The CT data were uploaded to a surgical modeling software system (Medical Modeling, Inc., Golden, Colo.), and CSP was initiated (Fig. 5A). Orbitozygomatic skeletal tangential osteotomies were designed along a steep angle to a superficial coronal plane, and a genioplasty osteotomy was planned. Donor facial skeletal subunits were virtually superimposed onto the recipient-planned osteotomies. Cutting guides were designed, prefabricated, and sterilized for operative use. Recipient preparation, allograft procurement and transfer, and rigid skeletal nasal, mandibular, and bilateral zygomatic subunit fixation were performed.6,10,12 Postoperative CT was performed on POD 8 to evaluate adherence of the surgical outcomes to the preoperative CSP and demonstrated mean positional differences ranging 0.78–1.67 mm and volumetric differences ranging 151–493 mm3 (Table 3) (see figure, Supplemental digital content 2, which demonstrates patient 2. Computerized representation of translational movements of the allograft’s skeletal elements in relation to the recipient skeleton. Distances are measured at 4 reference points on each skeletal element of the allograft relative to the recipient skeleton. Averages are calculated and compared between the CSP and postoperative CT results, http://links.lww.com/PRSGO/B178).
Patient 3 is a 25-year-old man who had sustained extensive ballistic facial injury in June 2016, requiring multiple procedures including maxillary, mandibular, zygomatic, and right orbital floor open reduction and internal fixation with debilitating functional deficits and exposed hardware. In preparation for FT, the patient underwent hardware removal, bilateral naso-orbito-ethmoid osteotomies, medial canthal tendons repositioning, and bilateral orbital floor reconstruction with alloplastic titanium implants. A matching 23-year-old male brain-dead donor was identified, and partial face and double jaw transplantation was performed in January 201813 (Figs. 7–9).
Three-dimensional craniofacial CT scans and formal angiography were obtained for both subjects. The donor CT was obtained after performing dental impressions and placing titanium skeletal anchorage screws. CT data were uploaded to the surgical planning software (ProPlan CMF; Materialise, Inc., Plymouth, Mich.) (Fig. 7) (See Video [online], which demonstrates patient 3. Preoperative CSP for partial face and double jaw transplant). Osteotomies were planned, and customized donor- and recipient-specific cutting guides were designed, 3D-printed, and sterilized.
Bilateral mandibular sagittal split and Le Fort III osteotomies in both the donor and recipient were completed based on the preoperative CSP and using the prefabricated cutting guides. The donor maxillary segment was tailored to a stereolithographic model of the planned recipient skeletal defect. The dentition was placed in a prefabricated dental splint, and the previously placed maxillary and mandibular anchorage screws were used to secure the occlusion. Intraoperative navigation was used to confirm appropriate position of the skeletal segments (Brainlab, Inc., Chicago, Ill.). Rigid fixation of the allograft was completed at the bilateral zygomatic bodies and mandibular segments. Intraoperative and postoperative CT scan on POD2 confirmed allograft positioning and skeletal contact (Fig. 8); sella-nasion–A point angle was 83.3 degrees, a difference of −6.2 degrees from preoperative CSP, sella-nasion–B point angle was 81.8 degrees, a difference of −7.1 from CSP, and the Frankfort–occlusal plane angle was 3.5 degrees, a difference of 8.3 from CSP (Table 4). On POD 11, the patient underwent hyoid and genioglossus advancement for floor of mouth dehiscence and palatal wound dehiscence repair. Despite normal occlusion at the conclusion of FT, he developed class II malocclusion and an open bite. Orthodontic treatment was initiated with successful correction of the occlusion by 10 months post-FT. On POD 108, he underwent left coronoidectomy and open reduction and internal fixation of left mandibular nonunion with appropriate recovery.
