In a retrospective review of 207 craniofacial implants placed in 72 patients, the overall survival rate for uncovered craniofacial implants was 80 %, which was significantly lower than that of the conventional dental implant. The detailed evaluation revealed that the cumulative 6-year survival rate for auricular, piriform/nasal, and orbital implants varied and was found to be 92, 87, and 59 %, respectively (Roumanas et al. 2002). Another study reported similar findings in their craniofacial prostheses patient population, i.e., that the survival of total implants in ­auricular, nasal, and orbital in congenital, trauma, and oncologic patients was 94.1, 88.9, and 80.8 %, respectively (Visser et al. 2008). Furthermore, in a study involving 153 implants placed to retain 44 orbital prostheses and followed for a mean period of 52.6 months, the overall integration survival rate was 73.2 % (Toljanic et al. 2005). From these studies, while craniofacial implants may offer significant benefits to patients, orbital bone presents a disadvantageous site for implant treatment, and the long-term stability of orbital implants can be unpredictable. No obvious causes to explain the decreased orbital implant survival were found among factors such as radiation, hyperbaric oxygen therapy, or implant location within the orbital bone (Toljanic et al. 2005).

In a study of craniofacial prostheses retained by implants and adhesive in 10 patients, 5 patients had prostheses retained with implants utilizing a total of 14 implants, and 2 had failed that were placed in irradiated areas (Pekkan et al. 2011). In another study, implants placed in nonirradiated tissues and irradiated tissues were evaluated. In nonirradiated tissues, survival rates for auricular, piriform, and orbital were 94, 80, and 70 %, respectively, with overall survival rate of 85 %. Survival rates in irradiated tissues for auricular, piriform, and orbital were 100, 83, and 27 %, respectively, with overall survival rate of 52 %. Limited numbers of orbital implants in patients were noted (Roumanas et al. 2002). Others reported in nonirradiated fields of the ear, nose, and orbit as 96, 87.5, and 94.1 %, respectively, with overall survival rate of 95.2 %. Implants in irradiated bone were reported as placement pre-irradiation and postirradiation. Survival of implant placement postirradiation of auricular, nasal, and orbital bone was 100, 100, and 28.6 %, respectively. Implant placement survival with pre-irradiation of the auricular, nasal, and orbital bone was 81.8, 83.3, and 79.3 %, respectively (Visser et al. 2008).

These studies indicated that reduction of implant survival in irradiated tissues was rather limited to orbital bone (Fig. 3.4). In particular, orbital implants placed after XRT showed a significantly decreased survival, whereas the orbital implant placed before the radiation therapy sustained their survival albeit at a lower rate than those placed in other craniofacial bones. Case reports describing the failure of orbital bone implants after XRT began to emerge in the 1990s. An increased loss of implants with time after irradiation was noted (Granström et al. 1993).

To date, the cause of the selective reduction in the survival of orbital implants, particularly after XRT, has not been elucidated.

The human orbit is composed of the frontal, maxilla, zygomatic, and lacrimal bones. The orbital plate of the frontal bone occupies the superior margin where orbital implants are placed. This bone is a dense cortical bone. This location is also adjacent to the frontal sinus, which limits the implant length and direction.

Craniofacial implants are shorter then dental implants; typically these implants are about 3–4 mm in length. Craniofacial implants are also different from intraoral implants in that the implant surfaces of currently available craniofacial implants are untreated. Many of these implants have collars on the neck of the implant to prevent overseating during placement. Too deep placement would create damage to the underlying dura mater. The short length may partially explain why survival rates are less than that of intraoral dental implants.

Furthermore, the biomechanics of facial prostheses are different from the cyclic loading that intraoral implants undergo. However, implants placed in orbital bone are often subjected to off axial loading with loads applied perpendicular to the implants. These mechanical and anatomical issues may contribute to the loss of orbital implants.

Anatomically, craniofacial bones are composed of relatively thick cortical bone and undeveloped trabecular bone. The initial implant stability may be established by the mechanical interaction between implant threads and cortical bone. It is believed that craniofacial implants may largely depend on this mechanical fixation. The orbital implants placed in the thin cortical bone may lack this initial fixation.

Bone remodeling is believed to be higher in trabecular bone than cortical bone, and thus, the biological establishment of osseointegration should be active with trabecular bone. In a preclinical animal study, implants were placed in the frontal calvarial region of nine adult pigs, and it was found that the bone-implant contact reached approximately 31 % at 2 weeks, 39 % after 4 weeks, and 51 % at 8 weeks (Petrovic et al. 2007). Implants placed in craniofacial bones should achieve osseointegration. However, the small trabecular bone compartment in craniofacial bones may be less adaptive for biological implant fixation. It is conceivable that radiation therapy reduces the viability of cells in the trabecular space. This may further impair the establishment of biological osseointegration involving bone remodeling.

It is tempting to speculate that osseointegration of craniofacial implants may highly depend on the mechanical fixation with cortical bone, while the de novo osseointegration in the trabecular space may not be effectively established. Although the rate of cortical bone remodeling may be slow, once the bone resorption is initiated, the loss of mechanical fixation may occur.