Complications Associated With the Use of Recombinant Human Bone Morphogenic Protein–2 in Ridge Augmentation: A Case Report

Insufficient alveolar ridge dimensions due to bone resorption following extractions, periodontal disease, or trauma create significant challenges to implant-supported dental reconstruction. Traditionally, several different surgical techniques for bone regeneration have been applied using a wide variety of bone-grafting materials including autogenous, allograft, xenograft, and alloplasts. Recently, different growth factors have been introduced in an effort to enhance the regenerative potential of the host, improve bone quality, and accelerate the bone maturation process.¹  Bone morphogenic proteins are members of the transforming growth factor–β factor superfamily that have been shown to have the ability to induce de novo bone formation, eliminating the need for autogenous grafting.¹  The use of recombinant human bone morphogenic protein–2 (rhBMP-2) carried in an absorbable collagen sponge (ACS; INFUSE, Medtronic, Minneapolis, Minn) was approved in 2007 by the Food and Drug Administration for maxillary sinus and alveolar ridge augmentations associated with extraction sockets. Since then, the use of rhBMP-2 in dentistry has reached significant acceptance. Its combination with other bone-grafting materials, despite being considered an off-label use, has been applied in a variety of clinical scenarios in an effort to enhance its mechanical properties.^2,3  The aim of this case report is to describe in detail a complication associated with the combination of rhBMP-2 with allograft and xenograft in a ridge augmentation procedure, along with the treatment steps taken to address it.

A 64-year-old systemically healthy woman initially presented at the University of Illinois at Chicago College of Dentistry seeking comprehensive dental rehabilitation. Upon initial presentation (Figure 1a and b), the patient had numerous defective restorations with open margins, overhanging margins, and recurrent caries on multiple teeth. Periodontally, she had generalized gingivitis with no clinical attachment loss or periodontal pockets greater than 4 mm. For the purpose of this case report, the description of treatment will be focused on the posterior maxilla as this is the area of interest due to associated complications. However, all other areas were treated, with treatment being uneventful. Following deconstruction of previous restorations and crowns, teeth 4, 5, 12, and 13 were determined to have a questionable long-term prognosis because of previous incomplete endodontic treatments as well as recurrent caries extending subgingivally and were treatment planned for extraction. To rehabilitate these areas, the accepted treatment plan included cement-retained implant-supported complete crowns in sites 4, 5, 12, and 13. The patient was informed about the treatment process, including risks, benefits, and alternatives, and all applicable informed consent forms were signed.

All surgeries were performed by 2 residents (P.D., W.D.W.) at the Department of Periodontics/University of Illinois at Chicago College of Dentistry under the guidance of board-certified periodontists. At No. 4, 5, 12, and 13 sites, nonsurgical extractions were completed without complications. Socket grafting was not performed as implants were planned to be placed approximately 8–10 weeks postextraction, with bone grafting as needed, following an early implant placement protocol.⁴ 

Evaluation of a cone-beam computerized tomography (CBCT) taken 10 weeks postextraction revealed that, while the alveolar ridge in the maxillary premolar regions bilaterally was sufficient in width for dental implant placement, such placement would be so palatally biased it would not allow the implants to be placed in a restoratively driven position (Figure 2a and b). The decision was made to perform simultaneous bilateral maxillary sinus (Figure 3a and b) and ridge augmentation procedures (Figure 4a and b) at the No. 4, 5, 12, and 13 sites to restore physiologic bone dimensions and allow for ideal implant positioning in the future. Material selection included titanium mesh (T-mesh) and a combination of xenograft (0.25- to 1-mm particle size; Bio-Oss, Geistlich, Wolhusen, Switzerland) and allograft (0.6- to 1.25-mm particle size; Cortical and Cancellous MinerOss, Biohorizons, Birmingham, Ala) in a 1:1 ratio, both mixed with ACS soaked in 1.5 mg/mL rhBMP-2 (1 large INFUSE kit, Medtronic; Figure 5).

During flap development, the rhBMP-2 protein component was processed, and after its reconstitution, it was evenly expressed onto the ACS and allowed to bind for at least 15 minutes. It was then cut into small pieces, and all sides were coated with the mix of allograft and xenograft. This mix was then used to graft both the maxillary sinus and the alveolar ridge. The T-mesh was shaped to allow for a future ridge width of at least 7 mm and was kept at a 1-mm distance from the adjacent teeth. Cortical perforations were completed to enhance angiogenesis. Stabilization of the T-mesh was completed using 4 tacks (AutoTac System Kit, Biohorizons), with 2 of them at the crest and the other 2 more apically. No perforation of the Schneiderian membrane occurred intraoperatively. Passive primary closure was achieved using periosteal releasing incisions. The patient healed uneventfully for 5 months with no early or late T-mesh exposures (Figure 6a and b). Five months postoperatively, a second CBCT was taken that showed significant ridge regeneration (greater than 5 mm of horizontal bone growth), idealizing the ridge for implant placement in the correct prosthetically driven position.

