Effects of Concentrated Growth Factor on Gingival Thickness in Periodontal Accelerated Osteogenic Orthodontics: A 6-Month Randomized Controlled Trial

Background: Earlier studies have not given clear results of concentrated growth factor (CGF) on gingival thickness (GT) in periodontal accelerated osteogenic orthodontics (PAOO). This randomized controlled trial aimed to evaluate the effects of CGF on GT in patients with thin gingival phenotype undergoing PAOO. Methods: 44 patients with 264 surgical sites were recruited at baseline. All patients were randomly allocated to either a control (collagen matrix memberane) or test (CGF) group and received PAOO. GT, gingival height (GH), buccal alveolar bone thickness (BT), and buccal alveolar bone height (BH) were evaluated depending on cross-sectional CBCT images at t0 (before surgery) and t1(6 months after surgery). Results: GT were increased in both groups at t1 compared to t0. Yet, higher values were observed in the test group (from 0.95±0.26 to 1.29±0.34mm) compared to the control group (from 0.92±0.25 to 1.01±0.18mm) (p < 0.05). Moreover, in the intergroup comparison, GT at t1 in the test group was signicantly higher compared to the control group (p < 0.01). Furthermore, the GT of central incisors, lateral incisors and canine teeth all showed signicantly changes compared with baseline and the test group showed higher increase (p < 0.01). No statistically signicant difference were found in GH, BT, BH and all clinical parameters between two groups at t1 (p > 0.05). Conclusions: Within the limitation of this study, gingival thickness could be increased by using CGF in PAOO for the patients with thin gingival phenotype.

using collagen matrix memberane could not increase the gingival thickness (GT). Patients with thin gingival phenotype are more likely to suffer a gingival recession after surgery.
The connective tissue graft (CTG) with a coronally advanced ap (CAF) procedure is the standard treatment approach for obtaining gingival tissue. It can solve the gingival recession problem, leading to predictable treatment outcomes. 16,17 Nevertheless, this technique has several disadvantages, including insu cient palatal donor tissue and additional risks due to the presence of a second surgical site. 18 Furthermore, CTG and CAF can not increase the alveolar bone volume of patients with skeletal deformities. Furthermore, the therapy of CTG and PAOO will extend surgical procedure time and increased bleeding and pain postoperatively 19 . Therefore, modi ed PAOO and other methods are needed to treat the problems of thin gingival phenotype in patients with skeletal deformities.
Growth factor therapy has recently gained more attention since it has been established; it can improve the effect of periodontal surgery. 20 Concentrated growth factor (CGF) has been used as a promoter for repairing intra-bony defects, fat graft, and sinus augmentation during tissue regeneration. 21,22 Previous study found that CGF and CAF can improve the gingival recession by increasing the width of keratinized gingiva (KTW) and GT. 23 Our previous study has also demonstrated that CGF promotes gingival regeneration through the AKT/Wnt/β-catenin and YAP signaling pathways. 24 Although tissue regeneration using CGF yields promising clinical and experimental results, so far, no study evaluated the effect CGF in combination with PAOO regarding the GT in patients with skeletal deformities.
The aim of this randomized controlled trial was to evaluate the clinical effectiveness of CGF used in PAOO on GT in patients with skeletal deformities and to compare other clinical parameters with collagen matrix membrane.

Trial design
This study was designed as a randomized controlled clinical trial to the effects of compare collagen matrix memberane (control group) and CGF memberane (test group) in PAOO. Forty-four consecutive patients from the Department of Oral and Cranio-Maxillofacial Surgery, Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, who met the inclusion criteria were enrolled from February 2018 to April 2020. The study was registred at Chinese Clinical Trial Registry (ChiCTRINR17013346). The study protocol in this study was approved by the Ethics Committee of Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine (SH9H-018-164-T122).

Sample size
Before the initiation of the study, the power analysis for sample size calculation was performed. According to the results of our preliminary experiment, a minimum of 20 patients was needed for each group so as to obtain 80% power and to detect a minimum clinically signi cant difference in GT after considering possible dropouts 23 .

