YIJOM-3755; No of Pages 9

Int. J. Oral Maxillofac. Surg. 2017; xxx: xxx–xxx

http://dx.doi.org/10.1016/j.ijom.2017.07.010, available online at http://www.sciencedirect.com

Clinical Paper
Oral Surgery

The horizontal inclination angle S. Ishii, S. Abe, A. Moro,

N. Yokomizo, Y. Kobayashi
Department of Dentistry and Oral Surgery,

is associated with the risk of Tokyo Metropolitan Hiroo Hospital,

Shibuya-ku, Tokyo, Japan

inferior alveolar nerve injury

during the extraction of
mandibular third molars
S. Ishii, S. Abe, A. Moro, N. Yokomizo, Y. Kobayashi: The horizontal inclination angle
is associated with the risk of inferior alveolar nerve injury during the extraction of
mandibular third molars. Int. J. Oral Maxillofac. Surg. 2017; xxx: xxx–xxx. ã 2017
International Association of Oral and Maxillofacial Surgeons. Published by Elsevier
Ltd. All rights reserved.

Abstract. The extraction of mandibular third molars can lead to injury to the inferior
alveolar nerve. Hence, it is important to assess the proximity of the root to the
inferior alveolar canal before extraction. The classification system of Pell and
Gregory and the Winter classification are commonly used to evaluate the positional
relationship of the third molar based on radiographs. This retrospective study
involving 105 mandibular third molars was performed to assess whether these
systems reflect the proximity of the root to the canal (based on computed
tomography images), and to identify risk factors for nerve injury. Regarding the
prediction of computed tomography-verified canal invasion, the sensitivity,
specificity, and positive and negative predictive values were high for each Pell and
Gregory category when there was radiographic evidence. The mean distance of
invasion was significantly greater in class III than in class I. However, there were no
Key words: classification; computed tomogra-
significant differences between the Winter inclination categories. The mean phy; high risk; inferior alveolar nerve; mandib-
distance differed significantly between a horizontal inclination angle to the buccal ular third molar; panoramic radiograph.
side of >5 and an angle of 5 . Thus, a horizontal inclination angle >5 represents
a novel risk factor for nerve injury. Accepted for publication 12 July 2017

The extraction of an impacted third molar
(M3) is the most common type of oral
surgery1. Mandibular M3 extraction often
leads to postoperative complications, the
most serious of which is paresthesia of the
lower lip or tongue due to injury to the
inferior alveolar nerve (IAN) or lingual
nerve, respectively1,2. Previous reports
have indicated that the incidence of IAN
injury varies from 0.35% to 8.4% of cases,
while <1% of cases involve permanent
IAN injury1–5.
There are many suggested risk factors
for IAN injury during M3 extraction,
including age, sex, tooth morphology,

