INTRODUCTION

The relationship between cranial base configuration and facial prognathism has been of interest to anthropologists, particularly in relation to racial differences. Huxley1 used the basicranial axis on sagittal sections of dried skulls to elucidate racial variation, whereas Young2 was one of the first researchers to suggest the possibility of an association between this variable and malocclusion. Bjork,3 using cephalometric radiographs, demonstrated the existence of a relationship between cranial base morphology and jaw relationship.

The maxilla and mandible articulate with different limbs of the cranial base, and therefore it is possible that variations in growth and orientation of the cranial base region could lead to a differential movement of the mandible in relation to the maxilla.

MATERIALS AND METHODS

Clinical cephalometric records from the orthodontic department of King's College Hospital in London, taken in centric relation, were surveyed and classified by observation into the four categories of malocclusion on the basis of the British Standards Institute incisor classification22:

• Class I: The lower incisor edges lie on or below the cingulum plateau on the palatal surface of the upper incisors with a normal overjet.

• Class II, Division 1: The lower incisor edges lie palatal to the cingulum plateau of the upper incisors, and the upper incisors are proclined or of normal inclination with an increased overjet.

• Class II, Division 2: The lower incisor edges lie palatal to the cingulum plateau of the upper incisors with the upper incisors being retroclined. The overjet is usually minimal but may be increased.

• Class III: The lower incisor edges lie labial to the cingulum plateau of the upper incisors.

The British Standards Institute incisor classification represents a measure of the antero-posterior arch discrepancy and is, therefore, analogous to Angle's molar classification. Incisor classification has been used previously to subdivide samples on the basis of malocclusion.1021

A total of 200 cephalometric radiographs of Caucasian patients (50 for each malocclusion group, with an age range between 8–12 years) were selected for the study. Records were matched for age and sex as far as possible. Each group contained approximately similar numbers of males and females, and the mean ages for each group were 10.4 years for the class I group; 10.10 years for class II, division 1; 11.1 years for class II, division 2; and 10.2 years for the class III group.

Cephalometric data were collected and stored for analysis by the method of direct on-line digitization as described by Bhatia.24 Twenty-five landmarks (Figure 1) from each radiograph, as defined by Bhatia and Leighton,5 were digitized twice and the means of the two sets of coordinates were used for computation of the results. The computer program was designed to highlight any points with a discrepancy of greater than 1 mm between the two recordings in either the horizontal or vertical plane. Such discrepant points were redigitized if it was felt that the cause was likely to be operator error. From the coordinate means a number of linear and angular variables were calculated:

• Cranial base flexure: N–S–Ba, N–S–Art;

• Jaw position: SNA, SNB;

• Skeletal pattern: ANB, maxillary mandibular planes angle (MMA) measured as the angle between the ANS–PNS and Me–Go planes;

• Dento–alveolar pattern: Imx (upper incisors to maxillary plane angle), Imn (lower incisors to mandibular plane angle), I/I (interincisal angle);

• Cranial base lengths: N–S, S–Ba;

• Jaw lengths: Cd–ANS, Cd–Pog, Art–ANS, Art–Pog, ANS–PNS, Me–Go;

• ANS–PNS: Me–Go.

View Figure

• Download as Powerpoint
• Download Figure

The error of the method was estimated by using the Dahlberg formula:

Errors were comparable to and showed similar trends to other cephalometric studies (Gravely and Benzies25).

The between group variability was investigated using a one-way analysis of variance (ANOVA). The group variable means for the class II and III groups were compared with the class I group by means of an independent t-test. Because of the large number of comparisons and to reduce the probability that some mean differences might be significant by chance alone, the minimum probability level for accepting significance between the group means was set at 1%. Intervariable correlations were also compiled. For clarity, only relevant correlations are presented.

RESULTS

The results are presented in Tables 1–8. First, it was necessary to demonstrate that the data for each variable showed significant variance in the four malocclusion groups so as not to invalidate comparisons between individual malocclusion groups. This was done using a one-way ANOVA of relevant groups (Table 1). Tables 2 and 3 show values for all variables and also between-group comparisons of variables according to independent t-tests. Table 4 shows the salient and significant intervariable correlations compiled using pooled data from all four malocclusions. Tables 5–8 show intervariable correlations for each malocclusion group.

