three-dimensional facial morphometry of attractive ... · risultati. gli adolescenti attraenti...

Objectives. To identify reference standards and possible esthetic features of facial size and form of Italianadolescent boys and girls. Methods and Subjects. A three-dimensional electromagnetic digitizer was used to collect the coordinatesof 50 facial landmarks (forehead, eyes, nose, cheeks, mouth, jaw, ears) in 231 healthy, reference adole-scents (10-17 years old) and in 93 “attractive” adolescents selected by a commercial casting; soft tissuefacial angles, distances and volumes were computed.Results. Attractive adolescents had wider and shorter faces than reference subjects of the same age andsex, with relatively larger upper and middle facial thirds, and a reduced mandible relative to the maxilla.Lips were larger and more prominent, and the nasolabial angle was reduced, but in older boys the effectwas reversed. The prominence of the soft-tissue profile, and of the maxilla relative to the mandible, werelarger in attractive boys, but smaller in attractive girls than in their reference peers. Attractive adolescentshad a relatively less prominent chin relative to the lower lip, with increased values of the mentolabial an-gle. In the horizontal plane, attractive “young” adolescents had a more obtuse mandibular convexity an-gle, while the opposite pattern was observed in the “old” adolescent subjects, with a relatively more pro-minent chin. Attractive adolescents had smaller noses than reference subjects of the same age and sex.Conclusion. Esthetic reference values have been identified which could be used to determine optimal timingand goals in orthodontic treatment.

Sforza C, Laino A, D'Alessio R, Grandi G, Catti F, Ferrario VF. Three-dimensionalfacial morphometry of attractive adolescent boys and girls. Prog Orthod2007;8(2):268-81.

Three-dimensional facial morphometry ofattractive adolescent boys and girls

Chiarella Sforza*, Alberto Laino**, Raoul D'Alessio***, Gaia Grandi****, Francesca Catti****, Virgilio Ferruccio Ferrario*****

*Professor, Functional Anatomy Research Center, Department of Human Morphology, University of Milan, Milano, Italy**Associate Professor, Dental, Oral, and Maxillo-Facial Sciences, Section of Orthodontics, University Federico II, Naples, Italy***Private Practice, Rome, Italy****Postdoc, Functional Anatomy Research Center, Department of Human Morphology, University of Milano, Milan, Italy*****Professor and Department Chair, Functional Anatomy Research Center, Department of Human Morphology, Universityof Milan, Milano, Italy

Correspondence to:prof. Chiarella SforzaDipartimento di Morfologia Umana, via Mangiagalli 31, 20133 Milan, ItalyTelephone: +39.02.503.155.07Fax: +39.02.503.153.87E-mail: [email protected]

PROGRESS in ORTHODONTICS 2007; 8(2):268-281

1 Facial morphometry of attractive adolescents

268

Obiettivi. Identificare eventuali canoni estetici nella forma e dimensioni del-la faccia degli adolescenti italiani. Metodi e Materiali. Un digitalizzatoreelettromagnetico tridimensionale ha permesso di raccogliere le coordina-te di 50 punti di repere facciali (fronte, occhi, naso, guance, bocca, man-dibola, orecchie) in 231 adolescenti normali (10-17 anni) e in 93 prove-nienti da un casting pubblicitario (”attraenti”); sono stati calcolati angoli,distanze, e volumi facciali. Risultati. Gli adolescenti attraenti hanno la fac-cia più larga e corta dei loro coetanei, con un maggiore sviluppo di ma-scellare e fronte, ed una mandibola in proporzione più piccola. Le labbrasono di maggiori dimensioni negli adolescenti attraenti, e più sporgenti,con un angolo nasolabiale ridotto. Il naso degli adolescenti attraenti è didimensioni minori di quello dei coetanei. Nei maschi attraenti, il profilo èpiù sporgente, con una maggiore prominenza mascellare; nelle femminesi osserva l’opposto. La prominenza del mento aumenta negli adolescentipiù vecchi. Conclusione. La definizione di parametri estetici di riferimentopuò fornire indicazioni per individuare modalità e finalità di trattamentoortodontico ottimali.

Key words: face; adolescents; attractiveness; soft tissues.

Introduction

In the contemporary Western so-ciety, esthetics is becoming a mat-ter of concern at all ages and in allsocial strata. Among all parts of thebody, primary attention is given tothe face1-4. Facial appearance isfundamental for communication andinteract ion with the environ-ment3,5,6, and it carries informationthat allows the identification of asingle person7,8. Bones, muscles,cutaneous and subcutaneous layersall contribute to a unique morpho-logy in the single individual9; thismorphology, even continuouslychanging during growth, develop-ment and aging10, always main-tains its peculiar characteristics11.

The definition of esthetic standardsbrings together scientists, clini-cians, artists, who tried to codifywhich facial dimensions, anglesand ratios make a person look mo-re attractive. Indeed, no universalcanons seem to exist, and estheticcharacteristics are often peculiar toeach cultural background2,6,12-15.Therefore, the quest for the sacredGrail of facial attractiveness is stillopen. From a psychological pointof view, average facial features,symmetry, youthfulness, neoteny(baby like appearance) all need tobe taken into consideration, espe-cially for women1,2,16,17. In con-trast, facial attractiveness in adultmen mingle perceptions of ma-sculinity, health, developmental sta-bility and social dominance1,18,19. Currently, the perception of attrac-tiveness is extremely influenced bymedia: television, cinema, adver-tisements, fashion industries, all en-ter in our life bringing facial “stan-dards” that should convey percep-tions of beauty, healthiness, fitness,mixed with feelings of social achie-vement, intelligence, richness, andhappiness: a beautiful face beco-mes the key to the success3,20,21.The clinical specialists working inthe facial area are thereforeasked by patients to provide me-dical and surgical modificationsof non-at t ract ive dentofacialphysiognomies14,17,22. As a re-sult, orthodontists, maxillofacialand plastic surgeons should havea deep understanding of thosequantifiable, objective facial cha-racteristics that are considered bythe public as “attractive”2,17,23,24.

