American Association of Plastic Surgeons

Back to 2018 Program


Three-Dimensional Analysis of Donor Masks for Restoration of the Face After Allograft Procurement, A Case Study
Michael J. Cammarata, BS1, Nicole Wake, MS2, Rami S. Kantar, MD1, Margy Maroutsis, .1, Elise C. Schmidt, BS1, William J. Rifkin, BA1, G. Leslie Bernstein, MPA1, Alexes Hazen, MD1, J. Rodrigo Diaz-Siso, MD1, Lawrence E. Brecht, DDS3, Eduardo D. Rodriguez, MD, DDS1.
1Hansjörg Wyss Department of Plastic Surgery, NYU Langone Health, New York, NY, USA, 2Department of Radiology, Center for Advanced Imaging Innovation and Research (CAI2R) and Bernard and Irene Schwartz Center for Biomedical Imaging, New York University School of Medicine, New York, NY, USA, 3Hansjörg Wyss Department of Plastic Surgery, NYU Langone Health; Jonathan and Maxine Ferencz Advanced Education Program in Prosthodontics, New York University College of Dentistry, New York, NY, USA.

PURPOSE: Face transplant teams have an ethical responsibility to restore the donor’s likeness after allograft procurement. This has been achieved with masks constructed from facial impressions, and more recently, three-dimensional (3D) printing. Our aim was to compare the accuracy of conventional impression and 3D-printing technology.

METHODS: A 3D-printed mask was created using an advanced 3D imaging system and polyjet printing technology (Stratasys J750). A silicone mask was made using a conventional impression technique, whereby a mold requiring direct contact with the subject’s face is reinforced by plaster bands and filled with silicone, which then cures at 160°C. Digital 3D models of the subject’s face (Vectra H1 Imaging) and both masks (3dMD) were acquired. Each digital mask was overlaid on the 3D face image using a seven-point registration, and a part-comparison was performed (Materialise 3-matic). The mean linear deviation between each digital mask and face image was compared with a t-test.

RESULTS: The mean linear deviation of the 3D-printed mask compared to the face image was significantly smaller than that of the silicone mask (0.58±0.68mm vs. 3.09±4.69mm, p< 0.001). Material cost and production times were $665, 2 days, 9 hours for the 3D-printed mask, and $100, 10 hours for the silicone mask.

CONCLUSIONS: Despite increased cost and production time, histogram analysis illustrates that the 3D-printed mask offers greater surface accuracy than the silicone mask. This, combined with greater donor resemblance without additional risk to the allograft may make 3D-printed masks the superior choice for face transplant teams.


Back to 2018 Program