DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
“Periodontal Biotype: Gingival Thickness as it Relates to Probe Visibility and Buccal Plate Thickness” Natalie A. Frost D.D.S., M.S.*, Brian L. Mealey D.D.S., M.S.†, Archie A. Jones D.D.S., M.B.A.†, Guy Huynh-Ba D.D.S., M.S.† *
Private practice, Omaha, NE.
†
Department of Periodontics, University of Texas Health Science Center at San Antonio, San Antonio, TX.
Background: Probe visibility is the clinical gold standard to discriminate thick from thin biotype, but is prone to subjective interpretation. The primary objective was to determine at what objective gingival thickness the probe becomes invisible through the tissue. A secondary objective compared mean buccal plate thickness between thick and thin biotypes as determined by probe visibility. Methods: Maxillary anterior teeth (n=306) were studied in 56 human patients. Biotype was determined by probe visibility through the tissue. Gingival thickness was measured via transgingival sounding. Buccal plate thickness was measured (n=66 teeth) by cone beam computed tomography. For the primary objective, the gingival thickness that best corresponded with probe invisibility was selected using the Receiver Operating Characteristic and Area Under the Curve (AUC) with the highest combination of sensitivity and specificity. For the secondary objective, mean buccal plate thickness was compared between sites where the probe was visible and when it was not (Student’s t-test, α= 0.05). Results: The gingival thickness that most closely corresponded with probe invisibility was >0.8mm (AUC 0.666, sensitivity 67.7%, specificity 65.4%). When the probe was visible, mean gingival thickness was 0.17mm less (p0.8mm was associated with 100% root coverage with a coronally advanced flap.2 Less post-treatment recession was observed after guided tissue regeneration procedures with tissue thickness >1mm as compared to sites 1.1mm and complete root coverage following connective tissue grafting and guided tissue regeneration procedures.4 A thicker biotype has been correlated with greater tissue rebound following surgical crown lengthening.5,6 Thin periodontal biotypes are associated with slightly greater buccal marginal tissue recession around implants as compared to thick biotypes.7, 8 Spray et al. documented that as buccal bone thickness approached 1.8 to 2.0mm, bone loss decreased 1
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
significantly and evidence of bone gain after implant placement was seen.9 More recently, studies have documented greater mean bone loss occurring around implants in sites with thin as compared to thick overlying mucosa.10,11 Many methods exist, by which practitioners distinguish thick from thin biotypes. The simplest evaluation is by visual assessment of the patient without periodontal probing. 12 However, nearly half of the thin-scalloped patients are misclassified, which supports that simple visual assessment is of limited value to distinguish thick from thin biotypes. A recent study suggested that if the sum of the 5 papillae lengths comprised between the maxillary canines is greater than 24mm, the patient may be considered as having as thin biotype.13While interesting, this data remains to be confirmed and has yet to gain clinical acceptance. Likely, the most commonly used method for determining biotype is placement of a probe within the gingival sulcus and evaluating for probe visibility. If the probe can be seen through the gingival tissue, the biotype is classified as thin. Conversely, if the probe cannot be seen through the gingival tissue, the biotype is classified as thick. Although subjective, this method has been documented to be highly reproducible with 85% agreement between duplicate assessments.14 Moreover, multiple clinical studies have used this method to determine the periodontal biotype and , thus, will be considered as a gold standard in the present manuscript.8, 14-16
Although not commonly used, the most objective and repeatable assessment of gingival thickness is likely by direct measurement with calipers or transgingival sounding. Kan et al. compared probe visibility assessment of periodontal biotype, thick versus thin, to direct gingival thickness measurements.15 In that study, direct mid-facial gingival thickness measurements were taken 2 mm apical to the free gingival margin. After comparing probe visibility to direct gingival thickness measurements, the biotype was always thin (100%) when gingival thickness was 0.6mm and always thick (100%) when gingival thickness was >1.2mm. From these findings Kan et al. proposed a 1.0mm threshold to differentiate thick from thin biotypes. Although this 1mm threshold is commonly reported in the literature, the value is somewhat arbitrary. The question remains, if 1mm represents the best threshold for diagnostic purposes.15 Confusion may arise when reading the literature because, on one hand, the term of gingival biotype has been used interchangeably with tissue thickness 6- 8, 10 or height of keratinized tissue7 and was considered as a site specific factor. On the other hand, other studies have considered gingival biotype as a patient specific factor.12-14 A recent study by Cook and coworkers16, suggested that the periodontal biotype (thick or thin) at a patient level does translate into specific site level factors including probe visibility, papilla height, width of keratinized tissue and distance from the cemento-enamel junction to the alveolar crest. Considering the methods reviewed, it is clear that no universally accepted criterion to distinguish thick from thin gingival thickness or biotypes exists. The primary objective of this study was to determine the objective gingival thickness measurement above which the probe becomes invisible when placed within the gingival sulcus. This thickness could serve as the recommended measurement threshold to objectively discriminate thick from thin periodontal biotypes, which has useful applications for stratification of sites into treatment groups for clinical research. Therefore, in the present manuscript, probe visibility was considered as the “clinical” gold standard to determine gingival biotype at the site level. Probe visibility and biotype were used interchangeably within the manuscript and the diagnostic value of gingival thickness was 2
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
determined relative to probe visibility at the site level. A secondary objective compared mean buccal plate thickness between sites where the probe was visible (i.e. thin biotype) and those where it was not visible (i.e. thick biotype). The main hypothesis of this study is that an objective gingival thickness threshold exists above which, the periodontal probe becomes invisible through the tissue. A secondary hypothesis is that smaller measurements of buccal plate thickness will be associated with probe visibility, or a thin biotype.