Patient Inclusion: Informed Consent, Randomization and Allocation concealment
After the purposes, risks, bene ts, and monitoring of the study were explained, the patients were invited to sign an informed consent form. Patients were randomly assigned to the control group (collagen matrix membrane) or test group (CGF). Sealed envelopes were used to perform randomization with an equal number of envelopes for every group. Group allocation was revealed just before surgery and remained blinded for the evaluating investigator for data collection, analysis, and processing during the project's analytical stage. The location and distribution of subjects are shown in Figure 1.

Preoperative therapy
All the patients received full-mouth scaling 2 weeks before surgery. After clinical evaluations, all patients underwent CBCT with the same settings at t0 (before surgery). During the scan, the lip tissue and the gingival tissue were separated, thereby visualizing periodontal hard and soft tissue on every crosssectional slide. Then, the evaluating investigator would measure the GT preoperatively to determine whether the patients could participate in the following study.
Interventions CGF membrane preparation CGF was prepared from human venous blood obtained from the subjects in test group. All subjects enrolled in this study gave informed consent. Venous blood was collected in two sterile 9 ml tubes and was immediately centrifuged in a special centrifuge (Medifuge, Slifadent srl, So a, Italy) for 13 min. As shown in Supplementary Figure 1, the CGF layer was squeezed to obtain CGF membrane using a matched machine.

Surgical procedure
As presented in Figure 2, the surgical technique for PAOO performed in mandicular anterior teeth (33-43) included 4 steps: raising of ap, decortication, grafting, and closure. Brie y, after the cortical incision was performed, the bone graft (Bio-oss, Geistlich Biomaterials, Wolhusen, Switzerland was implanted in the thin area of the cortical bone. CGF membrane was then used to cover the bone graft. Consequently, the incision was sutured without tension. The control group received collagen matrix membrane (Biogide, Geistlich Biomaterials,Wolhusen, Switzerland instead of CGF membrane (Supplementary Figure 2).

Postoperative care
After surgery, the oral cavity was kept clean and oral hygiene instructions were provided at each postoperative visit. Two weeks after the operation, the patient revisited the hospital, and the doctor checked whether there was wound dehiscence, implant infection, and gingival recession; after 3 months, the gingival conditions were evaluated; after 6 months, the CBCT was acquired to evaluate the changes of GT and other clinical parameters by comparing them with the preoperative ones.

Data collection
Gingival thickness (GT) The GT and its stability over 6 months was the primary outcome of the study. The GT was evaluated using cone-beam computed tomography (CBCT: 120 kV, 5 mA, exposure time 26.9s and voxel size 0.125 mm). Brie y, the sagittal plane of CBCT slide perpendicular to the long axis of the tooth was selected and the long axis of the teeth was identi ed. The point was then chosen perpendicular to the long axis at the following level: 2 mm apical to the gingival margin. GT was measured at the level from the junction between gingival and alveolar bone tissue to the outline of the gingival tissues (Supplementary Figure 3).

Gingival margin height (GH), Buccal alveolar bone thickness (BT), and Buccal alveolar bone height (BH)
For each tooth, the GH, BT, and BH were measured at t0 and t1 perpendicular to the tooth's long axis. GH was measured from the gingival margin to the tooth cusp; BT from the tooth surface to the bone-soft tissue interface (2 mm apical to the alveolar ridge); BH from alveolar ridge to the level of tooth cusp (Supplementary Figure 3).

Pain and swelling scores
Pain and swelling were evaluated at 2 weeks, 3 months months, and 6 months after surgery. The pain was evaluated using the following score: 0: no pain; 1: mild pain; 2: moderate pain; 3: severe and continuous pain; 4: severe and continuous pain with changes in blood pressure and pulse. The swelling was evaluated using the following score: 0: no swelling; 1: mild swelling; 2: moderate swelling; 3: severe swelling. 25,26 Healing index (HI) The HI, including tissue color, bleeding at palpation, epithelization of incision margins, and presence of Absorbable thread dissolution, was evaluated at 2 weeks, 3 months, and 6 months after surgery. A score of 1 indicated very poor improvement, and 5 indicated an excellent recovery. 25,26 Statistical analysis In this study, all data were performed using statistical software (version 22.0; SPSS Inc., Chicago, IL, USA). First, the data were tested for normality using the Kolmogorov-Smirnov test. The Mann-Whitney U test was used for the intergroup evaluations after the normality of data failed. When making intragroup comparisons, the Wilcoxon signed-rank test was used to assess changes in the double measurements. The Friedman test was used to assess changes in the triple measurements. Descriptive statistics such as mean±standard deviation and discrete numeric variables (minimum-maximum) were calculated for the clinical and radiographic parameters. Intra-and inter-assessor reliability was assessed on the basis of ve cases that were randomly selected for duplicate registration of GT. A p value < 0.05 was considered to be statistically signi cant.

Results
Patient Characteristics 44 patients were recruited in this study. All 40 patients completed the study and 4 were lost to the followup. All the operations were performed uneventfully. No adverse events related to surgicial techniques were recorded in the 6 months follow-up (Supplementary Figure 4). In the control group, 10 males and 10 females with a mean age of 26 years in the control group, and 11 males and 9 females with a mean age of 25 years in the test group. Both the control and test group included 20 mandibular central incisors, 20 lateral incisors, and 20 canine teeth. The intergroup and intragroup comparisons between the t0 and t1 are shown in Table 1, Table 2 and Supplementary Table 1.
Primary outcome GT The intragroup and intergroup comparisons of radiographic parameters are presented in Figure 3, Table  1 and Supplementary Figure 5. No signi cant differences in GT were found between the two groups at t0 (p=0.508). As shown in Figure 4, at t1, the mean GT values in the test group were higher compared to the control group (in the test group, the values increased from 0.95±0.26 to 1.29±0.34 mm; in the control group, the values increased from 0.92±0.25 to 1.01±0.18 mm; p < 0.01). In the intergroup comparison at t1, GT in the test group was signi cantly higher compared to the control group (p < 0.01).
To further observe every surgical site's change, the GT values of central incisors, lateral incisors, and canine teeth were evaluated ( Table 2). The GT values of all teeth were not signi cantly different in the intergroup comparisons at t0. In the control group, the mean GT values of central incisors were increased from 0.92±0.26 to 1.05±0.17 mm (p=0.002 < 0.01), while the canine teeth were increased from 0.94±0.23 to 1.01±0.18 mm at t1 (p=0.041 < 0.05). However, the lateral incisors did not show a signi cant change in terms of the mean GT values compared to t0 (from 0.90±0.26 to 0.96±0.20 mm) (p=0.052 > 0.05). In the test group, a greater increase was observed at t1 compared to t0 for all teeth (P<0.05). Moreover, in the intergroup comparison at t1, GT in the test group was signi cantly higher compared to the control group.
Secondary outcome GH There was no statistically signi cant difference between control and test group in the intergroup comparison at t0 (p=0.706 > 0.05; Table 1) and t1 (p=0.138 > 0.05). In the control group, GH values were increased from 8.27±0.71 to 8.26±0.65 mm (p=0.074 > 0.05). However, a slight increase was observed at t1 compared to the values at t0 for the test group (from 8.27±0.66 to 8.23±0.68 mm; p=0.003 < 0.01).

BT
In the intergroup comparison at t0 and t1, there was no statistically signi cant difference between the control and test group (p > 0.05; Table 1). After 6 months, BT increased from 1.02±0.31 to 1.42±0.44 mm in the control group (p < 0.01) and from 1.04±0.36 to 1.42±0.52 mm in the test group (p < 0.01).

BH
In the intergroup comparison at baseline and 6 months, there was no statistically signi cant difference between the control and test group (p > 0.05; Table 1). After 6 months, BH increased from 13.16±0.64 to 12.07±0.70 in the control group (p < 0.01) and from 13.21±0.74 to 12.09±0.60 in the test group (p < 0.01).

Clinical outcomes
Between-group comparisons, statistical analysis showed no signi cant difference in the mean pain values between the control and test groups at 3 and 6 months post-operatively (p > 0.05; Supplementary Table 1). However, when comparing the mean pain values at 2 week observation period, the test group had a signi cantly lower value compared with the control group (p=0.024 < 0.05).
There was no statistically signi cant difference in terms of mean swelling and HI values in the intergroup comparison at 2 weeks, 3 months, and 6 months. In the intragroup comparisons, the decrease in the mean pain and swelling values in the following days was statistically signi cant in both groups when compared with the values at 2 weeks (p < 0.01). Moreover, mean HI values were signi cantly increased in both groups at 3 months and 6 months when compared with the second week (p < 0.01).

Discussion
Poor gingival condition was one of common problems of skeletal deformities. Anatomical factors, malocclusion, and dental compensation are the main causes of poor gingival conditions. 27 Hence, gingival risk assessment and multidisciplinary treatment are of great importance for clinicians when selecting best treatment techniques to avoid treatment failure. The implementation of PAOO could increase the alveolar bone volume, accelerate the movement of orthodontic teeth, and increase preorthodontic stability. 28 Although the collagen matrix membranes used in PAOO have many bene ts, including soft tissue in-growth exclusion in the defect region and stabilization of the grafting material, they have no clear positive effect on the increase of GT. 29,30 In addition, gingival recession and root exposure often occur in patients with thin gingival phenotype after PAOO. Hence, increasing GT and improving gingival stability in patients with skeletal deformities are important preoperative orthonathic treatment issues.
An autologous gingival graft is considered the gold standard treatment approach for gingival recession coverage. However, this approach has various drawbacks, such as donor-site damage, limited palatal mucosal tissue, and poor post-operative outcome. 31 Although CAF can quickly increase the GH to some extent, it does not solve the problem of thin gingival phenotype. Therefore, whether modi ed PAOO techniques could increase the GT and GH while increasing the amount of alveolar bone volume needs further exploration.
Autologous platelet concentrates containing different growth factors have an essential role in tissue regeneration. 32 CGF, which can be found in a liquid state or as gel or membrane, could adapt to different clinical needs. CGF can enhance bone regeneration over a longer period by promoting cell proliferation, migration, and vascularization. 33,34 Nevertheless, the underlying mechanisms of CGF-mediated gingival regeneration remain unclear. In our previous study, we found that CGF may directly promote the proliferation and migration of GMSCs in vitro and enhance microcirculation in vivo. Based on these results, in the present study, we further examined the effect of CGF on gingival thickness and other radiographic and clinical results after PAOO.
Our data indicated that either CGF or collagen matrix membrane yielded short-term stable results. Nevertheless, the increase in GT was signi cantly higher in the test group than the control group after 6 months, which is consistent with a previous study that reported an increase of 0.06±0.34mm for the CAF group 0.32±0.10mm for the CAF and CGF group (p = 0.000). Aroca et al also showed a signi cant increase in GT in the MCAF+PRF group. 35 The higher increase in GT in the test group may be because CGF contains much larger, denser, and richer growth factors and brin matrix. 36 Taken together, these results con rmed that CGF could effectively increase GT. To the best of our knowledge, this is the rst study that investigated the effect of CGF on GT of different surgical sites. As presented in Table 2, the GT values of all mandibular anterior teeth in the test group at t1 were signi cantly different compared to t0. In the control group, the mean GT values of the central incisors and canines at t1 were statistically signi cant compared with t0, while the lateral incisors did not show a signi cant change in terms of the mean GT values. This may be related to the complex anatomy and surgical sites.
The other major nding in the present study was the increase in gingival height (GH) in the test group. GH is an important factor for the gingival function and aesthetic, responsible for reducing the incidence of periodontal disease and the maintenance of periodontal health. A previous study suggested that CGF could increase KTW by 0.58mm, which was bene cial in terms of attachment gain and increased longterm stability. 23 In this study, the mean GH in the test group showed an increase of 0.04 mm at t1 compared to an increase of 0.01 mm in the control group. The differences from the previous study may lay in different surgical and study designs as we only included the anterior mandibular teeth. Muscles in the maxilla and mandible have different tensile strength, and the differences in ap thicknesses may produce different results. Also, molar teeth have a larger surgical eld and more complex anatomy, which was not considered in the present study. Hence, whether CGF could affect the increase of GH remains to be explored. However, no gingival recession during the PAOO occurred regardless of using CGF or collagen matrix membrane.
The increases in BT and BH were observed in the control and test group. When comparing the mean BT values at t1 with t0, higher values were observed in both groups. Still, there were no signi cant differences between the groups. Similar to BT, apart from the signi cant differences over time within each group, there were no signi cant differences between the groups. Since bone augmentation was performed in both groups, there was a logically signi cant difference over time in BT and BH, which helped solve the problem of root exposure. Previous studies have reported that CGF has a positive effect on bone regeneration and wound healing, as it stimulates the broblast-DNA synthesis and promotes the expression of chemotactic factors, thus attracting osteoblastic cells to the wound area. 37 This point could not be re ected in our present study due to the in uence of bone augmentation during PAOO, which needed further histological examination.
CGF can accelerate soft tissue healing because of its growth and brin network structure. Several studies have indicated that PRF could decrease post-operative pain, increase the mean HI values, and accelerate healing during the rst weeks. 38,39 In this study, a signi cant difference in the mean pain values was observed between the control and test groups at two weeks after surgery. However, there was no signi cant difference in mean swelling and HI values between control and test groups when performing the intergroup comparison. PAOO was a relatively large periodontal surgery that involved at least six teeth surgical sites. Avoiding excessive stripping of mucoperiosteum and reducing operation time may be essential to decrease the pain, swelling, and HI values. Therefore, post-operative discomfort, including pain and swelling after surgery, is common, especially in the rst three days. In this study, no signi cant difference in terms of swelling and HI was observed between the test and control group at 2 weeks after surgery.
This study was designed as a randomized and controlled clinical trial of patients with skeletal deformities. Our data suggested that CGF could increase the GT of mandibular anterior teeth after PAOO to some degree, which is consistent with our previous basic research results. It has to be pointed out that skeletal deformities may increase surgical di culties compared with normal patients, as the surgical eld has more anatomic variations. Moreover, the thin gingival phenotype is common in patients with skeletal deformities. The gingiva is more prone to recession after PAOO in these patients. Further, The use of CGF membrane can repair bone defects with bone grafting at the same time. However, when interpreting the present study results, the following limitations should be taken into account. First, the 6-month postoperative measurement period was chosen to evaluate the gingival and alveolar bone indicators' stability after PAOO. The long-term evaluation is of utmost importance and may have a relevant impact on clinical decision-making and promoting the application of this novel strategy. Second, this study only included mandibular anterior teeth. The effects of CGF on posterior and maxillary teeth need to be investigated by further clinical studies because of the anatomic variations of surgical elds that may affect the results. Third, although CBCT images are characterized by high spatial resolution, soft tissues' visibility has limitations. Hence, according to these outcome variables, further RCTs should be designed to verify reported ndings and establish an optimal treatment approach.

Conclusion
Within the limits of the study, it can be concluded that CGF could increase gingival thickness in patients with skeletal deformities who underwent PAOO. PAOO combined with CGF also seems to be an effective option for BT and BH improvement and alveolar bone reestablishment. The study protocol in this study was approved by the Ethics Committee of Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine (SH9H-018-164-T122) and all patients signed informed consent form. All participants were informed of the research procedure and signed the participation consent agreement.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.  The measurements of CBCT images of central incisors, lateral incisors and canine teeth at t0 ( preoperation) and t1( 6 months postoperation) in the test group. 43, 33: canine teeth; 42,32: lateral incisors; 41,31: central incisors.