The gingival biotype measurement of soft and hard tissue dimensions a radiographic morphometric stud by Carlos

J Clin Periodontol 2013; 40: 1132–1139 doi: 10.1111/jcpe.12169

The gingival biotype: measurement of soft and hard tissue dimensions - a radiographic morphometric study

€ ping1, Jamal M. Stein1,2, Nils Lintel-Ho Christian Hamma¨cher2, Adrian Kasaj3, Miriam Tamm4 and Oliver Hanisch5 1

Department of Operative Dentistry, Periodontology and Preventive Dentistry, University Hospital (RWTH), Aachen, Germany; 2Private Practice, Aachen, Germany; 3Department of Operative Dentistry and Periodontology, University Medical Center, Mainz, Germany; 4Department of Medical Statistics, University Hospital (RWTH), Aachen, Germany; 5Private Practice, Paris, France

Stein JM, Lintel-Ho¨ping N, Hamma¨cher C, Kasaj A, Tamm M, Hanisch O. The gingival biotype: measurement of soft and hard tissue dimensions - a radiographic morphometric study. J Clin Periodontol 2013; 40: 1132–1139. doi: 10.1111/ jcpe.12169.

Abstract Background: Gingival biotypes have been reported to influence the outcome of restorative therapies. The aim of this study was to evaluate the correlation of different morphometric parameters with the thickness of the buccal gingiva and alveolar bone at different apico-coronal levels. Methods: In 60 periodontally healthy subjects, the central maxillary incisor was examined. Clinical parameters included the crown width/crown length ratio (CW/ CL), gingival width (GW), gingival scallop (SC) and transparency of the periodontal probe through the gingival sulcus (TRAN). Gingival and alveolar bone dimensions were assessed on parallel profile radiographs. Results: Crown width/crown length ratio was positively correlated with the thickness of the gingiva at the cementoenamel junction (G3) (r = 0.47) and to the thickness of the alveolar crest (A1) (r = 0.46); whereas SC had a weak negative and GW had a moderate positive correlation with all radiographic measurements. TRAN had a stronger negative relation to the thickness at the free gingiva (r = 0.42) than to other tissue thicknesses. All gingival thickness values were correlated with A1 value. Multivariate models identified CW/CL and GW as significant predictors for G3 value, whereas CW/CL was a significant predictor for A1 value. Conclusion: Crown width/crown length ratio and GW could represent surrogate parameters to anticipate the gingival thickness at the cementoenamel junction, whereas CW/CL might also be an indicator for alveolar bone crest thickness. Periodontal probing has a limited prognostic value for these tissue dimensions.

Different tissue biotypes have been suggested to influence the outcome of Conflict of interest and source of funding statement The authors declare that they have no conflict of interests. No external funding, apart from the support of the author’s institution, was available for this study.

1132

restorative treatments (Pontoriero & Carnevale 2001, Evans & Chen 2008). Thereby, gingival thickness seems to play a decisive role. The so-called “thick-flat” gingival biotype has been reported to be a prognostic factor for aesthetically successful outcomes of implants (Kois 2004), predictable results after recession coverage (Baldi et al. 1999, Hwang & Wang 2006) and regain of gingiva after resective

Key words: alveolar bone; crown form; gingival biotype; gingival thickness Accepted for publication 8 September 2013

osseous surgery (Pontoriero & Carnevale 2001). In contrast, patients with a “thin-scalloped” gingiva had a higher risk for periodontal recessions after placement of immediate implants (Evans & Chen 2008), a smaller number of complete root coverage after root coverage procedures (Baldi et al. 1999) and a trend towards soft tissue loss in case of periodontal inflammation (Ericsson & Lindhe

Gingival and alveolar bone dimensions 1984, Wennstr€ om et al. 1987). One reason for these observations might be the correlation between gingival thickness and the dimensions of the underlying alveolar bone plate (Seibert & Lindhe 1989). In patients with a thin gingival biotype, the absence or deficiency (fenestration, dehiscence) of the buccal bone may cause a trend towards periodontal recessions after periodontal treatments (Larato 1970), increase the amount of ridge resorption after tooth extractions (Kao et al. 2008) and reduce the stability of the inter-dental papillae after the insertion of immediate implants (Romeo et al. 2008). However, the association between thickness of the gingiva and thickness of the underlying alveolar bone has only been analysed in one study and it was concluded to be moderate (Fu et al. 2010). To increase the predictability of reconstructive periodontal and (peri-) implant therapies, it is important to develop guiding criteria to differentiate thick and thin periodontal biotypes. Several methods have been described to classify the gingival thickness: direct measurement (Greenberg et al. 1976), ultrasonic device (M€ uller et al. 2000), cone beam computed tomography (Barriviera et al. 2009) and probe transparency through the free gingiva (Kan et al. 2010). While direct measurement of gingival thickness is an invasive method with limitations of reproducibility, non-invasive ultrasonic devices could not be established as routine devices probably due to technical reasons (Vandana & Savitha 2005) and costs. Similarly, extensive radiographic diagnostics such as cone beam computed tomography does not appear appropriate as a first choice method for defining gingival biotypes. Therefore, visual inspection of the transparency of the periodontal probe through the sulcus had become the most frequently used method for discrimination of thin and thick biotypes (De Rouck et al. 2009, Kan et al. 2010). Nevertheless, the prognostic value of probe transparency has not been analysed in detail until now. Furthermore, surrogate parameters such as the crown form (squared short versus tapered long) (Olsson & Lindhe 1991) and the height of the gingival scallop (low versus high) (Weisgold 1977, Seibert & Lindhe

