Richard E. Ward
Elizabeth S. Moore
James K. Hartsfield Jr.
The relationship between cleft lip and palate (CLP) and facial form is widely recognized, frequently studied, and poorly understood. Thus, after nearly a century of research, it remains uncertain which aspects of facial form are related to the primary underlying cause of the cleft and which aspects are secondary to the effects of the cleft or of its repair. Clearly, complete CLP will distort normal facial growth but so may its later surgical repair. It is also suspected that certain facial shapes are more predisposed to developing CLP than others.
To address these issues, facial form has been evaluated in three distinct groups of subjects: untreated individuals with oral facial clefts, affected individuals with repaired clefts, and unaffected relatives of individuals with clefts. An astonishing number of studies over the last 50 years have examined one or more of these populations and have generated considerable, frequently conflicting information. Variations in methodology, sample composition, and the degree to which the studies have controlled for differences in cleft type and etiology have made it difficult to generalize from this information. We make no pretense of comprehensively reviewing every one of the several hundred articles written on these topics. Instead, in this chapter, we summarize a select group of key publications and extract from them common themes that represent what is known at the start of a new century of research.
More specifically, we review the morphometric methods that have been used to assess facial form in clefts, summarize what is known about facial form in untreated persons as well as individuals who have had their clefts repaired, and finally discuss the current understanding of facial form in unaffected relatives of individuals with clefts. There can be little doubt that we stand on the threshold of a new era of understanding as emerging technologies and a revolution in our knowledge of the interplay between genes and the “environment of development” promise to resolve many of the questions that remain unanswered. Nevertheless, this progress must grow from a foundation established by the work of dentists, physicians, anatomists, anthropologists, and other “morphometricians” who have attempted to penetrate the complexities of this most frequently encountered form of craniofacial anomaly.
Morphometrics
Morphometrics describes a variety of methods used to measure and assess form. Since its introduction by Broadbent in 1931, the two-dimensional lateral cephalometric radiograph has been the morphometric medium of choice for assessing cranial facial form. Less often, facial measurements have been taken directly (anthropometries) (Farkas and Lindsay, 1971,1972a,b; Šmahel and Brejcha, 1983; Šmahel, 1984a,b,d; Farkas et al., 1993) or derived from photographs (photogrammetrics) (Vegter et al., 1997). More recently, computerized tomography and other forms of three-dimensional radiography (Baumrind et al., 1983; Cheverud et al., 1983; Grayson et al., 1983,1985; Savara et al., 1985; Ras et al., 1994a,b) have augmented these methods.
Data derived from any of these methods must be adjusted mathematically so that individuals of different ages, sizes, and sexes can be compared. To create equations that control for size, age, and sex differences, investigators convert original data to standard deviation units (z-scores) or perform regression analysis. However, such adjustments are seldom completely effective. For example, regression equations are sensitive to the variation within the samples from which they are generated, and z-scores must be calculated against some normative or control population that may or may not be well matched to the study population. Therefore, even adjusted data are likely to retain some variation due to age, sex, and size, an often overlooked fact that can confound comparisons of results from different studies.
There is also considerable debate over the efficacy of morphometric techniques derived from fixed but often arbitrarily defined landmarks. For example, Moyers and Bookstein (1979) noted that traditional cephalometrics provides limited or even misleading information regarding the true shape and size of craniofacial structures. Compressing a complex three-dimensional structure into two dimensions distorts form, occludes anatomy, and creates artifacts. Furthermore, two-dimensional measurements cannot reveal differences clearly visible in three dimensions. Analyzing lines and angles to measure the shape and size of such a distorted image compounds the problem. Moyers and Bookstein (1979) noted that the standard measure of mandibular length (gonion to menton) is likely to be highly inaccurate because it uses a straight line to measure a complex, curved surface. Finally, they argue that classic serial cephalometric studies are inadequate because there is little relationship between the superimposed points moving in apparent succession and the actual geometry of growth.
In response to these challenges, new technologies and analytical techniques have been proposed to extract more accurate information from radiographic images. Geometric analytical techniques originally proposed by Thompson (1917) have come of age with the development of computer-assisted programs that are able to carry out the necessary complex equations (Bookstein, 1982; Grayson et al., 1987; Trotman and Ross, 1993; Lestrel et al., 1999). These techniques use geometric forms to assess change in shape over time or to compare form between individuals.
Despite the availability of new technologies and analytical techniques most studies of facial form in clefting continue to rely on traditional cephalometric approaches. It is likely that this reflects less a resistance to innovation and precision than the reality that most investigators do not have ready access to the technology, software, or training necessary to carry out the more sophisticated analyses. Moreover, while the more advanced imaging techniques generate stunningly detailed images, they cost more, increase radiation exposure, and generally lack adequate control or reference data. In addition, many investigators find that traditional analyses continue to yield useful results. Nevertheless, there appears to be little reason to doubt that our understanding of oral facial clefting will improve as we find more effective means to employ these more refined approaches and to couple them with the exploding technologies in developmental and molecular genetics (McAlarney and Chiu, 1997; Mossey et al., 1997; Peltonen and McKusick, 2001).
Facial form in Individuals with Oral Facial Clefts
As noted above, inconsistencies in study design, including method of evaluation (judgment vs. measurement), demographics of the patient population, selection of cephalometric or anthropometric landmarks, and methods of statistical analysis, make it difficult to compare results. In addition, as Bishara et al. (1985) and Gaggl et al. (1999) have pointed out, many studies have limitations such as small sample size; inconsistent age distribution; unspecified combinations of untreated, partially treated, and late treated individuals; as well as a mixture of individuals with different types of cleft. Despite the difficulties, we summarize in the following sections relevant morphometric literature and distill common findings regarding the craniofacial phenotype of the various forms of cleft both pre-and posttreatment. The discussion on cleft lip with or without cleft palate (CL/P) and cleft palate (CP) is separated because these are best regarded as etiologically and developmentally distinct anomalies. The review of the CL/P phenotype is divided into sections based on degree of facial disruption, beginning with the most severe (untreated bilateral complete CLP) and proceeding through the least severe (repaired CL without cleft alveolus). Where possible, facial form in untreated individuals is contrasted to that in individuals whose clefts have been repaired. The often confusing use of anatomical landmarks as descriptors has been avoided. Instead, we discuss anatomical areas and what individual measurements reveal about these structures.
Cleft Lip with or without Cleft Palate
Untreated Bilateral Cleft Lip and Palate
The most dramatic effects on facial structures occur in individuals with untreated complete bilateral CLP. Most structural abnormalities are obvious at birth (Huddart, 1970), and some, unless treated, worsen with age. In addition to the anterior facial areas directly affected by the cleft, adjacent and posterior areas (including the neurocranium and cranial base) often deviate from normal morphology.
Neurocranium and cranial base
Smaller craniofacial dimensions have been found in many different cleft types (Dahl, 1970; Ross and Johnston, 1972; Semb, 1991a,b; da Silva Filho et al., 1992; Capelozza Filho et al., 1993), including untreated bilateral cleft patients (da Silva Filho, 1998). In the latter study (da Silva Filho, 1998), there were generally no differences in the cranial base angle compared to controls, in spite of the finding that cranial base linear (sagittal) dimensions were smaller in the untreated CLP group. However, Šmahel et al. (1985) reported that maximum head width was significantly increased in untreated bilateral CLP patients relative to controls (without a cleft).
da Silva Filho et al. (1998) studied a sample of adult (Caucasian) Brazilians with untreated, complete bilateral CLP (20 males and 8 females) and compared these to a matched control, noncleft population. There was a general reduction in cranial and facial size in the untreated group with the means for most linear dimensions significantly reduced compared to controls.
Facial widths
da Silva Filho et al. (1998) reported significantly greater cranial and facial widths, including maximum head width, and bizygomatic, interorbital, and nasal widths. There was a significant sex effect, with males showing greater average mandibular retrognathia and females showing greater average facial convexity. Farkas et al. (1993) found that most untreated bilateral CLP patients had borderline-large to abnormally large facial widths and all had abnormally large nasal widths.
Craniofacial heights and profiles
An extremely prominent premaxilla combined with a smaller mandible results in marked facial convexity and an extreme imbalance between the jaws (Šmahel et al., 1985; da Silva Filho et al., 1998). da Silva Filho et al. (1998) also found significantly greater facial convexity in the untreated group. The bilateral group had a combination of a smaller nasomaxillary-pharyngeal complex and reduced upper posterior facial height (Brader, 1957; Levin, 1963; Dahl, 1970).
Maxillary, palatal, and mandibular structures
In infancy prior to surgery, the overall maxillary size tends to be larger and the maxilla retrognathic (Huddart, 1970). Infants with unrepaired bilateral CLP also appear to have a surprising degree of asymmetry in left and right maxillary arch size (Huddart, 1970). In addition, the premaxilla is rotated and protrusive (Huddart, 1970). The projection of the premaxilla continues if left untreated (Handelman and Pruzansky, 1968; Friede and Pruzansky, 1972; Bishara et al., 1978; da Silva Filho et al., 1998), contributing to the extreme premaxillary prominence that is the most striking feature in adult untreated bilateral CLP patients (da Silva Filho et al., 1998).
The mandible tends to be small and retruded in unrepaired bilateral CLP individuals (Bishara et al., 1985; Dahl et al., 1989; da Silva et al., 1998), a characteristic found in untreated adult patients with other types of cleft (Mars and Houston, 1990; da Silva Filho et al., 1992, 1993). In addition, there is a tendency for the mandibular plane to be steep (Bishara et al., 1985). Most studies have also found that untreated individuals with bilateral CLP have significantly more obtuse gonial angles (Brader, 1957; Levin, 1963; Dahl, 1970; Horowitz et al., 1976,1980; da Silva Filho et al., 1998).
Dental arches and relationships
Incisors in both jaws tend to be retroclined (da Silva Filho et al., 1998), and the two maxillary segments rotate medially, causing various degrees of arch collapse. Decreased maxillary intercanine width and increased incidence of crossbite in the canine region are found (Bishara et al., 1985). Patients with unoperated CLP tend to have greater maxillary alveolar widths than operated cleft subjects (Motohashi et al., 1994; Trotman et al., 1997).
Untreated Unilateral Cleft Lip and Palate
Individuals with unrepaired unilateral CLP exhibit many of the same features seen in bilateral CLP, such as a tendency for wide nasal cavities (Atherton, 1967a; Farkas et al., 1993), midface retrusion, acute nasaolabial angle, and relatively excessive lower facial height (Dahl, 1970; Bishara et al., 1985, 1986; Šmahel et al., 1985; Mars and Houston, 1990). In addition, individuals with unrepaired unilateral CLP tend to have significantly more obtuse gonial angles as well as a smaller nasomaxillary-pharyngeal complex and reduced upper posterior facial height, as reported in bilateral cases (Brader, 1957; Levin, 1963; Dahl, 1970; Horowitz et al., 1976, 1980).
In general, the cranial base of subjects with unrepaired unilateral CLP is similar to that of controls with the exception of reduced cranial base length (Bishara et al., 1976; Mars and Houston, 1990; Capelozza Filho et al., 1993). In individuals with unilateral CLP, the unoperated maxilla usually maintains a normal relationship to the cranial base, while the premaxilla tends to rotate forward on the noncleft side. In contrast, the mandible is generally rotated backward (retruded), which some researchers have suggested is secondary to the biomechanical effect of the cleft on the mandible (Atherton, 1967b; Dahl, 1970; Chierici et al., 1973; Bishara et al., 1985; Mars and Houston, 1990; Capelozza Filho et al, 1993). Capelozza Filho et al. (1993) also reported a smaller mandibular body and ramus height in subjects with untreated unilateral CLP. Several researchers have concluded that these subjects also had significant increases in the angle of the mandibular plane (Ortiz-Monasterio et al., 1966; Bishara et al., 1976; Mars and Houston, 1990; Capelozza Filho et al., 1993).
