Background
The mandibular third molar (M3) typically erupts last among the permanent teeth due to the lack of available space and thick soft tissues covering its surface [
1]. In many cases, impacted M3s require surgical procedures including alveoloplasty and tooth hemisection. Some clinical research has focused on the classification method for these impacted M3s, and Pell and Gregory classification [
2] is still considered one of the most effective methods. This classification categorizes M3 based on the relative positions of the ramus of the mandible and the occlusal surface of the adjacent M2 [
3,
4]. Statistically, M3 impaction occurs at a high rate of 66%, and a study of 3799 patients over the age of 25 reported that horizontal impaction was most prevalent among angulation types [
3,
5]. Among the lesions associated with impacted M3, dental caries occurs in the mandible three times more frequently than in the maxilla [
3]. One study found that the incidence of dental caries in the distal surface of M2 associated with M3 was 37.5%, most of which occurred in Pell and Gregory class I and position B [
6,
7].
Impacted M3 often causes suppurative inflammation such as chronic periodontitis and odontogenic cysts [
1]. In addition, when M3 is extracted, bone absorption, periodontal pocket formation, cementum exposure, and gingival recession may occur in the adjacent second molar [
8]. Several studies comparing groups with and without M3 extraction have found significant periodontal tissue destruction at the distal aspect of M2, including increase in probing depth and radiographic alveolar bone loss [
9]. Previous retrospective studies with a follow-up of more than 2 years reported that surgical extraction leaves deep infrabony defects but superior bone regeneration capacity in younger age groups [
10‐
12].
Most M3-associated lesions can occur in various forms on the distal surface of M2 and often require additional treatments. For most conservative and periodontal therapies, bone regeneration within the extraction socket should be completed in advance. However, there is a lack of clinical guidance and evidence for the optimal timing of treatment considering bone regeneration of the distal aspect of M2 after extraction of M3. The purpose of this study is to compare and analyze the degree of bone regeneration with respect to time and impaction depth in the extraction socket of mandibular third molars in reference to the distal aspect of adjacent second molars using panoramic radiography.
Discussion
In the Pell and Gregory classification, the position of the third molar is determined by the relationship between the ramus of the mandible and the mandibular adjacent M2. Classes I, II, and III specifies the mesiodistal width between M2 and the ramus, while positions A, B, and C refers to the vertical depth with respect to the M2 occlusal plane. DI scores combine the Pell and Gregory classification and the Winter classification, which defines the angulation of M3. In this study, only horizontally impacted samples were collected; therefore, two points were added equally to each DI score. The difficulty index assigns 1, 2, or 3 points for position A, B, or C, respectively, and 1, 2, or 3 points for class I, II, or II. The final DI score can be obtained by adding the scores of Pell and Gregory classification and Winter classification. A DI score of 3 or 4 points is categorized as minimally difficult, 5 to 7 points as moderately difficult, and 7 to 10 points as very difficult [
2,
4].
The samples collected for this study were homogeneous in nature because all were horizontally impacted M3s with a DI score between 5 and 8 points. In addition, the panoramic radiograph images were collected on the basis of patient histories that did not include complications or diseases that may affect bone regeneration. Kugelberg et al. [
10‐
12] reported that bone regeneration after M3 extraction is affected by age and is more likely to occur in younger patients under 25 years of age. However, this study was performed independent of the age of patients (38.7 ± 11.1 years).
In addition to clinical exams, radiographic exams are one of the major determinants of clinical bone regeneration and recovery following M3 extraction [
10,
17‐
19]. Time is an important variable in the analysis of radiographic images and has a direct effect on other measured variables in the image. Many previous retrospective studies focused mainly on bony changes over time after M3 extraction [
10,
17‐
20]. In spite of periapical radiograph is recommended to measure the bone level and bone margin evaluation with panoramic radiograph is not a standard method, the strength of this study was its homogenous collection of horizontally impacted M3s and inclusion of DI as an analyzed variable while focusing on bone regeneration over time.
Panoramic radiography is widely used in routine dental procedures such as implant placement, and it has the advantage of showing surrounding anatomical structures as well as the teeth. However, the panoramic image is magnified and distorted beyond actual size when the patient is out of the focal trough. Even if screened using a variate procedure, panoramic radiograph has an average magnification of 15 to 25% depending on the patient’s position [
21]. The magnification rate can be affected by the shape and size of the patient’s jaw and is greatest at the canine and premolar regions and lowest at the third molar region [
22,
23]. Therefore, it is difficult to position the patient accurately in the focal trough, even with the help of an aiming light. According to an ideal experimental study, the vertical magnification ratio showed less variation and more consistent results than horizontal magnification ratio [
24]. In a study comparing the reliability of cone-beam computed tomography and panoramic radiography, although a vertical overestimation of 0.87 mm occurred as the alveolar process moved 1 mm toward the lingual side, it concluded that such errors are acceptable for clinical use [
25].
A reliable and standardized diagnostic method, such as assessment of infrabony defects recovery after M3 extraction, is required to assess bony changes over time. However, many existing studies used various types of images with different measuring tools, and it was difficult to compare the data or results [
9,
10,
17,
18,
20,
26]. In this study, the existing method proposed by Faria et al. [
13] was employed to minimize deviating from the recent research standards. Bone regeneration after M3 extraction occurred constantly over time. The RID was 9.58 ± 2.25 mm at baseline, 6.41 ± 2.53 mm after 6 weeks, and 3.21 ± 1.39 mm after 6 months. In Faria et al. [
13], the initial RID0 was 4.54 ± 1.87 mm, and RID6M was 2.59 ± 1.85 mm. Bone regeneration was 1.40 ± 2.00 mm and 0.56 ± 1.19 mm at 3- and 6-month follow-ups, respectively. Another study by Faria et al. [
26] showed a 1.62 ± 2.44 mm recovery of periodontal pocket depth during the first 3 months after extraction, and there was no significant change in pocket depth between the 3-month and 12-month follow-ups. Although there was a difference in RID values, such a difference was considered reasonable in the present study because the samples were all M3s with deep horizontal impaction. In this study, a 3-month follow-up image was not included, but active bone regeneration was observed initially over the short period following extraction.