1989) have also been associated with the presence of a thick or thin gingiva respectively. The findings, however, were not consistent (Olsson et al. 1993, De Rouck et al. 2009) and should be verified. Therefore, the objective of this study was to examine how previously reported surrogate parameters correlate with gingival thickness using a radiographic morphometric method. In particular, the aim was to:

•

• •

evaluate the correlation between crown form, height of gingival scallop, width of keratinized gingiva and periodontal probe transparency through the sulcus with gingival thickness and thickness of the buccal alveolar bone plate at different apico-coronal levels; assess the relationship between gingival thickness and buccal alveolar bone thickness; and define the diagnostic parameter(s) with the highest correlation with gingival thickness at the supracrestal attachment and alveolar bone crest in multivariate models.

Material and Methods

1133

enrolled at the Department of Operative Dentistry, Periodontology and Preventive Dentistry, University Hospital Aachen between January and August 2012. Of this cohort, subjects were excluded if they met one of the following criteria: (i) intake of medicaments known to increase gingival overgrowth, (ii) systemic diseases having gingival manifestations and/or influence the bone metabolism, (iii) pregnancy, (iv) presence of periodontal probing depths ≥4 mm, (v) periodontal recessions, (vi) signs of incisal abrasions and (vii) crown restorations or fillings in the upper central incisor area. All study participants not meeting those exclusion criteria underwent a clinical oral and radiographic examination. The study protocol was approved by the Ethics Committee of the University of Aachen and written consent was obtained from all subjects before their examinations. After clinical and radiographic examination, quality control of the radiographs led to a second exclusion of subjects not meeting the quality criteria (see Quality Control). Finally, the final study cohort consisted of 60 participants. Figure 1 shows a flow diagram on selection of all subjects according to the described criteria.

Study population

The study was designed as descriptive cross-sectional study. A total of 85 volunteers without known periodontal or dental diseases had initially been

Clinical examination and photographic analysis

All clinical oral examinations have been performed on the left central

Potentially appropriate subjects without known dental diseases (N = 85) Exclusion of subjects because of medical history (N = 2) (Crohn’s disease: 1, Diabetes mellitus Type I: 1)

Potentially appropriate, systemically healthy subjects (N = 83) Exclusion of subjects with dental or periodontal lesions on the upper front teeth (N = 12) (periodontitis = 3, periodontal recessions = 4, restorations = 3, abrasions = 2)

Subjects considered for clinical and radiologic examination (N = 71) Exclusion of subjects due to quality deficiency of radiographic images (N = 11) (superimpositions = 6; not identifiable anatomic landmarks = 5)

Final number of subjects included in the study (N = 60)

Fig. 1. Flow diagram on the selection of study participants according to the inclusion and exclusion criteria.

1134

Stein et al.

incisor (index tooth) both with direct measurements and analyses of a clinical photograph taken from the region of the index tooth. Prior to the photograph, a lead plate (5.0 9 1.0 9 0.1 mm) was used as reference for all measurements on the photograph and the radiograph (Fig. 2a). It was positioned over the buccal gingival surface with its most coronal margin aligned with the edge of the gingival margin at the midbuccal position. It was also well aligned with the long axis of the tooth and delimited the profile of the gingiva from a lateral perspective. After positioning the lead plate on the gingival surface, tissue adhesive (Histoacrylâ; B.Braun, Melsungen, Germany) was applied over the margins of the plate in order to avoid movement. The following assessments were made directly on the patients using a periodontal probe (CP 15 UNC; Hu-Friedy, Rotterdam, Netherlands):

Photoshop CS3; Adobe Systems, San Jose, CA, USA):

•

To analyse the dimensions of the soft and hard tissue structures in the coronal aspect of the periodontium around the index tooth, parallel profile radiographs were obtained from a lateral position with the use of the above-described lead plate according to the method reported by AlpisteIllueca (2004). Regarding the principles of the traditional longcone parallel technique for periapical radiographs, the film was placed parallel to the long axis of the tooth to minimize distortion of the image. A long distance between the generator and the object as well as a minimal distance between the film and the

•

Width of the keratinized gingiva (GW) measured from the midbuccal position of the marginal gingiva to the mucogingival junction, to the nearest 0.5 mm. Transparency of the periodontal probe through the sulcus (TRAN) was determined after insertion of the probe into the sulcus on the midbuccal position and recorded as categorical variable (yes = probe visible; no = probe not visible).