MATERIALS AND METHODS Patient Enrollment The study protocol was approved by The University of Texas Health Science Center at San Antonio (UTHSCSA) Institutional Review Board for human patients (protocol number: HSC20120112H) and was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2000. Fifty-six patients, 33 females and 23 males, were recruited between April 2012 and August 2013 from a population reporting for comprehensive treatment at the UTHSCSA Dental School. Patient age ranged from 20-78 with an average 53 years. In this study, the tooth was utilized as the unit of measurement (n=306). Patients were enrolled sequentially by convenience sample until data for at least 300 teeth was acquired. All patients were treated either by dental students or periodontics residents as the primary provider. Patients were selected if they met all of the following criteria: 1) at least one tooth present in the anterior maxilla: including first premolars and incisors, 2) healthy (inflammation free) gingiva, 3) at least one tooth site to be anesthetized as part of a procedure not related to the study, and 4) ≥18 years of age. Patients were excluded from the study if they presented with any of the following conditions: 1) pregnancy, 2) facial probing pocket depth (PPD > 4mm) in the region of interest 3) history of orthodontic therapy, 4) previous surgery in the region of interest, 5) grossly malaligned teeth, 6) past or present use of medication known to increase risk for gingival enlargement (phenytoin, cyclosporine, nifedipine, amlodipine), 7) gingival enlargement, or 8) restorations involving the gingival margin in the region of interest. Probe Visibility Assessment Once enrollment was complete and written consent obtained, clinical data was gathered. Prior to administration of local anesthetic, a probe‡ was inserted into the mid-facial gingival sulcus and a digital photograph was taken in standardized fashion: 1) only one camera§ was used with the same settings, 2) prior to image acquisition, the tooth and gingiva were dried, 3) no light from the dental unit was directed on the patient’s mouth, 4) the tooth of interest was centered and surrounded on each side by half of the neighboring tooth, 5) photos were obtained with a background black photo contrast medium‖, and 6) all photos were taken by one investigator. Using the digital images, three examiners (NF, GH, BM) determined independently whether the probe was visible (Yes or No) through the gingival tissue when placed within the gingival sulcus. A tooth was categorized as thin biotype if the probe could be seen (Y) through the soft tissue. (Fig. 1) A tooth was categorized as thick biotype if the probe could not be seen (N) through the soft tissue (Fig. 1) Prior to independent assignment of probe visibility, two separate calibration sessions were performed to establish baseline agreement between examiners.
