Variance Components of Gingival Thickness

Objectives: Distinct periodontal phenotypes have been identified by cluster analysis which is an explorative method with very low external validity. The aim of the present study was to investigate variance components of facial gingival thickness in young adults with mild gingivitis using random intercept multilevel models. Methods: 33 females, 18-23 yr of age, with mild plaque-induced gingivitis participated. Gingival thickness (GTH) was measured facially (midbuccally) at every tooth present with an ultrasonic measuring device (Krupp SDM, Austenal Inc., Germany) to the next 0.1 mm with a lowest measurement of 0.5 mm. In addition, periodontal probing depth (PPD) and clinical attachment level were measured with a pressure-controlled probe (ClickProbe 1395, KerrHawe, Switzerland) to the next mm. Gingival bleeding index (GBI) after probing was assessed on a 0-2 scale, where 1 was slight, and 2 profuse bleeding after probing. Presence of calculus, and the Silness-Loe plaque index (PI) were also recorded. Multi-level models of GTH were constructed using special software (MLwiN 2.0). Results: A 2-level (subject, tooth) variance component model for GTH without any explanatory variable revealed an intercept (mean) of 0.93 ± 0.02 mm. Subject variation of GTH amounted to only 4.2% of the total variance. Addition of tooth- and subject-specific clinical variables revealed, after adjusting for tooth type, association with PPD (0.067 ± 0.025 mm), and considerable association with average GBI (-0.395 ± 0.149) and PI (0.125 ± 0.048). Variation at the tooth level was dramatically reduced whereas subject variation amounted to 5.2%. Conclusions: Multilevel modeling revealed that GTH is mainly associated with tooth-related variables. Subject variability related to periodontal phenotype may add to the total variance, however, to a very low extent. Supported by KerrHawe, Switzerland, and Kuwait University Research Grant DS0202.