ORIGINAL ARTICLE

Factors influencing the successful

eruption of the maxillary third molar after
extraction of the maxillary second molar
Insil Kim,a Sun-Hyung Park,b Yoon Jeong Choi,b Ji Hyun Lee,c Chooryung J. Chung,c and Kyung-Ho Kimc
Seoul, South Korea

Introduction: This study evaluated the occlusal status of the maxillary third molars that erupted spontaneously
after extraction of the maxillary second molars and investigated the factors that influenced the occlusal status of
the maxillary third molars. Methods: We assessed 136 maxillary third molars in 87 patients. Alignment, marginal
ridge discrepancy, occlusal contact, interproximal contact, and buccal overjet were used to score the occlusal
status. Occlusal status was classified as good (G group), acceptable (A group), and poor (P group) for the maxil-
lary third molar at its full eruption (T1). The Nolla’s stage, long axis angle, the vertical and horizontal position of
the maxillary third molar, and the maxillary tuberosity space were assessed at the time of maxillary second molar
extraction (T0) and T1 to identify factors influencing the eruption of the maxillary third molar. Results: G, A, and P
groups comprised 47.8%, 17.6,% and 34.6% of the sample, respectively. Age was the lowest in the G group at
both T0 and T1. The maxillary tuberosity space at T1 and the amount of the change of maxillary tuberosity space
were the largest in the G group. There was a significant difference in the distribution of the Nolla’s stage at T0.
The proportions of the G group were 60.0% in stage 4, 46.8% in stages 5 and 6, 70.4% in stage 7, and 15.0% in
stages 8-10. According to multiple logistic regression analysis, stages 8-10 for the maxillary third molar at T0 and
the amount of the change of maxillary tuberosity were negatively associated with the G group. Conclusions:
Good-to-acceptable occlusion was seen in 65.4% of the maxillary third molars after maxillary second molar
extraction. Insufficient increase in the maxillary tuberosity space and Nolla stage 8 or higher at T0 negatively
influenced the maxillary third molar eruption. (Am J Orthod Dentofacial Orthop 2023;164:636-45)

E
xtraction of permanent teeth is a common treat- malposition, posterior crowding, and mild crowding
ment strategy in orthodontics. Although maxillary with good facial profile, have been reported.4,17,18 An
second molar extraction is not as common as pre- adequately placed and normally shaped maxillary third
molar extraction, it is an option for solving various skel- molar is essential in such patients.2
etal and dental abnormalities.1-15 Because Chapin first Among the maxillary third molar that erupted after
mentioned maxillary second molar extraction,16 indica- maxillary second molar extraction, a satisfactory state
tions, such as severe dental caries, ectopic eruption, or good occlusion was reported in 59%-100% of the
patients.8,18-20 However, Bishara et al1 mentioned in
a
their review on maxillary second molar extraction that
Department of Orthodontics, College of Dentistry, Yonsei University, Seoul,
South Korea. comparisons among those studies were difficult because
b
Department of Orthodontics, The Institute of Craniofacial Deformity, Yonsei of varied definitions of good occlusion in the studies. In
University, College of Dentistry, Seoul, South Korea. addition, although many studies investigated the posi-
c
Department of Orthodontics, The Institute of Craniofacial Deformity, Gangnam
Severance Dental Hospital, Yonsei University, College of Dentistry, Seoul, South tion of the maxillary third molar after maxillary second
Korea. molar extraction, few have evaluated the occlusal status
Insil Kim and Sun-Hyung Park are joint first authors and contributed equally to of the maxillary third molar using dental models; more-
this work.
All authors have completed and submitted the ICMJE Form for Disclosure of Po- over, the criteria used for ascribing a satisfactory state
tential Conflicts of Interest, and none were reported. were not clearly defined.8,19,21,22
Address correspondence to: Kyung-Ho Kim, Department of Orthodontics, The Some studies indicated that the optimal timing of
Institute of Craniofacial Deformity, Gangnam Severance Dental Hospital, College
of Dentistry, Yonsei University, 211, Eonju-ro, Gangnam-gu, Seoul 06273, South maxillary second molar extraction is essential for the
Korea; e-mail, [email protected]. successful spontaneous eruption of the maxillary third
Submitted, December 2022; revised and accepted, March 2023. molar.4,6,17,20,23 It is recommended to consider the
0889-5406/$36.00
Ó 2023 by the American Association of Orthodontists. All rights reserved. developmental stage, position, and angulation of the
https://doi.org/10.1016/j.ajodo.2023.03.021 maxillary third molar when determining the timing for
636
Kim et al 637

