ORIGINAL ARTICLES. Three-Dimensional Analysis of the Child Cleft Face

Transcription

1 ORIGINAL ARTICLES Three-Dimensional Analysis of the Child Cleft Face S. DUFFY, B.CH.D., F.D.S.R.C.S., M.SC., M.ORTH. J. H. NOAR, B.D.S., F.D.S.R.C.S., M.SC., M.ORTH. R. D. EVANS, B.D.S., F.D.S.R.C.S., M.SC., M.ORTH. R. SANDERS, B.SC., M.B., B.S., F.R.C.S. Objective: This study examined the facial surfaces of cleft children and unaffected children aged 8 11 years with the aim of identifying and assessing differences in their facial surface morphology. The investigation was carried out using an Optical Surface Scanner, an instrument that utilizes laser light to construct and archive a three-dimensional image of the face suitable for linear measurement and direct surface comparisons. Design, Setting, and Patients: Thirty-nine cleft lip and palate (CLP) patients and 25 unaffected subjects were voluntarily recruited from two southeast England hospitals. A range of linear facial measurements was compared. Threedimensional differences between the cleft subgroups and the control group were visualized by superimposition of averaged cleft scans over the averaged control group images. Results: Statistically significant dimensional differences (p.05) in interocular width, nose base widths, mouth widths, and nose base/mouth width ratios were found between the cleft group and the control group. Qualitative differences over the whole of the face were readily demonstrated between the groups by superimposition. Face width and submandibular area depth differed consistently between the groups, the cleft face appearing narrower with a deeper submandibular area. Conclusion: Significant differences exist between the facial surface morphology of CLP patients and control subjects. KEY WORDS: child, cleft lip and palate, optical surface scanner, three-dimensional Few studies have examined the characteristics and dimensions of facial soft tissues in cleft lip and palate (CLP) children, particularly away from the midline (Prahl-Andersen et al., 1995). Methods that have been reported include photography, direct facial measurement and soft tissue radiography (Sadowsky et al., 1973; Susami et al., 1993; Zhu et al., 1994), but all have suffered problems with accuracy, overcomplexity, or possible radiation risk to the patient, and have thus failed to enter into routine clinical use. In comparison with unaf- S. Duffy is the Registrar in Orthodontics, Mount Vernon and Watford Hospital Trust, Northwood, Middlesex, and the Eastman Dental Hospital, London, England. J. H. Noar is Consultant/Senior Lecturer in Orthodontics, Mount Vernon and the Eastman Dental Hospital, London, England. R. D. Evans is Consultant/Senior Lecturer in Orthodontics, Great Ormond Street and the Eastman Dental Hospital, London, England. R. Sanders is Professor of Plastic Surgery, Mount Vernon Hospital, Northwood, England. Previously presented orally at the Craniofacial Society of Great Britain Annual Scientific Meeting, March 1998, Keeble College, Oxford, England. Submitted August 1998; Accepted May Reprint requests: Mr. J. H. Noar, Orthodontic department, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, England. fected controls, increased CLP lower face height, reduced mouth width, longer nose, wider nose base widths and variable upper lip changes have all been noted (Larson and Nilsson, 1983; Ross, 1987). The aim of this study was to compare objectively facial surface dimensions and characteristics of a group of children having cleft lip and/or palate with a group of unaffected children using the Optical Surface Scanner (OSS), a tool recently developed to provide quantitative three-dimensional information on the facial surface (Cutting et al., 1988; Moss et al., 1989). This instrument utilizes low-level laser light to construct and archive a three-dimensional image of the face suitable for computer analysis. The system has proved reliable and accurate in the identification and measurement of facial landmarks and dimensions (Moss et al., 1989; Bush and Antonyshyn, 1996). The scanning procedure takes approximately 10 seconds and is safe, noninvasive, and well tolerated by children (Fig. 1). METHODS Volunteer caucasian, nonsyndromic cleft lip and/or hard palate patients aged 8 11 years were recruited from Mount Ver- 137

2 138 Cleft Palate Craniofacial Journal, March 2000, Vol. 37 No. 2 FIGURE 1 Diagrammatic plan view of the Optical Surface Scanner. FIGURE 2 Anthropometric landmarks. non Hospital and the Eastman Dental Hospital, London. Any patient who had undergone alveolar bone grafting procedures was excluded. Unaffected volunteer caucasian control children aged 8 11 were recruited from orthodontic clinics at the same centers, excluding any child in whom skeletal disproportion was a feature of the malocclusion. Each subject was scanned with the Frankfort plane raised anteriorly 10 degrees to the horizontal with the mandible in the rest position and the lips lightly opposed if this was possible without undue muscular effort. Eight patients were rescanned 30 minutes after the original scan to assess image reproducibility. Analysis Linear Measurements and Ratios Anthropometric landmarks related to the eyes, nose, lips, and chin were identified on each scan (Fig. 2) by a single operator (SD). Linear facial measurements related to these landmarks were calculated by the scanner software (Tables 1 and 2). In addition, two ratios were recorded, including the percentage contribution of the lower face height (LFH) to the total face height (TFH) and the ratio of alar base width to mouth width. In subjects with unilateral defects, the images were first manipulated by the OSS software, if necessary, to position all clefts on the left side. For the purpose of analysis, the cleft subjects were subdivided into the following groups: bilateral complete cleft lip and palate (BCLP), unilateral complete cleft lip and palate (UCLP), unilateral cleft lip and alveolus (UCLA), and cleft palate (CP). Method Error Eight subjects were selected at random and rescanned 30 minutes after their initial scan. The landmarks and measurements obtained were compared with the original to assess the reliability of measurements from repeat scans. At a later date, a new set of landmarks and measurements were obtained for each of the 16 scans to assess the reliability of repeat landmark identification. In each case, coefficients of measurement variation were obtained, expressed as percentage differences between measurements. Inter- and intraimage landmark reproducibility lie within the range of 0.47 and 5.4%, with no discernible differences between inter- and intraimage landmarking. Statistical Analysis A Shapiro-Wilk W test established all data as normally distributed. A one-way analysis of variance (ANOVA) for significant differences between the means of the five groups (one control and four cleft subgroups) for each variable at the p.05 level was conducted. Where significant differences be-

3 Duffy et al., THREE-DIMENSIONAL ANALYSIS OF THE CHILD CLEFT FACE 139 TABLE 1 Mean Values (mm) for Linear Variables, with Percentage LFH and Nose/Mouth Width Ratio (Standard Deviations in Parentheses) Clefts (n 39) Variable Control n 25 (10 males, 15 females) BCLP n 7 (4 males, 3 females) UCLP n 10 (4 males, 6 females) UCLA n 9 (4 males, 5 females) CP n 13 (6 males, 7 females) Biocular width: ex-ex Left ocular width: ex l-en l Right ocular width: ex r-en r Intercanthal width: en l-en r Left endocanthon to nasion: en l-n Right endocanthon to nasion: en r-n Alar base width: al-al Pronasale to left alar base: al l-prn Pronasale to right alar base: al r-prn Alar base root width: sbal-sbal Subnasale to left alar base root: sbal l-sn Subnasale to right alar base root: sbal r-sn Nose dorsum length: n-prn Mouth width: ch-ch Upper lip length: sn-sto Lower lip length: sto-b Upper vermillion thickness: ls-sto Lower vermillion thickness: li-sto Lower face height: sn-gn Left upper lateral lip length: sbal l-cph l Right upper lateral lip length: sbal r-cph r Upper face height: n-sn