Journal Logo

Reconstructive: Head and Neck: Original Articles

Reconstruction of Posterior Mandibulectomy Defects in the Modern Era of Virtual Planning and Three-Dimensional Modeling

Chang, Edward I. M.D.; Boukovalas, Stefanos M.D.; Liu, Jun Ph.D.; Largo, Rene D. M.D.; Hanasono, Matthew M. M.D.; Garvey, Patrick B. M.D.

Author Information
Plastic and Reconstructive Surgery: September 2019 - Volume 144 - Issue 3 - p 453e-462e
doi: 10.1097/PRS.0000000000005954
  • Free

Abstract

Reconstruction of the mandible following oncologic mandibulectomy has evolved tremendously, with an increasing trend toward using vascularized bone rather than soft tissue for mandible reconstruction in an effort to maximize function and aesthetics. Since the free fibula osteocutaneous flap was first described by Hidalgo for bony reconstruction of composite mandibulectomy defects, the free fibula flap has become the workhorse flap for mandible reconstruction and has witnessed significant advances in flap design, particularly with regard to the skin paddle.1,2 Anatomical studies coupled with vascular imaging and increased comfort with perforator flaps have significantly increased the reliability of the skin paddle of the free fibula osteocutaneous flap.3–5 Current microsurgical techniques have even allowed the design of multiple skin islands based on separate perforators.6,7 Most recently, computer-aided design and manufacturing (CAD-CAM) techniques have been used to further optimize fibula flap harvest, which has truly revolutionized mandible reconstruction, providing customized, patient-specific cutting guides and milled titanium plates.8–10

Although studies have advocated its use in craniofacial reconstruction, none has specifically examined its impact on reconstruction of defects where the condyle is resected. In such scenarios, many surgeons still advocate for routine or selective use of soft-tissue flaps such as the anterolateral thigh or vertical rectus abdominis musculocutaneous flap for reconstruction of posterior mandibulectomy rather than using a vascularized bone flap. Studies using soft-tissue flaps, particularly when the condyle has been sacrificed, have demonstrated functional outcomes comparable to bony reconstruction.11–13 However, these studies were performed before the age of virtual planning and rapid prototype modeling, which may offer additional advantages in challenging defects with exophytic tumors and pathologic fractures requiring multiple osteotomies, improving the aesthetic outcomes and optimizing the potential for dental restoration and occlusion. We hypothesize that reconstruction using soft-tissue remains comparable to bony reconstruction, with the null hypothesis that medical modeling and virtual planning have no significant benefit in planning bony reconstruction of posterior mandible defects. The present study aims to evaluate outcomes for patients undergoing microvascular reconstruction of posterior mandible defects, comparing soft-tissue free flaps versus free fibula flaps with and without CAD-CAM.

PATIENTS AND METHODS

A retrospective review of a prospectively maintained database was performed for all patients who underwent reconstruction of posterior mandible defects from 2005 to 2016 after institutional review board approval was obtained. A posterior mandibulectomy was defined as any mandible resection where the condyle was removed; however, all defects involving the midline were preferentially reconstructed with bone flaps. All soft-tissue flaps were performed without the use of titanium plates. Patients who underwent reconstruction using a pedicle flap were excluded. Patients’ electronic medical records were reviewed for pathologic diagnosis, demographics, comorbidities, smoking status, radiation treatment, surgical treatment, timing of reconstruction, revisions, and complications. Postoperative functional outcomes including speech intelligibility, diet status, malocclusion, and mouth opening were also recorded. Speech and swallowing function and diet were determined based on evaluations by certified speech and swallow therapists monthly for the first 3 months, then every 3 months for the first year, and then as needed. Functional assessments were scored as described previously.14,15 Briefly, speech outcomes were graded on a scale of 0 to 3 (where 0 = unable to speak, 1 = dysarthric, 2 = 50 to 80 percent intelligible, and 3 = 100 percent intelligible), and eating and swallowing function were also graded on a scale of 0 to 3 (where 0 = tube feed dependent, 1 = tube feed supplementation, 2 = soft diet only, and 3 = regular diet). Patients with less than 6 months’ follow-up were excluded from functional analysis.