maxillary second molar extraction. Magness suggested Five criteria were used to evaluate the occlusal status
extracting the maxillary third molar when it reaches of the maxillary third molar at T1: (1) alignment—the
the cementoenamel junction of the maxillary second mesiodistal central groove of the maxillary first and third
molar.23 de-la-Rosa-Gay et al20 reported that the higher molars in the same plane; (2) marginal ridge—the mar-
the developmental stage (Nolla stage 8 or higher) of the ginal ridge of adjacent teeth at the same level or within
third molar at the time of the second molar extraction, 0.5 mm; (3) occlusal contact—the buccal cusp of the
the less ideal the final position of the third molar.20 mandibular molar and the lingual cusp of the maxillary
This study aimed to (1) assess the final position of the third molar should be contacting the occlusal surfaces
maxillary third molar that erupted after maxillary second of the antagonists; (4) interproximal contact—there
molar extraction on study models using the Objective should be no space between the maxillary first and third
Grading System of the American Board of Orthodontics molars; and (5) buccal overjet—the buccal cusp of the
(ABO)24 as a clear evaluation criterion and (2) investigate mandibular antagonist should contact the center of
the factors that influenced the occlusion of the maxillary the occlusal surfaces, buccolingually, of the maxillary
third molar using panoramic radiographs (PANs), lateral third molar. The measurements were scored according
cephalograms (CEPHs), and study models. to the ABO grading system.24 An ABO measuring gauge
was used to measure displacement (Fig 1).
Occlusal statuses were classified as “good,” “accept-
MATERIAL AND METHODS able,” and “poor” occlusion groups. The Good occlusion
The study included 87 patients (31 men and 56 group (G group) was defined when all 5 of the following
women) aged 12.9-17.7 years who had undergone conditions were satisfied; maxillary first and third molars
maxillary second molar extraction at the Department were in alignment or within 0.5 mm of proper alignment,
of Orthodontics, Gangnam Severance Dental Hospital. marginal ridges of maxillary first and third molars were
Of these, 49 patients underwent bilateral maxillary sec- at the same level or within 0.5 mm, cusps in contact
ond molar extractions, whereas 38 had unilateral extrac- with the opposing arch, no interproximal space between
tions. A total of 136 maxillary third molars (69 of right the maxillary first and third molars, and the buccal cusp
third molars and 67 of left third molars) were observed; of the mandibular molar contacting the center of the
48 were observed in males and 88 in females. The inclu- occlusal surface of the maxillary third molar. If at least
sion criteria for the participants were (1) no orthodontic 1 of the 5 following conditions were met, it was classi-
treatment for the maxillary third molar, (2) records avail- fied into the acceptable occlusion group (A group): (1)
able (PANs, CEPHs, and study models) from the time of when the contact point of the maxillary first and third
maxillary second molar extraction (T0) and the complete molars was 0.5-0.0 mm deviated from proper alignment,
eruption of the maxillary third molar (T1). When contact (2) marginal ridges between maxillary first and third mo-
with the antagonist or the marginal ridge discrepancy lars deviated between 0.5-1.0 mm, and (3) a cusp was
with the adjacent maxillary first molar was within 0.5 out of contact with the opposing arch, space up to 1.0
mm, it was defined as a complete eruption of the maxil- mm between the maxillary first and third molars, or
lary third molar. the mandibular buccal cusp deviated #1.0 mm from
This study was approved by the Institutional Review the center of the maxillary third molar. When the criteria
Board Gangnam Severance Dental Hospital (no. 3- for neither G nor A groups were satisfied, the patient was
2015-0327). classified into the poor occlusion group (P group)
The patients regularly visited the hospital and verified (Table I).
the time of the complete eruption of the maxillary third The position of the maxillary third molar (Fig 2) was
molar. An experienced radiologist obtained PANs and evaluated with PAN. The outline of the maxillary first
CEPHs with the same machine (Planmeca ProMax; Plan- and third molars was traced on the PAN. Eleven land-
meca, Helsinki, Finland) throughout the study. Measure- marks were designated: (1) the distobuccal cusp of the
ments of the PANs and CEPHs were conducted using maxillary third molar, (2) the mesiobuccal cusp of the
V-ceph software (version 5.5, Osstem Implant, Seoul, maxillary third molar, (3) the midpoint between the
South Korea), allowing distance measurements of up maxillary third molar and the mesiobuccal cusp of the
to 0.1 mm and angle measurements of up to 0.1
. Using maxillary third molar, (4) the height of contour on the
CEPHs at T0 and T1, the pretreatment factors affecting mesial surface of the maxillary third molar, root (5)
the successful eruption of the maxillary third molar and furcation of the maxillary third molar, (6) distobuccal
the differences between the 2 time periods were as- cusp of the maxillary first molar, (7) mesiobuccal cusp
sessed: (1) occlusal status of the erupted maxillary third of the maxillary first molar, (8) midpoint between the
molar and (2) position of the maxillary third molar. distobuccal cusp of the maxillary first molar and the

American Journal of Orthodontics and Dentofacial Orthopedics November 2023  Vol 164  Issue 5
638 Kim et al

Fig 1. ABO measuring gauge.

