262

Variations in Facial Soft Tissues of Italian Individuals With EctodermalDysplasia

CHIARELLA SFORZA, M.D.CLAUDIA DELLAVIA, D.D.S.

LAURA VIZZOTTO, PH.D.VIRGILIO F. FERRARIO, M.D.

Objective: To supply quantitative information about the facial soft tissues ofpatients with ectodermal dysplasia (ED).

Design, Setting, and Patients: The three-dimensional coordinates of soft tis-sue facial landmarks were obtained using an electromagnetic digitizer in 20Caucasian patients with ED (11 male and 9 female patients aged 7 to 41 years)and 391 healthy controls (163 female and 228 male subjects of comparableages). From the landmarks, 15 facial dimensions and two angles were calcu-lated. Data were compared with those collected in healthy individuals by com-puting z-scores. Two summary anthropometric measurements for quantifyingcraniofacial variations were assessed in both the patients and reference sub-jects: the mean z-score (an index of overall facial size) and its standard devi-ation, called the craniofacial variability index (an index of facial harmony).

Results: In patients with ED, a large variability was found for both the meanz-score and the craniofacial variability index. On average, the patients had asomewhat smaller facial size than the normal individuals, with a global dis-harmonious appearance. Overall, 70% of patients fell outside the quantitativedefinitions for normal facial size, harmony or both. When examining the dis-tribution of the single z-scores of the 17 selected measurements, variable pat-terns were found, without consistent effects of age or sex.

Conclusions: The facial soft tissue structures of patients with ED differedboth in size and shape from those of normal controls of the same age, sex,and ethnic group.

KEY WORDS: ectodermal dysplasia, face, soft tissues, three-dimensional

Ectodermal dysplasia (ED) is a common name for a largeand heterogeneous group of heritable disorders characterizedby abnormal development of ectodermal tissues. In the cranio-facial complex, the teeth, hair, and sweat glands, together withthe mesoectodermal layer constituted by the neural crest, arethe most commonly involved structures, thus producing an ab-normal morphology (Hudson and Witkop, 1975; Ward andBixler, 1987; Saksena and Bixler, 1990; Pinheiro and Freire-Maia, 1994; Chitty et al., 1996; Bondarets and McDonald,2000; Hickey and Vergo, 2001; Ruhin et al., 2001; Bondaretset al., 2002; Johnson et al., 2002).

The quantitative assessment of the craniofacial characteris-tics of patients with ED is mandatory for a correct diagnosis,

Submitted March 2003; Accepted May 2003.Address correspondence to: Prof. Virgilio F. Ferrario, Dipartimento di Ana-

tomia Umana, via Mangiagalli 31, I-20133 Milano, Italy. E-mail farc@unimi.it.

treatment planning, and evaluation of results. Both the skeletaland soft tissue structures should be analyzed to provide a com-plete evaluation of any patient and should be compared withthose of healthy subjects of the same age, sex, race, and ethnicgroup (Ward and Bixler, 1987; Saksena and Bixler, 1990; Ha-jnis et al., 1994; Bondarets and McDonald, 2000; Dibbets andNolte, 2002; Johnson et al., 2002).

The skeletal structures can be assessed using radiographicanalyses. Unfortunately, they have several limitations. Theyuse ionizing radiations, thus being invasive, and they providea two-dimensional assessment of the skeletal configuration, ne-glecting most of the soft tissues and projecting all structureson a single (usually midsagittal) plane. On ethical grounds,radiographic analyses cannot be performed on healthy subjectswithout a medical indication.

In contrast, anthropometry is noninvasive and three-dimen-sional, and it considers all the facial structures, thus providinga more complete evaluation of the single patient (Ward andBixler, 1987; Farkas, 1994b; Ferrario et al., 1998; Ward et al.,2000; Farkas et al., 2001). The collection of normative datadoes not infringe any current ethical consideration.