Ortiz-Monasterio et al. (1959) and Mestre et al. (1960) reported that the size of the maxilla in patients with unoperated clefts was within normal limits, while Capelozza Filho et al. (1993) found that the maxilla was smaller and protruded at the alveolar level. Capelozza Filho et al. (1993) employed posterior-anterior cephalograms in their study, while other researchers used lateral cephalograms. This may account for the variability in results. In the maxillary arch, the relationship of the cleft segment to the noncleft segment varied from normal to degrees of medial collapse, particularly in the canine area. This resulted in an increased incidence of crossbite (Atherton, 1967a; Bishara et al., 1976,1985). There was also an increased incidence of buccal overjet of the maxillary premolars (Bishara et al., 1985). The side of the cleft was not found to cause significant differences in craniofacial morphology in either pre-or postsurgery individuals (Jain and Krogman, 1983a; Hermann et al., 1999, 2000). However, there appeared to be a greater incidence of lingual crossbite on the cleft side compared to individuals with unilateral CL and alveolus but not CP (Bishara et al., 1985). Tang and So (1992) noted that the majority of individuals with unilateral CLP had severe malocclusion at an early age. In two samples of unoperated adults with unilateral CLP, the upper incisors were found to be similar to those of noncleft subjects in one study (Mars and Houston, 1990) and proclined in another (Capelozza Filho et al., 1993).
Repaired Bilateral Cleft Lip and Palate Compared to Repaired Unilateral Cleft Lip and Palate and Noncleft Individuals
Most morphometric studies of bilateral CLP have reported that with treatment the severe facial dysmorphology was greatly improved, although some residual effects remained (Trotman and Ross, 1993). However, several cephalometric studies have reported only minimal differences between subjects who had surgically repaired bilateral clefts and their nonrepaired counterparts (Dahl, 1970; Narula and Ross, 1970; Hanada and Krogman, 1975; Wepner and Hollman, 1975). In general, the facial differences found in the treated bilateral CLP group were similar to those found in treated unilateral CLP (Dahl, 1970; Ishiguro et al., 1976; Šmahel and Brejcha, 1983; Farkas et al., 1993).
Neurocranium, and cranial base
Šmahel (1984b) compared lateral cephalograms of surgically repaired adult males with complete bilateral CLP to unaffected controls as well as surgically repaired adult males with unilateral CLP. In the bilateral CLP group, the height of the neurocranium was reduced, the supraorbital frontal width was increased, and the posterior cranial base was significantly shorter compared to noncleft controls. In comparison to subjects with unilateral CLP, the cranial base was slightly but significantly flatter and the clivus significantly shorter. In a related study on the same group of treated adult males using anthropometries, Šmahel (1984c) noted that head circumference, head length, head width, and bizygomatic and bitragal (cranial base) widths were not significantly different compared to controls.
There has been less agreement regarding the cranial base in studies of repaired CLP. Aduss (1971) reported that there were no significant differences in the cranial base of patients with unilateral CLP compared with noncleft patients, while Krogman et al. (1975), in a longitudinal study of infants and children, found that all linear cranial base measurements were significantly longer and the cranial base angle was smaller in the unilateral CLP group compared to a noncleft control group. In contrast to both of these studies, Hayashi et al. (1976) reported no significant differences in the length of the anterior cranial base but noted that the cranial base angle was more obtuse in their unilateral CLP group compared to the noncleft group. Horswell and Gallup (1992) found that the age at which the linear dimensions became significantly reduced varied but that, in general, both anterior and posterior dimensions became significantly reduced between ages 8 and 9 (and remained so thereafter). No significant differences were found in the cranial base angle of either the unilateral or bilateral CLP individuals. In contrast, Semb (199la) noted that cranial base angulation was greater in unilateral CLP subjects when compared to noncleft controls but also found a sex effect, with the angle being significantly greater for affected females compared to affected males. Failure to consider variations due to age and sex may help to explain the lack of agreement in previous studies, as may the use of inappropriate control samples and other variations in methodology (Horswell and Gallup, 1992).
Facial widths
As was the case for untreated individuals, most studies of treated subjects with CLP have described excessive widths across the midface. For example, Šmahel and Brejcha (1983) and Šmahel (1984b)reported a widening of the nasal cavity and the interocular distance in both unilateral and bilateral subjects. In a longitudinal study comparing three different cleft types (bilateral CLP, unilateral CLP, and CP), Ishiguro et al. (1976) found that, in general, the repaired bilateral CLP group had significantly greater facial breadths than the other two cleft groups or the noncleft group. Athanasiou et al. (1990) found that individuals with repaired bilateral CLP also exhibited increases in nasal and intergonial widths compared to noncleft controls. However, Farkas et al. (1993) suggested that this might not be universal among affected individuals. Using direct anthropometry, they reported that 71% of patients with repaired unilateral CLP had normal soft tissue nasal width, while only 48% of repaired patients with bilateral CLP had normal soft tissue nasal width. Surprisingly, the mean values for these variables did not differ significantly between the two groups. They also noted that individuals with surgically repaired CLP tended to have normal or, in some cases, abnormally narrow bizygomatic width. Part of the difference between the outcomes of the radiographic and direct methods may be that the direct method measures the soft tissue, which may not absolutely correlate with the underlying skeletal structures. On the other hand, an uncritical reliance on group mean values can obscure meaningful differences between study populations.
Craniofacial heights and profiles
Both bilateral and unilateral surgically repaired groups had a combination of a smaller nasomaxillary-pharyngeal complex and reduced upper posterior facial height (Brader, 1957; Levin, 1963; Dahl, 1970; Friede and Johanson, 1974, 1977; Horowitz et al., 1976, 1980; Semb, 1991a,b). In contrast, both groups tended to exhibit relatively excessive anterior lower facial height (Dahl, 1970; Farkas and Lindsay, 1971; Krogman et al., 1975; Ross, 1987; Enemark et al., 1990; Semb, 1991a,1991b). Krogman et al. (1975) reported that both upper and lower anterior facial heights were increased in unilateral CLP children compared to controls, while Hayashi and colleagues (1976) found that the upper anterior facial height was reduced and the lower anterior facial height was increased, resulting in normal total facial height. Šmahel and Brejcha (1983) supported these findings. In comparison to individuals with unilateral CLP, individuals with bilateral CLP tended to have an increased length of the face as a whole but normal upper facial measurements (Šmahel, 1984b). A midfacial deficit was reported in the sagittal plane in patients with repaired bilateral CLP, as was a moderate reduction in height of the midface in favor of the lower face (Gaggl et al., 1999).
In contrasting repaired unilateral and bilateral CLP groups, the relatively greater increase in facial height in the bilateral cleft group was associated with an even more noticeable posterior growth rotation of the face, retroclination of the palatal plane, and greater retrusion of the mandible. This was reflected by the significant difference between the angulation of the midface relative to the cranial base in these two series (Šmahel, 1984b). The configuration of the soft tissue profile tended to correspond to its skeletal framework (Šmahel and Brejcha, 1983). Again, however, Farkas and colleagues (1993)demonstrated considerable individual variability in the expression of facial height deviations. They reported that in both unilateral and bilateral CLP groups the anthropometric height of the upper face was normal in two-thirds of individuals, in spite of the differences in mean values for facial height between the CLP and control groups.
Maxillary, palatal, and mandibular structures
In patients with repaired complete bilateral CLP, Narula and Ross (1970) found severe protrusion of the premaxilla at 6 years of age that reduced to almost normal by 16 years of age. They also found superior and posterior positioning of the lateral maxillary segments, although the length of the segments was normal. Similarly, Šmahel and Brejcha, (1983) found that in repaired bilateral cases there was displacement of the premaxilla forward and more marked displacement of the maxilla backward. This caused retrusion of the upper face similar to that found in cases with complete unilateral CLP, although the retrusion of lateral dentoalveolar segments and maxilla was even greater than in individuals with unilateral clefts. In addition, there was posterior displacement of the zygomatic bones and orbits in the repaired bilateral cases that reflected the more posterior positioning of the maxilla (Šmahel and Brejcha 1983; Schultes et al., 2000). Midfacial growth deficiencies and posterior displacement of the upper jaw have been reported in both unilateral and bilateral groups (Ross, 1987; Mars and Houston, 1990; Chen and So, 1997). Several studies have found that the palatal plane was rotated in a clockwise direction in treated CLP subjects (Dahl, 1970; Friede and Johanson, 1974, 1977; Horowitz et al., 1980).
Trotman and Ross (1993) concluded that the greatest effects of growth and treatment on the face of individuals with CLP were in the area of the maxilla. This region was consistently smaller and all its landmarks were located closer to the cranial base, implying a less developed midface and oropharynx. As noted, the premaxilla began in a grossly protruded position in the treated bilateral cases but gradually resolved until the relative protrusion was decreased in the adult. The nasal bones were longer and more protruded relative to the cranial base, and the posterior segments of the maxilla were hypoplastic, as evidenced by their posterior displacement and deficiency (Trotman and Ross, 1993). Gaggl et al. (1999) used cephalometric and model analyses to assess the long-term growth effects of surgical repair and orthodontic treatment in adults with complete bilateral CLP. Compared with standard values, all patients in their study retained a small extent of maxillary retrognathia and retroposition of the midface with a retrognathic type of face.
Schultes et al. (2000) used the same methods to assess the surgical and orthodontic treatment of adults with unilateral CLP or isolated CP. They found that, similar to the bilateral CLP group, the unilateral group had maxillary growth disturbances. On average, the unilateral CLP group had no growth disturbances in the higher midface region.
The maxillary deficiencies associated with surgical repair of CLP are associated with skeletal discrepancies between the upper and lower jaws (Pruzansky and Aduss, 1967; Ross and Johnston, 1972; Nordén et al., 1973; Bergland and Sidhu, 1974; Dahl and Hanusardottir, 1979; Šmahel and Müllerová, 1986). However, Friede and Pruzansky (1972) suggested that it was the palatoplasty and not the lip repair that most affected midfacial development. In their patients with bilateral CLP who had surgical closure of the lip without a premaxillary setback, facial profile measurements in early adolescence approximated those of the averages of noncleft individuals. Other morphometric studies have suggested that surgically induced scar tissue was a major contributing factor to the facial growth problems seen in patients with clefts (Graber 1949, 1954; Ortiz-Monasterio et al., 1959, 1966; Mestre et al., 1960; Glass, 1970; Narula and Ross 1970, Boo-Chai, 1971; Friede and Pruzansky, 1972).
Trotman and Ross (1993), in their study of individuals with repaired bilateral CLP, suggested that, compared to controls, the main shape differences in the mandible were concentrated at the gonial angle. They found, as had many others, that patients with treated bilateral CLP tended to have a significantly more obtuse gonial angle than noncleft controls (Narula and Ross, 1970; Horowitz et al., 1980; Šmahel, 1984b). Trotman and Ross (1993) also found that the mandibular length was smaller in surgically repaired patients at 6 years of age but not different in adults. In contrast, Narula and Ross (1970) found normal mandibular length in children with bilateral CLP, and Šmahel (1984b) reported that the mandibular length in an adult male study population was significantly shorter than in noncleft controls. Šmahel and Brejcha (1983) found that a higher degree of posterior growth rotation of the mandible was accompanied by an even more marked compensatory increase of the anterior height compared to individuals with unilateral clefts. They also found that subjects with unilateral CLP exhibited deficient mandibular growth, which was associated with changes in its shape, specifically a decrease in gonial angle and shortening of the mandibular body. Finally, several studies have reported that individuals with repaired complete unilateral CLP tend to have persistent facial asymmetries and nasal deformities (Harvold, 1954; Aduss and Pruzansky, 1967; Dahl, 1970; Mars and Houston, 1990; Molsted and Dahl, 1990; Molsted et al., 1992; Sandham and Murray, 1993; Trotman et al., 1993; Motohashi et al., 1994; Ras et al., 1994b; Kyrkanides et al., 1995).
Dental arches and relationships
Handelman and Pruzansky (1968) noted the presence of a significant over jet by the age of 4 in individuals with complete repaired bilateral CLP; however, Ross and Johnston (1972) found that significant overjet was rare but that 38% of their CLP sample exhibited incisor crossbite. According to Trotman and Ross (1993), the differences reported in the aforementioned studies were likely the result of differences in the surgical management of the two samples. Other differences in dental and arch relationships between individuals with repaired bilateral and unilateral clefts included a more pronounced retroclination of the upper incisors and of the alveolar process in the bilateral group (Šmahel and Brejcha, 1983; Trotman and Ross, 1993). In contrast, Šmahel and Brejcha (1983) noted a tendency toward overeruption of the posterior teeth in individuals with repaired unilateral CLP. They suggested that this could occur as compensatory growth due to the vertical posterior maxillary hypoplasia.
