Journal of Plastic, Reconstructive & Aesthetic Surgery (2009) 62, 1437e1441

Development of a method for rating nasalappearance after cleft lip repair

Xing He a, Bing Shi a,c,*, Mehul Kamdar b, Qian Zheng a,c, Sheng Li a,Yan Wang a

a Department of Oral and Maxillofacial Surgery, West China College of Stomatology, Sichuan University, Chengdu, Chinab Division of Plastic Surgery, Columbia-Presbyterian Medical Center, 161 Fort Washington Avenue,New York, NY 10032, USAc State Key Laboratory pf Oral Diseases, West China College of Stomatology, Sichuan University, Chengdu, China

Received 21 February 2008; accepted 2 May 2008

KEYWORDSCleft lip;Reliability;Rating method;Nasal appearance

* Corresponding author. DepartmenSurgery, West China Dental School,Section 3. Ren Min Nan Road, Chengdu61153005.

E-mail address: shibingcn@sina.co

1748-6815/$ - see front matter ª 2008 Briti

doi:10.1016/j.bjps.2008.05.018

Summary The aim of this study was to develop a new method for evaluation of nasal appear-ance in patients after cleft lip repair. A method is described in which the nasal region is eval-uated without the influence of the labial repair. Frontal, submental and profile viewphotographs of 45 patients after cleft lip repair were obtained. For each patient, two setsof images, one of the entire nasolabial region and one of the nasal complex in isolation, wereassessed with a five-point scale by a panel of seven judges. Repeat evaluation at 1 weekallowed calculation and comparison of intra-judge and inter-judge reproducibility. The inter-class correlation coefficient (ICC) values were higher in the assessment of the isolated nasalcomplex when compared to the nasolabial region with the exception of the cleft side lateralview. The ICC value of the assessment based on all three views was the highest. The levelof intra-judges and inter-judges was good, thus the reliability and sensibility of this newmethod is acceptable. This method is credible because acceptable pooled levels of reliabilitywere obtained. In the future, this rating system may be used to assess nasal appearance afterdifferent treatments for cleft lip patients.ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published byElsevier Ltd. All rights reserved.

t of Oral and MaxillofacialSichuan University, No. 14,610041, China. Tel.: þ86 28

m (B. Shi).

sh Association of Plastic, Reconstruct

As the technique of cleft lip repair has advanced withcorresponding improved results, there has been anincreased focus on the associated nasal deformity. Thecleft nasal deformity is now the most visible reminder ofthe cleft deformity.1 In patients who have had a cleft liprepair, the nose, rather than the lip, most reflects theoriginal deformity,2 especially those patients who did nothave primary nasal correction at the time of the primary lip

ive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.

Table 1 Interpretation of Kappa value

Kappa value Strength of agreement

<0.2 Poor0.21e0.40 Fair0.41e0.60 Moderate0.61e0.80 Substantial0.81e1.00 Perfect

Table 3 Intra-judge Kappa values in validity test ofaesthetic index

Numberof judges

WeightingKappa

95% confidenceinterval

1438 X. He et al.

repair. The adolescent or young adult with a cleft lip oftendisplays more concern about his nose than about theresidual lip deformity.

A major goal of the repair for cleft patients is to improvethe aesthetics and function which enhance socialacceptability. Balance, symmetry and proportion are thecomponents of the aesthetic appearance.3 The nose islocated in the middle of the face and is the most protu-berent part so more and more surgeons focus on nasalreconstruction and types of rhinoplasties performed.

In order to compare the effect of different surgicalprocedures, many methods have been developed to assessappearance. These methods can be broadly divided intoqualitative and quantitative. The latter analyse the extentof abnormal morphology and the degree of disproportionthrough facial measurement.4,5 They abstract the appear-ance based on numerical data without evaluating theoverall appearance in a vivid image. So the qualitativemethod better reflects both the patient’s and the public’sperception.6

One of the qualitative methods is to assess the appear-ance with a photograph. Most methods were used todescribe both nose and lip morphology together while a fewfocused on the nose only. To accurately assess the nasaldeformity, we should reduce the influence of the lip. So it ishelpful to develop a method for rating the isolated nasalappearance in a patient after cleft lip repair, especiallywhen researching a correlation between subjective nasalassessment and the objective nasal measurement, orcomparing the outcomes of different rhinoplasties.

