Discussion

Periodontal plastic and implant surgeries are complex, technique-

sensitive procedures that require a high level of expertise [36].
Hence, various factors should be addressed in the preoperative stage
to ensure optimal outcomes [37]. For a soft tissue graft, the planning
phase of a mucogingival surgery procedure should con sider the
tissue quantity and the optimal donor site [14]. T he present study
aimed to use the data typically col lected in backward planning or
navigated oral surgery (CBCTs and surface scans) for performing a
volumet ric digital soft tissue analysis. The currently established
measurement methods for determining intraoral tissue thickness
include direct and indirect procedures. The "classic" method is a
direct, invasive bone sounding, with a needle or periodontal probe
[15, 38]. However, the data obtained are limited, because only a few
punc tual linear measurements are made, and those may be biased
by the uneven structure of the alveolar bone or a non-perpendicular
position [26]. Due to the neces sary anaesthesia, bone sounding is
typically performed intraoperatively; consequently, preoperative
planning based on tissue thickness is not applicable. Later stud ies
pointed out the need for an indirect technique, with radiological
imaging or ultrasound examination, which could be performed
preoperatively [23, 24, 27, 28, 39–41]. T he present clinical study
developed a method for obtaining a volumetric, 3D representation of
the palatal soft tissues. Moreover, with this method, patients were
spared unnecessary anaesthesia. In the present study, no direct
measurement was performed as a control that could be compared to
the obtained data, because only two-dimensional data (linear
measurements) were avail able for comparisons. Indeed, to date, no
other 3D evalu ation of palatal soft tissue volume has been described
in the literature. Therefore, to put our MeanDist results (Table 2) into
the context of the linear measurements found in the literature,
appropriate articles were selected manually (Table 3) to provide an
overview of the typical 2D measurements performed and mean
results for the palatal mucosa. T he results of our linear
measurements (MeanDist, Table 2) were consistent with the trend of
palatal soft tissue thicknesses shown in Table 3. Our linear meas
urements of 1.9–4.0 mm were comparable to the mean values of
direct measurements (2.0–3.7 mm) reported in a study that
performed direct probing in 62 subjects [38]. Our proposed 3D virtual
analysis (Fig. 4) was consistent with reports in the literature that
showed that the soft tissue thicknesses tended to increase from the
anterior to the posterior palate [27, 38]. It should be noted that our
evaluations included the masticatory mucosa of both the lateral
palate and the posterior palate. Consistent with the findings of Studer
et al. [15], our results showed that the keratinised mucosa was
significantly thicker at the tuberosity and posterior palate (ROI 5,
Table 2; Fig. 3) than in the rest of the hard palate. Furthermore, our
results showed that the standard deviations of both the MeanDist and
the Vol increased from anterior to posterior. This result suggested that
tissue thicknesses showed greater anatomical variation near the
tuberosity and the entry of the vascular-nerve bundle, compared to
other regions. This observation was reported in most of the studies
shown in Table 3 (compare the standard val ues for first premolar [P1]
region to those of the second molar [M2] region). Wara-aswapati et al.
[38] found that the overall thickness of the palatal masticatory
mucosa increased from the gingival margin to the lateral pal ate, and
then, decreased to the midline. That finding explained the variation in
our results within a single seg ment, because the soft tissue volume
at the midpalate is lower than the volume at the lateral palate, as
shown in the colour-coded distance maps (Fig. 3), where the red
colour indicates the largest volume and the green colour indicates no
volume (i.e., contacting surfaces). A spatial concept of each patient’s
individual anatomy, as displayed in Fig. 3, prior to surgery has
advantages. Intraoperative arterial bleeding and postoperative haem
orrhage are risks that may arise when autologous grafts are extracted
from the lateral palate, due to the proxim ity of the greater palatine
artery. The digital analysis pre sented here showed that the highest
volume was found in the most dorsal part of the hard palate on both
sides of the maxilla (Fig. 3). This site coincides with the location of the
vascular-nerve bundle that exits the greater pala tine foramen. That
bundle is typically described as distal to the third molar or between
third and second molars, about 10–14 mm from the gingival margin
[42]. These anatomical features were also found in the measure
ments acquired in this study; however, the findings must be
interpreted with caution. For example, ROI 5 showed particularly high
volume (Table 2), because the measure ment included both the actual
soft tissue volume and the volume of the palatal vascular-nerve
bundle. T here are indications that may require a particularly thick
soft tissue graft, such as implant-related or pre prosthetic scenarios or
to compensate tissue volume loss after a tooth extraction. In these
cases, a presurgi cal soft tissue volume analysis might be beneficial.