The occlusal plane in surgically repaired CLP appeared not to differ significantly from that of controls (Narula and Ross, 1970; Trotman and Ross, 1993). However, Gaggl et al. (1999) employed model analysis to show that for most patients with repaired CLP either the maxillary or the mandibular arch was too wide and that all had transverse space deficits as well as a reduction in sagittal measurements, even after termination of orthodontic treatment. Interestingly, patients with bilateral CLP tended to possess a betterformed anterior dental arch compared to patients with unilateral CLP or CP. According to Gaggl et al. (1999), sagittal and transverse space deficits in bilateral CLP patients were consistent with the earlier observations reported by Steinhauser and Rudzki-Janson (1994). A sagittal space deficit was observed in half of the bilateral CLP group studied by Gaggl and colleagues (1999), but positional abnormalities of the upper incisors were rare. The majority of patients with bilateral CLP also had alveolar midline displacement of the maxilla and mandible (Gaggl et al., 1999).
In a similar study of individuals with repaired unilateral CLP, Schultes et al. (2000) noted that an even higher percentage of patients had a negative sagittal space available compared to individuals with bilateral CLP (88% vs. 50%, respectively). In most cases with unilateral CLP, there was a reduction in sagittal length, implying a sagittal space deficit. This may reflect the reduction of maxillary growth in an anteroposterior direction that has been reported in other CLP types (Bishara et al., 1979; da Silva Filho et al., 1998; Mars and Houston, 1990). Schultes et al. (2000) also found a unilateral persistent transverse space deficit in the premolar and molar regions (also reported by Bishara et al., 1985).
Cleft Lip with or without Cleft Alveolus
Most researchers agree that the craniofacial morphology of individuals with cleft lip with or without cleft alveolus (CL/A) is very similar to that found in noncleft controls (Dahl, 1970; Nakamura et al., 1972; Hirschfeld and Aduss, 1974; Cronin and Hunter, 1980; Šmahel, 1984d; Friede et al., 1986; Horswell and Gallup, 1992). However, Šmahel et al. (1985) noted several small but significant differences between individuals with unrepaired CL/A and normal controls. Specifically, head width, bizygomatic width, interocular width, nasal width, mouth width, total facial height and upper facial height, were significantly increased in the untreated CL group (although the degree of difference was generally less than that seen in CL/P). Several studies have agreed that the effects in this group seemed to center around the cleft area (Innes, 1962; Dahl, 1970; Bishara et al., 1976, 1985), although this does not account for the greater interocular and cranial widths often reported in this group.
Morphometric analysis of the craniofacial complex of individuals with CL/A after surgery showed little residual effect from lip and alveolar repair and suggested that the skeletal craniofacial relationship of CL/A patients posttreatment is similar to that in noncleft faces (Dahl, 1970; Bishara et al., 1985). However, only a few investigators have examined the craniofacial morphology of individuals with CLA (but not palate) independently from individuals with CL only (Bishara et al., 1976, 1985; Friede et al., 1986). Friede et al. (1986) argued that there were important distinctions between these two groups that accounted for some of the variability in results reported in the literature and suggested further that using CL/A individuals as normal controls (Dahl, 1970; Dahl et al., 1982; Nakamura et al., 1972) would be imprudent given the small but notable differences consistently described in these populations. In the following sections, the term CL/A is used only for those studies where it is clear that such distinctions have been noted.
Cranial base and facial widths
Dahl et al. (1982) found that adults with unrepaired CL had a (lateral) widening of the cranial base and increased interocular distance. This finding has been documented by numerous investigators in both pre- and postoperative CL/A patients (Ross and Coupe, 1965; Dahl, 1970; Aduss, 1971; Farkas and Lindsay, 1972a; Nakamura et al., 1972; Hirschfeld and Aduss, 1974; Šmahel, 1984d; Friede et al., 1986; Šmahel et al., 1985).
Craniofacial heights and profiles
In addition to the increase in anterior facial heights previously reported by Šmahel et al. (1985), Casal et al. (1997) reported an increase in facial convexity in treated CL/A, which was secondary to mandible retrognathism, and a decrease in mandibular body length together with a degree of mandibular rotation. Cronin and Hunter (1980) also noted a slight clockwise rotation of the mandible in individuals with repaired CL.
Maxillary, palatal, and mandibular structures
In their study of individuals with repaired CL/A, Nakamura et al. (1972) found a greater transverse maxillary dimension. The following features were also reported: decreased inclination of the nasal bones (Dahl, 1970; Nakamura et al., 1972), more superiorly placed anterior nasal spine (Ross and Coupe, 1965; Dahl, 1970; Nakamura et al., 1972), and increased length of the maxilla and mandible (Dahl et al., 1982). However, there has been some disagreement among investigators in regard to maxillary width. Dahl (1970) found a decreased maxillary width, while Friede et al. (1986) reported that the maxillary width of CL patients closely approximated noncleft values. The latter study also did not find the clockwise rotation of the mandible described by Cronin and Hunter (1980). It is possible that the differences were secondary to age discrepancies, treatment differences, and differential inclusion of alveolar clefts. Thus, Friede et al. (1986) employed a mixed longitudinal study of CL patients (excluding CLA patients) from infancy to 6 years of age, while most of the other studies were conducted on adult CL (possibly including CLA), in which most had undergone some type of orthodontic treatment.
Casal et al. (1997) found that in a small sample (n = 6) of young children with repaired isolated CL, mandibular body length was significantly shorter and mandibular position significantly more retruded than in the control population, leading to greater convexity. These findings were similar to those reported for CLP, but other investigators did not find these features in their generally older populations (Dahl et al., 1982). Thus, Casal et al. (1997) may have described a transient manifestation of CL that disappeared as the more normal growth trajectory of this group occurred or as orthodontic treatment proceeded.
Dental arches and relationships
Individuals with unrepaired CLA tended to have an increased overjet and canines that were edge-to-edge and sometimes in crossbite. The teeth on either side of the cleft tended to roll superiorly, resulting in infraocclusion, with a localized open bite tendency. Bishara et al. (1985) used lateral cephalometric analysis to compare unoperated CLA individuals to unoperated CLP individuals. They found that the mandibular incisors were significantly more lingually inclined in the CLP group and more labially inclined in the CLA group. This appeared to be different from what was seen in individuals with repaired CL alone. Casal et al. (1997) reported that such children had lingual inclination of the upper and lower teeth, which resulted in widening of the interincisal angle and a reduction in the incisor overjet prior to orthodontic treatment.
Isolated Cleft Palate
Debate persists about the degree to which isolated CP and/or its surgical repair affect facial form. Some researchers have reported significant craniofacial differences in children with unrepaired CP compared to controls (Šmahel et al., 1987) and to children with unilateral CL (Dahl et al., 1982). Significant differences were described in adult females with CP (both pre- and postpalatoplasty combined) compared to a noncleft control sample (Bishara, 1973a; Bishara and Iverson, 1974). However, Mestre et al. (1960) found no significant differences in facial morphology in adult Puerto Ricans with unoperated CP compared to a noncleft group. Bishara (1973a) suggested that the disagreement in results might be due to differences in the populations studied and to differences in the methods of study. Bishara (1973b) also found that sex and the extent of the cleft were not correlated with significant changes in the craniofacial complex. However, Dahl (1970) found a “systematic tendency for the morphological changes to be most marked in extensive clefts.”
In regard to the effects of surgical repair, several investigators have found that the craniofacial morphology of patients with surgically repaired CP differed significantly from that of noncleft patients (Shibaski and Ross, 1969; Dahl, 1970; Farkas and Lindsay, 1972b; Bishara, 1973a; Bishara and Iverson, 1974; Krogman et al., 1975). However, Bishara (1973a,b) and Bishara and Iverson (1974) found no significant differences in the craniofacial morphology of unrepaired and repaired (but obturated) CP adult females, which suggested that the morphometric features were not the result of surgical repair. However, Šmahel et al. (1987) argued that surgical repair did affect later growth, particularly in the restraint of further widening of the palate and surrounding structures. They also suggested that the relative maxillary shortening was progressive after palatoplasty. Finally, some deviations which occurred in adults, including changes in mandibular shape, were mostly insignificant before palatal surgery, as documented by Dahl et al. (1982) as well as Šmahel (1987). The phenotypic differences between individuals with untreated CP and individuals with untreated CLP reinforce the point that CP differs both etiologically and morphologically and, in general, is believed to result in less overall facial abnormality. Nevertheless, many phenotypic similarities between individuals with isolated CP and other forms of oral facial cleft occur.
Neurocranium and cranial base
Dahl et al. (1982) used lateral, anteroposterior, and axial cephalometry to demonstrate that infants with untreated CP compared to untreated CL infants (used as a “normal” control) had shorter anterior cranial bases. Šmahel et al. (1987) also found evidence for shortening of the cranial base in children before their palates were repaired.
Facial widths
Widening of the nasal cavity and of neighboring structures in all types of cleft with involvement of the palate prior to palatoplasty was demonstrated early by Subtelny (1955) and Coupe and Subtelny (1960). The finding of widening of the nasal cavity was also in agreement with the observations of Dahl et al. (1982). However, Šmahel et al. (1987) could not demonstrate a significant difference in interocular dimensions in their unrepaired CP subjects compared to controls. Šmahel (1984a) also reported that the increased nasal cavity width almost disappeared in adulthood, probably as a result of growth modifications caused by the repair. Farkas and Lindsay (1972b) used direct anthropometry to assess facial morphology in adult males and females with CP who underwent palate surgery in childhood. They found that both male and female CP patients had on average a significantly narrower bizygomatic diameter than noncleft age- and sex-matched controls. The soft tissue width of the nose was also significantly smaller in adults with repaired CP. Similarly, Jain and Krogman (1983b) could not demonstrate an increase in facial widths in their treated CP subjects, at least compared to their CLP groups. These findings contrast with those for CLP (and in some cases CL), where increased orbital and nasal widths were commonly reported in both untreated and treated subjects.
Craniofacial height and profile
Šmahel et al. (1987) found a reduction of the posterior height of the upper face in pretreated children but not in adults with surgically repaired palates. Individuals with unrepaired palatal clefts also failed to exhibit the shortening of the anterior height of the upper face or the elongation of the lower face typical of the CLP group. Farkas and Lindsay (1972b), however, found the total height of the face to be greater in adults with repaired CP than in controls, but the difference was significant in males only. The difference could be attributed almost entirely to an increase in anterior mandibular height, which was significantly greater in both male and female adults with repaired CP compared to controls. The anterior heights of the upper and middle parts of the face were similar in both groups (Farkas and Lindsay, 1972a).
Maxillary, palatal, and mandibular structures
Dahl et al. (1982) demonstrated that children with CP prior to surgical repair exhibited a short maxilla and reduced posterior maxillary height in addition to reduced dimensions of the mandible (especially mandibular length), narrow naso- and pharyngeal airways, and retrognathia. They noted shortening of both jaws as early as age 2 to 3 months in infants with isolated CP compared to infants with incomplete CL. Dahl (1970) found that in adult males with CP, untreated and treated combined, there was a tendency for the maxilla to be smaller in length and retrognathic. The maxilla was retruded in relation to the cranial base and to the mandible compared to a noncleft control group. In a study that combined untreated and treated adult females with CP, Bishara (1973a) reported similar results. There was a tendency toward posterior positioning of the maxilla and mandible in relation to the cranial base, while the maxillary-mandibular relation was not significantly different from that of noncleft controls. However, Šmahel et al. (1987) found no evidence of a posterior positioning of the maxilla in their study of 60 children prior to palatal repair.
Several studies have found serious aberrations in the craniofacial morphology of CP patients with repaired clefts, although it is unclear whether these are due to the repair or to the cleft. For example, two groups of investigators reported a relatively posterior position of the maxilla and mandible in relation to the cranial base in operated patients with CP compared to controls (Shibaski and Ross, 1969; Dahl, 1970). Reduction in the size of the mandible of children with repaired CP was reported by Nakamura et al. (1972). Šmahel et al. (1987) noted shortening of both jaws in adults who had undergone palatal repair as children, although Dahl et al. (1982) also found this feature in children prior to palatal repair. Šmahel et al. (1987) did not find retrognathia, commonly associated with CLP, or significant differences in other areas of the face except a reduction in the mandibular plane inclination in their sample of individuals with repaired CP. Casal et al. (1997) found no evidence of retrognathia in their study of young children with reparied CP but suggested that this may be secondary to the early age at reconstructive surgery.
