ORIGINAL CONTRIBUTIONS
Does 3-dimensional imaging of
the third molar reduce the risk of
experiencing inferior alveolar nerve
injury owing to extraction?
A meta-analysis
Adri Cl-Ovejero, DDS; Alba Snchez-Torres, DDS, MS;
Octavi Camps-Font, DDS, MS; Cosme Gay-Escoda, MD,
DDS, MS, PhD, EBOS, OMFS; Rui Figueiredo, DDS, MS,
PhD; Eduard Valmaseda-Castelln, DDS, MS, PhD, EBOS
I
nferior alveolar nerve (IAN) injuries are 1 of the
most critical complications that can occur as a
result of mandibular third-molar (M3M) extrac-
tion1-5; IAN injuries can cause neurosensory
impairment of the lower lip and chin area, which clearly
affects the patients quality of life.1 This complication,
which affects from 0.4% to 5.5%6 of patients, is usually
temporary, but on occasion, it also can lead to perma-
nent symptoms. The risk of experiencing nerve injury is
higher in cases in which the neurovascular bundle is
exposed during surgery.3,7
Some factors related to the surgical technique and
the surgeons experience could have an impact on the
patients risk of experiencing IAN injuries.3,7 In
addition, some investigators have described radio-
graphic warning signs.2 The most important predictor
seems to be the proximity of the M3M roots to the
mandibular canal (MC).1,6
Traditionally, clinicians have used panoramic
radiographic (PR) images to assess the relationship
between a patients M3M roots and the MC. In
patients who have a high risk of experiencing IAN
injuries, owing to the M3M being in close proximity
with the MC, clinicians usually recommend obtain-
ing computed tomographic (CT) images so that the
surgeon can have a preoperative, 3-dimensional (3D)
Copyright 2017 American Dental Association. All rights reserved.
ABSTRACT
Background. Clinicians generally use panoramic radio-
graphic (PR) images to assess the proximity of the mandibular
third molar to the inferior alveolar nerve (IAN). However, in
cases in which a patient needs to undergo a third-molar
extraction, many clinicians also assess computed tomographic
(CT) images to prevent nerve damage.
Types of Studies Reviewed. Two of the authors inde-
pendently searched MEDLINE (through PubMed), Cochrane
Library, Scopus, and Ovid. The authors included randomized
or nonrandomized longitudinal studies whose investigators
had compared the number of IAN injuries after third-molar
extraction in patients who had undergone preoperative CT
with patients who had undergone only PR.
Results. The authors analyzed the full text of 26 of the 745
articles they initially selected. They included 6 studies in the
meta-analysis. Four of the studies had a high risk of bias, and
the investigators of only 1 study had used blinding with the
patients. The authors observed no statistically signicant dif-
ferences between groups related to the total number of nerve
injuries (risk ratio, 0.96; 95% condence interval, 0.50 to 1.85;
P .91). The prognosis of the injuries was similar for both
groups.
Conclusions and Practical Implications. Although
having preoperative CT images might be useful for clinicians
in terms of diagnosing and extracting mandibular third mo-
lars, having these CT images does not reduce patients risk of
experiencing IAN injuries nor does it affect their prognosis.
Key Words. Third molar; computed tomography; pano-
ramic radiography; mandibular nerve; alveolar nerve.
JADA 2017:-(-):---
http://dx.doi.org/10.1016/j.adaj.2017.04.001
JADA ( )
- - http://jada.ada.org - 2017 1
 ORIGINAL CONTRIBUTIONS
view of the area.1 Nevertheless, the investigators of
some studies have concluded that the use of 3D imaging
does not seem to reduce the number of nerve injuries.1-
4,6
In addition, 3D imaging is associated with higher
costs1,6 and higher levels of radiation exposure
compared with PR.1,4 In spite of these facts, some cli-
nicians systematically indicate obtaining preoperative
CTs before performing M3M extraction to avoid legal
issues. Therefore, a meta-analysis of the published data
would be of great interest to clinicians. Consequently,
the aim of this study was to determine whether
obtaining preoperative CT images reduces either the
risk of experiencing or the severity of IAN injuries after
M3M extraction in comparison with obtaining PR
images.
METHODS
We ensured that the methodology of our study adhered
to the Preferred Reporting Items for Systematic Reviews
and Meta-Analyses statement.8
Study selection criteria. We included randomized
