A common and fundamental objective in Class II orthodontic correction is the establishment of a Class I molar relationship.1,2 To this end, it is important to identify whether higher occlusal contacts are achieved more with a Class I than Class II molar relationship. Furthermore, it is essential to determine whether distal displacement of molars within the maxilla can be achieved by the Class II correction strategy, or whether premolar extraction is necessary.

Molar relationships are generally analyzed from the buccal, rather than the lingual, aspect, because the latter is difficult to do without splitting dental casts. However, American Board of Orthodontics (ABO) guidelines have emphasized the importance of occlusal contact of the molar and premolar lingual cusps on the occlusal surface of the opposing teeth.2 In the classical text on tooth form, for instance, Russell3 demonstrated how important it is for the palatal cusp of the upper molar to be seated in the central fossa of the lower first molar. The position of the palatal cusp of the maxillary first molar was recently emphasized by Liu and Melsen,1 who performed examinations from both the buccal and lingual aspects when assessing Class II molar relationships. In order to avoid splitting study casts that had been made of the lingual surfaces of the first molars, the authors placed a dental mirror at a 45-degree angle to each cast when classifying relationships.

In general, orthodontic diagnosis of occlusion is mainly morphological and static, and is not directly related to normal function or functional disorders. Measuring the occlusal contact areas may serve as a pre- and post-treatment record of the functional relationship of occlusion, and, further, may facilitate the assessment of functional changes following active orthodontic treatment. In turn, a more differentiated diagnosis of molar relationships may help identify the true nature of the malocclusion, thus enabling the application of more suitable treatment approaches.

Until now, few studies have evaluated the occlusal contact areas of different levels of Class II molar relationships nor compared these with Class I molar relationships. Therefore, this study was performed with 2 main goals: firstly, to evaluate the consistency of molar relationship assessments performed from both the buccal and lingual aspects; and, secondly, to compare the occlusal contact areas of different severities of Class II and Class I molar relationships in 3-dimensional (3D) models.

We used 3D modeling to evaluate the consistency of buccal and lingual analyses of molar relationships. In addition to revealing that this categorization technique is associated with high intra-examiner reliability (p > 0.05), our results indicate that classifications derived from buccal and lingual molar aspect analyses are consistent only when the upper and lower first molars are in a Class I relationship (189 sides, 89.6%). We have considered lingual 0, 1, 2, and 3 as mild lingual class III, lingual Class I, mild, and severe lingual Class II, respectively.

Conflicting buccal and lingual categorizations were made for 61.5% (104 of 169 molars) of molar relationships determined, from buccal analyses, Class II. This level of buccal-lingual consistency (e.g., in only 38.5% of cases) was markedly lower than that found in molar Class I (89.6% of cases). Surprisingly, the majority (86.7%) of cases categorized as Class II from buccal data were determined to be Class I when observed from the lingual aspect (lingual 1). In other words, only a small percentage of the models reported as having a Class II molar relationship were truly Class II (lingual 2, 3). Indeed, cases classified as Class II from the lingual aspect (lingual 2, 3) always tended to be less severe than their buccal counterparts at each sublevel (Table 1).

These patterns result from the anatomical features of the upper and lower first molars in occlusion. Specifically, the lower first molar erupts into dentition prior to its upper counterpart, thereby facilitating "cusp-to-fossa" occlusion. Three fossae - the mesial triangular, central, and distal triangular - are functionally available on the occlusal surfaces of the lower molars to occlude with the upper palatal cusps.3 In the majority (89.9%) of Class II cases examined here, the upper mesiopalatal cusp was seated in either the mesial triangular (28.4%) or central (61.5%) fossae.

Molar rotation6-10 is another common cause of inconsistencies in buccal and lingual classifications of molar relationships.1 For example, Braun et al.7 reported that rotation of the upper first permanent molar existed in 85% of all orthodontic conditions, 90% of all Class II division 1 cases, and 100% of all maxillary constrictions studied. Because it is triple-rooted and has a trapezoidal configuration, the upper first molar is more prone to rotation than its lower counterpart.

Our findings suggest that the upper mesiopalatal cusp functions as a pivotal axis around which the upper first molar rotates. This cusp occludes most frequently in the central fossa, less frequently in the mesial triangular fossa, and rarely anterior to the mesial triangular fossa. The mild buccal Class II molar relationship is clinically referred to as "half cusp Class II"; among our study group, a large portion of these cases were more comparable to buccal Class I than severe buccal Class II (or "full cusp Class II") when analyzed from the lingual perspective. Given that the palatal cusp is the functional cusp of the upper molar, this implies that the functional properties of half cusp Class II are more similar to buccal Class I - a hypothesis that was partially verified by the occlusal contact areas measured in our samples.

Functional evaluation of occlusion is commonly based on measurements of occlusal contact areas.11,12 For instance, Luke and Lucas13 reported that masticatory performance is highly correlated with the occlusal area of teeth posterior to the canine. An even more important factor controlling the masticatory performance of human subjects with natural teeth is the amount of occlusal contact area between the molar and premolar teeth, which is on average, one-fifth of the total occlusal surfaces.14 Therefore, in this study, the occlusal contact area was used as a parameter for comparing the occlusion of Class I and Class II molar relationships.