Table I. Evaluation criteria for the occlusal state of the maxillary third molar
Alignment Marginal ridge Occlusal contact Interproximal contact Buccal overjet
Good Discrepancy #0.5 mm Discrepancy #0.5 mm Contact with the No space between Cusp of the opposing
occlusiony opposing arch the maxillary first and mandibular tooth
third molars contact with the
buccolingual center of
the maxillary third
molar
Acceptable Discrepancy #1.0 mm Discrepancy #1.0 mm Out of contact #1.0 mm The space between Deviation #1.0 mm
occlusionz maxillary first and
third molars #1.0 mm
Note. Patients not falling into the G and A groups were defined as the P group.
y
All 5 criteria should be satisfied; zIf at least 1 of the criteria of the good occlusion group met the following, it was classified as the acceptable oc-
clusion group.

mesiobuccal cusp of the maxillary first molar, (9) the molar was defined as the shortest distance between
height of contour on the distal surface of maxillary first PTv and Mx1Hc. In bilateral extraction of the maxillary
molar (Mx1Hc), (10) root furcation of the maxillary first second molar, the distance between the midpoints of
molar, and (11) the buccal cusp of the second premolar. the right and left pterygomaxillary fissure and Mx1Hc
The plane joining the premolar and molar cusps was was used, and in unilateral extraction, the distance
defined as the occlusal plane (OP). The vertical position was measured only on the extracted side.
of the maxillary third molar was measured as the shortest The developmental stage of the maxillary third molar
distance from the mesiobuccal cusp of the maxillary at T0 was classified into 4 categories (stage 4, stages 5
third molar to the OP (VMx3-OP), and the horizontal po- and 6, stage 7, and stages 8-10) according to Nolla’s
sition was calculated as the distance between the feet of stage25 identified on PAN (Table II). Stages 5 and 6
the perpendicular drawn from the height of contour on were difficult to distinguish clearly and were grouped
the mesial surface of the maxillary third molar to Mx1Hc into 1 group (stages 5 and 6); similarly, stages 8-10
on OP. To evaluate the axial change of the maxillary were grouped (stages 8-10).
third molar, 2 angles were measured; the angle formed Furthermore, the factors affecting the final occlusal
by the long axes of the maxillary third and first molars status of the maxillary third molar among the above
(:Mx3-Mx1) and the angle formed by the long axis measurements were investigated.
of the maxillary third molar and the perpendicular
from OP (:Mx3-vOP). The long axis of each tooth
was defined by connecting the midpoint between the Statistical analysis
mesiobuccal cusp, distobuccal cusp, and furcation point. A single examiner (I.K.) performed all measurements.
The maxillary tuberosity space (Fig 3) was measured Twenty samples were randomly selected and measured
using CEPHs at T0 and T1. The horizontal reference twice at 2-week intervals to evaluate the intraexaminer
line was constructed 7
below the sella-nasion line, error, assessed using the intraclass correlation coeffi-
and the vertical reference line (PTv) was drawn perpen- cient. The measured data were described as means and
dicular to the horizontal reference line passing the pter- standard deviations. An independent 2-sample t test
ygomaxillary fissure. The space for the maxillary third was used to compare male and female patients on age

November 2023  Vol 164  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Kim et al 639

Fig 2. A, Landmarks and reference lines were traced on the panoramic radiograph. Eleven landmarks
were used: (1) Mx3dc—the distobuccal cusp of the maxillary third molar, (2) Mx3mc—the mesiobuccal
cusp of the maxillary third molar, (3) Mx3m—the midpoint between Mx3dc and Mx3mc, (4) Mx3Hc—the
height of contour on the mesial surface of the maxillary third molar), (5) Mx3fur—the root furcation of the
maxillary third molar, (6) Mx1dc—the distobuccal cusp of the maxillary first molar, (7) Mx1mc—the me-
siobuccal cusp of the maxillary first molar; (8) Mx1m—the midpoint between Mx1dc and Mx1mc; (9)
Mx1Hc; Mx1fur—the root furcation of the maxillary first molar, (10) Pmc—the buccal cusp of the second
premolar, and (11) OP; B, Vertical and horizontal measurements: (1) VMx3-OP and (2) HMx3-Mx1—
the distance between the feet of the perpendicular drawn from Mx3Hc and Mx1Hc on OP; C, Angular
measurements: :Mx3-Mx1 and :Mx3-vOP.

American Journal of Orthodontics and Dentofacial Orthopedics November 2023  Vol 164  Issue 5
640 Kim et al

Fig 3. A, The horizontal reference line was constructed 7
below the sella-nasion line, and PTv was
drawn perpendicular to the horizontal reference line passing the pterygomaxillary fissure point (ptm).
PTv-Mx1 was defined as the shortest distance between PTv and Mx1Hc; B, In the case of bilateral
extraction of the maxillary second molar, the distance between the midpoints of right and left ptm
and Mx1Hc was used; C, In the case of unilateral extraction, the distance was measured on the ex-
tracted side.