and nonrandomized controlled trials and prospective
and retrospective cohort studies whose investigators had
compared the number of IAN injuries that patients
had experienced after undergoing M3M extraction with
whether the patients had undergone preoperative CT or
whether they had undergone only preoperative PR. We
applied no restrictions regarding language or publication
date. We excluded all of the articles that did not meet
these criteria.
The main outcome variable was the number of IAN
injuries for each group; we dened an IAN injury as a
loss of sensation in the lower lip or chin areas, either
subjectively reported by the patient or assessed by means
of clinical testing.
The secondary variables were:
- type of lesion: We considered the lesion to be
persistent if the symptoms lasted longer than 6 months;
- risk of experiencing nerve injury, estimated by using
a previously obtained radiographic assessment: We
classied the level of risk of experiencing IAN injury as
being moderate if there was a superimposition of the
M3M root and the MC, if there was at least 1 radio-
graphic warning sign (according to Rood and Shehabs
criteria9), or both. If we noted that there was more than
1 radiographic warning sign, we considered that the
patient had a high risk of experiencing IAN injury as a
result of undergoing M3M extraction.
Search strategy. We conducted an electronic search
of articles published up to March 12, 2017, in MEDLINE
(PubMed), Cochrane Library, Scopus, and Ovid
databases.
We used the following search strategies for each
database: (((Tomography, X-Ray Computed[Mesh])
OR Cone-Beam Computed Tomography[Mesh]) AND
(Molar, Third[Mesh]) OR Mandibular Nerve/
2 JADA ( )
- - http://jada.ada.org - 2017
injuries[Mesh]) for MEDLINE (PubMed), (X Ray
Computed Tomography Scanner OR Cone Beam
Computed Tomography) AND (Third Molar OR
Mandibular Nerve) for the Cochrane Library,
(Computed Tomography, X Ray OR Cone Beam
Computed Tomography) AND (Third Molar OR
Mandibular Nerve) for Scopus, and (X Ray
computed tomography or Cone Beam computed to-
mography) and (Third Molar or Mandibular nerve
injury) for Ovid.
We completed the search by manually screening the
references cited in the selected articles and reviews.
Selection of studies. Two reviewers (A.C.-O.,
A.S.-T.) independently screened the title and abstract of
each article to decide its eligibility. They then assessed
the full text of the selected articles. Figure 18 lists the
studies we removed at this stage and the reasons for
exclusion. A third reviewer (O.C.-F.) resolved any dis-
agreements. We calculated the k statistic to measure the
reviewers level of agreement.
In situations in which the reviewers identied multi-
ple reports with the same sets of patients, they included
only the study with the longest follow-up time.
Data extraction and method of analysis. Two re-
viewers (A.C.-O., A.S.-T.) independently extracted the
data using data extraction tables. Whenever possible,
they retrieved the following information from the
selected articles: names of authors, year of publication,
country in which the study was conducted, study
design, and details associated with the participants,
interventions, and outcomes.
We considered the number of IAN injuries to be the
primary outcome variable. The secondary outcomes
comprised the type of lesion and the preoperative esti-
mation of the risk of experiencing nerve injury.
Risk of bias assessment. We assessed the risk of bias
according to the guidelines provided in the Cochrane
Handbook for Systematic Reviews of Interventions,
Version 5.1.0,10 and we performed the data extraction
and meta-analysis with Review Manager software,
Version 5.3 (Cochrane Collaboration). We used the
Newcastle-Ottawa Scale11 to assess the cohort studies.
Statistical analysis. We carried out the statistical
analysis using Review Manager software. For dichoto-
mous outcomes, we used risk ratios (RR) with 95%
condence intervals (CI) to estimate the effect of the
operation. We used parametric and nonparametric tests
(Pearson c2 test and Fisher exact test) to compare the
groups. We set the level of signicance at a P value of less
than .05.
ABBREVIATION KEY. 3D: 3-dimensional. CT: Computed
tomographic. IAN: Inferior alveolar nerve. M3M: Mandibular
third molar. MC: Mandibular canal. NR: Not reported. PR:
Panoramic radiographic. RCT: Randomized controlled trial.
 ORIGINAL CONTRIBUTIONS