On the clinical photograph the following parameters were recorded using a photo software (Adobeâ

(a)

•

Crown width/crown length ratio (CW/CL): This measurement was performed according to the method of Olsson & Lindhe (1991). Crown length was measured from the incisal edge to the margin of the buccal gingiva. For assessment of the width, the crown length was divided into three equal portions; the width was recorded as the distance between the approximal crown surfaces at the border between the middle and the cervical portion. Height of the gingival scallop (SC): widest distance between the line formed by the connection of the peaks of the two adjacent inter-dental papillae and the most apical position of the buccal marginal gingiva.

Parallel profile radiographs

(b)

object was established to produce a clear image and minimize the magnification effect. The X-ray beam was oriented perpendicular to the axis of the tooth and the film. The paralleling system by Rinn for posterior quadrants was used for this study. The bite block was fixed with the anterior teeth so that the film was positioned on the lateral vestibule. Particular care was paid to the parallel orientation of the film towards the long axis of the tooth. This was achieved by viewing the lead plate through the aiming ring: only the profile of the lead plate had to be seen. From each patient, a digital radiograph (Heliodent DS; Sirona, Wals, Austria) using an intra-oral sensor was obtained (RVG 6100; Carestream Health, Toronto, ON, Canada) (Fig. 2b). Analyses of the radiographs

All images of the digital radiographs have been analysed using a photo software (Adobe Photoshop CS3; Adobe Systems). The following measurements were made on the radiographs:

•

thickness of the free gingiva: distance between the enamel surface to the palatal side of the lead plate measured at the coronal margin (G1) and the base (G2) of the free gingiva thickness of the gingiva at the supracrestal attachment: distance between the root surface and the palatal side of the lead plate measured at the cementoenamel junction (CEJ) (G3), the middle third (midpoint between the

(c)

Fig. 2. Index tooth with fixed transfer lead plate. Clinical view (a), radiographic view (b), radiographic measurement points for assessment of gingival (G1–G6) and alveolar bone (A1–A3) thickness values (c). © 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Gingival and alveolar bone dimensions

• •

distance CEJ - bone crest) (G4) and directly above the bone crest level (G5) thickness of the attached gingiva: distance between the buccal margin of the bone crest and the palatal side of the lead plate (G6) thickness of the buccal alveolar bone plate: distance between the buccal surface and the palatal side (Lamina dura) of the buccal bone plate measured at the bone crest level (A1), at the border between the coronal and middle third (A2) and between the middle and apical third (A3) of the root length

For all measurements, the length of the lead plate in the X-ray was used as a reference for the calculation of all measurements (Fig. 2c).

subgroup of 10 study participants. Therefore, all clinical parameters and those measured on photographs and radiographs were re-measured 1 week after the first recording. The intra-class correlation according to Gwet (2008) was assessed for all parameters and ranged from 0.98 to 1.00 indicating an excellent level of reproducibility. As the visual estimation of probe transparency (TRAN) may underlie a higher degree of subjectivity, for this parameter a second calibration was performed. In the same subgroup, TRAN was estimated by the examiner (NL) and two other dentists (JMS, CH). Interexaminer reliability was calculated using the method of Light (1971) and resulted in an average kappa of 0.86 indicating a good reliability for this assessment.

Quality control

Statistical analysis

The radiographs had to meet several criteria to be considered for the study. First, the lead plate had to be detectable on all X-rays delimiting the gingival profile in each subject. It was important that only the profile of the plate was visible with a minimal thickness over the entire length ensuring the correct tangential position of the index tooth. Second, it had to be assured that the following anatomic landmarks could be clearly identified on the X-rays without superimpositions: lead plate, CEJ, bone crest, buccal surface of the bone plate, buccal root surface. If these criteria were not fulfilled, measurements could not be performed and the subjects were excluded from the study (Fig. 1). Furthermore, the residual potential error due to a non-tangential positioning of the lead plate was calculated by measuring the minimal and maximal deviation of the thickness of the plate in the radiograph from the real thickness (0.10 mm) in all subjects. The mean deviation was 0.10 0.04 mm, whereas maximal and minimal deviation was 0.13 0.04 and 0.07 0.05 mm respectively.

