Alveolar ridge regeneration of damaged extraction sockets using deproteinized porcine versus bovine bone minerals: A randomized clinical trial

Abstract Backgrounds Clinical benefits in bone grafting of intact extraction socket have been widely known, but limited evidence is available for the procedure in damaged extraction sockets due to periodontal disease. Purpose This study aimed to determine the dimensional alteration of alveolar ridge following bone grafting of damaged extraction sockets, and compare the outcomes of using deproteinized bovine (DBBM) versus porcine bone mineral (DPBM) in the damaged sockets. Materials and Methods One hundred patients (n = 50 for each group) with periodontitis‐induced damaged extraction socket were included in this randomized, single‐blind trial. After removal of tooth and granulation tissue, sites were grafted with either DBBM (DBBM group) or DPBM (DPBM group), and covered with collagen membrane. Linear/volumetric analyses of hard and soft‐tissue dimensions were performed on reconstructed/superimposed computed tomography and scanned cast images, taken immediately and 4 months after surgery. Results The two groups showed comparable hard tissue augmentation with minimal reductions in the grafted volume, as well as in vertical (1.22 ± 2.16 and 1.45 ± 1.92 mm for DPBM and DBBM group, respectively) and horizontal (1.43 ± 3.40 and 1.83 ± 2.85 mm on the central section, respectively) dimensions at 4 months after surgery. However, several cases showed large variations in maintenance of the grafted volume. None of the measured parameters in hard and soft tissue dimensions differed significantly between DBBM and DPBM sites. Conclusions DBBM and DPBM can comparably augment damaged extraction sockets with minimal postoperative reduction of the grafted volume. However, the large variations in the results should be further evaluated for application in routine dental clinics.


| INTRODUCTION
The clinical procedure of grafting bone substitutes in the extraction tooth socket or peri-implant circumferential defect around an immediately placed implant is based on preclinical studies that have shown dimensional shrinkage of the alveolar ridge after tooth extraction and the dimensional compensation achieved by placing slowly resorbed biomaterials in the extraction sockets. 1,2 Clinical studies have also demonstrated maintenance of the horizontal width of the alveolar ridge and favorable esthetic results achieved by bone grafting of extraction sockets and immediate implant placement with filling of bone gaps. 3,4 However, the preclinical and clinical studies supporting these techniques were all based on experimental models or clinical indications of a single tooth gap and an intact extraction socket in the anterior region.
The above-mentioned indications are uncommon in the clinical setting, with a damaged extraction socket (missing one or more bony walls) occurring frequently due to periodontal and endodontic problems. This situation has prompted studies into extending the indications of extraction-socket grafting, 5,6 as well as the use of biomaterials in these techniques. 7,8 Barone et al. demonstrated that the horizontal dimension of the alveolar ridge could be maintained after grafting biomaterials with coverage by a collagen membrane in buccal-bone-deficient extraction sockets in a single cohort study. 5 Lee et al. also showed dimensional augmentation radiographically and successful new bone formation histologically by grafting bone substitutes and coverage with a collagen membrane in damaged extraction sockets in a preclinical animal model. 6 These findings support the grafting of biomaterials even in cases of damaged extraction sockets with a deficient bony wall, which can be caused by periodontitis or a combined endodontic and periodontal lesion.
Most previous studies of ridge preservation techniques showed successful dimensional preservation using deproteinized bovine bone mineral (DBBM), whereas the use of other grafting materials having characteristics of rapid resorbability (eg, autogenous bone) resulted in relatively poor clinical outcomes in terms of dimensional alterations. 9,10 DBBM has, therefore, been a gold-standard biomaterial for managing extraction sockets over the past 10 years based on the findings of both clinical and preclinical studies. However, recent clinical and preclinical studies have evaluated other types of grafting biomaterials as candidates for ridge preservation in order to minimize dimensional alterations of the alveolar ridge, and they have demonstrated comparably successful results both radiographically and histologically. 11 These biomaterials can provide space maintenance that facilitates bone formation even when the thin buccal bone plate is resorbed, and can also compensate for dimensional alterations. 1,12 Xenografts from swine and equine origin were also introduced in the field of implant dentistry 13,14 due to a potential risk of bovinespecific disease transmission by prions. 15 Among these, xenografts from porcine were used in the clinical trials and animal studies based on structural similarities to human bone, with the report of successful bone formation and volume maintenance comparable to DBBM. 16 The aims of this study were (i) to characterize dimensional alterations in alveolar ridge following grafting deproteinized bone mineral combined with coverage by a collagen membrane in damaged extraction sockets, and (ii) to compare the outcomes of using DBBM versus deproteinized porcine bone mineral (DPBM) in damaged extraction sockets in a randomized clinical trial. Patients who were planned to receive tooth extraction due to severe periodontitis were included in the present trial. The patients were required to be at least 20 years old, able to understand the protocols, and provide voluntary informed consent to participate for this study. Patients with the following characteristics were excluded: smoking more than 10 cigarettes daily, uncontrolled diabetes, hypertension, or coagulopathy, osteomyelitis, any history of alcoholism, taking steroids or immunosuppressors within the previous 2 weeks, allergic reactions to porcine bone or bovine bone, any history of radiotherapy, or taking bisphosphonate medication.