Total face height: n-gn LFH as a percentage of total face height Nose base/mouth width ratio: al-al/ch-ch (5.67) (2.8) (2.56) (3.24) (2.72) (2.52) (2.65) (3.2) (3.63) (2.62) (1.77) (1.93) (5.04) (3.22) (2.08) (2.41) 7.44 (1.87) 7.94 (1.96) (3.91) (2.38) 15.3 (2.35) (5.06) 100 (7.11) (2.81) 0.72 (.065) (4.69) (2.38) (3.7) (2.16) (2.22) (2.37) (4.04) 28.7 (4.08) (2.92) (3.54) (2.61) (2.2) (2.59) (2.81) (2.21) (2.68) 7.37 (1.7) 7.39 (2.84) (4.75) (2.12) (1.99) (2.52) 104 (5.87) (1.66) 0.89 (.078) (4.53) (3.59) 32.2 (2.89) (2.33) (2.08) (2.02) (2.25) (4.21) (3.21) (3.17) (2.14) (1.95) (3.6) (3.93) (1.51) (2.1) 5.46 (1.3) 7.21 (1.89) (5.39) (2.74) (2.29) (3.91) 102 (7.75) (2.51) 0.78 (.064) (7.03) (3.51) (4.32) (2.54) (1.63) (2.27) (2.55) (2.49) 27.1 (2.97) (2.74) 13.5 (1.96) (1.86) (3.59) (3.99) 18.6 (1.84) (1.86) 7.69 (4.82) 8.93 (2.39) (4.15) (1.69) (1.95) (4.23) (6.56) (1.95) 0.78 (.07) (6.34) (3.42) (3.4) (2.76) 20.8 (2.34) (3.12) 32.1 (3.25) (4.07) (3.49) 20.3 (4.07) (2.81) 11.4 (2.21) (5.15) (3.23) (2.63) (2.28) 7.67 (1.97) 7.87 (1.44) (4.95) (2.34) (2.21) (4.9) (7.58) (3.08) 0.78 (.056) TABLE 2 ANOVA, Student s t Tests and Sheffé Comparisons for Differences Between the Linear Variables Variable Initial one-way ANOVA for Differences Between the 5 Groups (d.f. 4, 59) F Ratio p ANOVA for Differences Between the 4 cleft Subgroups (d.f. 3, 34) Where Differences Exist in Initial ANOVA F Ratio p Two-Sample Student s t Test: Total Cleft Group Versus Control T Value p Scheffé Test Group Comparisons Significant at the p.05 level Biocular width: ex-ex Left ocular width: ex l-en l Right ocular width: ex r-en r Intercanthal width: en l-en r Left endocanthon to nasion: en l-n Right endocanthon to nasion: en r-n Alar base width: al-al Pronasale to left alar base: al l-prn Pronasale to right alar base: al r-prn Alar base root width: sbal-sbal Subnasale to left alar base root: sbal l-sn Subnasale to right alar base root: sbal r-sn Nose dorsum length: n-prn Mouth width: ch-ch Upper lip length: sn-sto Lower lip length: sto-b Upper vermillion thickness: ls-sto Lower vermillion thickness: li-sto Lower face height: sn-gn Left upper lateral lip length: sbal l-cph l Right upper lateral lip length: sbal r-cph r Upper face height: n-sn Total face height: n-gn LFH as a percentage of total face height Nose base/mouth width ratio: al-al/ch-ch Control UCLP and BCLP Control UCLP BCLP control UCLA control; BCLP control, CP, and UCLP BCLP control, CP, and UCLP BCLP control, CP, UCLA, and UCLP BCLP control, UCLA, and UCLP BCLP control, CP, UCLA, and UCLP BCLP control, CP, and UCLA

4 140 Cleft Palate Craniofacial Journal, March 2000, Vol. 37 No. 2 landmarked points across the averaged images using a leastmeans-squared, best-fit method. Differences are represented by changes of color in a format similar to contours on a map. The height that is measured is the radial distance from the center of rotation of the scanner chair to the facial surface. In this way, the averaged cleft subgroup images were superimposed, in turn, upon the averaged control image to show differences between the control group and the cleft subgroups. RESULTS FIGURE 3 Averaged control image. Linear Dimensions and Ratios Mean values of the linear variables and ratios in the control group and cleft subgroups, with standard deviations, are shown in Table 1. Table 2 shows the results of ANOVA tests, the Student s t test, and the Scheffé test. Significant differences between the controls and the cleft subgroups were found for the following variables. Intercanthal width (en-en) Intercanthal width was significantly narrower in the cleft group overall (p.05), and UCLP and BCLP subgroups compared separately with the controls. Left endocanthon to nasion (en l-r) This dimension was significantly narrower in the cleft group overall (p.05), and in the UCLP subgroup compared with the controls. Alar base width (al-al) tween the groups were demonstrated, the following were conducted: a one-way ANOVA for significant differences between the means of the four