0901-5027/000001+09 ã 2017 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
YIJOM-3755; No of Pages 9
2 Ishii et al.
surgeon experience, institutional setting,
surgical technique, method of anaesthesia,
traumatic tissue damage, post-surgical oe-
dema, and wound infection1,5,6. However,
Ghaeminia et al. reported no significant
association between IAN injury and age,
sex, or surgeon experience7. It has been
suggested that the proximity of the man-
dibular M3 to the IAN and their positional
relationship are the most important risk
factors1,5–9. Hence, in order to avoid dam-
age to the IAN, it is necessary to accurate-
ly evaluate the positional relationship
between the M3 and IAN, and to deter-
mine the most appropriate surgical tech-
nique before surgery10.
Panoramic radiographs are currently the
most common type of image used to assess
the risk of IAN injury during mandibular
M3 extraction1,3,11. Although the advan-
tages of panoramic radiographs include
coverage of the skeletal structures of the
entire mouth, low radiation exposure, and
low cost, the drawbacks include low image
resolution, high distortion, and a lack of
information on the buccolingual relation-
ship2. Hence, the exact anatomical relation-
ship (including the buccolingual
relationship) between the mandibular M3
and inferior alveolar canal (IAC) cannot be
detected using panoramic radiographs5.
Nevertheless, panoramic radiographs are
important for gauging the proximity of
the M3 to the IAC5,12. Three-dimensional
imaging using conventional computed to-
mography (CT) or cone beam CT (CBCT) is
recommended when there is radiographic
evidence of an overlap between the root and
IAC in order to verify the exact anatomical
relationship5,12.
Rood and Shehab proposed the use of
seven radiographic signs to predict the risk
of IAN injury during M3 extraction: dark
and bifid root apex, darkening, deflection,
and narrowing of the root, division and
narrowing of the IAC, and interruption of
the white line of the IAC13. If these signs
are present, it is recommended that a CT
scan be performed to verify the positional
relationship between the M3 and IAC1,14.
In addition, several classification sys-
tems have been developed for the assess-
ment of the difficulty of M3 extraction2.
The classification system of Pell and Gre-
gory and the Winter classification are the
most popular such systems. The Pell and
Gregory classification is used to catego-
rize the amount of tooth covered by the
anterior border of the ramus and the depth
of the impaction, and the Winter classifi-
cation is used to categorize the inclination
of the M32,15,16. These systems have been
used in the clinical setting, but there have
been few studies on the relationship
Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
between the M3 and IAC for each of the
categories set out in these classification
systems17–19.
The aim of this study was to assess
whether the Pell and Gregory and Winter
classification systems (which rely on evi-
dence from panoramic radiographs) reflect
the proximity of the M3 to the IAC (veri-
fied using CT images), and to identify risk
factors for IAN injury. These results may
contribute to more accurate diagnosis, im-
proved selection of surgical techniques,
and prevention of IAN injury during M3
extraction. A retrospective study design
was used, because a retrospective dataset
was readily available; this included de-
tailed information on potential risk factors
for IAN injury.
Materials and methods
Study subjects
The institutional review board of Tokyo
Metropolitan Hiroo Hospital approved the
study design. This study involved the in-
vestigation of 105 teeth in 68 randomly
selected patients who had undergone both
panoramic radiography and CT scans at
Tokyo Metropolitan Hiroo Hospital from
April 2015 to March 2016. Patients who
were undergoing mandibular M3 extrac-
tions and those who were not undergoing
this procedure were included. Patients
with cystic lesions, odontogenic tumours,
and other lesions around the mandibular
M3 and ramus were excluded.
Pell and Gregory classification system
The panoramic radiographs were used to
classify the mandibular M3 according to
the Pell and Gregory system15. This sys-
tem classifies the teeth based on the
Table 1. Patient characteristics.
Variable
Age, years, mean  SD (range)
Sex, n
Male
Female
Site of tooth, n
Right
Left
Pell and Gregory class, n (%)
I 23 (21.9)
II
III
Pell and Gregory position, n (%)
A 20 (19.0)
B 72 (68.6)
C 13 (12.4)
SD, standard deviation.
amount of tooth covered by the anterior
border of the ramus (class I–III) and based
on the depth of the impaction relative to
the adjacent tooth (position A, B, or C)
(Table 1).
Winter classification system
The panoramic radiographs were used to
assess the mesiodistal inclination of the
mandibular M3 based on a modified ver-
sion of the Winter system (Fig. 1A)16,20.
The inclination of the M3 was classified
with reference to the angle formed be-
tween the dental longitudinal axis (which
is perpendicular to the occlusal plane) and
the M3 axis (Fig. 1A), as measured using
the high-resolution picture archiving and
communication system SYNAPSE ver-
sion 3.2.1 SR-356 (Fujifilm Medical
Co., Ltd, Tokyo, Japan)20. The inclination
was classified as distoangular ( 6 ),
vertical ( 5 to 4 ), mesioangular (5 to
84 ), horizontal (85 to 94 ), or inverted
(95 )20. The measurements were taken
three times, and the mean values were
used for the analysis.
Distance from the root to the IAC
Using continuous dental CT images, the
proximity of the root to the IAC, or the
invasion of the IAC, was measured. When
the root did not overlap with the IAC, the
distance from the tangential line to the
white line of the IAC (on the CT image,
with the smallest distance) was measured.
When the root invaded the IAC, the dis-
tance of root invasion was calculated by
measuring the distance between the tan-
gential line and the perpendicular line
(Fig. 1B). These measurements were taken
using SYNAPSE.
Value
36.1  10.7 (18–65)
25
43
55
50
58 (55.2)
24 (22.9)
YIJOM-3755; No of Pages 9

Third molar inclination and IAN injury 3

Fig. 1. Measurements based on panoramic radiographs and CT images. (A) Mesiodistal angle of the third molar (on panoramic radiographs) in
terms of the angle between the dental longitudinal axis (which is perpendicular to the occlusal plane) and the third molar axis. (B) Distance from
the root to the IAC (on the consecutive dental CT images showing the smallest distance or greatest invasion) in terms of the distance between the
tangential line and the white line of the IAC or the perpendicular line, if the root overlaps or does not overlap, respectively. (C) Horizontal
inclination angle between the molar line (from the median of the first premolar (or canine in the case of an absent premolar) to the first molar) and
M3 axis line (from the median of the crown contour to the M3 root apex). IAC, inferior alveolar canal.