TABLE 1. Results of One-way ANOVA of Relevant Groups

View Fullscreen

TABLE 2. Comparison of Skeletal and Dental Variables Between the Class I and Class II Sample Groups

View Fullscreen

TABLE 3. Comparison of Skeletal and Dental Variables Between the Class I and Class III Sample Groups

View Fullscreen

TABLE 4. Correlation Coefficients With Pooled Group Data

View Fullscreen

TABLE 5. Class I Correlation Coefficients

View Fullscreen

DISCUSSION

The sample consisted of 200 cephalometric radiographs selected retrospectively on the basis of the observed incisor occlusion. The class I sample showed good agreement with published cephalometric norms for both dental and skeletal base relationships.26 Interestingly these figures also agree very closely with those recently published by Hamdan and Rock27 who calculated the mean of means for major cephalometric variables using data from 14 prominent studies. The sample, therefore, represents a valid reference group for the purposes of this study.

Data for the other malocclusion groups showed the expected variations in terms of dentoalveolar and skeletal base patterns with the skeletal pattern for each group matching the incisor relationship. The dental angular variations were as would be expected in order to compensate for the underlying skeletal discrepancies in the respective classes of malocclusion. For example, when compared with class I values the lower incisors were retroclined in class III subjects.

Although the N–S length is usually used as a measure of the anterior cranial base, there is some disagreement as to whether the posterior base should be measured from basion or articulare. Bjork3 advocated the use of articulare, rather than basion, because it is easier to identify. Subsequently other studies have used articulare to define the posterior limit of the cranial base.7

Varjanne and Koski28 have discouraged the use of articulare because of its remoteness from the cranial base and suggested basion as the more appropriate choice because of its anatomic significance despite potential difficulties in identification. Kerr and Adams10 used basion to measure the cranial base angle. Bhatia and Leighton5 have published figures for N–S–Ba and N–S–Art angles as well as the S–Ba and S–Art distances. Interestingly, they found the growth patterns, as described by use of basion or articulare, to be very similar.

The present results, using both measures of the posterior cranial base, do not support the concept that the cranial base angle, by providing a variation in the antero-posterior position of the mandibular articulation, is a major determinant in establishing the main classes of malocclusion.7 Indeed, only the class II division 1 group showed a significant difference in parameters N–S–Ba and N–S–Art in comparison to class I. Therefore, it is not possible to corroborate the assertion of Dibbets12 that the three Angle classes II, I, and III represent arbitrary markers on a morphological continuum.

Enlow29 has shown growth of the maxilla to be under the influence of the cranial base, which in turn is influenced by growth of the brain. The mandible, by virtue of its remoteness from the region, acts in a more independent way although its articulation at the glenoid fossa does provide potential for influence from the cranial base.

It should be noted that the temporo-mandibular joint is positioned at the lateral edges of the cranial base and is, in fact, considerably separated spatially from the midsagittal plane on which cephalometric analyses are based. It is likely, therefore, that changes in the cranial base angle may not be directly translated to the mandibular articulation. The trends found in the present study with regard to the maxillary and mandibular skeletal parameters agree with those reported by other researchers.2021

The finding from the pooled sample that cranial base angle was correlated to the angles SNA and SNB is in agreement with those of Bjork3 and Kasai et al,30 who demonstrated a relationship between the cranial base angle and facial prognathism.

The correlation analysis also suggests a relationship between mandibular position and the magnitude of cranial base flexure. The smaller the cranial base angle, the more forward the mandibular position, as indicated by angle SNB. However, it is impossible to establish cause or effect from these results since other factors may also be involved. The maxilla is equally affected by the angle, which suggests a more fundamental role for the cranial base in determining facial prognathism than that indicated by mere geometry. When interpreting the present results, it is also important to realize that angular correlations, which share a line or point, are not devoid of topographical influence.

Little correlation was evident between angle ANB and cranial base parameters when looking at pooled or individual group data (Tables 4–8). This suggests that there is no direct relationship between the cranial base and class of malocclusion. The positive and significant correlations between S–Ba and N–S distances with maxillary and mandibular lengths are likely to be topographical and of little biological significance. The results, however, do suggest a link between jaw length and facial prognathism. This relationship was most pronounced in class III subjects where a longer mandible was associated with increased mandibular prognathism.

Wide fluctuations in the magnitude of the cranial base angle have been remarked upon by a number of authors31 and make it difficult to interpret some of the present results. Differences between these results and those of other workers may be related, in part, to differences in case selection procedures. The cases for this study were selected on the basis of the unmodified British Standards Institute incisor classification.22 Kerr and Adams,10 for example, selected cases with more definite malocclusion, ie, class II division 1 cases with overjet ≥10 mm and class III subjects with maxillary incisors in lingual occlusion. It may well be that cranial base morphology has a more prominent role in establishing malocclusions at the extremes of the scale.

CONCLUSIONS

From the present study the following conclusions may be drawn:

1. The cranial base angle alone does not appear to play a pivotal part in the establishment of malocclusion.

2. Jaw lengths are significantly different between the malocclusion groups. The maxillary length is increased in class II malocclusions and the mandibular length is greater in class III.