Craniofacial data collected from syndromic patients have

Sforza et al., FACIAL SOFT TISSUES IN ECTODERMAL DYSPLASIA 263

been analyzed using two main approaches. Cephalometric andanthropometric measurements can be listed and the distance ofeach from a reference assessed, thus providing a detailed de-scription of the abnormal morphology (Ward and Bixler, 1987;Bondarets and McDonald, 2000; Farkas et al., 2001; Johnsonet al., 2002).

In addition, anatomic features can be combined and globalindices obtained (Saksena and Bixler, 1990; Ward et al., 2000).In particular, Ward et al. (2000) recently proposed two sum-mary anthropometric measurements for quantifying craniofa-cial variations: the mean z-score (an index of overall facialsize, compared with a reference population), and its standarddeviation, called the craniofacial variability index (CVI). Asillustrated by Ward et al. (2000), CVI represents the amountof variability across the analyzed z-scores: it will be low inpersons whose z-scores are all of analogous dimensions andin the same direction (increment or decrement) relative to thereference means. In contrast, CVI will be high in persons whopossess a disharmonious relationship between individual facialmeasurements. From this point of view, CVI can be consideredan index of facial harmony and an approximate assessment offacial shape. Both measurements have been found to provideuseful information in the classification of some craniofacialsyndromes (Ward et al., 2000).

According to previous cephalometric and anthropometric in-vestigations, patients with ED have a generalized reduction offacial and head dimensions as compared with normal subjects,but the reduction is nonuniform (Ward and Bixler, 1987; Sak-sena and Bixler, 1990; Bondarets and McDonald, 2000; John-son et al., 2002). This kind of modification seems worthy ofbeing investigated by the two indices introduced by Ward etal. (2000).

In the present study, the facial soft tissues of a group ofpatients with ED have been measured in three-dimensionalspace. The global facial characteristics of patients were quan-tified using the mean z-score and the CVI of selected anthro-pometric measurements (Ferrario et al., 2003). The aim of thisinvestigation was to measure the difference between patientswith ED and normal subjects of the same age and sex.

MATERIALS AND METHODS

Patients

Data from 20 patients with ED (11 boys and men, 9 girlsand women) aged 7 to 41 years were collected (Table 1). Allpatients were white northern Italians and came from the As-sociazione Nazionale Displasie Ectodermiche (ANDE, Italy).Diagnosis was provided from clinical assessments of hypohi-drosis, hypotrichosis, and hypodontia. None of the patients hadclefting, syndactyly, or other alterations.

No patient had undergone any craniofacial surgical proce-dure. The number of maxillary and mandibular teeth (bothdeciduous and permanent) actually present in the mouth is re-ported in Table 1. In half of the patients, present teeth weresymmetric in both number and type, and in the other half, left-

right differences within dental arch were always of one toothat most. Adolescent and adult patients all had some kind ofdental prosthesis, either fixed or removable, and prosthetic re-constructions were very variable in children.

Reference data were collected in previous investigationsperformed on normal subjects of the same ethnic group, age,and sex (Ferrario et al., 1999, 2001a, 2001b, 2001c). Eachreference group (for each age and sex) comprised at least 30subjects, and a total of 163 female and 228 male normal in-dividuals were considered in the current study. In the referencegroups, no subjects with a previous history of craniofacial trau-ma or congenital anomalies were included.

All the analyzed individuals, and the parents or legal guard-ians of individuals under 18 years, gave informed consent be-fore participating in the experiment.

Collection of Three-Dimensional Facial Landmarks

A detailed description of the data collection procedure canbe found elsewhere (Ferrario et al., 1998). In brief, for eachsubject, a single experienced operator located and marked 50landmarks on the cutaneous surface. Three-dimensional coor-dinates of the facial landmarks were then obtained with a com-puterized electromagnetic digitizer (3Draw, Polhemus Inc.,Colchester, VT). Data collection took less than 60 seconds.Three-dimensional (x, y, z) coordinates were recorded and an-alyzed using customized computer algorithms written by oneof the authors. The reproducibility of landmark identification,marker positioning, and data collection procedure were pre-viously reported and found to be reliable (Ferrario et al., 1998).

The following soft tissue landmarks were used in the presentstudy (Fig. 1):

• Midline landmarks: tr, trichion; n, nasion; prn, pronasale; sn,subnasale; ls, labiale superius; pg, pogonion.

• Paired landmarks (right and left side noted, r and l): exr, exl,exocanthion; zyr, zyl, zygion; tr, tl, tragion; alr, all, alare; cphr,cphl, crista philtri; chr, chl, cheilion; and gor, gol, gonion.

Landmark positions were defined according to Farkas(1994b). Midlandmarks were also mathematically derived asthe midpoint between two homologous landmarks and notedas landmarkm.

Data Analysis

The x, y, and z coordinates of the landmarks obtained oneach subject were used to calculate the following 15 lineardistances (in millimeters) and two angles (in degrees) (Ferrarioet al., 1998, 2003):

• Head: forehead height (tr-n); skull base width (t-t).• Face: face height (n-pg); lower face height (sn-pg); face

width (zy-zy); width of the mandible (go-go); upper facedepth (n-tm); midface depth (sn-tm); lower face depth (pg-tm).

• Orbits: biorbital width (ex-ex).

264 Cleft Palate–Craniofacial Journal, May 2004, Vol. 41 No. 3

TABLE 1 Analyzed Patients and Sex- and Age-Matched Reference Population

Patient

ED Patients†

Age (y)Teeth

MaxillaTeeth

MandibleMean

Z-Score CVI

Reference Population†

MeanZ-Score62 SD

CVI22 SD

CVI12 SD

Males

M02M03M01M04M05M06M07M09M08M10M11

MeanSD

7.347.377.418.029.219.39

10.6412.7013.4515.3730.9211.98

6.84

656674528885.9*1.9

814364727585.02.4

20.2620.06

0.3620.15

0.960.120.43

20.5021.1421.0820.8420.20

0.66

1.331.611.731.722.131.531.601.691.020.970.921.480.38

6 1.046 1.046 1.046 1.046 1.046 1.046 1.026 1.026 1.076 1.286 0.91

0.490.490.490.480.480.480.440.440.550.360.45

1.211.211.211.221.221.221.241.241.151.161.31

Females

F02F01F03F04F06F05F09F07F08

MeanSD

7.177.198.448.91

12.3213.5131.6936.3640.9418.5013.73

386

12‡10‡9

10788.1*2.6

245

12‡12‡106667.03.5

1.280.400.280.65

21.370.16

20.2920.41

0.110.090.74

1.341.430.780.611.121.900.920.941.921.220.47

6 0.926 0.926 1.086 1.086 1.086 0.956 1.046 1.046 1.04

0.450.450.400.400.400.550.480.480.48

1.311.311.261.261.261.211.221.221.22

TotalMeanSD

14.9210.73

6.92.4

5.93.1

20.070.69

1.360.43

† Mean z-score computed for each patient or reference population over 17 facial soft tissue measurements; ED 5 ectodermal dysplasia; CVI 5 craniofacial variability index, standard deviationof the mean z-scores computed for each patient or normal subject. The mean z-score in the reference population is always 0.

‡ Mixed dentition.Male versus female patient data: * p 5 .041 (Student’s t test for independent samples, 18 degrees of freedom).

• Nose: nose height (n-sn); length of the nasal bridge (n-prn);nose width (al-al); angle of facial convexity (n-prn-pg).

• Labio-oral: mouth width (ch-ch); width of the philtrum (cph-cph); nasolabial angle (ls-sn-prn).

As indicated by Ward et al. (2000), a mean z-score and itsstandard deviation (CVI) were computed from the 17 mea-surements as follows. Individual measurements obtained in the20 patients were transformed to z-scores by subtracting its sexand age reference mean value from it and dividing by therelevant reference standard deviation. For each patient, de-scriptive statistics of his/her 17 z-scores were computed, thusobtaining a mean z-score and its standard deviation (CVI).Individual z-scores as well as mean z-scores and CVI werealso calculated for the normal subjects. The two variables wereaveraged within each sex and age group (Ferrario et al., 2003).For each subject, computer time to perform all calculationswas about 3 minutes.

