Evaluation of the effectiveness of piezocision-assisted flapless corticotomy in the retraction of four upper incisors: A randomized controlled clinical trial Ocena skuteczności bezpłatowej piezokortykotomii w retrakcji czterech górnych siekaczy – randomizowane kontrolowane badanie kliniczne

Background. Comprehensive orthodontic treatment may last for 20–24 months. Reducing the treatment time has become a common demand from both patients and orthodontists. Minimally invasive piezocision is one of the suggested approaches to accelerate the orthodontic tooth movement. Objectives. The aim of this randomized controlled clinical trial was to assess the effectiveness of the flapless piezocision procedure in accelerating the retraction of upper incisors. Material and methods. A single-blinded, parallel-group controlled clinical The patients had class II division I malocclusion and were treated with fixed appliances using the two-step retraction technique. With an allocation ratio of 1:1, the participants were randomly assigned to either the experimental group (n = 21) or the control group (n = 21) using a computer-generated list of random numbers. Allocation was concealed due to the use of sequentially numbered, opaque, sealed envelopes. The primary outcomes were the rate of incisor retraction (RIR) and the time required for retraction. The outcome assessor was blinded. in the control group than in the experimental group. Skeletal measurements showed insignificant changes following retraction between the 2 groups. Conclusions. The piezocision procedure was found to be effective in accelerating the retraction of 4 upper incisors, reducing the retraction time, preserving anchorage and enhancing root torque control during retraction.


Introduction
The number of patients seeking improvement in den tofacial function and esthetics has increased over the past years. 1 The overall time of comprehensive ortho dontic treatment with fixed appliances may amount to 24 months, 2 depending on the patient's characteristics and the complexity of malocclusion. Prolonged orth odontic treatment may lead to external root resorption, hypomine ralization, dental caries, periodontal disease, pain, and discomfort. 3 Lengthy orthodontic treatment is considered a drawback; it is usually associated with poor patient compliance and may result in dissatisfac tion. 4 Therefore, the acceleration of the orthodontic tooth movement has become one of the primary concerns among orthodontists worldwide. To date, several methods have been proposed to shorten the treatment time, such as the use of lowfriction selfligating brackets (SLBs), pharmacological approaches, physical stimuli, and surgi cal methods. 5 Surgical interventions have been found to be the most effective in enhancing the tooth movement and the most widely used, with predictable outcomes. 6 These surgical interventions include conventional cor ticotomy, interseptal alveolar surgery, accelerated os teogenic orthodontics, dentoalveolar distraction, and periodontal distraction. 7 It has been demonstrated that surgical injury to the alveolar bone can temporarily ac celerate the orthodontic tooth movement by evoking the socalled 'regional acceleratory phenomenon' (RAP), which is a physiological healing response that decreases the resistance of the alveolar bone to orthodontic forces and reduces the treatment time. 8 Even though the conventional corticotomy procedures have been proven to be effective in reducing the orth odontic treatment time, 9 they have adverse sequelae, such as interdental bone loss, the loss of the attached gingiva, periodontal defects, and hematomas in the neck and face. These documented complications are due to the inva siveness of the traditional procedures with the need for elevating fullthickness periosteal flaps. 10 Consequently, several researchers has tended to investigate less invasive surgical acceleration modalities, such as laserassisted flapless corticotomy, piezocision, corticision, and micro osteoperforations. 6 Piezocision has recently evolved as a novel approach of manipulating the cortical bone with minimal damage, less discomfort and greater patient acceptance, 11 and was first introduced by Vercellotti and Podesta. 12 In 2009, Dibart et al. used a piezoelectric knife to achieve flapless alveolar decortication, subsequently inducing RAP with the possibility of hard or soft tissue grafting, using selec tive tunneling procedures. 13,14 Piezocisionassisted corti cotomy procedures have been investigated with several types of orthodontic tooth movement, such as resolving crowded lower anterior teeth with or without the extrac tion of premolars, 15 the retraction of maxillary canines, 16 the enmasse retraction of maxillary anterior teeth, 17 and leveling and alignment in moderately crowded arches us ing SLBs. 11 The camouflage treatment of class II malocclusion in nongrowing patients includes the removal of upper first premolars with the subsequent retraction of upper front teeth. 18 There are 2 main strategies to retract the anterior teeth into the retraction space -enmasse retraction and twostep retraction. The most common technique is the sequential method, in which canines are independently retracted, followed by the retraction of 4 incisors in the 2 nd stage. 19 Several studies have reported speeding up canine re traction in the twostep retraction technique. 16,20 In the study of Alfawal et al., canines moved 1.5-2 times faster than those retracted in the conventional way, which meant a reduction of approx. 2 months or 10% with regard to the average total orthodontic treatment time. 16 Such a reduc tion is not clinically sufficient and it does not significantly decrease the overall treatment time. There is a need to accelerate incisor retraction in order to gain a consider able overall time reduction; however, this dimension has not yet been evaluated in the literature. 6 To the best of our knowledge, there is no randomized controlled trial (RCT) assessing the efficacy of flapless piezosurgery in the re traction of 4 upper incisors in the twostep retraction technique.
Therefore, the objectives of this trial were to investigate this treatment modality in terms of rate of incisor retrac tion (RIR), time required for retraction and molar anchor age loss, and to assess dental and skeletal changes follow ing the treatment.

