Objectives: To evaluate and compare the enamel microcracks (EMCs) characteristics (qualitative and quantitative) in the form of tooth damage before and after debonding from human teeth of in vitro studies.
Eligibility criteria: Laboratorial studies evaluating EMCs characteristics before and after debonding metal and ceramic brackets from human teeth with intact buccal enamel.
Information sources: An electronic search of four databases (all databases of the Cochrane Library, CA Web of Science, MEDLINE via PubMed, and Google Scholar) and additional manual searches were carried out, without language restrictions. Studies published between 2000 and 2017 years were selected. Reference lists of the included articles were screened, and authors were contacted when necessary.
Risk of bias: The following six parameters were analyzed: blinding of examiner and outcome assessment, incomplete outcome data before bonding and after debonding, selective outcome reporting, and incomplete reporting of EMCs assessment.
Included studies: Out of 430 potentially eligible studies, 259 were screened by title and abstract, 180 were selected for full-text analysis, 14 were included in the systematic review. Seven studies were selected for the meta-analysis.
Synthesis of results: The results for EMCs characteristics were expressed as mean differences (MDs) with their 95 per cent confidence intervals (CIs), and calculated from random-effects meta-analyses. Debonding was associated with the increase in number (three studies, MD = 3.50, 95% CI, 2.13 to 4.87, P < 0.00001), length (seven studies, MD = 3.09 mm, 95% CI, 0.75-5.43, P < 0.00001), and width (three studies, MD = 0.39 µm, 95% CI, -0.01 to 0.79, P = 0.06) of EMCs. Considerable statistical heterogeneity was found for two forest plots evaluating the changes of number and length characteristics during debonding.
Conclusions: There is weak evidence indicating length and width of EMCs increase following bracket removal and the scientific evidence concerning quantitative evaluation of the number parameter before and after debonding is insufficient. However, there is a strong evidence that after debonding the number of EMCs is likely to increase.
Registration: No registration was performed.