Dental arches and relationships
Shortening of the dentoalveolar arch prior to palatoplasty was demonstrated by Peterka (1979). Shibasaki and Ross (1969), in a cross-sectional growth study of children with operated isolated CP, found that the maxillary underdevelopment progressed with age but that an acceptable facial balance was achieved because of the positional changes of the mandible. In addition, both Bishara (1973a) and Casal et al. (1997) found that, despite the posterior positioning of the maxilla and mandible, the maxillary-mandibular relationship was normal and similar to the noncleft group.
Discussion
Given the diversity of research designs, sample composition, treatment protocols, and analytical approaches evident in studies of facial form associated with oral clefts, disagreement is not surprising. Nevertheless, some common findings can be discerned (Table 7.1). Investigators have demonstrated consistently that the facial morphology of infants, children, adolescents, and adults with various types of cleft and various stages of treatment deviated from normal (or control) populations (Ross and Coupe, 1965; Dahl, 1970; Aduss, 1971; Friede and Johanson, 1977; Dahl et al., 1982, 1989; Friede et al., 1987; Molsted et al., 1987; Kilpeläinen and Laine-Alava, 1996; Kilpeläinen et al., 1996; Hermann et al., 1999). There is greater facial disruption in bilateral than in unilateral CLP and in unilateral CLP than in isolated CP or CL, in that order. Researchers have shown consistently that individuals with isolated CL have fairly normal facial development in comparison to subjects with some form of CLP or isolated CP (Dahl, 1970). However, even these CL individuals have been shown to differ, on average, from control values and in a fashion consistent with other forms of CLP.
|
TABLE 7.1. What We Know about Facial Shape in Cleft Lip with or without Cleft Palate and in Cleft Palate Alone |
|
|
Šmahel (1984a) concluded that the configuration of the face in complete unilateral and bilateral CLP was characterized by essentially identical patterns. This pattern included retroclination of the dentoalveolar component of the upper jaw, retroposition of the mandible and of the maxilla per se, and posterior growth rotation of the face. There was impairment of vertical growth within the lateral parts of the upper face in both groups. Šmahel (1984a) further suggested that increased intraorbital widths might be an intrinsic characteristic that predisposed the individual to develop the cleft.
Among the many unresolved questions is the effect of surgery on the growth and development of the face in individuals with oral facial clefts. In a recent series of studies that compared infants with early surgical repair of complete unilateral CLP with a group who had repaired incomplete unilateral CL, Hermann et al. (1999, 2000) argued that bimaxillary retrognathia, short posterior height of the maxilla, and short mandible were intrinsic in the CLP population and not the consequence of surgery. However, the maxillarymandibular discrepancies reported in operated unilateral CLP patients were not frequently observed in the unrepaired faces of individuals with unilateral CLP, indicating that these features may be secondary to the treatment (Bishara, 1973a,b; Bishara et al., 1976,1985).
In regard to isolated CP, debate centers around the merits of early palatoplasty, which increases the likelihood of normal speech, vs. later closure, which may maximize maxillary growth potential (and hence reduce the impact of surgery on growth). Rohrich and Byrd (1990) suggested that four key variables determine the outcome of cleft surgery: cleft type, surgeon's expertise, operative technique, and timing of the repair. They concluded from a careful review of the literature that the type of surgery and the skill with which it was carried out predicted outcome better than the timing of the palatal closure. Šmahel et al. (1999) argued in a similar vein, noting that surgical impact on growth was greater in their sample of 187 adult men with CLP, CP, and CL, all of whom had their surgical repairs before the “contemporary state of treatment,” which includes both alveolar repair and treatment with fixed orthodontic appliances. They found no difference in the basic bony facial characteristics of CP vs. CLP, a fact that underscores the overriding impact of the palatal repair in these two otherwise distinct populations.
Finally, there is great phenotypic variation within any group of individuals with oral facial clefts, even those who would be grouped together on the basis of cleft location or severity (Farkas et al., 1993; Ishikawa et al., 2000). This fact is seldom discussed but readily evident in studies where standard deviations are reported. Consistently, these measures of variation within the sample are larger in the cleft groups than in the controls. This is significant because nearly every study of facial morphometrics among individuals with oral clefts utilizes mean values to characterize cleft types (e.g., bilateral vs. unilateral.) However, these mean phenotypes are probably a poor representation of the individual cases comprising the category.
Facial form in Relatives of Affected individuals
Family Members of Individuals with Cleft Lip or Cleft Lip and Palate
The interest in a possible relationship between facial shape and increased risk for CL/P can be traced to studies on differences in the susceptibility of various strains of mice to CL/P. Trasler (1968) suggested that variations in embryonic facial shape and growth dynamics were factors in the increased susceptibility to CL in the A/J mouse compared to the C57BL strain. Specifically, she proposed that the more prominent and centrally placed medial nasal process characteristic of the A/J strain led to greater rates of failure in the fusion of medial and lateral nasal processes and in subsequent breakdown of the isthmus of tissue connecting medial, lateral, and maxillary processes. In effect, these mice had a heritable facial shape that led to a lower threshold (and hence a greater incidence) for CL.
As a result of Trasler's work, Fraser and Pashayan (1970) suggested the following: “if the shape of the embryonic face is related to the shape of the postnatal face, and if the face shape is at least in part genetically determined, and if face shape is indeed related to the predisposition to cleft lip, it follows that the parents of children with cleft lip should have faces that are, on the average, of a different shape than those of the general population.” These authors tested their hypothesis by examining 50 parents of children with CL/P (25 males and 25 females) and contrasting this sample with normal controls drawn from other clinical (noncleft) populations and hospital staff. Using direct measurements, measurements from photographs, and “physioprints” (after Sassouni, 1962), they concluded that the parental group tended to have, on average, longer facial heights (intraocular to chin dimensions), narrower bizygomatic widths, underdeveloped maxilla, thin upper lips, and more rectangular or trapezoidal facial shapes than ovoid (Fraser and Pashayan, 1970). Moreover, they argued that these deviations were more pronounced in parents who had more than one affected child. Finally, they suggested the following: “Though the genetic basis for the quantitative differences demonstrated in this study is [sic] likely to be complex, it is possible that specific traits showing simple Mendelian inheritance can be distinguished. Their identification would help to clarify the biological basis for the genetically determined susceptibility to cleft lip.”
Subsequent research can be divided into three categories, based on research design and theoretical assumptions. In the majority of studies, researchers have followed the basic design and assumptions first outlined by Fraser and Pashayan (1970). Thus, a multifactorial model was assumed and a population of parents was contrasted en masse to a population of controls. Differences were then documented using straightforward statistical tests of means. A second category of research design reflects the growing awareness of the etiological and genetic variability present within and between CLP and CP populations. This approach requires more sophisticated statistical models and the acceptance of the possibility that risk factors are not evenly distributed among or between parental pairs. The third category of research has emerged in the last several years as investigators have attempted to link suspected genetic markers associated with oral facial clefts to specific morphological patterns seen in the parents. This approach combines our greatly increased understanding of craniofacial morphogenetics with our knowledge of craniofacial morphometrics.
Multifactorial Models
Coccaro et al. (1972) built on the results of Fraser and Pashayan (1970), comparing cephalometric radiographs from a group of 40 parents (20 males and 20 females) of children with CL/P to a group of 40 control individuals. Their experimental group consistently displayed a more acute cranial base angle, smaller upper facial heights, shorter palatal lengths, shorter anterior lengths of the palate, and shorter noses. Mandibular length was greater in the experimental group than in the controls. They speculated that it was the “unfavorable variation” or disharmony between the upper and lower facial sizes that was passed on to their children, disrupting normal development and leading to increased frequency of clefting. Their research again supported the idea that parents of children with CL/P demonstrate a facial form that distinguishes them from the general population, implying a heritable predisposition to the condition.
Erickson (1974), citing Trasler (1968) and Fraser and Pashayan (1970), examined siblings of affected children and showed that they too had distinctive facial features. They compared palatal form, dental arch shape, and facial profile in the siblings of children with CL/P and siblings from families with no immediate history (first- or second-degree relatives) of CL/P. While only one of the three traits (palatal form) differed significantly between the study siblings and controls, consistent differences were found that tended in the direction of those previously reported in parents. Thus, the facial profile in the sibs of cleft children tended to be less convex than in controls and palatal form was tapered and highly arched. As in previous studies (and most of those that have followed), the presumptive genetic model being tested was the multifactorial threshold (MFT) model. Therefore, when calculating values for sibs, Erickson (1974) used the mean value derived from combining the scores of each (unaffected) child in the sibship for each variable. The “midsib” values were then compared between treatment and control groups. If another model is presumed (major gene) or if, as Fraser and Pashayan (1970) suggested, some pertinent aspects of facial form exhibit Mendelian patterns of inheritance, then it might be expected that the associated phenotype would not be equally distributed between members of a sibship. Therefore, any attempt to combine or average their values would weaken the ability to define the relevant phenotypic pattern.
Nakasima and Ichinose (1983) demonstrated that the unique facial phenotype transcended ethnicity. They examined lateral and posterior-anterior cephalometric head plates from parents of 251 children in Japan with oral facial clefts and an equal number of control pairs. These investigators included parents of children with CP as well as CLP and broke the latter sample into subgroups based on severity (CL/P and CL). They utilized a total of 53 cephalometric variables as well as multivariate statistical procedures (analysis of variance and discriminant function analysis) to further explore the data. As in all previous studies, an MFT mode of inheritance was presumed. Thus, the investigators averaged the values for both parents for each variable and compared the group means of these midparental scores to the means derived from the averaged scores of randomly selected pairs of male and female controls.
Their findings were largely consistent with those reported by previous researchers. Parents of CL/P children had on average significantly shorter heads and maxillary lengths as well as shorter anterior middle or upper facial heights coupled with a more open, rotated mandible (greater mandibular angle) and a significantly less convex face than controls. The midparental average for head width in the CL/P parent group was significantly smaller than in the control group. However, outer orbital width was significantly greater, as were forehead and nasal widths. The researchers also examined left-right asymmetry from the frontal radiographs and found that, for the most part, this was not increased in any of the study groups (the exceptions were alveolar width asymmetry and nasal floor height asymmetry among parents of children with CLP).
Given the smaller cranial dimensions in the study group, it would be reasonable to ask to what extent the facial measurements were affected by this general reduction in size. Nakasima and Ichinose (1983)addressed this question by examining proportional relationships between cranial dimensions (maximum head width) and five midfacial widths (outer orbital width, forehead width, nasal width, zygomatic width, and maxillary width). With the exception of the cranial index, the proportions were always larger in the study groups, indicating that the facial measurements were not simply reflections of a generalized reduction in body or head size. Nakasima and Ichinose (1984) investigated this question further in a study that compared cranial area in four groups: children with oral facial clefts, their parents, a control sample of children without clefts, and their parents. Cranial area (brain case) was significantly smaller in both the affected group and their parents (again expressed as a midparental average) compared to the control children and their parents (midparental average). They also found higher correlation coefficients between the control children and their parents for cranial area than between the affected children and their parents and suggested that this might reflect varied disruptions to normal growth caused by the cleft and its subsequent surgical repair. They argued as well that the small brain case appeared to be an inherited factor passed from the parents to the children with clefts and suggested that it may be related to susceptibility to oral facial clefting (Nakasima and Ichinose, 1983).
In the same study, the authors found a correlation between cleft severity in the child and facial form in the parents. Hence, they demonstrated that parents of children with CL alone consistently displayed less pronounced deviations from control values than did parents of children with CLP. This was noted both in mean values of the various measurements and in the number of these values that differed significantly from control values. However, these differences were minor compared to the differences with controls, as indicated by discriminant function analysis. This multivariate technique was unable to successfully separate the three study groups (CL, CL/P, and CP) from one another, but each could be separated from the controls, as could the study group as a whole. The authors proposed that these differences were great enough to serve as a possible mechanism for identifying parents at greater risk for having an affected child (Nakasima and Ichinose, 1983).
