Subperiosteal Implants: A Lost Art Worth Revisiting?

ABSTRACT

Background

Subperiosteal implants were commonly used for dentally rehabilitating atrophic maxillae and mandibles in the 1940s–1980s, losing popularity following the introduction of endosseous root‐formed osseointegrated implants.

Results

Historically, subperiosteal implants had regular complications of hardware exposure, implant mobility, and pain, resulting in the removal of the implant. The transmucosal posts appear to be the primary cause of failure due to bacterial colonization and propagation down the implant substructure. These implants are currently regaining interest due to their applications for dentally rehabilitating patients following an oncological ablation.

Conclusion

For these implants to return to the dental and maxillofacial industry, contemporary techniques of bone grafting and implant materials should be explored. This review discusses the historical issues with subperiosteal implants and avenues for the improvement of long‐term outcomes in the 21st century.

1. Introduction

Over the past 100 years, dental implants have evolved from the basket‐style endosseous implant, subperiosteal implant, blade implant, and transosteal implant to the well‐known contemporary endosseous root‐formed osseointegrated implant by Brånemark [1, 2, 3, 4, 5, 6, 7] (see Figure 1).

FIGURE 1.

Various historical implant designs, demonstrating the blade‐vent implant (left), contemporary endosseous root‐formed osseointegrated implant (middle), and transosteal implant (right).

Despite the success of the endosseous root‐formed osseointegrated implant, it still has limitations due to the surrounding anatomy restraining the ability for safe and predictable insertion (especially when considering the gold standard implant measurements of ≥ 3.3 mm diameter and ≥ 8 mm length) [8, 9]. Typically, the following dimensions need to be maintained to facilitate their success [10]:

• ≥ 1 mm inferior to the maxillary sinus or nasal floor

• Avoid the maxillary incisive canal

• 2 mm superior to the inferior alveolar canal

• 3 mm from adjacent implants

• 1.5 mm from the roots of adjacent teeth

As these requirements cannot always be met, endosseous osseointegrated implants may not be appropriate for some patients, particularly in the setting of maxillary and mandibular atrophy. Maxillary atrophy has the benefits of utilizing sinus lifts to facilitate contemporary endosseous implant insertion or the use of regional anchorage including zygomatic and pterygoid endosseous osseointegrated implants. Unfortunately, the mandible does not offer similar alternatives for regional anchorage, and staged bone augmentation for large volume horizontal and, more particularly, vertical bone defects, which can be technically challenging and unpredictable [11, 12].

The oncological ablation of oral cancer can also initiate the development of atrophic maxillae and mandibles. A maxillary tumor can involve a maxillectomy requiring the removal of the entire maxillary alveolus with or without the involvement of the nasal cavity or orbit. A mandibular tumor can involve a segmental mandibulectomy (full thickness of the mandible) or a marginal mandibulectomy (partial thickness of the mandible). The residual bone height and width commonly complicate the placement of endosseous implants, particularly in the mandible, due to the regional anatomy of the inferior alveolar nerve or the remaining bone stock (see Figure 2). The reconstruction of oral cancer ablative defects is commonly achieved through the use of septocutaneous (soft tissue) free flaps (e.g., radial artery forearm free flap), which, in conjunction with the loss of the alveolus, complicates the stability of a removable denture if a more conservative route is chosen. Due to the challenges, these patients are routinely not dentally rehabilitated, which can complicate the nutritional perspective of their post‐oncological recovery.

FIGURE 2.

Orthopantomogram (OPG) of two patients with oral squamous cell carcinoma, requiring a marginal mandibulectomy to achieve disease clearance. Subsequent residual mandibular bone stock limits the options for fixed prosthodontic appliances with endosseous implants.

These situations raise the possibility of subperiosteal implants being a reasonable alternative to endosseous osseointegrated implants for rehabilitating patients with severe mandibular atrophy or in the maxilla where there is insufficient bone, and grafting is problematic. Contemporary designs, in addition to the use of CAD‐CAM (computer‐aided design and computer‐aided manufacture) technologies to fabricate patient‐specific implants, also allow the possibility of a single‐stage operation (rather than a multi‐staged procedure), reducing morbidity and treatment time for the patient. In this context, these implants have regained interest in the dental and maxillofacial community. This review describes the history, development, and outcomes of subperiosteal implants and also elaborates on their current designs, manufacturing, and potential improvements for enhanced clinical outcomes.

