INTRODUCTION

Craniofacial growth anomalies are often accompanied by different patterns of mandibular growth that are generally described as hyperdivergent, normodivergent, and hypodivergent.1–3 The hyperdivergent pattern, or long face, is typically associated with a decreased posterior to anterior facial height ratio, an increased lower facial height, and a steep mandibular plane.4 The hypodivergent pattern, or short face, exhibits opposite features to the hyperdivergent pattern.5 The clinical features of these growth patterns are well described in the literature and often manifest as either a skeletal open or deep bite.6–9

It has been suggested that head form is closely associated with both vertical and transverse growth patterns.10,11 Dolichocephalic head forms have been described as having a narrow face and hyperdivergent mandible, while brachycephalic skulls exhibit a broader face and hypodivergent mandible.12 These stereotypic descriptions of head form suggest a close association between skeletal divergence and vertical and transverse facial growth. It is noteworthy, however, that these classical descriptions are mainly based on the two-dimensional analysis of lateral cephalograms.10,11 These techniques have some important drawbacks such as a high incidence of artifacts and errors and the lack of a third dimension.13 Moreover, the type and severity of different skeletal patterns are often masked by the facial soft tissues. Until the end of the 1950s, a common perception was that the integumental profile followed passively the underlying hard tissue, although later studies demonstrated that the soft tissues have an independent growth potential.14,15

Recent advances in biomedical imaging have made it possible to assess the facial hard and soft tissues in three dimensions (3D). Cone-beam computed tomography is a useful method for the simultaneous evaluation of both the soft and hard tissues but is often not suitable for retrospective samples because of the unnecessary dose of radiation.16 An alternative method to capture this information involves the combined use of pretreatment head films to evaluate the hard tissues and 3D scanners to assess the soft tissues. The latter method has the advantages of being noninvasive, free of radiation, and able to be used on large numbers of healthy people.17

The aim of this study was therefore to evaluate the relationship between mandibular divergence and vertical and transverse dimensions of the face through the combined use of lateral cephalograms and a white-light structured 3D scanner.

MATERIALS AND METHODS

The study was designed to detect a large effect size, as only major differences in craniofacial width between different jaw divergence groups were considered to be clinically relevant. To detect an effect size ≥0.8, with α  =  .05 and β  =  .80 (one-tail test), it was found that at least 20 participants per group were needed. Sixty participants were therefore recruited from the orthodontic clinic of the University of Otago. The sample consisted of 31 females and 29 males of New Zealand European origin, with a mean age of 21.1 ± 5.0 years (Table 1). Participants were included in the study if they were willing to participate, had a good-quality cephalogram, and provided informed consent. Exclusion criteria included having greater than four missing permanent teeth (excluding third molars), inflammatory or degenerative diseases of the temporomandibular joint, cleft lip and/or palate, craniofacial syndromes, facial asymmetry, ongoing orthodontic treatment, or a history of facial fractures or orthognathic surgery. The study was approved by the University of Otago Ethics Committee (12/054).

Table 1. Descriptive Statistics of the Sample

View Fullscreen

Statistical Analysis

The facial measurements used in the present study consisted of three vertical and four transverse variables (Figure 1). Descriptive statistics were computed using the Statistical Package for Social Sciences (SPSS version 20.0, Chicago, Ill). Univariate and multivariate analysis of variance were used to test for differences between groups after adjusting for age, gender, and BMI. Similar models unadjusted for BMI and height (data not shown) were also produced. Bonferroni correction was carried out for post hoc tests.

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

A descriptive summary of the study sample is presented in Table 1. There were no significant differences in BMI for either the jaw divergence group (P  =  .178) or gender (P  =  .768).

In the 3D soft tissue analysis (Table 2), the total facial height (NaMe) was significantly larger in the hyperdivergent group (124.6 ± 7.6 mm) than in the hypodivergent (115.6 ± 8.5 mm; P < .001) and normodivergent groups (120.4 ± 6.4 mm; P  =  .019). The hyperdivergent group also had an increased lower anterior facial height (SnMe  =  76.5 ± 5.4 mm) in comparison with both hypodivergent (68.6 ± 6.9 mm; P < .001) and normodivergent groups (72.0 ± 6.0 mm; P  =  .007). On the other hand, middle facial height (NaSn) was very similar in all three groups (P > .05).

Table 2. Soft Tissue Dimensions in the Vertical and Transversal Planes by Mandibular Divergence (Mean ± Standard Deviation) in Millimeters

View Fullscreen

With respect to the transverse dimension, there were no significant differences among the three groups except for mandibular width (GoGo), which was slightly larger in hyperdivergent (114.3 ± 8.2 mm) than hypodivergent individuals (112.8 ± 9.9 mm; P  =  .032). As expected, males had significantly larger vertical and transverse measurements than females (Table 3). All of the facial measurements were significantly associated with gender, even after adjusting for BMI (P < .05).

