A 71-year-old patient was successfully rehabilitated by means of a 3D model-derived, hydroxyapatite-coated titanium subperiosteal mandibular implant. The implant was specifically designed to allow bone augmentation. The deficient bone was simultaneously grafted with mineralized bone allograft and recombinant bone morphogenetic protein −2 (rhBMP–2). The 32-month postoperative cone beam computerized tomography follow-up showed vertical bone augmentation beneath the implant frame.
Resorption of the alveolar ridge represents a challenge for dental implant placement. Through the years, different techniques have been used to allow dental implant placement in sites with deficient alveolar bone volume. In previous decades, subperiosteal implants were extensively utilized in patients with severe bone resorption.1,2,3 The subperiosteal implant was first described in the mid-1940s,4 then further refined by Goldberg5 and Linkow;1 throughout the years it has undergone significant changes and improvements in terms of materials and shape.
One of the major problems attributed to the use of the early subperiosteal implants was the fibrous encapsulation that led to implant movement during occlusal load and eventually to bone loss and implant failure.6 Starting with the early 1970s, the researchers' attention focused on how to prevent bone resorption, improve bone volume, and allow the integration of the implant with the surrounding bone.
In 1972, a preliminary human study demonstrated the feasibility of autogenous bone graft performed simultaneously with a subperiosteal implant.7 A few years later, the same authors showed in a Macaca mulatta monkey study that grafting the atrophic mandible with autogenous bone and hydroxyapatite, while simultaneously placing a subperiosteal implant, led to stable bone formation underneath the implant at the 6-month follow-up.8 During the same time period, similar studies showed encouraging results.9 The rationale of using a bone graft underneath a subperiosteal implant was based on the possibility of obtaining a more resorption-resistant alveolar ridge,2 allowing osseointegration of the implant,2 and protecting the alveolar nerve which, in these cases, is usually dehiscent.7 However, in the early 1980s, the utility of bone grafts performed alongside subperiosteal implants remained questionable, since long-term results had not shown substantial differences between grafted and nongrafted mandibles after implant placement.6
In recent years, interest in bone morphogenetic proteins (BMPs) has grown considerably. Bone morphogenetic proteins are multifunctional proteins with a wide range of biologic activities, involving a variety of cell types. Bone morphogenetic proteins belong to the super-family of transforming growth factor–β.10 BMPs bind specific receptors to a variety of different cell types, including mesenchymal stem cells, osteoblasts, and osteoclasts. Subsequently, these receptors activate second-messenger systems within the cellular cytoplasm, which, in turn, leads to intramembranous bone formation when a high concentration of BMPs is present.11 Although more than 20 BMPs have been discovered, only BMP −2, −4, −6, −7 and −9 have proved to be capable of driving multipotent cells into an osteoblastic phenotype culture.12,13
The aim of this paper is to present a case in which a severely atrophic mandible was successfully treated with a 3D model-derived titanium-hydroxyapatite (HA)-coated subperiosteal implant and simultaneous grafting with rhBMP–2/mineralized allograft.
A 71-year-old healthy female patient presented for implant treatment for her mandibular edentulism at the Center for Implant Dentistry, Loma Linda University, California, USA.
The preoperative cone-beam computerized tomography (CBCT) showed severe bone resorption that included not only all of the alveolar bone, but also part of the basal bone (Figure 1a and b).
Because of severe bone resorption and the patient's own wish to avoid extensive bone grafting, the placement of a subperiosteal implant was planned.
The subperiosteal implant was fabricated by utilizing a modification of the technique described previously by Truitt.14 Briefly, a stereolithographic model (Figure 2) was fabricated based on the digital imaging and communication in medicine (DICOM) information acquired by a CBCT acquisition (I-cat Classic, Imaging Sciences International, Hatfield, Penn). The stereolithographic model reproducing patient mandibular anatomy was built up to the desired alveolar dimensions, utilizing a moldable artificial gingival material (GI-Mask, Coltène Whaledent, Switzerland; Figure 3). After the ideal contour was achieved, an impression was taken of the stereolithographic model and a refractory stone model was created. Subsequently, the design of the future subperiosteal implant was created in wax (Figure 4) and then cast in titanium alloy and coated with HA particles.
The surgical treatment was performed under local anesthesia. After a crestal incision from retromolar to retromolar area, the flap was elevated and the titanium type-IV hydroxyapatite-coated subperiosteal implant (Implantlab, San Diego, Calif) was inserted.