| Sample-size determination
The required sample size was calculated using G*Power 3.1 software 17 for a comparison of two experimental groups with a one-sided alpha level of 5% and a statistical power of 90%. Based on previous results for the mean and standard deviation of horizontal changes in the ridge dimensions in two experimental groups, 18 the effect size was calculated as 0.74. About 40 sites per group were required in the present study, and so 50 sites per group were used on the assumption of a dropout rate of 25%.

| Randomization and blinding
Informed consents were obtained from all included participants after providing a detailed explanation of the protocols and the benefits of risks related to participation. Each participant was given an enrollment number and then randomly assigned to the DBBM or DPBM group using a sequentially numbered sealed-envelope protocol. In brief, the participants were given enrollment-number-matched sealed envelopes containing group assignments based on block randomization.
At the next visit for experimental surgery, each participant was allocated immediately before applying the biomaterials after tooth extraction, by opening a sealed envelope. This trial was designed using single blinding, with all researchers blinded before opening the envelope, but then unblinded thereafter. Participants were blinded to their allocation. However, researchers were reblinded when performing measurements and analysis: a clinical research coordinator marked a concealed number on the radiographic and histologic data rather than the patient's information prior to performing the measurements.

| Surgical protocols and timetable
Preoperative treatment for inflammation and infection control was performed at the first visit for screening, which involved subgingival scaling/root planning and the application of local antibiotics (minocycline hydrochloride; Periocline, Sunstar, Osaka, Japan) to the apical end of the lesion. These were also applied to other sites involved with periodontitis to subside pathologic signs and symptoms. Broad-spectrum antibiotics were administered for 5-7 days before the extraction.
On the second visit, a dental impression was taken for fabricating a cast, and the tooth was extracted gently under local anesthesia. Thorough curettage of granulation tissue was performed in the extraction socket so as to expose the bony surface in all regions of the socket.
An incision was made from the adjacent dental papilla to an adjacent tooth to elevate a mucoperiosteal flap over the margin of the lesion, and the flap was elevated. After the exposed bony surface was clearly confirmed by performing further curettage, the DPBM and DBBM were grafted according to the random group allocation to fill the extraction socket so that the contour of the ridge was consistent with the adjacent ridge shape; the biomaterials were grafted up to the level at the extended outlines between the most coronal region of mesial and distal bony wall (vertically) and between the outermost regions of the adjacent alveolar ridge (horizontally). Collagen membrane (Bio-Gide, Geistlich Pharma) was applied to cover the defect margin and the grafted biomaterials, and the mucoperiosteal flaps were repositioned and fixed with a 6-0 nylon suture (Monosyn, B. Braun, Aesculap, Center Valley, PA, USA). Primary closure was achieved throughout the incised region except over the entrance of the extraction socket, which exposed the underlying collagen membrane but was also sutured to ensure wound stability. Gentle pressure was applied buccolingually on the grafted area for 3-5 minutes, and postoperative computed tomography (CT) was taken. Broad-spectrum antibiotics were administered for 7 days. The suture was removed after a healing period of 1 week, and checkup visits were performed at 1 and 3 months after surgery ( Figure 1A-J).
CT and a dental impression were taken, and dental implants were placed 4 months after surgery. The bone was prepared for the implant using a trephine bur, and a tissue sample within the trephine bur was analyzed histologically (the histologic results are reported elsewhere).

| Outcome variables 2.5.1 | Linear measurements
The alveolar ridge dimension was analyzed quantitatively on the CT images obtained immediately and 4 months after surgery (designated as the first and second CT images, respectively), and these images were automatically superimposed using computer software (OnDemand 3D, CyberMed, Seoul, Korea). The horizontal width and vertical height were measured as the primary outcomes in the same cross-sectional plane as the superimposed view. The acquired CT images were processed in DICOM format, and the first and second CT images were fused automatically based on neighboring anatomic structures such as teeth, the mandibular border, and the sinus floor.
The resulting superimposed CT image was aligned with the long axis of the adjacent tooth, and the grafted area was marked based on differences in radiopacity relative to the preexisting neighboring bone ( Figure 3A-E). An imaginary bottom plane was drawn that included the lowermost point of the grafted area on the first CT image and was perpendicular to the previously set axis. The vertical height was measured as a distance from the bottom plane to the uppermost point of regenerated alveolar ridge on the superimposed CT image. The horizontal width was measured at three lines that divided the grafted area equally into four sections ( Figure 4A and B). The changes in the vertical height and horizontal width were calculated by subtracting the measurements from the first and second CT images.

