Determining the Perceptibility and Acceptability Threshold of the Relative Translucency Parameter

Translucency is considered one of the most critical components to imitating the natural appearance of teeth (1). Translucency is defined by a material's optical properties that dictate how light interacts inside of it. The Translucency Parameter (2) is a measure of translucency which is obtained via spectrophotometry of a single specimen (3) with pure white and black backings. The Relative Translucency Parameter (RTP) is the Translucency Parameter when the black and white backings are not pure. In our study, the Relative Translucency Parameter was calculated as the ∆TP between specimens, but it is important to note that the Translucency Parameter is universal and can be used for any thickness. When using the Relative Translucency Parameter, thickness must be well-defined within narrow limits. A clinically relevant measurement of translucency can be determined via perceptibility and acceptability thresholds. Although these thresholds have been established for color difference, there is limited agreement on the perceptibility and acceptability thresholds of the Relative Translucency Parameter determined in a dental clinical setting. Five groups, and a total of ten lithium disilicate CAD specimens were fabricated from highest to lowest translucency, controlling thickness and translucency in the following specifications: 0.5mm High Translucency (HT), 1mm Medium Translucency (MT), 1.5mm Low Translucency (LT), 2mm Medium Translucency, and 2mm Medium Opacity (MO). All specimens were shade A2. Using a glycerin solution, the specimens were optically joined to a black and white backing, both of which were at least 84 CIEDE2000 color difference units apart. The spectral radiance was measured for each of the specimens, from which the Translucency Parameter and Relative Translucency Parameter values were calculated. Teeth #8 and #9 (maxillary central incisors) on a typodont were hand prepared for lithium disilicate crowns using a putty cutback method to create a facial thickness of 0.5mm, 1mm, 1.5mm, and 2mm. The preparations were digitally scanned with the Trios 3 Scanner (3 Shape, Copenhagen, Denmark) and then the crowns were digitally designed, milled, and finished. Utilizing 0˚ observation and D65 lighting at 45 degree illumination in a Pantone light booth, a digital image was taken of all pairs of #8 and #9 crowns at the same angulation and lighting condition for every specimen using a camera tripod. These images were digitally reproduced and paired at random (#8 and #9) to create 100 paired samples. 30 members of the Department of Restorative Dentistry at the University of Illinois Chicago were recruited to participate and take survey. The inclusion criteria were membership in the Department of Restorative Dentistry at the University of Illinois Chicago, and the exclusion criteria included scoring 1.5 or less on the Ridgevue Contrast Sensitivity test (4). No participants were excluded. A photo of two specimens simultaneously appeared on the provided digital monitor with specifications: 1920x1080 pixels (full high definition) and 10-bit RGB. Participants were given instructions before beginning the survey. Participants were asked to select “Yes” or “No” to the question, “Is there a difference in translucency between these two crowns?” If the answer was no, the next specimens were displayed. If the answer was yes, the next question was: “Is the translucency difference between the two crowns clinically acceptable?” Participants were asked to complete as many randomized specimen comparisons as possible for 30 minutes before taking a break for 10 minutes, to reduce visual fatigue. A full repeated-measures logistical regression model was used for statistical analysis. A generalized linear mixed model (GLIMMIX) was used as a fitting method to model the probability of a Preferred or Not Preferred judgement and this model was subjected to a repeated-measures ANOVA to account for repeated questions. The 50:50% acceptability and perceptibility thresholds were obtained from the observed relative frequencies plots at 0.50. No statistically significant difference was found between the slopes of the covariate for the acceptability and perceptibility judgements (P = 0.2467). However, the slope of the covariate was found to be significantly different from zero (P <.0001) and the intercept was found to be dependent on the judgement (P = 0.0003). The 50:50 acceptability threshold was determined to be 12.60, p = 0.0003. A valid 50:50 perceptibility threshold value was not determined, which was -17.4965, p = 0.0003. The acceptability threshold of the Relative Translucency Parameter for lithium disilicate crowns was determined to be 12.60, which is much larger than the established acceptability threshold for color difference of 1.8.