Facial morphometry of attractive adolescents 2

PROGRESS in ORTHODONTICS 2007; 8(2):268-281 269

Acknowledgement: The authorsare greatly indebted to all theStaff of the Laboratorio di Anato-mia Funzionale dell'ApparatoStomatognatico, Università degliStudi di Milano, who participatedin data collection and analysis(Dr’s Miriam Binelli, Claudia Del-lavia, Fabrizio Mian, GianlucaM. Tartaglia; Ms. Iolanda Deco-rato, Mr. Luca Pisoni, Mr. MatteoRossini). Dr Patrizia Frangella col-laborated to the organization ofthe data collection in attractiveadolescents. The expert secreta-rial assistance of Ms. Cinzia Lo-zio is gratefully acknowledged.Financial support was obtainedfrom University of Milan (FIRST,2006), and from the Board of Di-rectors of the Società Italiana DiOrtodonzia (SIDO).



270

In a previous investigation, weanalyzed attractive and reference,normal children with deciduousand early mixed dentition, andfound that attractive children main-tained several facial characteristicsof baby like appearance: a largeface, with relatively large maxillaand forehead, and reduced verti-cal dimensions. Lips were morevoluminous, mouth was bigger,and the soft-tissue facial profile wasmore convex in attractive children,with a more prominent maxilla re-lative to the mandible, than in thereference coetaneous24,25.In the current investigation, the three-dimensional facial characteristics ofadolescent boys and girls conside-red “attractive” were measured with

a non-invasive computerized instru-ment, and several measurements ob-tained. Data were compared to tho-se collected in healthy children ofthe same age, sex and ethnicity, se-lected using criteria of dentofacialnormality6,12,14. The possible pre-sence of measurable esthetic cha-racteristics was assessed.

Materials and subjects

Subjects

Three hundred and 24 white Cau-casian, Northern Italian adolescentboys and girls aged 10-17 yearswere analyzed.A first group, “reference” adole-

scents, was made of 141 boysand 90 girls. All subjects werehealthy, they had normal dento-facial dimensions and propor-tions; no subjects with a previoushistory of craniofacial trauma orwith congenital anomalies wereincluded. These adolescents we-re attending several schools inMilan and surroundings, and partof their data had already beenpublished10.A second group, “beautiful” or“attractive” adolescents, compri-sed 46 boys and 47 girls selec-ted by a commercial castingagency. The same selection cri-teria used in the previous investi-gation performed on young chil-dren24,25 were used: the agency

Résumé. Pour identifier des normes de référence esthétiques possibles de dimension et forme faciale degarçons et de filles adolescents italiens. Méthodes et objets. Un convertisseur analogique/numérique élec-tromagnétique tridimensionnel a été utilisé pour rassembler les coordonnées de 50 loints faciales (front,yeux, nez, joues, bouche, mâchoire, oreilles) dans 231 adolescents en bonne santé, de référence (10-17ans de ) et dans 93 adolescents "attirants" choisis par un group commercial; ont a calculé des angles fa-ciaux, les distances et les volumes de tissus mous. Résultats. Les adolescents attirants ont eu des visagesplus larges et plus courts que des sujets de référence du mêmes âge et sexe, avec des troisièmes faciauxsupérieurs et moyens relativement plus grands, et une mâchoire inférieure réduite relativement au maxil-laire supérieur. Les lèvres étaient plus grandes et plus en avant, et l'angle nasolabial a été réduit, mais dans des garçonsplus âgés l'effet a été renversé. La prominence du profil des tissu doux, et du maxillaire supérieur relati-vement à la mâchoire inférieure, étaient plus grande dans les garçons attirants, mais plus petite dans lesfilles attirantes que dans leur référence. Les adolescents attirants ont eu un menton relativement moins enavant relativement à la lèvre inférieure, avec des plus grandes valeurs de l'angle mentolabial. Dans le planhorizontal, les "jeunes" adolescents attirants ont eu un angle mandibulaire plus obtu de convexité, alorsqu'on observait le modèle opposé dans les "vieux" sujets adolescents, avec un menton relativement plusen avant. Les adolescents attirants ont eu de plus petits nez que des sujets de référence du mêmes âge etsexe. Conclusion. On a identifié des valeurs de référence esthétiques qui pourraient être employées pourdéterminer la temps et les buts optimaux dans le traitement orthodontique.

Traduit par Maria Giacinta Paolone



was asked to provide adole-scents of both sexes wi th a“beautiful”, “attractive” face, wi-thin a well defined age rangeand ethnicity. These boys andgirls normally act in cinema andtelevision, are used for adverti-sing, and the fashion industry;overall, their faces could be con-sidered “trendy”, and should con-vey “positive” feelings4. All the analyzed adolescents,and their parents/legal guar-dians gave their informed con-sent to the experiment. All pro-cedures were non-invasive, didnot provoke damages, risks ordiscomfort to the subjects, andwere preventively approved bythe local ethic committee.

