ and the CBCT-study by Barriviera et al did not reveal significant gender effects either.¹⁴ Other investigators, however, found significantly thinner palatal masticatory mucosa in females using biopsies,²³ ultrasound⁸ and conventional CT.¹³ In consideration of our data and previous reports, it can be concluded that the palatal masticatory mucosa is slightly thinner in females on average, although statistical significance is not always reached.

Although we found a consistent increase of palatal masticatory mucosa thickness with age, it should be stressed that there was a high intra- and interindividual variability among all age groups for both females and males. This finding is in line with previous studies reporting high levels of variance of the palatal masticatory mucosa thickness.^7,8 Therefore, the palatal mucosa thickness cannot be estimated by “age dependent expected values”. Due to the high variance within all age groups, an individual pre-surgical assessment of the palatal masticatory mucosa thickness can be beneficial for treatment planning and treatment outcome. Interestingly, we observed the largest variance in the older age groups (50–59 and 60–69 years), which are more likely to have dental diseases requiring soft tissue grafts.

At this point in time, the clinical use of dental MRI is associated with concerns about costs and availability. Consequently, the applicability of dental MRI to assess the palatal masticatory mucosa thickness is limited. However, the dental MRI technique applied here is not restricted to quantitative assessment of the palatal masticatory mucosa, as it covers all teeth and periodontal structures of the maxilla as well as the mandible with high contrast and high resolution. Further potential fields of application of dental MRI have been published recently, including implantology,²⁴ periodontology,²⁵ endodontics^26,27 and orthodontics.²⁸ This opens a wide range of potential indications for dental MRI including the assessment of periodontal diseases as well as preoperative planning in Oral Surgery and Implant Dentistry. Thus, whenever a MRI examination is performed, individual measurements of the palatal masticatory mucosa thickness could be used as an add-on, resulting in additional information without increasing costs.

We developed a dedicated dental MRI technique for the present study. The acquired high-resolution 3D images allowed a clear depiction of all predefined measurement points. Based on this, the resulting measurements turned out to be highly reliable in all measurement points among all subjects which is reflected by high levels of intraobserver and interobserver agreement. A high-resolution 2D sequence, however, was not evaluated for the determination of the palatal masticatory mucosa thickness. Future studies should include 2D sequences as well, as this could further decrease acquisition time. When using 2D sequences, however, slice direction must be chosen carefully to reduce the potential impact of partial-volume effects. In this regard, especially the application of intraoral markers (e.g. water with or without gadolinium) could allow for accurate acquisition planning.

Apart from the finding that dental MRI allows for non-invasive and radiation-free determination of palatal masticatory mucosa thickness, another advantage of dental MRI is its excellent visualization of soft tissues. This feature is a major benefit in comparison to CBCT, which does not provide adequate soft tissue contrast.²⁹ Future studies should assess whether this could lead to an improvement of preoperative planning and reduction of complication rates. Therefore, it would be desirable to develop MRI sequences that allow for tracking the course of the palatine vessels, which might assist in identifying the ideal harvesting site. Furthermore, it would be beneficial to evaluate whether dental MRI can determine the palatal masticatory mucosa thickness without the use of contrast media (e.g. by using optimized proton density or T2 weighted sequences) to reduce costs and exclude the risk of intolerance reactions.

A limitation of this study was that measurement accuracy could not be proven directly. However, high geometric accuracy was proven for the dental MRI technique applied in the study.³⁰ Thus, high measurement accuracy can be assumed. This is in line with the observation that the present study confirmed the characteristic pattern of region-dependent thickness of the palatal masticatory mucosa which has been shown before by using different accurate methods such as bone sounding,^10,11 conventional CT¹³ and CBCT.¹⁴

A further limitation of the here presented radiation-free MRI approach is that it can be affected by artefacts caused by dental restorations and implants. Highly paramagnetic dental materials containing stainless steel and or non-precious alloys are known to cause considerable signal-loss artefacts.^31,32 Therefore, image quality in patients presenting with those materials can be reduced in dental MRI, similarly to metal artefacts in conventional CT or CBCT.

Conclusions

This prospective study shows that the determination of the palatal masticatory mucosa thickness is feasible based on dental MRI. Our results revealed an age-dependent increase of palatal masticatory mucosa thickness with a significant difference between the age groups of 30–39 and 40–49 years. Gender had no significant effect on average palatal masticatory mucosa thickness.

Acknowledgements

The authors kindly thank the Dietmar Hopp Foundation for supporting this project and NORAS MRI products GmbH for providing the 16-channel multipurpose coil used in this study.

References

Author notes