PDF
Citation
Print
INTRODUCTION
Extraction of the mandibular second molar (L7) can be effective in various situations—including ectopic eruption, severe dental caries, or posterior crowding—as long as the mandibular third molar (L8) is developing normally.1-10 This approach avoids a reduction in the number of functioning teeth.3 By extracting L7 and allowing L8 to erupt in that place, space is created for aligning other teeth in the arch, potentially reducing treatment duration and preventing future crowding in the lower jaw.1,11 However, L8 is known for possessing the most variable shape, size, and position among all teeth.5,6,12
Several studies have investigated the methods for predicting L8 eruption.2,3,6,13,14 Spontaneous eruption of L8 occurs in approximately 59–73% of young adults (age range, 14–26 years), with some variation based on race but not sex.15-17 The eruption and impaction of L8 are influenced by numerous factors, including skeletal growth, facial patterns, and space availability.13,14,18 A lack of adequate retromolar space (RMS) is a key factor contributing to L8 impaction.19 For normal eruption, the distance between the back of L7 and Xi point20 should ideally be at least 25 mm on a lateral cephalogram (CEPH) by adulthood.21
Successful spontaneous eruption of L8 with developing roots following L7 extraction have been reported, with eruption rates ranging 66.2–96.0% in good or acceptable positions.2-5,22-25 However, Asai et al.26 noted that 85% of erupted L8 required additional orthodontic treatment. Bishara et al.27 observed that L8 frequently erupt with poor angulation or inadequate contact with the mandibular first molar (L6), necessitating further “late” period orthodontic intervention to achieve good occlusion.
Definitions of eruption success and measurement methods to predict L8 eruption following L7 extraction are varied. Previous studies have employed diverse approaches to access L8 angulation, developmental stage, rotation, and occlusion using various tools: panoramic radiographs (PAN),2,25,28 dental casts,22,23,25,29 and CEPH analysis.5,21-23,29 These methods were used either individually2,5,21,28 or in combination.22,23,25,29 Additionally, some studies incorporated orthodontic treatment after L7 extraction,2,25,28,29 whereas others focused on spontaneous eruption.21,23,24
Despite numerous studies on the timing of L7 extraction for L8 development, differing opinions and conflicting findings exist in the field.4,29,30
While prior research explored L8 eruption using various criteria and methods, few studies objectively evaluated final occlusion. Furthermore, limitations existed in predicting the eruption of L8 before L7 extraction and in studying their spontaneous eruptions without orthodontic intervention. To determine the timing of L7 extraction, predictive factors such as the position, angulation, size, available space, and tooth development stage of L8 must be comprehensively considered. The objectives of this study were (1) to evaluate the final alignment (AL) of L8 after L7 extraction using the American Board of Orthodontics (ABO) index, a standardized evaluation method, and (2) to examine the factors affecting the successful eruption of L8 through the use of PAN, lateral CEPH, and dental study models.
MATERIALS AND METHODS
Participants
This study analyzed the records of patients treated at the Gangnam Severance Dental Hospital between 1998 and 2008. Ethical approval was granted by the Institutional Review Board of Gangnam Severance Hospital (Approval No. 3-2017-0096) before data collection commenced. The need for informed consent was waived by the Ethics Committee owing to the retrospective nature of the study.
This study included 46 L8 teeth from a sample of 28 patients (Table 1). Bilateral L7 extractions were performed on 18 patients, whereas the remaining 10 patients had undergone unilateral extractions. The reasons for L7 extraction included severe caries, malposition, severe rotation, or the need for distalization of the mandibular dentition. The inclusion criteria were as follows: (1) spontaneous eruption of L8; (2) no orthodontic treatment for L8; (3) regular follow-up visits; and (4) complete set of records with PAN, CEPH, and study models at the time of L7 extraction (T1) and completion of L8 eruption (T2). Spontaneous eruption was defined as exposure of L8 into the oral cavity without the use of orthodontic forces. T2 was defined as the stage at which L8 either established occlusal contact (OC) with opposing teeth or exhibited no observable positional change, confirmed through routine radiographic and clinical evaluations conducted at 6-month intervals.
The 46 L8 teeth were divided into two groups based on their occlusion at T2: acceptable (A-group) and unacceptable (U-group) categories. Following the ABO objective grading system,31 five criteria (AL, marginal ridge [MR], OC, interproximal contact [IC], and buccal overjet [BO]) were evaluated at T2. Each sample was classified into A-group only when the tooth satisfied all the five criteria (Supplementary Figures 1 and 2).