Relationship between radiographic
evidence of root–IAC overlap and
invasion of the IAC verified in CT images
For each of Pell and Gregory category, this
study evaluated whether the root and IAC
overlapped based on the panoramic radio-
graphs3. The sensitivity, specificity, posi-
tive predictive value (PPV), and negative
predictive value (NPV) of using radio-
graphic evidence of root and IAC overlap
to predict IAC invasion (verified using CT
images) were calculated.
Horizontal inclination angle of the M3
Using SYNAPSE, the horizontal inclina-
tion angle (HIA) of the M3 to the buccal
side of the root was measured with refer-
ence to the angle between the molar line
(from the median of the lower first premo-
lar (or canine in the case of an absent
premolar) to the first molar) and the M3
axis line (from the median of the crown
contour to the M3 root apex) (Fig. 1C).
This measurement was taken three times,
and the mean HIA was used in the
analysis.
Statistical analysis
Bar graphs showing the mean  standard
error of the mean (SEM) distances from
the root to the IAC, and box-and-whisker
plots showing the median and 10th and
90th percentiles, were plotted.
The statistical significance of the differ-
ences in the distance from the root to the
IAC between (1) the classes and positions
of the Pell and Gregory classification sys-
tem, and (2) the inclination categories
based on the Winter classification system
were determined using the Kruskal–
Wallis test, followed by Steel–Dwass
post-hoc test for multiple comparisons.
The Student t-test was used to determine
the statistical significance of the differ-
ences in the distance from the root to
the IAC between (3) classes I + II and
class III, (4) the HIA groups (5 or
>5 ), and (5) the HIA groups (5 or
>5 ) for classes I + II and class III sepa-
rately.
P-values of <0.05 were considered sta-
tistically significant, while those in the
range of 0.05–0.1 were considered to rep-
resent a tendency towards significance.
All of the statistical analyses were per-
formed using EZR (Saitama Medical Cen-
ter, Jichi Medical University, Shimotsuke,
Japan), which is a graphical user interface
for R (version 2.13.0; R Foundation for
Statistical Computing, Vienna, Austria)21.
Results
Study subjects
The patient characteristics are summa-
rized in Table 1. The study involved
105 teeth in 68 patients. The mean age
of the patients was 36.1  10.7 years
(range 18–65 years). There were 25 male
subjects (36.8%) and 43 female subjects
(63.2%). Fifty-five teeth (52.4%) were on

Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
YIJOM-3755; No of Pages 9
4 Ishii et al.
Fig. 2. Relationship between class and position in the Pell and Gregory classification. (A) Percentage of teeth in positions A–C by class. (B)
Percentage of teeth in classes I–III by position.
the right side and 50 (47.6%) were on the
left side.
Pell and Gregory classification system
Based on the Pell and Gregory classifica-
tion system, there were 23 (21.9%) teeth in
class I, 58 (55.2%) in class II, and 24
(22.9%) in class III, and there were 20
(19.0%) in position A, 72 (68.6%) in
position B, and 13 (12.4%) in position C.
For each position of the Pell and Gre-
gory classification system, this study ex-
amined the relationship with class
(according to the Pell and Gregory classi-
fication system), and vice versa. It was
found that the percentage of teeth in posi-
tion A decreased from class I to class III,
while the percentage of teeth in position C
increased (Fig. 2A). However, there was
no trend in the percentage of teeth in
position B (Fig. 2A). Furthermore, the
percentage of teeth in class I decreased
from position A to position C, while the
percentage of teeth in class III increased
(Fig. 2B). However, there was no trend in
the percentage of teeth in class II
(Fig. 2B).
Winter classification system
Most of the teeth were in the mesioangular
position (n = 72, 68.6%), followed by the
horizontal (n = 15, 14.3%), inverted
(n = 12, 11.4%), vertical (n = 3, 2.9%),
and distoangular (n = 3, 2.9%) positions.
For each category of inclination in the
Winter classification system, this study
examined the relationship with both the
class and position according to the Pell
and Gregory classification system. There
Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
was no trend in the percentage of teeth in
each category of inclination from class I to
class III of the Pell and Gregory system
(Fig. 3A). In contrast, the percentage of
teeth in the mesioangular position de-
creased, and the percentages in the hori-
zontal and inverted positions increased
from position A to position C of the Pell
and Gregory system (Fig. 3B). There was
no significant difference between the cat-
egories of inclination in terms of the dis-
tance from the root to the IAC (Fig. 3C).
Relationship between radiographic
evidence of root–IAC overlap and
invasion of the IAC verified in CT images
This study evaluated the sensitivity, spec-
ificity, PPV, and NPV of using radio-
graphic evidence of root and IAC
overlap to predict IAC invasion verified
in CT images. In most cases in which there
was radiographic evidence of root inva-
sion of the IAC, or a suspicion of contact
with the IAC, there was IAC invasion. The
values for class III (sensitivity 82.6%,
specificity 100%, PPV 100%, and NPV
80%) were higher than those for classes I
and II (Table 2).
Table 2. Relationship between radiographic evidence of root–IAC overlap and invasion of the
IAC in CT images for each Pell and Gregory category.
Pell and Gregory class/position Sensitivity
I 60.0
II 57.1
III 82.6
A 63.6
B 64.4
C 66.7
CT, computed tomography; IAC, inferior alveolar canal; NPV, negative predictive value; PPV,
positive predictive value.
Distance from the root to the IAC
according to the Pell and Gregory
classification system
The distance from the root to the IAC was
measured for each class and position of the
Pell and Gregory classification system.
The distance of invasion increased from
class I to class III. There was a significant
difference between class I and class III
(p = 0.004), and a tendency towards a
significant difference between class II
and class III (p = 0.055) (Fig. 4A). There
was also a significant difference in the
comparison of classes I + II to class III
(p = 0.007) (Fig. 4C). Although the dis-
tance of invasion increased from position
A to position C, there was no significant
difference between the positions (Fig. 4B).
Distance from the root to the IAC
according to the horizontal inclination
angle of the M3
It was hypothesized that invasion of the
mandibular M3 into the mandibular ramus
reflects IAC invasion. As the mandibular
bone normally curves to the buccal side at
the leading anterior border of the ramus, it
Specificity PPV NPV
92.3 85.7 25.0
100 100 70.0
100 100 80.0
88.9 87.5 33.3
100 100 61.8
100 100 80
YIJOM-3755; No of Pages 9

Third molar inclination and IAN injury 5

Fig. 3. Winter classification and distance from the root to the IAC. (A) Percentage of teeth in each category of inclination by class. (B) Percentage
of teeth in each category of inclination by position. (C) Distance from the root to the IAC in each category of inclination. The box-and-whisker
plots show the median and the 10th and 90th percentiles. IAC, inferior alveolar canal.

was assumed that the HIA and IAC inva-
sion are related.
Surprisingly, the mean distance from
the root to the IAC in the HIA 5 group
was significantly different to that in the
>5 group: the mean distance in the >5
group was negative (indicating invasion),
while the mean distance in the 5 group
was positive (p = 0.0002) (Fig. 5A). The
percentage of teeth with an HIA >5
increased and the percentage with an
HIA 5 decreased from class I to class
III. In contrast, there was no trend in the
percentage of teeth with an HIA >5 or
5 from position A to position C (data
not shown). Furthermore, in both HIA
groups, the distance of invasion increased
from class I to class III and, particularly in
the >5 group, All classes had invasion
averagely (data not shown).
In both HIA groups, the mean distance
from the root to the IAC was more nega-
tive (indicating invasion) in class III than
in classes I + II (Fig. 5B). However, the
differences were not significant, although
there was tendency towards a significant
difference in the >5 group (p = 0.058)
(Fig. 5B). There was a significant differ-
ence in the mean distance from the root to
the IAC between the 5 and >5 groups
among classes I + II (p = 0.0001), but not
among class III (Fig. 5B).
Discussion
This study found no significant differences
in the distance from the root to the IAC
between the inclination categories (based
on the Winter classification system). How-
ever, the distance of invasion increased
from class I to class III (based on the Pell
and Gregory classification system), and
there was a significant difference between
class I and class III. Moreover, the mean
distance from the root to the IAC in the
HIA 5 group was significantly different
to that in the >5 group: the mean distance
in the >5 group was negative (indicating
invasion), while the mean distance in the
5 group was positive. Therefore, an
HIA >5 (based on CT images) represents
a novel risk factor for IAN injury.
The proportions of teeth in each Pell and
Gregory category in the samples of previ-
ous studies have varied widely: the range
is 10.9–50% for class I, 47.6–72.2% for
class II, and 2.38–18.3% for class III, and
the range is 21.3–51.7% for position A,
35.8–48.4% for position B, and 12.5–
41.8% for position C1,17,19. As the indi-
viduals investigated in this study were
those with both panoramic radiographs
and CT images available, the proportions
do not reflect mandibular M3 overall. It is
suspected that the proportions of teeth in
the various study samples differ depend-
ing on the country and institutional setting
of the respective studies.
Using the Winter classification system,
Costa et al. reported that most of the teeth
in their sample were in the vertical posi-
tion (46.0%), followed by the mesioangu-

Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
YIJOM-3755; No of Pages 9

6 Ishii et al.

Fig. 4. Pell and Gregory classification and distance from the root to the IAC. Distance from the root to the IAC in each (A) class and (B) position.
(C) Comparison of the distance from the root to the IAC between classes I + II and class III. The box-and-whisker plots show the median and the
10th and 90th percentiles. The bar graph shows the mean  standard error of the mean values. *p < 0.05; IAC, inferior alveolar canal.

Fig. 5. HIA and distance from the root to the IAC. (A) Distance from the root to the IAC for each HIA group (5 and >5 ). (B) Comparison of the
distance from the root to the IAC between classes I + II and class III for each HIA group. The bar graph shows the mean  standard error of the
mean values. *p < 0.05; IAC, inferior alveolar canal; HIA, horizontal inclination angle.

Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
YIJOM-3755; No of Pages 9
lar (35.7%) and horizontal (17.5%) posi-
tions19. In contrast, many other studies
have shown that most of the teeth in their
samples were in the mesioangular position
(45.8–56.7%), followed by the vertical
(21.3–35.2%) and horizontal (8.8–
11.6%) positions17,19,20,22. As for the
patients studied in the Pell and Gregory
classification analysis, the patients in the
Winter classification analysis underwent
both panoramic radiography and CT im-
aging, so they do not reflect the entire
population of individuals with mandibular
M3. Blondeau and Daniel reported that
almost all IAN paresthesia cases involved
the extraction of a M3 in the mesioangular
position17. However, this study found no
significant differences in the distance from
the root to the IAC between the mesioan-
gular position and other positions (data not
shown). Other studies have suggested that
there is no association between the incli-
nation of the M3 and the risk of IAN
injury17,23,24, which is in agreement with
the results of this study.
After Rood and Shehab proposed the
use of seven radiographic signs for the
prediction of IAN injury during M3 ex-
traction13, several studies were performed
to explore the risk factors further. Huang
et al. reported that three of the radiograph-
ic signs (interruption of the white line of
the IAC and division and narrowing of the
IAC) were significantly associated with
IAN sensory impairment, but a dark and
bifid root apex and darkening, deflection,
and narrowing of the root were not1. In
contrast, Kim et al. reported that a dark
and bifid root apex, darkening, deflection,
and narrowing of the root, and narrowing
of the IAC were all significant risk fac-
tors3. However, Gomes et al. reported no
significant associations between the radio-
graphic signs and IAN injury after man-
dibular M3 extraction25. In a previous
study using CT images, Peker et al.
reported that darkening of the root and
interruption of the white line of the IAC on
radiographs were significantly associated
with contact between the M3 and IAC on
CT Images5. Similarly, Nakagawa et al.
reported that interruption of the white line
was significantly associated with contact
between the M3 and IAC on CT Images26.
The conflicting results among previous
studies indicate that the seven radio-
graphic signs do not accurately predict
the likelihood of IAN injury. As pan-
oramic radiographs do not allow accurate
diagnoses to be made (in order to accu-
rately determine the risk), this study only
used them to examine whether the root
and IAC overlapped. In most cases in
which there was overlap on the panoram-
Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
Third molar inclination and IAN injury
ic radiographs, the CT images revealed
invasion of the IAC (Table 2). Rud
reported that the incidence of paralysis
in the cases involving no radiographic
evidence of overlap between the root
and the IAC was 0.4%, but when there
was overlap, the incidence was 2.4%, and
when the root pressed against the IAC,
the incidence was 23%27. As radiographic
evidence of the presence or absence of
overlap between the root and IAC is very
important, conventional CT or CBCT is
recommended in cases involving overlap
in order to verify the proximity of the root
to the IAC.
Previous reports have suggested that the
occurrence of paresthesia due to IAN in-
jury during M3 extraction increases with
the impaction depth (based on the posi-