Male and female patient data were compared by two-tailStudent’s t tests for independent samples, with a level of sig-nificance set at 5% (p , 0.05).

RESULTS

All patients had some kind of hypodontia, but the patternwas very variable, ranging from a total of five teeth in childF02 to a total of 24 teeth in child F04. Indeed, the number ofteeth reported in Table 1 refers only to the erupted teeth (bothdeciduous and permanent) because no radiographic evaluationswere performed. Overall, female patients had more teeth thanmale patients, a difference significant in the maxillary arch(Student’s t test for independent samples, p , .05). Twelvepatients had more maxillary than mandibular teeth, but thereverse was found in four patients. The number of maxillaryversus mandibular teeth was not significantly different (p ..05).

Table 1 also reports the mean z-score and CVI (calculatedusing 17 measurements) of each of the analyzed patients, to-gether with the relevant descriptive statistics of the z-score andCVI of the reference population matched for sex and age (Fer-rario et al., 1999, 2001a, 2001b, 2001c). In the reference sub-jects, the average of the individual mean z-scores was zero,with standard deviations ranging from 0.456 to 0.64. By def-inition, in each age and sex group, the z-score has a mean of

Sforza et al., FACIAL SOFT TISSUES IN ECTODERMAL DYSPLASIA 265

FIGURE 1 Soft tissue facial landmarks digitized on all patients. Thelandmarks are defined in the text (from Ferrario et al., 2003).

FIGURE 2 Bivariate plot of mean z-score (x-axis, an assessment of facialsize) and craniofacial variability index (CVI; y-axis, an evaluation of facialshape and harmony) in the 20 analyzed patients. The black square rep-resents the normal individuals, with horizontal and vertical lines drawnfor reference purposes. Patients with either z-score or CVI that exceed thenormal range (mean 6 2 SD) of individuals of the same age and sex arecircled.

0 and standard deviation of 1. The standard deviation of theaverage mean z-scores was less than 1 because it was calcu-lated on a distribution of means, not individual z-scores (Wardet al., 2000).

The patients with ED had a mean z-score of 20.068 (SD0.694). That is, they had, on average, a somewhat smaller fa-cial size than the normal individuals. Indeed, a large variabilitywas found, with 10 patients scoring a positive z-score (facelarger than reference) and 10 patients scoring a negative z-score. Seven of the 11 boys and men had a negative z-score,and six of the nine girls and women had a positive z-score.Nevertheless, the mean z-score did not significantly differ be-tween sexes (Student’s t test for independent samples, p ..05). Apparently, no consistent effects of age were observed.Only three patients (M08, F02, and F06) had a mean z-scorefalling outside the normal interval (mean 6 2 SD). Child F02was larger, and the adolescents M11 and F06 were smaller thantheir age-related norms.

In the reference population, the CVI ranged between 0.76and 0.88, with standard deviations up to 0.216. The mean CVIin the patients was 1.36 (SD 0.43), and 12 patients of 20 hada CVI larger than the normal interval, thus indicating a globaldisharmonious appearance. In addition, this parameter wasvery variable among the patients, and three of them had a CVIlower than their reference mean. No significant sex differenceswere found (p . .05), nor were there consistent age effects.Overall, 70% of patients fell outside the quantitative defini-tions for normal facial size (z-score), harmony (CVI), or both(Fig. 2).

When examining the distribution of the single z-scores ofthe 17 selected measurements, variable patterns were found.For example, 11 patients had a higher forehead (trichion-na-

sion) than the reference population, with a retruded positionof landmark trichion. The lower third of the face was smallerthan the norm in 14 patients and somewhat higher in fourpatients, with z-scores up to 0.29. The soft tissue facial profilewas more prominent in 13 patients (mostly for a backwardposition of chin and a saddle nose) and less prominent in otherthree, with an altered position of pronasale. Thirteen patientshad a less prominent upper lip, and two had an opposite mor-phology. Facial widths were increased in six female subjectsand six male subjects, and decreased in two female subjectsand four male subjects. No consistent effects of age or sexcould be observed. Additionally, the number of maxillary ormandibular teeth did not seem to be related to the sign (incre-ment or decrement) or magnitude of the z-scores of facial mea-surements.