After implant insertion, the absorbable collagen sponge (ACS) carrier was cut in small pieces and soaked with rhBMP-2 (Infuse bone graft, Medtronic, Minneapolis, Minn). Subsequently, the pieces were mixed together with mineralized bone allograft (Puros allograft, Zimmer Dental, Carlsbad, Calif) and the resulting mixture was used to graft the subperiosteal implant. Primary closure was achieved with GORE-TEX sutures.
Healing was uneventful and sutures were removed after 15 days. The implant was loaded 3 months later. CBCT was taken immediately after surgery (Figures 5a and b) and at 32 months after implant insertion (Figures 6a and b). The 32-month CBCT follow-up showed maturation of the bone graft and vertical bone gain (Figures 6a and b). At the 32-month follow-up, the implant in fully functional occlusion was clinically stable (Figures 7 and 8).
This report is the first documented case in which rh-BMP2 (Medtronic) and mineralized allograft (Zimmer Dental) were used with the intent to promote bone regeneration underneath a subperiosteal implant and osseointegration of the implant itself. The shape of our implant followed the guidelines proposed by James.15,16 The implant rested mainly on areas that tend to resist the resorption of the alveolar ridge: the genial tubercle and the two retromolars. This even distribution of forces may result in good alveolar bone maintenance. Furthermore, the implant design allowed grafting in those areas that are usually deficient of bone, like the parasymphyseal and molar/premolar area. The implant was HA coated, in order to promote possible osseointegration. The available literature on subperiosteal implants placed simultaneously alongside bone grafts is limited and controversial. While different authors agreed on the usefulness of the bone graft,2,7–9 in the only long-term study available on the subject, the author concluded that there were no significant clinical results, in terms of failure rate and complications, when comparing chromium-cobaltum subperiosteal implants placed with or without autogenous bone graft harvested from the ilium.6 The same article also pointed out that some bone gain was evident at 1-year postoperative panoramic X rays, but still the implant success rate was around 68% at the 5-year follow-up. A more recent retrospective study17 showed that HA-coated subperiosteal implants placed over a 10-year period had a 91% success rate. However, during those 10 years, 36% of implants needed additional corrective interventions. In the same study, the subperiosteal implants constructed using a 3D model showed a 100% success rate up to the 9-year follow-up. In our case, we also used a 3D model-derived, HA-coated titanium implant. The 32-month CBCT follow-up showed extensive bone formation. Bone growth phenomenon in severely resorbed mandibles has been previously documented in a large number of cases in which transmandibular implant systems were used.18–20 However, those studies were based on panoramic images and no standardization technique was implemented in order to prevent possible measurement errors.21,22 Bone growth was also reported for ramus frame implants.23 A case of bone growth after using a subperiosteal implant was reported by Fish.24 In this case, a tripodal HA-coated subperiosteal implant placed in 1985 showed bone growth above the mandibular canal in a panoramic X ray taken 14 years later. Bone growth was confirmed also at clinical re-entry. All previous reports of bone growth in severely resorbed mandibles speculated that the insertion and the loading of those different implants may have induced a physiologic “positive” microstrain to the bone, which, over time and in accordance with Wolff's law,25 promoted bone apposition. In our case, as in the case described by Fish,24 the HA-coating of the implant may have played a significant role in bone apposition; presumably, bone apposition could have happened over time without the additional use of rhBMP–2 and mineralized allograft, but it is very likely that these actively promoted bone formation.26–30 RhBMP–2 (Infuse bone graft) were used “off-label” in the attempt to achieve extensive bone formation without resorting to the “osteoinductive” autogenous bone. The rationale for combining them with a mineralized allograft was to integrate an osteoinductive material (BMPs) with an osteoconductive material (allograft) in order to possibly enhance bone regeneration. A human, randomized, controlled clinical and histomorphometric study showed that combining BMP–2 with an osteoconductive bone substitute material (xenograft) in the test group enhanced the maturation process of bone regeneration and increased the graft-to-bone contact, compared with controls, where the same material was used alone.31 In a rabbit calvarium model on vertical guided bone regeneration, the use of rhBMP–2/ACS combined with osteoconductive bone substitute materials resulted in a greater amount of bone formation than the one produced with the osteoconductive bone substitute materials alone or rhBMP–2/ACS and blood clot.32 Furthermore, in a case report where a maxillary sinus was grafted with the same graft materials used in our case (Infuse bone graft and Puros allograft), the bone core biopsy showed new bone formation in direct contact with the allogenic bone, which appeared to have acted as scaffold.33 A combination of Infuse bone graft and Puros allograft was used also to attain successful bone regeneration in an alveolar cleft palate patient.34 In our case report no jig was used to standardize the i-CAT images. However, we tried to be as accurate as possible. We used as landmark references the area immediately distal to the radiographic image of the digastric fossae (for the pre-op and post-op i-CAT images) and the canine abutments of the subperiosteal implant (for the post-op i-CAT follow-up). The cross-sectional images were closely approximated to the area corresponding to the canine position. The authors are fully aware of the fact that the images presented in this article are not entirely superimposable; therefore, they decided not to include any linear measurements in this report. However, in spite of the fact that some margins of error may exist, the preoperative, immediate postoperative and 32-month postoperative images clearly show visual evidence of the amount of bone regeneration achieved as well as the preoperative bone level status. Postoperative bone formation appears to be the result of a regenerative process, rather than just an adaptation of the bone under the stimulus of the implant load. In all probability, the subperiosteal implant frame acted as a rigid barrier, which allowed bone regeneration. Therefore, the subperiosteal implant shape may also play an important role in these types of cases.