cleft subgroups at the p.05 level, a prespecified contrast between the control group and the cleft group as a whole using the two-sample Student s t test (p.05) for each variable that showed no significant ANOVA differences between the four cleft subgroups, and a multiple range Scheffé test for significant differences between paired groups (one control versus one cleft subgroup) at the p.05 level. Image Superimposition Differences between the averaged subgroup images and the control group were assessed qualitatively by image superimposition. The OSS software constructs an average image of each group from the scans available by registration of the images in a common coordinate system of anatomical landmarks (Fig. 3). The averaged images obtained can then be superimposed upon one another to give a color graduated image showing average differences in facial contour between groups (Figs. 4 through 7). The computer superimposition is achieved by matching a range of mathematically constructed and visually The alar base was significantly broader in the BCLP subgroup compared with the controls. Alar base root width (sbal l-sbal r) The alar base root was significantly broader in the BCLP and UCLA subgroups compared with the controls. The BCLP subgroup width was also significantly broader than the UCLP and CP subgroups. Alar base roots to midline (sbal l-sn, sbal r-sn) The BCLP subgroup dimensions were significantly larger than the control group and remaining cleft subgroups with the exception of the UCLA group on the cleft side. Lateral lip lengths (sbal l-cph l, sbal r-cph r) There were significant intergroup differences, with the BCLP subgroup having significantly longer lateral lip lengths on each side than all other groups with the exception of the CP group on the left side.

5 Duffy et al., THREE-DIMENSIONAL ANALYSIS OF THE CHILD CLEFT FACE 141 FIGURE 4. Control versus CP. FIGURE 5. Control versus UCLA. FIGURE 6. Control versus UCLP. FIGURE 7. Control versus BCLP. FIGURES 4 7. Cleft groups superimposed on control group average. Positive differences control face area more prominent. Negative differences cleft face area more prominent. Color contour scale:

6 142 Cleft Palate Craniofacial Journal, March 2000, Vol. 37 No. 2 Upper vermillion thickness (ls-sto) There were significant intergroup differences, although no single paired group showed domination in the Scheffé test. Mouth width (ch-ch) Mouth width was significantly narrower in the cleft group overall (p.05), although no single paired group showed domination in the Scheffé test. Nose width/mouth width ratio The BCLP ratio was significantly greater than the control, UCLA, and CP ratios. Image Superimpositions Figures 4 7 show captured photographic images of the averaged cleft subgroups superimposed upon the averaged control image. Each color change represents 2 mm of difference. (See color photos on the online version of The Cleft Palate Craniofacial Journal at The warm yellows, browns, and reds represent a positive difference the control image is more prominent in these areas. The cold greens and blues represent a negative difference the cleft superimposition is more prominent in these areas. Maintenance of the original computer constructed flesh tone indicates areas of little or no depth difference, 1 to 1 mm apart. Control Versus CP (Fig. 4) The cheeks, upper lip, and part of the chin appear generally more prominent in the control image, with over 5 mm difference at the left midcheek point. Large areas of the chin, nose, periorbital region, and forehead are closely approximated. The left nose dorsum, left nostril, and submandibular region are more prominent in the CP image. Control Versus UCLA (Fig. 5) Most of the cheek area and the left side of the nose appears more prominent in the control image. Large areas of each image approximate well, particularly over the forehead, periorbital regions, lips, and chin. The submandibular region is more prominent in the UCLA image. Control Versus UCLP (Fig. 6) The nose tip, cleft side