Table 3. Soft Tissue Dimensions in the Vertical and Transversal Planes by Sex (Mean ± SD) in Millimeters

View Fullscreen

Examples of hyperdivergent and hypodivergent facial types are given in Figure 2.

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

Several important findings were identified from the comparison between mandibular divergence and facial soft tissue dimensions. In particular, the three study groups differed significantly with respect to the vertical dimension but not for any of the four transverse measurements investigated.

Mandibular width, however, was slightly increased in the hyperdivergent group after adjusting for BMI, although this was not considered to be clinically significant (≅1.5 mm). Nonetheless, this finding was somewhat surprising since hypodivergent facial types are generally considered as having broader faces.12 One plausible explanation for this misconception is that a significant increase in BMI can sometimes obscure the underlying vertical morphology. In other words, hyperdivergent individuals may appear to have a flat mandibular plane that resembles a hypodivergent head form because of the overlying fat tissue. Indeed, overweight/obese people accumulate fat in the cheek areas22 but not around soft tissue menton, subnasion, and nasion.23,24 Increased cheek thickness may affect the correct localization of soft tissue gonion,19 thus resulting in an overall appearance of a square face (ie, hypodivergent).

In general, the findings of the present study did not support the notion that the transverse dimension of the face is associated with vertical facial features. Although hyperdivergent individuals have previously been reported to have a low width to height ratio of the underlying skeletal structures of the head,11 the data collected on soft tissue dimensions suggested that this is mainly because of an increase in facial height rather than width. A recent study on facial dimensions also reported a similar transverse dimension in individuals with different cephalic indices.25 In contrast to the current findings, vertical measurements in that study were also similar between the various cephalic groups. It is noteworthy, however, that the age group of the two study samples differed, and this may have accounted for the differences in vertical dimension. Since transverse growth ceases relatively early in life,26,27 it is not surprising that both studies identified a difference in this dimension irrespective of the sample's age. Vertical facial development, on the other hand, is often the last dimension to cease growth27,28 and is therefore more likely to exhibit greater variation in older samples such as the present one.

Sexual dimorphism was also evident for all of the vertical and transverse measurements in the present sample. In general, females had a narrower face than males, especially in the lower third of the face (ie, GoGo). Males, however, exhibited larger vertical measurements than females. These findings are consistent with a previous anthropometric study.29

The current findings provide data describing the vertical and transverse features of the face in individuals with different jaw divergence. This information could be helpful in the treatment planning of orthognathic surgery cases presenting with either hyperdivergent or hypodivergent jaw patterns. Indeed, both mandibular advancement and setback procedures have a direct effect on the transverse dimension of the face, especially the intergonial distance.30,31

The present study has a number of limitations that should be noted. Some of the landmarks were difficult to localize on the facial scans through visual inspection alone (eg, Gonion, Menton, Tragion, and Nasion). In general, landmarks located on smooth surfaces are often difficult to identify on 3D images without the use of reference markers.32,33 To improve landmark identification, some of the more difficult landmarks were directly labeled with round markers.33,34 Nonetheless, it is possible that some of the unmarked landmarks were associated with a higher degree of measurement error. Second, patients were instructed to look straight ahead while recording the scans, and this was assumed to replicate the NHP. A consistent head posture is particularly important in this type of study, since stretching of the soft tissues can occur as a result of postural changes.35 Although no specific aids (eg, mirrors) were used to position the head, it would be expected that most participants would have exhibited a NHP since this is a highly reproducible head posture.36,37 Finally, the findings are restricted to the soft tissues of the face and may not necessarily be valid for making assumptions about the underlying skeletal pattern.

On the other hand, stereophotogrammetric techniques have a number of important advantages, especially in these types of studies. The low cost, lack of radiation, and quick exposure time permit it to be used in large clinical and nonclinical samples. Moreover, these new imaging techniques allow researchers to evaluate multiple planes, both cross-sectionally and longitudinally. Finally, the present study demonstrated that this 3D imaging technique is both valid and reliable for evaluating facial form. Future research in the field of craniofacial growth and development may therefore greatly benefit from this simple and noninvasive imaging technique.

CONCLUSIONS

• The soft tissue facial form of hyperdivergent individuals is associated with a long face pattern but not necessarily a narrow one. Variation in the transverse dimension is more likely to be due to gender and BMI.

• The relatively simple and noninvasive 3D imaging technique used in this study has been shown to be reliable. Findings from this study support its use for investigating facial form in larger samples.