Outcome values of all continuous parameters were given as mean and standard deviation (SD). Using the Pearson correlation coefficient with the corresponding 95% confidence interval, correlations of CW/CL, SC and GW with the thickness of the gingiva at different apico-coronal levels (G1–G6) as well as thickness of the buccal alveolar bone plate at different apico-coronal levels (A1, A2, A3) were calculated. The relationship between TRAN and all thickness measurements was evaluated with the point biserial correlation. Furthermore, multiple linear regression analyses were performed to test the association of CW/CL,

Examiner reliability

All measurements were performed by one examiner (NL). Intra-examiner reliability was evaluated in a

1135

SC, GW and TRAN with the thickness of the supracrestal attachment at the CEJ level (G3) and buccal alveolar bone crest (A1). The requirements of normally distributed residuals, homoscedasticity and linearity were checked using normal probability and residual plots. Furthermore, sensitivity analyses were made to investigate the robustness of the findings. In case where relevant changes were observed, they are stated in the results section. All analyses were done in an explorative manner. Thus, p-values <0.05 were considered statistically significant. The statistical analyses were carried out with SAS Version 9.2 (SAS Institute Inc., Cary, NC, USA). Results

The study population consisted of 60 Caucasian subjects (24 men, 36 women) with a mean age of 31.53 years (range, 18–61 years). Thirty-one per cent of the subjects were light smokers (<5 cigarettes/ day). Table 1 shows the descriptive data of all clinical and radiographic measurements. The subjects comprised crown forms that ranged from a tapered long form with a very low CW/CL of 0.55 to a squared short shape with a maximum CW/CL of 1.0 and an average of 0.72. The mean values for SC and GW were 4.24 and 4.92 mm respectively. The mean thickness of the free gingiva was 0.59 mm at the coronal margin (G1) and increased to 0.96 mm at its base (G2). For the gingival thickness at the supracrestal attachment, the

Table 1. Clinical and radiographic measurements Mean SD or number (%)

Range

0.72 0.09 4.24 0.85 4.92 1.01

0.55–1.00 1.93–6.12 2.50–7.00

0.17 0.24 0.35 0.35 0.37 0.21

0.22–1.13 0.43–1.42 0.52–2.06 0.68–2.13 0.52–2.39 0.47–1.57

0.57 0.23 0.77 0.30 0.85 0.45 28 (46.7%)

0.20–1.58 0.23–1.78 0.23–3.01

Crown width/crown length ratio (CW/CL) Height of gingival scallop (SC) (mm) Gingival width (GW) (mm) Gingival thickness G1 (mm) G2 (mm) G3 (mm) G4 (mm) G5 (mm) G6 (mm) Alveolar bone plate thickness A1 (mm) A2 (mm) A3 (mm) Transparency of the periodontal probe (TRAN) SD, standard deviation.

0.59 0.96 1.25 1.43 1.46 0.79

1136

Stein et al.

mean values in the middle third (G4; 1.43 mm) and directly coronal to the bone crest (G5; 1.46 mm) were minimally higher than at the CEJ (G3; 1.25 mm), whereas the thickness of the attached gingiva over the bone crest (G6; 0.79 mm) was smaller than the gingival thickness at the supracrestal attachment (G3–G5). The mean thickness of the alveolar bone plate increased from 0.57 mm at the crest to 0.85 mm at the apical third of the root length. The periodontal probe that was inserted at the midbuccal aspect of the sulcus was visible in 28 subjects (46.7%), whereas in the other 32 participants (53.3%) it could not be detected. Correlation analyses revealed a positive relationship between CW/ CL and all thickness parameters (Table 2). Hereby, G3 (r = 0.47) and A1 (r = 0.43) showed a stronger correlation than the other parameters (r < 0.4). On the contrary, for SC a negative correlation with all thickness measurements was found. However, all relationships were weak (r > 0.25). The correlation between GW and all thickness parameters was moderately positive for G3, G4,

G5 and A3 (r > 0.25, r < 0.3). Moreover, TRAN was negatively correlated with all thickness values (Table 2). Hereby, TRAN had the strongest correlation with G2 (r = 0.42), whereas all other relationships were weaker (r > 0.4). Besides, the relation between the thickness of the gingiva and the alveolar bone plate was examined (Table 3). There was a positive correlation between the gingival thickness at all levels (G1–G6) and the thickness of the alveolar bone plate (A1–A3). However, the correlations between A1 and thickness of the gingiva at the supracrestal attachment (G3–G5) were the strongest (r: 0.51– 0.56), whereas the correlations between the attached gingiva thickness at the bone crest (G6) and all bone thickness values (A1–A3) were weak (r ≤ 0.25). To further analyse the association of the parameters CW/CL, SC, GW and TRAN with gingival thickness at the CEJ (G3) and alveolar bone crest thickness (A1), these parameters were included in two multiple regression models. CW/CL and GW were revealed as significant predic-

tors for G3, whereas only CW/CL was a significant predictor for A1. All other parameters had no significant impact on G3 and A1 (Tables 4, 5). Discussion