Collection of facial landmarks

All the procedures were non-in-vasive, not potentially harmful,did not provoke pain and did notuse any instrument or energy cur-rently considered to be potentiallydangerous to the present or futu-re health of the subjects or of her/his offspring. The data collection procedure tookplace in two separate steps, and itwas followed by off-line calcula-tions26. At first, for each subject, asingle experienced operator locateda set of 50 soft-tissue landmarks byinspection and/or palpation10, andmarked them on the cutaneous sur-face using a liquid eye-liner. Theeye-liner can be easily washed from

the face with soap and water. Du-ring landmark marking, the subjectssat relaxed in a position suitablefor a correct identification of facialfeatures. For each subject, this pha-se lasted less than 5 minutes. Theprevious marking of the landmarksallowed a subsequent faster datacollection procedure, and providedto the operator all the time neces-sary for a correct identification,even of those landmarks that mustbe palpated for their individuation(for instance, gonion).In the second step, the three-dimen-sional (x, y, z) coordinates of the fa-cial landmarks were obtained with acomputerized electromagnetic digiti-zer (3Draw, Polhemus Inc., Colche-ster, VT) that supplies real metric data

Objetivos. Fueron tomados en consideración un grupo de adolescentes Italianos “hombres y mujeres” , paraidentificar estándares posibles del tamaño y forma facial en referencia a las características estéticas. Mé-todos y temas. Fue utilizado un digitador electromagnético tridimensional para acumular las coordenadasde 50 puntos faciales (frente, ojos, nariz, mejillas, boca, quijada, oídos) en 231 adolescentes sanos, entre(10-17 años) y en 93 adolescentes “atractivos” seleccionados para un comercial; de esta forma fueron com-putarizados los ángulos faciales, las distancias y los volúmenes del tejido blandos. Resultados. Los adole-scentes atractivos tuvieron caras más anchas y más cortas en referencia a los sujetos de la misma edad ysexo, con el tercio facial superior y medios relativamente más grandes. Además presentaron una mandí-bula mas pequeña con respecto al maxilar. Los labios eran más grandes y más prominentes, y el ángulonasolabial era mas reducido; mientras que en los adolescentes mas viejos el efecto fue inverso. La promi-nencia del perfil de los tejidos blandos y del maxilla con respecto a la mandíbula, fue más grande en losadolescentes atractivos; y más pequeño en las adolescentes atractivas con referencia al mismo. Los adole-scentes atractivos tenían una barbilla relativamente menos prominente concerniente al labio inferior con va-lores aumentados del ángulo mentolabial. En el plano horizontal, los adolescentes “jóvenes” atractivos te-nían un ángulo de la convexidad mandíbular más obtuso, mientras que en los adolescentes mas “viejos”se observo todo lo contrario, ose a una barbilla relativamente más prominente. Los adolescentes atractivostenían narices más pequeñas que los sujetos de referencia de la misma edad y sexo. Conclusión. Fueronidentificados los valores de referencia estéticos que se podrían ser utilizar para determinar los objetivos ymetas óptimas para el tratamiento ortodóntico.

Traducido por Santiago Isaza Penco



272

independent from external referencesystems. The digitizer has a resolutionof 0.005 mm/mm of range, and anaccuracy of 0.08 mm, with the re-ceivers located within 76 cm of thetransmitter. During data collection, thesubjects sit in a natural head positionin a chair with a backrest, where acephalostat fixes the subject’s head.Vertically and horizontally movablesystems accommodation for differentsitting heights and head dimensions.To obtain the natural head position,the subjects were asked to look at thereflected image of their eyes in a mir-ror positioned at eye level at appro-ximately 2 m of distance. They wereasked to close their eyes, to keeptheir teeth in contact, and to not mo-ve for the duration of the data col-lection26. Using the instrument stylus, a singleoperator gently touched the faciallandmarks in a standardized se-quence. Data collection takes ap-

proximately 1 minute. Before dis-charging the subject, the computerperformed a fast reconstruction of fa-cial morphology using the three-di-mensional coordinates of the collec-ted landmarks, and a check bet-ween the video image and the faceof the subject was made to assessthe correct sequence of landmarks,and any motion artifact. The proce-dure was repeated immediately in1% of acquisitions.The reproducibility of landmarkidentification, marker positioningand the reproducibility of the datacollection procedure have been re-ported26, with random errors of1.20 mm (adult men) and 0.95mm (adult women), correspondingto 1.04-1.05% of the relevant na-sion-mid tragion distance. Files ofthe three-dimensional coordinateswere obtained, and original com-puter programs were used for allthe subsequent off-line calculations.

Data analysis

Fifty facial soft tissue landmarkswere collected on each subject(Fig. 1):• midline landmarks: tr, trichion; g,

glabella; n, nasion; prn, prona-sale; c’, columella; sn, subnasale;ls, labiale superius; sto, stomion;li, labiale inferius; sl, sublabiale;pg, pogonion; me, menton;

• paired landmarks (right andleft side noted r and l): exr,exl, exocanthion; enr, enl, en-docanthion; osr, osl, orbitalesuperius; orr, orl, orbitale; ftr,ftl: frontotemporale; chkr, chkl,cheek; zyr, zyl, zygion; tr, tl,tragion; alr, al-l, alare; acr,acl, nasal alar crest; itnr, itnl,inferior point of the nostrilaxis; stnr, stnl, superior pointof the nostril axis; cphr, cphl,crista philtri; chr, chl, cheilion;gor, gol, gonion; prar, pral,

Fig. 1 Digitized facial landmarks: tr, trichion; g, glabella; n, na-sion; prn, pronasale; c’, columella; sn, subnasale; ls, labiale su-perius; sto, stomion; li, labiale inferius; sl, sublabiale; pg, po-gonion; me, menton; ex, exocanthion; en, endocanthion; os,orbitale superius; or, orbitale; ft, frontotemporale; zy, zygion;chk, cheek; t, tragion; pra, preaurale; sa, superaurale; pa, po-sturale; sba, subaurale; al, alare; ac, nasal alar crest; itn, in-ferior point of the nostril axis; stn, superior point of the nostril axis;cph, crista philtri; ch, cheilion; go, gonion; pra, preaurale; sa,superaurale; pa, postaurale; sba, subaurale.



preaurale; sar, sal, superau-rale; par, pal, postaurale;sbar, sbal, subaurale.