Raghavan et al. (1994) analyzed a set of 38 parents of children with CL/P from India and compared these individuals to a control set of 24 parents of healthy children from the same area. They utilized both lateral and posterior-anterior cephalograms to assess mean differences between the experimental group and the control group for 30 variables, Midparental values were used to generate mean values for each group, once again reflecting the assumption of a multifactorial mode of inheritance. Seven of the 22 means derived from the lateral films were significantly different between the two samples, with the experimental group demonstrating on average a significantly more obtuse cranial base angle, a more acute articular angle, shorter upper facial height, shorter posterior facial height, a greater palatal length, a more projecting midface, and a more obtuse mandibular angle. In contrast, seven of the eight width dimensions differed significantly between the control and experimental groups, with maximum head width, bizygomaticofrontal suture width, zygomatic width, alveolar width, gonial width, and total facial height being smaller in the experimental group. Only nasal cavity width was greater in the experimental group. These results are somewhat at odds with those reported by previous researchers. Examination of their published data, however, offers a possible explanation for this apparent discrepancy: head size appears to be reduced on average in the experimental group. Had these investigators followed Nakasima and Ichinose (1983) and used ratios to explore the possibility that the experimental group might exhibit facial widths that were broad compared to head size, a different set of conclusions would have resulted. Ratios can be calculated from the mean values in their tables, and these indicate that for inner orbital width, frontal width, and bigonial width (in addition to nasal width) the values from the experimental group ranged from 3% to 16% larger than those for the controls, a finding that is much more in keeping with those reported earlier.
Investigating Phenotypic Heterogeneity
Kurisu et al. (1974) expanded on all previous studies in a landmark work that still stands as a model of methodological rigor. Their study used a greatly increased sample size and included parents of children with isolated CP as well as CL/P. In addition, they utilized two independent control samples, one from Lancaster, Pennsylvania, and the other from Ann Arbor, Michigan. They also utilized posterior-anterior cephalograms, in addition to the more traditional lateral head films, to obtain a set of 20 linear and angular variables. Theirs was the first study to analyze fathers separately from mothers. Their methods included both factor analysis to generate a better understanding of the underlying biological (anatomical) differences between the groups and a form of cluster analysis (Q-mode correlations in conjunction with principal components analysis) to study the multivariate or global differences between the study and control groups.
In general, Kurisu et al. (1974) found the same phenotypic associations reported by Coccaro et al. (1972). Thus, parents of children with CL/P tended to have more concave facial profiles, more relative mandibular prognathism, and shorter vertical and wider horizontal dimensions in the upper face. However, fathers demonstrated these features more dramatically than mothers in regard to both individual variables and factor scores. (Their results relative to parents of children with CP are discussed in a later section of this chapter, Parental Facial Form in Isolated Cleft Palate.) The researchers found no correlation between cleft severity and degree of facial abnormality in either parent and called into question the MFT model of inheritance, which would predict such a correlation. Interestingly, they also found significant differences between the two normal control populations and cautioned that such variation is likely to account for some of the discrepancies reported between studies using different control populations.
Ward et al. (1989, 1994) criticized previous studies, including that of Kurisu et al. (1974), because parental samples were treated as uniform entities. They argued that in any large sample, whether parents of sporadic or of familial CLP children, a variety of etiologies would likely be represented. For example, among the sporadic cases, some would have a greater familial or genetic component than others (it is well known that a percentage of sporadic cases become familial cases after a subsequent affected child is born). Even in those cases with a stronger familial component, there were likely to be multiple genetic loci associated with increased risk for clefting, not all associated with the same or any effect on facial form in the parents. Furthermore, Ward et al. (1989) noted that in most previous studies the MFT model had been assumed, in which both parents contributed to the risk of an affected child. Hence, parental phenotypes were considered en masse even when values were separated by sex. There is nothing in the MFT model that requires the parental contributions to be equal, and moreover, there is ample evidence that the MFT model is inadequate for explaining the inheritance of many cases of oral facial clefting. Finally, when relying on mean values alone, little attention is paid to variation within samples.
To overcome these methodological shortcomings, these investigators sorted their parental sample (lateral cephalograms from 82 parents of children with sporadic CL/P) using multivariate cluster analysis. This technique generated multivariate distance or similarity measures to identify individuals with similar phenotypes. A dendrogram, or tree diagram, was produced from these results, which displayed the pattern of similarity between groups of individuals. Resulting clusters were compared to one another using linear regression analysis, which measures the correlation, or similarity in the pattern of variation, between two clusters (with a value of 1 representing pattern identity).
The results of this analysis supported the contention that there was considerable variability in phenotypes among parents of children with CL/P. While half of the parents sorted into a large group (n = 39) with mean values that differed only in being consistently (but insignificantly) a little larger than the published norms, the remaining 43 parents sorted into a series of smaller groupings that had major deviations from the published norms. All clusters were compared (using regression analysis) to an independent and unrelated sample of 16 individuals with repaired CLP. Interestingly, the largest two phenotypically unusual groupings (n = 17 and n =12) exhibited patterns of variation that were surprisingly similar to the phenotype seen in individuals with CLP. Correlations between the facial pattern of these two clusters and that of the group with overt clefts were (r = 0.88) and (r = 0.68), respectively. In contrast, there was little correlation between the pattern in the cleft group and the largest (nominally normal) cluster of parents (r = 0.37).
Another outcome of Ward et al.'s (1989) different methodological approach was the documentation that, at least in this sample, it was rare for both parents to be in one of the “phenotypically unusual” groups. Thus, in only 15% of the pairs were both parents unusual. In over half of the pairs (54%), only one member showed an unusual facial phenotype, and in 31% of the pairs neither parent was unusual. These findings suggest that some sporadic cases have a genetic or familial component while others do not.
Mossey et al. (1998b) examined 40 fathers and 43 mothers of children with CL, CLP, and isolated CP. They analyzed fathers and mothers separately, echoing the concerns of Ward et al. (1989) that combining parents in a single group ignored the likelihood that parental contributions to risk in children are unequal. The parents were compared to a carefully selected group of 100 control individuals from the same general population (Glasgow). Thirty-seven measurements of the head and face were derived from lateral cephalograms and included area measurements as well as angles and linear measures. Univariate analysis revealed that both fathers and mothers differed significantly from controls but displayed little concordance in the specific traits that were significantly different. In fathers, the mean mandibular area measurement was significantly smaller than in controls, as was the mean palatal length. The mean cranial base angle was significantly more acute and the mean cross-sectional area of the cranium significantly smaller than those derived from control males. Only the mean area of occipital subtenuce was larger in fathers than control males. In contrast, means for mothers exhibited significant increases in mandibular length, total facial height, anterior cranial base, and clivus length. Mothers resembled fathers in having significantly reduced means for cranial area.
AlEmran et al. (1999) also found differences in fathers and mothers of children with clefts. Specifically, utilizing frontal cephalograms, they reported that fathers tended to exhibit significantly greater nasal widths and narrower maxillary widths compared to controls. Mothers tended to display significantly smaller head, bizygomatic, alveolar, and bigonial widths (although these differences were not apparent when considered as ratios to head width). Both mothers and fathers had significantly increased facial asymmetry in orbital and alveolar measurements to a constructed midline compared to controls. However, these results are difficult to interpret because the investigators apparently did not separate parents of children with CP from those with other types of cleft.
Some investigators have examined populations with familial forms of isolated CLP on the assumption that these may be more likely to display a heritable facial phenotype associated with the risk for oral facial clefts. Thus, Ward et al. (1994) studied a single multiplex family using lateral and posterior-anterior cephalograms to assess seven craniofacial variables. They demonstrated that the four individuals in this family who would be classified as likely gene carriers by pedigree analysis (i.e., they had an affected parent or a sib and an affected child but were not cleft themselves) exhibited a distinct craniofacial phenotype, similar in some ways to that reported in previous studies (Fraser and Pashayan, 1970; Cocarro et al., 1972; Nakasima and Ichinose, 1983). Thus, these individuals had significantly larger means for (outer) orbital widths and nasal cavity widths. However, in contrast to previous studies, the carriers had significantly shorter lower faces than either the affected or the obligate normal group. Carriers also did not show the decreased facial convexity or increased lower facial heights frequently described in previous studies. Multivariate discriminant function analysis showed that two functions comprised of four variables (nasal cavity width, facial profile angle, palatal length, and lower facial height) correctly classified 89% of the affected, carrier, and obligate normal family members (no carriers were misclassified, one affected individual was misclassified as obligate normal, and one obligate normal was classified as an obligate carrier). The fact that carriers had even greater mean values for nasal cavity width and outer orbital width than affected (though treated) individuals suggests that these phenotypic characteristics do not simply result from the oral cleft (since the carriers have no cleft).
Suzuki et al. (1999) examined 65 individuals (25 fathers and 40 mothers) who were parents of children with CL/P but who also had at least one (blood) relative with CL/P within the last three generations. These familial cases were compared to an unspecified control sample of 413 parental pairs using 19 ratios (to adjust for the effects of sex) derived from lateral and posterior-anterior cephalograms. Their results indicated that on average the adjusted measurements for interorbital width, bicoronoid width, nasal cavity width, and anterior cranial base length were significantly increased in the parental sample compared to the controls. Discriminant function analysis indicated that parents and controls could be correctly classified 68% of the time. This value is considerably less than that reported by Ward et al. (1994), but Suzuki et al. (1999) used a sample that was nearly 45 times larger.
Linking Morphometrics and Morphogenetics
Mossey et al. (1998a) combined molecular and morphometric techniques. Specifically, they examined polymorphisms in the transforming growth factor alpha (TGF-α) locus in 83 parents (the same sample reported in Mossey et al., 1998b). They found that one polymorphism (the C2 allele of the TGF-α Taql polymorphism), although rare in general, had a significantly higher occurrence in the parents of both CL/P and CP children than in controls. Also, by combining these genotypic predictors of risk status with cephalometric predictors, discriminant function classification improved. They correctly classified 76% of the CP and 94% of the CL/P parents (compared to controls). The authors acknowledged that it is unlikely that the TGF-α gene plays a major role in the susceptibility to oral facial clefts but speculated that it might be a modifying factor. Beiraghi et al. (1994), who linked the risk of CL/P to the 4p-region of chromosome 4, examined the same family as Ward et al. (1994). In an unpublished study, Ward and Beiraghi (personal communication) found that there was 100% correspondence between the individuals identified as carriers by genetic marker analysis and morphometric analysis.
Parental Facial Form in Isolated Cleft Palate
Craniofacial form in parents of children with isolated CP has not been as extensively studied, and the few studies that can be found exhibit many of the same deficiencies in research design described in the previous section. Nevertheless, the same tendency for parents of affected children to exhibit unusual patterns of facial form has been reported in each of these studies. The nature of this pattern differed in significant ways from that described in parents of children with CLP. Thus, Kurisu et al. (1974) and Nakasima and Ichinose (1983) found that the parents of children with CP evidenced fewer deviations from normal compared to the parents of children with CL/P; however, the general direction for the differences was similar. Kurisu et al. (1974) also noted that while fathers and mothers tended to have the same pattern of deviation from normal, the specific list of variables that differed significantly from controls varied by sex. For example, both fathers and mothers had, on average, larger maxillary widths and shorter anterior mid-facial measurements, but the former reached a value that was significant in the mothers while the latter reached significance in the fathers. In general, fathers differed more than mothers compared to control values. Prochazkova and Tolarová (1986) also examined mothers and fathers of CP children separately (20 mothers and 20 fathers). These subjects were compared to a control sample of 75 university students for a series of 34 characteristics taken from lateral cephalograms and dental casts. They reported that parents on average had a significantly longer anterior cranial base, longer palate, and shorter mandibular body. Soft tissue lower facial heights were greater in parents compared to controls. Again, there were a few minor differences in significance levels for variables between mothers and fathers, and, in general, the fathers showed greater deviation from normal in most measurements. In a later study, Prochazkova and Vinsova (1995) found that males tended to have significantly larger mandibles compared to controls, but this feature was not characteristic of females. Both males and females however, differed from controls in a number of other cephalometric and anthropometric variables.