and Nolla’s stage. The age, eruption duration, and patients in the G, A, and P groups was 47.8% (n 5
measured values among G, A, and P groups were 65), 17.6% (n 5 24), and 34.6% (n 5 47), respec-
compared using analysis of variance and post-hoc anal- tively, and showed a significant difference (P 5
ysis with Bonferroni correction. The Fisher exact test and 0.006) (Table IV). The G group had a higher proportion
post-hoc comparison with Bonferroni correction were of Nolla stages 4-7 and a lower proportion of stages
used to observe the association of the Nolla’s stage 8-10, whereas it was the opposite in the P group.
and the occlusal status of the maxillary third molar at The proportion of the G group was highest for Nolla
T1. To further analyze the effect of the developmental stage 7 (70.4%) and the lowest for stages 8-10
stage, the samples were divided into 2 groups: stages (15.0%). Because there was a sharp decrease in the
4-7 and 8-10. Simple and multiple logistic regression proportion of the G group between Nolla stage 7
was performed to find factors related to good occlusion and stages 8-10, we divided the samples into 2 groups
in the maxillary third molar among the variables. The (stages 4-7 and stages 8-10) for comparison. The re-
odds ratio with a 95% confidence interval was used, sults showed that the proportion of the G group in
and the significance level was tested. The significance Nolla stages 4-7 and 8-10 was 53.4% and 15.0%,
level was set at P \0.05. SAS software (version 9.2; respectively.
SAS, Cary, NC) was used for all statistical analyses. Table V shows the following 6 measurements showed
significant differences among 3 groups: age at T0 and
RESULTS T1, maxillary tuberosity space (PTV-Mx1) at T1, PTv-
The intraclass correlation coefficient values for the Mx1 change, :Mx3-vOP, and VMx3-OP at T1. The vari-
measurements ranged between 0.93-0.99, indicating a ables could be grouped into 3 categories: age, the maxil-
high level of intraexaminer reliability. The age of the par- lary tuberosity space, and the position of the maxillary
ticipants is shown in Table III. The independent 2- third molar.
sample t test showed no significant difference between Patients in the G group were significantly younger
male and female patients in age at T0 and T1 and Nolla’s than A and P groups at T0 (P 5 0.01) and T1
stage at T0 (Table III). (P \0.001). However, the 3 groups showed no signifi-
The study models at T1 were analyzed when all cant difference in the mean duration for eruption
maxillary third molars erupted. The proportion of the completion of the maxillary third molar.

November 2023  Vol 164  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Kim et al 641

Table II. Nolla’s developmental stages Table III. Mean and standard deviation of age at T0
and T1, duration of eruption and Nolla’s stage of the
Stage Development of the tooth Tooth
maxillary third molar at T0 and mesiodistal width of
0 Absence of crypt
1 Presence of crypt the maxillary molars
Male Female Total
Variables (n 5 48) (n 5 88) P value* (n 5 136)
T0 (y) 15.3 6 2.6 15.6 6 2.6 0.3930 15.5 6 2.6
2 Initial calcification T1 (y) 19.2 6 1.9 19.6 6 2.0 0.1732 19.5 6 2.0
Duration (y) 3.9 6 1.6 4.0 6 1.8 0.7895 4.0 6 1.7
Nolla stage at T0 6.0 6 1.2 6.3 6 1.3 0.2042 6.2 6 1.3

3 One-third of the crown formed Note. Values are presented as mean 6 standard deviation.
n, number of the maxillary third molar; T0, the time the maxillary
second molar was extracted; TF, the time of complete eruption of
the maxillary third molar.
*Independent 2-sample t test was used to compare males and
4 Two-thirds of the crown formed
females.

5 Crown almost fully formed differences were found between the G and P groups in
angulation :Mx3-vOP (P \0.05) and VMx3-OP
(P\0.01) of the maxillary third molar. The maxillary third
6 Crown fully formed molar in the G group was more upright to OP and more
vertically erupted than the P group at T1.
Simple and multiple logistic regression analyses were
used to identify the factors related to the G group. Sim-
7 One-third of the root formed ple logistic regression analysis resulted in significant dif-
ferences among the 3 groups in age at T0 and T1, Nolla’s
stage at T0, PTv-Mx1 at T1, PTv-Mx1 change, :Mx3-
Mx1 at T1, :Mx3-vOP at T1, and VMx3-OP at T1
8 Two-thirds of the root formed (Table VI).
As the age increased by 1 year, the odds of belonging
to the G group decreased by 20.4% at T0 and 35.8% at
T1 (P \0.001).
The odds of belonging to the G group decreased by
9 Root almost formed
28.6% as Nolla’s stage increased at T0 (P \0.05). The
patients with Nolla stages 8-10 had 11.8% (P \0.05)
and 15.4% (P \0.01) odds of being in the G group
compared with stages 4 and 4-7, respectively.
The odds of belonging to the G group increased by
10 Closed apex 1.248 times (P \0.001) and 2.871 times (P \0.001)
with every millimeter increase in PTv-Mx1 at T1 and
PTv-Mx1 change, respectively.
Position of the maxillary third molar: There was no
association with the maxillary third molar position at
T0; however, as the :Mx3-Mx1, :Mx3-vOP, and
At T0, there was no difference in PTv-Mx1 among the VMx3-OP at T1 increased, the odds of belonging to
3 groups. However, the G group showed the largest the G group decreased.
change in PTv-Mx1 than A and P groups (P 5 0.001). According to the multiple logistic regression analysis,
At T1, the G group showed larger PTv-Mx1 than the P the odds of belonging to the G group were 2.577 times
group (P \0.001). higher for every mm increase in PTv-Mx1 change
Comparison of the vertical, horizontal, and angular (P \0.001). For Nolla stages 8-10, the odds of the G
positions of the maxillary third molar showed no differ- group were 0.132 times lower than stages 4-7
ence among the groups at T0. However, at T1, significant (P \0.05) (Table VII).