Patient Outcomes

The overall complication rate was defined as the incidence of patients who experienced at least one postoperative surgical complication. Complications were divided into reconstructive failures (i.e., flap loss and removal of construct), recipient-site complications (i.e., infection, delayed wound healing, hematoma, seroma, fistula), and donor-site complications (i.e., infection, delayed wound healing, hematoma, seroma). Complications not related to surgery and one of the operative sites were classified as medical complications. Patients were followed postoperatively at least monthly after discharge for 3 months, every 3 to 6 months until 1 year, and then at least yearly thereafter.

Surgical Technique

All mandible reconstructions were performed by 24 different faculty over the 12-year period, and the reconstructive modality was based on the discretion of the reconstructive surgeon. All defects involving the midline were reconstructed with bone. A second soft-tissue flap was typically performed if the posterior mandibulectomy also included an extensive soft-tissue defect and for through-and-through defects. Although there is some variability in surgical technique, the harvest of the fibula was relatively uniform among the group as described previously.4,6 Briefly, the anterior skin incision was made, and the skin was elevated in a suprafascial plane until the peroneus longus tendon and then the fascia was incised to identify the perforators to the skin paddle. The peroneus longus muscle was released from the fibula until the anterior septum, which was then divided to enter the anterior compartment. The anterior compartment muscles were dissected away from the fibula, with careful attention to identify the anterior tibial neurovascular bundle. The interosseus membrane was then isolated and the distal and proximal osteotomies were performed. After release of the interosseus membrane, the distal peroneal vessels were ligated, and the muscles were dissected away from the pedicle, which was dissected to the takeoff from the tibioperoneal trunk.

For the free-hand technique, a template was made of the native mandible and a titanium plate was bent accordingly. Sterile rulers or tongue depressors were cut to the appropriate lengths and angles for each fibula segment, and the osteotomies were performed, making certain that malleable retractors were placed to protect the pedicle. Additional burring may be performed to obtain a more precise fit to the titanium plate. For the CAD-CAM approach, milled titanium plates were provided with the cutting guides for the mandible and the fibula. The cutting guide was secured to the fibula using monocortical screws, and the osteotomies were performed also protecting the pedicle with malleable retractors. All plates were placed at the inferior border of the reconstruction, and no double-barrel fibulas were performed.

Statistical Analysis

Descriptive statistics such as means, standard deviations, median, and range were used to summarize age, body mass index, operative time, and functional scores. Frequencies and percentages were used to summarize the categorical clinical characteristics and outcomes. The Pearson chi-square test and Fisher’s exact test were used to assess associations between categorical variables. The Kruskal-Wallis test was used to compare operative time and functional scores among the three cohorts. Overall survival time was defined as the time interval from operation to the death or the last contact date, whichever occurred first. Cumulative survival rates were estimated using the Kaplan-Meier product limit method. A logistic regression model was used to evaluate the risk factors of binary outcomes. An ordinal logistic regression model was used to assess the effects on speech and diet. The Tukey-Kramer method was used to adjust multiple comparisons. The Cox proportional hazard regression model was used to estimate the adjusted hazard ratios. A backward selection algorithm was used to fit the best multivariable model. All tests were two-sided. A value of p < 0.05 was considered significant. The analyses were performed in SAS 9.2 (SAS Institute, Inc., Cary, N.C.) by a senior statistician (J.L.).

RESULTS

A total of 291 patients underwent reconstruction of a posterior mandibulectomy defect during the study period. The average follow-up time was 41.8 months. The average patient age was 56.9 years and the average body mass index was 26.2 kg/m2. More than half of the patients had a history of smoking [n = 164 (56.4 percent)], 119 (40.9 percent) patients received radiation therapy, 107 (36.8 percent) received chemotherapy, and 63 patients (21.6 percent) received both chemotherapy and radiation therapy. Fifty-nine patients received radiation therapy and 32 patients had chemotherapy before the resection and reconstruction. Overall, 122 patients (41.9 percent) underwent bony reconstruction, whereas 169 patients (58.1 percent) underwent soft-tissue reconstruction. The free fibula flap was used in 95.9 percent of the bony reconstructions (n = 117). The free anterolateral thigh flap was the most popular of the soft-tissue flap reconstructions (n = 144). Thirty-two of the 122 bony reconstruction patients (26.2 percent) used CAD-CAM, and all of these were free fibula flaps (Fig. 1). There were significant differences in average age, hypertension, and diabetes among the three cohorts (Table 1).