Records identified Records identified Records identified Records identified

Identification

through MEDLINE through Cochrane through Scopus through Ovid

(PubMed) (n = 298) Library (n = 10) (n = 297) (n = 140)

Records after duplicates removed

(n = 533)
Screening

Records screened Records excluded

(n = 533) (n = 507)

Full-text articles assessed Full-text articles excluded,

for eligibility with reasons (n = 20)
(n = 26) Number of injuries not
Eligibility

specified (n = 4)5,14-16
CT image obtained for all
patients (n = 13)17-29
Case series study
(n = 1)30
Studies included in
Ecological study (n = 1)31
qualitative synthesis
No comparisons made
(n = 6)
between CT and
panoramic radiographic
groups (n = 1)7
Included

Studies included in
quantitative synthesis
(meta-analysis)
(n = 6)

Figure 1. Flow chart of the article selection process for the systematic review and meta-analysis, according to the Preferred Reporting Items for
Systematic Reviews and Meta-Analyses statement.8 CT: Computed tomographic.

We performed a meta-analysis only for the studies
whose investigators had compared similar techniques
and reported the same outcome measures. We conducted
a subgroup analysis taking into consideration the pre-
operative risk of experiencing IAN damage.
We estimated statistical heterogeneity by means of
using the c2 test (Cochran Q test value) and I2 analyses.
We interpreted a c2 test P value of less than .10 and
an I2 value of greater than 50% as having a signicant
level of heterogeneity.12 We selected a xed-effects
model or a random-effects model according to these
values.
If we had found a sufcient number of meta-analyzed
trials (more than 10), we would have performed an
analysis of publication bias, an assessment of clinical
heterogeneity, and an analysis of sensitivity, in accor-
dance with guidelines for analysis by Higgins and
Thompson.12
RESULTS
Study selection and description. As shown in
Figure 1,1-8,13-31 the initial electronic search yielded 745
references. After we removed duplicates, we screened the
abstracts of 533 articles. We selected 26 articles1-7,13-31

JADA ( )
- - http://jada.ada.org - 2017 3
 ORIGINAL CONTRIBUTIONS

of some studies included

Guerrero and colleagues,2 2012 + + + + + +
patients that, on the basis
of the results of preop-
Guerrero and colleagues, 2014 3 + + + + + ? erative testing, had a
high risk of experiencing
Ghaeminia and colleagues,1 2015 + + + + + IAN injury.1,6,13 Sanmart-
Garca and colleagues6
Petersen and colleagues, 2016 4 + + + + + + + selected patients whose
PR images showed su-
Korkmaz and colleagues,13 2017 ? + + + + + perimposition of the MC
and the M3M roots, fully

tio as)

po ta ( tion s)

s)