Trial design and settings
This study was a twoarm, parallelgroup RCT, and there were no changes regarding its published protocol follow ing the trial commencement. This study was conducted at the Department of Orthodontics at the University of Damascus Dental School, Syria, between September 2016 and November 2017. Ethical approval was obtained from the related Local Ethics Committee of the Univer sity of Damascus Dental School (UDDS293822112015/ SRC5927). This trial was registered in the Clinical Trials database with the identification number NCT03149016.

Sample size calculation
The sample size was calculated using Minitab ® , v. 17 (Minitab Inc., State College, USA) considering that the two sample ttest was intended at a significance level of 0.05 and a power of 85%. It was assumed that the piezocision intervention would decrease the overall treatment time by 40%, whereas the variability of this outcome measure in a previous study was 9.27 ±2.55 months 17 ; therefore, the number of participants required in each group was 19. In order to compensate for sample attrition, 2 partici pants were added to each group with a total sample size of 42 patients.

Participants and eligibility criteria
Recruitment was done by screening patients who had visited the Department of Orthodontics at the University of Damascus Dental School and were seeking orthodontic treatment. Out of the 109 patients who were initially exa mined, 42 participants were identified to be eligible for the study. The included patients were randomly assigned to 2 equal groups with a 1:1 allocation ratio: the piezocision group (PG; n = 21), which received a surgical interven tion, and the control group (CG; n = 21), in which incisors were retracted in the conventional manner. All patients were chosen to meet the following inclusion criteria: class II division I malocclusion requiring the extraction of up per first premolars and the retraction of the upper ante rior teeth; class II skeletal relationship (4° < ANB < 10°); the clinical and radiological diagnosis of an average to vertical anterior facial height; an overjet not exceeding 10 mm and a normal overbite of 0-50%; wellaligned maxillary incisors with mild crowding (≤3.5 mm); age of 15-26 years with a skeletal maturity stage of MP 3U or R U depending on the handwrist radiograph; complete per manent dentition; no congenitally missing teeth (except for third molars); no orthodontic treatment received be fore; the absence of systemic disorders that could contra indicate oral surgery or affect the tooth movement; and good oral hygiene. Informed consent was obtained after a full explanation to patients and/or their parents about the purpose of the research, methods and procedures.

Randomization, allocation concealment and blinding
Simple randomization was done by one member of the academic staff not involved in this trial, using a computer generated list of random numbers (Minitab, v. 17). The participants were randomly assigned to either PG or CG. Allocation was concealed using sequentially numbered, opaque, sealed envelopes, which were opened only after the end of the canine retraction phase. The blinding of the principal investigator and the patients was impossible, and thus blinding was only employed in the data analysis.