3
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
For final probe visibility assessment, all images were projected digitally¶ onto a screen# where they measured approximately 3-4 ft x 3-4 ft. No overhead lighting was utilized. All three examiners were present at the same time within the same room for probe visibility categorization. Each examiner was blinded to the assessments of the other examiners. Direct Measurements of Gingival Thickness Measurements of gingival thickness were taken after delivery of topical and/or local anesthetic which was deposited at the depth of the vestibule to avoid any inadvertent increase in gingival thickness at the sites to be measured. No teeth were anesthetized for the sole purpose of this research study, but were anesthetized as part of the patient’s comprehensive periodontal or restorative therapies. Gingival thickness measurements were taken at the mid-facial location that corresponded to the same location for probe visibility assessment. On the mid-facial of aspect of each tooth, measurements were obtained 1mm coronal to the gingival probing pocket depth (PPD) within free and keratinized gingiva, and by the same clinical investigator. An endodontic file** was inserted transgingivally until contact with the tooth structure was felt. Flowable composite††was applied around the file, on the external surface of the gingiva, and cured‡‡ to create a fixed reference point (Fig 2). After curing, the endodontic file was removed and a digital photograph was obtained with a measurement gauge§§in the captured field as a reference for software calibration (Fig 2). Using photographs containing the file and measurement gauge, the transgingival portion of each file was measured using digital measuring software‖‖. A total of 306 gingival thickness measurements were determined. A random list of 60 digital images containing both the file and measurement gauge, was generated by a computer for repeat measurements. These were completed by the initial investigator (NF) who was blinded to the identity of the chosen images and were completed 10 days after the first measurements. The digital images from the random 60 files were also measured twice by a second investigator (GH), for a total of four separate measurements to assess the reliability of the gingival thickness measurements using the Pearson Correlation Coefficient and percentage of absolute differences. Buccal Plate Thickness Measurements- CBCT After patient enrollment was closed, a retrospective chart review of enrolled patients was completed to access previously acquired CBCT scans. These scans were made as part of the patient’s comprehensive dental treatment and were not acquired for the purpose of this study. All CBCT scans were acquired with the same scanning device¶¶. Measurements of buccal plate thickness were made using digital three-dimensional imaging software## to the nearest 0.01mm. A total of 66 tooth sites were available for CBCT measurements of buccal plate thickness. Buccal plate thickness was measured at the mid-facial aspect of the identical tooth site within the same subject from which assessment of probe visibility, thick versus thin biotype, was determined. Prior to measurement, all scans were aligned with a standardized protocol. This was done with the following three-dimensional guidelines: 1) along the bucco-lingual axis, the sagittal plane was placed in the middle of the selected tooth, 2) along the mesio-distal axis, the frontal plane was placed in the center of the selected tooth, and 3) along the apico-coronal axis, the axial plane was perpendicular to the long axis of the selected tooth at the level of the 4
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
cementoenamel junction (Fig. 3). Buccal plate thickness was measured from the inner aspect of the buccal plate, identified by the periodontal ligament (PDL) space to the external surface of the buccal plate perpendicular to the long axis of the PDL space (Fig. 3). All measurements were made 1mm apical to the alveolar crest by a single examiner (NF) and repeated in duplicate to determine the Pearson Correlation Coefficient between initial and repeat measurements as well as percentage of absolute differences. Statistical Analyses Prior to initiation of the study, a power analysis was completed. Based on previous studies, it was anticipated that 60% of teeth would be classified as having thick biotype, with the remaining % having thin biotype.15,16 Using these criteria, power analyses were performed***and a proposed sample size of 300 teeth was determined. For each tooth, measurements of gingival thickness were matched with the consensus value for probe visibility based on agreement among at least two out of three examiners. When the examiner consensus was that the probe was not visible, the tooth was classified as having thick biotype and predicted to have a gingival thickness greater than a threshold value. When the consensus was that the probe was visible, the tooth was classified as having thin biotype and predicted to have a gingival thickness less than or equal to a threshold value. Teeth with gingival thickness above the threshold value were considered diagnostic positives; otherwise, the teeth were considered diagnostic negatives. The threshold gingival thickness for which the diagnostic classification of thick or thin biotype maximized the area under the Receiver Operating Characteristic curve (ROC) and which achieved the combination of highest sensitivity and specificity for the final sample of 306 teeth was determined. For the secondary study objective, the mean buccal plate thickness was compared between thick and thin biotypes, based on probe visibility, using a Student’s t-test at a significance level of α= 0.05.
RESULTS Biotype Prevalence Based on the consensus among the three examiners for probe visibility, the overall prevalence of a thick biotype was 83% or 254/306 tooth sites. The prevalence of a thin biotype was 17% or 52/306 tooth sites. Relative to tooth type, the prevalence of a thin biotype was found in 11% premolars (7/62), 4% canines (20/83), 23% laterals (20/86), and 7% centrals (5/75). Accuracy of Digital Gingival Thickness Measurements For the random sample of 60 digital images of files selected for repeat measurements of gingival thickness made by two independent examiners, the Pearson Correlation Coefficient was r=0.964, 95% CI (0.940,0.978). Comparing percentage difference between the two repeat measurements, 71% differed by
Journal of Periodontology; Copyright 2010
“Periodontal Biotype: Gingival Thickness as it Relates to Probe Visibility and Buccal Plate Thickness” Natalie A. Frost D.D.S., M.S.*, Brian L. Mealey D.D.S., M.S.†, Archie A. Jones D.D.S., M.B.A.†, Guy Huynh-Ba D.D.S., M.S.† *
Private practice, Omaha, NE.
