Thank you for giving us the chance to reply to the Letter to the Editor critiquing certain aspects of our article “Does the piezoelectric surgical technique produce fewer postoperative sequelae after lower third molar surgery than conventional rotary instruments? A systematic review and meta-analysis”. The claims made in the Letter are not correct, and we would like to address each one, objectively and briefly, in order to clarify the results provided in our review in regard to the critique. We think that if the methodology presented in our article had been read carefully and understood, this critique would have been unnecessary. The points outlined below refute their claims.
First, concerning the inclusion of the study by Chang et al., this study fulfilled the inclusion criteria of our study. It is important to state that all of the studies included used a Piezotome not to perform tooth sectioning, but just to perform the bony guttering around the impacted tooth to facilitate luxation and delivery of the tooth. The study by Chang et al. used a high speed hand-piece to perform tooth or crown sectioning, and then a Piezotome was used to remove bone around the tooth. Thus it was appropriate to include that study in our analysis.
Second, the claims concerning the inclusion and exclusion criteria of our study being very narrow and the participants not falling within the specified age range or not meeting the stated indications for extraction are not correct. Our study used specific and very clear inclusion criteria with the identification of patients aged 18–25 years being treated for orthodontic, prophylactic purposes.
Third, concerning the claim that conducting an analysis of studies with variations in follow-up times (ranging from 1 day to 30 days) is inappropriate, we wish to state that a principle of meta-analysis is based on the pooling of data from the studies included, so heterogeneity could be the end result. This is why an assessment of heterogeneity among included studies is an essential part of every meta-analysis, and such an assessment was performed. In the case of diversity within the studies included and this heterogeneity being statistically significant ( P < 0.10), the random-effects pooled estimate is necessary; this was applied for all of our analyses. Therefore, these analyses are valid and appropriate statistically.
Fourth, concerning the analysis for the outcome of trismus, there was no statistically significant heterogeneity ( P = 0.0001; I 2 = 73%). Thus, we used the mean difference instead of the standard mean difference (SMD). Concerning the analysis of facial swelling, some studies reported this in millimetres and others in centimetres; in such cases (when there is a difference in the instrument used to assess an outcome), the SMD is required, and this was used.
Fifth, for the outcome of facial swelling, after correction through the exclusion of the data from the study by Sivolella et al. and the preoperative measurement from the study by Sortino et al., the statistically significant advantage of piezoelectric surgery over conventional rotary surgery remained, as reported in our study ( Fig. 1 ).