CATEGORY 94 AUSTRALASIAN DENTIST CLINICAL At 12 weeks after placement, both the implant and the fixation screws showed signs of ossification. The same study also analysed the von Mises stress distribution, highlighting how the implant can disperse the physiological load delivered by chewing on the teeth surfaces without areas of implant structure overload. Only Mangano et al.28 have presented results for additively manufactured subperiosteal implants used to rehabilitate atrophic posterior mandibular sectors. The authors reported a 1-year survival rate of 100% in a series of 10 patients. However, the implants used still featured a first-generation design, with the majority of the structure lying just below the surgical wound, rigid fixation placed in the alveolar bone, and no crestal preparation, with the abutments resting on the residual bone profile. Over time the alveolar bone may undergo further resorption with the loss of screws or decubitus of the implant below the mucosa and subsequent exposure. These were the factors behind the significant rate of loss of first-generation subperiosteal implants over the long term15,21. For this reason, certain factors should be particularly taken into account during implant design planning and surgery. The buccal portion of the implant should run as far from the surgical wound as possible to prevent bacterial colonisation and possible exposure. This is a well-established concept in mandibular fracture fixation, where the complication rate goes from more than 10% for the upper plate to less than 1% for the inferior29,30. Furthermore, in this way it is possible to position the osteosynthesis screws in the areas of greatest resistance of the jaw, following the physiological distribution lines of the masticatory and muscular forces specified by Le Fort and Champy31. For this reason, if the positioning of the abutments allows it, the implant should always run below the emergence of the mental nerve. With this configuration of the implant, complete isolation of the nerve and its adequate protection minimise the risk of neurological deficits, which, if present, recover within a few weeks of the operation. In order to avoid traction damage, tissue retraction should be stopped in the downtime of surgery and the closest screw holes should be placed at least 5–8 mm from the mental foramen. The literature does not provide data on the minimum number of osteosynthesis screws required. Not less than four screws per implant were used in this series, which is the number that is sufficient for rigid fixation of favourable mandibular fractures. In any case, the osteosynthesis screws only provide primary stability to the implant and allow for immediate loading21. The masticatory load is distributed and supported by the bone underlying the plate, as demonstrated by finite element analysis18. The abutments should always be housed in special slots created in the alveolar crest and rest on the basal bone, which is less subject to resorption over time. At present, it is not known what effect the distribution of masticatory loads has on the maintenance of bone trophism in subperiosteal implants in the long term. De Moor et al.18, in a prospective 1-year follow- up study, detected an average 0.2 mm of crestal bone resorption around subperiosteal implants in the maxilla; this is similar to the resorption reported in the literature for endosseous implants. Furthermore, removal of the residual alveolar ridge is often necessary to create a regular support base for the abutment, especially in the case of purely horizontal defects. Bone resorption beneath the abutments was evaluated in this study, and no significant variations in the implant/ bone gap were observed over time. The gap detected in the postoperative CT scan at 10 days could be due to minor inaccuracies in the CT processing or implant production, the production of the crestal template, or, more likely, a slight over-preparation of the crest during the creation of the slots for the abutments. This gap was larger in the first implants of the series, where a ball bur was used. In the more recent cases, a larger cylindrical bur was used, which could rest on the edges of the template, allowing more precise preparation. Interestingly, the average gap reduced during the first year after surgery, although not significantly. This could be related to the fact that the implant itself, beneath the abutment, may act as a titanium membrane, preventing the invasion of the gap by connective tissue and allowing the regeneration of the underlying bone. The absence of significant resorption beneath the implants in this study aligns with findings reported by other authors regarding the upper jaw, where porous titanium implants and screws were used16,18–20,28. Similarly, in the cases included in this study, both the implant and the screws were made of grade V porous titanium. This consistency may support the hypothesis that the use of porous titanium devices can mitigate stress shielding effects20. On the lingual side, the connection of the abutments is not possible if the mylohyoid line is very superficial. In fact, its positioning would make it necessary to dissect the muscle from the jaw and, above all, an undercut would be created that could not be crossed by the implant. In the present series, no complications related to the absence of the lingual connection were identified in the short or medium term, but studies on the distribution of loads and finite element analysis will be necessary in the future to evaluate the reliability of this design. Regarding the possible indications, subperiosteal implants are primarily indicated for graftless rehabilitation of severe atrophy of the posterior mandibular sectors, particularly when the residual bone volumes are insufficient for the use of short implants. Short implants have proven to be a safe and reliable technique, yielding satisfactory long-term results9,10,32. Compared to subperiosteal implants, they can be placed using guided and less invasive techniques. Like short implants, subperiosteal implants do not provide for a restoration of the bone height, often necessitating compromises in prosthetic aesthetics and requiring the filling of a larger prosthetic space with longer teeth. However, in many cases, despite a significant vertical bone deficit, the prosthetic space is reduced due to the hyper-eruption of the upper teeth (Fig. 5). In such scenarios, the use of regenerative techniques would result in further reduction of the prosthetic space, making it impossible to accommodate a prosthesis without encroaching on the upper opposing teeth. The primary limitation of this study is the retrospective design and the relatively small number of patients. In the future, prospective studies and randomised trials will be needed to effectively determine the ideal implant design and the number and location of Fig. 5. Case 14 (female, age 49 years): (A) during digital planning, the CT scan showed severe atrophy of the mandible; (B) preoperative view; (C) the definitive prosthesis; note the height of the teeth which, despite the severe vertical atrophy, is not increased, due to the hyper-extrusion of the upper opposing teeth. C B A
RkJQdWJsaXNoZXIy MTc3NDk3Mw==