†
Department of Periodontics, University of Texas Health Science Center at San Antonio, San Antonio, TX.
Background: Probe visibility is the clinical gold standard to discriminate thick from thin biotype, but is prone to subjective interpretation. The primary objective was to determine at what objective gingival thickness the probe becomes invisible through the tissue. A secondary objective compared mean buccal plate thickness between thick and thin biotypes as determined by probe visibility. Methods: Maxillary anterior teeth (n=306) were studied in 56 human patients. Biotype was determined by probe visibility through the tissue. Gingival thickness was measured via transgingival sounding. Buccal plate thickness was measured (n=66 teeth) by cone beam computed tomography. For the primary objective, the gingival thickness that best corresponded with probe invisibility was selected using the Receiver Operating Characteristic and Area Under the Curve (AUC) with the highest combination of sensitivity and specificity. For the secondary objective, mean buccal plate thickness was compared between sites where the probe was visible and when it was not (Student’s t-test, α= 0.05). Results: The gingival thickness that most closely corresponded with probe invisibility was >0.8mm (AUC 0.666, sensitivity 67.7%, specificity 65.4%). When the probe was visible, mean gingival thickness was 0.17mm less (p0.8mm was associated with 100% root coverage with a coronally advanced flap.2 Less post-treatment recession was observed after guided tissue regeneration procedures with tissue thickness >1mm as compared to sites 1.1mm and complete root coverage following connective tissue grafting and guided tissue regeneration procedures.4 A thicker biotype has been correlated with greater tissue rebound following surgical crown lengthening.5,6 Thin periodontal biotypes are associated with slightly greater buccal marginal tissue recession around implants as compared to thick biotypes.7, 8 Spray et al. documented that as buccal bone thickness approached 1.8 to 2.0mm, bone loss decreased 1
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
significantly and evidence of bone gain after implant placement was seen.9 More recently, studies have documented greater mean bone loss occurring around implants in sites with thin as compared to thick overlying mucosa.10,11 Many methods exist, by which practitioners distinguish thick from thin biotypes. The simplest evaluation is by visual assessment of the patient without periodontal probing. 12 However, nearly half of the thin-scalloped patients are misclassified, which supports that simple visual assessment is of limited value to distinguish thick from thin biotypes. A recent study suggested that if the sum of the 5 papillae lengths comprised between the maxillary canines is greater than 24mm, the patient may be considered as having as thin biotype.13While interesting, this data remains to be confirmed and has yet to gain clinical acceptance. Likely, the most commonly used method for determining biotype is placement of a probe within the gingival sulcus and evaluating for probe visibility. If the probe can be seen through the gingival tissue, the biotype is classified as thin. Conversely, if the probe cannot be seen through the gingival tissue, the biotype is classified as thick. Although subjective, this method has been documented to be highly reproducible with 85% agreement between duplicate assessments.14 Moreover, multiple clinical studies have used this method to determine the periodontal biotype and , thus, will be considered as a gold standard in the present manuscript.8, 14-16
Although not commonly used, the most objective and repeatable assessment of gingival thickness is likely by direct measurement with calipers or transgingival sounding. Kan et al. compared probe visibility assessment of periodontal biotype, thick versus thin, to direct gingival thickness measurements.15 In that study, direct mid-facial gingival thickness measurements were taken 2 mm apical to the free gingival margin. After comparing probe visibility to direct gingival thickness measurements, the biotype was always thin (100%) when gingival thickness was 0.6mm and always thick (100%) when gingival thickness was >1.2mm. From these findings Kan et al. proposed a 1.0mm threshold to differentiate thick from thin biotypes. Although this 1mm threshold is commonly reported in the literature, the value is somewhat arbitrary. The question remains, if 1mm represents the best threshold for diagnostic purposes.15 Confusion may arise when reading the literature because, on one hand, the term of gingival biotype has been used interchangeably with tissue thickness 6- 8, 10 or height of keratinized tissue7 and was considered as a site specific factor. On the other hand, other studies have considered gingival biotype as a patient specific factor.12-14 A recent study by Cook and coworkers16, suggested that the periodontal biotype (thick or thin) at a patient level does translate into specific site level factors including probe visibility, papilla height, width of keratinized tissue and distance from the cemento-enamel junction to the alveolar crest. Considering the methods reviewed, it is clear that no universally accepted criterion to distinguish thick from thin gingival thickness or biotypes exists. The primary objective of this study was to determine the objective gingival thickness measurement above which the probe becomes invisible when placed within the gingival sulcus. This thickness could serve as the recommended measurement threshold to objectively discriminate thick from thin periodontal biotypes, which has useful applications for stratification of sites into treatment groups for clinical research. Therefore, in the present manuscript, probe visibility was considered as the “clinical” gold standard to determine gingival biotype at the site level. Probe visibility and biotype were used interchangeably within the manuscript and the diagnostic value of gingival thickness was 2
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
determined relative to probe visibility at the site level. A secondary objective compared mean buccal plate thickness between sites where the probe was visible (i.e. thin biotype) and those where it was not visible (i.e. thick biotype). The main hypothesis of this study is that an objective gingival thickness threshold exists above which, the periodontal probe becomes invisible through the tissue. A secondary hypothesis is that smaller measurements of buccal plate thickness will be associated with probe visibility, or a thin biotype.