Orthodontic treatment in both groups
After setting up usual diagnostic records, orthodontic treatment in both groups was established using MBT pre adjusted appliances with 0.022inch slot size (American Orthodontics, Sheboygan, USA). A soldered transpalatal arch was used to reinforce anchorage at the beginning of treatment. The teeth were leveled and aligned using the following archwire sequence: 0.014inch NiTi (nickel titanium); 0.016inch NiTi; 0.016 × 0.022inch NiTi; 0.017 × 0.025inch NiTi; and finally 0.019 × 0.025inch SS (stainless steel) (American Orthodontics). 21 After the leveling phase had been finished, upper first premolars were extracted and upper canines were retracted us ing elastic chains until class I canine relationship was achieved. The participants were then randomized into CG or PG. Incisor retraction was initiated by soldering 5millimeterlong power arm hooks to the basal arch distal to the brackets of lateral incisors and NiTi closed coil springs were used to deliver a continuous force of the load of 150 g each side to retract upper incisors. 22 The force level was checked and measured every other week using a force gauge to keep it unchanged during the entire retraction stage. The incisors retraction stage was started by applying coil springs (T 0 = start of obser vation) and considered complete (T f = end of observa tion) when one of the 2 possible events occurred: spaces lateral to incisors were closed, or a contact between up per incisors and lower incisors or the brackets on lower incisors was observed.

Surgical procedure in the experimental group
Piezosurgery was carried out at the Department of Oral and Maxillofacial Surgery at the University of Damascus Dental School, and was performed by an Oral and Maxil lofacial Surgery (OMFS) Master's degree student (E.M.) under the supervision of an OMFS consultant (Y.M.). Af ter rigorous rinsing with 0.12% chlorhexidine gluconate for 1 min, local infiltrative anesthesia was induced in the buccal and palatal aspects of the upper anterior segment. Using a blade size 15, vertical interproximal microinci sions were made through the periosteum between the particular teeth, 4 mm above the interdental papillae, and extended along the middle third of the root (Fig. 1).

Fig. 1. Piezoelectric knife inserted to perform vertical incisions
the middle point of either the right or left central incisor; the medial end of the third palatal ruga as a stable landmark for assessing the anteroposterior movements of incisors and molars; and the central fossa of first molar (Fig. 4). 23 The standardized digital photographs of the poured study models were taken using Nikon D80 macro twin flash 18-55millimeter lens (Nikon, Shinagawa, Tokyo, Japan). The camera was attached to a holder specifically designed for this research project, with a model-lens dis tance of 30 cm, employing perpendicular projection. The models were placed against a dark background to aid vi sualization. A steel millimeter ruler was inserted into each image for size correction. After digitizing the marked landmarks, linear measurements were calculated on the digital photographs using the AudaxCeph ® , v. 3.4.2.2710 orthodontic software (Audax d.o.o., Ljubljana, Slovenia). The displacement of the incisor segment was assessed by measuring the anteroposterior distance between the in cisal edges of maxillary central incisors to the medial end of the third palatal ruga (Fig. 5). The time required for incisor retraction was recorded for both groups.

Molar anchorage loss
The amount and rate of the mesial displacement of up per first molar was assessed by measuring the distance from the medial end of the third palatal ruga to the cen tral fossa of first molar on the digital photographs of the related dental casts (Fig. 5).
A piezosurgical microsaw was used to create buccal and palatal cortical alveolar incisions to a depth of approx. 3 mm, which was verified by a millimetric gradation on the piezosurgical knife (Fig. 2,3). The postoperative in structions given to the patients were as follows: analgesics to relieve pain (acetaminophen 500 mg tablets thrice daily (TID) for a week); antibiotics (Augmentin 1,000 mg twice daily (BID) for a week); nonsteroidal antiinflammatory drugs (NSAIDs) were banned, as they were expected to interfere with RAP; the application of ice packs for the next 12 h; the avoidance of irritating food for 2-3 days after surgery; maintaining ideal oral hygiene; and rinsing with 0.12% chlorhexidine gluconate BID for a week.

Primary outcome measures: Rate of incisor retraction and time required for retraction
The primary outcome was RIR in the 3 rd week (T 1 ), 6 th week (T 2 ), 9 th week (T 3 ), and 12 th week (T 4 ), which was calculated as the distance incisors moved each week. Maxillary alginate impressions were taken for each pa tient at 3week intervals starting from the onset of retrac tion (T 0 ) until the end of retraction (T 4 ).
Using a 0.3millimeter graphite pencil, the following landmarks were marked on the casts by the principal researcher and rechecked by the research supervisor:

