Our article was not intended to prove or disprove any mechanism to explain how or why bleaching agents frequently cause dentinal hypersensitivity, as it is well known that they do. Whether that may be due to osmotic effects is not fully established.
Our citation of Bränström’s1 article was intended to note that we believe it plausibly explains the etiology of many cases of hypersensitivity, while giving credit to the architect of that model. The fact that Tung and Eichmiller’s2 1999 ADAF article directs the reader to another study by the same authors does not merit further comment, beyond adding that they discuss a 72 percent reduction in thermal (cold air) sensitivity and a 76 percent reduction in tactile sensitivity subsequent to application of ACP. The fact that this was found in a small study does not negate the veracity of their results.
We were very careful to point out that the ability of ACP to depolarize nerve endings is a hypothesis requiring further study. Dr. Kanca correctly points out that Yates and colleagues3 found disappointing results with their particular ACP preparation, but we stand by our data, which showed reduced sensitivity scores in the test group exposed to the gel containing ACP.
With respect to our use of terminology to describe tests, or the level of precision we used to report our results, we acknowledge the editorial accuracy, but remind the reader that we do not claim to have found an unarguable result and continue to encourage further study on the effect of any dental bleaching product on dentinal hypersensitivity.
When we designed this study, we looked for the most similar commercially available product we could find to use as a control. Since we did not have access to the formulations of any manufacturer other than Discus, we necessarily excluded all non-Discus products, and we made that quite clear. The ingredients, and relative ratios of those ingredients, are nearly identical in the test and control gels, with the exception of the changes made to accommodate the ACP in the test gel.
We believe Dr. Kanca’s contention that one of the byproducts of ACP is potassium nitrate (KNO3) is misleading. When calcium nitrate and potassium phosphate react to form ACP, potassium and nitrate ions are indeed left in solution, but their concentration would amount to no more than 0.25 percent, an amount so far below the usual 3 to 5 percent amount that is generally employed for mitigating sensitivity as to make it clinically irrelevant. Regardless of whether or not KNO3 is formed, the control material had no KNO3 or ACP; hence, we believe that any desensitizing effect was fairly attributed to the test gel containing the ACP constituents.
Dr. Tung has, in fact, stated that the mechanism of ACP-induced desensitization is via blockage of dentinal tubules.4 However, Dr. Kanca has correctly pointed out a flaw, not in the protocol, but in our description of it, in that patients with recession or exposed dentin were actually not excluded. In fact, Figure 1 on page 387 clearly shows a patient with clinical attachment loss. We appreciate his editorial diligence.
For both the shade-change and dentinal hypersensitivity analyses, we employed the t test for paired samples, which is appropriate for testing the average of differences between series of paired observations; that is, pre-procedure versus postprocedure data. We are aware that previous authors have employed the Kruskal-Wallis test for evaluating shade change data that we presume are based solely on the fact that the shade and sensitivity data appear to be ordinal. They actually are not.
Our understanding of Kruskal-Wallis is that it is appropriate for examining unpaired ordinal data originating from the same population and presented in a validated, logical sequence. The sensitivity scales we employed may be logical; our discussion makes it clear they are not validated, and the analysis of paired data therefore makes use of Kruskal-Wallis inappropriate.
Although it can be debated whether the group of patients who would consent to a dental whitening procedure is a truly random sample, the distribution of sensitivity scores for both groups is within 2 standard errors of kurtosis at all evaluation points, indicating a near normal distribution. We, therefore, believe the parametric analyses we employed are appropriate and do not mislead the reader.
We do not concur with the suggestions that we have made egregious or blatant errors, but we respect that reasonable minds can find disagreement on the same topic. We believe Dr. Kanca’s suggestions would be more appropriate for larger studies that are beyond the scope and intended purpose of this article. A study of tooth bleaching that could serve as a universal clinical predictor of shade change or dentinal hypersensitivity would be of tremendous benefit to the professional community, but it would require demographic and shade-matched enrollment of several hundred patients, and a validated evaluation method for each.
Although we would strongly endorse such a study, we are not in a position to sponsor one. Instead, we remind the reader that the intent of this study was to compare a new whitening product to one that is already commercially available, and we believe our article is appropriate for that purpose.
TOP
REFERENCES
