3D longitudinal evaluation of facial morphology of the surgically managed unilateral cleft lip and palate (UCLP) cases.

Introduction Modern society is passionate about beauty and aesthetics. According to a 2016 survey by the International Society of Aesthetic and Plastic Surgery, the demand for aesthetic surgery is more than ever. People’s perception and awareness of facial aesthetics has increased. Orofacial cleft is the most common facial dysmorphology, with prevalence about 1.46:1000 in Scotland. The aim of the initial surgical repair of cleft lip is to improve facial aesthetics and function, without interrupting facial growth. Nevertheless, facial asymmetry is a stigma in cleft patients, and revision surgery due to a patient or their parents’ dissatisfaction with the outcomes is not uncommon. Objective evaluation of facial asymmetry after primary surgical repair is valuable. It is an indication of the success of surgery, and it informs the surgeon of the magnitude and location of residual asymmetry. The evaluation of facial asymmetry has evolved significantly from landmark-based assessment to surface-based analysis. The latter provides a comprehensive evaluation of facial asymmetry by superimposing the original 3D model on its mirror copy. This permits the quantification and the visualisation of the disparity between the two halves of the face. Many studies evaluated facial asymmetry a few years after primary surgical repair. Longitudinal monitoring and quantifying of facial shape changes can potentially guide the surgeon to the optimal surgical technique. Only a few studies evaluated facial asymmetry before and after primary surgery, and their analysis was dependent on a set of facial landmarks that did not describe the asymmetry of facial surfaces between these landmarks. Unfortunately, the existing literature does not provide comprehensive longitudinal evaluation of facial asymmetry of cleft patients, and the impact of facial expression on residual facial asymmetry has not been fully investigated. Aim The aim of this study was the longitudinal evaluation of facial asymmetry of UCLP patients using an advanced facial analysis tool, and to compare the postoperative residual asymmetry with the control group. The study was carried out to assess the impact of growth and facial expression on residual facial asymmetry. Methodology This study was carried out on 30 UCLP patients. All the patients were Caucasian and underwent the same surgical protocol, which was carried out by the same surgeon at the Royal Hospital of Sick Children, Edinburgh. 3D facial images were captured for the patients, before surgery, at about 4 months after surgery and at four-year follow-up, at rest and at maximum smile using 3dMDface system. Historical data of 70 3D facial images of six-month-old non-cleft infants were also analysed in this study. Facial asymmetry was evaluated using a generic mesh. A generic facial mesh is a mathematical facial mask that consists of 7,190 vertices. The mesh was conformed on each 3D facial image. The conformed meshes were utilised to evaluate facial asymmetry using two methods: the average asymmetry, the total and regional facial asymmetry. The average asymmetry method involved the creation of four average faces for cleft patients: an average preoperative face, an average postoperative face, and two average faces at the four-year follow-up (one at rest and one at a maximum smile). The fifth average face was that of six-month-old non-cleft infants. A mirror copy for each average 3D facial model was created by reflecting it on a lateral arbitrary plane. The original and mirror models were superimposed, the absolute distances between corresponding points on the two surfaces were calculated and analysed in three directions (mediolateral, vertical and anteroposterior), to quantify facial asymmetry. The results were displayed in colour-coded maps. Asymmetry scores were obtained by calculating the median of the absolute distances between corresponding points for the total face, upper lip and nose. The asymmetry scores in the mediolateral, vertical and anteroposterior directions were also quantified. Statistical tests were applied to detect significant differences in asymmetry scores of the total face and each facial region between study groups (before surgery, after surgery and at four-year follow-up), and between surgically managed cleft group and the control group. The correlations of asymmetry scores of the total face, nose and upper lip before surgery, after surgery and at four-year follow-up were also investigated. Results Facial asymmetry in cleft patients was dramatically improved after surgery. However, the postoperative residual asymmetry of UCLP patients was significantly higher than the non-cleft infants in the three directions. Furthermore, facial asymmetry increased during growth, with main impact on the nose. Facial expressions accentuated the residual asymmetry. Specifically, there was considerable shifting of the upper lip toward the scar tissue of the affected side. The residual asymmetry of the nose at the four-year follow-up was correlated to initial nasal asymmetry and residual nasal postoperative asymmetry. The anteroposterior deficiency of the upper lip, nose and paranasal area was pronounced in the cleft group at all time intervals due to insufficient bony support of the cleft maxilla. Conclusions Cleft patients and their parents should be informed of the likelihood of residual asymmetry following surgery. Refinements in primary surgery are necessary. The superficial and deep fibres of the orbicularis muscle have to be accurately repaired according to the direction of the muscle fibres to avoid the shifting of the philtrum of the upper lip toward the scar tissue on the affected side. The orbicularis oris muscle has to be adequately dissected and rotated in the downward direction to eliminate the residual vertical deficiency at the corner of the mouth on the affected side. An incision in the internal lateral side of the nose should be considered to reduce this deficiency. The levator labii superioris alaeque nasi muscle of the cleft side has to be reflected and sutured to the corresponding muscle fibres on the other side, to avoid the residual shifting of the nose to the non-cleft side, and to eliminate the residual vertical deficiency of the alar base on the cleft side during smiling. Revision surgery should be delayed until completion of growth. Before lip revision surgery, it is necessary to evaluate the residual asymmetry when the face is at rest and during facial expressions. Consideration should be given to initial nasal asymmetry and residual postoperative nasal asymmetry. Patients should be informed about the expected need for revision surgery including rhinoplasty. We were able for the first time to quantify facial asymmetry in three directions which provided an insight into the cause of the residual facial asymmetry at rest and at maximum smile. The generic mesh is an innovative tool for the assessment of facial asymmetry