Lateral cephalometric analysis
Standardized lateral cephalometric radiographs were ob tained at 2 assessment times: T 0 (at the beginning of incisor retraction) and T 4 (after the end of retraction). The rate of in cisor retraction was not assessed on the lateral cephalometric radiographs for ethical reasons, i.e., to avoid exposing the pa tients to radiation every 3 weeks. Lateral cephalometric ra diographs were taken using a Planmeca cephalometer (PM 2002 EC Proline ® ; Planmeca, Helsinki, Finland). All subjects were positioned in the cephalostat with the path of Xrays at a right angle to the sagittal plane, the Frankfort plane parallel to the horizontal plane, the teeth in centric occlusion, and the lips relaxed. All radiographs were digitized and traced by the principal researcher. Twelve angular and 5 linear mea surements were made to evaluate skeletal, dental and soft tissue changes following the retraction of 4 upper incisors. The definitions of these variables are given in Table 1, and the cephalometric landmarks, planes and measurements are presented in Fig. 6 and 7. There were no outcome changes after the trial commencement.

Statistical analysis
The IBM SPSS Statistics for Windows , v. 20 software (IBM Corp., Armonk, USA) was used to perform all statis tical analyses. Parametric tests were used, as the data distri bution was deemed normal according to the Shapiro-Wilk tests. In each group, changes that occurred between every 2 time points were calculated. The twosample ttest was used to detect significant differences between the 2 groups regarding the observed changes in the primary and second ary outcomes. The level of significance was set at 0.05. The outcome data assessor was blinded to all patients' data.

Error of the method
To assess the reliability of measurements, 20 models and 20 cephalograms were randomly chosen and reana lyzed 1 month after initial assessment. The pairedsample ttests were used to detect systematic errors, whereas intraclass correlation coefficients (ICCs) were used to evaluate the intraexaminer reliability (i.e., random error).

Patient recruitment, follow-up, entry to data analysis, and baseline sample characteristics
Fortytwo patients were enrolled in this trial and were randomly assigned to either PG or CG. One patient with drew from CG for personal reasons and another patient was excluded from the experimental group, because she did not follow the given oral hygiene instructions thor oughly, which caused acute postsurgical inflammation at the palatal side between upper central incisors. There fore, the data analysis stage included 40 patients who were treated between September 2016 and November 2017. The Consolidated Standards of Reporting Trials (CONSORT) flow diagram is shown in Fig. 8.
The overall sample age ranged from 16 to 31 years (mean age: 19.15 ±3.4 years; 19.8 ±4.17 years for the intervention group and 18.5 ±2.32 years for CG). The difference in mean age between the 2 groups was insignificant (p = 0.231). Ba sic sample characteristics are given in Table 2.

Changes in the retraction distance, the rate of incisor retraction and the time required for retraction
The mean incisor retraction distance was 6.48 ±0.51 mm and 4.21 ±0.38 mm in the experimental and control groups, respectively. In addition, the mean dif ference between the 2 groups regarding the palatal displacement of incisors was significant (p < 0.001) ( Table 3). Statistical significance was found between the 2 groups regarding RIR, which was greater in the experimental group (0.74 ±0.09 mm/week) compared to CG (0.35 ±0.04 mm/week) through the whole obser vation period (p < 0.001) ( Table 4). In the experimental group, RIR was increased by about 53% in comparison with the controls through the whole observation period. The mean difference in the retraction rate between the 2 groups through the entire 12week assessment period was 0.39 ±0.02 mm/week. The mean retraction time for the experimental group was 8.80 ±0.89 weeks and 11.95 ±0.68 weeks for CG. Sta tistical significance was observed between the 2 groups (p < 0.001). Therefore, the upper incisor retraction time was decreased by 27%.