Forty-four procedures performed in 43 patients have established the feasibility of FT as a comprehensive reconstructive solution for patients with composite defects not amenable to satisfactory autologous reconstruction.1,27,28 The procedure continues to adapt to increasingly complex clinical scenarios, with opportunities for improvement in technical, functional, and esthetic outcomes. CSP and CAD/CAM technology have been adopted in craniomaxillofacial surgery including oncologic reconstruction, craniosynostosis, and implant-based procedures.29,30 CSP allows for increased accuracy and decreased intraoperative time and decision-making despite potential limitations in cost and availability.29,30 Detailed reports of CSP use in FT are scarce. Our overall evolving experience with FT planning and execution has resulted in a gradual decrease in reliance on cadaveric simulation, from 10 mock transplants and a research procurement before the senior author’s first FT in 2012 to 6 mock transplants and no research procurement before the third FT in 2018. Additionally, our systematic incorporation of CSP into FT has allowed for improved operative efficiency and accuracy by providing valuable information to supplement intraoperative decision-making. It has also allowed for improved communication and a more objective evaluation of surgical outcomes based on cephalometric analysis.
Since the first case in 2005, FT teams have used magnetic resonance imaging and/or 3D CT scans for preoperative evaluation. Devauchelle et al17 delineated the contour of their allograft based on a rigid metallic pattern manufactured on the recipient to match the exact dimensions and shape of the defect. In 2008, Siemionow et al8,18 used a defect template they had pretested in mock cadaveric FTs to facilitate allograft procurement. Reconfirmation of the 3D graft architecture and size was performed using a stereolithographic model based on the patient’s CT. Pomahac et al19 used a 3D skull model to plan donor zygomatic osteotomies corresponding to the anticipated recipient defect. In 2011, Roche, Blondeel, and colleagues used CSP software and 3D-printed cutting guides in a partial FT in a 54 year-old man with ballistic injury.20 For CSP and cutting guide prefabrication, the team used 3D CT images of the patient and his son (chosen for morphologic resemblance) rather than the actual prospective donor. This was due to concerns that donor-derived CSP and printing of customized cutting guides would be time consuming in a potential scenario of hemodynamic instability interfering with the procurement of vital organs. After planning of osteotomies, cutting guides and models of the recipient’s missing facial bones were 3D-printed. The patient had class II malocclusion posttransplantation.20
Our protocol includes both recipient and donor CT data for CSP and CAD/CAM. The process was optimized and quantitatively evaluated through cadaveric simulation before successful translation to the clinical setting. Patients 1 and 3 received maxillomandibular transplants. For patient 1, recipient-specific cutting guides were generated while intraoperative navigation was used to gauge donor osteotomies. Postoperative CT confirmed adherence to the CSP. In patient 3’s case, cutting guides were generated for both donor and recipient, whereas intraoperative CT and navigation were used to guide and confirm skeletal fixation in accordance with the CSP. In both FTs, vascular anastomoses were performed after osteosynthesis with similar ischemia times (4 hours 26 minutes and 4 hours 35 minutes). Operative time was significantly reduced from 36 to 25 hours for patients 1 and 3, respectively. This improvement is partly due to the less extensive nature of patient 3’s procedure, but it can also represent the learning curve with a move toward prefabrication of customized cutting guides for both donor and recipient and the addition of confirmatory intraoperative imaging. Intraoperative judgment and considerations related to scarring, soft tissue dissection, preservation of functional structures, and maximization of bone-to-bone contact required some deviation from the original CSP. Importantly, CSP cannot replace clinical judgment and experience, but it can contribute to increased accuracy and efficiency. Fitting a maxillomandibular allograft to a recipient’s cranial base is particularly challenging, but the inclusion of both jaws and the use of CSP and cephalometric analysis have allowed us to achieve normal occlusion at the completion of FT.4 Yet, both patients 1 and 3 subsequently developed malocclusion. Posttransplant malocclusion is not an uncommon complication in maxillomandibular FT, and its gradual postoperative development has also been observed by other teams.31 A possible contributing factor could be the absence of proprioceptive feedback and motor tone during the early postoperative months as nerves regenerate. During that period, condylar-glenoid