tions in the Pell and Gregory classification
system)3,17. Kim et al. suggested that the
impaction depth (which correlates with the
amount of bone shaved off and the diffi-
culty in obtaining sufficient space to per-
form the extraction) could increase the
risk of IAN injury because it is associated
with surgical complexity3. However, this
study found no significant differences in
the distance from the root to the IAC
between positions A–C. This study evalu-
ated the root invasion of the IAC using CT
images but did not explore the rate of
postoperative paralysis or the complexity
of the surgical techniques. On the other
hand, the distance of invasion increased
from class I to class III, and there was a
significant difference between class I and
class III (p = 0.004) and a tendency to-
wards a significant difference between
class II and class III (p = 0.055). This
may occur, in part, because the mandibu-
lar M3 is positioned at the distal end of the
mandibular body where it is connected to
the thin mandibular ramus2, and the M3 is
more deeply embedded in the mandibular
ramus from class I to class III.
Beaty and Le used CT images to evalu-
ate the mandibular thickness in young
dentate men and women at several points:
symphysis, halfway between symphysis
and the mental nerve, the mental nerve,
halfway between the mental nerve and
antegonial notch, the antegonial notch,
gonion, and the ramus–condylar neck bor-
der28. They found that the thickness was
acceptable at the point halfway between
the mental nerve and the antegonial notch
(men 10.33 mm, women 9.45 mm), but it
decreased at the antegonial notch (men
7.27 mm, women 7.10 mm) and gonion
(men 5.42 mm, women 5.39 mm)28. This
suggests that from class I to class III, the
impacted tooth invades further into the
narrower mandibular ramus, which may
7
influence the likelihood of the root invad-
ing the IAC.
Additionally, Kwon et al. used CBCT
images to investigate the position of the
IAC in the mandibular ramus, and they
found that at the contact point between the
occlusal plane and anterior border of the
mandibular ramus, the horizontal, vertical,
and nearest distances to the IAC were
13.07  2.45 mm, 14.24  2.41 mm, and
10.12  1.76 mm, respectively29. These
results indicate that the distance between
the mandibular ramus and IAC is short.
Taken together, the results of previous
studies suggest that the more posterior
the impacted M3 is in the mandibular
ramus, the narrower the region, which
increases the likelihood of root invasion
of the IAC from class I to class III.
It was found in the present study that
when the HIA to the buccal side of the root
was >5 , the mean distance was negative
(indicating invasion), while the mean dis-
tance in the 5 group was positive. In
addition, in classes I + II, the distance was
significantly different in the 5 group
(i.e., it was positive) compared to that in
the >5 group (i.e., it was negative, indi-
cating invasion) (p = 0.0001). Moreover,
in the >5 group, the M3 in class III had a
greater distance of invasion, with a ten-
dency towards significance (p = 0.058),
compared to the M3 in classes I + II
(Fig. 5). The mandibular bone curves to
the buccal side at the leading anterior
border of the mandibular ramus. The
HIA to the buccal side suggests that the
impacted M3 exists along with the man-
dibular ramus. Although this study did not
evaluate factors such as thickness, length,
and number of M3 roots, it appears that the
risk of invasion of the IAC is greater if the
M3 is angled in the direction of the man-
dibular ramus.
The results of this study indicate that
patients with an HIA 5 , radiographic
evidence of no overlap between the root
and IAC, and a M3 in class I or II have a
low risk of IAN injury. In contrast,
patients with an HIA >5 and a M3 in
class I or II have a moderate risk, and those
with a M3 in class III and radiographic
evidence of overlap between the root and
IAC have a high risk of IAN injury. How-
ever, it is proposed that CT scans should
be obtained for any cases in which the root
and IAC overlap.
Recently, a two-step method and a cor-
onectomy technique have been suggested
as ways to avoid IAN injury when extract-
ing teeth4,30,31. The two-step technique
involves the removal of the crown of
the M3 to create a space for root migration
and, after 3–4 months, the extraction of the
