© The authors, 2025

Introduction

The management of maxillary and facial defects resulting from tumour excision surgery poses a challenge for maxillofacial surgeons. Among these, class VI defects according to Brown et al. (2010), involving the naso-maxillary region, are considered some of the most difficult to treat [1].

For these complex cases, prosthetic obturation offers clear advantages, including immediate restoration of appearance and the possibility of monitoring the cavity. This epithesis can be retained using tissue adhesives or attached to spectacles; however, these methods have notable limitations. Its stability can be improved with bone-anchored retention devices. The placement of standard-length implants in the midfacial region remains the standard approach, with implants typically placed in the nasal floor and glabella, even in potentially irradiated bone [2–6]. Meanwhile, plate anchorage systems, such as epiplating, represent an interesting alternative for partial rhinectomies [7,8].

In cases of total or extended rhinectomies, the use of zygomatic implants (ZI) may provide a particularly reliable retention solution, positioned away from irradiated bone. Their efficacy is well documented for the rehabilitation of maxillary defects, with high survival rates (77–100%) with a low complication rate [9–11]. However, their exclusive use, in a horizontal position, to support nasal prostheses is a more recent concept. First reported by Bowden et al. [12] after conventional implant failure, this technique has yet to find its place in clinical practice. Available studies primarily concern two retrospective, single-centre studies [13,14]. The first, carried out by Scott et al., involved the rehabilitation of 28 patients using ZI [13]. The second, conducted in the Salisbury department, addressed the rehabilitation of 34 patients with bone-anchored implants, including 8 ZI [14].

Although these reports have provided valuable data, there is still a lack of detailed documentation on long-term outcomes, technical variations, and the adjustments required in different clinical contexts. By presenting recent cases with prolonged follow-up, this report aims to identify potential improvements and guide practitioners towards the optimal application of this innovative technique. Finally, the writing of this article followed the CARE guidelines [15], which are part of the EQUATOR network.

Case study

Observation and therapeutic decision

All clinical information and observations for both patients are summarised in Table I. The first patient complained of chronic epithesis mobility and was reluctant to undergo further surgery, so enhancing his existing prosthesis met his needs. In contrast, the second patient, who had developed extensive facial dermatitis due to tissue adhesives, was willing to consider reconstruction. However, given his ongoing smoking habit, reconstructive surgery seemed high-risk, making an improved epithesis the more suitable option. Consequently, both patients were offered an implant-retained epithesis. After reviewing the isodose curves (indicating less than 50Gy to the zygomatic bones) and assessing bone volume on CT scan, the decision was made to place two horizontally positioned ZI emerging at the level of the piriform rim. Both patients had been in remission for 27 and 17 months, respectively, so there was no contraindication to implant placement. Because patient1 underwent a less radical tumour resection, 15mm of nasal bone and the nasal septum were preserved.

Pre-operative assessment

Implant planning was carried out using 3D reconstruction software (DTX-Studio) to position two implants horizontally, ensuring that the zygomatic bone was fully engaged. The main objectives were to ensure parallelism between the two implants, as well as an emergence offset of 3 to 4 mm in height, to facilitate their placement during surgery. The department used the Zygoma Branemark system (Nobel Biocare, Kloten, Sweden), with a TiUnite or machined surface, a diameter of 3.9 mm, and a length ranging from 30 mm to 52.5 mm. Furthermore, a 3D craniofacial model was 3D-printed from the CT data using ITK-SNAP (v3.8.0, University of Pennsylvania, USA) and 3D Slicer (v4.11, Harvard Medical School, USA), then printed via a stereolithography resin printer (Fig. 1). On this physical set-up, pilot holes were created to simulate the implant drills, helping us validate implant axes and refine the intraoperative approach.

Fig. 1 Setup on a 3D model of the upper facial structure for patient 1.

Operating protocol

Surgery was performed under general anaesthesia. Access was achieved via the nasal route, at the site of the pyramid amputation, and the incision was made at the junction of the mucosa and skin. A progressive subperiosteal dissection was carried out to identify the residual bony edges of the piriform orifice and to locate the emergence of the infraorbital nerve. The infraorbital rim and the zygomatic bone were exposed. A pilot drill (2.9 mm) was used to create the entry point in the medial wall of the maxillary sinus and to drill the zygomatic bone to the full length of the implant. The drilling was then widened to 3.5 mm to allow the insertion of the ZI (50 N/cm). Near-perfect parallelism was achieved between the two implants (Figs. 2–5), with the head of the implant inclined at 45° to the coronal plane. Healing abutments were placed to prevent soft tissue proliferation during the healing phase.