and maxillomandibular relationships can change in response to gravity, forces imparted by the soft tissue envelope, and speech and masticatory rehabilitation. For patient 1, orthodontic treatment was started at 5 months post-FT and failed. Le Fort III advancement using CSP was thus performed a month later with successful results. Patient 3’s malocclusion was diagnosed in the second week post-FT and successfully corrected with early initiation of orthodontic treatment, avoiding major orthognathic revision surgery. In the future, preemptive application of orthodontic elastics immediately posttransplantation may be crucial in preventing the gradual development of malocclusion during the slow sensory-motor recovery that occurs in the first 6–9 months. Patient 2’s defect did not involve bone loss, but his facial allograft was designed to include skeletal subunits to augment facial projection while preserving retaining ligaments and muscular insertion sites. CSP was instrumental to the design and safe execution of nasal, zygomatic, and genial osteotomies. Minimal translational and volumetric differences were noted between CSP and postoperative result. Few other teams have used CSP in bone including FTs. In 2016, Lassus et al21 used donor and recipient 3D CT data for CSP and 3D printing of customized cutting guides for a Lefort II–based FT, also completing skeletal fixation before vascular anastomoses. Total surgical time was 19 hours and ischemia time of 3 hours 15 minutes. The patient had temporary temporomandibular joint pain and motion restriction as a result of imperfect mandible positioning. More recent transplants have reportedly used CSP and CAD/CAM, with no detailed description of their CSP methodology in the peer-reviewed literature.22–25
To the best of our knowledge, this is the largest and most detailed series describing CSP and CAD/CAM use in FT. Further validation remains to be performed. As CSP was integrated in our cadaveric rehearsals, comparison to control procedures performed without CSP was not available. Furthermore, the limited worldwide experience in FT includes heterogeneous defects and allograft designs with inconsistent reporting of preoperative preparation, operative execution, and postoperative outcomes, precluding comparative analyses on a larger scale. Two of the cases described in this series represent the only documented use of intraoperative navigation in clinical FT. The use of surgical navigation systems has expanded from intracranial and spinal applications to craniomaxillofacial surgery, providing real-time intraoperative guidance with 1–2 mm precision.32–40 Intraoperative CT has been shown to facilitate immediate revision during orbital, zygomaticomaxillary complex, and mandibular angle fracture repair and bimaxillary repositioning osteotomies.41–43 For patient 1, navigation was used to guide the donor Le Fort III osteotomies. For patient 3, it was combined with intraoperative CT following recipient defect creation for real-time image-guided allograft inset and skeletal fixation. A Computer-Assisted Planning and Execution workstation has been described by Gordon et al44 and piloted in swine and human cadaveric simulations of Le Fort-type face transplants with satisfactory accuracy.45,46 The experimental system seeks to dynamically integrate CSP, CAD/CAM, and surgical navigation. It enables intraoperative revision of CSP based on real-time cephalometric and occlusion analyses and positional tracking of cutting guides, combined with the use of custom prebent fixation plates and palatal splints. While that system has not reached clinical applicability, future iterations may provide more cohesive integration of the technologic elements described in our series for further advances in computer-assisted clinical FT.
A CSP protocol developed through cadaveric simulation and clinical implementation allows for refinement of operative flow, technique, and outcomes in partial and full FT. Standards for functional and esthetic outcomes are bound to evolve with the field’s growth, and computerized planning and execution offer a reproducible approach to FT through objective quality assurance.
Statement of Conformity: This study was conducted in accordance with the Declaration of Helsinki.
1. Rifkin WJ, David JA, Plana NM, et al. Achievements and challenges in facial transplantation. Ann Surg. 2018;268:260–270.
2. Mohan R, Borsuk DE, Dorafshar AH, et al. Aesthetic and functional facial transplantation: a classification system and treatment algorithm. Plast Reconstr Surg. 2014;133:386–397.
3. Lantieri L, Grimbert P, Ortonne N, et al. Face transplant: long-term follow-up and results of a prospective open study. Lancet. 2016;388:1398–1407.
4. Dorafshar AH, Brazio PS, Mundinger GS, Mohan R, Brown EN, Rodriguez ED. Found in space: computer-assisted orthognathic alignment of a total face allograft in six degrees of freedom. J Oral Maxillofac Surg. 2014;72:1788–1800.
