Full Facial Allotransplantation Including the Temporomandibular Joints: An Anatomical Study and Surgical Protocol

Introduction: Face allotransplantation cardinally shifted the reconstruction paradigm of unmanageable facial defects. Incorporation of temporomandibular joint (TMJ) in total face allotransplant can further improve functional outcome in these patients. The purpose of our study was to evaluate the vascular supply of TMJ and develop a surgical protocol of a total face allotransplantation including TMJ.

Methods: 100 skulls and mandibles from the Hamann-Todd collection (Cleveland Museum of Natural History) and 100 facial CT scans were examined to evaluate the variability of TMJ dimensions. Intercondylar distance, width, length and height of condyles, mandibular ramus and body dimensions, condylar fossa breadth and anteroposterior dimension, and gonial (GA), intercondylar (ICA) and condyle-symphysis (CSA) angles were measured on dry skulls. Bilateral frontal ramal inclination (FRI) and lateral ramal inclination (LRI) were measured in 3D reconstructed CT scans. Injection of the common carotid arteries with red-colored latex was performed in eight fresh cadaver heads. They were then dissected for visualization of the vascular branches entering the TMJ.  Four fresh cadavers were then used to perform a TMJ-included total face transplantation. Donor allografts were harvested using an extracranial Le Fort III approach combined with a temporal craniotomy to include the total TMJ. Internal carotid and middle meningeal arteries were well preserved. Bilateral sagittal split osteotomy was performed to address any anticipated discrepancy between donor and recipient intercondylar distance. The allograft was transferred to the recipient face and secured with miniplates at the orbital rims and nasal bridge. TMJs were secured on the zygomatic processes of temporal bones by long miniscrews.

Results: Statistically significant difference between genders were observed in all parameters measured on dry skulls except for ICA  and CSA. There was a statistically significant difference of FRI between genders in CT measurements. The mean values, standard deviation and the range of all measurement are provided bellow in male vs female format.

Intercondylar distance: 118.1mm±5 (range 104.6-128mm) vs 112.2mm±6.5 (range 98-123.3mm) p<0.05

Gonion-gonion distance: 99.0mm±6.3 (range 86.6-115.3mm) vs 91.2mm±4.9 (range 82.7-104.5mm) p<0.05

Lateral fossa-lateral fossa points: 120.1mm±4.6 (range 109.5-129.1mm) vs 113.3mm±4.9 (range 101.7-123.2mm) p<0.05

ICA: 141.1º ±10.2 (range 118.9-162.5º) vs 139.7º±10.0 (range 119.3-160.3º), p>0.05

CSA: 64.8º±5.6 (range 53.4-74.6º) vs 64.1º±5.6 (range 54-77.1º), p>0.05

GA: 122.4º±7.8 (range 104.2-147.9º) vs 126.4º±6.2 (range 114.2-144.6º), p<0.05

FRI: 79.3º±4.1 (range 69.7-90.2º) vs 77.1º±3.6 (range 71.3-96.9º), p<0.05

LRI: 82.1º±5.4 (range 69.2-96.8º) vs 82.3º±5.8 (range 114.2-144.6º), p>0.05

The TMJ received 1 – 3 direct branches (maximal diameter 0.7-0.8 mm) from the maxillary and middle meningeal arteries medially, and the superficial temporal artery laterally. Fixation of TMJ on zygomatic processes minimally increased the intercondylar distance on recipient face. However, transplanted joints were located more inferior and anterior compared to their normal anatomical position. Class 1 original donor occlusion was achieved with normal ramal inclination and mandibular range of motion.

Conclusion: We demonstrated that TMJ-included total face allograft procurement and transplantation is technically and functionally feasible and reasonable occlusion, range of motion and lateral excursions is achievable.