American Journal of Orthodontics and Dentofacial Orthopedics November 2023  Vol 164  Issue 5
642 Kim et al

Table IV. The Distribution of the Nolla’s stage of the maxillary third molar at the time the maxillary second molar was
extracted, according to the occlusal state of the maxillary third molar at the time of complete eruption of the maxillary
third molar
Good Acceptable Poor Total P valuey
Stage 4 6 (60.0) 2 (20.0) 2 (20.0) 10 (100) 0.006**
Stage 5 and 6 37 (46.8) 12 (15.2) 30 (38.0) 79 (100)
Stage 7 19 (70.4) 4 (14.8) 4 (14.8) 27 (100)
Stage 8-10 3 (15.0) 6 (30.0) 11 (55.0) 20 (100)
P valuez a ab b
Stage 4-7 62 (53.4) 18 (15.5) 36 (31.1) 116 (100) 0.003**
Stage 8-10 3 (15.0) 6 (30.0) 11 (55.0) 20 (100)
P valuez a b b
Total (%) 65 (47.8) 24 (17.6) 47 (34.6) 136 (100)

Note. Values are presented as n (%). Fisher exact test and post-hoc pairwise comparison with Bonferroni correction were performed at a significance
level of P \0.05.
y
Comparison among G, A, and P groups for the distribution of samples using the Fisher exact test; zResult of post-hoc pairwise comparison with
Bonferroni correction. The letters indicate the Bonferroni post-hoc results, with different letters representing statistically significant differences
(P \0.05); **P \0.01.

DISCUSSION molar relationship was not relevant to this study, and

This study focused on the occlusal status of the the morphology of some of the maxillary third molars
maxillary third molar that erupted after maxillary second was inappropriate for measuring the buccolingual incli-
molar extraction and found that 65.4% of the maxillary nation. Although the prerequisite for maxillary second
third molar erupted at good-to-acceptable occlusal sta- molar extraction was the normal shape and size of the
tus after the maxillary second molar extraction. maxillary third molar, it was difficult to identify the exact
Orton-Gibbs et al19 investigated the eruption status form of the maxillary third molar using 2-dimensional
of the maxillary third molar after maxillary second molar radiographic images.
extraction using PAN and a study model. They reported Several factors influence the eruption of the maxillary
that 100% of the maxillary third molar and 99% of the third molar, such as individual growth patterns, insuffi-
mandibular third molar showed good-to-acceptable oc- cient space for eruption, angulation of the maxillary
clusion. De-la-Rosa-Gay et al20 identified that 96.2% of third molar, and permanent tooth extraction. This study
the maxillary third molar and 66.2% of the mandibular aimed to identify the predictors of the successful erup-
third molar were in good occlusion after second molar tion of the maxillary third molar after maxillary second
extraction; moreover, higher the developmental stage molar extraction. The patients were classified into G,
(Nolla stage 8 or higher) of the third molar, the less suc- A, and P groups by occlusal status. Subsequently, the
cessful its eruption. Asai et al18 reported a lower success variables were compared among the 3 groups as follows:
rate of a third molar eruption than previously mentioned (1) age at T0 and T1, (2) PTv-Mx1 at T1, (3) PTv-Mx1
studies. Of the 35 patients who underwent extraction of change, (4) the angulation :Mx3-vOP, and (5) vertical
the second molar, 19 (54%) required orthodontic treat- position (VMx3-OP) of the maxillary third molar at T1.
ment (41% of the maxillary third molar and 85% of the According to the simple logistic regression analysis,
mandibular third molar). This study showed 65.4% of age and Nolla’s stage at T0 were associated with good
the maxillary third molar showed good-to-acceptable occlusion. In multiple logistic regression analysis, PTv-
occlusion. This difference in success rate may have Mx3 change and Nolla stages 8-10 were essential for
been due to the different definitions of good occlusion good occlusion. Findings regarding each variable are
in the studies. The ABO grading system used in this study described below.
was more comprehensive than the grading used in pre- The G group participants were younger at T0 and T1
vious reports. Of the 8 criteria in the ABO system (align- than A and P groups. As the age at T0 and T1 increased,
ment, marginal ridges, buccolingual inclination, occlusal the odds of a good occlusion were diminished.
relationships, occlusal contacts, overjet, interproximal The distribution of Nolla’s stage of the maxillary third
contacts, and root angulation), this study excluded molar at T0 showed that 53.4% of patients lower than
occlusal relationship and buccolingual inclination. The stage 7 were included in the G group. In contrast,

November 2023  Vol 164  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Kim et al 643

Table V. Comparison of extraction time (T0), full eruption time (T1), eruption duration, developmental stages, and
measurements according to the occlusal condition
Groups

