All patients undergoing mandibular reconstruction with an FFF between January 2013 and December 2018 were consulted. Of these, only those who agreed to participate (written consent) in the study were included ( Figure 1 ). All clinical investigations and procedures were conducted according to the principles expressed in the Declaration of Helsinki. This cross-sectional study was approved by the Ethical Committee of the Technical University of Munich, TUM School of Medicine and Health (Approval No. 459/18S-KK).

Mandibular reconstruction with the FFF was principally performed in a standardized way with regard to surgical approach and raising using cutting guides (either CAD/CAM or ReconGuide) (21). Further, we only use individually preformed 2.0 miniplates for osteosynthesis, which were removed prior to implantation. A volume reduction of the intraoral skin island and vestibuloplasty was performed during plate removal if necessary.

Patients were concomitantly interviewed with two structured questionnaires, designed for this study in our clinic, which focus mainly on the oral rehabilitation itself and are not yet validated. The first questionnaire captured the current medical and especially dental treatment situation aiming at the oral restoration stage, since patients did not attend the clinic for clinical examination and the potential prosthetic restauration is provided ex house. The second one highlighted social participation, including physical impairment and aesthetic satisfaction ( Table 1 ). In addition, the patient’s wishes, treatment course, and satisfaction with the oral rehabilitation were questioned. In accordance with the nature of this cross-sectional study, patients were interviewed and included at different points in their treatment history.

In addition, medical records and general information were registered. Further, the time interval for endosseous implant placement [Straumann SLActive standard or standard plus implants (Straumann GmbH; Freiburg, Germany) or Xive implants (Dentsply Sirona Deutschland GmbH; Bensheim, Germany)], and the duration until consecutive prosthetic restoration were recorded. In this way, analyses of implant success, implant survival, and status of oral rehabilitation were recorded. In coordination with the respective dentist, a removable prosthesis on an implant-supported bar construction was chosen (22). In this study, implant quality of health was defined on the basis of the Pisa consensus classification. Accordingly, success required “optimum conditions” namely <2 mm crestal bone loss and the implant being prosthetically loaded. Survival was defined with a crestal bone loss to range from >2 mm to less than half of the implant length. Whereby failure was defined as an implant impossible to load or with crestal bone loss >4 mm and >50% of implant length. Implants that were lost prior to this study (August 2020) were included in the failure group (23).

Peri-implant bone resorption was assessed according to Kniha et al. at the defined regions of interests D 1–4 using available radiographs ( Figure 2 ) (24) and implant-specific complications were also addressed. The gold standard of evaluation, the single-tooth peri-apical radiograph, was mostly not applicable because of the challenging anatomy in the reconstructed oral cavity. This necessitated the use of panoramic radiographs, which were calibrated using the known metrics of the dental implants used. For comparability and equality in the evaluation, we decided to only use panoramic radiographs.

For categorical data absolute and relative frequencies are shown, and quantitative data are described by mean and standard deviation or median and interquartile range [1^st to 3^rd quartile]. For the analysis of pre- and postoperative differences in patient assessments, the Wilcoxon signed-rank test was used. For the assessment of mean differences between relevant groups, linear regression models were fitted to the data. All statistical tests were performed on an exploratory two-sided 5% significance level. No adjustments were made for multiple testing. For relevant effect sizes, 95% -confidence intervals are presented. Analysis was done with IBM SPSS 27 for Mac software (IBM Corp, Armonk; New York, United States).

The final population consisted of 59 patients, excluding those with maxillary reconstruction or who did not return the questionnaires. The gender distribution, age, and operation as well as hospitalization data of the enrolled patients are shown in Table 2 .

Patients were interviewed with two structured questionnaires, as described in section 3.2. The results of the first questionnaire are following coherently as a part of section 3.7, stating the oral rehabilitation status. The second questionnaire had 5 options to answer per question, punctuating 0–4 points and ascending from the worst to the best possible answer. Questions 4 and 5 asked how confident patients were in familiar and unfamiliar environments. The possible answers aimed at the physical impairment after FFF raising, including lack of proprioception and mobility. Question number 6 sought to determine how satisfied patients were with their appearance and if it impairs their choice of acquaintances or localities they see and go to. The results were generally satisfying since the mode was the best answer to all three questions. While Questions 4 and 6, including moving in familiar environments and the satisfaction of appearance, were only answered with the worst option in 5.3 and 5.5% respectively, question 5, with moving in unfamiliar environments, was answered in 17,5% with the worst option ( Table 3 ).