2. Subperiosteal Implants

Subperiosteal implants are fitted between the periosteum and bony alveolar crest and held in place via the overlying mucoperiosteum. Each implant is bespoke and traditionally created over multiple appointments. The first appointment was similar to impressions for a full denture of an edentulous maxilla or mandible [13]. A cast model was then poured from this impression, enabling the dentist to trim or scrape back the model to what they believe would represent the morphology of the patient's bony alveolar ridge [13]. The modified cast model then facilitated the subperiosteal framework to be made. The second appointment involved the surgical exposure of the bony alveolar ridge, insertion of the implant, and closure of the mucoperiosteal flap with sutures. Unfortunately, the process of arbitrarily trimming the cast models of the mucosal alveolar ridge impressions created many prosthetic inaccuracies and instability [13].

From the 1960s, this issue was circumvented by surgically exposing the bony alveolar ridge for the dental impressions in the first appointment. Subsequently, a more stable subperiosteal implant would be fabricated and inserted during a following surgery. This created a painful intermediate step due to a large area of subperiosteal dissection to facilitate impressions [13].

3. Subperiosteal Implant Design

Subperiosteal implants consist of primary struts, secondary struts, and transmucosal abutment posts. The primary struts are the peripheral metal framework that supports the implant substructure, while the secondary struts interlink and reinforce the primary struts, which can include a lattice‐like structure (see Figure 3). The aim of the lattice design is to facilitate mucoperiosteal reattachment onto the underlying bone to anchor the prosthesis, while reducing the overall weight of the implant (particularly important for maxillary subperiosteal implants, to reduce overlying gingival erosion).

FIGURE 3.

Subperiosteal implant with a minimalistic design, demonstrating the primary struts surrounding the periphery (blue arrows), secondary struts (red arrowheads), transmucosal posts (green arrows) and prosthetic bar to facilitate an over denture attachment (black arrowhead). Reproduced with permission from reference [14].

Alongside this subperiosteal framework, these implants consist of transmucosal abutment posts that enable the attachment of various components and oral prostheses. This can include a Dolder Bar or Hader Bar for retaining an overdenture.

Subperiosteal implants must be designed to be seated on the densest cortical bone available, while taking into consideration that the remodeled alveolar ridge is typically cancellous bone [15]. The struts of subperiosteal implants should be seated against prominent bony structures to reduce the amount of lateral movement following insertion [15]. Struts should not be placed over sharp bony edges to avoid overlying gingival ischemia and subsequent gingival erosion, thereby requiring an alveoloplasty (pre‐prosthetic alveolar ridge adjustment) prior to impressions of the underlying bone [15]. Finally, the subperiosteal transmucosal abutment posts should perforate the mucosa at sites of keratinized tissue (typically located more palatally or lingually in an atrophic maxilla and mandible) [15].

Conventional subperiosteal implants were fabricated from Vitallium, an alloy composed of 60% cobalt, 30% chromium, and 10% molybdenum [16]. In the 1940s, Vitallium was the metalware of choice for bone surgery, as it was electrically inert (in the presence of bodily fluids), corrosion‐free, and did not cause any pathological changes to the bone, enabling it to be widely used in Orthopedic Surgery for the fixation of fractures and joint replacements [17]. Following the success rates of titanium‐based endosseous root‐formed osseointegrated implants in the 1980s, later generation subperiosteal implants followed suit and shifted their composition to titanium implants [18].

4. Outcomes of Subperiosteal Implants

Early studies of subperiosteal implants led to their increased uptake around the world. Obwegeser [19] was one of the first clinicians to publish a prospective study in 1959 with a promising 97% survival rate of subperiosteal implants, 12 months after insertion. However, of significant note, two‐thirds of the 35 subperiosteal implants inserted had complications (including pain, recurrent infections, paraesthesias, and fistulas) despite remaining in situ for the duration of the study.