MATERIALS AND METHODS Patient Enrollment The study protocol was approved by The University of Texas Health Science Center at San Antonio (UTHSCSA) Institutional Review Board for human patients (protocol number: HSC20120112H) and was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2000. Fifty-six patients, 33 females and 23 males, were recruited between April 2012 and August 2013 from a population reporting for comprehensive treatment at the UTHSCSA Dental School. Patient age ranged from 20-78 with an average 53 years. In this study, the tooth was utilized as the unit of measurement (n=306). Patients were enrolled sequentially by convenience sample until data for at least 300 teeth was acquired. All patients were treated either by dental students or periodontics residents as the primary provider. Patients were selected if they met all of the following criteria: 1) at least one tooth present in the anterior maxilla: including first premolars and incisors, 2) healthy (inflammation free) gingiva, 3) at least one tooth site to be anesthetized as part of a procedure not related to the study, and 4) ≥18 years of age. Patients were excluded from the study if they presented with any of the following conditions: 1) pregnancy, 2) facial probing pocket depth (PPD > 4mm) in the region of interest 3) history of orthodontic therapy, 4) previous surgery in the region of interest, 5) grossly malaligned teeth, 6) past or present use of medication known to increase risk for gingival enlargement (phenytoin, cyclosporine, nifedipine, amlodipine), 7) gingival enlargement, or 8) restorations involving the gingival margin in the region of interest. Probe Visibility Assessment Once enrollment was complete and written consent obtained, clinical data was gathered. Prior to administration of local anesthetic, a probe‡ was inserted into the mid-facial gingival sulcus and a digital photograph was taken in standardized fashion: 1) only one camera§ was used with the same settings, 2) prior to image acquisition, the tooth and gingiva were dried, 3) no light from the dental unit was directed on the patient’s mouth, 4) the tooth of interest was centered and surrounded on each side by half of the neighboring tooth, 5) photos were obtained with a background black photo contrast medium‖, and 6) all photos were taken by one investigator. Using the digital images, three examiners (NF, GH, BM) determined independently whether the probe was visible (Yes or No) through the gingival tissue when placed within the gingival sulcus. A tooth was categorized as thin biotype if the probe could be seen (Y) through the soft tissue. (Fig. 1) A tooth was categorized as thick biotype if the probe could not be seen (N) through the soft tissue (Fig. 1) Prior to independent assignment of probe visibility, two separate calibration sessions were performed to establish baseline agreement between examiners.