nostril, complete upper lip, forehead, and cleft side outer cheek are more prominent in the control image, the nose tip by over 5 mm. The periorbital regions, nose bridge areas, and noncleft side cheek correlate well. The chin, orbits, submandibular region, and noncleft side nostril are more prominent in the UCLP image. Control Versus BCLP (Fig. 7) The left and right cheeks, left chin, and parts of the forehead are more prominent in the control image. Small areas of the chin, lips, nose tip, orbits, and forehead approximate. The nose bridge, dorsum, nostrils, and area under the chin are more prominent in the BCLP image. DISCUSSION Linear Dimensions and Ratios Although there have been few studies providing anthropometric data for this age range, the upper lip length, upper vermillion thickness, lower vermillion thickness, mouth width, lower lip length, and nose base widths in this study were in general agreement with previous reports (Larson and Nilson, 1983; Nanda et al.,1990; Susami et al., 1993). Control nose dorsum length, at mm, was over 20% shorter than in previously reported studies in spite of broad similarities in group composition (Nanda et al., 1990; Prahl- Anderson et al., 1995). In this study, nasion was calculated as the true point of maximum frontonasal concavity, as opposed to a horizontal cephalometric projection of skeletal nasion favored by some authors. Intercanthal Width It is tempting to postulate that the growth retardant effect of CLP surgery can be manifested transversely as high as orbital level in view of the findings that UCLP and BCLP subjects in this study have narrower intercanthal widths than the controls. The UCLP group also showed a reduced endocanthon to nasion dimension on the cleft side compared with the controls, adding weight to this suggestion. No other orbital dimensions, however, differed significantly between the groups and therefore these findings in isolation must be viewed with caution. Alar Base, Alar Base Root, and Mouth Widths The findings that BCLP subjects have broad alar base widths are in keeping with several previous studies (Farkas and Lindsay, 1971; Larson and Nilson, 1983; Pigott, 1985; Vegter et al., 1997). Farkas and Lindsay (1971) examined a group of unilateral and bilateral cleft neonates prior to primary repair, finding nose width disparity much greater before cleft repair, and it is presumed the surgery goes some way to correcting this discrepancy but generally fails to restore normal dimensions. Tissue deficiency, marginal excisions, scar contraction, and growth disruption are all likely to have contributed to the reduced mouth width dimensions in cleft subjects in this study. Susami et al. (1993) found no significant differences in mouth width of cleft subjects and controls aged 9 12 years, whereas

7 Duffy et al., THREE-DIMENSIONAL ANALYSIS OF THE CHILD CLEFT FACE 143 Larson and Nilsson (1983) and Zhu et al. (1994) noted significantly narrower mouths in cleft subjects of a similar age. Nose Width/Mouth Width Ratio The nose/mouth width ratio has recently been suggested as particularly important in the assessment of the CLP face (Vegter et al., 1997), with a ratio significantly above normal considered potentially unattractive. The ratio is significantly greater in the BCLP subjects than the control, CP, and UCLA groups in this study. Thus, the undesirable visual impact of a broad nose base is likely to be exacerbated in the BCLP subject by the anticipated narrow mouth width. Lateral Lip Lengths The BCLP subgroup had significantly longer lateral lip lengths than any other group. No such findings were observed in the midline lip measurement (sn-st), so the major differences in the vertical plane in this sample were limited to localized disruption of the regions lateral to the philtrum. No significant differences between the sides were observed in spite of the obvious potential for one-sided surgical disruption in unilateral cleft subjects. The effect of surgery on lip length is intimately related to operation type, with