Prediction of optical efficacy of vital tooth bleaching using regression analysis

Establishing a colorimetric guideline to predict the effectiveness of tooth bleaching could produce a more reliable dental treatment. The purpose of this study was to evaluate the effectiveness of tooth bleaching and to test the predictability of tooth color changes. A 10% carbamide peroxide bleaching system was used in studies at Harvard University and at Iwate Medical University in Japan. L* a*, and b* values (CIELAB) for pre- and postbleaching were obtained and color differences (DeltaE) were calculated. The b* and L* values of the original tooth color indicated a relatively strong to moderate correlation with AE values, whereas a* showed a weak correlation. The multiple-regression equation obtained from the color data of Harvard subjects performed better than the predictive model. The predicted AE correlated strongly with the observed DeltaE (r = 0.78). The validation of this equation on data collected from Iwate confirmed the strong correlation (r = 0.74). (C) 2004 Wiley Periodicals, Inc.