Mossey et al. (1997) examined 35 parents of children with CP as well as 12 with CL children and 36 with children who had CL/P derived from a population in western Scotland. Using 37 linear, angular, and area measurements from lateral cephalograms, they concluded that were no significant differences between parents of children with CL and CL/P but a few significant differences between parents of children with CP and CL/P. However, these differences were apparent only if the sample was first subdivided by sex. In this case, means for mothers of children with CP were consistently greater for mandibular length, mandibular ramus length, mandibular area, and cranial area when compared to the means for mothers of children with CL/P. Stepwise discriminant function analysis could correctly classify 80% of the CL/P mothers and 75% of the CP mothers using just mandibular ramus length and cranial height. Mandibular ramus length alone could correctly classify 71% of the CP parents and 63% of the CL/P parents regardless of sex.
All of these research efforts suffered from the same methodological assumption of a multifactorial mode of inheritance seen in most of the studies relatives CLP cases. In an ongoing research effort in our lab, we are using a different approach but observing similar sex specific morphometric differences in parents of children with CP. More specifically, Sammons (1999) examined 15 parental pairs who had a child with isolated CP. Seventeen variables from lateral cephalograms and 25 posterior-anterior variables were analyzed and compared to published data derived from the same geographic region (Saksena et al., 1987, 1990). Twenty of the 42 variables differed significantly from the reference means. Most values were smaller in the parental group (14/20), although mean values for inner orbital width, nasal shelf width, mandibular angle, mandibular protrusion, lower facial height, and palatal length were larger in the parental group. However, Sammons (1999) went the additional step of using cluster analysis to determine if distinct groups could be identified with in this parental sample. Two groupings could be defined, one comprised largely of females and the other of males. The predominantly female group (9/12 or 75%) was characterized by relatively greater reductions in facial height, shorter palates, longer mandibular bodies, and greater mandibular prognathism compared to the cluster that was predominantly male (12/18 or 67%). While these findings do not correspond directly with those reported by Mossey et al. (1998b), they do support the contention that sex has an important influence on the expression of facial features that differentiate parents of CP children from the general population.
Discussion
What has been learned in the 30 years since Fraser and Pashayan (1970) first suggested that facial form in parents might relate to the risk of oral facial clefts in their offspring? The answer to this question is complex. As Mossey and coworkers (1998a,b) have demonstrated in their reviews of this subject, there is a lack of agreement on the specific craniofacial features that distinguish parents with such risk factors from the general population. However, we can make some generalizations (Table 7.2).
In spite of the differences in study design, variables assessed, and ethnic origin of the populations under study, investigators have consistently found that, taken as a group, parents of children with isolated CL/P and CP display morphometric features that distinguish them from the general population (as represented by control subjects). This in and of itself would not be surprising because, as Kurisu et al. (1974) demonstrated, even two reference normal populations can exhibit an alarming number of statistically significant differences if they are drawn from different regions of the country and/or are measured by different investigators. However, the differences exhibited by the parents of children with oral clefts are quite similar to those shown by individuals with oral facial clefts (as demonstrated by Ward et al., 1989). Two classes of shared traits stand out. First, the repeated finding of excess intraorbital widths in both affected individuals and noncleft parents of children with CL/P reinforces the original observations of Fraser and Pashayan (1970) and suggests that increased facial width may be a heritable trait that predisposes one to an oral cleft by reducing the threshold for its occurrence. Second, the excessive width through the orbits (and in some cases other regions of the midface) may indicate a general disturbance in growth, perhaps caused by a major gene operating in the craniofacial developmental pathway. In this case, altered facial form could be seen as a forme frust of the full-blown cleft.
|
TABLE 7.2. What We Have Learned about Facial Shape in Relatives of Individuals with Orofacial Clefts |
|
|
One possible developmental anomaly that may explain the peculiar set of features shared by parents and affected individuals, at least for CLP, is the size and shape of the cartilaginous cranial base. In studies using axial projection instead of standard cephalometrics, Molsted and colleagues (1993) found that there were significant differences in the width of the sphenooccipital synchondrosis and cranial base dimensions between a group of infants with untreated CLP and a group of infants with untreated CL with minor incomplete clefts (used as a normal control). They also found that the CLP group exhibited significantly greater widths in the superior as well as the inferior part of the sphenooccipital synchondrosis. No difference between the groups was found in the cranial base length. However, the distance from the synchondrosis to the sella point was significantly shorter in the CLP group (Molsted et al., 1993). In their 1995 study, Molsted et al. reported that the CLP group had an increased width of the cranial base and that the angle between the right and left ala major of the sphenoid bone and the angle between the petrous part on the right and left side was also increased. Also, the posterior width of the maxilla was greater. Others have found that anterior-posterior cranial base dimensions are reduced in individuals with oral clefts (Bishara and Iversen, 1974; Horswell and Gallup, 1992). Dahl et al. (1982) suggested that the reduced anterior-posterior dimensions of the cranial base might be an intrinsic (predisposing) factor in CP. We do not know whether such specific morphometric features are heritable, but if so, the primary defect in some forms of CLP could be an abnormally wide but short cranial base. This would account for the greater facial widths in both a parent and the affected child, as well as abnormal midfacial depths (retruded maxilla, etc.), and could, following Trasler's (1968) original suggestion, lower the threshold for expression of the cleft.
The second area where common morphometric abnormalities can be seen is the mandible, which tends to be posteriorly rotated both in people with CL/P and in (some) otherwise normal parents of these individuals. This rotation also gives rise to either an increased total facial height or lower facial height and often an associated decrease in facial convexity. Mossey and coworkers (1997) suggested that in parents expressing some primary defect in maxillary growth (a finding in many studies of parents of children with CL/P) the mandibular abnormalities may reflect functional adaptation to the deficiency in the midface. They further suggested, as did Cocarro et al. (1972), that these differences in mandibular growth might interact with embryonic tongue positioning in a susceptible offspring and subsequently interfere with lip and/or palatal closure. However, the fact that an increased intraorbital width and a posterior rotation of the mandible occur together in otherwise normal parents of children with CL/P argues for a more generalized disturbance in craniofacial growth.
The suggestion that variations in the shape of the cranial base or in any other region of the face correlate with morphological changes in structures far removed from the primary defect is a logical consequence of the functional matrix theory, especially in its most recent incarnations (Moss, 1997a,b). It also may indicate that many cases of oral facial clefts are best seen as part of larger malformation sequences, as suggested by Shprintzen et al. (1985) and Horswell and Gallup (1992).
It is interesting that in most studies of parents of children with CP the morphometric differences with the control population are less pronounced and seldom include significant differences in facial width or cranial size. However, variations in mandibular morphology may be more common in some relatives of individuals with CP. There also appears to be a stronger correlation between facial morphology and sex in parents of CP children. These differences suggest that heritable differences in facial form may be less important in producing a child with CP than some as yet unspecified influence of sex or imprinting coming from the parents.
Conclusions
Facial form associated with CL/P reflects a complex mix of primary and secondary factors. Thus, in both noncleft relatives and individuals with CL/P prior to treatment, excessive width through the midface (orbits and nasal cavity) and posterior rotation of the mandible suggest that facial form itself may be a causal factor or a minimal expression of the cleft. It is equally clear that an untreated cleft, whether of an isolated palate, lip and palate, or primary palate, interferes with facial growth and distorts facial form. Finally, treatment and surgical repair can also disrupt the normal pattern of facial growth. However, one of the great remaining difficulties faced by morphometricians, developmental biologists, and clinicians is that presented by the variability of expression within the population of individuals with oral clefts (and their families). The standard approach to studying facial form as it relates to CL/P has always been to compare groups of individuals with a certain cleft phenotype to groups lacking this trait. This of course ignores the likelihood of genetic and etiological heterogeneity within the cleft population (Shprintzen et al., 1985). This intrinsic variability is further obscured by the practice of summarizing the phenotype through the use of mean values for the traits in question. The consequent tendency to rely on phenotypic typologies that may poorly reflect underlying etiological typologies remains the biggest hindrance to a meaningful linkage between morphometrics and morphogenetics. Finally, Moss (1997a,b) suggested that variant genes in the developmental pathway interact in an as yet unrecognized but undoubtedly complex fashion with epigenetic (environmental) factors to produce a given phenotype. From this perspective, variability is an inevitable expression of developmental complexity, and it is unlikely that we will ever understand the morphometrics of individuals with oral clefts without equal attention to both genetic and epigenetic factors that can influence facial form. This problem is underscored by research in the last two decades that has identified a number of genetic loci that may be linked to the expression of oral clefts in different individuals (Mossey et al., 1998a).
It appears that, as Fraser and Pashayan (1972) first predicted, facial shape may be an important aid in sorting out the genetic contributions to oral facial clefts. Specific morphometric patterns may serve as a marker for identifying individuals or families in which there may be a greater risk for inheriting this condition. The possibility of utilizing this information in combination with emerging technologies of genetic linkage analysis has not gone unnoticed (Ward et al., 1989, 1994; Mossey et al., 1998a). However, future morphometric research must recognize the lessons already learned, and a more sophisticated research design must be used than has generally been employed in the past. First, morphometric research must allow for variable expression of the key features in individuals affected by oral clefts and their otherwise unaffected relatives. Indeed, it is likely that there are multiple rather than singular characteristic phenotypes within the traditional cleft typologies of bilateral, unilateral, complete, and incomplete CL/P. For example, individuals with CLP caused primarily by intrauterine exposure to a teratogen may have a different phenotypic pattern than those who inherited a cleftsusceptible facial form. Research has shown that among parents of affected individuals, significant phenotypic differences exist within and between parental pairs as well as by sex. Second, future research must be guided by our increasing understanding of the complex developmental pathways that can lead to an oral cleft. Mossey and colleagues (1998a) were the first to utilize such a combined approach, but it may be expected that more productive lines of inquiry will emerge as our understanding of the genetics of craniofacial development improves; i.e., it may be possible to associate specific genetic variants with different patterns of facial form as well as with the probability of the appearance of CL/P in one's offspring. Third, whether we are studying facial form as it relates to the effects of the oral facial clefts, their surgical repair, or subtle differences among family members, the new three-dimensional technologies and geometrically based analyses of form must be harnessed to create a more precise understanding of facial morphology and its inheritance. If these new techniques become as readily accessible as the two-dimensional head plate and associated cephalometric analysis have been, the new century can expect to see a resolution of those questions that escaped the last.
References
Aduss, H (1971). Craniofacial growth in complete unilateral cleft lip and cleft palate. Angle Orthod 41: 202–213.
Aduss, H, Pruzansky, S (1967). The nasal cavity in complete unilateral cleft lip and palate. Arch Otolaryngol 85: 53–61.
AlEmran, SE, Fatani, E, Hassanain, JE (1999). Craniofacial variability in parents of children with cleft lip and palate. J Clin Pediatr Dent 23: 337–341.
Athanasiou, AE, Tseng, CY, Zarrinnia, K, Mazaheri, M (1990).Frontal cephalometric study of dentofacial morphology in children with bilateral clefts of lip, alveolus and palate. J Craniomaxillofac Surg 18: 49–54.
Atherton, JD (1967a). Morphology of facial bones in skulls with unoperated unilateral cleft palate. Cleft Palate J 4: 18–30.
Atherton, JD (1967b). A descriptive anatomy of the face in human fetuses with unilateral cleft lip and palate. Cleft Palate J 4: 104–114.
Baumrind, S, Moffit, F, Curry, S (1983). Three-dimensional x-ray stereometry from paired coplanar images. A progress report. Am J Orthod 84: 313.
Beiraghi, S, Foroud, T, Diouhy, S, et al. (1994). Possible localization of a major gene for cleft lip and palate to 4q. Clin Genet 46: 255–256.
Bergland, O, Sidhu, SS (1974). Occlusal changes from the deciduous to the early mixed dentition in unilateral complete clefts. Cleft Palate J 11: 317–326.
Bishara, SE (1973a). Cephalometric evaluation of facial growth in operated and non-operated individuals with isolated clefts of the palate. Cleft Palate J 10: 239–246.
Bishara, SE (1973b). The influence of palatoplasty and cleft length on facial development. Cleft Palate J 10: 390–398.
Bishara, SE, de Arrendondo, RS, Vales, HP, Jakobsen, JR (1985). Dentofacial relationships in persons with unoperated clefts: comparisons between three cleft types. Am J Orthod 87: 481–507.
