Facial Symmetry in Unilateral Cleft Lip and Palate Following Alar Base Augmentation With Bone Graft: A Three-Dimensional Assessment

Transcription

1 Facial Symmetry in Unilateral Cleft Lip and Palate Following Alar Base Augmentation With Bone Graft: A Three-Dimensional Assessment Mark F. Devlin, M.B.Ch.B., B.D.S., F.R.C.S.Ed. (O.M.F.S.), F.R.C.S.Ed., F.D.S.R.C.P.S., Arup Ray, M.B.B.S., M.S., F.R.C.S. (Plastic), Peter Raine, F.R.C.S., Adrian Bowman, B.Sc., Dip. Math. Stat., Ph.D., F.R.S.E., Ashraf F. Ayoub, Ph.D., B.D.S., M.D.S., F.D.S.R.C.P.S., F.D.S.R.C.S. Objective: The aim of this study was to assess the outcome of bone grafting using a corticocancellous block of iliac crest to reconstruct the support for the deformed, volume-deficient alar base in treated patients with unilateral cleft lip and palate (UCLP). The main outcome being measured was nasal symmetry. Design: This was a prospective study using a noninvasive three-dimensional stereophotogrammetry system (C3D) to assess the position of the alar base. Images were captured immediately preoperatively and at 6 months following the augmentation of the alar base with a block of bone graft. These images were used to calculate facial symmetry scores and were compared using a two sample Student s t test to assess the efficacy of the surgical method in reducing facial/nasal asymmetry. Patients: This investigation was conducted on 18 patients with one patient failing to attend for follow-up. The results for 17 patients are presented. Results: Facial symmetry scores improved significantly following the insertion of the bone graft at the deficient alar base (p 0.005). Conclusions: 3D stereophotogrammetry is a noninvasive, accurate, and archiveable method of assessing facial form and surgical change. Nasal symmetry can be quantified and measured reliably with this tool. Bone grafting to the alar base region of treated UCLP patients with volume deficiency produces improvement in nasal symmetry. KEY WORDS: cleft nose, nasal symmetry, stereophotogrammetry Assessment of facial deformity should ideally be quantitative, objective, and easily archived. The Clinical Standards Advisory Group report commissioned by the government to assess the standard of cleft care in the United Kingdom was published in 1998 (Sandy et al., 1998). This highlighted the need for objective outcome assessments following surgery. Recently the results of the Eurocleft multicenter study have been published. One of the findings of the study was that adolescent Dr. Devlin is Consultant Cleft Surgeon, Regional Maxillofacial Unit, Southern General Hospital, Glasgow, Scotland, United Kingdom. Dr. Ray is Consultant Plastic/Cleft Surgeon, Canniesburn Unit, Glasgow Royal Infirmary, Scotland, United Kingdom. Dr. Raine is Consultant Pediatric/Cleft Surgeon, Royal Hospital for Sick Children, Glasgow, Scotland, United Kingdom. Dr. Bowman is Professor of Statistics, University of Glasgow, Scotland, United Kingdom. Dr. Ayoub is Professor of Oral and Maxillofacial Surgery, Dental School, University of Glasgow, Scotland, United Kingdom. This research has been funded by a grant from the Chief Scientist Office of the United Kingdom. Presented at Craniofacial Society of Great Britain and Ireland, Annual Scientific Meeting, Swansea, April Submitted September 2006; Accepted November Address correspondence to: Dr. Mark F. Devlin, Consultant Cleft Surgeon, Regional Maxillofacial Unit, Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, Scotland, United Kingdom. Mark.Devlin@sgh. scot.nhs.uk. DOI: / patients with cleft deformity rate their nasal appearance as being poor. It was also suggested that the assessment of nasal appearance is an area that requires further research. (Semb et al., 2004; Shaw et al., 2005). Aesthetic outcome of cleft surgery is important and previous papers have described methods of objectively assessing aesthetics. Methods of two-dimensional (2D) assessment have included photography and video clips, which are inadequate in assessing the three-dimensional (3D) soft tissue volume deficiencies which often occur in residual cleft deformity (Morrant and Shaw, 1996). These methods also require significant patient cooperation, and standardization can be difficult. Coghlan et al. (1993) and McComb and Coghlan (1996) measured nasal symmetry using a computerized photographic system, which was both objective and quantifiable. Standardization of clinical photographs, however, was a problem. 3D assessment is possible using various methods, such as laser scanning (McCance et al., 1997), direct surface measurement (anthropometry) (Farkas, 1996), surface topography (Leivesley, 1983; Kawai et al., 1990), computed tomography/ magnetic resonance imaging (CT/MRI) (Fisher et al., 1999), and stereophotogrammetry (Ayoub et al., 2003). Direct surface measurement is accurate but time consuming, whereas surface topography may not be accurate enough to detect small changes. 391