as
formed roots, and the

an ias

iti ias

as
a

or bia

bi
bi

bi

bi
presence of at least 1 of

er
se on

on

g
tin
c

th
7 radiographic signs
ti

O
ec

ec

tr
en rm

te

ep
at
el

de
described by Rood and
o
(s

(r
f
t(

t(
Shehab9 regarding the
es per
n

g
en

da
io

in
(
t
lm

rt
ou sm
el
ra

e
m proximity between these
nn
ea
ne

le tco

re
nc

o
ge

ss

structures. In addition,
rs

e
co

a
pe

iv
ce

Petersen and colleagues4

e

ct
n

e
en

om
d

et
tio

an
qu

Se
pl
tc

included patients whose

ca

m
se

ts

ou
lo

co
an

PR images showed con-

Al
om

of

In
ip
ng
nd

ic

tact or overlap between

rt

di
Ra

pa

in

the tooth or root complex

Bl
of

and the MC. We consid-

ng
di

ered the patientss

in
Bl

included in the study by

Petersen and colleagues4
Figure 2. Risk of bias assessment, according to the Cochrane Handbook for Systematic Reviews of Interventions, to have a moderate risk of
Version 5.1.0.10 : Low risk of bias. : High risk of bias. ?: Unclear risk of bias. experiencing IAN injury,
although these authors
for full-text analysis, and we included only 6 articles1-4,6,13 did not classify the patients risk of experiencing injury.
in the meta-analysis. The level of agreement between the In both of their studies, Guerrero and colleagues2,3
reviewers was good, with a k statistic index of 0.909. We selected patients who had a moderate risk of experi-
selected 5 randomized controlled trials (RCT)1-4,13 and 1 encing IAN injury. They excluded patients who they
retrospective cohort study.6 considered to have a high risk (with a high probability
After examining the full text of the 26 articles, we of experiencing harm to the neurovascular bundle) and
excluded 20 of the articles for the following reasons: the low risk (no clear radiographic relationship between
investigators did not specify the number of IAN injuries the M3M and the MC) of experiencing IAN injury.
in each group,5,14-16 the investigators performed CT for Quantitative synthesis. The prevalence of IAN
all patients,17-29 the investigators study design did not neurosensory disturbance was 9.3% (52 of 560) for the
comply with the inclusion criteria (case series30 and CT group and 8.3% (44 of 532) for the PR group. We
ecological design31), and the investigators did not detected persistent neurosensory disturbances in 1.5% of
compare the CT and the PR groups.7 patients (7 of 477) for the CT group and 0.9% (4 of 445)
Risk of bias assessment. We classied 41-3,13 of the 5 for the PR group. Table 2 shows the prevalence of IAN
RCTs1-4,13 as having a high risk of bias, owing to the disturbances according to different preoperative classi-
investigators lack of blinding regarding participants and cations of risk.
personnel1-3,13 and lack of providing incomplete outcome We did not nd statistically signicant differences in
data,1 as shown in Figure 2.1-4,13 the prevalence of IAN injury between the CT and PR
We awarded the retrospective cohort study6 4 points groups (RR, 0.96; 95% CI, 0.50 to 1.85; P .91) or
for the selection category, 1 point for the comparability between the studies whose investigators found that pa-
category, and 3 points for the outcome category, totaling tients had a moderate risk (RR, 1.02; 95% CI, 0.27 to 3.86;
8 points; therefore, we classied the results of the cohort P .97) or a high risk (RR, 0.85; 95% CI, 0.33 to 2.23;
study6 as being of high quality. P .75) of experiencing potential damage (Figure 31,3,4,6,13).
Data extraction. Qualitative synthesis. The in- The investigators of 3 studies found persistent in-
vestigators of the selected studies included different types juries. Ghaeminia and colleagues1 observed 5 lesions in
of patients (Table 11-4,6,13). For example, the investigators the CT group and 2 in the PR group that persisted

4 JADA ( )
- - http://jada.ada.org - 2017
 ORIGINAL CONTRIBUTIONS

TABLE 1
Characteristics of the studies included in the meta-analysis regarding
participants, interventions, and outcomes.
STUDY PATIENTS INTERVENTIONS OUTCOMES STUDY
DESIGN
Preoperative Assessment Criteria Radiographic Sample Nerve Persistent Persistent
Risk Technique Size (N) Lesions, Lesions (n) Lesion
No. (%) Follow-up
(mo)
Guerrero and Moderate risk Cases that were not considered CT 43 1 (2.3) 0
Colleagues,2 to be high risk (with a high PR 43 1 (2.3) 0
2012 probability of experiencing
harm to the neurovascular
RCT#
bundle) or low risk (no clear NR
(parallel)
radiographic relationship
between the mandibular third
molar and the MC*) of
experiencing IAN injury
Sanmart- High risk Superimposition of roots and CT 95 15 (15.8) 1
Garcia and MC and any 1 of the 7 PR 55 6 (10.9) 1 Cohort
12
Colleagues,6 radiographic warning signs, (retrospective)
2012 cited by Rood and Shehab**
Guerrero and Moderate risk Cases that were not considered CT 126 2 (1.5) NR
Colleagues,3 to be high risk (with a high PR 130 5 (3.8) NR
2014 probability of experiencing
harm to the neurovascular
RCT (parallel,
bundle) or low risk (no clear NR
multicenter)
radiographic relationship
between the mandibular third
molar and the MC) of
experiencing IAN injury
Ghaeminia High risk Superimposition of roots and CT 156 11 (7.1) 5
and MC covering more than one-half PR 164 9 (5.5) 2 8 RCT (parallel)
Colleagues,1 the height of the MC
2015
Petersen and Moderate risk Superimposition of roots and CT 111 21 (18.9) 1
Colleagues,4 MC PR 116 13 (11.2) 1 6 RCT (parallel)
2016
Korkmaz and High risk Close relationship between the CT 72 3 (4.2) 0
Colleagues,13 MC and the third molar, dened PR 67 11 (16.4) 0
2017 as the presence of at least 1 of
the following 6 radiographic
markers: interruption of the
6 RCT (parallel)
white line of the MC, darkening
of the root, narrowing of the MC
or roots, dark and bid roots,
deected roots, diversion of the
MC
* MC: Mandibular canal.
IAN: Inferior alveolar nerve.
CT: Computed tomographic.
PR: Panoramic radiographic.
NR: Not reported.
# RCT: Randomized controlled trial.
** Source: Rood and Shehab.9

