Quantitative Facial Asymmetry: Using Three-Dimensional Photogrammetry to Measure Baseline Facial Surface Symmetry

Background: Although symmetry is hailed as a fundamental goal of aesthetic and reconstructive surgery, our tools for measuring this outcome have been limited and subjective. With the advent of three-dimensional photogrammetry, surface geometry can be captured, manipulated, and measured quantitatively. Until now, few normative data existed with regard to facial surface symmetry. Here, we present a method for reproducibly calculating overall facial symmetry and present normative data on 100 subjects. Methods: We enrolled 100 volunteers who underwent three-dimensional photogrammetry of their faces in repose. We collected demographic data on age, sex, and race and subjectively scored facial symmetry. We calculated the root mean square deviation (RMSD) between the native and reflected faces, reflecting about a plane of maximum symmetry. We analyzed the interobserver reliability of the subjective assessment of facial asymmetry and the quantitative measurements and compared the subjective and objective values. We also classified areas of greatest asymmetry as localized to the upper, middle, or lower facial thirds. This cluster of normative data was compared with a group of patients with subtle but increasing amounts of facial asymmetry. Results: We imaged 100 subjects by three-dimensional photogrammetry. There was a poor interobserver correlation between subjective assessments of asymmetry (r = 0.56). There was a high interobserver reliability for quantitative measurements of facial symmetry RMSD calculations (r = 0.91-0.95). The mean RMSD for this normative population was found to be 0.80 +/- 0.24 mm. Areas of greatest asymmetry were distributed as follows: 10% upper facial third, 49% central facial third, and 41% lower facial third. Precise measurement permitted discrimination of subtle facial asymmetry within this normative group and distinguished norms from patients with subtle facial asymmetry, with placement of RMSDs along an asymmetry ruler. Conclusions: Facial surface symmetry, which is poorly assessed subjectively, can be easily and reproducibly measured using three-dimensional photogrammetry. The RMSD for facial asymmetry of healthy volunteers clusters at approximately 0.80 +/- 0.24 mm. Patients with facial asymmetry due to a pathologic process can be differentiated from normative facial asymmetry based on their RMSDs. Clinical Question/Level of Evidence: Diagnostic, II.