Over the following 20–30 years, with more subperiosteal implants being in service, larger sample sizes were critically analyzed surrounding their success. The literature overwhelmingly reported high rates of failures and complications [16, 20, 21, 22, 23]. These failures of subperiosteal implants were attributed torecurrent infections, fistula formation, pain, mental nerve paraesthesias, implant substructure exposure, extensive bone resorption, and mobility. Bailey et al. [23] termed the “degenerative triad” of (1) wound dehiscence, (2) pain with vertical force (i.e., biting), and (3) recurrent inflammation as the most likely causes for a subperiosteal implant to fail and require its removal.

In review of the literature, the survival rates of standard subperiosteal implant placement (without augmentation with particulate bone grafts) appear to be 97% at 12 months [19], 90%–96% at 5 years [16, 20, 21], 67%–86% at 10 years [20, 21, 22, 23], 52%–60% at 15 years [20, 21, 22], 50% at 20 years [20], 26% at 30 years [20], and 12% at 35 years [20] (see Table 1). Unfortunately, there are very limited long‐term studies due to subperiosteal implants losing popularity to the contemporary endosseous root‐formed osseointegrated implant.

TABLE 1.

Historical implant survival rates.

Author	Year	Study type	No. subperiosteal implants	Implant survival rates		Complications	Comments
Obwegeser [19]	1959	Prospective. 1 year.	25 mandibular, 10 maxillary.	1 year	97%	N/A	2/3 had complications: pain, recurrent infections, paraesthesia, fistulas.
Bodine [21]	1974	Prospective. 22 years.	27 mandibular.	5 year 10 year 16 year	96% 67% 52%	Chronic inflammation. Fistula formation. Metal exposure. Bone resorption.
Young et al. [16]	1983	Retrospective. 20 years.	25 mandibular.	5 year 10 year	90% 75%	Chronic inflammation. Bone resorption.	50% had paraesthesia or anesthesia. Partial implant removal was not considered a failure.
Bailey et al. [23]	1988	Prospective. 14 years.	74 mandibular.	10 year	86%	Pain with vertical loading of posts. Wound dehiscence.	51% had paraesthesia. 41% had pain with vertical loading of posts.
Yanase et al. [22]	1994	Prospective. 21 years.	81 mandibular.	10 year 15 year	79% 60%	Chronic inflammation. Pain. Paraesthesia.	8 free palatal autografts to chronic inflamed abutments failed.
Bodine et al. [20]	1996	Retrospective. 41 years.	41 mandibular.	5 year 10 year 15 year 20 year 25 year 30 year 35 year	95% 76% 60% 50% 33% 26% 12%	Recurrent chronic inflammation. Bone resorption (particularly in posterior mandibles).	Implants prematurely terminated (patient death, mandible fracture, etc), removed from dataset.
Linkow et al. [18]	1998	Prospective. 14 years	317 “tripodal” designed mandibular. 271 participants followed up.	14 year	98%	Metalware exposure (confounded by radiation therapy for throat cancer). Pain. Bone resorption.	Tripodal subperiosteal implant design, avoiding the severely atrophic body of mandible.
Moore & Hansen [24]	2004	Retrospective. 18 years	39 mandibular.	8 year	97%	Chronic inflammation	Tripodal subperiosteal implant design. Close follow up.

A 1998 retrospective analysis by Linkow et al. [18] demonstrated significantly different results from the remaining literature, likely attributed to their modified subperiosteal implant design, titanium composition (compared to Vitallium alloy previously utilized), and minimally‐invasive surgical approach (see Figure 4). The Linkow subperiosteal implant was tripodal, with posterior rests on bilateral external oblique ridges of the mandible, and anteriorly at the mandibular symphysis. These chosen sites reflected the most amount of muscular attachments onto the mandible. With mechanical activity, the alternating tension‐compression forces help to maintain bone stock, particularly after the loss of teeth. Therefore, implants resting on these sites would potentially be most resistant to bone resorption. In their 14‐year multi‐centered study, 317 tripodal mandibular subperiosteal implants were inserted, with 271 participants followed up. During this period, any implant that demonstrated secondary strut hardware exposure due to erosion of the overlying mucosa underwent the removal of the secondary strut to encourage mucosal coverage. Localized infections occurred immediately post‐implantation, likely due to foreign debris, whereby all patients were successfully treated with antibiotics, and “improved oral hygiene.” After the 14‐year study, 98% (267) of implants survived. Four implants were removed due to a combination of pain, bone resorption and metalware exposure. One of the cases was confounded by previous radiation therapy to the head and neck for the treatment of throat cancer.