The desire for a prosthetic reconstruction, the query shown in Tables 1 , 3 , increased significantly in the course of the treatment in a comparison of the pre- and postoperative motivation in our questionnaires (p=0.013). The motivation for oral rehabilitation before and after radiotherapy showed no significant difference (p=0.444) but pointed to an increase. Generally, the planning was welcomed by the patients, although the interval between the mandibular reconstruction and the start of the planning, as well as its duration, were often criticized. Additionally, patients with malignancies (and consecutive adjuvant radiotherapy) were significantly less satisfied with the handling (mean difference of 30 points, 95% confidence interval 6 to 60 points, p=0.046) and care (33 points, 95% ci 3 to 63 points, p=0.031) of the prosthesis.

Dental implant insertion was accomplished in accumulative 22 patients (37.3%) at the time point of analysis, with 15 being implanted in our institution and 7 elsewhere. The median time from reconstructive surgery to dental implantation was 366 days (0–1,262). Completed oral rehabilitation was reported in 63.6% (14/22) of the implanted cases and overall 23.7% (14/59) of enrolled patients at the time of data collection in this cross-sectional study. The median time from reconstructive surgery to oral rehabilitation with a removable prosthesis on an implant-supported bar construction was 750 days (224–1,447). After eight implant losses, a total of 52 dental implants (median 4; range 1–6) in 15 patients were evaluated radiographically. There was no data available for seven out of 22 patients (see below). The incidence of dental implant insertion and oral rehabilitation in context with gender, age, and clinical data is illustrated in Table 4 .

There were 60 implants placed in 15 patients. The other seven out of 22 patients were implanted elsewhere and/or without radiography and consequently without an implant count and evaluation in this chapter. The median duration of surveillance was 32 months (15–56). A total of eight implant losses occurred in three patients, with one patient losing all three implants. So, 52 out of 60 (88.6%) implants were still in situ in August 2020.

We defined survival and success of implants one the basis of the 2008 Pisa consensus in four groups as follows: I (= success), II (= satisfactory survival), III (= compromised survival), and IV (failure) (23). Forty-two implants were classified in group I, seven in group II, one in group III, and two in group IV. Considering that an implant should be loaded and in group I to be completely successful, a total of 28 out of 52 (53.8%) matched those criteria. Including those, that makes a total of 50 out of 60 (83.3%) implants surviving, when looking at the implants only and not considering the loading of them with prosthetics.

The radiographs of the 15 patients with 52 surviving dental implants were assessed according to Kniha et al. (24). Due to incomplete radiography 52 implants and 104 crestal bone levels (D 1) were evaluated in total. Crestal bone loss (D1) was <2 mm in 83, 2–4 mm in 16, and >4 mm in five measurements. Looking at D1 mesial and distal of single implants, 37 of them showed a crestal bone loss <2 mm, 23 ranged from 2–4 mm, and two cases had a bone loss of >4 mm.

Table 5 shows the results of the binary logistic regression analyses of potential confounding factors (age, gender, BMI, indication, flap type, number of FFF segments and adjuvant radiotherapy) on implant survival and incidence of oral rehabilitation.

The fibular bone quality is reported as good and reliable for dental implant insertion and the FFF transfer for mandibular reconstruction has reached meanwhile a highly standardized level. This leads consequently to an increasing number of studies thematizing oral rehabilitation. But this final step is because of diverse reasons surgically demanding in a heterogenous and challenging cohort. An implant-borne prosthesis is in most cases a favorable solution for oral rehabilitation due to the complex anatomy and function of the oral cavity following ablative surgery and bony reconstruction (26). Yet it remains more problematic in the management of the soft tissue since the FFF is often necessary in tumor cases, and therefore a skin island is also required. This is often too voluminous, remains mobile and levels the vestibulum. Ultimately, at least one pre-implantological correction of this unfortunate situation is required (volume reduction and vestibuloplasty) to reduce the secondary occurrence of mucosal or osseous peri-implantitis (27). This situation requires a high level of patient compliance and surgical expertise, as the soft tissues may react differently due to perceived irradiation in the case of malignancies, and one risks a wound-healing disorder with secondary loss of the vascularized bone transplant.

The literature on this topic is consecutively very heterogeneous (in terms of indication, positive radiation history, nicotine abuse, extent of resection, number of bone segments, etc.) and also uses different definitions for survival and success. For this reason, we have opted for the Pisa consensus definition in order to enable a certain degree of comparability in our cross-sectional study, which analyzed 59 patients with a female/male 33.9%/66.1% distribution. This is comparable to a systematic review by Wijbenga et al. (28). But in terms of age, Wijbenga et al. describe a mean of 50.9 years, which is lower than the 60.6 years mean in our study (28). The indication for the bony reconstruction was similar as described by Hundepool et al., with the largest proportion of patients suffering from oral squamous cell carcinoma (45.8% vs. 63%) and the second largest suffering from osteoradionecrosis (28.8% vs. 20%) (29).

Since we designed the questionnaires, the comparability to other literature is limited. We defined the quality of life with the satisfaction with their appearance as well as the feasibility and confidence in participation in daily life with different settings and companions. Schliephake and Jamil pointed out that oncologic surgery for oral cancer led to a “significant decrease in oral function with reduced body image and reduced ability and willingness for social contact” (30). This is endorsed by Löfstrand et al., who described a significantly worsened social functioning and role functioning in the SF-36 compared to reference populations (31). Besides the psychosocial impairment physical problems like ankle instability, weakness, toe contraction and decreased range of motion are occurring (32). Our study suggests increased satisfaction with those parameters. It needs to be pointed out that the oncologic surgery took place years before our questionnaires and study, and so patients have already adapted and accepted the new situation. On the one hand this fact might lead to better results in this questionnaire regarding the will and interest for oral rehabilitation, because the negative memories of the exhausting and debilitating ablative and reconstructive surgery with adjuvant therapy have faded over time. At the time of the study, there was a significant increase in the desire for oral rehabilitation in the included population. On the other hand, it could have been that there was no interest in further surgical interventions that are necessary due to dental implantation (vestibuloplasty, removal of the osteosynthesis plates, implantation and secondary exposure, etc.), the associated surgical risks including loss of the FFF and further hospitalization. Patients with a bony reconstruction of the mandible are scarce. This could lead to insufficient prosthodontic treatment by the private dental practitioner. Fierz et al. pointed out that prosthodontic treatment in tumor patients is close interdisciplinary cooperation, with 13 appointments common (26). The significance of oral rehabilitation is underlined by Hundepool et al., who found statistical significance when it came to social participation in the H&N35 pre- and post-oral rehabilitation (29). Although there is no significant difference in quality of life in terms of flap type in the literature, there is a tendency toward the better in patients with FFF, according to Moubayed et al. (33).

The interval of secondary dental implantation in FFF is in line with the literature in our study (mean time of 12.9 months). Some others reported a shorter interval with 9.4 (34), and others had a longer interval with 17.5 months (29) until dental implant placement. Also, the duration of treatment from bony reconstruction to oral rehabilitation was nearly similar at 23.9 months compared to the 22.4 months described by Parbo et al. (34).

Survival rates reported in the literature range from 87–97% in five- to ten-year surveillance (19, 35, 36). In our cross-sectional study, the survival rate of implants in the median duration of surveillance of 32 months (15–56) was 83.3% which is lower. This could be the result of an older cohort as stated above, although Sendyk et al. meta-analyzed that age is not a risk factor for dental implant loss, at least in the original bone (37). As reported by others, radiotherapy did not negatively influence the success rate (38).