Variables G (n 5 65) A (n 5 24) P (n 5 47) P value

Age at T0 (y) 14.8 6 2.2a 16.3 6 2.8b 16.1 6 2.8b 0.01*
Age at T1 (y) 18.8 6 1.5a 20.1 6 2.1b 20.2 6 2.1b \0.001***
Duration T0-T1 (y) 3.9 6 1.6 3.7 6 1.0 4.1 6 1.9 0.737
Nolla’s stage at T0 6.0 6 1.2 6.5 6 1.4 6.5 6 1.3 0.057
PTv-Mx1 at T0 (mm) 11.5 6 2.4 11.0 6 2.9 11.0 6 3.5 0.56
PTv-Mx1 at T1 (mm) 13.5 6 2.5a 12.1 6 2.4ac 11.5 6 3.5bc 0.001**
PTv-Mx1 change (mm) 2.0 6 1.4a 1.1 6 1.4b 0.5 6 0.8b \0.001***
T1
:Mx3-Mx1 (
) 22.7 6 13.7 21.5 6 16.7 21.9 6 17.2 0.936
:Mx3-vOP (
) 21.1 6 13.3 22.0 6 15.6 21.6 6 15.7 0.967
VMx3-OP (mm) 17.5 6 2.5 18.9 6 2.8 17.6 6 2.7 0.085
HMx3-Mx1 (mm) 6.3 6 1.6 5.3 6 2.1 5.7 6 2.7 0.117
T2
:Mx3-Mx1 (
) 10.3 6 4.9 12.5 6 7.1 13.1 6 7.7 0.058
:Mx3-vOP (
) 8.6 6 5.6a 11.6 6 7.6ac 12.4 6 9.8bc 0.024*
VMx3-OP (mm) 2.9 6 1.2a 3.5 6 1.8ac 3.8 6 1.6bc 0.005**
HMx3-Mx1 (mm) 0.2 6 0.2 0.1 6 0.6 0.2 6 0.4 0.404

Note. Analysis of variance and post-hoc analysis with Bonferroni correction were performed. The superscripted letters indicate the Bonferroni post
hoc results, with different letters representing statistically significant differences (P \0.05).
T0, at the time of maxillary second molar extracted; T1, the time of complete eruption of maxillary third molar; PTv, vertical reference line passing
through the pterygomaxillary fissure point; vOP, the line perpendicular to the occlusal plane; OP, occlusal plane; VMx3-OP, vertical distance from
OP to mid-point of the maxillary third molar; HMx3-Mx1, the horizontal distance between the mesial surface of the maxillary third molar to the
distal surface of the maxillary first molar.
*P \0.05; **P \ 0.01; ***P \0.001.

55.0% of patients greater than stage 8 were in the P the maxillary tuberosity space is essential for successful
group. In the G group, stage 7 (one-third of the root for- eruption. Clinically, if distalization of the maxillary
mation) comprised the highest proportion, and this ratio dentition after maxillary second molar extraction is
drastically decreased with stages 8-10. According to planned, security of the space for the maxillary third
simple logistic regression analysis, the odds for the G molar should be considered.
group decreased by 29% as the stage at T0 increased The change of the maxillary tuberosity was the
by 1 stage. On comparing the odds ratio for each Nolla’s largest in the G group. When it increases by 1 mm, the
stage beginning at stage 4, the odds of good occlusion in odds for the G group increase by 2.9 and 2.6 times,
stages 8-10 decreased by 89%, and when compared with respectively, resulting from simple and multiple logistic
stages 4-7, it decreased by 85%. According to the mul- regression analysis. This study initially showed no signif-
tiple logistic regression analysis, the odds of good occlu- icant difference in maxillary tuberosity space among the
sion in stages 8-10 compared with stages 4-7 decreased 3 groups. However, the G group showed an apparent in-
by 86.8%. Hence, extracting the maxillary second molar crease in space, whereas the P group showed a slight in-
could be recommended before Nolla’s stage of the crease during the observation period. Previous studies
maxillary third molar reaches stage 8. In addition to also reported that the growth of maxillary tuberosity
the good occlusion of the maxillary third molar, starting affected the successful eruption of the maxillary third
treatment at Nolla stage 7 would be advantageous for molar.17,28 Vardimon et al28 investigated the growth of
shortening the overall treatment time. the maxillary tuberosity and reported interaction be-
The PTv-Mx1 area corresponds to the maxillary tu- tween the dental and skeletal components. They
berosity space. Chipman demonstrated that there must described that if the growth of maxillary tuberosity
be an adequate maxillary tuberosity space for normal was insufficient, the maxillary third molar was distally
eruption of maxillary molars.17 In this study, PTv-Mx1 angulated to compensate for the deficient space, specif-
at T1 was the smallest in the P group. Similarly, previous ically by the presence of the curve of Spee. This study
studies reported that insufficient space for eruption also reported that the maxillary third molar in the P
caused maxillary third molar impaction.26,27 Security of group showed a larger distal angulation than the

American Journal of Orthodontics and Dentofacial Orthopedics November 2023  Vol 164  Issue 5
644 Kim et al