some procedures resulting in a shorter or longer lip on the affected side (Assuncao, 1992; Roberts-Harry and Sandy, 1992). Original surgery notes were not examined for these subjects, and it must be assumed that the net effect of averaging this dimension was to cancel such potential differences out or that the differences were too small to be of significance. Upper Vermillion Thickness In contrast with total lip length, the BCLP group had the least upper vermillion thickness and, while the ANOVA analysis showed intergroup differences, the paired Scheffé comparisons did not highlight any particular group. Susami et al. (1993) reported a reduced upper vermillion thickness in a similarly aged sample of BCLP and UCLP subjects, although the differences were only significant in the UCLP group. The effects of developmental disruption and subsequent surgical repair of the lip have been described above, and it is assumed that the range of techniques for lip closure have a variable effect on the vermillion, tending to result in a reduced thickness. General Comments Three individual surgeons were involved in the treatment of the cleft patients in this sample, undertaking a range of broadly similar procedures, but the small sample size prevents meaningful comparisons of surgical techniques. In spite of obvious nasal asymmetry in many of the unilateral cleft subjects, no significant dimensional differences between the sides were noted other than around nasion. As the dimensions calculated only took into account nose tip and base variation, perhaps the most marked differences are manifest between these areas around the nostrils and lateral nasal perimeters (Coghlan et al., 1993). There were no significant differences between the face heights of control subjects and any cleft subgroups as measured from midline profile landmarks. This differs from previous child and adult cephalometric studies of CLP subjects, which have shown increased face heights compared with controls (Smahel and Mullerova, 1986; Ross, 1987). Superimpositions Certain features were common to each superimposition. The cheek and zygomatic areas were invariably more prominent in the control group, especially on the left side. (Not all unilateral clefts were left sided, but to permit comparison, their images were manipulated to appear on the left by the OSS software to standardize the cleft side.) A general narrowness of the face in CLP patients has been noted in other studies (Han et al., 1995), but these findings are not unanimous. It does seem reasonable to suggest, however, that the disturbance in lateral growth of the face in the CLP subject is often detectable on the surface, especially on the affected side in unilateral clefts. In all four group superimpositions, the CLP submandibular region was markedly more prominent than the control, with contour bands showing a 5 7 mm difference over a wide area. The significance of this difference is unknown. Markus et al. (1992) describe the posterior part of the tongue in clefts involving the soft palate lying at a lower level as a result of the defect before repair. Ross (1987) cites mandibular posture being significantly affected by the normally sized tongue s quest for space in the presence of a narrow, short maxilla. This may be consistently manifest on the facial surface by a general fullness under the chin. Other differences were specific to the cleft subgroup studied and are discussed in the following sections. UCLA Nose asymmetry was well demonstrated by the superimposition. Most of the face approximated well to the control, suggesting the relatively minor effect on facial profile when the cleft defect is less severe. UCLP Again, nose asymmetry was an obvious feature of the superimposition. The whole of the upper lip was retrusive relative to the control, showing that maxillary hypoplasia, even at this age, is affecting the soft tissue profile. BCLP The whole of the nose area differs markedly from the control, being wider and more prominent, until the tip is reached.