FIGURE 4.

Linkow tripodal subperiosteal implant design, demonstrating the placement on the resilient bony landmarks of the mandibular symphysis and external oblique ridges. Reproduced with permission from reference [18].

Moore and Hansen [24] also published a paper in 2004 with outlying results, whereby 40 Vitallium subperiosteal implants were placed between 1982 and 2000. All implants followed the Linkow tripodal mandibular subperiosteal implant design. Of the 40 implants, 39 were included in the study (as one patient passed away due to unrelated causes), with a mean time of implant service of 8 years. Thirty‐eight of the 39 implants did not show any signs of overlying mucosal inflammation or implant mobility. One of the implants demonstrated chronic mucosal inflammation, though this was confounded by the patient's background of diabetes mellitus. Interestingly, all patients were closely followed up post‐operatively in 6–12 month intervals for complementary hygiene and maintenance of their prostheses (including denture relines and exchanging retentive O‐ring components), likely contributing to their overall implant success.

Table 1 summarizes the historical landmark studies and highlights their clinical outcomes, indicating a global trend of increasing failure of these implants over time.

5. Advances in Subperiosteal Implants

Despite the initial loss of popularity of subperiosteal implants in the 1980s, due to the more reliable long‐term outcomes of endosseous osseointegrated implants, subperiosteal implants have started to be re‐explored in the recent decade for the reconstruction and rehabilitation of maxillofacial oncological patients and atrophic jaws (where regional anchorage with zygomatic and pterygoid implants are not feasible options).

In recent years, subperiosteal implants have utilized high‐quality computed tomography (CT) images with CAD‐CAM to fabricate custom patient‐specific implants with a high degree of accuracy through direct metal laser sintering (DMLS; a method of three‐dimensional printing or additive manufacturing) [25, 26, 27, 28, 29]. This eliminates the morbid surgical exposure of the alveolar ridge to capture an impression, while improving the overall fit of the implant. The use of CAD‐CAM also facilitates a prosthodontically‐driven, reverse‐engineered approach to designing the subperiosteal implants based on where the final prosthetic teeth will be located, and the subsequent required angulations and positionings of the abutments of the underlying subperiosteal implant [29].

Table 2 summarizes the technological advancements in the recent few decades, which contribute to the application of custom patient‐specific implants and surgical guides, and the re‐exploration of subperiosteal implants.

TABLE 2.

Technological advances applicable to the pre‐, intra‐, and post‐operative care of patients with subperiosteal implants.

Technological advances applicable to subperiosteal implants
Pre‐operative
High quality or fine slice CT
Intraoral scanning (IOS), and generation of transferrable data with various clinicians
Medical engineering software (e.g., Materialise)—for virtual surgical planning, utilizing DICOM data from CT and STL‐files from IOS
3D printing of stereolithic models, custom implants, and surgical guides
Intraoperative
Intraoperative CT (e.g., O‐arm)
Navigation systems (e.g., StealthStation)
Post‐operative
Telehealth consultations (phone or video)—for earlier identification of post‐operative complications for geographically distant patients.

Abbreviations: 3D, three‐dimensional; CT, computed tomography; DICOM, digital imaging and communications in medicine; STL, standard template library.

Additionally, extensive research in endosseous root‐formed osseointegrated implant materials has led to titanium being the preferred option due to its mechanical resilience, biocompatibility, and predictable osseointegration properties. As such, it is only logical for subperiosteal implants to follow suit.