Table 1. - Patient Characteristics and Demographics
Variable Bone without CAD-CAM (%) Bone with CAD-CAM (%) Soft Tissue (%) p * p
No. of patients 90 32 169
Age, yr <0.001 <0.001
 Mean ± SD 53.3 ± 17.1 44.1 ± 20 61.3 ± 14.5
 Median 56.2 46.4 63.1
 Range 3–89.3 8.3–77.7 13.5–90
BMI, kg/m2 0.588 0.684
 Mean ± SD 25.5 ± 5.1 27.6 ± 8.3 26.4 ± 6.7
 Median 25 25 25.5
 Range 15–40 16–55 13–61
Length of follow-up, mo 0.679 0.454
 Mean ± SD 42.5 ± 39.8 35.9 ± 36.5 42.5 ± 37.4
 Median 27.7 20.7 27.1
 Range 0.3–140.7 2.1–146.3 1.3–152.1
Sex
 Male 35 (38.5) 13 (40.6) 50 (29.6)
 Female 56 (61.5) 19 (59.4) 119 (70.4) 0.235 0.110
HTN 59 (64.8) 25 (78.1) 76 (45) <0.001 <0.001
CVA 5 (5.6) 1 (3.1) 6 (3.6) 0.822 0.566
CHF 1 (1.1) 0 (0) 2 (1.2) 0.999 0.999
Coronary artery disease 16 (17.8) 1 (3.1) 29 (17.2) 0.097 0.457
DM 7 (7.8) 2 (6.3) 31 (18.3) 0.031 0.009
PVD 3 (3.3) 0 (0) 5 (3) 0.880 0.999
MO 1 (1.1) 3 (9.4) 7 (4.1) 0.090 0.703
Smoker 50 (55.6) 14 (43.8) 104 (61.5) 0.148 0.122
XRT 29 (32.2) 12 (37.5) 78 (46.2) 0.087 0.032
Chemotherapy 28 (31.1) 14 (43.8) 65 (38.5) 0.347 0.481
Both XRT and chemotherapy 13 (14.4) 7 (21.9) 43 (25.4) 0.103 0.174
CAD-CAM, computer-aided design and manufacturing; HTN, hypertension; CVA, cerebrovascular accident; CHF, congestive heart failure; DM, diabetes mellitus; PVD, peripheral vascular disease; MO, morbid obesity; XRT, radiation therapy.
*Calculated by comparing variables among the three groups.
Calculated by comparing the variable between patients with and without bone used in reconstruction.

Fig. 1.
Fig. 1.:
Virtual planning for a patient undergoing posterior mandibulectomy demonstrating resection and planned reconstruction with a free fibula osteocutaneous flap.

Regarding operative factors, there were significant differences in the total average operative times that included the resection and reconstruction (Table 2). Of the three groups, CAD-CAM bony reconstruction took the longest (p < 0.001). When comparing total operative time for bony reconstruction with and without CAD-CAM, CAD-CAM bony reconstruction was significantly longer (CAD-CAM, 635.3 ± 132.7 minutes; free hand, 564.7 ± 182.2 minutes; p = 0.022). Further analysis of the number of osteotomies performed between patients undergoing bony reconstruction demonstrated an equivalent average number of osteotomies performed regardless of whether medical modeling was used (p = 0.84). Eighty-two patients (28.1 percent) were able to undergo placement of dental implants following reconstruction. Although there was no significant difference among the three cohorts, patients who underwent a CAD-CAM free fibula reconstruction had the highest rate of dental rehabilitation.

Table 2. - Details of Operation and Surgical Factors
Variable Bone without CAD-CAM (%) Bone with CAD-CAM (%) Soft Tissue (%) p * p
Flap type
 Single flap 64 (71.1) 24 (75) 163 (96.4)
 Double flap 26 (28.9) 8 (25) 6 (3.6) <0.001 <0.001
Dental implants
 No 61 (67.8) 19 (59.4) 124 (73.4)
 Yes 26 (28.9) 13 (40.6) 43 (25.4) 0.224 0.195
Osteotomies 0.840
 Mean ± SD 1.9 ± 1.2 1.8 ± 0.8
 Median 1 2
 Range 0–6 0–4
Operative time, min 564.7 ± 182.2 635.3 ± 132.7 498.6 ± 173.6 <0.001 0.022
CAD-CAM, computer-aided design and manufacturing.
*Calculated by comparing variables among the three groups.
Calculated by comparing the variable between patients with and without bone used in reconstruction.