Soft and hard tissue dimensions are important parameters that affect the outcome of periodontal and restorative treatments. Knowledge about these factors may help to better assess the need for soft or hard tissue augmentations and to avoid failures or complications, in particular in the aesthetically critical area. Therefore, it would be useful to have reliable guidelines or surrogate objective parameters for the identification of critical cases with thin gingival and/or alveolar bone thickness, which might compromise the success of the treatment. Different parameters have been used to assess the gingival thickness or the so-called gingival biotype. However, the results and recommendations are controversial and none of the described parameters can be considered as best or most reliable. Up until now, there is no precise

Table 2. Correlation between CW/CL, gingival scallop, width of keratinized gingiva and probe transparency with gingival and alveolar bone thickness

CW/CL

TRAN

0.2373 ( 0.0197, 0.4617) 0.1905 ( 0.4226, 0.0683) 0.0180 ( 0.2705, 0.2371) 0.2406 ( 0.4644, 0.0163)

0.3846 (0.1416, 0.5795) 0.2510 ( 0.4730, 0.0053) 0.1653 ( 0.0939, 0.4011) 0.4169 ( 0.6043, 0.1788)

0.4663 (0.2371, 0.6417) 0.2004 ( 0.4310, 0.0580) 0.2607 (0.0050, 0.4810) 0.2990 ( 0.5123, 0.0465)

0.3855 (0.1426, 0.5802) 0.2302 ( 0.4558, 0.0272) 0.2763 (0.0218, 0.4937) 0.3465 ( 0.5498, 0.0986)

0.2298 ( 0.0276, 0.4555) 0.1510 ( 0.3889, 0.1082) 0.2744 (0.0196, 0.4921) 0.3283 ( 0.5354,

0.0783)

0.1063 ( 0.1526, 0.3500) 0.1549 ( 0.3923, 0.1044) 0.20196 ( 0.0565, 0.4323) 0.3805 ( 0.5764, 0.1370)

0.4260 (0.1894, 0.6113) 0.1718 ( 0.4067, 0.0873) 0.18372 ( 0.0752, 0.4169) 0.21329 ( 0.4418, 0.0448)

0.2624 (0.0068, 0.4823) 0.1663 ( 0.4021, 0.0928) 0.20926 ( 0.0489, 0.4384) 0.24240 ( 0.4659, 0.0143)

0.1010 ( 0.1578, 0.3453) 0.0666 ( 0.3147, 0.1911) 0.28808 (0.0344, 0.5032) 0.17892 ( 0.4128, 0.0801)

Pearson correlation coefficients (CW/CL, SC, GW) and point biserial correlation (TRAN) given with 95% confidence interval. CW/CL, crown width/crown length ratio; GW, gingival width; SC, gingival scallop.

Table 3. Correlation between thickness of the gingiva and alveolar bone plate

A1 A2 A3

0.4757 (0.2485, 0.6487) 0.4212 (0.1838, 0.6076) 0.3880 (0.1450, 0.5819)

0.4315 (0.1958, 0.6154) 0.3652 (0.1196, 0.5645) 0.2465 ( 0.0100, 0.4693)

0.5087 (0.2886, 0.6730) 0.3225 (0.0720, 0.5308) 0.2110 ( 0.0475, 0.4396)

0.5830 (0.3820, 0.7267) 0.3603 (0.1140, 0.5606) 0.3118 (0.0602, 0.5223)

0.5644 (0.3583, 0.7134) 0.3850 (0.1420, 0.5798) 0.3638 (0.1180, 0.56340)

0.2546 ( 0.0014 0.4760) 0.1438 ( 0.1155, 0.3826) 0.1027 ( 0.1561, 0.3468)

Gingival and alveolar bone dimensions Table 4. Multiple regression model for the association between CW/CL, probe transparency, gingival width, gingival scallop and the thickness of the attached gingiva at the CEJ level (G3) (R2 = 0.3024) DF Intercept CW/CL TRAN GW SC

1 1 1 1 1

Parameter estimate 0.65866 1.79769 0.10522 0.08887 0.05263

t-value

p-value

0.61178 0.53585 0.08489 0.04013 0.05625

1.08 3.35 1.24 2.21 0.94

0.2863 0.0014 0.2204 0.0310 0.3535

CEJ, cementoenamel junction; CW/CL, crown width/crown length ratio; DF, degree of freedom; GW, gingival width; SC, gingival scallop. Values in bold are statistically significant (p< 0.05). Table 5. Multiple regression model for the association between CW/CL, probe transparency, gingival width, gingival scallop and the thickness of the buccal alveolar bone crest (A1) (R2 = 0.2202) DF Intercept CW/CL TRAN GW SC