Following radiologic evaluation, 4 implants (3.5 × 13 mm; Astra Osseospeed TX, Dentsply, Molndal, Sweden) were placed in the No. 4, 5, 12, and 13 positions with 2 mm of buccal bone thickness present (Figure 7a and b) to ensure long-term tissue stability.⁵  Implants were overlaid with Acellular Dermal Matrix (ADM; AlloDerm RTM, Biohorizons) to enhance soft-tissue thickness (Figure 8a and b). The ADM was prepared following the manufacturer's instructions for rehydration and application. Passive primary closure was achieved again, and the patient healed uneventfully. Four months later, these implants were scheduled for stage 2 with simultaneous extraction of No. 14 due to vertical root fracture. Upon flap elevation, dehiscence was noted in all 4 osseointegrated implants extending close to the implant apex (Figure 9a and b). Two available options were discussed: (1) complete guided bone regeneration (GBR) over the exposed implant surfaces or (2) implant removal, completion of a second GBR throughout these regions, and replacement of these implants 6 months postexplantation and GBR. As grafting over the exposed implant surfaces was considered unpredictable because of the lack of vascular supply to the graft and extent of the dehiscences, option 2 was elected.

To provide a tenting effect and prevent horizontal ridge collapse, 3 tenting screws were placed bilaterally for space maintenance (Figure 10a and b). The screws were allowed to protrude enough to overcorrect the defect and restore alveolar width dimensions. The GBR was completed using xenograft (0.25- to 1-mm particle size; Bio-Oss, Geistlich) and non–cross-linked resorbable collagen membranes (Bio-Gide, Geistlich) in 2 layers to enhance stabilization of the augmentation material. Primary closure was achieved using releasing incisions to ensure tension-free closure.

Seven months postsurgery, a third CBCT was taken showing bone augmentation covering the head of the tenting screws (Figure 11a and b). The Digital Imaging Communication in Medicine data were then converted to a Simplant (Dentsply) file for use in executing CT-guided implant surgery using Straumann guided surgery. After radiologic review, 5 implants (Straumann SLActive Bone Level, Basel, Switzerland) were placed in the No. 4, 5, 12, 13, and 14 sites using a computer-generated tooth-supported implant guide (Figure 12a). A full-thickness mucoperiosteal flap was elevated with a palatally biased crestal incision to allow for an increase in the amount of buccal keratinized gingiva around the implants. Drilling and implant placement were completed through the guide, ensuring implant positioning as planned on the CBCT (Figure 12b and c). Healing abutments were placed at the time of implant placement as adequate primary stability was achieved (insertion torque >25 N/cm). The maxillary anterior teeth were definitively restored prior to completion of the posterior teeth to prevent further caries progression. Because of shallow buccal vestibule depth and lack of keratinized gingiva (KG; Figure 13a and b), free gingival grafts were performed bilaterally (Figure 14a and b). Upon reflection of a split-thickness flap, maintenance of the bone augmentation was noted on the buccal of the implants. Twelve weeks after soft-tissue augmentation, approximately 6–7 mm of KG was noted. An open tray final impression was made with custom impression copings. The final delivery of crowns and bridges was completed, resulting in a pleasing outcome for the patient (Figure 15a and b).

In the authors' opinion, there were 2 major complications that changed the course of treatment. The first was the extent of the ridge resorption in the premolar sites despite uncomplicated, nonsurgical extractions. The decision not to graft the extraction sockets was made with the intention to place the implants 8–10 weeks postextraction as early implant placement with contour augmentation has been shown to produce long-term graft stability and pleasing esthetic outcomes.⁴  Given that implant placement in the available, palatally biased bone would result in buccal cantilevers and potentially plaque retention issues, a ridge augmentation procedure was considered a more appropriate approach to achieve better long-term outcomes.

The second and most significant complication was the extent of buccal dehiscence and resultant thread exposure noted on all implants at stage 2 surgery. Retrospectively, it is plausible that multiple factors could have contributed to this complication. The most likely cause is that the 5 months allowed between ridge augmentation and implant placement was an insufficient time frame to allow the regenerated bone to mature, heal, and vascularize, resulting in bone resorption after implant placement.