Another area of progression in dental implantology is our greater understanding of risk factors associated with endosseous dental implant failures. This particularly involves a history of periodontal disease, smoking, bruxism, poor glycaemic control, antiresorptive therapy, and a history of radiation therapy to the head and neck. Of this list, the modifiable risk factors include smoking, glycemic control, and bruxism. These risk factors can have a significant impact on implant failure, whereby cigarette smokers have a hazard ratio of 4.36 for implant loss [30, 31], and bruxers have an odds ratio of 60.95 for mechanical complications [32].

As subperiosteal implants have only very recently been re‐explored, the literature on contemporary subperiosteal implants (using CAD‐CAM and DMLS) is still very limited (see Table 3). Mommaerts [29], in 2017, has been one of the pioneers in the re‐exploration of subperiosteal implants in the 21st Century, describing his technique for fabricating a CAD‐CAM additively manufactured titanium implant for atrophic edentulous maxillae. His designs included minimal hardware overlying the alveolar crest, and subsequent mini‐screw fixation along the nasomaxillary and zygomaticomaxillary buttresses (where bone stock and bone quality is best). A case series of 15 patients by Van den Borre et al. [33] utilizing the Mommaerts subperiosteal implant were followed up over 12 months, demonstrating promising results including nil clinical and radiological findings of rhinosinusitis and nil mobility of the subperiosteal implant (100% survival rate, 100% success rate).

TABLE 3.

Contemporary outcomes on subperiosteal implant survival rates (remaining in situ) and success rates (free of any complication).

Author	Year	Study type	No. subperiosteal implants	Implant survival rates		Complications	Implant success rates
Gellrich et al. [38]	2017	Prospective	1 maxillary, 2 mandibular. (2 oncological under native mucosa)	18 months (mean)	100%	N/A	18 months (mean)	100%
Van den Borre et al. [33]	2022	Prospective	15 maxillary	12 months	100%	N/A	12 months	100%
Dimitroulis et al.	2022	Retrospective	18 maxillary, 3 mandibular	22 months (mean)	80.9%	Mobility. Implant hardware exposure	22 months (mean)	66.7%
Carretero et al. [36]	2022	Prospective	4 maxillary oncological implants (under septocutaneous flap)	19 months (mean)	100%	N/A	19 months (mean)	100%
Korn et al. [37]	2022	Prospective	13 maxillary implants (into 10 patients)	8.2 months (mean)	100%	Infection/abscess. Osteosynthesis mini‐screw fracture. Implant hardware exposure.	8.2 months (mean)	69%
Van den Borre et al. [34]	2024	Retrospective	40 maxillary	2.5 years	100%	Implant hardware exposure. Recurrent infections.	2.5 years	35%
Onica et al. [40]	2024	Prospective	48 maxillary, 13 mandibular	6 years	54%	Implant hardware exposure. Recurrent infections. Mobility. Implant fracture.	6 years	19.7%

Another study by Mommaerts and his team assessed the mucosal health of 40 patients with subperiosteal implants over a mean period of 2.5 years [34]. Twelve patients had post‐operative infections, of which six (6) required further surgical interventions (including exploration and debridement). Similar to endosseous root‐formed osseointegrated implants, keratinization around the transmucosal posts is crucial for the prevention of peri‐implant mucositis and hardware exposure. The Mommaerts subperiosteal implant design incorporated contingencies for amputating transmucosal posts and secondary struts without complicating the stability and function of the subperiosteal implant. This was conducted in 3 of the 40 patients with recurrent infections, which subsequently did not impact the prosthodontic restoration. Despite these interventions, 26 of the 40 patients still had varying degrees of hardware exposure (65%), though all remained functional and were left in situ at the time of the study (100% survival rate, 35% success rate).

A recent 2022 study by Dimitroulis et al. [35] followed the insertion of 21 subperiosteal implants (18 maxillary, 3 mandibular) using CAD‐CAM techniques over a mean period of 22 months. Seven implants had recorded complications, all within 6 months. One implant was removed secondary to unrelated psychosocial distress in the participant. Three implants were completely revised with a different subperiosteal implant due to hardware exposure from bulky implants from their original design. One implant was salvaged through additional mini‐screw fixation for stabilization of a mobile frame. Two implants had varying degrees of hardware exposure and were left in situ at the time of the study. This results in an 80.9% (17/21) survival rate, with a 66.7% success rate.