1 1 1 1 1

Parameter estimate 0.57709 1.14439 0.03169 0.04098 0.03137

t-value

p-value

0.42352 0.37096 0.05877 0.02778 0.03894

1.36 3.08 0.54 1.48 0.81

0.1786 0.0032 0.5919 0.1459 0.4239

CW/CL, crown width/crown length ratio; DF, degree of freedom; GW, gingival width; SC, gingival scallop. Values in bold are statistically significant (p< 0.05).

definition of how thick a thick biotype should be compared to a thin one. One of the reasons may be seen in the fact that thickness of the gingiva has been assessed at different vertical levels (Olsson et al. 1993, Fu et al. 2010, Kan et al. 2010). The present data clearly show that gingival thicknesses measured at different levels (G1â&#x20AC;&#x201C;G6) differ from each other and notably increase from the level of the margin (G1) towards the level directly coronal to the bone crest (G5), whereas the attached gingiva over the bone crest (G6) was remarkably smaller. Therefore, the results of the cited studies were hardly comparable. Another reason might be the method of measuring gingival thickness. Manual assessment using a calliper after tooth extraction (Fu et al. 2010, Kan et al. 2010), a syringe with endodontic depth marker (Olsson et al. 1993) or cone beam radiographs without reference objects (Fu et al. 2010) might have limitations of their accuracy. In contrast to these reports, in this study, a modified radiographic technique described by Alpiste-Illueca (2004) was used. To obtain maximally precise measurements, quality control was established regarding the (i) exactly parallel orientation of the film towards the long axis of the tooth, (ii) exclusion of cases with insufficient

contrast of anatomic landmarks, (iii) reproducibility of the real tissue dimensions using a standardized lead plate allowing the calculation of the magnification effect and (iv) recordings of the thickness values to 0.1 mm using Photoshop software. Nevertheless, there are two potential limitations. First, measurements at the base of the free gingiva (G2) comprise the sulcus width, which might be considered as bias. However, as all participants did not have any signs of inflammatory gingival diseases and no pathologic attachment loss that could be associated with remarkably increased gingival sulcus, this bias seems to be negligible. Second, despite the exact parallel positioning, a strictly tangential projection over the entire length of the plate is difficult. Nonetheless, the potential error due to the deviation of the projected thickness from the real thickness of the lead plate was not more than 0.1 mm in average. This amount of bias can be regarded as minimal and supposed to be no larger than errors occurring in previously reported techniques such as direct measurements with invasive techniques (Olsson et al. 1993, Kan et al. 2010). One of the main results of this study was the positive correlation of

ÂŠ 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

1137

gingival thickness at all levels with the thickness of the buccal alveolar bone. This confirms and extends the results of Fu et al. (2010) who recognized a moderate correlation between gingiva and bone thickness on cadaver teeth. In this case, this relationship was differentiated so far that the correlations between the gingival thickness at the supracrestal attachment (G3â&#x20AC;&#x201C;G5) and the alveolar bone crest (A1) were stronger than correlations between the thickness of other parts of the gingiva (G1, G2, G6) and more apical parts of the alveolar bone plate (A2, A3). Furthermore, data from the present investigation demonstrates a positive correlation of the crown form (CW/CL) and the width of the keratinized gingiva (GW) with all thickness parameters. Thereby, the strongest correlation was revealed between CW/CL and the gingiva over the CEJ (G3) as well as the alveolar crest (A1). Contrary, the height of the gingival scallop had only a weak negative correlation with all parameters. The latter was in accordance with the observations of De Rouck et al. (2009) that thin or thick gingival biotypes are not necessarily associated with high or low scallop. Therefore, gingival scallop does not seem to be an appropriate indicator for gingival biotypes. One of the most frequently used methods for identifying gingival thickness was the transparency of the periodontal probe (De Rouck et al. 2009, Kan et al. 2010, Fu et al. 2010, Cook et al. 2011). However, probe transparency has not always been correlated with measurements of the gingival thickness (De Rouck et al. 2009, Eghbali et al. 2009, Cook et al. 2011). In those who did, visibility of the probe was related to the thickness of the gingiva, either 2 mm from the gingival margin (Kan et al. 2010) or 2 mm apically to the alveolar crest (Fu et al. 2010). Although Kan et al. (2010) proposed this instrument as appropriate for the differentiation of thick and thin gingival biotypes, Fu et al. (2010) and Eghbali et al. (2009) reported it not as useful. In this study, the results differed according to the reference point of the gingiva. Due to the mild negative correlation (r < 0.3), the findings suggest that probe transparency is not the best