8 144 Cleft Palate Craniofacial Journal, March 2000, Vol. 37 No. 2 Differences in the prominence of the upper lip were surprisingly small in view of the high probability of underlying maxillary hypoplasia. A bulging of the incomplete orbicularis oris under the surface of the upper lip is often visible in cleft lip subjects, even with the lips at rest, and this may account for some of the apparent lip fullness. CP Most of the face was retrusive relative to the control. The nose area was asymmetrical and appreciably wider on the left side, but any reason for this must be attributed to disruption in the area of the hard and soft palate and not more anteriorly. Similar findings have been reported in CP adults (Farkas and Lindsay, 1972). They suggest a combination of the embryonal disruption and surgical repair accounted for the narrow face and commented that the unexpected nasal asymmetry may indicate the primary defect was in fact a cleft lip and palate microform. CONCLUSIONS The conclusions are as follows: (1) Significant differences in interocular widths, nose base widths, lateral lip lengths, vermillion thickness, mouth width, and nose/mouth width ratios were found between the control group and the cleft subgroups. (2) The BCLP subgroup differed most from the control group, reflecting the extent of disruption in this condition. (3) Local differences in facial contour were readily apparent on superimpositions of the control group average with the cleft subgroup averages. (4) Face width and submandibular depth differed consistently between the groups, the cleft face appearing narrower, with a more prominent submandibular area. (5) The OSS proved able to effectively disclose and quantify surface characteristics of the child cleft face. FURTHER STUDY This study is to provide the basis for a longitudinal study into the effects of growth and surgery such as alveolar bone grafting and orthognathic procedures on the CLP facial contour. Acknowledgments. This study was made possible by a grant from the Restoration of Appearance and Function Trust, Mount Vernon Hospital. We are grateful for the assistance of Mr. C. Foy, medical statistician, Mount Vernon Hospital. REFERENCES Assuncao AGA. The VLS classification for secondary deformities in the unilateral cleft lip. Br J Plast Surg. 1992;45: Bush K, Antonyshyn O. Three-dimensional facial anthropometry using a laser scanner: validation of the technique. Plast Reconstr Surg. 1996;98: Coghlan BA, Laitung JK, Pigot RW. A computer aided method of measuring nasal symmetry in the cleft nose. Br J Plast Surg. 1993;46: Cutting CB, McCarthy JG, Karron DB. Three-dimensional input of body surface data using a laser light scanner. Ann Plast Surg. 1988;21(1): Farkas LG, Lindsay WK. Morphology of adult face after repair of bilateral cleft lip and palate in childhood. Plast Reconstr Surg. 1971;47: Farkas LG, Lindsay WK. Morphology of adult face after repair of isolated cleft palate in childhood. Cleft Palate J. 1972; Han BJ, Suzuki A, Tashiro H. Longitudinal study of craniofacial growth in subjects with cleft lip and palate: from cheiloplasty to 8 years of age. Cleft Palate Craniofac J. 1995;32(2): Larson O, Nilsson B. Early bone grafting in complete cleft lip and palate cases following maxillofacial orthopaedics. Scand J Plast Reconstr Surg. 1983; 17: Markus AF, Delaire J, Smith WP. Facial balance in cleft lip and palate. 1: Normal development and cleft palate. Br J Oral Maxillofac Surg. 1992;30: Moss JP, Linney AD, Grindrod SR, Mosse CA. A laser scanning system for the measurement of facial surface morphology. Optics Lasers Eng. 1989; 10: Nanda RS, Meng H, Kapila S, Goorhuis J. Growth changes in the soft tissue profile. Angle Orthod. 1990;60(3): Pigott RW. Alar leapfrog a technique for repositioning the total alar cartilage at primary cleft lip repair. Clin Plast Surg. 1985;12: Prahl-Andersen B, Ligthelm-Bakker ASWMR, Wattel E, Nanda R. Adolescent growth changes in soft tissue profile. Am J Orthod Dentofacial Orthop. 1995;107: Roberts-Harry D, Sandy JR. Repair of cleft lip and palate: 1. Surgical techniques. Dent Update. 1992; Ross RB. Treatment variables affecting facial growth in complete unilateral cleft lip and palate. Cleft Palate J. 1987;24:5 64. Sadowsky C, Aduss H, Pruzansky S. The soft tissue profile in unilateral clefts. Angle Orthod. 1973;43: Smahel Z, Mullerova Z. Craniofacial morphology in unilateral cleft lip and palate prior to palatoplasty. Cleft Palate J. 1986;23(3): Susami T, Kamiyama H, Uji M, Motohashi N, Kuroda T. Quantitative evaluation of the shape and elasticity of the repaired cleft lip. Cleft Palate Craniofac J. 1993;30(3): Vegter F, Mulder JW, Hage JJ. Major residual deformities in cleft patients: a new anthropometric approach. Cleft Palate Craniofac J. 1997;34: Zhu NW, Senewiratne S, Pigott RW. Lip posture and mouth width in children with unilateral cleft lip and palate. Br J Plast Surg. 1994;47: