This parallel randomized clinical trial was approved by the Research Ethics Committee of the Faculty of Dentistry at King Abdulaziz University (No. 078 − 16) and the ethical standards of the Declaration of Helsinki. The research was registered under the Clinical Trial Registry (NCT03180151) and the updated CONSORT statement for reporting randomized clinical trials was followed.
The study sample was taken from a previous clinical trial . Briefly, Orthodontic patients were recruited from the Faculty of Dentistry at King Abdulaziz University according to the following inclusion criteria: 1) patients age between 16–26 years, 2) half cusp or more Class II division 1 malocclusion, 3) mild or no crowding, 4) maxillary arches requiring bilateral first premolar extractions as part of the orthodontic treatment plan, and 5) good oral hygiene and healthy periodontal condition. The exclusion criteria were: 1) history of systemic disease, 2) presence of dental or facial anomalies, 3) patients receiving treatment that may affect bone biology and/or tooth movement, and 4) history of periodontal surgery. After obtaining informed consent, a simple randomization method was used as follows: patients were randomly assigned into two equal groups: 1) piezocision group (n = 13) and 2) control group (n = 13), using the opaque sealed envelope technique.
All patients were treated using a modified bidimensional bracket system 0.018-inch slots in the six anterior teeth and 0.022-inch slots in the posterior teeth (Gemini Series Bracket; 3M Unitek, Monrovia, CA, USA). Also, all patients had miniscrews (3M Unitek, Monrovia, Calif, USA) inserted bilaterally between the maxillary first molar and second premolar. After leveling and alignment, all patients had their upper first premolars extracted. In the piezocision group, a canine to canine piezocision decortication was performed only labially during the extraction procedure and no grafting was done. The en-masse retraction was started a week later for both groups on a 0.018 × 0.025-inch stainless steel arch-wire through nickel-titanium active coil springs delivering a bilateral 250grams of force. To monitor the force level, a strain caliper was used (Dentaurum, Ispringen, Germany).
During the study, one patient from the piezocision group and two patients from the control group were excluded due to miniscrew failure. Hence, 23 participants completed the study, 12 in the piezocision group, and 11 in the control group.
CBCT images were acquired using the i-CAT next generation CBCT machine (I-CAT Cone Beam 3D Imaging, Imaging Sciences International Hatfield, PA, USA). CBCT images were taken during the initial orthodontic records visit (T0) and at the end of en-masse retraction, mean = 122.74 ± 3.06 days (T1). Pulp volume of each of maxillary anterior teeth (canine to canine) was measured by one blinded investigator using an image processing software (Mimics Version 14 on Windows; Materialize, Leuven, Belgium). Measurements were performed by image segmentation and separating the selected teeth which were carried out by grayscale threshold. Briefly, on the axial, sagittal, and coronal views a mask was built to segment the pulp cavity of the selected tooth, then an optimum separating grayscale threshold was selected. The threshold values were established separately with the same Hounsfield Units (HU) for each tooth at T0 and T1, then the calculation of the 3-dimensional image was performed. The mask of the pulp cavity in 3-dimensional allowed the operator to calculate the pulp volume (Fig. 1). The root lengths of each maxillary six anterior teeth were measured along the long axis of the teeth using the CBCT frontal view. The measurements were recorded from the middle of the cementoenamel junction to the root apex.
Intra-examiner reliability was tested using the intra-class correlation coefficient (ICC). Measurements of 10 participants were repeated after 2 weeks to assess measurement errors. The ICC coefficient ranged between 0.79 and 0.98%, indicating excellent reliability.
Descriptive statistics (means and standard deviations) were calculated. The normality of the main dependent variables tested using the Shapiro-Wilks test (P > 0.05) showed that they were almost normally distributed in both groups. Paired t-tests were used to compare mean changes (T0 vs T1) within groups and independent sample t-tests were used to compare mean differences (T0-T1) between groups. Chi-square was used to test the distribution of males and females between and within groups. Pearson’s correlation coefficient was used to assess the relationship between the mean changes of each tooth pulp volume and its root length. Data were analyzed using the Statistical Package for Social Sciences (IBM SPSS Statistics for Macintosh, Version 26.0. Armonk, NY, USA). The significance level was set at α = 0.05 by a statistician who was blinded to the results.