Partial-mouth plaque recording revisited: evaluation of tooth- and surface-based subsets using a data-driven benchmark

OBJECTIVES: To evaluate the validity of established and newly developed partial-mouth recording subsets for approximating full-mouth plaque across different plaque metrics and oral hygiene states, and to determine whether surface-based subsets derived from the observed plaque distribution offer advantages over established recording schemes. MATERIALS AND METHODS: Disclosed-plaque intraoral scans from 30 adults (24.0 ± 4.1 years) were obtained at baseline, after 72 h plaque accumulation, and after toothbrushing. Plaque was quantified planimetrically (percentage plaque coverage, P%) and using two plaque indices (Turesky modified Quigley-Hein Plaque Index (TQHPI); Rustogi modified Navy Plaque Index (RMNPI)). Established Ramfjord (6 teeth/12 surfaces) and Community-Periodontal-Index-of-Treatment-Needs (CPITN; 6 teeth/12 surfaces or 10 teeth/20 surfaces) subsets, optimised subsets derived from the observed plaque distribution to represent overall plaque levels, and additional subsets representing plaque-prone sites with elevated plaque levels were compared with full-mouth recordings. Performance was assessed using bias, root-mean-square error, intraclass correlation coefficients (ICC), and mixed-effects modelling. RESULTS: An Optimised-12 surface subset showed near-perfect agreement with P% full-mouth recordings (bias 0.03, ICC 0.962); the 12-surface CPITN revealed similar results (bias -0.07, ICC 0.967). The Ramfjord subset showed slight underestimation but high agreement (bias -1.23, ICC 0.970), whereas the 20-surface CPITN consistently overestimated full-mouth plaque (bias 4.33, ICC 0.839). A reduced 6-surface Ramfjord subset also preserved high agreement with full-mouth recordings (bias -0.76, ICC 0.954). For surfaces with elevated plaque levels, newly developed 12- and 6-surface subsets closely matched the elevated-plaque reference. Subset performance patterns were consistent across all oral hygiene stages and across plaque metrics (P%, TQHPI, RMNPI). CONCLUSIONS: Except for the 20-surface CPITN, all subsets provided a robust approximation of full-mouth plaque. Surface-based subsets derived from the observed plaque distribution offer accurate alternatives for assessing overall plaque, while the elevated-plaque subsets provided a useful approach for monitoring plaque-prone sites. CLINICAL RELEVANCE: Time-efficient partial-mouth recording may support feasible plaque monitoring in clinical research and preventive care, particularly when full-mouth assessment is impractical.