YIJOM-3755; No of Pages 9
8 Ishii et al.
M3 (which will have detached and migrat-
ed from the IAC)30. The second technique,
coronectomy31–33, involves only remov-
ing the crown of the M3, and leaving the
residual root to ankylose. The advantage
of this technique is that a second operation
is generally not required.
In high-risk cases in which the root
invades the IAC, surgeons should extract
the M3 (after informed consent is obtained
from the patient) using an appropriate
technique (i.e., the two-step technique or
coronectomy) based on evidence from CT
images, or they should leave completely
embedded teeth where they are in cases
involving asymptomatic patients. This
study revealed that there were no signifi-
cant differences in the distance from the
root to the IAC between the Winter incli-
nation categories. However, the invasion
distance was greatest for class III (com-
pared to class I). In addition, the mean
distance from the root to the IAC in the
HIA 5 group was significantly different
to that in the >5 group, which makes an
HIA >5 (based on CT images) a novel
risk factor for IAN injury.
In most cases involving radiographic
evidence of root and IAC overlap, there
was evidence of IAC invasion. When there
is radiographic evidence of an overlap of
the root and IAC, conventional CT or
CBCT is recommended to verify the exact
proximity of the root to the IAC.
Funding
None.
Ethical approval
This study design was approved by the
institutional review board of Tokyo
Metropolitan Hiroo Hospital (expedited
review approval number 2015-11).
Patient consent
Not required.
Competing interests
None.
Acknowledgements. We thank Editage
(http://www.editage.com) for English
language editing.
References
1. Huang CK, Lui MT, Cheng DH. Use of
panoramic radiography to predict postsurgical
sensory impairment following extraction of
Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
impacted mandibular third molars. J Chin
Med Assoc 2015;78:617–22.
2. Juodzbalys G, Daugela P. Mandibular third
molar impaction: review of literature and a
proposal of a classification. J Oral Maxillofac
Res 2013;4:e1.
3. Kim JW, Cha IH, Kim SJ, Kim MR. Which
risk factors are associated with neurosensory
deficits of inferior alveolar nerve after man-
dibular third molar extraction? J Oral Max-
iilofac Surg 2012;70:2508–14.
4. Sarikov R, Judzbalys G. Inferior alveolar
nerve injury after mandibular third molar
extraction: a literature review. J Oral Max-
illofac Res 2014;5:e1.
5. Peker I, Sarikir C, Alkurt MT, Zor ZF. Pan-
oramic radiography and cone-beam computed
tomography findings in preoperative exami-
nation of impacted mandibular third molars.
BMC Oral Health 2014;14:71.
6. Xu GZ, Yang C, Fan XD, Yu CQ, Cai XY,
Wang Y, He D. Anatomic relationship be-
tween impacted third mandibular molar and
the mandibular canal as the risk factor of
inferior alveolar nerve injury. Br J Oral Max-
illofac Surg 2013;51:e215–9.
7. Ghaeminia H, Gerlach NL, Hoppenreijs TJ,
Kicken M, Dings JP, Borstlap WA, de Haan
T, Bergé SJ, Meijer GJ, Maal TJ. Clinical
relevance of cone beam computed tomog-
raphy in mandibular third molar removal: a
multicentre, randomised, controlled trial. J
Craniomaxillofac Surg 2015;43:2158–67.
8. Kipp DP, Goldstein BH, Weiss WW. Dys-
esthesia after mandibular third molar surgery:
a retrospective study and analysis of 1,377
surgical procedures. J Am Dent Assoc
1980;100:185–92.
9. Monaco G, Montevecchi M, Bonetti GA,
Gatto MR, Checchi L. Reliability of panoram-
ic radiography in evaluating the topographic
relationship between the mandibular canal
and impacted third molars. J Am Dent Assoc
2004;135:312–8.
10. Flygare L, Ohman A. Preoperative imaging
procedures for lower wisdom teeth removal.
Clin Oral Investig 2008;12:291–302.
11. Ghaeminia H, Meijer GJ, Soehardi A, Bors-
tlap WA, Mulder J, Bergé SJ. Position of the
impacted third molar in relation to the man-
dibular canal. Diagnostic accuracy of cone
beam computed tomography compared with
panoramic radiography. Int J Oral Maxillo-
fac Surg 2009;38:964–71.
12. Feifel H, Riediger D, Gustorf-Aeckerle R.
High resolution computed tomography of the
inferior alveolar and lingual nerves. Neuro-
radiology 1994;36:236–8.
13. Rood JP, Shehab BA. The radiological pre-
diction of inferior alveolar nerve injury dur-
ing third molar surgery. Br J Oral Maxillofac
Surg 1990;28:20–5.
14. Peixoto LR, Gonzaga AK, Melo SL, Pontual