5. Dorafshar AH, Bojovic B, Christy MR, et al. Total face, double jaw, and tongue transplantation: an evolutionary concept. Plast Reconstr Surg. 2013;131:241–251.
6. Sosin M, Ceradini DJ, Levine JP, et al. Total face, eyelids, ears, scalp, and skeletal subunit transplant: a reconstructive solution for the full face and total scalp burn. Plast Reconstr Surg. 2016;138:205–219.
7. Kollar B, Pomahac B. Facial restoration by transplantation. Surgeon. 2018;16:245–249.
8. Siemionow MZ, Papay F, Djohan R, et al. First U.S. Near-total human face transplantation: a paradigm shift for massive complex injuries. Plast Reconstr Surg. 2010;125:111–122.
9. Brown EN, Dorafshar AH, Bojovic B, et al. Total face, double jaw, and tongue transplant simulation: a cadaveric study using computer-assisted techniques. Plast Reconstr Surg. 2012;130:815–823.
10. Sosin M, Ceradini DJ, Hazen A, et al. Total face, eyelids, ears, scalp, and skeletal subunit transplant cadaver simulation: the culmination of aesthetic, craniofacial, and microsurgery principles. Plast Reconstr Surg. 2016;137:1569–1581.
11. Bojovic B, Dorafshar AH, Brown EN, et al. Total face, double jaw, and tongue transplant research procurement: an educational model. Plast Reconstr Surg. 2012;130:824–834.
12. Sosin M, Ceradini DJ, Hazen A, et al. Total face, eyelids, ears, scalp, and skeletal subunit transplant research procurement: a translational simulation model. Plast Reconstr Surg. 2016;137:845e–854e.
13. Kantar RS, Ceradini DJ, Gelb BE, et al. Facial transplantation for an irreparable central and lower face injury: a modernized approach to a classic challenge. Plast Reconstr Surg. 2019. Forthcoming.
14. Kantar RS, Rifkin WJ, Diaz-Siso JR, et al. Quality improvement in facial transplantation: standard approach for novel procedures. Plast Reconstr Surg Glob Open. 2018;6:e1653.
15. Kantar RS, Gelb BE, Hazen A, et al. Facial transplantation: highlighting the importance of clinical vigilance in donor selection. Plast Reconstr Surg. 2018;142:611e–612e.
16. Gelb BE, Diaz-Siso JR, Plana NM, et al. Absence of rejection in a facial allograft recipient with a positive flow crossmatch 24 months after induction with rabbit anti-thymocyte globulin and anti-CD20 monoclonal antibody. Case Rep Transplant. 2018;2018:7691072.
17. Devauchelle B, Badet L, Lengelé B, et al. First human face allograft: early report. Lancet. 2006;368:203–209.
18. Siemionow M, Papay F, Alam D, et al. Near-total human face transplantation for a severely disfigured patient in the USA. Lancet. 2009;374:203–209.
19. Pomahac B, Lengele B, Ridgway EB, et al. Vascular considerations in composite midfacial allotransplantation. Plast Reconstr Surg. 2010;125:517–522.
20. Roche NA, Vermeersch HF, Stillaert FB, et al. Complex facial reconstruction by vascularized composite allotransplantation: the first Belgian case. J Plast Reconstr Aesthet Surg. 2015;68:362–371.
21. Lassus P, Lindford A, Vuola J, et al. The Helsinki face transplantation: surgical aspects and 1-year outcome. J Plast Reconstr Aesthet Surg. 2018;71:132–139.
25. Amer H, Jowsey-Gregoire SG, Rosen CB, et al. Mayo clinic’s first face transplant: one year outcome. Transplantation. 2018;102:S433.
26. Mohan R, Fisher M, Dorafshar A, et al. Principles of face transplant revision: beyond primary repair. Plast Reconstr Surg. 2014;134:1295–1304.
27. Lindford AJ, Mäkisalo H, Jalanko H, et al. The Helsinki approach to face transplantation. J Plast Reconstr Aesthet Surg. 2019;72:173–180.