1138

Stein et al.

method for evaluation of the thickness of the gingiva at the supracrestal attachment (G3–G5), the attached gingiva (G6) or alveolar bone thickness (A1–A3). Only for the free gingiva (G2), the negative correlation was stronger (r < 0.4). As G2 was on average located 0.5– 1 mm apically from the gingival margin, transparency of the probe may be a helpful indicator for the thickness of the free gingiva when restorations are placed slightly (0.5 mm) subgingival with the need to conceal alloy margins. However, periodontal probing does not appear to be recommended in evaluating the thickness of the gingiva located at the CEJ (G3) or more apically (G4– G6), which might, for example, be of interest for coverage of implant abutments, teeth treated with endodontic posts having the risk of corrosion and, therefore, impairing gingival colour or planning of implant insertion. In addition to the correlation analyses, the association of CW/CL, GW, SC and TRAN with G3 and A1 has been tested in two multiple regression models. G3 and A1 were chosen as target parameters because gingival thickness at the CEJ (G3) corresponds well to the reference points used in most other studies (Olsson et al. 1993, Kan et al. 2010), whereas the thickness of the alveolar crest (A1) has a particular meaning for the maintenance of the gingival stability and might be of strategic value before restorative and (immediate) implant therapies. The analyses revealed that CW/CL and GW were significant predictors for G3 value and only CW/CL was significantly associated with A1 value. This supports the hypothesis that teeth with squared crown forms are more likely to have a thicker marginal gingiva and alveolar bone crest than teeth with tapered crown forms. This is in disagreement to the data of Olsson et al. (1993) who did not find a significant association between CW/ CL and gingival thickness. The above-mentioned differences in the methods of measuring gingival thickness might possibly explain the discrepancies. Another reason could be seen in the fact that Olsson et al. (1993) only compared a small number of subjects with the highest (N = 10) and those with the lowest

(N = 10) CW/CL values, whereas in this study all participants (N = 60) had been included in the analyses. Conclusions

The data of the present study show that a clear distinction between a “thin” and a “thick” gingival biotype is very difficult. Differences in the tissue thickness at different apicocoronal levels and lack of consensus as to which anatomic landmark should be used as reference point might explain the inconsistency of previous studies in the definition of gingiva biotypes. Our findings suggest that crown form (CW/CL) and width of keratinized gingiva (GW) are helpful indicators for the thickness of the gingiva over the cementoenamel junction (G3), whereas CW/ CL also represents a predictor for the thickness of the buccal alveolar crest. All thickness parameters of the gingiva (G1–G6) were positively related to alveolar crest thickness. Transparency of the periodontal probe was negatively correlated with thickness of free gingiva (G2), but had only minimal prognostic value for the thickness of the attached gingiva or alveolar crest. Finally, the data indicated that the height of the gingival scallop cannot be recommended as an indicator for tissue thickness. However, due to the limited sample size the results should not be generalized. Therefore, it is recommended in future studies to verify the predictive potential of crown form and gingival width (GW) on tissue thickness considering the thickness parameters suggested here. References Alpiste-Illueca, F. (2004) Dimensions of the dentogingival unit in maxillary anterior teeth: a new exploration technique (parallel profile radiograph). International Journal of Periodontics and Restorative Dentistry 24, 386–396. Baldi, C., Pini-Prato, G., Pagliaro, U., Nieri, M., Saletta, D., Muzzi, L. & Cortellini, P. (1999) Coronally advanced flap procedure for root coverage. Is flap thickness a relevant predictor to achieve root coverage? A 19-case series. Journal of Periodontology 70, 1077–1084. Barriviera, M., Duarte, W. R., Januario, A. L., Faber, J. & Bezerra, A. C. (2009) A new method to assess and measure palatal masticatory mucosa by cone-beam computerized tomography. Journal of Clinical Periodontology 36, 564–568. Cook, D. R., Mealey, B. L., Verrett, R. G., Mills, M. P., Noujeim, M. E., Lasho, D. J. & Cronin,