Implant success is defined very heterogeneously in the literature and there is yet to find a gold-standard (39). We defined the success of implants as the implant being in situ and loaded, and furthermore with a bone loss <2 mm, but not exceeding half of the length of the dental implant, according to the 2008 Pisa consensus definition (23). It needs to be underlined that the cohort in this study was known to be challenging in terms of oral rehabilitation. The lower implant success in this study compared to others is mainly due to a lack of oral rehabilitation, which was partly in progress in some enrolled participants. Having this in mind, the duration of surveillance was shorter than in comparable studies like Hundepool et al., which is a cause of the lower prosthetic restoration lowering the success rate (29). Besides that, the duration of treatment from bony reconstruction to oral rehabilitation was nearly similar at 23.9 months compared to 22.4 months described by Parbo et al. (34).

We did not collect the past and current status of smoking and oral hygiene. So apart from the missing clinical examination, the lack of information impedes the determination of possible reasons for the impaired implant survival. However, most implants that were surveilled and did not fail were healing properly, since only two out of 52 did not match the 2008 Pisa criteria (23). Pellegrino et al. described an overall implant success of 95.4% at 12-month follow-up and 73.5% at 60-month follow-up (14), which is an indicator, that long-term success in implant placement in FFF is challenging. In our study, the timing with regard to primary or secondary implantation showed no difference in healing and success. This differs from Panchal et al., who reported in a meta-analysis about implant survival in vascularized bone flaps an estimated implant success rate of 97% in implants placed immediately and 89.9% in delayed implant placement, with a mean follow-up ranging from 14 to 40 months. Also, Hessling et al. noted that implantation in hard tissue reconstruction significantly decreased implant survival compared to the mandible, with the FFF being, in their study, statistically significantly more likely to fail compared to other bone grafts for dental implants (40).

Some limitations of this study include the heterogeneity of the cohort regarding indication and a retrospective analysis of data without clinical examination. The implant survival rate may therefore be overestimated since clinical signs of failure were not evaluated. Furthermore, data is derived exclusively from the questionnaires. Patients could have misunderstood the questions or just did not know what their private dental practitioner is planning. Another problem is the use of panoramic radiographs instead of single-tooth peri-apical radiographs, which is due to the anatomical obstacles after reconstructive surgery. Also, there is no structured radiographic recall protocol that defines marks of when X-ray has been used to inspect the implant. Furthermore, there is a German bureaucracy and administration issue as there exist two types of insurance: public and private insurance (41). This leads in our case to an inadequacy of treatment, as the costs of privately insured patients are usually covered by the insurance provider more quickly. In publicly insured patients, the costs are not always covered and need to be applied for by the surgeon and the dentist beforehand. An application for an exemption must be made in accordance with §28 of the Social Insurance Act V for the covering of costs associated with oral rehabilitation by the insurance. We therefore believe that the direct costs are not necessarily the reason for such a low rate in Germany, but rather the associated bureaucratic and time-consuming effort, as it is not uncommon for expert opinions to have to be prepared by third parties. Although the outcome is physically not impaired (42), the results of our survey indicate a significantly rising will for prosthesis after surgery, therefore resulting in a delay in therapy. In conclusion, the complete oral rehabilitation is mostly just being delayed by this administrative process. Although it does not impair the outcome of the implants, patients are being led down a prolonged path of therapy toward oral rehabilitation, which negatively affects their quality of life. Nowadays bony reconstruction is mostly CAD/CAM planned. The ideal position of the neomandibula (in terms of shape of the original mandible, relation to the maxilla, and crestal bone height for implant placement) is therefore predestined and mostly achieved (43). This is underlined by Tran et al., who stated that with in-house virtual surgical planning FFF were more amenable to dental implants as compared to freehand surgery (44).

Mandibular reconstruction is today a very standardized procedure with good postoperative results regarding symmetry and accuracy. Based on the increased use of CAD/CAM technology and the preoperative planning of a close-to-perfect reconstruction of the tooth-bearing mandibula, the chances of improvements in oral rehabilitation can cautiously be seen positively. Within the limits highlighted in this study, the reconstructive surgeon can achieve a good base for oral rehabilitation when using the free fibula flap, since patient motivation for a prosthesis is increasing significantly. Furthermore, non-OSCC patients are significantly more content with their implant-borne prosthesis. An interdisciplinary approach to ensure a satisfactory process and outcome of oral rehabilitation is necessary. Good guidance for patients after a long therapy path should be the goal of the physician.