after an 8-month follow-up period, whereas Sanmart-
Garca and colleagues6 and Petersen and colleagues4 each
classied only 1 lesion in each group as being persistent
(at least 12 and 6 months of evolution, respectively).
We did not detect statistically signicant differences in
persistent lesions (RR, 1.64; 95% CI, 0.50 to 5.41; P .42).
In the subgroup analysis of these persistent injuries, no
differences were found in patients who had a high risk
of experiencing injury (RR, 1.82; 95% CI, 0.48 to 6.90;
P .38) and a moderate risk of experiencing injury (RR,
1.05; 95% CI, 0.07 to 16.50; P .98) (Figure 41,2,4,6,13).
DISCUSSION
In our study, we failed to nd signicant differences in
the rate of nerve injuries after M3M extraction between
patients who had undergone preoperative CT and those
who had not undergone preoperative CT. Unfortunately,
there is no consensus on the criteria for the preoperative

JADA ( )
- - http://jada.ada.org - 2017 5
 ORIGINAL CONTRIBUTIONS

TABLE 2 support the hypothesis that obtaining preoperative

Percentages of nerve disturbances. CT images does not appear to lead to a signicant
decrease in the rate of IAN injury after M3M
RISK IAN* INJURIES IAN INJURIES extraction.

CT GROUP PR GROUP
Investigators have identied several risk factors
Overall Persistent Overall Persistent
that can predict a patients likelihood of experi-
Moderate 9.7% (23/237) 0.6% (1/154) 7.3% (18/246) 0.6% (1/159) encing IAN injury after M3M extraction. Absence of
High 9.0% (29/323) 1.9% (6/323) 9.1% (26/286) 1.0% (3/286) cortication,20,28 a dumb-bellshaped MC,23,28 and a
Overall 9.3% (52/560) 1.5% (7/477) 8.3% (44/532) 0.9% (4/445) lingual1,22,29 or interradicular29 position of the MC
* IAN: Inferior alveolar nerve. are among these risk factors. Some investigators
CT: Computed tomographic.
PR: Panoramic radiographic. have stated that injuries occur more easily in MCs
that have a narrow conguration.1 In addition, when
the MC is positioned lingually, the IAN may receive
unfavorable forces if the surgeon starts his approach
assessment of the risk of experiencing IAN injury after by luxating on the buccal side.29 For this reason, it is
undergoing third-molar extraction. Indeed, all of the thought that most IAN injuries are the result of
investigators had established their own criteria to classify compression and traction movements during M3M
patients risk, which made it difcult for us to make surgery.17 In cases such as these, preoperative CT images
comparisons. In addition, the time needed to consider a might provide useful information to the surgeon con-
lesion to be permanent also varied. Many investigators cerning where to apply the elevator.
used the term permanent to dene a nerve injury that A signicant proportion (45.5%) of M3Ms have some
had not recovered by the time of the patients nal degree of superimposition over the MC in PR.7
follow-up visit. For this reason, we determined that using According to Nakamori and colleagues,7 this superim-
the term persistent would be more appropriate, position, darkening of the root, and narrowing of
because investigators could not assess the evolution of the MC on PR images correlate with an absence of
the patients injury after the period of data recording cortication7 and suggest close contact between the
concluded. In our meta-analysis, we considered lesions tooth and the nerve. This direct contact, which can be
that lasted more than 6 months to be persistent. We observed in a CT image, seems to be associated with an
selected this time frame because Cheung and colleagues32 increased risk of experiencing IAN injury.19,23 However,
reported that lesions that lasted more than 6 months had owing to this exposure, only those patients who have
a low probability for recovery. Also, Valmaseda-Castelln a true anatomic relationship between the tooth and
and colleagues,33 in the report of their prospective the nerve have a higher risk of experiencing an IAN
cohort study of 1,117 M3M extractions, observed that le- impairment.26
sions that had not recovered 6 months after surgery were Identifying risk factors for IAN injuries is a key issue
very likely to be permanent. involved in a clinicians decision-making process
Drawbacks related to the results of our report include regarding M3M extraction, because identifying these
the fact that both the number of studies and the size risk factors allows the clinician to identify patients who
of the samples were limited and that the results of 4 of have a high risk of experiencing IAN injuries. In-
the RCTs were associated with a high risk of bias. The vestigators have considered the patients age and, espe-
small number of participants included in these RCTs cially, the anatomic proximity of the roots to the MC, to
might have led to a type 2 error (a false-negative result). be the most relevant variables for predicting this
These drawbacks clearly are related to the low incidence complication.17 Clinicians still consider PR to be the
of IAN complications. Indeed, if we dene an IAN injury criterion standard examination for M3M extraction.
after M3M as being the primary outcome and if we Clinicians do not consider the simple superimposition of
consider a 0.5 difference between groups to be clinically the third-molar roots over the MC to be a sign of a close
signicant, favoring the CT group (as proposed by relationship. In fact, the positive predictive value of IAN
Petersen and colleagues4), we nd that none of the injury in cases of superimposition without additional
studies had a statistical power of greater than 70%. features is low. Even when the clinician nds more
Therefore, the results of our meta-analysis strongly specic images, the positive predictive value still remains
indicate the need for researchers to perform RCTs for small, probably owing to the low incidence of this
which they have established a correct calculation of the complication.29,31 Indeed, only approximately 15% of the
statistical power of the study. Clinicians must take patients who undergo M3M extraction and who have
these limitations into account when considering the had preoperative PR images that suggested that the
results of our study. Even with these limitations, the patient has a high risk of experiencing IAN injuries will
outcomes of our meta-analysis and the results of most experience a neurosensory impairment.6 Although CT
of the studies included in our meta-analysis seem to images are more specic for clinicians to detect the true