| Volumetric measurements
The secondary outcomes were changes in the grafted volume, papilla height, and soft-tissue dimensions. The grafted volume within the damaged extraction socket was measured on the reconstructed CT images taken immediately after surgery. The volume change was also measured on the superimposed CT image. An imaginary cube was set that included the bottom plane and which contacted the most-prominent contour of the grafted area, and the volume change was FIGURE 1 Clinical and radiographic photographs of the representative case in the present clinical trial (DPBM group). Initial periapical radiograph A, showed severe periradicular radiolucency related to vertical root fracture, and preoperative photograph B, revealed gingival inflammation around the tooth. After mucoperiosteal flap elevation and tooth removal (C and D), bony destruction extending to buccal/lingual bony walls and an interradicular septum could be found. Test bone substitute was grafted up to the extension line between the adjacent bony walls (E), and collagen membrane covered the graft (F). Delayed healing at the entrance of extraction socket was observed in the serial photographs; G, immediately, H, 1 week, I, 1 month, J, 4 months after the graft surgery. Regenerated alveolar ridge could be seen at the time of implantation with relatively poor density of bone (K and L) calculated on the superimposed CT image using computer software.
The percentage volumetric change relative to the grafted volume was calculated.  Figure 3F).

| Statistical analysis
The mean ± standard-deviation values of all parameters in the radiographic and cast analyses were calculated for the DBBM and DPBM groups. Unpaired t-tests were used to determine the significance of any differences between the two groups, with the cutoff for statistical significance set at P < .05.

| Participants
In total, 132 volunteers were screened in this study from December 2015 to October 2016, with 100 patients finally included. In cases where multiple teeth or sites were extracted, the most severely damaged site was chosen for inclusion in this clinical trial. The 100 sites were randomly assigned to the DBBM (n = 50) and DPBM (n = 50) groups, and they received ridge regeneration with a collagen membrane and the applicable biomaterial. Six participants were lost during the follow-up period before implant surgery, and nine participants withdrew their permission for implant installation; however, CT and cast impressions were still taken in the latter nine participants. Therefore, the procedures for this trial were ultimately completed in The demographic data of the finally included participants are presented in Table 1; none of their characteristics differed significantly between groups. Most of the sites included in this study were in the molar area, and the site distribution was similar in the two groups. In total, sockets involving the extraction of 79 single tooth sites and 15 multiple teeth sites were included, with no significant difference in the proportion between the two groups.

| Clinical findings
The included sites exhibited defects with various morphologies, including circumferential defects and the destruction of one or two bony walls. Some parts of gingival margin had receded in the exposed area of the extraction-socket entrance, and sloughing tissues covered the entire expanded socket entrance at 1 week after the surgery.

| Linear measurements
The radiographic alterations of the horizontal and vertical dimensions are presented in Table 2

| Volumetric measurements
There were similarly small volumetric changes in the DBBM and DPBM groups (Table 2); the proportional volumetric changes relative to the grafted volumes were 8.14% ± 22.23% and 6.48% ± 24.23% in the DBBM and DPBM groups, with a very large variation in range FIGURE 3 Representative radiographs and overlapped 3D-scanned study casts (the same case with Figure 1) and linear measurements. Crosssectional view from computed tomography taken at the time of surgery (A and C) and 4 months after the surgery (B and D). In the automatically superimposed images from the two time-points (E), vertical and horizontal changes of the alveolar bone dimension were measured. Surface information from the 3D-scanned study casts (S1 at the time of surgery; S2 at 4 months after the surgery) were superimposed with the computed tomography at 4 months after the surgery (F). In this view, horizontal soft tissue dimension was measured at the most-crestal level of the regenerated alveolar bone at two time-points (SH1 and SH2), and horizontal reduction was calculated by their subtraction. Vertical change of soft tissue dimension was measured at both buccal and lingual gingival margins around the extracted tooth (SVB and SVL)   There was no significant difference in distribution of all samples between control and test groups. DBBM, deproteinized bovine bone mineral; DPBM, deproteinized porcine bone mineral; Smoker, smoking less than 10 cigarettes daily. periodontally compromised sites to levels comparable to those for an intact extraction socket. Therefore, grafting immediately after extraction is expected to prevent dimensional collapse in a totally or partially damaged extraction socket, and also to reduce the need for additional surgery such as guided bone regeneration or sinus augmentation (Supporting Information Figure S1). 20 , and yet, this requires further studies to confirm the clinical necessity of these procedures in damaged extraction sockets. In addition, the issues for appropriate protocol such as intentional membrane exposure via socket entrance and healing time for implant installation should be FIGURE 5 Box plots of the vertical height and volume reduction in alveolar ridge and changes in soft tissue dimension. Box plot revealed the first and the third quartiles, median, and maximum/minimum of the data evaluated in further studies. Therefore, this study should be interpreted conservatively, as a pilot study for the extended indication of extraction socket grafting.
This study found minimal alterations in alveolar ridge volume both horizontally and vertically after grafting either type of deproteinized bone mineral in damaged extraction sockets induced by periodontitisinvolved lesion, which was comparable to the findings of previous investigations of intact extraction sockets. It is therefore suggested that DBBM and DPBM can be similarly used for grafting in damaged extraction sockets that lack bony walls. However, the presence of very wide ranges of values for all of the dimension-related parameters means that further studies are needed to narrow the indications for damaged extraction sockets in order to enhance the predictability of extraction-socket grafting techniques.