Pre-surgical CT/FEA for craniofacial distraction: I. Methodology, development, and validation of the cranial finite...

Recently, surgeons have begun to treat serious congenital craniofacial deformities including craniosynostoses with mechanical devices that gradually distract the skull. As a prospective means of treatment planning for such complex deformities, FE models derived from routine preoperative CT scans (CT/FEA) would provide ideal patient specific engineering analyses. The purpose of this study was to assess the dimensional and predictive accuracy of the CT/FEA process through the development of a 3D model of a dry human calvarium subjected to two-point distraction ex vivo. Comparative skull measurements revealed that CT/FEA construction error did not exceed 1% for transcranial dimensions, and the thickness error did not exceed 8.66% or 0.31 mm. CT/FEA strain predictions for the central region of the skull, between the distraction posts, were not statistically different from homologous gage values at P < 0.05. Peripherally, however, the strain fields were less well behaved and the FE predictions showed only general qualitative agreement with gage recordings.