Table VI. Factors affecting the good occlusion of the Table VII. Factors affecting the good occlusion of the
maxillary third molars maxillary third molars
Odd ratio (95% confidence Odd ratio (95% confidence
Variables interval) P value Variables interval) P value
Age at T0 (y) 0.796 (0.679-0.933) 0.00048** PTv-Mx1 change 2.577 (1.579-4.206) 0.0002**
Age at T1 (y) 0.642 (0.511-0.807) 0.0001*** (mm)
Duration T0-T1 (y) 0.986 (0.810-1.201) 0.8876 :Mx3-vOP at T2 (
) 0.936 (0.0875-1.001) 0.0545
Nolla’s stage at T0 0.714 (0.537-0.948) 0.02* VMx3-OP at T2 (mm) 0.724 (0.513-1.022) 0.0667
PTv-Mx1 at T0 (mm) 1.067 (0.949-1.201) 0.2794 Nolla stage at T1 0.718 (0.508-1.015) 0.0606
PTv-Mx1 at T1 (mm) 1.248 (1.096-1.422) 0.0008*** Nolla stage 8-10 y 0.132 (0.028-0.637) 0.0116*
PTv-Mx1 change 2.871 (1.812-4.55) \0.0001***
(mm) Note. A multiple logistic regression analysis with good occlusion as
T1 the dependent variable was used.
:Mx3-Mx1 (
) 1.004 (0.982-1.026) 0.7285 PTv, vertical reference line passing through the pterygomaxillary
:Mx3-vOP (
) 0.997 (0.974-1.021) 0.8143 fissure point; T0, at the time of maxillary second molar extracted;
VMx3-OP (mm) 0.924 (0.814-1.05) 0.2271 T1, the time of complete eruption of maxillary third molar; OP,
HMx3-Mx1 (mm) 1.174 (0.999-1.38) 0.0519 occlusal plane; Mx3, maxillary third molar; vOP, the line perpendic-
T2 ular to the OP; VMx3-OP, vertical distance from OP to mid-point of
:Mx3-Mx1 (
) 0.937 (0.886-0.99) 0.0211* the maxillary third molar.
y
:Mx3-vOP (
) 0.938 (0.895-0.984) 0.009** The reference for the odd ratio of Nolla stages 8-10 was Nolla stages
VMx3-OP (mm) 0.671 (0.519-0.867) 0.0023** 4-7; *P \0.05; **P \ 0.01.
HMx3-Mx1 (mm) 1.122 (0.436-2.891) 0.8112
T1
Nolla stage 4 Reference could be related to the insufficient space for the maxil-
Nolla stages 5 and 6 0.587 (0.154-2.243) 0.4363 lary third molar in the P group at T1.
Nolla stage 7 1.583 (0.349-7.174) 0.5511 The findings show that securing sufficient space for the
Nolla stages 8-10 0.118 (0.02-0.686) 0.0173* maxillary third molar eruption and Nolla’s stage were the
Nolla stages 4-7 Reference
most influential factors for good occlusion of the maxillary
Nolla stages 8-10 0.154 (0.043-0.553) 0.0042**
third molar. From the developmental stage of the maxillary
Note. A simple logistic regression analysis with good occlusion as the
third molar, maxillary second molar extraction could be
dependent variable was used.
T0, at the time of maxillary second molar extracted; T1, the time of recommended before Nolla stage 8. However, it must be
complete eruption of maxillary third molar; PTv, vertical reference remembered that 31.1% of the maxillary third molar at
line passing through the pterygomaxillary fissure point; vOP, the stages 4-7 showed poor occlusion, which needed
line perpendicular to the occlusal plane; OP, occlusal plane; additional orthodontic treatment. The limitation of this
VMx3-OP, vertical distance from OP to mid-point of the maxillary
study is that 2-dimensional radiographs, which have
third molar; HMx3-Mx1, the horizontal distance between the mesial
surface of the maxillary third molar (Mx3Hc) to the distal surface of inherent errors such as distortion and magnification,
maxillary first molar (Mx1Hc). were used. More accurate measurements could be obtained
*P \0.05; **P \ 0.01; ***P \0.001. using 3D images from cone-beam computed tomography.

CONCLUSIONS
maxillary third molar in the G group. Hence, the maxil-
This study evaluated the occlusal status of the maxil-
lary third molar could be aligned better in G and A
lary third molar that erupted after the maxillary second
groups because of the larger maxillary tuberosity space
molar extraction. Good-to-acceptable occlusion was
than in the P group. Sufficient growth at the maxillary
achieved in 65.4% of the maxillary third molar after
tuberosity played an important role in the occlusion of
the maxillary second molar extraction. The maxillary
the maxillary third molar.
third molar in Nolla stage 8 or higher at the time of
Chipman reported that if the maxillary third molar
maxillary second molar extraction and insufficient in-
was at a higher position, the possibility of its impaction
crease in maxillary tuberosity space were negatively
could increase. He suggested that the optimal angula-
related to the achievement of good occlusal status.
tion of the long axis of the maxillary third molar to the
OP should range between 0
and 30
, with the maxillary
AUTHOR CREDIT STATEMENT
third molar tipped distally.17 However, in this study, the
initial position of the maxillary third molar did not affect Insil Kim contributed to conceptualization, method-
occlusion after the eruption. The maxillary third molar in ology, formal analysis, investigation, original draft
the P group was more distally tipped and higher posi- preparation; Sun-Hyung Park contributed to conceptu-
tioned to the OP at T1. As mentioned previously, these alization, validation, manuscript review and editing, and