6 JADA ( )
- - http://jada.ada.org - 2017
 ORIGINAL CONTRIBUTIONS

CT PR Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight MH, Random, 95% CI MH, Random, 95% CI

1.1.1 Moderate risk of experiencing potential damage to the IAN

Guerrero and colleagues,3 2014 2 126 5 130 11.3% 0.41 (0.08 to 2.09)
Petersen and colleagues,4 2016 21 111 13 116 27.5% 1.69 (0.89 to 3.20)
Subtotal (95% CI) 237 246 38.9% 1.02 (0.27 to 3.86)
Total events 23 18
2 2 2
Heterogeneity: = 0.61; 1 = 2.53, P = .11; I = 60%
Test for overall effect: z = 0.03 (P = .97)

1.1.2 High risk of experiencing potential damage to the IAN

Sanmarti-Garcia and colleagues,6 2012 15 95 6 55 22.2% 1.45 (0.60 to 3.51)
Ghaeminia and colleagues,1 2015 11 156 9 164 22.9% 1.28 (0.55 to 3.02)
Korkmaz and colleagues,13 2017 3 72 11 67 16.1% 0.25 (0.07 to 0.87)
Subtotal (95% CI) 323 286 61.1% 0.85 (0.33 to 2.23)
Total events 29 26
2 2 2
Heterogeneity: = 0.47; 2 = 5.81, P = .05; I = 66%
Test for overall effect: z = 0.32 (P = .75)

Total (95% CI) 560 532 100.0% 0.96 (0.50 to 1.85)

Total events 52 44
2 2 2
Heterogeneity: = 0.30; 4 = 9.12, P = .06; I = 56%
0.01 0.1 1 10 100
Test for overall effect: z = 0.11 (P = .91)
2 2 Favors CT Favors PR
Test for subgroup differences: 1 = 0.05, P = .83; I = 0%

Figure 3. Meta-analysis: inferior alveolar nerve injury after mandibular third-molar extraction. CI: Condence interval. CT: Computed tomographic.
IAN: Inferior alveolar nerve. M-H: MantelHaenszel. PR: Panoramic radiographic.