DISCUSSION

Anthropometry gives an objective aid to the qualitative ap-praisal of soft tissue anatomy of the head and face, supplyingthe clinician with useful indications about the anatomical struc-tures and regions that differ the most from the norm (Wardand Bixler, 1987; Farkas et al., 2001).

In the present study, a three-dimensional noninvasive systemallowed the quantitative analysis of the soft tissue facial char-acteristics in a group of patients with various types of ED ina wide age range. Data were compared with those collected innormal subjects of the same age, sex, and ethnic group byusing two indices, namely the mean z-score and craniofacialvariability index (Ward et al., 2000). These two indices shouldprovide a global assessment of the standardized deviations offacial size (mean z-score) and shape (CVI).

The two indices allowed for discrimination of 70% of pa-tients with ED when compared with normal subjects. Most ofthese patients were not harmonious, with CVIs larger than thenormality threshold (mean plus 2 SDs; Ward et al., 2000).

266 Cleft Palate–Craniofacial Journal, May 2004, Vol. 41 No. 3

Ward et al. (2000) found the two indices to provide somewhatmore useful information (correct classification up to 87%) inthe definition of five other craniofacial syndromes (Brach-mann-de Lange, Prader-Willi, Rubinstein-Taybi, Smith-Ma-genis, and Sotos).

As detailed in a previous study (Ferrario et al., 2003), wecould not use the same anthropometric measurements em-ployed by Ward et al. (2000) because our digital method doesnot provide head circumference or facial curvature. Alternativemeasurements were selected considering those facial charac-teristics that have been reported to differ the more in patientswith ED (Saksena and Bixler, 1990; Bondarets and McDonald,2000; Hickey and Vergo, 2001; Ruhin et al., 2001; Johnson etal., 2002). The selected measurements covered the entire face,focusing on the eyes, nose, lips, i.e. those parts of the face thatare usually considered while looking at another person (Farkas,1994a).

Previous investigations (Saksena and Bixler, 1990; Bondar-ets and McDonald, 2000; Johnson et al., 2002) detailed themain differences between the craniofacial structures of normalsubjects and patients with ED as depicted by cephalograms.Overall, the patients had a generalized reduction of facial andhead dimensions, with midface hypoplasia, a reduced lowerfacial height, a class III tendency with flat or concave facialprofile, an acute nasiolabial angle, a deeper supramental sul-cus, and protuberant and everted lips.

Anthropometric data on patients with ED were provided byWard and Bixler (1987), who found a nonuniform reductionin facial dimensions. The largest deviations from the normwere found in facial height and anterior-posterior dimensionsof facial lower two thirds, and facial widths were more similarto the reference values. In the investigation by Ward and Bixler(1987), anthropometric data from 22 months to 31 years ofage were considered as a whole, and no detailed analysis ofsex- and age-related patterns were provided.

Some of the quantitative findings of the previous studieswere confirmed in the present group of patients: reduced facialdimensions, decreased lower facial height, variable pattern infacial widths, and nonharmonious modification of facial struc-tures.

When the single facial measurements were analyzed, thepresent patients had very variable patterns, and the breakdownof single facial areas was less informative than the assessmentof the global indices proposed by Ward et al. (2000). Also,Ruhin et al. (2001) stated that patients with ED do not possessa specific anthropometric profile. The remarkable variability infacial dimensions and harmony found in patients with EDprobably corresponds to the different kinds of dysplasia, withdifferent expression of the interested genes (Pinheiro and Fre-ire-Maia, 1994; Ruhin et al., 2001).

As underlined by Saksena and Bixler (1990), single facialvariables should not be considered as isolated items becausethey interrelate with all the other structures. Indeed, the dis-criminant analysis performed on their cephalometric data cor-rectly classified 100% of the measured ED gene carriers (Sak-sena and Bixler, 1990). A somewhat poorer performance

(97%) was found for anthropometric data (Ward and Bixler,1987).