Specifically, we found that mean occlusal contact area was significantly higher in Class I cases than in Class II cases (79.77 ± 11.62 mm² vs. 62.58 ± 11.09 mm² for Class II-s and 36.99 ± 13.95 mm² for Class II-m) (Table 2). Although differences between Class I and mild Class II values were not significant, there were significant differences between severe Class II values and those of other groups.

In our sample, severe Class II cases exhibited good interdigitation from the buccal aspect, but had the smallest occlusal contact areas. Furthermore, these findings were significantly different from those measured in Class I cases. In contrast, mild Class II cases had poor interdigitation from the buccal aspect, but occlusal contact areas were similar to those documented in Class I cases. Moreover, these values were significantly higher for mild Class II molar relationships than for severe Class II relationships (Table 2). Thus, we conclude that functional occlusions are better for mild buccal Class II cases than severe Class II cases.

Our measurements of mean occlusal contact areas also highlighted the importance of the cusp-to-central fossa relationships (lingual 1). The mean value (73.51 ± 8.37 mm²) was associated with significantly (p < 0.05) greater mean occlusal contact area than was measured for any other lingual relationship except cusp-to-distal triangular fossa (lingual 0). However, the differences between cusp-to-mesial triangular fossa (lingual 2) and cusp-to-marginal ridge relationships (lingual 3) were not statistically significant (p > 0.05). In general, our findings were in accordance with those previously reported by others. For instance, in a study on associations between occlusal parameters and comminution of solid food during chewing, Sierpinska et al.15 found that the mean area of the chewing platform was 125.12 ± 46.5 mm² in subjects with full dentitions (in 25 completely dentate subjects; 28 teeth, 14 functional dental units).

When considering the results of research on occlusal contact areas, it is important to take note of the methods used to quantify this variable, and whether they are appropriate for each particular analysis. Contact area has frequently been measured with T-Scan II equipment (Tekscan Inc., Boston, MA, USA),16,17 which is a reliable method for the analysis and evaluation of occlusal contact distribution in maximum intercuspation. This value is calculated by multiplying the number of tooth contacts recorded by the T-Scan sensor by the size of a single pixel (1.6 mm²), thus reflecting occlusal conditions. However, one limitation of the system is that values are recorded during real-time occlusal contacts generated by occlusal forces in the oral cavity. Both maximum occlusal force (N) and occlusal contact area (mm²) have also been evaluated using horseshoe-shaped pressure-sensitive films (Dental Prescale 50H, R type; Fujifilm, Tokyo, Japan), which exhibit color variation and yield an area dependent on the amount of applied pressure within a range of 5 - 120 MPa. The system then automatically estimates the occlusal pressure and contact area by measuring color density and area data. However, because this method is associated with variations in occlusal pressure, all subjects must cooperate if records are to be obtained in a uniform fashion.18-20 Since the aim of the current study was to analyze the occlusal contact areas without taking occlusal forces into consideration, we felt it was most appropriate to use 3D digitalization, which is not associated with mastication components that are difficult to control.

Overall, the majority of cases categorized as mild Class II included lingual Class I (lingual 1) classifications, while the majority of cases categorized as severe Class II included lingual Class II (lingual 2, 3) classifications. Thus, from the buccal aspect, individuals with severe Class II seem to have better interdigitation than those with mild Class II, though the occlusal contact area of the former may not necessarily be greater than that of the latter. These differences between buccal and lingual classifications can have significant clinical impacts, as recently emphasized by Korn.21 For example, it is necessary to identify the category of the molar relationships during the treatment planning stage to correct a malposed upper first molar. Class I lingual molar relationships (lingual 1) can most likely be corrected by rotation around the lingual cusp. This displacement will generate sufficient space for the correction of the mild Class II tendency observed in the molar region. By contrast, molar relationships categorized as Class II (lingual 2, 3) from lingual aspect analyses are more likely to have a skeletal discrepancy and reflect a true discrepancy in the sagittal jaw relationship. A skeletal correction is required in such cases.

Although our statistical analyses revealed convincing, and consistent patterns, there were some limitations to our study. First, our strict inclusion criteria caused us to exclude many samples, particularly those belonging to the Class II-s group; as a result, we did not have even numbers of samples for each category. Secondly, deviation from mean values was sometimes considerable, possibly as a result of differences in tooth size or errors made in the process of creating impressions and casts; there may also have been some errors made during the 3D scanning procedure. In order to verify the accuracy of our findings, and to further our understanding of the occlusal contact between the upper and lower teeth and how this can be evaluated from both the buccal and lingual perspectives, we encourage future studies with larger sample sizes, measurements of dynamic occlusal force, and prospective study designs.

The classifications of buccal (particularly buccal Class II) and lingual molar relationships were not always consistent. A large portion of molars that were categorized buccally as Class II were categorized lingually as Class I. The smallest occlusal contact area was observed in cases of severe buccal Class II molar relationships, which could be inferred as having the worst molar relationship among these groups-especially from the lingual aspect.