3
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
For final probe visibility assessment, all images were projected digitally¶ onto a screen# where they measured approximately 3-4 ft x 3-4 ft. No overhead lighting was utilized. All three examiners were present at the same time within the same room for probe visibility categorization. Each examiner was blinded to the assessments of the other examiners. Direct Measurements of Gingival Thickness Measurements of gingival thickness were taken after delivery of topical and/or local anesthetic which was deposited at the depth of the vestibule to avoid any inadvertent increase in gingival thickness at the sites to be measured. No teeth were anesthetized for the sole purpose of this research study, but were anesthetized as part of the patient’s comprehensive periodontal or restorative therapies. Gingival thickness measurements were taken at the mid-facial location that corresponded to the same location for probe visibility assessment. On the mid-facial of aspect of each tooth, measurements were obtained 1mm coronal to the gingival probing pocket depth (PPD) within free and keratinized gingiva, and by the same clinical investigator. An endodontic file** was inserted transgingivally until contact with the tooth structure was felt. Flowable composite††was applied around the file, on the external surface of the gingiva, and cured‡‡ to create a fixed reference point (Fig 2). After curing, the endodontic file was removed and a digital photograph was obtained with a measurement gauge§§in the captured field as a reference for software calibration (Fig 2). Using photographs containing the file and measurement gauge, the transgingival portion of each file was measured using digital measuring software‖‖. A total of 306 gingival thickness measurements were determined. A random list of 60 digital images containing both the file and measurement gauge, was generated by a computer for repeat measurements. These were completed by the initial investigator (NF) who was blinded to the identity of the chosen images and were completed 10 days after the first measurements. The digital images from the random 60 files were also measured twice by a second investigator (GH), for a total of four separate measurements to assess the reliability of the gingival thickness measurements using the Pearson Correlation Coefficient and percentage of absolute differences. Buccal Plate Thickness Measurements- CBCT After patient enrollment was closed, a retrospective chart review of enrolled patients was completed to access previously acquired CBCT scans. These scans were made as part of the patient’s comprehensive dental treatment and were not acquired for the purpose of this study. All CBCT scans were acquired with the same scanning device¶¶. Measurements of buccal plate thickness were made using digital three-dimensional imaging software## to the nearest 0.01mm. A total of 66 tooth sites were available for CBCT measurements of buccal plate thickness. Buccal plate thickness was measured at the mid-facial aspect of the identical tooth site within the same subject from which assessment of probe visibility, thick versus thin biotype, was determined. Prior to measurement, all scans were aligned with a standardized protocol. This was done with the following three-dimensional guidelines: 1) along the bucco-lingual axis, the sagittal plane was placed in the middle of the selected tooth, 2) along the mesio-distal axis, the frontal plane was placed in the center of the selected tooth, and 3) along the apico-coronal axis, the axial plane was perpendicular to the long axis of the selected tooth at the level of the 4
DOI: 10.1902/jop.2015.140394
Journal of Periodontology; Copyright 2010
cementoenamel junction (Fig. 3). Buccal plate thickness was measured from the inner aspect of the buccal plate, identified by the periodontal ligament (PDL) space to the external surface of the buccal plate perpendicular to the long axis of the PDL space (Fig. 3). All measurements were made 1mm apical to the alveolar crest by a single examiner (NF) and repeated in duplicate to determine the Pearson Correlation Coefficient between initial and repeat measurements as well as percentage of absolute differences. Statistical Analyses Prior to initiation of the study, a power analysis was completed. Based on previous studies, it was anticipated that 60% of teeth would be classified as having thick biotype, with the remaining % having thin biotype.15,16 Using these criteria, power analyses were performed***and a proposed sample size of 300 teeth was determined. For each tooth, measurements of gingival thickness were matched with the consensus value for probe visibility based on agreement among at least two out of three examiners. When the examiner consensus was that the probe was not visible, the tooth was classified as having thick biotype and predicted to have a gingival thickness greater than a threshold value. When the consensus was that the probe was visible, the tooth was classified as having thin biotype and predicted to have a gingival thickness less than or equal to a threshold value. Teeth with gingival thickness above the threshold value were considered diagnostic positives; otherwise, the teeth were considered diagnostic negatives. The threshold gingival thickness for which the diagnostic classification of thick or thin biotype maximized the area under the Receiver Operating Characteristic curve (ROC) and which achieved the combination of highest sensitivity and specificity for the final sample of 306 teeth was determined. For the secondary study objective, the mean buccal plate thickness was compared between thick and thin biotypes, based on probe visibility, using a Student’s t-test at a significance level of α= 0.05.
RESULTS Biotype Prevalence Based on the consensus among the three examiners for probe visibility, the overall prevalence of a thick biotype was 83% or 254/306 tooth sites. The prevalence of a thin biotype was 17% or 52/306 tooth sites. Relative to tooth type, the prevalence of a thin biotype was found in 11% premolars (7/62), 4% canines (20/83), 23% laterals (20/86), and 7% centrals (5/75). Accuracy of Digital Gingival Thickness Measurements For the random sample of 60 digital images of files selected for repeat measurements of gingival thickness made by two independent examiners, the Pearson Correlation Coefficient was r=0.964, 95% CI (0.940,0.978). Comparing percentage difference between the two repeat measurements, 71% differed by