The three-dimensional coordinatesof the 50 landmarks were used toestimate a set of linear distances, an-gles, facial volumes and areas asdetailed elsewhere10,13,24-28. Eucli-dean geometry was used for all cal-culations; the volumes of facial struc-tures were estimated from the sum ofseveral tetrahedra, with the 50 land-marks serving as nodes (vertices ofthe tetrahedra). In particular, the fol-lowing were computed:• distances (unit, mm): facial

height (n-pg); upper facialwidth (ex-ex); lower facialwidth (go-go); middle facialdepth (sn-t); mandibular cor-pus length (pg-go); mouthwidth (ch-ch); vermilion height(ls-li); upper lip to E-line dis-tance, ls-(prn-pg); lower lip toE-line distance, li-(prn-pg);

• angles (unit, degrees): facialconvexity including the nose(n-prn-pg); lower face conve-xity (go-pg-go); maxillary pro-minence, soft tissue analog ofskeletal ANB angle (sl-n-sn);nasolabial (prn-sn-ls); mento-labial (li-sl-pg); interlabial (sn-ls^sl-pg);

• areas (unit: cm2): area of thevermilion of the upper and lowerlips (between landmarks chr, ls,chl, sto; and chr, li, chl, sto);

• volumes (unit: mm3): total fa-cial volume (volumes of all fa-cial structures from the exter-nal cutaneous surface up to aquasi-frontal plane passingthrough trichion, tragi and go-nia), subdivided into facialupper thi rd volume ( fore-head), facial middle third vo-lume (maxilla), facial lowerthird volume (mandible); na-sal volume.

Statistical calculations

“Reference” and “attractive” boysand girls were divided into twoage groups for each sex. For girls,24 attractive and 39 referencegirls were aged 10 to 12 years(“young” adolescent girls); 23 at-tractive and 51 reference girls we-re aged 13 to 15 years (“old”adolescent girls). For boys, 22 at-tractive and 87 reference boyswere aged 12 to 14 years(“young” adolescent boys); 24 at-tractive and 54 reference boyswere aged 15 to 17 years (“old”adolescent boys). Different agegroups were used for the two se-xes because of the different timingof pubertal growth spurt10.Descriptive statistics (mean andstandard deviation) were compu-ted for each group, angular datawere assessed using their rectan-gular components (sine and cosi-

Fig. 2 Upper facial width (ex-ex) in the analyzed adolescentsubjects. Mean values for male (light blue) and female (pink)attractive (solid colors) and reference (hatched colors) ado-lescents in the two age groups. Darker shades correspond toolder adolescents.

Fig. 3 Lower facial width (go-go) in the analyzed adolescentsubjects. Mean values for male (light blue) and female (pink)attractive (solid colors) and reference (hatched colors) ado-lescents in the two age groups. Darker shades correspondto older adolescents.

ne). Comparisons between the twogroups of adolescents (reference,attractive) were performed withineach sex and age group using in-dependent Student’s tests with two-tailed distributions, and a signifi-cance level set at 5% (p < 0.05).

Results

Upper facial width (ex-ex) resultedlarger in boys than in girls in both“young” and “old” adolescent agegroups, and increased with age(Fig. 2). In all groups, attractive

adolescents had a wider upper fa-ce than reference adolescents ofthe same age and sex; the diffe-rences were statistically significantin both female groups, and in the“young” adolescent males (p <0.02, Student’s t for independent



274

Fig. 4 Mandibular corpus length (pg-go) in the analyzedadolescent subjects. Mean values for male (light blue) and fe-male (pink) attractive (solid colors) and reference (hatched co-lors) adolescents in the two age groups. Darker shades cor-respond to older adolescents.

Fig. 5 Facial height (n-pg) in the analyzed adolescent sub-jects. Mean values for male (light blue) and female (pink) at-tractive (solid colors) and reference (hatched colors) adole-scents in the two age groups. Darker shades correspond toolder adolescents.

Fig. 6 Middle facial depth (sn-t) in the analyzed adolescentsubjects. Mean values for male (light blue) and female (pink)attractive (solid colors) and reference (hatched colors) ado-lescents in the two age groups. Darker shades correspond toolder adolescents.

Fig. 7 Facial volume in the analyzed adolescent subjects.Mean values for male (light blue) and female (pink) attracti-ve (solid colors) and reference (hatched colors) adolescentsin the two age groups. Darker shades correspond to olderadolescents.



samples). Similar effects of sex andage were observed for lower fa-cial width (go-go) and mandibularcorpus length (pg-go), with maleslarger than females, and olderadolescents larger than youngeradolescents (Figs 3, 4). Overall,reference adolescents had largerfaces and longer mandibles thanattractive adolescents, with the ex-ception of the “young” girls, but inno occasion the trends were sub-stantiated by statistical significance(p > 0.05). In both sexes and in both agegroups, facial height (n-pg) andmiddle facial depth (sn-t) were smal-ler in the attractive adolescent thanin their reference peers (Figs 5, 6);the differences were statistically si-gnificant in the “old” male adole-scents (n-pg, p = 0.003) and in the“old” female adolescents (sn-t, p =0.003). Age and sex influencedboth measurements, with longer

and deeper faces in males than infemales, and in “old” than in“young” adolescents. The modifications in facial dimen-sions resulted in variations in facialvolumes: total facial volume waslarger in attractive girls in both agegroups, and in attractive “young”adolescent males than in the refe-rence subjects, with significant dif-ferences in the “young” adolescents(p = 0.021 in girls, p = 0.009 inboys). In contrast, “old” adolescentattractive boys had a significantlysmaller face then their referencepeers (p = 0.009). Age-related in-crements in facial volume were mo-re evident in boys than in girls; se-xual dimorphism was present inboth age groups (Fig. 7). A different arrangement of facialthirds was observed in the attractivesubjects when compared to the re-ference ones, with relative largerupper and middle thirds. Indeed,

the forehead (facial upper third) oc-cupied a larger part of the face inboth attractive boys and girls, andat both ages (Fig. 8). The effectwas particularly evident in the “old”adolescent girls and in the “young”adolescent boys (p < 0.001). Ove-rall, the ratio decreased with age(older adolescents had a relativelysmaller forehead than younger ado-lescents), and, in each age group,it was larger in girls than in boys. Inattractive adolescents, the mandi-ble was reduced relatively to themaxilla, with mandibular/ maxil-lary volume ratios all less than 94%(Fig. 9). The differences were stati-stically significant in the girls, and inthe “young” adolescent males (p <0.05). In reference girls, the ratio in-creased with age, with a relativelarger grow in the facial lower thirdthan in the middle facial third.In the middle facial third, nasalvolume increased with age, and it

Fig. 8 Forehead volume as a percentage of facial volumein the analyzed adolescent subjects. Mean values for male(light blue) and female (pink) attractive (solid colors) and re-ference (hatched colors) adolescents in the two age groups.Darker shades correspond to older adolescents.

Fig. 9 Mandibular volume as a percentage of maxillary vo-lume in the analyzed adolescent subjects. Mean values formale (light blue) and female (pink) attractive (solid colors) andreference (hatched colors) adolescents in the two age groups.Darker shades correspond to older adolescents.

was larger in boys than in girls(Fig. 10). Attractive adolescentshad smaller noses than referencesubjects of the same age and sex,with significant differences in the“old” adolescents (p < 0.002).The area of the vermilion of the to-tal (upper plus lower) lip was lar-ger in the attractive than in the re-ference adolescents (Fig. 11),

with a larger contribution of theupper lip (Fig. 12). In the attracti-ve adolescents, vermilion height(ls-li) was a larger percentage ofmouth width (ch-ch) than in the re-ference subjects (Fig. 13); at bothages, it was larger in attractivegirls than in attractive boys. In theattractive adolescents, both lipswere nearer to the esthetic E-line

(prn-pg) than in the reference ado-lescents (Fig. 14); the observedtrend was not substantiated by sta-tistically significant differences. Inboth sexes, the distances increa-sed as a function of age; in the lo-wer lip, values were somewhatsmaller (that is, the lip was moreprominent) in girls than in boys.Attractive boys had more acute soft



276

Fig. 10 External nasal volume in the analyzed adolescentsubjects. Mean values for male (light blue) and female (pink)attractive (solid colors) and reference (hatched colors) ado-lescents in the two age groups. Darker shades correspond toolder adolescents.

Fig. 11 Area of the vermilion of the total lip in the analyzedadolescent subjects. Mean values for male (light blue) and fe-male (pink) attractive (solid colors) and reference (hatched co-lors) adolescents in the two age groups. Darker shades cor-respond to older adolescents.

Fig. 12 Percentage areas of the vermilion of the upper andlower lips in the analyzed adolescent subjects.

Fig. 13 Mouth vermilion height (ls-li) as a percentage of mouthwidth (ch-ch) in the analyzed adolescent subjects. Mean valuesfor male (light blue) and female (pink) attractive (solid colors) andreference (hatched colors) adolescents in the two age groups.Darker shades correspond to older adolescents.



tissue profiles than reference boys,with smaller angles of facial con-vexity including the nose (n-prn-pg,Fig. 15), a difference significant inthe 13-to-15 year old age group(p = 0.014). Accordingly, theyhad a larger maxillary prominencerelative to the mandible, with asmaller soft tissue analog of skele-tal ANB angle (sl-n-sn, Fig. 16). Ingirls, the reverse pattern wasfound, with less acute facial profi-les (p = 0.049 in 10-to-12 yearold girls), and reduced maxillaryprominence (p = 0.008 in 14-to-16 year old girls) in the attractivethan in the reference subjects.In the middle and lower parts ofthe face, reduced nasolabial (prn-sn-ls, Fig. 17) and interlabial (sn-ls; sl-pg, Fig. 18) angles charac-terized attractive adolescents,who presented with relatively mo-re prominent lips, with the ex-ception of the 15-to-17 year oldboys. In the attractive boys, andin the attractive 14-to-16 year oldgirls, the chin was relatively less

prominent relative to the lower lip,with increased values of the men-tolabial angle (li-sl-pg, Fig. 19). Inthe “young” adolescent attractiveboys and girls, the horizontal pla-ne convexity of the lower facialthird (go-pg-go) was somewhatlarger than in their reference peers(Fig. 20), while the opposite pat-tern was observed in the “old”adolescent subjects, with a relati-vely more prominent chin.

Discussion

Analysis of the three-dimensionalarrangement of facial soft tissuesshould be made with a completemorphological and functional eva-luation, aimed at a global asses-sment of all elements classically for-ming beauty: precision, symmetry,coordination and functional struc-ture9. The first elements to be con-sidered are those describing themorphological structure, that formsthe base for function. Attractive per-

sons (like cinema and television ac-tors and actresses, fashion and ad-vertising models) are often belie-ved to possess peculiar esthetic re-lationships, but these do not seemto be a necessary condition for at-tractiveness17,28-30. Indeed, not only do not culturalbackground and secular trends in-fluence influence the perception ofbeauty6,12-15,20, but also a well-de-fined sexual dimorphism seems toemerge1,2,4,19,31. In adult Cauca-sian women, current opinions seemto prefer a prominent facial middlethird, with full lips; a relatively largeface with a reduced mandible andincreased forehead; reduced verti-cal development1,2,4,6,12,13,31,32.Overall, attractive women seem tomaintain several facial characteri-stics of baby like appearance. Incontrast, in adult men attractivenessseems to be positively influencedby facial markers of high testostero-ne levels, with a relative incrementof the facial lower third1,2,4,18,19,31. Further factors that may influence

Fig. 14 Upper lip to E-line distance, ls-(prn-pg), and lowerlip to E-line distance, li-(prn-pg) in the analyzed adolescentsubjects. Mean values for male (light blue) and female (pink)attractive (solid colors) and reference (hatched colors) ado-lescents in the two age groups. Darker shades correspond toolder adolescents.

Fig. 15 Facial convexity including the nose (n-prn-pg) in theanalyzed adolescent subjects. Mean values for male (lightblue) and female (pink) attractive (solid colors) and referen-ce (hatched colors) adolescents in the two age groups. Dar-ker shades correspond to older adolescents.



278

Fig. 16 Maxillary prominence angle (sl-n-sn) in the analyzedadolescent subjects. Mean values for male (light blue) and fe-male (pink) attractive (solid colors) and reference (hatched co-lors) adolescents in the two age groups. Darker shades cor-respond to older adolescents.

Fig. 17 Nasolabial angle (prn-sn-ls) in the analyzed adole-scent subjects. Mean values for male (light blue) and fema-le (pink) attractive (solid colors) and reference (hatched colors)adolescents in the two age groups. Darker shades correspondto older adolescents.

Fig. 18 Interlabial angle (sn-ls; sl-pg) in the analyzed ado-lescent subjects. Mean values for male (light blue) and female(pink) attractive (solid colors) and reference (hatched colors)adolescents in the two age groups. Darker shades correspondto older adolescents.

Fig. 19 Mentolabial angle (li-sl-pg) in the analyzed adole-scent subjects. Mean values for male (light blue) and fema-le (pink) attractive (solid colors) and reference (hatched colors)adolescents in the two age groups. Darker shades correspondto older adolescents.

the perception of beauty aregrowth and development: attrac-tiveness is becoming a matter ofconcern also in growing indivi-duals. Currently, children andadolescents are widely employedat the cinema and television, playan important part in the fashion in-dustry, and are a key element foradvertising. Children with a non-

attractive face are likely to be con-sidered less intelligent, and to beisolated and underscored thanchildren with an attractive fa-ce3,21,23. A beautiful face is be-lieved the key to the success, withparents and children looking formodifications of non-attractive den-tofacial physiognomies14,22. Previous studies on facial attracti-

veness in children and adolescentsmostly focused on dentolabial cha-racteristics, where subjects with ma-locclusion, irregular dental archesand thin lips were considered lessattractive than children with nor-mal occlusion, well-arranged den-tal arches, and medium or thicklips22,23. The global soft-tissue fa-cial esthetic characteristics in three



dimensions were scarcely asses-sed: Ferrario et al.33 analyzed therelationship between attractivenessand facial dimensions in 8-to-9year old children, Sforza et al.24

measured attractive and normalchildren aged 4 to 9 years. Ove-rall, attractive young children sha-red several of the facial characte-ristics found in attractive wo-men12,13,32: relatively large fore-head and more prominent maxilla,reduced vertical dimensions, volu-minous lips and more prominentsoft-tissue facial profile24.In the current investigation, some ofthe previous esthetic characteristicswere maintained, but also somesex- and age-related differencesemerged. In accord with previousfindings24, attractive subjects hadwider and shorter faces than refe-rence subjects of the same age andsex, with relatively larger upperand middle facial thirds, and a re-

duced mandible relative to the ma-xilla. Lips were larger and moreprominent, as previously found byMatoula and Pancherz6 for 14-to-24 years old attractive women; thenasolabial angle was reduced ingirls and younger boys, but in olderboys the effect was reversed. Ver-million height was a larger percen-tage of mouth width than in the re-ference subjects, thus confirming thefindings reported by Scott et al.23. The prominence of the soft-tissue pro-file, and of the maxilla relative to themandible, were larger in attractiveboys, but smaller in attractive girlsthan in their reference peers. An in-creased facial convexity was also re-ported by Kiekens et al.22 as theprincipal anterior-posterior charac-teristics for 10-to-16 years old ado-lescents considered attractive by lay-people, and by Matoula and Pan-cherz6. Attractive adolescents had arelatively less prominent chin relative

to the lower lip, with increased va-lues of the mentolabial angle, whe-reas the attractive children analy-zed by Sforza et al.24 had the op-posite pattern. In the horizontal pla-ne, attractive children and “young”adolescents had a more obtusemandibular convexity angle, whilethe opposite pattern was observedin the “old” adolescent subjects,with a relatively more prominentchin. This feature is in good accordwith the reports by Fink et al.34 andby Schaefer et al.19: high prenataltestosterone levels, and high circu-lating testosterone produce a moreprominent lower face, which is po-sitively linked to male attractiveness1.Attractive adolescents had smallernoses than reference subjects of thesame age and sex, in accord withprevious findings in adult women13,but in contrast with the patterns ob-served in attractive children24. Evenif the reduced nasal volume in at-tractive women may be an effect ofsurgical interventions, the current fin-ding seems to depend from actualesthetic preferences.Among the limitations of the currentstudy there is the analysis of onlytwo age groups for each sex. In-deed, the current attractive adole-scents were selected by a castingagency among those already invol-ved into cinema, television, and ad-vertising industry; additionally, theyshould move to the laboratory formeasurement, and this limited theselection to persons living in Milanand surroundings, thus reducing theirnumber. Possibly, this limited the num-ber of significant differences, andfor several measurements only trends

Fig. 20 Lower face convexity in the horizontal plane (go-pg-go) in the analyzed adolescent subjects. Mean values for ma-le (light blue) and female (pink) attractive (solid colors) and re-ference (hatched colors) adolescents in the two age groups.Darker shades correspond to older adolescents.