2 392 Cleft Palate Craniofacial Journal, July 2007, Vol. 44 No. 4 study has been validated for use in assessment of primary cleft deformity and is applicable to this age group due to its short capture time of 30 to 50 milliseconds. Using this system of stereophotogrammetry (C3D) (Ayoub et al., 1998, 2003; Hajeer et al., 2005) we aimed to quantify the facial symmetry as a whole, and of the nasolabial complex in particular, in patients with residual nasal deformity following previous treatment for unilateral cleft lip and palate (UCLP) deformity (Fig. 1). The aim of the study was to quantify the improvement in this deformity following the augmentation of the alar base with corticocancellous bone graft from the iliac crest. It was an objective to explore the relationship between the volume of bone used to the subsequent improvement in symmetry. Ethical approval for this project was granted before commencement. MATERIALS AND METHODS Patients with previously treated UCLP, seen on the multidisciplinary cleft clinic in Glasgow, were recruited if they had residual nasal deformity and clinical deficiency of underlying bone presenting as alar base asymmetry. 3D Capture and Assessment Protocol FIGURE 1 Typical alar base volume deficiency to be corrected. The use of laser scanning to capture facial morphology has certain disadvantages for use in the pediatric population, as they require a patient to cooperate with the assessment for a prolonged period of time. CT scanning has the disadvantage of exposure to ionizing radiation. The method used in this FIGURE 2 Constructed 3D model within the facial analysis software. The landmarks have been placed. The patient s face was captured using the 3D imaging system before surgery and at 6 months postoperatively. The images were taken in the rest position. Patients were instructed to allow their teeth to gently touch together with their lips relaxed. The method of imaging was based on the use of stereo pairs of digital cameras and a special textured illumination flash providing capture times of 50 milliseconds. This made the system appropriate for use in the pediatric patient. A digital color image was also taken at the same time from either side of the patient and was appended to each stereo pair. This was draped around the 3D model resulting in a natural skin surface appearance. This ultimately produced a photorealistic, life-like model that could be manipulated on-screen in three dimensions using a separate software program (Fig. 2). To carry out the quantitative assessment of facial symmetry 21 anthropometric landmarks (Table 1) were placed on each model. Each of these landmarks was located and the x-, y-, and z- coordinates were registered (Hajeer et al., 2005). The main source of error was in the manual location of the anthropometric landmarks on each model. The error of repeatedly locating the landmarks was calculated by one of the authors (MFD). To assess symmetry, a reflected 3D image was constructed and the degree of mismatch, or asymmetry, between the image and mirror image was measured. The actual difference between each landmark and its corresponding mirror image was measured, then squared. The resulting measurements were added together and an average overall displacement from the original to the reflected image was calculated (Hajeer et al., 2004).

3 Devlin et al., STEREOPHOTOGRAMMETRIC ASSESSMENT OF CLEFT NASAL DEFORMITY 393 TABLE 1 Landmarks Used in the Study Landmark 1 Exocanthion (left) 2 Endocanthion (left) 3 Glabella 4 Soft tissue nasion 5 Endocanthion (right) 6 Exocanthion (right) 7 Superior nostril (right) 8 Alar crest (right) 9 Alar base (right) 10 Subnasion 11 Alar base (left) 12 Superior nostril (left) 13 Alar crest (left) 14 Superior columella (left) 15 Superior columella (right) 16 Pronasion 17 Base of Cupid s bow 18 Peak of Cupid s bow (left) 19 Cheilion (left) 20 Peak of Cupid s bow (right) 21 Cheilion (right) Surgical Correction At the time of the operation, under general anesthetic, a block of corticocancellous bone was harvested from the medial iliac crest. This was applied as an onlay graft to the pyriform rim of the deficient side to augment the alar base via an intraoral incision. This incision varied to include the closure of an oronasal fistula if present. The bone graft was adjusted in both shape and size and placed such that the repositioned alar base of the patient appeared symmetrical with the opposite side. Once the alar bases were judged to be symmetrical the bone volume was measured using a technique based on water displacement (Devlin et al., 2003) and the bone was placed and fixed with a suitably long positional screw from the 1.7-mm maxillofacial plating kit (Fig. 3). Intraoral closure was completed using a 4/0 resorbable suture and the hip wound was closed in layers with placement of an epidural catheter (Epidural minipack, 16G, Portex Ltd., Hythe, UK) in the wound to allow administration of a local anesthetic agent (0.25% marcaine) for postoperative analgesia. A standard regime of perioperative antibiotics was used (200 mg metronidazole three times per day and 500 mg phenoxymethyl penicillin four times per day) and the patients were discharged when they were mobile and comfortable. Followup at the outpatient clinic was on a weekly basis for the first 2 weeks to monitor wound healing and screen for complications. RESULTS Eighteen patients entered the study (10 females, 8 males). One of the male patients failed to return for any follow-up appointments and was not included in the results. The mean age of the patients was 14.8 years (range 11 to 27 years). FIGURE 3 screw. The bone graft is secured at the alar base with a positional Three patients had postoperative complications. Two had minor wound dehiscence and exposure of bone graft, which was resolved using mouthwash and irrigation. In the other patient, the bone graft was noticed to have shifted