Fig. 2 Patient 1, Post-operative panoramic radiograph.

Fig. 3 Patient 1, post-operative teleradiograph, after maxillary edentation and placement of the prosthetic bar.

Fig. 4 Patient 2, post-operative panoramic radiograph.

Fig. 5 Patient 2, post-operative frontal radiograph (Blondeau view).

Prosthesis fabrication

The final prosthesis, made by an epithesist, was completed and fixed to the implants six months after surgery (Fig. 6), during which the patient wore a conventional epithesis. Impressions of the nasal cavity were taken using a custom tray with clip-on implant transfers. A Dolder-type gold bar was designed to connect the two implants and distribute stress. Circular in shape, this bar also incorporated a central vertical anti-rotational component to enhance stability and function as a stabilising tripod. Once passive adaptation of the bar was confirmed, it was attached to the epithesis by means of a retentive rider. This fixation system included an intermediate structure made of self-curing acrylic resin, firmly anchored to the silicone elastomer. The prosthesis was fabricated in appropriate shades, based on CT scan data and photographs taken before the onset of the disease. The follow-up, provided twice a year by the epithesist, aimed to maintain retention and make any necessary aesthetic adjustments.

Fig. 6 Patient 2, Placement of the prosthetic bar 6 months after surgery, Intrados of the nasal epithesis, Nasal epithesis in place, clipped to the implant-supported bar.

Post-operative follow-up

Patient 1 experienced an immediate failure of osseointegration, with mobility and rotation of the implant, as well as migration into the zygomatic space, visible on the CT scan. The implant was replaced with a new one, positioned further forward in the body of the zygoma. A few weeks later, maxillary sinusitis developed on the same side but was effectively treated with antibiotics, and the new implant integrated successfully. No complications were observed for patient 2. Issues with crusting and endonasal dryness, related to the absence of nocturnal use of the epithesis, were resolved with regular wear. Long-term implant follow-up was conducted separately from the oncological follow-up of the nasal cavity. Appointments took place every three months in the first year, every six months in the second year, and then annually. At the two-year follow-up, all implants demonstrated perfect osseointegration and excellent prosthetic retention. The results therefore showed an implant survival rate of 75% (3/4 implants), with one implant subsequently replaced successfully.

Discussion

The results of this study confirm the effectiveness of ZI in improving nasal epithesis retention following total or subtotal rhinectomies. These findings align with those of Scott et al. (overall survival rate of 98%) and Ethunandan (87%) [13,14]. In our two cases, despite one initial implant failure requiring replacement, the long-term outcomes were excellent. Nasal prosthetic rehabilitation was performed independently of oral rehabilitation, as both patients were satisfied with their removable maxillary prostheses. However, had implant-supported oral rehabilitation been considered, zygomatic implantation could have been an option depending on the severity of maxillary bone atrophy. In this context, the use of an epiplating system or conventional mid-facial implants could also have been discussed to support the nasal epithesis [7,8].

The implants were placed after radiotherapy, which was feasible because the cumulative dose to the zygoma was below 50 Gy [16]. Implants were placed after rhinectomy, as initial treatment was performed in another centre where bonded epithesis had been preferred. Several studies currently recommend placing ZI at the time of tumour resection to expedite prosthetic rehabilitation and reduce the risks of invasive surgery in irradiated areas [4]. According to Butterworth et al., this approach has a 4% failure rate compared to 11% for secondary implants, although the difference is not significant [10]. Scott et al. (Morriston experience), placed 56 ZI (46%) prior to radiotherapy, and reported only one failure, emphasizing the advantages of early implantation [13]. However, our two implant surgeries were performed within the recommended one-year post-radiotherapy period, which has been associated with significantly improved implant survival rates [4,16,17].

In oncological resection, nasal bones are preserved as far as possible to provide a robust tissue base for the prosthesis and allow the nasal bridge to support glasses. This was not possible for the second patient; depending on the connection system, a glabellar implant could have been considered [14,18]. Nevertheless, Scott et al., who rehabilitated 28 patients using ZI, opted for an additional nasal implant whenever the nasal cavity width allowed. In their work, Scott and Ethunandan used guided surgery (stereolithographic or chrome-cobalt guide), manufactured in two parts and resting on the opposing piriform rim, to ensure accurate implant positioning [13,14]. Recently, modified oncology ZI (Southern Implants, Nobel) have been developed, featuring a polished midsection and a 15 mm apical thread to minimize debris accumulation and facilitate cleaning. In Boyes-Varley's study, on post-maxillectomy defects, they achieved a 100% survival rate [19]. However, Butterworth's study showed these implants doubled the failure risk (HR = 0.56) compared to conventional ZI [10], indicating further research is needed to assess their equivalence.