280

not substantiated by statistically si-gnificant values were found. Selection was made using the sameprocedure followed for attractive chil-dren24 and women12,13,32,35: boysand girls were independently selec-ted by professionals in a castingagency who were unaware of theactual scope of the investigation;they were asked to provide “attrac-tive” subjects of both sexes within awell defined age range and ethni-city. These faces were to be consi-dered “positive” and “acceptable”for mass medias4. This procedureavoided the specialized opinions ofdental and surgical professionals,that are often relatively more criticalin their assessment of facial estheticsthan non-professionals3,4,14,20,36,37,even if different findings about pla-stic surgeons have recently been re-ported23. Additionally, estheticsshould be evaluated by the layper-sons, who actual seek (and finallyjudge) orthodontic or maxillofacialtreatment22,37. The existence of so-me genetic background to the esthe-tic perception of attractiveness hasbeen reported1, as demonstrated ininfants as young as 3 months ofage17, but the effect is likely to bemodulated by several social fac-tors4,6,12,19,36.Male and female attractive ado-lescents were subdivided into dif-ferent age groups, girls being 2-3 years young than boys. Thisdifference should take the sex-re-lated discrepancies in the timingof pubertal growth spurt into ac-count10, thus allowing the asses-sment of more homogenous bio-logical ages. Nevertheless, theextension of the study to other

age groups could allow a deeperunderstanding of the actual bio-logical processes.A further limitation resides in the se-lected measurements: the analysisof soft tissue facial dimensions andangles should be implemented withthe assessment of symmetry17,20,35,and of facial shape (independentlyfrom dimensions19,27,34). Also, a wi-der set of angles and distancesshould be analyzed, with the inclu-sion of all facial structures, includingeyes and ears1,2. In conclusion, when compared to re-ference subjects of the same ageand sex, attractive adolescents had:• wider and shorter faces with re-

latively larger upper and midd-le facial thirds, and a reducedmandible relative to the maxilla;

• larger and more prominentlips, with a reduced nasola-bial angle;

• in boys, more prominent soft-tis-sue profile, and maxilla relativeto the mandible;

• smaller noses.

Clinical implications

Facial esthetics is one of the prin-cipal concerns of orthodontists andmaxillofacial surgeons2,11,12,17,38.The creation of a harmonic occlu-sion, within a well functioning sto-matognathic apparatus14, must al-ways consider the effect of teethposition on facial soft tissues28. Theclinician should therefore be provi-ded with esthetic guidelines refer-red to subjects of the same age,sex and ethnic group of their pa-tients; the guidelines should also

be updated, considering the evo-lution of the esthetic canons withina given society12,37. These guide-lines may offer useful indicationsfor the best kind, timing and goalsof orthodontic treatment, with thebest cost/ benefit ratio.

References

1. Bashour M. History and current con-cepts in the analysis of facial attracti-veness. Plast Reconstr Surg 2006Sep;118(3):741-56.

2. Bashour M. An objective systemfor measuring facial attractiveness.P las t Recons t r Su rg 2006Sep;118(3):757-74.

3. Johnson N, Sandy J. An aesthetic in-dex for evaluation of cleft repair. EurJ Orthod 2003 Jun;25(3):243-9.

4. Orsini MG et al. Methods to evalua-te profile preferences for the antero-posterior position of the mandible.Am J Orthod Dentofacial Orthop2006 Sep;130(3):283-91.

5. Hennessy RJ, McLearie S, KinsellaA, Waddington JL. Facial surfaceanalysis by 3D laser scanning andgeometric morphometrics in relationto sexual dimorphism in cerebral-craniofacial morphogenesis and co-gnitive function. J Anat 2005Sep;207(3):283-95.

6. Matoula S, Pancherz H. Skeletofa-cial morphology of attractive and non-attractive faces. Angle Orthod 2006Mar;76(2):204-10.

7. DeCarlo D, Metaxas D, Stone M.An anthropometric face model usingvariational techniques. Proc. SIG-GRAPH ’98, 1998;32:67-74.

8. Shi J, Samal A, Marx D. How effec-tive are landmarks and their geometryfor face recognition? Comput Vis Ima-ge Understand 2006;102:117-33.

9. Breitsprecher L et al. The influence of themuscles of facial expression on the de-velopment of the midface and the nosein cleft lip and palate patients. A re-flection of functional anatomy, facialesthetics and physiology of the nose.Ann Anat 1999 Jan;181(1):19-25.



10. Ferrario VF, Sforza C, Serrao G, Ciu-sa V, Dellavia C. Growth and agingof facial soft-tissues: a computerisedthree–dimensional mesh diagramanalysis. Clin Anat 2003Sep;16(5):420-33.

11. Vidarsdottir US, O'Higgins P, Strin-ger C. A geometric morphometricstudy of regional differences in the on-togeny of the modern human facialskeleton. J Anat 2002;201:211-29.

12. Auger TA, Turley PK. The female softtissue profile as presented in fashionmagazines during the 1900s : aphotographic analysis. Int J Adult Or-thodon Orthognath Surg1999;14(1):7-18.