November 2023  Vol 164  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Kim et al 645

visualization; Yoon Jeong Choi contributed to supervi- 13. Iijima S, Yoshida M, Terada K. Treatment of dentoskeletal bimax-
sion; Ji Hyun Lee contributed to supervision; Chooryung illary protrusion: additional extraction of maxillary second molars.
Odontology 2009;97:115-9.
J. Chung contributed to supervision; and Kyung-Ho Kim
14. Kojima K, Endo T, Shimooka S. Effects of maxillary second molar
contributed to conceptualization and project adminis- extraction on dentofacial morphology before and after anterior open-
tration. bite treatment: a cephalometric study. Odontology 2009;97:43-50.
15. Chung KR, Choo H, Lee JH, Kim SH. Atypical orthodontic extrac-
tion pattern managed by differential en-masse retraction
against a temporary skeletal anchorage device in the treatment
SUPPLEMENTARY DATA
of bimaxillary protrusion. Am J Orthod Dentofacial Orthop 2011;
Supplementary data associated with this article can 140:423-32.
be found, in the online version, at https://doi.org/10. 16. Chapin WC. The extraction of maxillary second molars to reduce
growth stimulation. Am J Orthod Oral Surg 1939;25:1072-8.
1016/j.ajodo.2023.03.021.
17. Chipman MR. Second and third molars: their role in orthodontic
therapy. Am J Orthod 1961;47:498-520.
REFERENCES 18. Asai M, Tomochika A, Asai Y. Molar extraction: a perspective on a
clinical study. Orthod Waves 2007;66:41-51.
1. Bishara SE, Ortho D, Burkey PS. Second molar extractions: a re- 19. Orton-Gibbs S, Crow V, Orton HS. Eruption of third permanent
view. Am J Orthod 1986;89:415-24. molars after the extraction of second permanent molars. Part 1:
2. Graber TM. Maxillary second molar extraction in Class II malocclu- assessment of third molar position and size. Am J Orthod Dento-
sion. Am J Orthod 1969;56:331-53. facial Orthop 2001;119:226-38.
3. Haas AJ. Let’s take a rational look at permanent second molar 20. De-La-Rosa-Gay C, Valmaseda-Castell on E, Gay-Escoda C. Sponta-
extraction. Am J Orthod Dentofacial Orthop 1986;90:361-3. neous third-molar eruption after second-molar extraction in ortho-
4. Lehman R. A consideration of the advantages of second molar ex- dontic patients. Am J Orthod Dentofacial Orthop 2006;129:337-44.
tractions in orthodontics. Eur J Orthod 1979;1:119-24. 21. Moffitt AH. Eruption and function of maxillary third molars after
5. Liddle DW. Second molar extraction in orthodontic treatment. Am extraction of second molars. Angle Orthod 1998;68:147-52.
J Orthod 1977;72:599-616. 22. Smith R. The effects of extracting upper second permanent molars on
6. Wilson HE. Extraction of second permanent molars in orthodontic lower second permanent molar position. Br J Orthod 1996;23:109-14.
treatment. Orthodontist 1971;3:18-24. 23. Magness WB. Extraction of second molars. J Clin Orthod 1986;20:
7. Isaacson RJ, Lindauer SJ, Rubenstein LK. Case report: four perma- 519-22.
nent second molar extractions. Angle Orthod 1993;63:87-90. 24. Casko JS, Vaden JL, Kokich VG, Damone J, James RD,
8. Basdra EK, Stellzig A, Komposch G. Extraction of maxillary second Cangialosi TJ, et al. Objective grading system for dental casts
molars in the treatment of Class II malocclusion. Angle Orthod and panoramic radiographs. American Board of Orthodontics.
1996;66:287-92. Am J Orthod Dentofacial Orthop 1998;114:589-99.
9. Harnick DJ. Case report: Class II correction using a modified Wilson 25. Nolla C. The development of the human dentition. ASDC J Dent
bimetric distalizing arch and maxillary second molar extraction. Child 1960;27:254-66.
Angle Orthod 1998;68:275-80. 26. 
Artun J, Thalib L, Little RM. Third molar angulation during and af-
10. Aras A. Class II Correction with the modified sagittal appliance and ter treatment of adolescent orthodontic patients. Eur J Orthod
maxillary second molar extraction. Angle Orthod 2000;70:332-8. 2005;27:590-6.
11. Escoto M, O’Shaughnessy T, Yerkes I. Removal of second molars to 27. Almpani K, Kolokitha OE. Role of third molars in orthodontics.
alleviate crowding and to facilitate orthodontic treatment. Funct World J Clin Cases 2015;3:132-40.
Orthod 2000;17:32-6. 28. Vardimon AD, Shoshani K, Shpack N, Reimann S, Bourauel C,
12. Jones H. Second molar extraction therapy–two case reports. Funct Brosh T. Incremental growth of the maxillary tuberosity from 6 to
Orthod 2000;17:17-20. 20 years-a cross-sectional study. Arch Oral Biol 2010;55:655-62.

American Journal of Orthodontics and Dentofacial Orthopedics November 2023  Vol 164  Issue 5