The aim of this study was to develop a simple and flex-ible method to enable evaluation of the nasal aesthetic andto test its reliability and sensibility.

Material and methods

Sample

Forty-five patients affected by a non-syndromic unilateralcleft lip and palate (UCLP) were randomly selected for thisretrospective study. They comprised 26 complete UCLPpatients and 19 incomplete UCLP patients, 29 males and 16females. The mean age of the 26 complete UCLP patients was

Table 2 Interpretation of interclass correlation coefficient

Interclass correlation coefficient Strength of agreement

<0.40 Poor0.40e0.74 Moderate to good0.75e1.00 Perfect

13.75 years and the mean age of the 19 incomplete UCLPpatients was 12.6 years. These patients all underwentprimary lip repair, without primary nasal correction, atdifferent hospitals before 1 year of age. They did not undergosecondary nasal surgery prior to coming to our hospital.Frontal, submental and profile view photographs of 45patients were obtained when they came to our hospital.

Aesthetic assessment

A panel of seven judges was included in this assessment.Before judging, the judges were shown the colour slidesand photos all 45 patients to instruct on scaling. This fam-iliarised them with the views and showed the range of nasaldeformity.

Firstly, the frontal, submental and cleft side lateral slideswere presented, respectively, to the judges with the twodifferent areas: the nasal area only and the nasolabial area,so there were six two-dimensional slides for each patient.The judges evaluated the patients’ aesthetic nasal appear-ance. They were asked to rate the appearance of the cleftnose, using a five-point scale for each photo: 1 representingvery good appearance, 2 representing good, 3 representingfair, 4 is poor and 5 is very poor. One week later, judgers re-evaluated the photos of only the nasal region and scaled theirappearance. The procedure was similar to the previousdescription except the sequence of photos was changed.Then another person, rather than judges, tallied thesescores. He combined the scores of the frontal and submentalviews of each patient, the scores of the frontal and lateralviews of each patient, the scores of the submental andlateral views of each patient and the combined scores of thefrontal, submental and lateral views of each patient asdifferent methods for rating nasal appearance. Using themean scores from these seven judges, the summed assess-ment of the frontal view, submental view and lateral views ofeach patient allowed two subgroups to be developed: thebest and the worst patient outcomes.

Slides were presented to the judges for 10 seconds.The slides were projected on to a screen 6 feet in frontof the judge. The size of the projected photos approxi-mated the size of the human face. When rating, thejudges were forbidden discussion.

Statistical analysis

Statistical analysis was carried out using SPSS11.0. The levelof intra-judge agreement was assessed using the Kappa test

1 0.47 0.39e0.962 0.71 0.55e0.873 0.46 0.30e0.634 0.48 0.31e0.645 0.53 0.32e0.746 0.61 0.45e0.757 0.64 0.50e0.78

Table 4 Reliability of assessments in different methods for rating nasal appearance

Frontalview

Submentalview

Cleft sidelateral view

Frontal plussubmental view

Frontal pluslateral view

Submental andlateral view

Total

Nasal area only 0.539 0.642 0.601 0.700 0.648 0.674 0.728Nasolabial area 0.538 0.627 0.641 0.674 0.591 0.647 0.666

Development of a method for rating nasal appearance after cleft lip repair 1439

and the level of inter-judge agreement was assessed usingan interclass correlation coefficient (ICC). The Kappa valueand ICC classification are shown in Tables 1 and 2. Two-independent sample nonparametric test was used todetermine whether there was any significant differencebetween appearance of the best and the worst subgroups inthe frontal, submental and cleft side lateral views.

Results

The intra-judge agreement was only calculated for thephotos of the nasal area. The Kappa values of the sevenjudges are shown in Table 3. We found that the level ofintra-judge agreement was moderate to substantial.

The inter-judge reliability was calculated for bothphotos of the nasal area alone and photos of the nasolabialarea. In each of the two groups, the ICC values werecalculated using the different methods for rating nasalappearance (Table 4).

In Table 4, The ICC value of the assessment based onusing all three views is the highest, suggesting that the levelof agreement among the judges is the best. The level ofagreement is better with combined assessment of twodifferent views rather than a single view. The level ofagreement with the single frontal view is the poorest of all.The ICC values of the nasal only slides were higher than thenasolabial slides, except for the cleft side lateral view. Thissuggests that evaluating the nasal complex in isolation isbetter than as a combined nasolabial region.

Two subgroups, the best and worst nasal appearance,each comprising six patients, were created based on thesummation of the mean scores for frontal, submental andlateral views for each patient. A significant difference wasfound between the best and the worst subgroups for eachof three views (Table 5).

Discussion

In the literature, most studies analyse nasolabial appear-ance.7,8,9 However, when a study focuses specifically on the

Table 5 Significant differences in the aesthetic ratingsbetween the best and worst aesthetic subgroups in nasalonly slides

Slide view for the rating P value

Front 0.000< 0.05 (significantlydifferent)

Cleft side lateral 0.013< 0.05 (significantlydifferent)

Submental 0.000< 0.05 (significantlydifferent)

nasal area, a new scaling method should be developed, forexample, studies looking for a correlation between thesubjective nasal assessment and the objective nasalmeasurement, or those comparing the outcome of differentrhinoplasties. If the photographs included the lip area, thenasal subjective assessment is inevitably inaccurate becauseof the influence of the lip. As many papers noted, thejudgement of the nasolabial area is likely to be influenced bysurrounding features unrelated to the cleft,10,11 the judg-ment of the nasal area is likely to be influenced by the lip.

If a scaling method is to be a useful tool in analysis ofnasal appearance, it must satisfy the requirement ofscientific reliability. The inter-judge agreement on thenasal area ranged from moderate to substantial (0.54e0.73), and the inter-judge agreement on the nasolabialarea ranged from moderate to substantial (0.53e0.68). Sothe isolated nasal photos were generally better than thenasolabial photos. This assists the above hypothesis thatthe lip may influence the nasal assessment. Because of thelip, the attention of some judges is distracted, thus bias isintroduced. In order to avoid this error, the nasal onlyphotos were used. The unique exception was that, in thecleft side lateral views, the ICC value of nasal photos wasnot higher than the nasolabial slides. This may be becausethe labial scar and labial asymmetry cannot be seen in thelateral views, further assisting the above hypothesis. So thenew method is a useful tool.

Another important facet of the new method is decidingwhich views should be selected to evaluate nasal appear-ance. In former studies, Roberts12 and Kane13 used only thefrontal view to assess the aesthetics, Thomson14 used thefrontal and submental views, Johnson6 used the frontal and

Figure 1 The frontal view of the patient with the bestoutcome.

Figure 2 The submental view of the patient with the bestoutcome.

Figure 4 The frontal view of the patient with the worstoutcome.

1440 X. He et al.

cleft side lateral views and Russell15 used the frontal,submental, three-quarter and cleft side lateral views toevaluate appearance, so there were no unified criteriaregarding views. In this study, the ICC values of thedifferent views were calculated (Table 4). The frontal slidewas lowest and the slide including all three views was thehighest. Meanwhile, the degree of agreement was betterwhen two different views were used than a single view.

The nose is a three-dimensional organ and the singletwo-dimensional slide cannot accurately reflect the wholefeature. The frontal, submental and cleft side lateral viewsrepresent the coronal plane, horizontal plane and ante-roposterior plane separately. The combined assessment canreflect the three-dimensional general features. So thismethod’s content validity is good. Using only one view canlead to bias more easily than using two or three views. So inthe new method, three different views were combined toassess nasal appearance.

Figure 3 The lateral view of the patient with the bestoutcome.

The other factor for a useful tool in analysis of nasalappearance is sensibility. In order to test the sensibility, thebest and worst subgroups were developed based on thetotal score. All three views showed significant differencebetween the best and the worst subgroups (Table 5).Comparison of this sensibility with other published studiesis hampered by quoting reproducibility statistics or by theuse of different statistical methods. There is a study suit-able for comparison with this new method. Russell15

pointed out that differences in nasal aesthetic ratings, asassessed from the frontal, basal views, were not detectedbetween the best and worst aesthetic subgroups in hisstudy using his own scaling method. So, sensibility of thisnew method is better than Russell’s.