CT PR Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight MH, Fixed, 95% CI MH, Fixed, 95% CI

2.1.1 Moderate risk of experiencing potential damage to the IAN

Guerrero and colleagues,2 2014 0 43 0 43 Not estimable
Petersen and colleagues,4 2016 1 111 1 116 23.3% 1.05 (0.07 to 16.50)
Subtotal (95% CI) 154 159 23.3% 1.05 (0.07 to 16.50)

Total events 1 1
Heterogeneity: Not applicable
Test for overall effect: z = 0.03 (P = .98)

2.1.2 High risk of experiencing potential damage to the IAN

Sanmarti-Garcia and colleagues,6 2012 1 95 1 55 30.2% 0.58 (0.04 to 9.07)
Ghaeminia and colleagues,1 2015 5 156 2 164 46.5% 2.63 (0.52 to 13.35)
Korkmaz and colleagues,13 2017 0 72 0 67 Not estimable
Subtotal (95% CI) 323 286 76.7% 1.82 (0.48 to 6.90)
Total events 6 3
2 2
Heterogeneity: 1 = 0.86, P = .35; I = 0%
Test for overall effect: z = 0.88 (P = .38)

Total (95% CI) 477 445 100.0% 1.64 (0.50 to 5.41)

Total events 7 4
Heterogeneity: 22 = 0.98, P = .061; I2 = 0%
Test for overall effect: z = 0.81 (P = .42) 0.01 0.1 1 10 100
2 2
Test for subgroup differences: 1 = 0.13, P = .72; I = 0%
Favors CT Favors PR

Figure 4. Meta-analysis: persistent inferior alveolar nerve injury after mandibular third-molar extraction. CI: Condence interval. CT: Computed
tomographic. IAN: Inferior alveolar nerve. M-H: MantelHaenszel. PR: Panoramic radiographic.

anatomic relationship between M3Ms and MCs, the impairment when 3D techniques are used, compared
positive predictive value of CT images is still low and, to with conventional PR.
our knowledge, there are no data suggesting that there According to the results of this meta-analysis, clini-
would be any reduction in the prevalence of IAN cians should not perform CT routinely before M3M

JADA ( )
- - http://jada.ada.org - 2017 7
 ORIGINAL CONTRIBUTIONS
surgery. However, in specic cases in which the clinician
suspects a close relationship between the MC and the
M3M after observing PR images, we recommend
obtaining CT images.18
CONCLUSIONS
According to the results of our meta-analysis, clinicians
should not perform CT routinely before M3M surgery
because using CT images does not seem to reduce the
incidence or affect the patients prognosis of IAN injuries
in comparison with using PR images. n
Dr. Cl-Ovejero is a fellow, Master of Oral Surgery and Orofacial
Implantology, Faculty of Medicine and Health Sciences, University of
Barcelona, Barcelona, Spain.
Dr. Snchez-Torres is a graduate of the Master of Oral Surgery and
Orofacial Implantology, and an associate professor of oral surgery, Faculty
of Medicine and Health Sciences, University of Barcelona, Barcelona; and a
researcher, Bellvitge Biomedical Research Institute, Barcelona, Spain.
Dr. Camps-Font is a graduate of the Master of Oral Surgery and Orofacial
Implantology, and an associate professor of oral surgery, Faculty of Medi-
cine and Health Sciences, University of Barcelona, Barcelona; and a
researcher, Bellvitge Biomedical Research Institute, Barcelona, Spain.
Dr. Gay-Escoda is the chair and a professor, Oral and Maxillofacial
Surgery, Faculty of Medicine and Health Sciences, University of Barcelona,
Barcelona; the director of the Master Degree Program in Oral Surgery and
Implantology, EFHRE (European Foundation for Health Research and
Education) International University/FUCSO (Catalonian Foundation for
Oral Health); a coordinator and researcher, Bellvitge Biomedical Research
Institute, Barcelona; and the head, Department of Maxillofacial Surgery and
Implantology, Teknon Medical Center, Barcelona, Spain.
Dr. Figueiredo is an associate professor, Oral Surgery, Faculty of Medicine
and Health Sciences, University of Barcelona, Barcelona; and a researcher,
Bellvitge Biomedical Research Institute (IDIBELL), Barcelona, