Radiographic assessment of different autogenous bone grafts in the alveolar cleft: a retrospective longitudinal study

Purpose: To assess feasibility and maintenance of bone after alveolar cleft reconstructions using graft from iliac crest and mandibular synthesis. Methods: 51 alveolar clefts grafted with iliac crest and 51 ones grafted with mandibular symphysis bones were selected from patients aged between 7 and 12 years. At three (T1) and 12 months (T2) after surgeries, periapical radiographs were performed to measure the height of the grafted bone based on the Bergland scale. Chi-square and Mann-Whitney-Wilcoxon tests compared differences between T1 and T2 according to each bone graft. Results: From the clefts grafted with mandibular symphysis bone, 47 were classi�ed as type I (92.5%) and 04 as type II (7.84%) at T1. At T2, 36 were classi�ed as type II (25.49%) and 02 as type III (3.92%). In the analysis of the clefts grafted with iliac crest at T1, 48 were classi�ed as type I (94.11 %) and three as type II (5.88%). At T2, 37 classi�eds as type I (72.54%), 12 as type II (23.52%) and two as type III (3.92%). There was no statistically signi�cant difference between treatments. Conclusions: It was concluded that iliac crest and mandibular symphysis are adequate areas from which bone grafts can be obtained for reconstruction of alveolar cleft.


Introduction
Orofacial clefts, especially cleft lip and palate, are the most prevalent congenital craniofacial defects as they have an overall prevalence of 0.45 in every 1,000 births [1]. The affected individuals may have di culties in feeding and speaking as well as dental and hearing problems. The treatment of cleft lip and palate involves several craniofacial and dental surgeries associated with speech therapy. The surgical closure of the cleft lip and palate is performed in the rst months of life, provided that the patient is able to undergo the procedures. Reconstruction of the alveolar cleft plays an important role [2].
Although there are multiple indications for bone reconstruction of alveolar cleft, the primary reason is to provide amount and quality of bone in the area of the defect in order to allow adjacent teeth to erupt and orthodontic movement. In addition, bone grafting is indicated for closure of nasal stulas so that the base of the nose can be supported and to allow placement of osseous-integrated implants if teeth are missing [3,4].
In the literature, there are various bone grafting options for reconstruction of alveolar crest [5]. However, none of them was proved to be superior or even equivalent to autogenous bone grafting [3,6]. Iliac crest, mandibular symphysis, skull cap and rib can be the donor sites of autogenous bone, and they are chosen depending on several factors, such as experience and preference of the surgeon, amount of bone and morbidity associated with the bone removed from the donor site [7].
Despite the high rates of resorption and post-operative morbidity, the iliac crest bone is the most used as a grafting material [7,8]. However, some studies show that mandibular symphysis is a viable source of bone graft for alveolar reconstruction [6, [9][10][11][12][13]. Therefore, the objective of the present study was to assess bone feasibility and maintenance after alveolar cleft reconstructions by using bone grafts from iliac crest and mandibular synthesis. Clinical records of both male and female patients, all presenting unilateral or bilateral transforamen clefts, aged between 7 and 12 years old and who underwent bone grafting treatment, in which iliac crest or mandibular symphysis were the donor sites, were included. Patients with insu cient radiographic documentation as well as those presenting syndromic oral clefts were excluded.

Materials And Methods
Prior to surgery for reconstruction of the alveolar cleft, the patients were submitted to surgical closure of the clef lip and palate, which was performed in the rst years of life once they were able to undergo the procedures. The patients remained under multidisciplinary follow-up. In this context, each patient underwent orthodontic treatment when the rst permanent teeth erupted so that possible transverse discrepancies between maxillary and mandibular arches could be corrected. Whenever necessary, the patients underwent maxillary expansion before bone grafting surgery. In this step, the chronological age and dental development of the patient should be taken into consideration. Moreover, bone grating surgery should be performed before the canine erupts in the region of the alveolar cleft.

Surgical Procedure
The same reconstruction technique was used in all patients, regardless of the donor site, which consisted in separating the mouth from the nasal cavity and lling the alveolar cleft with bone tissue. The surgery was performed under general anesthesia with nasotracheal tube placed into the nostril opposite to the cleft in unilateral patients and with orotracheal intubation in the bilateral ones.
After in ltration of local anaesthetic with adrenaline at a ratio of 1:200,000 (Cristália, Itapira, São Paulo, Brazil) in the cleft region and oropharyngeal packing, two vertical incisions were made (one in each edge of the cleft) with a scalpel blade #15 (Solidor, Diadema, São Paulo, Brazil) bilaterally along the prolongations of the sulci. The size of the intra-sulcular incision ranged depending on the size of the cleft, which usually involves two neighbouring teeth on each side of the cleft. Next, sub-periosteal displacement was proceeded ( Fig. 1A-C).
In this moment, oral mucosa was separated from the nasal mucosa, which was cut inferiorly in order to obtain an adequate amount of tissue for suture with Vicryl 4.0 (Ethicon, São Paulo, São Paulo, Brazil) and to reconstruct the nasal mucosa oor with no tension (Fig. 1D). Next, the bone graft was placed in the alveolar cleft and suture performed (Fig. 1E).
Regardless of the origin of the bone graft, immobilisation was performed using less xation material as possible to keep the graft stable in position. Because both symphysis and iliac crest bone grafts were used as block, the form of xation was chosen according to the following order until reaching the necessary stability: 1) under pressure, 2) with 1.5-mm xation screw, 3) with two 1.5-mm xation screws, and 4) with plate and at least three 1.5-mm xation screws (MDT, Rio Claro, São Paulo, Brazil).
Some pointed edges were observed after xation of the graft, which might cause injury to the lining mucosa. In these cases, a spherical diamond bur was used to erode them. Next, relieving incisions were made in the periosteum and the oral mucosa was sutured over the graft with single stitches by using nylon thread 4 − 0 (Ethicon, São Paulo, São Paulo, Brazil) (Fig. 1F). Patients with bilateral cleft were submitted to the same grafting technique, with the premaxilla (anterior dental segment) being repositioned.
The patients remained hospitalised for 24 hours and were medicated with antibiotic, anti-in ammatory and analgesic drugs. After discharge, the patients were instructed on oral care and how to clean the wound, which should be performed with gauze moistened with 0.12% chlorhexidine solution. The medications used by the patients after discharge were amoxicillin 250ml/5ml every 8 hours for 7 days and sodium dipyrone 500 mg/1mL every 6 hours for 3 days. Sutures were removed after two weeks from the surgery.

Radiographic Evaluation
Periapical radiographs were acquired three months after the surgery (T1) to evaluate the condition of the graft and to plan the removal of the xation material. After 12 months from the surgery (T2), the patients were submitted again to periapical radiography, and the radiographs were digitalised with a HP scanner (Scanjet G4050, Hewlett-Packard Company, Palo Alto, CA, USA) at a resolution of 200 ppi and full scale.
The digitalised radiographs were evaluated by two surgeon-dentists (G.S. and D.A.A.M.) based on the Bergland scale, 14 which classi es the height of the grafted bone as type I (normal height), type II (up to ¾ of the normal height, type III (less than ¾ of the normal height) and type IV (no bone bridge) (Fig. 2). The surgeon-dentists were blinded to the identi cation of the patients and the group to which they belonged.
The upper central incisor on the side of the cleft was used as a reference to divide the granted bone into thirds for further classi cation, since T1 radiograph showed that the canine had not yet been erupted. For doing so, a line was drawn along the long axis of the central incisor and the thirds were divided equally and perpendicularly to this line from the enamel-cement boundary. This change was made in order to evaluate the absence of the upper canine and to compensate any error in the standardised acquisition of radiographs, although the radiographic examination had been performed with a positioner and the same X-ray unit (Gnatus Timex 66, Ribeirão Preto, São Paulo, Brazil).
The Dental Imaging Software, version 6.13.3, (Rochester, NY, USA) was used for measuring the full-size radiographs. The two post-operative radiographs (i.e. T1 and T2) of each patient were examined and given scores. They were compared to know whether there was any change in the height of iliac and symphysis bone grafts inserted in the clefts only, since the quanti cation of resorption was not the aim of this study.

Data Collection and Assessment
The following data were collected from the clinical records of the patients: gender, age, graft type, presence of impacted canine, date of the rst and last radiographs, follow-up length and Bergland scale [14]. Each patient received a number so that they could not be identi ed.
It was de ned that 20 radiographs should be evaluated for calibration, thus the rst 20 ones were selected and each examiner performed two evaluations with a 15-day interval between them. The examiners completed a table with data on type of graft and classi cation of the height of the alveolar bone septum. Next, the data were tabulated and submitted to weighted kappa test.

Statistical Analysis
Chi-square and Mann-Whitney-Wilcoxon tests were performed by using the Statgraphics Centurion XV software, version 15.1.02 (StatPoint Technologies, Inc, Warrenton, VA, USA), whereas kappa analysis was performed by using the statistical R software, version 3.0.0, (R Foundation for Statistical Computing, Vienna, Austria). Intra-and inter-rater agreements were calculated by using the intra-class correlation coe cient (ICC).
Because these data are categorical and do not follow a normal distribution (Kolmogorov-Smirnov test, P < 0.001), non-parametric tests were used for analysis.