Molar anchorage loss and cephalometric variables
The control group exhibited a significant increase in the mean of molar anchorage loss at all observation time points compared to the experimental group (p < 0.001) ( Table 5). Regarding the cephalometric measurements, changes that oc curred in PG between T 0 and T 4 were compared to those ob served in CG. The angles between upper central incisor and the anterior cranial base, the palatal plane, and the NA line showed a significant reduction in both groups (p < 0.001), and the mean reduction was significantly greater in CG compared to PG (1.52°, 1.4° and 1.53°, respectively; p < 0.001) ( Table 6). Furthermore, a significant increase was observed in the mean interincisor angle in CG compared to PG (−1.11°; p < 0.001) ( Table 6). The upper and lower lips moved backward in rela tion to the esthetic line (Eline) of Ricketts after the end of re traction in both groups, but the movement was insignificantly greater in CG (0.41 mm and 0.33 mm, respectively; p > 0.05) ( Table 6). Regarding the skeletal assessment, the mean val ues of SNA, SNB and ANB insignificantly decreased in the 2 groups after the end of retraction (p > 0.05); however, the mean difference between the 2 groups was also insignificant (−0.34°, −0.2° and −0.14°, respectively; p > 0.05) ( Table 6). All the other skeletal measurements changed insignificantly fol lowing retraction (p > 0.05), with the differences between the 2 groups being also insignificant (p > 0.05) ( Table 6). CG -control group; PG -piezocision group; M -males; F -females; SD -standard deviation; * the χ 2 test; ** the two-sample t-test. T0 -start of observation; T1 -3 rd week; T2 -6 th week; T3 -9 th week; T4 -12 th week; CI -confidence interval; * statistically significant (the two-sample t-test).