Carretero et al. [36] in 2022 had a small case series of four post‐oncological patients with a resultant 100% success rate over a mean follow‐up period of 19 months. All patients included in the study had an oncological maxillectomy with a failed bony fibula free flap reconstruction, while maintaining vitality to the overlying septocutaneous flap (soft tissue component of the fibula free flap). The dental rehabilitation was facilitated by a customized subperiosteal titanium maxillary implant, inserted under the septocutaneous flap and anchored onto the residual native maxillofacial skeleton with mini‐screws. Perhaps, the increased laxity and thicker overlying septocutaneous flap were protective of the implant, in comparison to a significantly thinner mucoperiosteal flap in a non‐oncological case. This correlates with the 2024 study by Van den Borre et al. [34] identifying the issues with a thin biotype contributing to mucositis and subperiosteal implant hardware exposure.

Gellrich and colleagues have also explored the use of DMLS patient‐specific subperiosteal implants for rehabilitating severe atrophic maxillae and mandibles [37, 38]. Their maxillary subperiosteal implant preference included a single‐unit prosthesis (incorporating the transmucosal posts), minimal hardware overlying the alveolar crest, bony anchorage with multiple osteosynthesis mini‐screws over the nasomaxillary and zygomaticomaxillary buttresses bilaterally (ideally ≥ 20 screws), with a total of four transmucosal abutments emerging from the alveolus overlying these regions [37, 39]. One of their studies (Korn et al. [37]) showed positive results over a short prospective period of a mean of 8.2 months (100% survival, 69% success). Interestingly, Onica et al. [40] had similar subperiosteal implant designs to Gellrich, though their 6‐year prospective study was complicated by hardware mobility and exposure, recurrent infections, and fracture of the implant (54% survival, 19.7% success).

6. Discussion

Subperiosteal implants were a popular dental implant from the 1940s to the 1980s, as an intuitive solution to dentally rehabilitate patients with atrophic maxillae and mandibles. Following the introduction of the Brånemark endosseous root‐formed osseointegrated implant in the 1980s, subperiosteal implants quickly dropped out of fashion due to endosseous implants having more favorable and predictable long‐term outcomes.

Similar to endosseous implants, subperiosteal implants are also subject to peri‐implantitis and have historically failed due to bacterial colonization of the substructure, reactive bone loss surrounding the implant, implant mobility, and implant hardware exposure (see Figure 5). These subsequently lead to recurrent infections, fistulation, pain, and a non‐functional overlying prosthesis, commonly requiring the removal of the subperiosteal implant. Although not clearly documented, the most likely cause of failure is the implant‐mucosa interface at the transmucosal posts. As there is no bony coverage of the implant substructure to facilitate a more robust biological barrier, bacterial colonization and biofilm formation at the transmucosal post can propagate along the entire implant substructure under the mucoperiosteum, causing localized chronic inflammatory responses, calculus formation, bone resorption, implant mobility, and hardware mucosal erosion.

FIGURE 5.

Peri‐implantitis and peri‐implant mucositis of a subperiosteal implant, clinically notable for hardware exposure, implant mobility, bone resorption, and overlying mucosal inflammation. Photograph courtesy of Dr. Peter Russell (Prosthodontist, Herston Oral Health Centre, Brisbane, Queensland, Australia).

Peri‐implantitis in subperiosteal implants is not as heavily researched as with endosseous implants. Endosseous implant peri‐implantitis management may involve a combination of: surgical exposure and debridement, laser and photodynamic therapy for surface decontamination, implantoplasty, guided tissue regeneration, guided bone regeneration, and improved oral hygiene measures [41, 42, 43]. Currently, the management of subperiosteal implant peri‐implantitis involves antimicrobial therapy and surgical debridement with/without the amputation of exposed secondary struts (without complicating the entire supragingival prosthesis).