Complications

Overall, 107 patients (36.8 percent) developed a complication following reconstruction (Tables 3 and 4). The majority of complications involved the head and neck recipient site, with only 31 patients (10.7 percent) suffering complications in the donor site. Comparing the three cohorts, patients without CAD-CAM had the highest number of overall complications, particularly hematoma of the recipient site. However, the number of donor-site complications was highest with patients undergoing CAD-CAM–assisted bone reconstruction. There were no significant differences in total flap losses between the soft-tissue and bony reconstructions. There were six total flap losses, four of which were soft-tissue flaps, and two were fibula osteocutaneous flaps. Multivariate analysis did demonstrate an increased risk of complications in patients undergoing a bony reconstruction (adjusted OR, 2.05; 95 percent CI, 1.25 to 3.38; p = 0.005).

Table 3. - Summary of Postoperative Complications
Variable Bone without CAD-CAM (%) Bone with CAD-CAM (%) Soft Tissue (%) p * p
Overall complication 42 (46.7) 13 (40.6) 52 (30.8) 0.037 0.012
Infection 16 (17.8) 2 (6.3) 16 (9.5) 0.084 0.166
Wound healing 11 (12.2) 4 (12.5) 16 (9.5) 0.742 0.440
Fistula 1 (1.1) 1 (3.1) 10 (5.9) 0.171 0.080
Seroma 4 (4.4) 0 (0) 3 (1.8) 0.352 0.458
Hematoma 7 (7.8) 1 (3.1) 2 (1.2) 0.017 0.019
Total flap loss 2 (2.2) 0 (0) 4 (2.4) 1.000 1.000
Partial flap loss 1 (1.1) 2 (6.3) 1 (0.6) 0.083 0.313
Reoperation 6 (6.7) 2 (6.3) 8 (4.7) 0.689 0.501
Donor site 10 (11.1) 8 (25) 13 (7.7) 0.014 0.054
Medical 1 (1.1) 0 (0) 6 (3.6) 0.433 0.245
Death within 30 days 2 (2.2) 0 (0) 0 (0) 0.175 0.175
CAD-CAM, computer-aided design and manufacturing.
*Calculated by comparing variables among the three groups.
Calculated by comparing the variable between patients with and without bone used in reconstruction.

Table 4. - Univariate and Multivariate Analysis of Complications
Variable Univariate Analysis Multivariable Analysis
OR (95% CI) p OR (95% CI) p
Bone without CAD-CAM vs. ST 1.97 (1.16–3.34) 0.012 2.18 (1.26–3.75) 0.005
Bone with CAD-CAM vs. ST 1.54 (0.71–3.35) 0.277 1.73 (0.78–3.84) 0.179
Bone vs. ST 1.85 (1.14–2.99) 0.013 2.05 (1.25–3.38) 0.005
Female vs. male 2.27 (1.32–3.91) 0.003 2.50 (1.43–4.36) 0.001
CAD-CAM, computer-aided design and manufacturing; ST, soft tissue.

Functional Outcomes

Certified speech and swallow therapists documented functional outcomes in 285 patients (97.6 percent) (Table 5). Overall, diet function was significantly different among the three study groups (p = 0.002). Multivariate analysis demonstrated that patients with CAD-CAM–assisted bony reconstruction had a significantly higher probability of tolerating a regular diet compared to patients with soft-tissue reconstruction (adjusted OR, 3.25; 95 percent CI, 1.35 to 7.83; p = 0.009). Diet function was not significantly different between bony reconstruction with and without CAD-CAM (p = 0.153). Radiation therapy and increased age were associated with significantly worse overall diet function on multivariate analysis [adjusted OR, 0.43 (95 percent CI, 0.26 to 0.71), p < 0.001; and adjusted OR, 0.97 (95 percent CI, 0.95 to 0.98), p < 0.001, respectively] (Table 6).

