Reliability of quantitative laser fluorescence analysis of smooth surface lesions adjacent to the gingival tissues.

The aim of this study was to determine the impact of analysis patch border inclusion or exclusion on the reliability of quantitative laser fluorescence (QLF) analyses of lesions close to the gingival margin. Ninety-three lesions on the maxillary anterior teeth were imaged using a fluorescent capturing system. All the lesions were located on the gingival third of the teeth. One examiner undertook two analyses of the images 1 week apart and another examiner undertook a single analysis. Analyses were undertaken using QLF 2.00 g with four patch borders. Each border was assessed as active or inactive. QLF metrics DeltaF, DeltaQ and area were exported. Kappa statistics were used to measure the agreement of border inclusion between all three analyses, and intra-class correlation coefficients (ICCs) were used to determine the intra- and inter-examiner reliability of the QLF metrics. Agreement on border exclusion was poor; with intra-examiner kappa of 0.48 and inter-examiner at 0.20. However, despite the inconsistencies in border exclusions the ICCs for each QLF metric were high; intra-examiner DeltaQ 0.91, DeltaF 0.80 and area 0.92; inter-examiner DeltaQ 0.86, DeltaF 0.68, area 0.88. Lesions adjacent to the gingival margin will often require a patch analysis border to be excluded in order to ensure a satisfactory reconstruction and thus accurate analysis. The decision to include or exclude a border and, if excluded, which border to select appears to be highly variable between examiners. Nevertheless, the QLF metrics appear to be robust as demonstrated by the high ICCs noted in this study.