In most previous studies, the nasal area was alwaysdivided into different parts: the nasal tip, the alar rim, alarbase, nostril, columella and so on.16,17 This new methodrequires consideration of the nose as a whole and allocation

Figure 5 The submental view of the patient with the worstoutcome.

Figure 6 The lateral view of patient with the worstoutcome.

Development of a method for rating nasal appearance after cleft lip repair 1441

of a score from 1 to 5. The nose is an integrated organ,divided evaluation of the individual parts cannot reflect thereal contour. In addition, a greater number of categoriesmay make the scoring system complicated and the repro-ducibility statistics are not overtly better.6

The present method allows sensitive and reliable ratingof the individual features of the nose and appears workablein practice. The convenience is another merit of thismethod. The best and worst patient outcomes assessed bythis method are shown in Figures 1e6.

References

1. Sundine MJ, Phillips JH. Treatment of the unilateral cleft lipnasal deformity. J Craniofac Surg 2004;15:69e76 [discussion76e77].

2. Lindsay WK, Farkas LG. The use of anthropometry in assessingthe cleft-lip nose. Plast Reconstr Surg 1972;49:287e93.

3. Peck S, Peck L, Kataja M. Skeletal asymmetry in estheticallypleasing faces. Angle Orthod 1991;61:43e7.

4. Whittle J. Preoperative anthropometric analysis of the cleftchildren’s face: a comparison between groups. Int J Surg 2004;2:91e5.

5. Farkas LG, Hajnis K, Posnick J. Anthropometric and anthropo-scopic findings of the nasal and facial region in cleft patientsbefore and after primary lip and palate repair. Cleft PalateCraniofac J 1993;30:1e12.

6. Johnson N, Sandy J. An aesthetic index for evaluation of cleftrepair. Eur J Orthod 2003;25:243e9.

7. Williams HB. A method of assessing cleft lip repair: comparisonof LeMesurier and Millard techniques. Plast Reconstr Surg 1968;41:103e7.

8. Rullo R, Carinic F, Mazzarella N, et al. Delaire’s cheilo-rhinoplasty: unilateral cleft aesthetic outcome scoredaccording to the Eurocleft guidelines. Int J Pediatr Oto-rhinolaryngol 2006;70:463e8.

9. Tobiasen JM, Hiebert JM, Boraz RA. Development of scales ofseverity of facial cleft impairment. Cleft Palate Craniofac J1991;28:419e25.

10. Asher-McDade C, Roberts C, Shaw WC, et al. Development ofa method for rating nasolabial appearance in patients withclefts of the lip and palate. Cleft Palate Craniofac J 1991;28:385e91.

11. Zhou ZB, Sun YG, Luo Y. Comparison of three methods ofassessment of cleft lip and nose deformity. J Modern Stomatol2005;19:139e41.

12. Roberts-Harry DP, Hathorn IS, Stephens CD. The ranking offacial attractiveness. Eur J Orthod 1992;14:483e8.

13. Kane AA, Pilgram TK, Moshiri M, et al. Long-term outcome ofcleft lip nasal reconstruction in childhood. Plast Reconstr Surg2000;105:1600e8.

14. Thomson HG, Reinders FX. A long-term appraisal of theunilateral complete cleft lip repair: one surgery’s experience.Plast Reconstr Surg 1995;96:549e63.

15. Russell KA, Waldman SD, Tompson B, et al. Nasal morphologyand shape parameters as predictors of nasal esthetics in indi-viduals with complete unilateral cleft lip and palate. CleftPalate Craniofac J 2001;38:476e85.

16. Asher-McDade C, Brattstrom V, Dahl E, et al. A six-centerinternational study of treatment outcome in patients withclefts of the lip and palate: part 4: assessment of nasolabialappearance. Cleft Palate Craniofac J 1992;29:409e12.

17. Morrant DG, Shaw WC. Use of standardized video recordings toassess cleft surgery outcome. Cleft Palate Craniofac J 1996;33:134e42.