Measurements
The position, angulation, available space, and Nolla stage of L8 were evaluated on PAN or CEPH using V-ceph program (version 5.5; OSSTEM, Seoul, Korea), while the size of the mandibular molars were measured at T1 or T2 using dental study models. All the measurements were performed by a single examiner (YN).
Position and angulation of L8
Eleven landmarks were identified, on the PAN obtained at T1, to evaluate the position of L8 (Figure 1A). A reference line—known as the occlusal plane (OP)—was established, connecting the buccal cusp of the mandibular second premolar (Pmc) to the midpoint between the mesiobuccal and distobuccal cusp tips of the L6. The vertical position of L8 (V8-OP) was determined by measuring the perpendicular distance from the midpoint (8m) of the mesiobuccal and distobuccal cusp tips of L8 to the OP. The horizontal position of L8 (H8-6) was defined as the distance between the perpendicular points of the contour height on the mesial surface of L8 and distal surface of L6 (8Hc and 6Hc, respectively) relative to the OP.
The angulation of L8 was determined by defining its long axis as a line connecting the furcation area to the midpoint of the mesial and distal contour heights on CEPH. In cases where the furcation area had not yet developed, the midpoint between the mesial and distal points at the cervical areas was used instead.
Four angulation measurements were taken from each CEPH at both T1 and T2 (Figure 1B): ∠8-6, the angle between the long axes of L8 and L6; ∠8-vOP, the angle between the long axis of L8 and a line perpendicular to the OP; and ∠8-MnP and ∠6-MnP, the angles between the long axes of L8 and L6 and the mandibular plane (MnP), respectively.
Available space for L8 eruption
The available space for L8 eruption was evaluated on both PAN and CEPH at T1 and T2. On PAN, the RMS was measured as the distance from the perpendicular point of 6Hc to the OP to the ascending ramus along the OP (Figure 2A). On CEPH, the distance between the Xi point and 6Hc (Xi-6) was measured (Figure 2B). The Xi point represents the geometric center of the mandibular ramus, located at the center of a rectangle formed by R1—the deepest point on the subcoronoid incisure; R2—directly opposite R1 on the posterior border of the ramus; R3—the deepest point on the sigmoid notch; and R4—directly below R3 on the lower border of the ramus.20,32
Size of the mandibular molars
The mesiodistal widths of the L6 and L7 crowns were measured on the study model at T1, whereas that of the L8 crown was measured at T2 using a digital caliper (ABS Digimatic Caliper; Mitutoyo, Kawasaki, Japan).
Nolla stage of L8
The developmental stage of L8 at T1 was evaluated using the Nolla stage33 on PAN (Supplementary Figure 3). As the Nolla stages of 5 and 6 were hard to distinguish, they were grouped together.
Statistical analysis
Statistical analyses were conducted using SPSS version 22 (IBM Corp., Armonk, NY, USA) and R software version 4.3.2 (R Core Team, Vienna, Austria). Before data collection, a sample size calculation was performed using G*Power software (Heinrich-Heine-Universität Düsseldorf, Dusseldorf, Germany).34 This calculation aimed to determine the variables that might influence the occlusal state of L8, with an effect size of 0.2, power of 80%, and significance level (α) of 0.05, based on the findings from previous studies.1,3,24 This analysis indicated a requirement of a minimum sample size of 42, which was met by the present study (46 samples) and we performed post-hoc power analyses.35
To assess measurement consistency, a single examiner repeated measurements on 20 randomly selected samples after a 2-week interval. Intraclass coefficients were calculated to evaluate measurement reliability. The Shapiro–Wilk test was employed to evaluate the normality of the distribution. Descriptive statistics, such as means, standard deviations, medians, and interquartile ranges, were calculated for all variables. Additionally, differences between T1 and T2 were calculated for the measurements of L8 angulation and available spaces.
Independent t tests or Mann–Whitney U tests were conducted to compare continuous variables between the A- and U-group categories, with Bonferroni correction for multiple comparisons. Intersexual differences were analyzed using independent t tests, while chi-square or Fisher’s exact tests were utilized to examine the Nolla stage of L8 based on the occlusal state at T2. A logistic regression analysis was performed to determine the factors that positively affect the final occlusal outcome of L8. The significance level for all tests was established at P < 0.05.