Table 5. - Summary of Postoperative Functional Outcomes
Variable Bone without CAD-CAM (%) Bone with CAD-CAM (%) Soft Tissue (%) p * p
Diet 0.002 0.012
 Mean ± SD 1.7 ± 1.2 2.3 ± 1.1 1.5 ± 1.1
 Median 2 3 2
 Range 0–3 0–3 0–3
Speech 0.051 0.120
 Mean ± SD 2.6 ± 0.8 2.9 ± 0.4 2.5 ± 0.8
 Median 3 3 3
 Range 0–3 1–3 0–3
Malocclusion 36 (40) 5 (15.6) 112 (66.3) <0.001 <0.001
 Mean ± SD 0.5 ± 0.5 0.1 ± 0.3 0.8 ± 0.5
 Median 0 0 1
 Range 0–2 0–1 0–2
Trismus 32 (35.6) 10 (31.2) 43 (25.4) 0.332 0.718
Survival 35 (38.9) 9 (28.1) 108 (63.9) 0.014 0.004
CAD-CAM, computer-aided design and manufacturing.
*Calculated by comparing variables among the three groups.
Calculated by comparing the variable between patients with and without bone used in reconstruction.

Table 6. - Univariate and Multivariable Ordinal Logistic Regression Model of Postoperative Diet
Variable Univariate Analysis Multivariable Analysis
OR (95% CI) p OR (95% CI) p
Bone without CAD-CAM vs. ST 1.37 (0.83–2.27) 0.215 1.16 (0.67–2.00) 0.593
Bone with CAD-CAM vs. ST 4.63 (2.06–10.39) <0.001 3.25 (1.35–7.83) 0.009
Bone vs. ST 1.83 (1.45–2.91) 0.011 1.47 (0.82–2.44) 0.140
Preoperative XRT 0.40 (0.25–0.65) <0.001 0.43 (0.26–0.71) <0.001
Dental implants 2.61 (1.38–3.71) 0.001 1.87 (1.11–3.14) 0.018
Age 0.96 (0.95–0.98) <0.001 0.97 (0.95–0.98) <0.001
CAD-CAM, computer-aided design and manufacturing; ST, soft tissue; XRT, radiation therapy.

Overall speech function also favored CAD-CAM–assisted reconstruction, although the difference did not reach statistical significance (p = 0.051). Multivariate analysis demonstrated significant benefits of bony reconstruction over soft-tissue reconstruction (adjusted OR, 2.16; 95 percent CI, 1.09 to 4.27; p = 0.028), in particular, CAD-CAM–assisted reconstruction over soft-tissue alone (adjusted OR, 8.69; 95 percent CI, 1.76 to 42.88; p = 0.008). Other factors associated with worse speech function are demonstrated in Table 7.

Table 7. - Univariate and Multivariable Ordinal Logistic Regression Model of Postoperative Speech Function
Variable Univariate Analysis Multivariable Analysis
OR (95% CI) p OR (95% CI) p
Bone without CAD-CAM vs. ST 1.22 (0.67–2.23) 0.519 1.67 (0.82–3.37) 0.156
Bone with CAD-CAM vs. ST 5.39 (1.23–23.55) 0.025 8.69 (1.76–42.88) 0.008
Bone vs. ST 1.58 (0.89–2.80) 0.119 2.16 (1.09–4.27) 0.028
Preoperative XRT 0.41 (0.24–0.72) 0.002 0.38 (0.21–0.69) 0.002
CAD 0.37 (0.19–0.73) 0.004 0.37 (0.18–0.76) 0.007
Total flap loss 0.16 (0.03–0.80) 0.026 0.14 (0.03–0.72) 0.019
Double flap 0.41 (0.20–0.84) 0.016 0.21 (0.09–0.51) <0.001
CAD-CAM, computer-aided design and manufacturing; ST, soft tissue; XRT, radiation therapy.

