The objective of this study was to verify, using periapical radiographs, whether a partially erupted mandibular third molar is a factor in the presence of dental caries on the distal surface of the adjacent second molar. Two-forty six high quality periapical radiographs were selected, each showing a partially erupted mandibular third molar. The variables analyzed were: tooth number; gender; age; radiographic presence of caries on the distal surface of the adjacent molar; Pell and Gregory classification; Winter classification; angulation and distance between the second and mandibular third molar. The examiners were previously calibrated to collect data (kappa statistics from 0.87 to 1.0). The prevalence rate of caries on the distal surface of the second molar was 13.4%. In the logistical multivariate regression analysis, the angulation (OR = 8.5; IC95%: 1.7–43.8; p = 0.011) and the gender (OR = 3.3; IC95%: 1.4–7.7; p = 0.005) remained statistically significant after an age adjustment was made. The results indicate that the presence of a partially erupted mandibular third molar with an angulation of 31 degrees or more, is a risk factor for caries on the distal surface of the mandibular second molars.
The prevalence of caries on the mandibular second molar due to the presence of a third, partially erupted molar, varies between 7% and 32%. Some studies have verified that the presence of caries on the distal surface of the mandibular second molar could be caused by the angulation of the partially mandibular third molar, the distance between the cementoenamel junction (CEJ), the level of impaction and the amount of contact between the second and third lower molar.
Studies linking caries on the distal surface of the mandibular second molar to a third molar have been carried out. These studies based their prevalence data on panoramic radiographs which they used to diagnose caries. This type of radiograph is adequate when planning extraction surgery for third molars but it is not as precise as periapical radiograph when diagnosing caries. The lack of sample characterization, the absence of sample calculation, the deficient or inadequate statistical analysis and the absence of a description of the eligibility criteria, discredits the scientific evidence of these previous studies. Research using more appropriate methodology is necessary to test the hypothesis.The objective of this study was to verify whether caries, when present on the distal surface of the mandibular second molars, is linked to the presence of partially erupted third molars and what radiographic characteristics are considered to be risk factors for this pathology.
Materials and methods
This was a sectional study to evaluate the link between a partially erupted mandibular third molar and the presence of caries on the distal surface of the second molar using periapical radiographs. A sample calculation was carried out for the selection of radiographs, which considered, as given, a prevalence rate of 20% and an error margin of 5%. Two-forty six radiographs were needed for the study. High quality periapical radiographs were included in the sample, providing good contrast and an adequate relationship between the second and third molars and the ramus of the mandible. Samples with inadequate contrast, exposure or setting were excluded, as were those with collimation, distortion, or overlapping enamel and restorations on the second molar, to facilitate the analysis of the variables. Radiographs which did not show the relationship between the mandibular ramus and the crown of the third molar, as well as third molars which were not semi-erupted, were also excluded. Radiograph selection was conducted using analysis of the medical records of patients, who had surgery to remove a third lower molar in the dental clinic referenced by the authors. Clinical records were assessed until the pre-determined total (sample calculation) of 246 radiographs was attained. Once this total of 246 radiographs was attained, clinical records analysis was terminated with a final total of 2734 clinical records assessed between the years 2010 to 1999. As the 246 radiographs that satisfied the inclusion criteria became the total of the sample, randomization was not necessary.
The dependent variable was the presence of caries on the distal surface of the second lower molar. The independent variables were: age; gender; Pell and Gregory classification ( Fig. 1 A and B ); distance from the CEJ of the third molar to CEJ of the second molar according to the method described by Leone et al. ; Winter’s classification; angulation of the mandibular third molar to the second molar. The team was made up of three previously calibrated examiners (kappa statistics from 0.87 to 1.00). The radiographs were evaluated using a light box. The first radiographic diagnosis was to establish the presence or lack of caries on the distal surface of the second molars ( Fig. 1 C). The radiographic characteristics of the third molars were collected. They were traced onto a sheet of tracing paper, superimposed on the radiograph and the light box using two lines, one touching the tips of the cusps of the second lower molar and the other the tips of the cusps of the mandibular third molar ( Fig. 1 D). The angle formed at the intersection of the two lines was evaluated using a protractor (Faber Castell ® ). The distance from the CEJ of the mandibular second molars distal to the CEJ of the mandibular third molar was measured in mm using a dry compass and a mm ruler (Faber Castell ® ) ( Fig. 1 E).
The data collected were typed and organized in a database, using Package for Social Science (SPSS) software, version 17.0. The process included coding, typing and editing. Frequency analysis was carried out and the variables were subjected to Pearson’s χ 2 -test to verify the association between variables. The continuous variables were categorized before using Pearson’s χ 2 -test. Angulation was categorized in tertiles and the distance between the CEJs was categorized at the median ( Table 1 ). The data were submitted for univariate and multivariate logistical regression analysis.
|Total||Caries on second molar distal surface||p *|
|n (%)||n (%)|
|Pell Gregory (Ramus)|
|Pell and Gregory (occlusal surface)|
|−32 to −1 degrees||78||73(93.6)||5(6.4)||<0.001|
|0 to 29 degrees||102||95(93.1)||7(6.9)|
|31 to 108 degrees||66||45(68.2)||21(31.8)|
246 radiographs were taken from the patient records. The average age of the patients was 24.17 years (range 16–57). One twenty-six of the radiographs (52.1%) were of the third left sided molar and 177 (72%) of female gender. The prevalence of distal caries on the mandibular second molars was 13.4%. With respect to the Pell and Gregory classification in relation to the mandibular ramus, 86 (35%) were given classification I, 123 (50%) classification II and 37 (15%) classification III. With respect to the Pell and Gregory classification in relation to the occlusal surface of the mandibular second molar, 180 (73.17%) were class A and 66 (26.83%) were class B. Regarding Winter classification, 13 (5.3%) of mandibular third molars were distoangular, 148 (60.2%) were vertical, 43 (17.5%) were mesioangular and 42 (17.1%) were horizontal. The average angulation between second and third molars was 16.44 degrees (range −32 to 108 degrees). The average distance between the CEJ of the second and mandibular third molars was 3.46 mm (range 1–10 mm). Pearson’s χ 2 -test was used to investigate the link between the presence of distal caries on the mandibular second molar and the independent variables. It was considered to be significant when p < 0.05 ( Table 1 ).
The univariate logistical regression analysis showed that male patients, and patients aged 23–57 years, are the most likely to be affected by distal caries on the second molar. With respect to radiographic characteristics, mandibular third molars with an angulation between 31 and 108 degrees, a distance of 3–10 mm to the second molar and those classified as horizontal or mesioangular, were most likely to be linked to distal caries on the second molar ( Table 2 ).