position on the first postoperative day. The patient was taken back to the operating room after 72 hours to replace the bone graft to its original position where it was fixed with a bicortical 1.7-mm screw. Errors of Method Landmark Placement Error The main source of error was in the accuracy and reproducibility of landmark placement. This was assessed for the author responsible for placing the landmarks (MFD). Each of the 21 landmarks used in the analyses were placed on a model on 10 occasions at least 72 hours apart. The actual distance in millimeters from the initially placed landmark was calculated for each subsequent placement. Thus, an assessment of the reproducibility of landmark placement was made for each individual landmark. The mean distance of each placed landmark from its original was calculated (Bock and Bowman, 2006). Figure 4 is a boxplot demonstrating the reproducibility of placement of each of the 21 landmarks used in the analysis. The landmarks were placed more than 2 mm away from the mean 2.43% of the time, and more than 1 mm from the mean 16.51% of the time. The overall median distance was mm from the mean. For the majority of the landmarks, this was acceptably within a reasonable distance for clinical assessment. Landmark Placement It was crucial to be able to place the landmarks on the 3D models repeatedly in the same position. The boxplot shown in Figure 4 reveals that certain landmarks were more difficult to repeatedly place in the same position. In particular, landmark 3, which represents the anthropometric point glabella, was difficult to place repeatedly.

4 394 Cleft Palate Craniofacial Journal, July 2007, Vol. 44 No. 4 FIGURE 4 Boxplot demonstrating the reproducibility of placement of each of the 21 landmarks used in the analysis. Symmetry Results Preoperative and postoperative symmetry results were calculated for each subject and plotted against each other in a scatterplot diagram (Fig. 5). The results were compared using a two sample Student s t test. The improvement in facial symmetry score postoperatively was statistically significant (p 0.005). Bone Volume Correlation to Symmetry Improvement There was no correlation between the volume of bone placed and the improvement in symmetry score (Spearman s rank test). DISCUSSION 3D cleft deformity ideally requires an accurate and quantifiable 3D method of assessment. It should be practical for use in the pediatric population and allow data to be archived and retrieved simply. The system of stereophotogrammetry used seems to be an ideal tool to assess facial deformity in an objective manner and measure 3D deficiencies (and reconstructions) in terms of the residual asymmetry present following surgical repair of cleft lip and palate. The causes of facial asymmetry following repair of UCLP are not known. It may develop following the inaccurate primary reconstruction of the nasolabial tissue or as a result of impaired normal growth of the hard and soft tissues of the face. This impairment of facial growth may occur secondarily to scarring or may represent an inherent growth deficiency. Whatever the cause of the asymmetry, it remains a common problem in this group of patients. Effective means of managing this firstly require an accurate assessment the deformity itself. 2D assessment of the underlying bony deformity is inaccurate and does not allow planning of the 3D reconstruction that is required. The method of streophotogrammetry described was able to quantify the surgical changes which took place in terms of FIGURE 5 Scatterplot showing the improvement in symmetry postoperatively (data points below the line represents improvement). facial symmetry in our group of patients with significant overall improvement (p 0.005). It may be that this will continue to improve over time as we have tended to over-reconstruct the volume deficiency in anticipation of some resorption of the bone graft. The results reveal one patient who was more asymmetric following treatment; this was due to over correction of the deformity. It was disappointing that the system was unable to calculate the soft tissue volume change which took place following treatment. The resolution of the cameras used in this study was about 1.5 megapixels which may not have been high enough to allow small volume changes to be accurately assessed. A new system using high resolution cameras is now available and may overcome this problem. The postoperative complication rate was within an acceptable range for this type of surgical intervention. The logic of this particular surgical intervention was twofold. Firstly, as we have shown, an improvement in symmetry of the nose and the face is achievable with alar base bone grafting. Secondly, if subsequent soft tissue nasal correction is to be undertaken, then this will have a sound and more anatomically accurate bony base on which to be carried out. The longer-term stability of this reconstruction cannot be commented on at this point. CONCLUSION 3D stereophotogrammetry provides a practical method of objectively assessing and quantifying facial deformity and surgical outcome, as measured by symmetry scores, for secondary nasal deformity in treated UCLP patients undergoing corticocancellous bone grafting to reconstruct the deficient alar base. Using the method of reconstruction described in this paper we were able to improve overall facial symmetry in this group of patients.