The literature shows considerable diversity in protocols and techniques for loading ZI. For “classic” maxillary atrophy, immediate loading is generally preferred. In oncologic post-maxillectomy contexts, most studies recommend 2–6 months of healing [9]. According to Scott et al.—who reported a 98% implant survival rate—the prosthesis was fabricated and loaded only six weeks postoperatively, though this early loading involved primary implantation, differing from our approach [13]. By comparison, Butterworth recommends an average loading period of 9.3 months for secondary implantations in combined facial defects [10]. We chose a shorter six-month interval, during which the implants were not subjected to mechanical stress. Ethunandan's study proposed a different protocol depending on the extent of the rhinectomy [14].

In both our cases, a bar retention system was used, and no failures were observed after loading. This bar, with its central vertical element, avoided the need for an additional implant in the nasal cavity while minimizing emergence-related stresses. However, Scott et al. preferred magnets, and according to Ethunandan, bar retention was associated with a higher failure rate than magnets (33% vs. 9%, p = 0.0564). These data suggest that while a bar system offers specific advantages, magnets may present a lower risk of failure [13,14].

Managing complications, such as the initial osseointegration failure seen in one of our patients, raises questions about long-term success factors. These were examined in the retrospective study conducted in Salisbury, which found that variables such as age, sex, and tumour histology had no significant influence on implant failure rates. Although failures were more frequent in smokers (13%) than in non-smokers (7%), the difference was not statistically significant. However, the study did show, a correlation between implant location and failure rate: implants placed in the orbital and maxillary rims failed more often than those positioned in the zygoma, nasal floor, or glabella [14]. No specific study evaluates hyperbaric oxygen therapy (HBO) for the survival of ZI after radiotherapy, but Granström et al. reported a significant improvement in extra-oral implant survival in irradiated patients treated with HBO [20]. Based on this Swedish experience, Abu-Serriah et al. proposed HBO before implant placement in irradiated periorbital sites, given its positive impact on irradiated bone, particularly in the absence of other therapeutic alternatives [3]. Mechanical stresses laterally, or the “lever arm” effect, are significantly greater than those on standard dental implants, where occlusal force is usually parallel to the implant axis. These stresses may explain failures in patients who have undergone extensive maxillectomy [13]. In mid-facial defects, however, the main forces involve nasal epithesis retention rather than occlusal loads, potentially accounting for the very low failure rate post-loading, as seen in the Morriston series by Scott et al., where only one failure occurred out of 56 zygomatic implants. Our own results, with long-term survival of all four implants — including one successfully replaced — are consistent with these findings.

Conclusion

Zygomatic implants offer a promising alternative to conventional implants for nasal prosthetic rehabilitation, thanks to their length, which ensures strong bone engagement and promotes osseointegration. Their positioning, often outside the irradiated zone, may reduce failure risk. Although postoperative complications such as osseointegration failure, sinusitis, or facial paraesthesia may occur, as illustrated by the manageable complications observed in our two cases, these did not compromise the overall effectiveness of the rehabilitation.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not for profit sectors.

Conflicts of interest

The authors declare that they have no conflicts of interest related to this study. No external funding, grants, or financial support were received for the conception, execution, or publication of this research. The authors alone are responsible for the content and writing of this article.

Data availability statement

All data analysed in this study were obtained through university access or interlibrary loan. All sources used are appropriately cited in the manuscript.

Author contribution statement

M. Bovis assisted with the surgical procedures, collected the clinical data, and wrote the manuscript. G. Raoul performed the surgeries and critically revised the manuscript. Both authors approved the final version.

Informed consent

Written informed consent was obtained from both patients for the publication of their anonymised data and medical images.

References

All Tables

All Figures

	Fig. 1 Setup on a 3D model of the upper facial structure for patient 1.
In the text

	Fig. 2 Patient 1, Post-operative panoramic radiograph.
In the text

	Fig. 3 Patient 1, post-operative teleradiograph, after maxillary edentation and placement of the prosthetic bar.
In the text

	Fig. 4 Patient 2, post-operative panoramic radiograph.
In the text

	Fig. 5 Patient 2, post-operative frontal radiograph (Blondeau view).
In the text

	Fig. 6 Patient 2, Placement of the prosthetic bar 6 months after surgery, Intrados of the nasal epithesis, Nasal epithesis in place, clipped to the implant-supported bar.
In the text