ML, Pontual Ados A, de Melo DP. The effect
of two enhancement tools on the assessment
of the relationship between third molars and
the inferior alveolar canal. J Craniomaxillo-
fac Surg 2015;43:637–42.
15. Pell GJ, Gregory BT. Impacted mandibular
third molars: classification and modified
techniques for removal. Dent Digest
1933;39:330–8.
16. Winter GB. Impacted mandibular third
molars. St Louis: American Medical Book
Co.; 1926: 241–79.
17. Blondeau F, Daniel NG. Extraction of im-
pacted mandibular third molars: postopera-
tive complications and their risk factors. J
Can Dent Assoc 2007;73:325.
18. Akadiri OA, Fasola AO, Arotiba JT. Inci-
dence and risk factors for nerve injuries in
mandibular third molar surgery. Niger J Med
2009;18:402–8.
19. Costa FW, Fontenele EH, Bezerra TP,
Ribeiro TR, Carneiro BG, Soares EC. Cor-
relation between radiographic signs of third
molar proximity with inferior alveolar nerve
and postoperative occurrence of neurosenso-
ry disorders: a prospective, double-blind
study. Acta Cir Bras 2013;28:221–7.
20. Abe H, Toma H, Hanyu Y, Kuroda K,
Takeyama K, Kamagata Y, Oka M. A case
of impacted and inverted mandibular third
molar causing anesthesia and paresthesia of
inferior alveolar nerve (in Japanese). Tokyo
Women’s Med Univ J 1989;59:1359–63.
21. Kanda Y. Free statistical software: EZR
(Easy R) on R commander. http://www.
jichi.ac.jp/saitama-sct/SaitamaHP.files/
statmedEN.html. [Accessibility verified
February 8, 2017].
22. Sedaghatfar M, August MA, Dodson TB.
Panoramic radiographic findings as predic-
tors of inferior alveolar nerve exposure fol-
lowing third molar extraction. J Oral
Maxillofac Surg 2005;63:3–7.
23. Blaeser BF, August MA, Donoff RB, Kaban
LB, Dodson TB. Panoramic radiographic
risk factors for inferior alveolar nerve injury
after third molar extraction. J Oral Maxillo-
fac Surg 2003;61:417–21.
24. Cheung LK, Leung YY, Chow LK, Wong
MC, Chan EK, Fok YH. Incidence of neu-
rosensory deficits and recovery after lower
third molar surgery: a prospective clinical
study of 4338 cases. Int J Oral Maxillofac
Surg 2010;39:320–6.
25. Caldas
Gomes Ade AC, JrVasconcelos
F, Pita NetoBC,
IC. Sensitivity
Silva ED,
and specificity of pantomography to predict
inferior alveolar nerve damage during ex-
traction of impacted lower third molars. J
Oral Maxillofac Surg 2008;66:256–9.
26. Nakagawa Y, Ishii H, Nomura Y, Watanabe
NY, Hoshiba D, Kobayashi K, Ishibashi K.
Third molar position: reliability of panoram-
ic radiography. J Oral Maxillofac Surg
2007;65:1303–8.
27. Rud J. Third molar surgery: relationship of
root to mandibular canal and injuries to the
inferior dental nerve. Tandlaegebladet
1983;87:619–31.
28. Beaty NB, Le TT. Mandibular thickness
measurements in young dentate adults. Arch
YIJOM-3755; No of Pages 9
Otolaryngol Head Neck Surg 2009;135:920–
3.
29. Kwon KH, Sim KB, Lee JM. Evaluation of
the course of the inferior alveolar canal in the
mandibular ramus using cone beam comput-
ed tomography. J Korean Assoc Oral Max-
illofac Surg 2012;38:231–9.
30. Landi L, Manicone PF, Piccinelli S, Raia A,
Raia R. A novel surgical approach to impact-
ed mandibular third molars to reduce the risk
of paresthesia: a case series. J Oral Max-
illofac Surg 2010;68:969–74.
31. Pogrel MA, Lee JS, Muff DF. Coronectomy:
a technique to protect the inferior alveolar
Please cite this article in press as: Ishii S, et al. The horizontal inclination angle is associated with the risk of inferior alveolar nerve
injury during the extraction of mandibular third molars, Int J Oral Maxillofac Surg (2017), http://dx.doi.org/10.1016/j.
Third molar inclination and IAN injury 9
nerve. J Oral Maxillofac Surg Address:
2004;62:1447–52. Shigehiro Abe
32. Renton T, Hankins M, Sproate C, McGurk Department of Dentistry and Oral Surgery
M. A randomised controlled clinical trial to Tokyo Metropolitan Hiroo Hospital
compare the incidence of injury to the infe- 2-34-10 Ebisu
rior alveolar nerve as a result of coronectomy Shibuya-ku
and removal of mandibular third molars. Br J Tokyo 150-0013
Oral Maxillofac Surg 2005;43:7–12.
Japan
Tel: +81 3 3444 1181
33. Kohara K, Kurita K, Kuroiwa Y, Goto S,
Fax: +81 3 3444 3196
Umemura E. Usefulness of mandibular third
E-mail: shigehiroabe@hotmail.com
molar coronectomy assessed through clinical
evaluation over three years of follow-up. Int
J Oral Maxillofac Surg 2015;44:259–66.