Three-dimensional analysis of donor masks for facial transplantation

Background: 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 three-dimensional printing. The authors compare the accuracy of conventional impression and three-dimensional printing technology. Methods: For three subjects, a three-dimensionally-printed mask was created using advanced three-dimensional imaging and PolyJet technology. Three silicone masks were made using an impression technique; a mold requiring direct contact with each subject's face was reinforced by plaster bands and filled with silicone. Digital models of the face and both masks of each subject were acquired with Vectra H1 Imaging or Artec scanners. Each digital mask model was overlaid onto its corresponding digital face model using a seven-landmark coregistration; part comparison was performed. The absolute deviation between each digital mask and digital face model was compared with the Mann-Whitney U test. Results: The absolute deviation (in millimeters) of each digitally printed mask model relative to the digital face model was significantly smaller than that of the digital silicone mask model (subject 1, 0.61 versus 1.29, p < 0.001; subject 2, 2.59 versus 2.87, p < 0.001; subject 3, 1.77 versus 4.20, p < 0.001). Mean cost and production times were $720 and 40.2 hours for three-dimensionally printed masks, and $735 and 11 hours for silicone masks. Conclusions: Surface analysis shows that three-dimensionally-printed masks offer greater surface accuracy than silicone masks. Greater donor resemblance without additional risk to the allograft may make three-dimensionally-printed masks the superior choice for face transplant teams.