RESULTS
Intraclass correlation coefficients varied from 0.910 to 0.977, reflecting a high degree of reliability. The ages of the study participants ranged 13.6–13.8 years at T1 and 17.9–18.5 years at T2. No statistically significant difference in age was found between males and females (P > 0.05; Table 1).
Out of the 46 L8 teeth evaluated, 27 (58.7%) exhibited adequate occlusion (A-group), while 19 (41.3%) exhibited unacceptable occlusion (U-group). The mean age at T1 was significantly higher in the U-group (14.3 ± 1.8 years) than that in the A-group (13.2 ± 1.3 years) (P = 0.033). The mean duration of L8 eruption (4.4–4.6 years) and mean age at T2 (17.9–18.7 years) did not differ significantly between the two groups (P > 0.05; Table 2).
No statistically significant differences were observed between the A- and U-groups in terms of L8 horizontal position, vertical position, or angulation at T1 (P > 0.05). At T2, ∠6-MnP and ∠8-MnP showed significant differences between the two groups (P < 0.01). These angles were larger in the A-group, indicating that both L6 and L8 were more upright and erupted more vertically compared with those in the U-group. During the observation period (ΔT2-T1), Δ∠6-MnP and Δ∠8-MnP were larger in the A-group than those in the U-group (P < 0.05), whereas Δ∠8-6 and Δ∠8-vOP showed no significant differences between the two groups (P > 0.05). While no significant difference in RMS was observed between the groups at T1 and T2, Xi-6 distance was significantly greater in the A-group than that in the U-group at both time points (P < 0.01).
There were no significant differences in the mesiodistal widths of L6, L7, and L8 between the two groups (P > 0.05; Table 3). Additionally, the Nolla stage of L8 at T1 did not differ significantly between the two groups (P > 0.05; Table 4).
In the U-group, the most frequently lacking criteria in the ABO index were AL, MR, and OC, each occurring in seven cases. Issues with BO and IC were less common, occurring in five and three cases, respectively. Nolla stages 5–7 were most often associated with deficiencies in these criteria, suggesting a developmental link to the absence of acceptable occlusion (Table 5).
Univariate logistic regression analysis revealed that both age and Xi-L6 at T1 significantly influenced the successful eruption of L8. Multivariate logistic regression analysis further indicated that Xi-L6 at T1 was a significant factor affecting the successful eruption of L8 (P < 0.05; Table 6). Specifically, as age at T1 increased, the probability of L8 being classified in the A-group decreased, whereas an increase in the distance from Xi to L6 at T1 was associated with a higher probability of L8 being classified in the A-group.
DISCUSSION
This study investigated the spontaneous eruption and final occlusion of L8 teeth following the extraction of L7 without any orthodontic intervention. The results demonstrated that over half (58.7%) of the L8 teeth erupted successfully into an acceptable occlusion, which was significantly affected by the age and Xi-L6 at T1. Younger age at the time of L7 extraction seems to be a key factor influencing the successful L8 eruption with proper occlusion.
The observed success rate (58.7%) of acceptable occlusion in this study is slightly lower than the range reported in previous studies (66.2–96.0%).2,5,6,12,28 This difference might be attributed to the stricter criteria employed in this study. The ABO index, encompassing five criteria was used to define acceptable occlusion.31 In contrast, prior studies often lacked a comprehensive definition, excluding factors like IC, buccal crossbite, or MR discrepancies.5,21,28
Additionally, some studies investigated the rate of successful eruption, instead of the acceptable occlusion, based on PAN images which may not accurately depict the actual tooth position.2,19,28 If one of the successful eruption criteria, which was less than 35° of ∠8-6, is addressed in our samples,28 a significant portion would be considered successful because our samples exhibited average ∠8-6 values of 29.3° and 18.3° at T1 and T2, respectively.
Previous studies have suggested that the angulation between L6 and L8 is a critical factor for successful eruption,7,28 and 15–30° of ∠8-6 between these teeth was proposed as a favorable condition for L7 extraction.7 Interestingly, while many L8 teeth in the A-group had > 30° of ∠8-6 at T1 (mean: 29.9°), all of them erupted successfully. Notably, the angulation between L8 and MnP decreased significantly by T2, to an average of 16.9°, indicating that L8 tends to self-upright during the eruption.22,25
The parameters of ∠6-MnP and ∠8-MnP can also explain the self-uprighting of L8. Although no significant differences were observed between the A- and U-groups at T1, both parameters were higher in the A-group compared with the U-group at T2. This indicates that more uprighting of L6 and L8 occurred in the A-group than that in the U-group during eruption. The changes in angulation relative to the MnP between T1 and T2, Δ∠6-MnP and Δ∠8-MnP, were also pronounced in the A-group than those in the U-group. Notably, the A-group experienced a 21.6° increase in ∠8-MnP, whereas the U-group had an increase of only 11.6°. This significant uprighting of L8 during the eruption period might contribute to successful eruption and acceptable occlusion. However, these observed changes in angulation may also reflect underlying skeletal growth or mandibular remodeling, emphasizing the need for further investigation to clarify their significance.
Our study identified two key factors influencing favorable L8 occlusion: patient age and the distance between L6 and Xi point at T1. Although the duration of L8 eruption and age at T2 were similar between the two groups, the A-group was, on average, about a year younger than the U-group at T1. Logistic regression analysis showed that for each year younger at T1, the likelihood of being in the A-group increased by 1.535 times. Regarding the distance of Xi-6, both groups showed 1.8–2.0 mm increase in Xi-L6 during the observation period. However, the distance from the Xi point to L6 was greater in the A-group compared with the U-group at both T1 and T2, indicating that insufficient RMS could be the biggest cause of L8 impaction.19 The odds of being in the A-group decreased by 23.3% for every millimeter decrease in Xi-6. Nevertheless, the RMS was not identified as a significant factor in the regression analysis, suggesting the potential for distortion in the PAN images. Surprisingly, the developmental stage of L8 at the initial assessment did not significantly influence the final occlusion, contrary to earlier research findings.2,28 This discrepancy might be explained by the inclusion criteria in our study, which considered only teeth with less than one-third root formation.
The initial angulation, and horizontal and vertical positions of L8 were not reliable predictors, which is consistent with the findings related to maxillary third molars following extraction of the second molar.9 However, the 58.7% rate of acceptable occlusion observed in this study was slightly lower than that reported in previous studies involving maxillary teeth, which indicated acceptable occlusion rates of 59–100%.2,9,25,26 This discrepancy may be attributed to the differing eruption and occlusion patterns of maxillary third molars. The mesial positioning of the root apex to the crown in maxillary molars, as opposed to the distal positioning of the root apex in mandibular molars, could account for this variation.32,36
The size and developmental stage of L8 did not show significant differences between the two groups. The dimensions of L8 were adequate to serve as replacements for L7 in both groups. This could be due to the decision to replace L7 with L8 based on the radiographic characteristics of the L8 tooth germ, which demonstrated a favorable shape and size. Additionally, there was no significant difference in the developmental stage of L8 at T1 between the A- and U-groups. However, as the developmental stage advanced, the proportion of teeth in the U-group increased. At Nolla stage 7, approximately half of the teeth were classified in the U-group. Moreover, the three L8 lacking IC with L6 were all in Nolla stage 7 at T1. Therefore, the absence of statistical significance between the Nolla stage and final occlusion might be attributed to our sample comprising only L8 in the Nolla stage below 8, following guidelines that recommend early L7 extraction once a viable L8 germ is detectable,30 but ideally before reaching stages 7 or 8.7,28
This study has few limitations. Owing to ethical considerations, this study could not employ a randomized controlled trial design. Since the study focused on spontaneous L8 eruption following L7 extraction at Nolla stages ≤ 7, comparisons with extractions at later stages were not possible.28,30 Future research is needed for the optimal timing of L7 extraction based on a broader range of Nolla stages. This analysis relied on two-dimensional radiographic images. PAN were used in this study for their ability to avoid bilateral overlap, enabling clear observation of third molar development with low radiation exposure.7,37-39 They offer reproducibility when taken by an experienced radiographer and are particularly reliable in the posterior region.37 Future studies could utilize three-dimensional imaging techniques to overcome the limitations of two-dimensional images—such as magnification and distortion—and to achieve more precise measurements.
CONCLUSIONS
This study assessed the final AL of L8 after L7 extraction, and the findings were as follows:
The L8 erupted spontaneously in all samples (46 teeth), and 27 teeth (58.7%) showed acceptable occlusion without intervention.
The key factors for achieving acceptable occlusion of L8 were a younger age at the time of L7 extraction and sufficient space (Xi-6) observed on CEPH.
No significant difference was found in the final occlusion, when L7 was extracted at Nolla stages 4–7.
The vertical and horizontal positions of the L8 at the time of L7 extraction did not affect the final occlusal status.
XML Download