Results
The results of ICC demonstrated that the examiners were calibrated as intra-rater reliability values were 0.82 and 0.80 for examines 1 and 2, respectively, and inter-rater reliability was 0.81, thus indicating excellent agreement.
A total of 186 patients were found after gathering data during the study period. Of these, clinical records of 40 patients were not found, 17 patients were older than the required age, 28 patients were excluded due to lack of radiographic documentation, 20 patients had incomplete pre-foramen clefts, ve patients had bilateral clefts and absence of teeth in the pre-maxilla, and one patient had complex facial cleft. Therefore, 75 patients were included for study in which 48 had unilateral cleft lip and palate (33 males and 15 females) and 27 had bilateral cleft lip and palate (18 males and 9 females). In the nal, 102 cases of cleft lip and palate treated with each type of graft were evaluated (i.e. 51 with symphysis bone graft and 51 with iliac bone graft). Statistically signi cant differences were found regarding the interval between the acquisitions of radiographs, in which the follow-up length was longer in patients with iliac graft (Table I).
With regard to alveolar clefts grafted with symphysis bone, T1 radiographs showed that 47 were of type I (92.15%) and four of type II (7.84%), whereas T2 radiographs showed that 36 were of type I (70.58%), 13 of type II (25.49%) and two of type III (3.92%). As for alveolar clefts grafted with iliac crest bone, on the other hand, T1 radiographs showed that 48 were of type I (94.11%) and three of type II (5.88%), whereas T2 radiographs showed that 37 were of type I (72.54%), 12 of type II (23.52%) and two of type III (3.92%).
By analysing both variables for differences in the measurements made at T1 and T2, it was found that there were signi cant differences in patients treated with mandibular symphysis bone graft as well as in those treated with iliac crest bone graft, thus indicating presence of resorption in both groups of patients. When the scores of the radiographs showing clefts grafted with symphysis bone were compared to those showing clefts grafted with iliac bone, no statistically signi cant differences were observed between both groups (Table II).
Scores of types I and II were dichotomised as being a success, whereas types III and IV as being a failure, meaning that a new graft was necessary. No statistically signi cant differences were found between the results of the measurements made at T1 and T2 in both groups of patients (Table III).
There was no correlation between time interval of acquisition and results obtained (P = 0.2824).

Discussion
Secondary alveoloplasty with autogenous bone graft is the most commonly procedure used to reconstruct alveolar clefts [15]. Despite the resorption rates and post-operative morbidity, the iliac crest bone is the most used as a grafting material [7,8], thus being the donor site preferred by the majority of the cleft treatment centres [16]. Nevertheless, the results presented in our study demonstrated that mandibular symphysis bone graft, when used to reconstruct alveolar clefts in patients aged between 7 and 12 years old, has a behavior similar to that of the iliac crest one.
Extraction of bone grafting material from the mandibular symphysis is described as being more advantageous than the iliac crest because of lower morbidity rate, shorter hospitalization and absence of skin scar [17], in addition to better bone incorporation as both mandible and maxilla have the same ectomesenchymal origin and intra-membranous ossi cation process [6]. Mandibular symphysis has a lower amount of bone available for grafting [6,7,17,18], but it is a donor site near the operative eld, which reduces the surgery and anaesthesia times [17].
Data on post-operative complications in the donor graft sites were not collected in the present study.
Possible complications associated with the extraction of bone graft from the mandibular symphysis are lower lip ptosis, lesion in teeth adjacent to the donor site and lesion to the mentual nerve. On the other hand, extraction of bone graft from the iliac crest results necessarily in skin scar and the patients may also complain of post-operative pain, di culty in walking [17,19], sensorial loss, seroma, haematoma, fracture, abdominal herniation, contour defects, peritoneal perforation and keloids [8,20].
Previous studies have determined that the age of 9-12 years old, i.e. before eruption of the canines, is the most suitable moment for performing alveolar grafting [14,17,21,22]. Although our results were positive for patients with similar age (7-12 years old), Dissaux et al. [23] performed a tomographic analysis to show that alveolar bone grafts are successful when children are surgically treated earlier (i.e. around 5 years old) compared to those aged around 10 years old. Therefore, the upper lateral incisors can erupt through the grafted bone, which ensures better results in terms of residual bone height.
The results demonstrated that bone grafts from iliac crest and mandibular symphysis interfered with the upper canine eruption at the same frequency, that is, 19.6% in both groups of patients. These values were close to those reported by Sindet-Pedersen and Enemark [17], who found 15% and 20% of impacted upper canines in patients submitted to mandibular symphysis and iliac crest grafts, respectively. Sindet-Pedersen and Enemark [22] studied a sample of 28 patients with mandibular symphysis grafts and showed that all patients had the canines erupted in the grafted area, but they did not describe whether the teeth were surgically exposed or orthodontically pulled. Desai et al. [24] evaluated the eruption stage and changes in the position and pattern of canine eruption after alveolar graft from iliac crest in 30 patients aged between 9 and 13 years old. The authors found that the canine did not erupt in only 20% of the cases, demonstrating that eruption was satisfactory and the root grew continuously in the region of the grafted alveolar cleft. The risk of impacted canines is increased if they are unfavourably, either vertically or laterally, positioned before the surgery as they erupt continuously in the same graft angle.
The literature has long been evaluating the resorption of autologous bone grafts in alveolar clefts. Sindet-Pedersen and Enemark [25] compared radiographically the results of alveoloplasty with iliac crest grafting performed in patients before and after the eruption of upper canine. The authors used an approach similar to that in our study, but they divided the grafted area into quarters rather than into thirds. Of the patients with the age group and cleft classi cation similar to ours, they found 89.4% of type I, 9.67% of type II and 1.07 of type III. Despite the change in the classi cation, the results are compatible to those found at T1.
Bergland et al. [14] evaluated periapical radiographs of patients submitted to iliac crest grafting after at least 12 months from the surgery. Of the 143 patients aged 8-11 years old who underwent grafting before eruption of the canine in the area of cleft, 96% of the cases were successful, which is a result similar to ours. Enemark et al. [21] repeated the methodology used in the study by Sindet-Pedersen and Enemark [25], but with a 7-year follow-up. Among the 95 patients with cleft classi cation and age similar to those of our study, 71.57% were classi ed as type I, 23.15% as type II, 4.2% as type III and 1.05 as type IV. Likewise, despite the different cleft classi cation, it should be noted that their result was worse than ours at T2. Sindet-Pedersen and Enemark [22] evaluated the radiographs of unilateral patients aged 8-15 years old for 8 months, on average, after mandibular symphysis bone grafting. Bergland scale was used and all 28 patients had their cleft categorised as type I (26 patients) or type II (2 patients). Sindet-Pedersen and Enemark [17] compared the radiographic images of unilateral cleft patients who were submitted to alveolar reconstruction. Twenty patients received the iliac crest bone graft and 20 received mandibular symphysis bone graft. The age of the patients ranged from 8 to 13 years old and the followup period lasted 19 months, on average. The alveolar clefts on the radiographs were classi ed in the same way as earlier and the authors found no signi cant difference in the bone height between both treatments, showing results similar to ours. One factor making it di cult to compare both groups of patients relies on the fact that the maxillary transverse discrepancies in patients with mandibular symphysis bone graft were not corrected prior to the surgery, differently from those patients with iliac crest bone graft. Such heterogeneity did not occur in our sample, thus making our assessment more reliable.
Trujillo et al. [12] used CBCT to compare the bone formation in cleft patients with mean age of 10 years old submitted to alveolar bone grafting using recombinant human bone morphogenetic protein-2 (rhBMP-2) and autogenous grafts obtained from iliac crest and mandibular symphysis. The higher mean amount of bone neoformation was achieved in patients with iliac crest bone graft (85.4%), followed by those with rhBMP-2 (81.22%) and with mandibular symphysis bone graft (80.56%). However, there was no statistically signi cant difference, which corroborates our ndings. Lundberg et al. [26] used the Bergland scale to assess the results of the alveolar bone grafting with iliac crest bone graft in 91 patients with mean age of 9.2 years old after seven years from the surgery. The authors found a high rate of success (91%) and reported that a poor oral hygiene increased signi cantly the risk of surgical failure. This suggests that perioperative measures to maintain a good oral health can reduce such risk.
The rates of success found in the present study are compatible to the literature. However, it is di cult to compare it with other studies because of the different classi cation criteria used and differences in previous treatments before the grafting procedure. Notably, the majority of the cleft lip and palate centers perform alveoloplasty with bone from the iliac crest instead of mandibular symphysis [7,16]. Nevertheless, the mandibular symphysis was shown to be a viable source of graft as the long-term results are compatible with those using iliac crest bones, in addition to the lower rates of complication [17,19,20]. Therefore, among the two options presented here, the choice of the graft donor site should be made on the basis of the necessary and available amount of bone.
Grafting procedures using bones from iliac crest and mandibular symphysis were considered to be viable when used for reconstruction of alveolar clefts in patients aged between 7 and 12 years old, since there were no statistically signi cant differences in the bone formation.  Type IV 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) * p value for Chi-square test.   Radiographic images showing mandibular symphysis and iliac crest bone graft used for reconstructing alveolar cleft to be classi ed according to Berglan, Semb e Abyholm Scale [14].