5 Devlin et al., STEREOPHOTOGRAMMETRIC ASSESSMENT OF CLEFT NASAL DEFORMITY 395 REFERENCES Ayoub AF, Garrahy A, Hood CA, White JE, Bock M, Siebert JP, Spencer R, Ray A. Validation of a vision based three-dimensional facial imaging system. Cleft Palate Craniofac J. 2003;40: Ayoub AF, Siebert JP, Moos KF, Wray D, Urquart C, Niblett TB. A visionbased three-dimensional capture system for maxillofacial assessment and surgical planning. Br J Oral Maxillofac Surg. 1998;36: Bock MT, Bowman AW. On the measurement and analysis of object asymmetry with application to facial modeling. Appl Stat. 2006;55: Coghlan BA, Laitung JK, Piggot RW. A computer-aided method of measuring nasal symmetry in the cleft lip nose. Br J Plast Surg. 1993;46: Devlin MF, Ray A, Jones R, Ayoub AF. A simple method of measuring bone graft volume, technical note. J Craniomaxillofac Surg. 2003;31: Farkas LG. Accuracy of anthropometric measurements: past, present and future. Cleft Palate Craniofac J. 1996;33: Fisher DM, Lo LJ, Chen YR, Noordhoff MS. Three-dimensional computed tomographic analysis of the primary nasal deformity in 3-month-old infants with complete unilateral cleft lip and palate. Plast Reconstr Surg. 1999;103: Hajeer MY, Ayoub AF, Millett DT. Three-dimensional assessment of facial softtissue asymmetry before and after orthognathic surgery. Br J Oral and Maxillofac Surg. 2004;42: Hajeer MY, Mao Z, Millett DT, Ayoub AF, Siebert JP. A new three-dimensional method of assessing facial volumetric changes after orthognathic treatment. Cleft Palate Craniofac J. 2005;42: Kawai T, Natsume N, Shibata T, Yamamoto T. Three-dimensional analysis of facial morphology using moiré stripes. Int J Oral and Maxillofac Surg. 1990;19: Leivesley WD. The reliability of contour photography for facial measurements. Br J Orthod. 1983;10: McCance AM, Moss JP, Fright WR, Linney AD, James DR. Three dimensional analysis techniques part 4: three-dimensional analysis of bone and soft tissue to bone ratio movements in 24 cleft palate patients following Le Fort osteotomy: a preliminary report. Cleft Palate Craniofac J. 1997;34: McComb HK, Coghlan BA. Primary repair of the unilateral cleft lip nose: completion of a longitudinal study. Cleft Palate Craniofac J. 1996;33: Morrant DG, Shaw WC. Use of standardized video recordings to assess cleft surgery outcome. Cleft Palate Craniofac J. 1996;33: Sandy J, Williams A, Mildinhall S, Murphy T, Bearn D, Shaw B, Sell D, Devlin B, Murray J. The Clinical Standards Advisory Group (CSAG) cleft lip and palate study. Br J Orthod. 1998;25: Semb G, Brattstrom V, Molsted K, Prahl-Andersen B, Rumsey N, Shaw WC. The Eurocleft study: longitudinal follow-up of patients with complete cleft lip and palate: part 4, relationship between outcome, patient/parent satisfaction and the burden of care. Cleft Palate Craniofac J. 2004;42: Shaw WC, Brattstrom V, Molsted K, Prahl-Andersen B, Roberts C, Semb G. The Eurocleft study: intercenter study of treatment outcome in patients with complete cleft lip and palate. Part 5: discussion and conclusions. Cleft Palate Craniofac J. 2005;42:93 98.