Regarding trismus, there was no significant difference among the three cohorts (p = 0.33), although patients undergoing soft-tissue reconstruction had the lowest incidence of trismus. Multivariate analysis demonstrated an increased risk of trismus with bony reconstruction, although the difference was not statistically significant (adjusted OR, 1.78; 94 percent CI, 0.99 to 3.22; p = 0.056). Pairwise comparison showed no significant difference in the incidence of trismus between patients who underwent a bony reconstruction with and without CAD-CAM (p = 0.61). However, radiation therapy and chemotherapy were both significantly associated with an increased risk of trismus (adjusted OR, 2.03; 95 percent CI, 1.10 to 3.75; p = 0.024; and adjusted OR, 2.02; 95 percent CI, 1.07 to 3.80; p = 0.030, respectively).

Analysis for malocclusion demonstrated the highest rates of malocclusion with soft-tissue reconstruction followed by bony reconstruction without CAD-CAM (p < 0.001). Multivariate analysis strongly favored bony over soft-tissue reconstruction in minimizing the risks of malocclusion (adjusted OR, 0.17; 95 percent CI, 0.10 to 0.31; p < 0.001). Further analysis demonstrated that, between the two bony reconstruction cohorts, CAD-CAM–assisted reconstruction significantly reduced the risk of malocclusion (adjusted OR, 0.07; 95 percent CI, 0.01 to 0.4; p = 0.001).

Survival

Overall survival was significantly different among the three cohorts (p = 0.013) (Fig. 2). Although there was no difference in overall survival between the two bony reconstruction cohorts, overall survival favored bony reconstruction over soft-tissue reconstruction (Fig. 3). However, using a multivariable Cox proportional hazards model, although patients undergoing bony reconstruction trended toward improved overall survival compared with patients undergoing soft-tissue reconstruction, the difference did not reach statistical significance (adjusted hazard ratio, 0.67; 95 percent CI, 0.45 to 1.01; p = 0.053).

Fig. 2.
Fig. 2.:
Kaplan-Meier curves demonstrating overall survival in the three cohorts. CADCAM, computer-aided design and manufacturing.
Fig. 3.
Fig. 3.:
Kaplan-Meier curves demonstrating improved survival in patients undergoing bony reconstruction compared with soft-tissue reconstruction for posterior mandible defects.

DISCUSSION

Reconstruction of complex head and neck defects aims to restore both form and function. This is especially important in patients undergoing reconstruction following a mandibulectomy; however, there is still some debate regarding the optimal means of reconstructing mandible defects when the condyle is sacrificed. Prior studies have also demonstrated equivalent outcomes using soft-tissue flaps in the setting of a posterior mandibulectomy, but these studies are not without limitations.8–10 Our own prior prospective experience demonstrated comparable speaking and swallowing function, but superior cosmetic outcomes were achieved with a bony reconstruction compared with soft-tissue flaps.11 Although the aesthetic scores favored bony reconstruction, the difference was not statistically significant. However, one of the limitations of our prior and current study is the lack of an objective assessment of aesthetic outcomes using a validated measure such as the FACE-Q.11 We also found that reconstruction using a free osteocutaneous flap was more commonly performed in younger patients with lower American Society of Anesthesiologists scores, suggesting the practice of reserving a more complex and potentially longer operation for younger, healthier patients.8 Prior studies have demonstrated that virtual surgical planning has allowed free fibula flaps to be performed more efficiently in more complex cases, without increased risk of complications.12,13 However, the present study demonstrated contradictory findings with the longest operative times, although functional outcomes were superior.14,15

Patients who underwent bony reconstruction using CAD-CAM demonstrated a higher likelihood for a regular diet without the need for long-term tube feed diet supplementation and also superior speech function compared with the soft-tissue flap and free-hand fibula cohorts. The precise correlation between medical modeling and speaking and eating is not entirely clear, but we hypothesize the superior function is attributable to less malocclusion and higher rates of dental rehabilitation in the CAD-CAM cohort. With the evolution of virtual planning, the placement of immediate dental implants at the time of reconstruction is currently an area of active investigation at our institution.

Interestingly, we found that the overall operative times were the longest with the use of CAD-CAM, which differs from other prior studies.10 With only 32 CAD-CAM reconstructions in this study, this may represent a type II (beta) error, as the time recorded in the retrospective study was total operative time, including both the time required for the head and neck surgeon to perform the tumor resection, mandibulectomy, neck dissections, tracheostomy, and placement of percutaneous feeding tubes, and for the plastic surgeon to perform the flap harvest and mandible reconstruction. Because of the retrospective nature of this study, we were unable to isolate the time for the flap harvest and mandible reconstruction alone. However, the increased operative time cannot be attributed to the number of osteotomies performed, which was equivalent between the two groups. Because the adoption of CAD-CAM virtual planning is a relatively new technology, there is definitely a learning curve associated with its use, and it is also more commonly adopted by junior faculty, which may account for the longer operative times. The decision to use medical modeling is at the surgeon’s discretion, so this may reflect more junior surgeons having adopted the technology, whereas more experienced surgeons are able to perform the reconstruction (even those including multiple osteotomies) with equivalent efficiency. Thus, the impact of surgeon experience and surgical technique cannot be discounted when evaluating surgical efficiency and operative time. The issue of performing multiple free flaps may also reflect the impact of surgeon experience, as senior surgeons may be more likely to perform a second flap compared with junior surgeons.

Another potential drawback of using CAD-CAM is the availability and costs of this technology. A significant limitation of this study is the inability to provide an accurate cost analysis between patients who underwent a bony reconstruction with and without medical modeling. The costs of the CAD-CAM approach should also consider the additional preoperative imaging that is obtained, both of the resection and of the donor site. We have routinely ordered preoperative angiograms to delineate not only the vascular and perforator anatomy, but also to orient the patient-specific cutting guides to the perforators supplying the fibula skin paddle.5,15 We are currently in the process of developing a methodology to solve this limitation and hope to provide these data in future studies. Other important issues to consider before adopting CAD-CAM is the turnover time needed to generate the models and customized titanium plate, and the inability to modify the plan intraoperatively. It is imperative that both the reconstructive surgeon and the resecting surgeon participate in the planning session to minimize the likelihood of intraoperative changes of the planned resection. With increased experience in data transfer, coordinated virtual planning, and manufacturing of the models, the customized cutting guides and milled titanium plate can typically be completed in 1 week.8

The study has a number of limitations, including fewer patients who underwent a CAD-CAM reconstruction for posterior mandibulectomies over the study survey period in comparison with the soft-tissue and free-hand fibulas, as the routine use of CAD-CAM free fibula osteocutaneous flap reconstruction is a relatively recent development at our institution. Consequently, there may have been a learning curve associated with using the CAD-CAM approach both for the resecting surgeons and for the reconstructive microsurgeons. The retrospective design of our study also cannot account for more subtle factors such as surgeon experience, technical skill, artistry, and expertise, which may offset the advantages of CAD-CAM technology.16 For more junior reconstructive surgeons and surgeons with less experience, three-dimensional planning and modeling may help to achieve results superior to those achievable with the traditional free-hand technique. However, regardless of the increased operative time and costs, we believe the final outcomes justify considering CAD-CAM as a useful adjunct for posterior mandible reconstruction. Unfortunately, we are not able to provide any comparisons to other techniques such as condylar implants or condylar transplantation, which have demonstrated some promising although conflicting results, because they are not the standard of care at the authors’ institution.17–20 Despite these limitations, the present study provides interesting findings that are relevant for plastic surgeons performing complex microvascular head and neck reconstruction and may help to achieve the best possible functional and aesthetic outcomes following a posterior mandible resection.21

The use of CAD-CAM technology is becoming more common in head and neck reconstruction and now appears to have ushered in a paradigm shift for reconstruction of posterior mandibulectomy defects. Although our practice previously favored the use of soft tissue for the majority of patients with such defects, our current practice has evolved to using CAD-CAM virtual planning and rapid prototype modeling for posterior mandible defects. Given the significant functional benefits, we believe this new approach offsets the added cost and logistical challenges of virtual planning for free fibula flap bony reconstruction of posterior mandible defects.

Conclusions

Either a soft-tissue or osteocutaneous flap can be used to reconstruct a posterior mandibulectomy defect with reasonable outcomes and reliability. However, the use of computer-aided design and modeling appears to significantly improve functional outcomes in comparison with soft-tissue or free-hand bony reconstruction of posterior mandible defects, which reconstructive microsurgeons may wish to consider for this complex patient population.