The number of subjects analyzed in the current study wasreduced, and their age range was wide, encompassing child-hood, adolescence, young adulthood, and the beginning of mi-dage. Even if comparable numbers of patients and age rangeswere reported in previous investigations (Ward and Bixler,1987; Saksena and Bixler, 1990; Ruhin et al., 2001), the pre-sent results must be considered with caution. Also, no apparentrelationship to the number of erupted teeth was found, in con-trast with Johnson et al. (2002), who reported that the numberof permanent maxillary teeth (erupted and pre-erupted) wassignificantly related to several craniofacial dimensions.

The analysis of adult patients with ED could provide quan-titative data that could be used for differential diagnosis (Sak-sena and Bixler, 1990; Ward and Bixler, 1987). Indeed, in Italy,ED has begun to be systematically recognized in childhoodonly since the mid-1990s. All the present adult patients andmost of the adolescents were recognized as ED affected onlyrecently, most often after the diagnosis of a strict relative.Therefore, quantitative analysis of craniofacial structures inadult patients with ED could also offer a practical help toclinicians.

In nontreated patients with ED, craniofacial deviations fromthe norm increased with advancing age (Johnson et al., 2002),with a tendency toward a Class III pattern with anterior growthrotation (Bondarets et al., 2002). Prosthodontic treatmentseemed to emphasize growth normalization, with a higher fa-cial balance after treatment (Ruhin et al., 2001). Indeed, im-plant-treated adult subjects were more similar to the norm thannontreated adults (Johnson et al., 2002). Unfortunately, dataon the three-dimensional arrangement of cutaneous facialstructures during growth, development, and aging of patientswith ED are still lacking.

The method used in the current investigation allowed a sim-ple, low-cost, fast, and noninvasive examination of the pa-tients, and provided a quantitative assessment of the deviationfrom the norm (Ward and Bixler, 1987). Furthermore, the useof a computerized digitizer allowed the creation of a data bankto be used for further analysis and the longitudinal comparisonof patients. From this point of view, the current investigationmay represent a development and expansion of the study byWard and Bixler (1987), made possible by the technology cur-rently available.

In conclusion, the facial soft tissue structures of child, ad-olescent, and adult patients with ED present a wide range ofdifferences. Overall, 70% of patients fell outside the quanti-tative definitions for normal facial size, harmony, or both, assuggested by two summary indices (the mean z-score and theCVI).

Only a larger group of patients with ED may allow a deeperinsight into the craniofacial characteristics of this syndrome,together with the longitudinal assessments of the patterns ofcraniofacial growth, development, and aging. The two indicesused in the present study may be a useful noninvasive andlow-cost tool for the identification of probands (e.g., relatives

Sforza et al., FACIAL SOFT TISSUES IN ECTODERMAL DYSPLASIA 267

of already recognized patients, individuals seeking dental treat-ment for hypodontia and presenting other features of ED).Moreover, the method could be used to assess the growth pat-terns and for the quantification of the effects of treatment. Dataon this matter are currently being collected in our laboratory.

Acknowledgments. The precious and indispensable collaboration of all the pa-tients and their families as well as the Associazione Nazionale Displasie Ec-todermiche (ANDE, Italy) is gratefully acknowledged.

REFERENCES

Bondarets N, Jones RM, McDonald F. Analysis of facial growth in subjectswith syndromic ectodermal dysplasia: a longitudinal analysis. Orthod Cran-iofac Res. 2002;5:71–84.

Bondarets N, McDonald F. Analysis of the vertical facial form in patients withsevere hypodontia. Am J Phys Anthropol. 2000;111:177–184.

Chitty LS, Dennis N, Baraitser M. Hidrotic ectodermal dysplasia of hair, teeth,and nails: case reports and review. J Med Genet. 1996;33:707–710.