How ever, it is not sufficient to analyse solely the quantity of soft
tissue; previous research has shown that good tissue quality (e.g.,
tissue rich in lamina propria) is crucial for surgical success. The layers
of palatal soft tissue include the epithelium (approximately 0.30 –
0.44 mm thick), the connective tissue with the lamina propria (0.8 –
1.5 mm thick), and the submucosal tissue with its fatty compo nent
[43, 44]. It should be highlighted that the present study focused on
the quantitative analysis of palatal soft tissue with 3D imaging to
provide information about the anatomical conditions of the entire
palate, but it did not investigate tissue quality. Table 3 shows that
examining radiological data in sec tional views is a popular tool for
soft tissue analysis. Some studies measured the palatal masticatory
mucosa thickness in 2D sectional views of CBCTs. Ogawa et al.
showed that punctual linear measurements on CBCTs were 0.34 (±
0.04) mm smaller than direct measure ments with a K-file [26].
Considering the low contrast resolution, in general, soft tissue
analyses are limited with CBCTs [45]; therefore, soft tissue
measurements are only suitable to a limited extent in CBCTs.
Consequently, we integrated a surface scan into our method of
measuring soft tissue, by superimposing an intraoral surface model
and a CBCT model. As stated, there is a need for an easy, indirect,
read ily available method for measuring the soft tissue of the palate.
The tissue volume measurement procedure performed in this study
was indirect and non-invasive. Although a CBCT was used, it was not
obtained for study purposes, but for navigated implant surgery or
another medically justified indication. CBCT scans have become an
indispensable part of everyday clinical practice in the f ield of oral and
craniofacial surgery, but of course, the cost–benefit for the patient
must be weighed, and radi ological exposure should only be applied
when justified [46]. Compared to CBCT, magnetic resonance imaging
(MRI) might be superior for visualising intraoral hard and soft tissues
while providing reliable data [41]. MRI has not been clinically
established for dental purposes, it is expensive and requires a long
examination time. There fore, MRI is not currently an option for
replacing the radiation emitting imaging of CBCT and MSCT. Mucosa
measurements performed with MRI provide results com parable to
those obtained with direct bone sounding [47], and MRI is the only
absolutely non-invasive method, with no radiation risk. T he individual
steps applied in this study were based on previous research that
examined the accuracy of each step. The precision of the high-
performance industrial optical scanner (Atos SO II, GOM GmbH) for
digitis ing the impressions was 3 µm [34], and the accuracy of the
converted 3D CBCT models was 400 (± 229) µm for 0.3 voxels [48].
The merging workflow was performed in a precise, two stage
approach (manually and with a best f it-algorithm), where images
were registered by the tooth surfaces. Accurate imaging and precise
superimposition are the foundation for the proposed workflow and
relia ble results. A recent study found that the matching accu racy
was 300 µm with manual alignments of data (CBCT scan + surface
scan) from patients without metallic res torations in the region of
interest [32]. Data merging via the tooth surfaces was particularly
accurate for patients without metallic restorations; therefore, the
presence of a metallic restoration was selected as an exclusion
criterion for participating in this study. A valid, reliable alternative to
taking impressions of patients with high-precision sili cone in this
study would have been a direct intraoral scan as intraoral scans can
correctly display both the teeth and the soft tissues [49, 50]. Merging
3D data provides information about anatomy, aesthetics, and
function, before therapy has begun. In the f ield of digital dentistry,
innovations arise at a fast pace, and they constantly lead to the
development of therapy concepts and materials for oral rehabilitation
based on 3D datasets. Recent innovations have included a fully digital
workflow for dental prostheses [51, 52] and the emergence of highly
functional materials, like oxide ceramics [53], fibre-reinforced
composites [54] and glass/ carbon fibres [55]. None of the studies
listed in Table 3 measured soft tis sue volume (in mm3). Therefore, no
direct comparisons could be made with previous findings.
Nevertheless, the results of this study should be interpreted with cau
tion, because the sample size limited the reliability of the obtained
measurements. It should be noted that this study was the first to test
the possibility of 3D volumetric measurements of palatal soft tissue.
However, a 3D examination of soft tissue changes has been
established and is currently used in clinical research [56]: That
investigation technique involves superimpos ing pre- and
postoperative intraoral surface scans or model scans, and then
comparing them to determine the average tissue thickness and
volume relative to baseline (i.e., preoperative values). For example, a
recent study investigated the changes in palatal soft tissue, in terms
of wound healing, after graft harvesting [57]. However, that technique
solely compared the gingival surface at differ ent time points, and
CBCT scans were not integrated; therefore, only alterations in soft
tissue relative to base line were analysed. In contrast, our study
showed that the proposed method could be used to determine,
volumetri cally, the absolute measurements of the soft tissue in the
palate.