REFERENCES

1. Hidalgo DA. Fibula free flap: A new method of mandible reconstruction. Plast Reconstr Surg. 1989;84:71–79.
2. Hidalgo DA, Rekow A. A review of 60 consecutive fibula free flap mandible reconstructions. Plast Reconstr Surg. 1995;96:585–596; discussion 597–602.
3. Cordeiro PG, Disa JJ, Hidalgo DA, Hu QY. Reconstruction of the mandible with osseous free flaps: A 10-year experience with 150 consecutive patients. Plast Reconstr Surg. 1999;104:1314–1320.
4. Yu P, Chang EI, Hanasono MM. Design of a reliable skin paddle for the fibula osteocutaneous flap: Perforator anatomy revisited. Plast Reconstr Surg. 2011;128:440–446.
5. Garvey PB, Chang EI, Selber JC, et al. A prospective study of preoperative computed tomographic angiographic mapping of free fibula osteocutaneous flaps for head and neck reconstruction. Plast Reconstr Surg. 2012;130:541e–549e.
6. Chang EI, Yu P. Prospective series of reconstruction of complex composite mandibulectomy defects with double island free fibula flap. J Surg Oncol. 2017;116:258–262.
7. Potter JK, Lee MR, Oxford L, Wong C, Saint-Cyr M. Proximal peroneal perforator in dual-skin paddle configuration of fibula free flap for composite oral reconstruction. Plast Reconstr Surg. 2014;133:1485–1492.
8. Largo RD, Garvey PB. Updates in head and neck reconstruction. Plast Reconstr Surg. 2017;141:271e–285e.
9. Levine JP, Bae JS, Soares M, et al. Jaw in a day: Total maxillofacial reconstruction using digital technology. Plast Reconstr Surg. 2013;131:1386–1391.
10. Chang EI, Jenkins MP, Patel SA, Topham NS. Long-term operative outcomes of preoperative computed tomography-guided virtual surgical planning for osteocutaneous free flap mandible reconstruction. Plast Reconstr Surg. 2016;137:619–623.
11. Hanasono MM, Zevallos JP, Skoracki RJ, Yu P. A prospective analysis of bony versus soft-tissue reconstruction for posterior mandibular defects. Plast Reconstr Surg. 2010;125:1413–1421.
12. King TW, Gallas MT, Robb GL, Lalani Z, Miller MJ. Aesthetic and functional outcomes using osseous or soft-tissue free flaps. J Reconstr Microsurg. 2002;18:365–371.
13. Mosahebi A, Chaudhry A, McCarthy CM, et al. Reconstruction of extensive composite posterolateral mandibular defects using nonosseous free tissue transfer. Plast Reconstr Surg. 2009;124:1571–1577.
14. Monaco C, Stranix JT, Avraham T, et al. Evolution of surgical techniques for mandibular reconstruction using free fibula flaps: The next generation. Head Neck 2016;38(Suppl 1):E2066–E2073.
15. Avraham T, Franco P, Brecht LE, et al. Functional outcomes of virtually planned free fibula flap reconstruction of the mandible. Plast Reconstr Surg. 2014;134:628e–634e.
16. Deek NF, Wei FC. Computer-assisted surgery for segmental mandibular reconstruction with the osteoseptocutaneous fibula flap: Can we instigate ideological and technological reforms? Plast Reconstr Surg. 2016;137:963–970.
17. Tarsitano A, Battaglia S, Ramieri V, et al. Short-term outcomes of mandibular reconstruction in oncological patients using a CAD/CAM prosthesis including a condyle supporting a fibular free flap. J Craniomaxillofac Surg. 2017;45:330–337.
18. Chao JW, Rohde CH, Chang MM, Kutler DI, Friedman J, Spector JA. Oral rehabilitation outcomes after free fibula reconstruction of the mandible without condylar restoration. J Craniofac Surg. 2014;25:415–417.
19. Hidalgo DA. Condyle transplantation in free flap mandible reconstruction. Plast Reconstr Surg. 1994;93:770–781; discussion 782–783.
20. Patel A, Maisel R. Condylar prostheses in head and neck cancer reconstruction. Arch Otolaryngol Head Neck Surg. 2001;127:842–846.
21. Hidalgo DA. Aesthetic improvements in free-flap mandible reconstruction. Plast Reconstr Surg. 1991;88:574–585; discussion 586–587.
Copyright © 2019 by the American Society of Plastic Surgeons