R. J., Jr (2011) Relationship between clinical periodontal biotype and labial plate thickness: an in vivo study. International Journal of Periodontics and Restorative Dentistry 31, 345–354. De Rouck, T., Eghbali, R., Collys, K., De Bruyn, H. & Cosyn, J. (2009) The gingival biotype revisited: transparency of the periodontal probe through the gingival margin as a method to discriminate thin from thick gingiva. Journal of Clinical Periodontology 36, 428–433. Eghbali, A., De Rouck, T., De Bruyn, H. & Cosyn, J. (2009) The gingival biotype assessed by experienced and inexperienced clinicians. Journal of Clinical Periodontology 36, 958–963. Ericsson, I. & Lindhe, J. (1984) Recession in sites with inadequate width of the keratinized gingiva. An experimental study in the dog. Journal of Clinical Periodontology 11, 95–103. Evans, C. D. & Chen, S. T. (2008) Esthetic outcomes of immediate implant placements. Clinical Oral Implants Research 19, 73–80. Fu, J. H., Yeh, C. Y., Chan, H. L., Tatarakis, N., Leong, D. J. & Wang, H. L. (2010) Tissue biotype and its relation to the underlying bone morphology. Journal of Periodontology 81, 569–574. Greenberg, J., Laster, L. & Listgarten, M. A. (1976) Transgingival probing as a potential estimator of alveolar bone level. Journal of Periodontology 47, 514–517. Gwet, K. L. (2008) Intrarater reliability. In: Hoboken, N. J. (ed). Wiley Encyclopedia of Clinical Trials, pp. 1–14, Gaithersburg: John Wiley & Sons. Hwang, D. & Wang, H. L. (2006) Flap thickness as a predictor of root coverage: A systematic review. Journal of Periodontology 77, 1625–1634. Kan, J. Y., Morimoto, T., Rungcharassaeng, K., Roe, P. & Smith, D. H. (2010) Gingival biotype assessment in the esthetic zone: visual versus direct measurement. International Journal of Periodontics and Restorative Dentistry 30, 237–243. Kao, R. T., Fagan, M. C. & Conte, G. J. (2008) Thick vs. thin gingival biotypes: a key determinant in treatment planning for dental implants. Journal of the California Dental Association 36, 193–198. Kois, J.C. (2004) Predictable single-tooth periimplant esthetics: five diagnostic keys. Compendium of Continuing Education in Dentistry 25, 895–896, 898, 900. Larato, D. C. (1970) Alveolar plate fenestrations and dehiscences of the human skull. Oral Surgery, Oral Medicine and Oral Pathology 29, 816–819. Light, R. J. (1971) Measures of response agreement for qualitative data: Some generalizations and alternatives. Psychological Bulletin 76, 365–377. M€ uller, H. P., Schaller, N., Eger, T. & Heinecke, A. (2000) Thickness of masticatory mucosa. Journal of Clinical Periodontology 27, 431–436. Olsson, M. & Lindhe, J. (1991) Periodontal characteristics in individuals with varying form of the upper central incisors. Journal of Clinical Periodontology 18, 78–82. Olsson, M., Lindhe, J. & Marinello, C. P. (1993) On the relationship between crown form and clinical features of the gingiva in adolescents. Journal of Clinical Periodontology 20, 570–577. Pontoriero, R. & Carnevale, G. (2001) Surgical crown lengthening: a 12-month clinical wound healing study. Journal of Periodontology 72, 841–848. Romeo, E., Lops, D., Rossi, A., Storelli, S., Rozza, R. & Chiapasco, M. (2008) Surgical and prosthetic management of interproximal region with single-implant restorations: 1-year prospective study. Journal of Periodontology 79, 1048–1055.

Gingival and alveolar bone dimensions Seibert, J. L. & Lindhe, J. (1989) Esthetics and periodontal therapy, Chapter 19. In: Lindhe, J. (ed). Textbook of Clinical Periodontology, 2nd edition, pp. 477–514, Copenhagen: Munksgaard. Vandana, K. L. & Savitha, B. (2005) Thickness of gingiva in association with age, gender and dental arch location. Journal of Clinical Periodontology 32, 828–830. Weisgold, A. S. (1977) Contours of the full crown restoration. Alpha Omegan 70, 77–89.

Clinical Relevance

Scientific rationale for the study: Gingival biotypes are supposed to influence the outcome of reconstructive therapies. Different methods have been reported to identify gingival thickness, however, without consistent results. Principal findings: Using multiple regression models, crown form

1139

Wennstr€ om, J. L., Lindhe, J., Sinclair, F. & Thilander, B. (1987) Some periodontal tissue reactions to orthodontic tooth movement in monkeys. Journal of Clinical Periodontology 14, 121–129.

Address: Jamal M. Stein Department of Operative Dentistry, Periodontology and Preventive Dentistry, University Hospital Aachen (RWTH), Pauwelsstrasse 30, D-52074 Aachen, Germany E-mail: jstein@ukaachen.de

revealed to be a predictor for the thickness of the gingiva at the cementoenamel junction and the alveolar bone crest. Transparency of the periodontal probe had a limited negative correlation with free gingiva thickness, but no significant influence on other tissue dimensions. Practical implications: Long tapered crown forms may indicate the pres-

ence of a thinner gingiva at the cementoenamel junction and a thinner buccal alveolar bone at the crest level. Periodontal probing cannot be recommended for evaluation of the gingival thickness at the supracrestal attachment.