Surgical correction of cranial abnormalities, including craniosynostosis, requires knowledge of normal skull shape to appreciate dysmorphic variations. However, the inability of current anthropometric techniques to adequately characterize three-dimensional cranial shape severely limits morphologic study. The authors previously introduced three-dimensional vector analysis, a quantitative method that maps cranial form from computed tomography data. In this article, the authors report its role in the development and validation of a normative database of pediatric cranial morphology and in clinical analysis of craniosynostosis.
Normal pediatric craniofacial computed tomography data sets were acquired retrospectively from the Duke University Picture Archive and Communications System. Age increments ranging from 1 to 72 months were predetermined for scan acquisition. Three-dimensional vector analysis was performed on individual data sets, generating a set of point clouds. Averages and standard deviations for the age and gender bins of point clouds were used to create normative three-dimensional models. Anthropometric measurements from three-dimensional vector analysis models were compared with published matched data. Preoperative and postoperative morphologies of a sagittal synostosis case were analyzed using three-dimensional vector analysis and the normative database.
Three- and two-dimensional representations were created to define age-incremental normative models. Length and width dimensions agreed with previously published data. Detailed morphologic analysis is provided for a case of sagittal synostosis by applying age- and gender-matched data.
Three-dimensional vector analysis provides accurate, comprehensive description of cranial morphology with quantitative graphic output. The method enables development of an extensive pediatric normative craniofacial database. Future application of these data will facilitate analysis of cranial anomalies and assist with clinical assessment.
From the Interdisciplinary Craniofacial Imaging Laboratory, the Department of Radiology, and the Division of Plastic and Reconstructive Surgery, Department of Surgery, Duke University Medical Center, and the School of Medicine and the Department of Biomedical Engineering, Duke University.
Received for publication February 11, 2009; accepted June 29, 2009.
This work was presented in part at the Third International Conference of Advanced Digital Technology in Head and Neck Reconstruction, in Cardiff, Wales, United Kingdom, June 29 through July 1, 2008; and at the 22nd International Congress of Computer Assisted Radiology and Surgery, in Barcelona, Spain, June 25 through 28, 2008.
Disclosure:Dr. Marcus is a consultant for Stryker Leibinger. Funding for this research came from research grants from the Southeastern Society of Plastic Surgeons and Stryker Leibinger, Inc.
Jeffrey R. Marcus, M.D., Duke University Medical Center, Box 3974, Durham, N.C. 27710, firstname.lastname@example.org