The sensitivity of digital intraoral scanners at measuring early erosive wear

Abstract

Objectives: To investigate the sensitivity of intraoral scanners to quantitatively detect early erosive tooth wear. Methods: Natural buccal enamel samples were mounted in acrylic and scanned at baseline with an intraoral scanner (3M True Definition Scanner, 3M, USA). Samples were then exposed to 0.3% citric acid pH 3.2 at intervals of 10 minutes up to a total of 120 minutes and scanned after each exposure resulting in analysis of 13 datapoints per sample. Each scan was aligned with the baseline and data points superimposed using an iterative closest point (ICP) algorithm on the acrylic surfaces (Geomagic Control Software, 3Dsystems, Darmstadt, Germany). Wear was measured using maximum profile loss, average profile loss and volume change. Data were normally distributed and Pearson correlations between erosion time and wear measurements assessed. Results: After each 10-minute exposure until 120 minutes, maximum profile loss (m) increased from 33.4 to 72.8 m, average profile loss from 9.1 to 18.6 m. Wear correlated with increasing acid exposure for both maximum profile loss wear (r=0.877 p<0.001) and average profile loss (r=0.663 p=0.019) respectively. Volume measurements were inconsistent at this level of wear. Conclusions: Using scan data obtained from the intra oral scanners (IOS), increasing step height changes were observed with increasing exposures to acid. This study indicates there is potential of scans taken with an IOS to be used to detect early erosive tooth wear. However, precision was low suggesting limitations for minimal changes. Clinical Significance: Although sub-visual wear was detected by intra-oral scanners on natural enamel surfaces, the accuracy was not sufficient to reliably diagnose that wear had occurred and interpretation of measurements should be done with caution. However, these results may be promising for detecting wear at more advanced stages.

Detecting early erosive tooth wear using an intraoral scanner system

“…By combining measurement error and uncertainty, surface profile differences>0.035mm represent true tooth substance loss. Taking into account that the accuracy of a long span 3D model, as featured in this study, is lower compared to single tooth models, we can foresee even smaller measurement error and uncertainty if the same method is tested on single tooth models [20,21,25,28]. Accordingly, larger measurement error and uncertainty are expected in full-arch 3D models [28].…”

“…Accordingly, larger measurement error and uncertainty are expected in full-arch 3D models [28]. Furthermore, it is worth mentioning that no powder was needed for scanning with the IOS system used, a notable point as a previous study reported part of the measurement error had resulted from the presence of powder on the surfaces of interest [21].…”

“…QLF and OCT, although they can potentially be applied directly in the mouth [17], show suboptimal reproducibility in longitudinal studies involving patients [2,3,5,18,19]. The use of IOS has been recently suggested for early detection, quantification, and monitoring of tooth wear based on in vitro [20][21][22] as well as in vivo studies examining 3D data obtained directly from patients [23] or cast models [24]. Studies focusing on visual detection of early erosive tooth wear demonstrated detection with higher sensitivity using meticulous visual examination (BEWE index) on full-arch digital 3D models compared to traditional visual examination on patients [23,24].…”

“…This finding was attributed to the larger magnification and clearer image of the dentition allowed by the digital 3D model compared to the traditional visual examination directly on the patient. Other studies went further by using specific software to align multiple 3D models, from which measurements of profile or volumetric differences were obtained [20][21][22]25,26]. In order to develop an objective, automated, and reliable in vivo method for early detection and monitoring of tooth wear, the use of an accurate IOS is essential [27,28], in addition to a reliable method for 3D model alignment with minimal error [20,21,25].…”

“…Other studies went further by using specific software to align multiple 3D models, from which measurements of profile or volumetric differences were obtained [20][21][22]25,26]. In order to develop an objective, automated, and reliable in vivo method for early detection and monitoring of tooth wear, the use of an accurate IOS is essential [27,28], in addition to a reliable method for 3D model alignment with minimal error [20,21,25]. This study builds upon the existing knowledge derived from in vitro models with single-tooth set-ups [20,24,25], in which the accumulation of errors related to scanning and tooth alignment is relatively small, and approaches the clinical scenario by investigating erosive tooth wear on a set-up designed to simulate the in vivo dentition.…”