Validating Single Site Measurements of Probing Depth and Attachment Loss

Objective: The purpose of this study is to examine the relationship between full mouth measurements using all 6 sites per tooth and single site measurements, for probing depth (PD) and attachment loss (AL) and to assess the degree of relative bias that occurs in using each site per tooth. Methods: Using a data set of 644 adults screened for a treatment study of Type III or early Type IV chronic periodontitis, 130 adults (76 Male/54 Female), averaging 49.9 ±10.03 years, with no periodontal therapy (in previous 6 months) or prophylaxis (in previous 30 days), constituted the study sample. The NC probe (one mm increments) was used to measure PD, gingival recession (to obtain AL) and bleeding upon probing. Paired t-test and regression analyses were used to compare PD and AL using all 6 sites per tooth to each site. Results: Mean differences (paired t-test) showed that two of the sites underestimated the severity of both PD and AL, while the remaining four sites overestimated severity (range: PD: -32.40% to 16.01%; AL: -11.58% to 9.43%). For PD, all of these differences were statistically significant (p<0.001), whereas none were statistically significant for AL. Using bivariate linear regression, each of the six sites was a significant predictor (p < 0.0001) of the six-site measure for PD and AL. However, the mesial lingual site gave the best fit for PD (R² = 0.88, p < 0.0001) and AL (R² = 0.91, p < 0.0001). For PD severity, the mesial buccal (MB) and distal buccal (DB) sites had the smallest relative biases (both 12.42%); for AL severity, the MB had the smallest relative bias (1.86%). Conclusions: In this sample, the MB site showed the most consistent pattern across the various analyses. Supported by HRSA Grant 5 D13 HP30009.