The historical landmark study by Linkow et al. [18] requires further attention, as despite reporting the clinical data of their subperiosteal implants, the article was particularly focused on detailing their surgical techniques and modification of the conventional subperiosteal implant design. Subsequently, there was very limited explanation on their methodology, patient follow up period, and their criteria for a “successful subperiosteal implant.” As such, there was no mention of patients with implant hardware exposure, recurrent pain or draining sinuses and fistulas, which could be rather prevalent when extrapolating the data from other historical studies [16, 19, 20, 21, 22, 23].

The follow‐on study by Moore and Hansen [24], replicating the Linkow tripodal subperiosteal implant, did reaffirm the successful subperiosteal implant design, with a 97% implant survival rate over a mean service period of 8 years. Furthermore, only one of their remaining 38 implants demonstrated signs of chronic inflammation around the transmucosal posts, confounded by their background of diabetes mellitus. Perhaps carefully designed subperiosteal implants can improve the overall survival, with particular attention to seating the primary struts on the most stable basal bone of the mandible (external oblique ridges and mandibular symphysis) and broadening the distribution of occlusal forces to minimize the transverse mobility of the subperiosteal implant.

Interestingly, despite the promising results from Linkow et al. [18] and Moore & Hansen [24], positive outcomes of these implants remain infrequent and were therefore rarely used in the 21st century until very recently. Perhaps subperiosteal implants are extremely technique‐sensitive, requiring significant attention to the implant framework design, implant positioning, coordination between the surgeon and the prosthodontist and dental laboratory, and patient compliance in order to achieve successful and reliable clinical outcomes.

As seen in the limited contemporary literature on subperiosteal implants (Table 3), despite the use of CAD‐CAM techniques and improved perioperative optimization, subperiosteal implants still have unpredictable outcomes in comparison to endosseous osseointegrated implants, with the most common failures continuing to be hardware exposure and mobility.

The authors of this current paper believe that subperiosteal implants are a viable option for a cohort of patients. Alterations that would improve survival and success rates include: mucosal grafting, bone grafting, adequate bony anchorage, carefully positioned implants, titanium implant composition, and implant surface modification to facilitate bone coverage and increased implant stability.

The design of subperiosteal implants is an aspect of the treatment planning pathway that requires a lot of consideration of patients' biological factors. As with endosseous implants, keratinized mucosa is highly protective against peri‐implantitis and peri‐implant mucositis. Subperiosteal implants should similarly aim to achieve the emergence of their transmucosal posts through a collar of keratinized mucosa. In an edentulous maxilla, this can be achieved with a more palatally placed incision. However, the mandible unfortunately does not retain the same amount of keratinization, being ≤ 3.5 mm wide in a resorbed alveolar ridge [44]. Pre‐operative planning and careful mucosal handling are crucial to preserve and incorporate this in the wound margin and subsequent transmucosal posts. If this cannot be achieved due to regional anatomy or prosthodontic considerations, keratinized mucosal grafts should be kept as an option.

In addition to the ability for osseointegration, titanium can maintain its rigidity with minimal thickness. This has benefits in subperiosteal implants, as the low‐profile hardware would cause less overlying tissue ischemia, and subsequently less likely for the implant to be exposed and contaminated.

Implant surface treatment and modification should be explored to increase the potential for osseointegration and oral hygiene. Similar to tissue‐level endosseous osseointegrated implants, the transmucosal posts of subperiosteal implants should maintain a highly polished machined surface to minimize plaque retention and improve oral hygiene. The surface of the implant substructure could be acid‐etched, surface‐blasted, or galvanized with an additional oxidized layer to improve the titanium implant's osseointegrative properties [45].

If considering a possibility for subperiosteal implant osseointegrating following insertion, primary stability is highly important. As described by contemporary subperiosteal implant leaders like Mommaerts and Gellrich [29, 38], the use of CAD‐CAM and DMLS can facilitate the incorporation of bony anchorage via osteosynthesis mini‐screws away from the alveolus, along the nasomaxillary and zygomaticomaxillary buttresses of the midface, where bone stock is most likely to be preserved (in an atrophic maxilla) to provide rigidity of the overall prosthesis. The current recommendation is to insert ≥ 20 mini‐screws for a full maxillary subperiosteal implant prosthesis. In an atrophic mandible, establishing ≥ 20 mini‐screw fixation away from the alveolus is unlikely to be easily achievable; though perhaps incorporating the Linkow tripodal mandibular subperiosteal implant design with fixation along the ramus and menton might be an avenue worth exploring in the future.