Dibbets JMH, Nolte K. Regional size differences in four commonly used ceph-alometric atlases: the Ann Arbor, Cleveland (Bolton), London (UK), andPhiladelphia atlases compared. Orthod Craniofac Res. 2002;5:51–58.

Farkas LG. Asymmetry of the head and face. In: Farkas LG, ed. Anthropometryof the Head and Face. 2nd ed. New York: Raven Press; 1994a:103–111.

Farkas LG. Examination. In: Farkas LG, ed. Anthropometry of the Head andFace. 2nd ed. New York: Raven Press; 1994b:3–56.

Farkas LG, Katic MJ, Forrest CR, Litsas L. Surface anatomy of the face inDown’s syndrome: linear and angular measurements in the craniofacial re-gions. J Craniofac Surg. 2001;12:373–379.

Ferrario VF, Sforza C, Dellavia C, Tartaglia GM, Sozzi D, Caru A. A quanti-tative three-dimensional assessment of abnormal variations in facial soft-tissues of cleft lip and palate adult patients. Cleft Palate Craniofac J. 2003;40:544–549.

Ferrario VF, Sforza C, Ciusa V, Serrao G, Tartaglia GM. Morphometry of thenormal human ear: a cross-sectional study from adolescence to mid-adult-hood. J Craniofac Genet Dev Biol. 1999;19:226–233.

Ferrario VF, Sforza C, Ciusa V, Dellavia C, Tartaglia GM. The effect of sexand age on facial asymmetry in healthy subjects: a cross-sectional studyfrom adolescence to mid-adulthood. J Oral Maxillofac Surg. 2001a;59:382–388.

Ferrario VF, Sforza C, Colombo A, Schmitz JH, Serrao G. Morphometry of theorbital region: a soft-tissue study from adolescence to mid-adulthood. PlasticReconstr Surg. 2001b;108:285–292.

Ferrario VF, Sforza C, Poggio CE, Cova M, Tartaglia G. Preliminary evaluationof an electromagnetic three-dimensional digitizer in facial anthropometry.Cleft Palate Craniofac. J. 1998;35:9–15.

Ferrario VF, Sforza C, Schmitz JH, Ciusa V, Dellavia C. Digitized three-di-mensional analysis of normal dento-labial relationships. Prog Orthod.2001c;2:14–23.

Hajnis K, Farkas LG, Ngim RCK, Lee ST, Venkatadri G. Racial and ethnicmorphometric differences in the craniofacial complex. In: Farkas LG, ed.Anthropometry of the Head and Face. 2nd ed. New York: Raven Press;1994:201–218.

Hickey AJ, Vergo TJ Jr. Prosthetic treatments for patients with ectodermaldysplasia. J Prosthet Dent. 2001;86:364–368.

Hudson CD, Witkop CJ. Autosomal dominant hypodontia with nail dysgenesis.Oral Surg. 1975;39:409–423.

Johnson ER, Roberts MW, Guckes AD, Bailey LJ, Phillips CL, Wright JT.Analysis of craniofacial development in children with hypohidrotic ecto-dermal dysplasia. Am J Med Genet. 2002;112:327–334.

Pinheiro M, Freire-Maia N. Ectodermal dysplasias: a clinical classification anda causal review. Am J Med Genet. 1994;53:153–162.

Ruhin B, Martinot V, Lafforgue P, Catteau B, Manouvrier-Hanu S, Ferri J. Pureectodermal dysplasia: retrospective study of 16 cases and literature review.Cleft Palate Craniofac J. 2001;38:504–518.

Saksena SS, Bixler D. Facial morphometrics in the identification of gene car-riers of X-linked hypohidrotic ectodermal dysplasia. Am J Med Genet. 1990;35:105–114.

Ward RE, Bixler D. Anthropometric analysis of the face in hypohidrotic ecto-dermal dysplasia: a family study. Am J Phys Anthropol. 1987;74:453–458.

Ward RE, Jamison PL, Allanson JE. Quantitative approach to identifying ab-normal variation in the human face exemplified by a study of 278 individualwith five craniofacial syndromes. Am J Med Genet. 2000;91:8–17.