In our case report, the postoperative follow-up was uneventful and the patient did not experience any adverse events. The most frequently reported adverse events with autogenous bone or BMPs grafting procedures are mainly pain, oral edema, face edema, and oral erythema. In certain cases, edema may have an important clinical relevance. Evaluation of studies comparing face edema as an adverse event after grafting procedures with autogenous bone or BMPs for oral and maxillofacial applications showed a higher incidence, although not statistically significant, of face edema when BMPs were used.35 It is worth noting that the collective data from the same studies mentioned above, showed that the BMPs group had fewer adverse events than the autogenous bone graft group.35 BMPs induce recruitment of inflammatory cells and fluids and may potentially cause severe soft tissue swelling. This edema, as shown in a rat model, is dependent on the dosage.36 To the authors' knowledge, in dental literature there are no documented cases of dangerous swelling after use of BMPs. However, recent orthopedic and pediatric surgery literature reports adverse soft tissue swelling, which is often associated with an “off-label” use of BMPs, leading to serious complications.37–42 It is believed that edema is the reason why BMPs are approved only for lumbar spine surgery and not for cervical spine surgery. As in our case, whenever large amounts of BMPs are used “off-label” in the mandible, a potential edema of the floor of the mouth, which may occur few days after the surgery, has to be taken into consideration before planning such a procedure, in order to prepare for potential complications.
Grafting the atrophic mandible with mineralized allograft and rh-BMP-2 at the time of subperiosteal implant insertion may lead to substantial bone formation.
Additional studies are needed to evaluate the extent to which rhBMP–2/mineralized allograft may be useful when used in conjunction with subperiosteal implants, and whether this technique can predictably induce new bone formation and influence the long-term success in those selected cases in which subperiosteal implants may be indicated.
Bone morphogenetic proteins (BMPs) show great potential in terms of regenerative therapy. However, some concerns still remain, especially regarding the long-term side effects.43 Human studies using rhBMP–2 have not demonstrated systemic toxicity.44–48 However, there are no adequate and well-controlled studies on pregnant women. There are rising concerns regarding the interference of possible maternal anti-BMP antibodies49 on BMP-dependent processes in a developing embryo50–54 as well the effects of BMPs on a developing skeleton.55 Both BMPs and BMP receptors have been isolated from human tumors.56,57 BMP-2 has stimulatory and inhibitory effects on different pancreatic tumor cell lines.58 Despite the evidence that some BMPs and their receptors can be found in tumors, there is no evidence that they are actually carcinogenic. Their presence does not mean that they induced neoplasia. They are most likely upregulated.59 Existing data is encouraging, and although long-term follow-up is not yet available, intermediate data reveals a benign side profile.43
BMPs represent a valuable treatment option for regenerative therapy. As previously stated, in the case presented in this article the patient did not experience any complications. However, further studies are needed to evaluate the “off-label” use of BMPs.
The US Food and Drug Administration approves rhBMP–2 (Infuse bone graft) as “an alternative to autogenous bone graft for sinus augmentation and for alveolar ridge augmentation for defects associated with extraction sockets.” Furthermore, the clinical trials used for rhBMP–2 (Infuse bone graft) evaluation used only rhBMP–2 by itself, without any addition of allograft or xenograft. In our clinical report, rhBMP–2 (Infuse bone graft) was used “off-label.” The patient signed a written consent form for off-label use prior to the surgical procedure. The authors declare no conflict of interest of any sort. Also, this case was not financially or otherwise supported by any company nor any other venture capitalist.