Bone grafting techniques used alongside subperiosteal implants could potentially modify the implant‐mucosa interface at the transmucosal posts. If successful with bone formation overlying the implant substructure, the mucoperiosteal fibers would be adherent directly onto the bone, creating a more familiar and hygienic junction as those with contemporary endosseous osseointegrated implants. A study by Kratochvil et al. [46] explored the placement of particulate cancellous bone grafts around subperiosteal implants in nine Macaca mulatta monkeys. Autologous bone from the posterior ileum and greater trochanter was grafted around the mandibular subperiosteal implants of edentulous monkeys. The overdentures were fitted 2 months after implantation and functioned for a total of 4 months prior to euthanization. The subsequent histological assessment of the mandibles demonstrated complete bone coverage in seven of the nine subperiosteal implants.

This concept was taken to human subjects by Bloomquist [47] in 1982, with 19 subperiosteal implants inserted with mini‐screw fixation of the implant and simultaneous autologous grafting of particulate cancellous bone. Unfortunately, six implants failed and were removed due to persistent facial pain, recurrent mucosal infections, and progressive exposure of the implant framework. Three of the remaining 13 implants had some degree of non‐progressive implant exposure, which remained in situ at the time of the study. It is likely that the six implants failed due to the short one‐month healing period between the implantation and grafting procedures prior to the surgical re‐exposure for the insertion of the transmucosal posts. Perhaps the outcomes could improve if the implants and bone graft were left uncovered for a longer duration, enabling adequate bone formation and remodeling.

Aaboe et al. [48] in 2000 conducted a histological descriptive study assessing the outcomes of bone grafting around a subperiosteal implant. Titanium subperiosteal implants were placed into the tibias of nine rabbits (two implants per rabbit), covered initially with particulate bovine xenograft (Bio‐Oss, Geistlich Pharma, Wolhusen, Switzerland), and finally with one of three randomly assigned membranes (a resorbable Polyglactin 910 mesh; a resorbable bilayered collagen membrane; or a non‐resorbable expanded polytetrafluoroethylene [ePTFE] membrane). After a 12‐week healing period, there was notable osseointegration of all implants, and all implants showed signs of bone formation. The two resorbable membranes showed signs of collapse but still persisted with bone formation due to the osteoconductive properties of the supporting Bio‐Oss. The non‐resorbable ePTFE membrane had no sign of collapse overlying the implant and subsequently was most successful in new bone formation overlying the titanium subperiosteal implant.

The Kratochvil et al. [46] and Aaboe et al. [48] studies propose the possibility of improved outcomes with successful burial of subperiosteal implants using bone grafts, and potentially, with advanced bone grafting techniques such as guided bone regeneration, more reliable results can be achieved.

7. Conclusion

Subperiosteal implants are a sound concept for dental and maxillofacial rehabilitation when endosseous implants and regional anchorage (zygomatic and pterygoid implants) are not possible. Unfortunately, most of the studies demonstrate unpredictable long‐term success, requiring the removal of the subperiosteal implant. As with contemporary endosseous osseointegrated implants, the transmucosal interface is one of the primary causes of subperiosteal implant failure, due to bacterial colonization and bacterial propagation down the implant substructure. Modification of these implants by combining advanced bone grafting techniques with implant surface modification, mini‐screw bony anchorage, and accurately contoured implants may improve the outcome for these implants in the future.

Author Contributions

Ryan Goh: drafting of the paper, data collection/literature review, analysis and interpretation of data, approval of the submitted and final versions. Cedryck Vaquette: critical revision of the article, approval of the submitted and final versions. Omar Breik: critical revision of the article, approval of the submitted and final versions. Saso Ivanovski: critical revision of the article, approval of the submitted and final versions. Martin Batstone: concept/idea, discussion, critical revision of the article, approval of the submitted and final versions.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.