Influence of study design on the impact of bleaching agents on dental enamel microhardness: a review.

OBJECTIVE Numerous studies investigated the impact of bleaching procedures on enamel microhardness. The outcomes of these studies reveal inconsistencies regarding the fact whether a microhardness reduction due to bleaching occurs or not. Aim of the present review was to summarize the existing literature of external bleaching therapies, which used microhardness tests for evaluation of possible effects on enamel and to weigh up different parameters of the study designs with respect to the outcome of these studies. METHODS The data from original scientific full papers listed in PubMed or ISI Web of Science (search term: enamel and (bleaching or peroxide) and (hardness or microhardness or Knoop or Vickers)) and received by additional hand-search meeting the inclusion criteria were included in the review. Influences of different parameters on the outcome of the bleaching treatments were analyzed with the Fisher's-exact-test. RESULTS A total of 55 studies were identified with 166 hardness measurements conducted directly after bleaching and 69 measurements performed after a post-treatment episode. Directly after bleaching, 84 (51%) treatments showed microhardness reduction compared to baseline, whereas 82 (49%) did not yield microhardness reduction. After the post-treatment episode, 20 (29%) treatments showed hardness reduction and 49 (71%) did not. A significant higher number of bleaching treatments resulting in enamel microhardness reduction were observed, when artificial instead of human saliva was used for storage of the enamel samples in the intervals between the bleaching applications and when no fluoridation measures were applied during or after the bleaching phase. SIGNIFICANCE The review shows that in those studies, which simulated the intraoral conditions as closely as possible, the risk of enamel microhardness decrease due to bleaching treatments seems to be reduced. Nevertheless more in situ- and in vivo-studies are needed to verify this observation.