In case of large extraction sockets, the proportion of RIDs > 6 mm decreased dramatically from 91.2% to 2.9% during the 6-month follow-up period. In addition, 61.8% of infrabony defects recovered nearly to the physiologic condition of RID ≤ 3 mm. An analysis of RBH between evaluation periods showed that a few cases exhibited bone loss in the early stages, but all eventually showed bone gain after the final follow-up of 6 months. Therefore, as with the in vivo study of mongrel dogs, it appears that transient bone loss was caused by osteoclast activity during the early stages of bone remodeling [
27,
28]. Because there was considerable individual variation in terms of bone-healing rate, it was difficult to predict the healing progress of a patient at a given time.
More recent research has focused on peripheral bone changes that occur with post-extraction healing. In vivo studies using mongrel dogs showed bone resorption through osteoclast activity during the first 8 weeks after extraction, causing a decrease in vertical height [
27]. Although 88% of the extraction socket was replaced with mineralized bone 30 days after extraction, the mineralized tissue decreased to 15% after 180 days, and the bone marrow increased to 85% over time [
27,
28]. In actual clinical settings, patients with M3 extraction showed periodontal problems to some extent during the first 3 months of follow-up, but the problems lessened remarkably after 1 year [
29]. Another study reported that bone healing did not occur during the first 3 months after extraction, but infrabony defects recovered to their original state after 12 months [
13].
The difficulty of impacted mandibular M3 extraction can be influenced by the shape of the tooth, the location within the arch, the depth of impaction, and the angulation of tooth. Above all, impaction depth and angulation are directly related to difficulty in extraction [
30]. In this regard, the DI using the Pell and Gregory classification and the Winter angulation classification can play an important role in diagnosis and preoperative planning. Among the total of 34 study samples, five had a DI score of 5, 11 samples were assigned a DI score of 6, 10 samples a DI score of 7, and eight samples a DI score of 8; all were classified as moderately difficult or very difficult. RID differences were analyzed with respect to DI scores, and only the difference between baseline RID and RID6W showed statistical significance (
p < 0.05). Within the RID6W_RID0, the group with a DI score of 8 had the highest average RID differences (− 5.37 ± 2.80 mm). A difference in the RID value is a measure of the degree of bone regeneration. These differences between RID values represented the amount of bone regeneration, and initial bone regeneration was observed during the early stage of the healing process. However, further study was needed to verify the current results and to reveal the contributing factors that might have affected bone regeneration.
The correlation coefficient between the DI score and the RID difference was only statistically significant in RID6M_RID0 (p < 0.05), which showed a relatively low positive correlation coefficient of 0.396. Thus, patients with higher initial DI scores would have a higher absolute amount of bone regeneration. In this context, the samples used in the present study also showed greater bone regeneration with a deeper initial RID0 and higher DI score, indicating a positive correlation. As a result, extraction difficulty had no significant effect on initial bone regeneration, although it might affect final bone regeneration, and the increase in initial RID could result in greater bone regeneration.
It is important to obtain standardized measurements and images in radiographic analysis as in the present study. Several studies mentioned that it is difficult to standardize panoramic images [
21,
24]. In Faria et al. [
13], a modified intraoral radiography device was used to reduce and standardize the error between images. However, in the present study, it was impossible to avoid distortions from characteristics of panoramic images. Instead, to compensate for the difference in distortion rate between images, a DF was calculated and applied to the RID values. Furthermore, to reduce human error and intra-examiner bias in analyzing panoramic images, it is necessary to use a radiopaque indicator, such as the dental probe used by Faria et al. [
13]. Without these devices, the present study was left with some limitations: pre-extraction RID could not be measured due to superimposition of teeth. Lastly, unlike most studies focusing on bone regeneration, which included a minimum of 1-year follow-up, the present study only had a 6-month follow-up period because of limitation in the research settings.
Large RIDs (> 6 mm) that developed immediately after extraction constantly decreased over time and recovered to a normal range (RID ≤ 3 mm) in more than half of the cases after 6 months of extraction. Although bone regeneration after tooth extraction occurred actively throughout the first 6 months, extraction difficulty was significantly affected within the first 6 weeks. Correlation analysis between extraction difficulty and bone regeneration showed that the increase in infrabony defects may lead to enhanced bone healing in the long term. While DI did not affect long-term bone healing from 6 weeks to 6 months, it did affect initial bone regeneration; therefore, further study will be needed to determine the specific factors associated with the initial bone-healing process.
As a result, if additional treatments of an adjacent M2 are required after M3 extraction, it is recommended that clinicians do not proceed with further treatment during the first 6 months after extraction. However, because bone regeneration patterns, rate, and recovery ability vary greatly among individuals, it is difficult to predict the absolute stage of bone regeneration in a patient. Clinicians must perform clinical and radiographic exams before proceeding with further treatments. Extraction difficulty appears to affect bone regeneration, but further research is needed on the related factors.