The decision to access the site earlier than 6 months was based on the use of rhBMP-2 and its effects on bone regeneration. More specifically, the use of rhBMP-2 was selected because of its osteoinductive properties.¹  rhBMP-2 has been shown to induce de novo bone formation, enhance vascularization, and accelerate the regenerative process.¹  It has been used successfully in bone augmentation procedures either alone⁶  or in combination with bone-grafting materials^7,8  with promising outcomes. rhBMP-2 is currently approved for sinus augmentation and localized alveolar ridge augmentation associated with extraction sockets. It is not approved for ridge augmentation in multiple and contiguous sites, and its use in combination with other bone-grafting materials is still considered off-label. The combination of rhBMP-2 with bone grafts aims to enhance graft rigidity and maintain volume, as rhBMP-2/ACS lacks space maintenance properties. However, there is very limited histologic/histomorphometric evidence on the true regenerative potential of this combination. A recent case report reported very favorable outcomes when rhBMP-2/ACS and allograft were combined in a ridge augmentation procedure, with 76.1% of new vital bone formation 9 months postsurgery.⁹  Jung et al³  reported that the addition of rhBMP-2 in xenograft was beneficial in bone regeneration of peri-implant defects, 6 months postaugmentation. Although there was no statistically significant difference between the percentage of newly formed bone between the control (xenograft) and the test sites (xenograft + rhBMP-2), a greater fraction of the mineralized bone was identified as lamellar in the test sites, with the authors concluding that rhBMP-2 could enhance and accelerate bone maturation. However, another recent randomized controlled study comparing new bone formation in grafted maxillary sinuses with rhBMP-2 + xenograft (80/20 ratio; test) or xenograft (control) showed a smaller percentage of new bone formation on the test side (16.04 ± 7.45 vs 24.85 ± 5.82).¹⁰  The authors concluded that the addition of rhBMP-2 to xenograft had a negative effect on bone formation, possibly attributed to an increase in osteoclast differentiation caused by the release of rhBMP-2. They indicated that the clinician should consider using rhBMP-2 and xenograft in a singular manner in the maxillary sinus and not in combination. The enhancement of osteoclast differentiation in the presence of BMP-2 has been reported in other studies too.^11,12  To the authors' knowledge, the literature is very scarce regarding evidence on the effects of rhBMP-2 in bone maturation when combined with other biomaterials. Whether it can actually enhance and accelerate bone regeneration, what are the best biomaterials to be combined with and what is the timing of the bone maturation process are questions that still have not been adequately addressed. Until then, it is the authors' opinion that it may be more prudent to allow for longer waiting periods for the regenerated bone to heal and mature before accessing it for implant placement.

In the presented case, direct conclusions regarding the positive or negative effects of rhBMP-2 in bone augmentation cannot be made, as no histology/histomorphometry of the regenerated bone was completed. However, based on the previous significant ridge resorption using this approach, the decision was made to complete the second ridge augmentation using a different regenerative approach, using only xenograft, which has been shown to maintain augmentation dimensions predictably over time.^13,14  The decision was made also for a longer waiting period (7 months) before implant placement following the second ridge augmentation, to allow for additional healing time for the regenerated bone. One could argue that completing GBR over the exposed implant surfaces would also be an appropriate approach and may even be more favorable, as it would reduce the number of surgeries and duration of treatment. In support of this, a wealth of research has shown that GBR is effective at correcting peri-implant dehiscences and fenestrations, with studies reporting up to 100% coverage.¹⁵  However, all of these studies report on GBR procedures completed at the time of implant placement, whereas no studies were found reporting on GBR to correct significant peri-implant dehiscences at stage 2 surgery. It is proposed that, because of the lack of a fresh wound and reduced vascular supply, grafting over dehisced and fenestrated implant surfaces at stage 2 surgery may not be as predictable, especially when the dehiscence is as significant as what was seen in this case. Also, the use of ADM to enhance tissue quality at the time of implant placement has been shown to be effective at improving tissue phenotype.¹⁶  The possibility that it may have impaired blood supply from the periosteum to the regenerated bone exists. Finally, the vertical root fracture on No. 14 could be associated with harboring pathogens and disrupting healing; however, the bilateral peri-implant bone loss makes it less likely as a causative factor.

In summary, all of the aforementioned factors may have contributed to the complications described in this case. Conclusions on the definite etiology of these complications cannot be made. Regardless, when we determined that the implants showed significant fenestrations and dehiscences at stage 2, we shifted our approach and pivoted to not using rhBMP-2 for the second round of ridge augmentation. While at time of publication this case is only out 18 months since final restorations, we assume that there has been no further bone loss around the newly restored implants. While a 3-dimensional image would be required to validate this claim, a panoramic X ray (Figure 15a and b) taken at 1 year postrestoration showed no detectable interproximal bone loss. The patient has been followed every 3 months for a 1½ years since completion of treatment with maintenance of therapy outcomes and no complications.

This case shows in detail a complication associated with rapid resorption of regenerated bone following ridge augmentation using rhBMP-2 in combination with allograft and xenograft. While rhBMP-2 use in full-mouth reconstruction is promising, the timing and the extent of the bone maturation process when rhBMP-2 is combined with other biomaterials may warrant further investigation. Practitioners should be well aware of the available evidence prior to the use of any biomaterial in an effort to enhance patient treatment outcomes and avoid surgical complications.

Abbreviations

ACS:

absorbable collagen sponge

ADM:

acellular dermal matrix

CBCT:

cone-beam computerized tomography

GBR:

guided bone regeneration

KG:

keratinized gingiva

rhBMP-2:

recombinant human bone morphogenic protein–2

T-mesh:

titanium mesh

There are no financial relationships between any author and the commercial products or companies mentioned in the article that may pose a conflict of interest.