Discussion
This singleblinded, twoarm, parallelgroup random ized controlled clinical trial was designed to investigate the efficacy of the piezocision technique in the retraction of 4 upper incisors using sliding mechanics. There was no significant difference between the 2 groups at T 0 in the dis tance between the incisal edges to the third palatal ruga, which assured the similarity of the compared groups.
In order to retract incisors, NiTi closed coil springs were used, as they exert force at a constant level and maintain good oral hygiene. The 5millimeterlong hooks enabled the adjustment of the force level so that the line of action of force could pass as much as possible through the center of resistance of upper central incisors, enabling the bodi ly movement of incisors during retraction. It has been demon strated that the center of resistance during the retraction of 4 upper incisors is located within the mid sagittal plane, approx. 6 mm apical and 4 mm posterior to the line perpendicular to the occlusal plane from the labial alveolar crest of central incisor, and 5 mm apical to the bracket position. 24 The results of the current trial cannot be compared with those of other published papers, since this is the first RCT evaluating the piezocisionassisted retraction of 4 upper incisors with sliding mechanics. In the present study, the piezocision procedure was found to be effective and shortened the overall time required for incisor retrac tion by 27% (approx. 4 weeks). Therefore, a 4week time reduction in a procedure that usually takes 12-16 weeks can be considered clinically important. It has been em phasized that orthodontists would embrace a new ad junctive intervention to accelerate the orthodontic tooth movement if it shortened the treatment time by 20-40%. 25 Accordingly, piezocisionassisted upper incisor retraction may be an acceptable treatment modality to reduce the treatment time. Even though incisor retraction is part of overall extraction treatment, flapless piezocision can be redone to accelerate several stages of the treatment, 26 i.e., the canine retraction phase and the incisor retraction phase. In the current study, we looked at a specific period of the treatment sequence. We are totally aware that we evaluated a small portion of the whole scenario, but when we add the time reduction obtained in the current work with other time reductions revealed by other researchers in other portions of the treatment sequence, an overall re duction in the treatment time becomes apparent.
We found that upper incisors in the experimental group moved backward to a greater extent than those in CG (6.48 ±0.51 mm and 4.21 ±0.38 mm in PG and CG, respectively). This can be explained by the effect of piezocision surgery on reducing dense cortical bone resistance to the orthodon tic tooth movement. The rate of upper incisor retraction in the experimental group during the first 3 weeks was ap prox. twice as big as that observed in CG (0.84 ±0.05 mm/ week and 0.39 ±0.07 mm/week in PG and CG, respective ly). These results indicate a 53% higher retraction rate. This mean rate slightly decreased to 52% and 41% between the 3 rd and the 6 th week and between the 6 th and the 9 th week, respectively. A recently published report on RCT conduct ed by Tunçer et al. addressed pie zocisionassisted enmasse retraction. 17 The researchers reported an insignificantly higher rate of retraction in the experimental group at all time points except for day 90 (i.e., the 13 th week), whereas the current trial showed a superior rate of retraction in the experimental group compared to the controls extending until the 9 th week following surgery. Additionally, Tunçer et al. did not find any significant difference in the retraction speed between the 2 groups. 17 This can be attributed to the insufficient amount of bone injury that was performed in their study, since they restricted their incisions to the labial cortex of the alveolus with no intervention on the palatal side. In the current study, the palatal intervention was un dertaken to initiate RAP at the sites where the teeth were moving forward. 27 Apparently, it is not logical to compare the enmasse retraction of the 6 anterior teeth with incisor retraction, as there are differences between the 2 treatment modalities, such as longer treatment time, higher anchor age requirements and a greater root surface area in the en masse retraction technique. Besides, in the 2 techniques, different orthodontic biomechanics is employed. Despite the conclusion of Tunçer et al. that piezoci sion was ineffective in accelerating the enmasse retrac tion of the maxillary anterior teeth, 17 2 other publications reported opposite findings. Bhattacharya et al. used the socalled 'accelerated osteogenic orthodontics' (AOO) to enhance the enmasse retraction of the upper anterior teeth and found that the treatment time in the experimen tal group was approx. 45% shorter than that in the control group, 28 whereas Sakthi et al. assessed the corticotomy assisted enmasse retraction of the anterior teeth in the treatment of bimaxillary protrusion and showed that the rate of space closure was significantly higher in the corti cotomy group. 29 However, these 2 aforementioned studies are not comparable to our study due to the different number of teeth involved in the retraction technique and the invasiveness of the procedures undertaken in these 2 studies, i.e., the traditional elevation of fullthickness labial and lingual periosteal flaps.
The piezocisionassisted twostep retraction technique is expected to play an important role in our daily prac tice and might be superior to piezocisionassisted en masse retraction, since it has been shown that a signifi cant improvement in the speed of canine retraction can be achieved by minimally invasive methods, 6,16 and the results of the current trial show that also the time of inci sor retraction can be significantly shortened when using these methods.
Regarding molar anchorage loss, CG exhibited a higher rate of molar mesial movement by about 44% compared to the experimental group in the whole observation pe riod. It has been postulated that corticotomy procedures have the advantage of enhancing anchorage by provid ing the differential movement of the anchoring and non anchoring teeth, i.e., the teeth located in the corticoto mized regions tend to move more readily than those in the noncorticotomized regions. 8 Thus, piezocision seems to facilitate the tooth movement with lighter ortho dontic forces and, therefore, reduces the load applied to the pos terior anchoring teeth. These results are in agreement with previous investigations which addressed the accele ration of enmasse retraction. 17,29 All skeletal measurements showed insignificant changes following the retraction of 4 upper incisors. It has been concluded that there are no significant differences in the changes in the skeletal measurements when com paring the accelerated to nonaccelerated retraction of the maxillary anterior teeth. 30 However, in this trial, the significant postretraction changes between CG and PG in the mean values of U1SN, U1SPP, U1NA, and U1L1 indicated that the type of movement of incisors after the end of retraction had a greater translational component in the experimental group, whereas in the controls, the type of movement was controlled tipping. This might be explained by the induced RAP effect, which may have facilitated the expression of the root torque during retraction.
There are some limitations of this study. A high per centage of female patients (77.5%) constituted this sam ple; therefore, intergender differences could not be in vestigated. Patientcentered outcomes, quality of life and the levels of acceptance were not evaluated in this trial. There was no postsurgical followup to investigate the effects of piezosurgical incisions on the labial gingivae (i.e., residual scars) as well as on the periodontal status. The blinding of the patients and the researcher was im possible; however, this could not be a source of bias, since no patientcentered selfassessed outcomes, such as the feeling of pain, discomfort or swelling, were evaluated.
One unexpected postsurgical complication in the ex perimental group was noticed in 1 patient who was ex cluded from the trial. One week after the piezocision procedure, an acute inflammatory response developed between central incisors in the midline at the palatal side. Appropriate care was provided for this patient, but a 2millimeter recession of the interdental papilla in the midline was documented and retraction was resumed after 2 months. In the literature, there are no previous reports on periodontal complications following flapless corticotomy in the retraction of the anterior teeth.

Conclusions
Piezocision proved to be effective in accelerating the retraction of 4 upper incisors, with the rate of retrac tion higher by 53% in PG compared to CG in the whole observation period. The incisor retraction period was shortened by 27% compared to the controls. The results of this study also showed a significant change in the type of incisor retraction movement, which was predominant ly translational movement in PG and controlled tipping movement in CG. The rate of molar anchorage loss was significantly higher in CG compared to the corticotomy group.