Bishara, SE, Iversen, WW (1974). Cephalometric comparisons on the cranial base and face in individuals with isolated clefts of the palate. Cleft Palate J 11: 162–175.
Bishara, SE, Jakobsen, JR, Krause, JC, Sosa-Martinez, R (1986). Cephalometric comparisons of individuals from India and Mexico with unoperated cleft lip and palate. Cleft Palate J 23:116-25.
Bishara, SE, Krause, CJ, Olin, WH, et al. (1976). Facial and dental relationships of individuals with unoperated clefts of the lip and/or palate. Cleft Palate J 13: 238–252.
Bishara, SE, Olin, WH, Krause, CJ (1978). Cephalometric findings in two cases with unrepaired bilateral cleft lip and palate. Cleft Palate J 15: 233–238.
Bishara, SE, Sierk, DL, Huang, KS (1979). A longitudinal cephalometric study on unilateral cleft lip and palate subjects. Cleft Palate J 26:59-71.
Boo-Chai, K (1971). The unoperated adult bilateral cleft of the lip and palate. Br J Plast Surg 24: 250–257.
Bookstein, FL (1982). On the cephalometrics of skeletal change. Am J Orthod 82: 177–198.
Brader, AC (1957). A cephalometric x-ray appraisal of morphological variations in cranial base and associated pharyngeal structures: implications in cleft palate therapy. Angle Orthod 27: 179–194.
Broadbent, BH (1931). A new x-ray technique and its application to orthodontia. Angle Orthod 1: 45–66.
Capelozza Filho, L, Taniguchi, SM, da Silva Filho, OG (1993). Craniofacial morphology of adult unoperated complete unilateral cleft lip and palate patients. Cleft Palate Craniofac J 30: 376–381.
Casal, C, Rivera, A, Rubio, G, et al. (1997). Examination of craniofacial morphology in 10-month to 5-year-old children with cleft lip and palate. Cleft Palate Craniofac J 34: 490–497.
Chen, KF, So, LL (1997). Soft tissue profile changes of reverse head-gear treatment in Chinese boys with complete unilateral cleft lip and palate. Angle Orthod 67: 31–38.
Cheverud, J, Lewis, JL, Bachrach, W, Lew, WD (1983). The mea surement of form and variation in form: an application three-dimensional quantitative morphology by finite-element methods. Am J Phys Anthropol 62: 151–165.
Chierici, G, Harvold, EP, Vargervik, K (1973). Morphogenetic experiments in cleft palate: mandibular response. Cleft Palate J 10: 51–61.
Coccaro, PJ, D'Amico, R, Chavoor, A (1972). Craniofacial morphology of parents with and without cleft lip and palate children. Cleft Palate J 9: 28–38.
Coupe, TB, Subtelny, JD (1960). Cleft palate—deficiency or displacement of tissue. Plast Reconstr Surg 26: 600–612.
Cronin, DG, Hunter, WS (1980). Craniofacial morphology in twins discordant for cleft lip and/or palate. Cleft Palate J 17: 116–126.
Dahl, E (1970). Craniofacial morphology in congenital clefts of the lip and palate. An x-ray cephalometric study of young adult males. Acta Odontol Scand 28: 11–164.
Dahl, E, Hanusardottir, B (1979). Prevalence of malocclusion in the primary and early mixed dentition in Danish children with complete cleft lip and palate. Eur J Orthod 1: 81–88.
Dahl, E, Kreiborg, S, Jensen, BL (1989). Roentgencephalometric studies of infants with untreated cleft lip and palate. In: What Is a Cleft Lip and Palate? A Multidisciplinary Update, edited by O Kriens. Stuttgart: Georg Thieme Verlag, pp. 113–115.
Dahl, E, Kreiborg, S, Jensen, BL, Fogh-Andersen, P (1982). Comparison of craniofacial morphology in infants with incomplete cleft lip and infants with isolated cleft palate. Cleft Palate J 19: 258–266.
da Silva Filho, OG, Carvalho Lauris, RC, Capelozza Filho, L, Semb, G (1998). Craniofacial morphology in adult patients with unoperated complete bilateral cleft lip and palate. Cleft Palate Craniofac J 35: 111–119.
da Silva Filho, OG, Normando, AD, Capelozza Filho, L (1992). Mandibular morphology and spatial position in patients with clefts: intrinsic or iatrogenic? Cleft Palate Craniofac J 29: 369–375.
da Silva Filho, OG, Normando, AD, Capelozza Filho, L (1993). Mandibular growth in patients with cleft lip and/or cleft palate—the influence of cleft type. Am J Orthod Dentofacial Orthop 104: 269–275.
Enemark, H, Bolund, S, Jorgensen, I (1990). Evaluation of unilateral cleft lip and palate treatment: long term results. Cleft Palate J 27: 354–361.
Erickson, JD (1974). Facial and oral form in sibs of children with cleft lip with or without cleft palate. Ann Hum Genet 38: 77–88.
Farkas, LG, Hajnis, K, Posnick, JC (1993). Anthropometric and anthroposcopic findings of the nasal and facial region in cleft patients before and after primary lip and palate repair. Cleft Palate Craniofac J 30: 1–12.
Farkas, LG, Lindsay, WK (1971). Morphology of the adult face following repair of bilateral cleft lip and palate in childhood. Plast Reconstr Surg 47: 25–32.
Farkas, LG, Lindsay, WK (1972a). Morphology of the orbital region in adults following the cleft lip-palate repair in childhood. Am J Phys Anthropol 37: 65–73.
Farkas, LG, Lindsay, WK (1972b). Morphology of adult face after repair of isolated cleft palate in childhood. Cleft Palate J 9: 132–142.
Fraser, FC, Pashayan, H (1970). Relation of face shape to susceptibility to congenital cleft lip. A preliminary report. J Med Genet 7: 112–117.
Friede, H, Johanson, B (1974). A follow-up study of cleft children treated with primary bone grafting. 1. Orthodontic aspects. Scand J Plast Reconstr Surg 8: 88–103.
Friede, H, Johanson, B (1977). A follow-up study of cleft children treated with vomer flap as part of a three-stage soft tissue surgical procedure. Facial morphology and dental occlusion. Scand J Plast Reconstr Surg 11: 45–57.
Friede, H, Figueroa, AA, Naegele, ML, et al. (1986). Craniofacial growth data for cleft lip patients infancy to 6 years of age: potential applications. Am J Orthod Dentofacial Orthop 90: 388–409.
Friede, H, Moller, M, Lilja, J, et al. (1987). Facial morphology and occlusion at the stage of early mixed dentition in cleft lip and palate patients treated with delayed closure of the hard palate. Scand J Plast Reconstr Surg Hand Surg21: 65–71.
Friede, H, Pruzansky, S (1972). Longitudinal study of growth in bilateral cleft lip and palate, from infancy to adolescence. Plast Reconstr Surg 49: 392–403.
Gaggl, A, Schultes, G, Karcher, H (1999). Aesthetic and functional outcome of surgical and orthodontic correction of bilateral clefts of lip, palate, and alveolus. Cleft Palate Craniofac J 36: 407–412.
Glass, D (1970). The early management of bilateral cleft of lip and palate. Br J Plast Surg 23: 130–141.
Graber, TM (1949). A cephalometric analysis of the developmental pattern and facial morphology in cleft palate. Angle Orthod 19: 91–100.
Graber, TM (1954). The congenital cleft palate deformity. J Am Dent Assoc 48: 375–395.
Grayson, BH, Bookstein, FL, McCarthy, JG, Mueeddin, T (1987). Mean tensor cephalometric analysis of a patient population with clefts of the palate and lip. Cleft Palate J 24: 267–277.
Grayson, BH, McCarthy, JG, Bookstein, F (1983). Analysis of craniofacial asymmetry by multiplane cephalometry. Am J Orthod 84: 217–224.
Grayson, BH, Weintraub, N, Bookstein, FL, McCarthy, JG (1985). A comparative cephalometric study of the cranial base in craniofacial anomalies. Part I: Tensor analysis. Cleft Palate J 22: 75–87.
Hanada, K, Krogman, WM (1975). A longitudinal study of postoperative changes in the soft-tissue profile in bilateral cleft lip and palate from birth to 6 years. Am J Orthod 67: 363–376.
Handelman, CS, Pruzansky, S (1968). Occlusion and dental profile with complete bilateral cleft lip and palate. Angle Orthod 38: 185–198.
Harvold, E (1954). Cleft lip and palate. Am J Orthod 40: 493–506.
Hayashi, I, Sakuda, M, Takimoto, K, Miyazaki, T (1976). Craniofacial growth in complete unilateral cleft lip and palate: a roentgeno-cephalometric study. Cleft Palate J 13: 215–237.
Hermann, NV, Jensen, BL, Dahl, E, et al. (1999). Craniofacial growth in subjects with unilateral complete cleft lip and palate, and unilateral incomplete cleft lip, from 2 to 22 months of age. J Craniofac Genet Dev Biol 19: 135–147.
Hermann, NV, Jensen, BL, Dahl, E, et al. (2000). Craniofacial comparisons in 22-month-old lip-operated children with unilateral complete cleft lip and palate and unilateral incomplete lip. Cleft Palate Craniofac J 37: 303–317.
Hirschfeld, WJ, Aduss, H (1974). Interorbital distance in cleft lip and palate: significant differences found by sign test. J Dent Res 53: 947.
Horowitz, SL, Graf, B, Bettex, M, et al. (1980). Factor analysis of Craniofacial morphology in complete bilateral cleft lip and palate. Cleft Palate J 17: 234–244.
Horowitz, SL, Graf-Pinthus, B, Bettex, M, et al. (1976). Factor analysis of Craniofacial morphology in cleft lip and palate in man. Arch Oral Biol 21: 465–472.
Horswell, BB, Gallup, BV (1992). Cranial base morphology in cleft lip and palate: a cephalometric study from 7 to 18 years of age. J Oral Maxillofac Surg 50: 681–686.
Huddart, AG (1970). Maxillary arch dimensions in bilateral cleft lip and palate subjects. Cleft Palate J 7: 139–155.
Innes, CO (1962). Some observations on unrepaired hare-lips and cleft palates in adult members of the Dusan tribes of north Borneo. Br J Plast Surg 15: 173–181.
Ishiguro, K, Krogman, WM, Mazaheri, M, Harding, RL (1976). A longitudinal study of morphological Craniofacial patterns via P-A x-ray headfilms in cleft patients from birth to six years of age. Cleft Palate J 13: 104–126.
Ishikawa, H, Kitazawa, S, Iwasaki, H, Nakamura, S (2000). Effects of maxillary protraction combined with chin-cap therapy in unilateral cleft lip and palate patients. Cleft Palate Craniofac J 37: 92–97.
Jain, RB, Krogman, WM (1983a). Cleft type and sex differences in Craniofacial growth in clefting from one month to ten years. Cleft Palate J 20: 238–245.
Jain, RB, Krogman, WM (1983b). Craniofacial growth in clefting from one month to ten years as studied by P-A headfilms. Cleft Palate J 20: 314–326.
Kilpeläinen, PV, Laine-Alava, MT (1996). Palatal asymmetry in cleft palate subjects. Cleft Palate Craniofac J 33: 483–488.
Kilpeläinen, PV, Laine-Alava, MT, Lammi, S (1996). Palatal morphology and type of clefting. Cleft Palate Craniofac J 33: 477–482.
Krogman, WM, Mazaheri, M, Harding, RL, et al (1975). A longitudinal study of the craniofacial growth pattern in children with clefts as compared to normal, birth to six years. Cleft Palate J 12: 59–84.
Kurisu, K, Niswander, JD, Johnston, MC, Mazaheri, M (1974). Facial morphology as an indicator of genetic predisposition to cleft lip and palate. Am J Hum Genet 26: 702–714.
Kyrkanides, S, Bellohusen, R, Subtelny, JD (1995). Skeletal asymmetries of the nasomaxillary complex in noncleft and postsurgical unilateral cleft lip and palate individuals. Cleft Palate Craniofac J 32:428-433.
Lestrel, PE, Berkowitz, S, Takahashi, O (1999). Shape changes in the cleft palate maxilla: a longitudinal study. Cleft Palate Craniofac J 36: 292–303.
Levin, HS (1963). A cephalometric analysis of cleft palate deficiencies in the middle third of the face. Angle Orthod 33: 186–194.
Mars, M, Houston, WJ (1990). A preliminary study of facial growth and morphology in unoperated male unilateral cleft lip and palate subjects over 13 years of age. Cleft Palate J 27: 7–10.
McAlarney, ME, Chiu, WK (1997). Comparison of numeric techniques in the analysis of cleft palate dental arch form change. Cleft Palate Craniofac J 34: 281–291.
Mestre, J, Dejesus, J, Subtelny, JD (1960). Unoperated oral clefts at maturation. Angle Orthod 30: 78–85.
Molsted, K, Dahl, E (1990). Asymmetry of the maxilla in children with complete unilateral cleft lip and palate [published erratum appears in Cleft Palate J 1990; 27: 444]. Cleft Palate J 27: 184–192.
Molsted, K, Asher-McDade, C, Brattstrom, V, et al. (1992). A six-center international study of treatment outcome in patients with clefts of the lip and palate: Part 2. Craniofacial form and soft tissue profile. Cleft Palate Craniofac J 29: 398–404.
Molsted, K, Kjaer, I, Dahl, E (1993). Sphenooccipital synchondrosis in three-month-old children with clefts of the lip and palate: a radiographic study. Cleft Palate Craniofac J 30: 569–573.
Molsted, K, Kjaer, I, Dahl, E (1995). Cranial base in newborns with complete cleft lip and palate: radiographic study. Cleft Palate Craniofac J 32: 199–205.
Molsted, K, Palmberg, A, Dahl, E, Fogh-Andersen, P (1987). Mai-occlusion in complete unilateral and bilateral cleft lip and palate. The results of a change in the surgical procedure. Scand J Plast Reconstr Surg Hand Surg 21: 81–85.
Moss, ML (1997a). The functional matrix hypothesis revisited. 3. The genomic thesis. Am J Orthod Dentofacial Orthop 112: 338–342.
Moss, ML (1997b). The functional matrix hypothesis revisited. 4. The epigenetic antithesis and the resolving synthesis. Am J Orthod Dentofacial Orthop 112: 410–417.
Mossey, PA, Arngrimsson, R, McColl, J, et al. (1998a). Prediction of liability to orofacial clefting using genetic and craniofacial data from parents. J Med Genet 35: 371–378.
Mossey, PA, McColl, J, O'Hara, M (1998b). Cephalometric features in the parents of children with orofacial clefting. Br J Oral Maxillofac Surg 36: 202–212.
Mossey, PA, McColl, JH, Stirrups, DR (1997). Differentiation between cleft lip with or without cleft palate and isolated cleft palate using parental cephalometric parameters. Cleft Palate Craniofac J 34: 27–35.
Motohashi, N, Kuroda, T, Capelozza Filho, L, Freitas, JA (1994). P-A cephalometric analysis of non-operated adult cleft lip and palate. Cleft Palate Craniofac J 31: 193–200.
Moyers, RE, Bookstein, FL (1979). The inappropriateness of conventional cephalometrics. Am J Orthod 75: 599–617.
Nakamura, S, Savara, BS, Thomas, DR (1972). Facial growth of children with cleft lip and/or palate. Cleft Palate J 9: 119–131.
Nakasima, A, Ichinose, M (1983). Characteristics of craniofacial structures of parents of children with cleft lip and/or palate. Am J Orthod 84: 140–146.
Nakasima, A, Ichinose, M (1984). Size of the cranium in parents and their children with cleft lip. Cleft Palate J 21: 193–203.
Narula, JK, Ross, RB (1970). Facial growth in children with complete bilateral cleft lip and palate. Cleft Palate J 7: 239–248.
Nordén, E, Linder-Aronson, S, Stenberg, T (1973). The deciduous dentition after only primary surgical operations for clefts of the lip, jaw, and palate. Am J Orthod 63: 229–236.
Ortiz-Monasterio, F, Afonso Serrano, R, Gustavo Barrera, P, et al. (1966). A study of untreated adult cleft palate patients. Plast Reconstr Surg 38: 36–41.
Ortiz-Monasterio, F, Rebiel, A, Valderrama, M, Crux, R (1959). Cephalometric measurements in adult patients with non-operated cleft palate. Plast Reconstr Surg 24: 53–61.
Peltonen, L, McKusick, V (2001). Dissecting human diseases in the postgenomic era. Science 291: 1224–1229.
Peterka, M (1979). Prenatal Growth and Damage of the Orofacial Complex [in Czech]. Prague: Czechoslovak Academy of Sciences. Dissertation.
Prochazkova, J, Tolarová, M (1986). Craniofacial morphological features in parents of children with isolated cleft palate. Acta Chir Plast 28: 194–204.
Prochazkova, J, Vinsova, J (1995). Craniofacial morphology as a marker of predisposition to isolated cleft palate. J Craniofac Genet Dev Biol 25: 162–168.
Pruzansky, S, Aduss, H (1967). Prevalence of arch collapse and malocclusion in complete unilateral cleft lip and palate. Rep Congr Eur Orthod Soc 365–382.
Raghavan, R, Sidhu, SS, Kharbanda, OP (1994). Craniofacial pattern of parents of children having cleft lip and/or cleft palate anomaly. Angle Orthod 64: 137–144.
Ras, F, Habets, LL, van Ginkel, FC, Prahl-Andersen, B (1994a). Three-dimensional evaluation of facial asymmetry in cleft lip and palate. Cleft Palate Craniofac J 31: 116–21.
Ras, F, Habets, LL, van Ginkel, FC, Prahl-Andersen, B (1994b). Facial left-right dominance in cleft lip and palate: three-dimension evaluation. Cleft Palate Craniofac J 31: 461–465.
Rohrich, RJ, Byrd, HS (1990). Optimal timing of cleft palate closure. Speech, facial growth, and hearing considerations. Clin Plast Surg 17: 27–36.
Ross, RB (1987). Treatment variables affecting facial growth in complete unilateral cleft lip and palate. Cleft Palate J 24: 3–89.
Ross, RB, Coupe, TB (1965). Craniofacial morphology in six pairs of monozygotic twins discordant for cleft lip and palate. J Can Dent Assoc 31: 149–157.
Ross, RB, Johnston, MC (1972). Cleft Lip and Palate. Baltimore: Williams 6c Wilkins.
Saksena, SS, Bixler, D, Yu, P (1990). A Clinical Atlas of Roentgencephalometry in Norma Frontalis. New York: Alan R. Liss.
Saksena, SS, Walker, GF, Bixler, D, Yu, P (1987). A Clinical Atlas of Roentgencephalometry in Norma Lateralis. New York: Alan R. Liss.
Sammons, E (1999). Cephalometric Similarity among Parents of Individuals with Sporadic Isolated Cleft Palate: Is There Evidence for an Inherited Predisposition? Indianapolis: Indiana University School of Dentistry. Dissertation.
Sandham, A, Murray, JA (1993). Nasal septal deformity in unilateral cleft lip and palate. Cleft Palate Craniofac J 30: 222–226.
Sassouni, V (1962). The Face in Five Dimensions, 2nd ed. Morgan-town: West Virginia University Press.
Savara, BS, Miller, SH, Demuth, RJ, Kawamoto, HK (1985). Biostereometrics and computer graphics for patients with craniofacial malformation: diagnosis and treatment planning. Plast Reconstr Surg 75: 495.
Schultes, G, Gaggl, A, Karcher, H (2000). A comparison of growth impairment and orthodontic results in adult patients with clefts of palate and unilateral clefts of lip, palate and alveolus. Br J Oral Maxillofac Surg 38: 26–32.
Semb, G (1991a). A study of facial growth in patients with unilateral cleft lip and palate treated by the Oslo CLP Team. Cleft Palate Craniofac J 28: 1–21.
Semb, G (1991b). A study of facial growth in patients with bilateral cleft lip and palate treated by the Oslo CLP Team. Cleft Palate Craniofac J 28: 22–39.
Shibasaki, Y, Ross, RB (1969). Facial growth in children with isolated cleft palate. Cleft Palate J 6: 290–302.
Shprintzen, RJ, Siegel-Sadewitz, VL, Amato, J, Goldberg, RB (1985). Anomalies associated with cleft lip, cleft palate, or both. Am J Med Genet 20: 585–595.
Šmahel, Z (1984a). Variations in craniofacial morphology with severity of isolated cleft palate. Cleft Palate J 21: 140–158.
Šmahel, Z (1984b). Craniofacial morphology in adults with bilateral complete cleft lip and palate. Cleft Palate J 21: 159–169.
Šmahel, Z (1984c). Cephalometric and morphologic facial changes in adults with complete bilateral cleft lip and palate. Acta Chir Plast 26.- 200–215.
Šmahel, Z (1984d). Craniofacial changes in unilateral cleft lip in adults. Acta Chir Plast 26: 129–149.
Šmahel, Z, Brejcha, M (1983). Differences in craniofacial morphology between complete and incomplete unilateral cleft lip and palate in adults. Cleft Palate J 20: 113–127.
Šmahel, Z, Brousilova, M, Müllerová, Z (1987). Craniofacial morphology in isolated cleft palate prior to palatoplasty. Cleft Palate J 24: 200–208.
Šmahel, Z, Hradisky, D, Müllerová, Z (1999). Multivariate comparison of craniofacial morphology in different types of facial clefts. Acta Chir Plast 41: 59–65.
Šmahel, Z, Müllerová, Z (1986). Craniofacial morphology in unilateral cleft lip and palate prior to palatoplasty. Cleft Palate J 23:225-232.
Šmahel, Z, Pobisova, Z, Figalova, P (1985). Basic cephalometric facial characteristics in cleft lip and/or cleft palate prior to the first surgical repair. Acta Chir Plast 27: 131–144.
Steinhauser, EW, Rudzki-Janson, JM (1994). Kieferorthopadische Chirurgie: Eine interdisziplindre Aufgabe. Berlin: Quintessenz Verlag.
Subtelny, JD (1955). Width of the nasopharynx and related anatomic structures in normal and unoperated cleft palate children. Am J Orthod 41: 889–909.
Suzuki, A, Takenoshita, Y, Honda, Y, Matsuura, C (1999). Den tocraniofacial morphology in parents of children with cleft lip and/or palate. Cleft Palate Craniofac J 36: 131–138.
Tang, EL, So, LL (1992). Prevalence and severity of malocclusion in children with cleft lip and/or palate in Hong Kong. Cleft Palate Craniofac J 29: 287–291.
Thompson, DW (1917). On Growth and Form. Cambridge: Cam bridge University Press.
Trasler, DG (1968). Pathogenesis of cleft lip and its relation to embryonic face shape in A-J and C57BL mice. Teratology 1: 33–49.
Trotman, CA, Collett, AR, McNamara, JA, Jr, Cohen, SR (1993). Analyses of craniofacial and dental morphology in monozygotic twins discordant for cleft lip and unilateral cleft lip and palate. Angle Orthod 63: 135–139.
Trotman, CA, Papillon, F, Ross, RB, et al. (1997). A retrospective comparison of frontal facial dimensions in alveolar bone-grafted and nongrafted unilateral cleft lip and palate patients. Angle Orthod 67: 389–394.
Trotman, CA, Ross, RB (1993). Craniofacial growth in bilateral cleft lip and palate: ages six years to adulthood. Cleft Palate Craniofac J 30: 261–273.
Vegter, F, Mulder, JW, Hage, JJ (1997). Major residual deformities in cleft patients: a new anthropometric approach. Cleft Palate Craniofac J 34: 106–110.
Ward, RE, Bixler, D, Jamison, PL (1994). Cephalometric evidence for a dominantly inherited predisposition to cleft lip-cleft palate in a single large kindred. Am J Med Genet 50: 57–63.
Ward, RE, Bixler, D, Raywood, ER (1989). A study of cephalometric features in cleft lip-cleft palate families. I: Phenotypic heterogeneity and genetic predisposition in parents of sporadic cases. Cleft Palate J 26: 318–326.
Wepner, F, Hollmann, K (1975). Mid-face anthropometry on the cephalometric radiograph in cleft lip and palate cases. J Maxillofac Surg 3: 188–197.