29. Suchyta M, Mardini S. Innovations and future directions in head and neck microsurgical reconstruction. Clin Plast Surg. 2017;44:325–344.
30. Rodby KA, Turin S, Jacobs RJ, et al. Advances in oncologic head and neck reconstruction: systematic review and future considerations of virtual surgical planning and computer aided design/computer aided modeling. J Plast Reconstr Aesthet Surg. 2014;67:1171–1185.
31. Krezdorn N, Alhefzi M, Perry B, et al. Trismus in face transplantation following ballistic trauma. J Craniofac Surg. 2018;29:843–847.
32. Zhang JS, Qu L, Wang Q, et al. Intraoperative visualisation of functional structures facilitates safe frameless stereotactic biopsy in the motor eloquent regions of the brain. Br J Neurosurg. 2018;32:372–380.
33. Zhang S, Gui H, Lin Y, et al. Navigation-guided correction of midfacial post-traumatic deformities (Shanghai experience with 40 cases). J Oral Maxillofac Surg. 2012;70:1426–1433.
34. Du JP, Fan Y, Wu QN, et al. Accuracy of pedicle screw insertion among 3 image-guided navigation systems: systematic review and meta-analysis. World Neurosurg. 2018;109:24–30.
35. He Y, Zhang Y, An JG, et al. Zygomatic surface marker-assisted surgical navigation: a new computer-assisted navigation method for accurate treatment of delayed zygomatic fractures. J Oral Maxillofac Surg. 2013;71:2101–2114.
36. Andrews BT, Surek CC, Tanna N, et al. Utilization of computed tomography image-guided navigation in orbit fracture repair. Laryngoscope. 2013;123:1389–1393.
37. Yu H, Shen SG, Wang X, et al. The indication and application of computer-assisted navigation in oral and maxillofacial surgery-shanghai’s experience based on 104 cases. J Craniomaxillofac Surg. 2013;41:770–774.
38. Azarmehr I, Stokbro K, Bell RB, et al. Surgical navigation: A systematic review of indications, treatments, and outcomes in oral and maxillofacial surgery. J Oral Maxillofac Surg. 2017;75:1987–2005.
39. Marmulla R, Hilbert M, Niederdellmann H. Inherent precision of mechanical, infrared and laser-guided navigation systems for computer-assisted surgery. J Craniomaxillofac Surg. 1997;25:192–197.
40. Strong EB, Rafii A, Holhweg-Majert B, et al. Comparison of 3 optical navigation systems for computer-aided maxillofacial surgery. Arch Otolaryngol Head Neck Surg. 2008;134:1080–1084.
41. Cuddy K, Khatib B, Bell RB, et al. Use of intraoperative computed tomography in craniomaxillofacial trauma surgery. J Oral Maxillofac Surg. 2018;76:1016–1025.
42. Heiland M, Schmelzle R, Hebecker A, et al. Intraoperative 3D imaging of the facial skeleton using the SIREMOBIL iso-C3D. Dentomaxillofac Radiol. 2004;33:130–132.
43. Klatt JC, Heiland M, Marx S, et al. Clinical indication for intraoperative 3D imaging during open reduction of fractures of the mandibular angle. J Craniomaxillofac Surg. 2013;41:e87–e90.
44. Gordon CR, Murphy RJ, Coon D, et al. Preliminary development of a workstation for craniomaxillofacial surgical procedures: introducing a computer-assisted planning and execution system. J Craniofac Surg. 2014;25:273–283.
45. Murphy RJ, Gordon CR, Basafa E, et al. Computer-assisted, le fort-based, face-jaw-teeth transplantation: a pilot study on system feasiblity and translational assessment. Int J Comput Assist Radiol Surg. 2015;10:1117–1126.
46. Hashemi S, Armand M, Gordon CR. Development and refinement of computer-assisted planning and execution system for use in face-jaw-teeth transplantation to improve skeletal and dento-occlusal outcomes. Curr Opin Organ Transplant. 2016;21:523–529.