Abstract
The scientific literature is sparse on reports that evaluate facial asymmetry after mandibular reconstructive surgery objectively. The aim of this study is to introduce and validate a new method, using three dimensional (3D) stereophotogrammetry, that quantifies soft-tissue facial asymmetry in patients who have undergone mandibular reconstruction. To validate the new method, two observers applied the method on 3D photographs of five patients and five controls. An inter-observer difference of 0.04 mm (−0.08 to 0.17) was found with a measurement error of 0.13 mm. 15 3D photographs of the mandibular reconstructed patients were compared with 24 3D photographs of healthy controls. A significant difference (1.19 mm) in asymmetry was found between patients and controls. It is concluded that this new measuring method is a valid, fast and clinically applicable technique to quantify soft-tissue facial asymmetry. It is concluded that facial symmetry in patients is not restored to the level of the control group with the mandibular reconstruction method applied.
The success of mandibular reconstructive surgery in terms of functional outcome has reached a high level of predictability during the last decade. The shift towards optimizing and analysing aesthetic outcome is gaining more importance. Publications that report aesthetic outcome after mandibular reconstructive surgery are scarce. The available publications report subjective outcomes using a visual analogue scoring scale. Developments in three-dimensional (3D) computer technology now make it possible to assess and quantify aesthetic outcome objectively. 3D computer techniques give surgeons opportunities to plan reconstructive procedures with a more predictive, aesthetic and functional outcome. Before new surgical techniques are deemed ‘successful’, instruments have to be developed to test them objectively. The aim of this study is to introduce and validate a new method, based on 3D stereophotogrammetry, that objectively quantifies soft-tissue asymmetry, and to apply this method to a group of patients who had undergone segmental mandibular reconstruction.
Materials and methods
20 patients who underwent immediate or delayed mandibular reconstruction because of a segmental resection in the period 2000–2009 were included in this study. In all 20 patients the reconstruction was performed with two preshaped titanium 2.3 mm plates (Smart plates; KLS Martin Group, Tuttlingen, Germany), particulate autogenous cancellous bone graft and platelet-rich plasma. A group of 24 healthy volunteers with no history of facial surgery or existing facial deformities served as controls. The controls were selected from a database to match the exact age, sex and age distribution of the patients. Ethical approval was granted and all patients gave their informed consent to this study.
3D photographs were acquired for all patients using a stereophotogrammatrical camera set-up (3dMD face™ System, 3dMD LLC, Atlanta, GA, USA). Patients were positioned in a natural head position and were asked to keep their eyes open and relax their facial musculature. All 3D photographs were taken by an experienced photographer.
A new method to quantify soft-tissue facial asymmetry was applied to the acquired 3D stereophotogrammetry data set. This method consists of four consecutive steps performed in a digital environment.
- Step 1
Removal of confounding regions
The neck, ears and hair were removed in 3dMD patient v3.1.0.3 (3dMD patient™ Software Platform, 3dMD LLC) to exclude confounding regions. The neck was included from the thyroid cartilage up and medially to the sternocleidomastoid muscle ( Fig. 1 ). This adjusted 3D photograph was imported into Maxilim ® (Medicim NV, Mechelen, Belgium).
Fig. 1 Step 1. Removal of confounding regions. The upper two images are raw images. The lower two images are the result after the removal of the confounding regions. - Step 2
Creation of a mirrored 3D photograph
In Maxilim ® , four soft-tissue landmarks were manually identified: left (ExL) and right (ExR) exocanthion, subnasale (Sn) and sellion (Se). A transversal plane was constructed through ExL, ExR and Se. Perpendicular to the transversal plane and through both exocanthi a coronal plane was constructed. Perpendicular to the coronal plane and through Se and Sn, a sagittal plane was constructed ( Fig. 2 ). The sagittal plane was used to create a mirrored 3D photograph.
Fig. 2 Step 2. Creation of a mirrored 3D photograph. ExR (exocanthion right), ExL (exocanthion left), Sn (subnasale), Se (sellion). A (sagittal mirror plane). The mirrored surface is orange. - Step 3
Surface registration of the original and mirrored 3D photograph
The original and the mirrored 3D photograph were matched using a complex surface registration algorithm (Iterative Closest Point Algorithm). In order to be able to perform this registration procedure, on both the original and the mirrored3D photograph the forehead, upper nasal dorsum and zygoma were selected. Using the surface registration tool in Maxilim ® the selected regions of both 3D photographs were matched ( Fig. 3 ).