13. Ferrario VF, Sforza C, Poggio CE,Tartaglia G. Facial morphometry oftelevision actresses compared withnormal women. J Oral MaxillofacSurg 1995 Sep;53(9):1008-14.

14. Isiksal E, Hazar S, Akyalcin S. Smileesthetics: perception and compari-son of treated and untreated smiles.Am J Orthod Dentofacial Orthop2006 Jan;129(1):8-16.

15. Peck S, Peck L. Selected aspects ofthe art and science of facial esthetics.Semin Orthod 1995 Jun;1(2):105-26.

16. Edler R, Agarwal P, Wertheim D,Greenhill D. The use of anthropome-tric proportion indices in the measu-rement of facial attractiveness. Eur JOrthod 2006 Jun;28(3):274-81.

17. Naini FB, Moss JP, Gill DS. The enig-ma of facial beauty: esthetics, pro-portions, deformity, and controversy.Am J Orthod Dentofacial Orthop2006 Sep;130(3):277-82.

18. Fink B, Neave N, Seydel H. Male fa-cial appearance signals physicalstrength to women. Am J Hum Biol2007 Jan-Feb;19(1):82-7.

19. Schaefer K, Fink B, Mitteroecker P,Neave N, Bookstein FL. Visualizingfacial shape regression upon 2nd to4th digit ratio and testosterone. CollAntropol 2005 Dec;29(2):415-9.

20. Kokich VO, Kokich VG, Kiyak HA.Perceptions of dental professionalsand laypersons to altered dental esthe-tics: asymmetric and symmetric situa-tions. Am J Orthod Dentofacial Or-thop 2006 Aug;130(2):141-51.

21. Shaw WC. The influence of childre-n's dentofacial appearance on theirsocial attractiveness as judged bypeers and lay adults. Am J Orthod1981 Apr;79(4):399-415.

22. Kiekens RM, Maltha JC, van't HofMA, Kuijpers-Jagtman AM. Objectivemeasures as indicators for facial esthe-tics in white adolescents. Angle Or-thod 2006 Jul;76(4):551-6.

23. Scott CR, Goonewardene MS, Mur-ray K. Influence of lips on the per-ception of malocclusion. Am J OrthodDentofacial Orthop. 2006Aug;130(2):152-62.

24. Sforza C., Laino A, D’Alessio R, Del-lavia C, Grandi G, Ferrario VF.Three-dimensional facial morpho-metry of attractive children and nor-mal children in the deciduous andearly mixed dentition. Angle Orthod2007; 77(6):1025-33.

25. Sforza C. Analisi tridimensionale noninvasiva della crescita craniofaciale indentizione decidua e mista precoce.Ortognatodonzia Italiana2006;13:53-62.

26. Ferrario VF, Sforza C, Poggio CE,Cova M, Tartaglia G. Preliminary eva-luation of an electromagnetic three-dimensional digitizer in facial anthro-pometry. Cleft Palate-Craniofac J1998 Jan;35(1):9-15.

27. Sforza C, Peretta R, Grandi G, Fer-ronato G, Ferrario VF. Soft tissue facialvolumes and shape in skeletal class IIIpatients before and after orthogna-thic surgery treatment. J Plast ReconstrAesthet Surg 2007;60:130-8.

28. Sforza C, Peretta R, Grandi G, Fer-ronato G, Ferrario VF. Three-dimen-sional facial morphometry in skeletalclass III patients. A non-invasive studyof soft-tissue changes before and afterorthognathic surgery. Br J Oral Ma-xillofac Surg 2007;45:138-44.

29. Mew J. Suggestions for forecastingand monitoring facial growth. Am JOrthod Dentofacial Orthop 1993Aug;104(2):105-20.

30. Ricketts RM. Perspectives in the clini-cal application of cephalometrics.The first fifty years. Angle Orthod1981 Apr;51(2):115-50.

31. Ramieri G, Spada MC, Nasi A, Ta-volaccini A, Berrone S. Facial anth-ropometry and aesthetic perception inyoung italian subjects: their use fororthognatic surgery (Antropometria epercezione estetica del volto in uncampione di popolazione italiana).Minerva Stomatol 2002 Dec;51(11-12):479-93.

32. Bisson M, Grobbelaar A. The esthe-tic properties of lips: a comparison ofmodels and nonmodels. Angle Or-thod 2004 Apr;74(2):162-6.

33. Ferrario VF, Sforza C, Poggio CE,Colombo A, Tartaglia G. The rela-tionship between facial 3-d mor-phometry and the perception of at-tractiveness in children. Int J AdultOr thodon Or thognath Surg1997;12(2):145-52.

34. Fink B et al. Second to fourth digit ra-tio and face shape. Proc Biol Sci2005 Oct;272(1576):1995-2001.

35. Peck S, Peck L, Kataja M. Skeletalasymmetry in esthetically pleasingfaces. Angle Or thod 1991Spring;67(1):43-8.

36. Kiekens RM, van 't Hof MA, Straat-man H, Kuijpers-Jagtman AM, Mal-tha JC. Influence of panel composi-tion on aesthetic evaluation of ado-lescent faces. Eur J Orthod 2007Feb;29(1):95-9.

37. Todd SA, Hammond P, Hutton T,Cochrane S, Cunningham S. Per-ceptions of facial aesthetics in twoand three dimensions. Eur J Orthod2005 Aug;27(4):363-9.

38. Bergman RT. Cephalometric soft tissuefacial analysis. Am J Orthod DentofacOrthop 1999 Oct;116(4): 373-89.

three-dimensional facial morphometry of attractive ... · risultati. gli adolescenti attraenti...

Documents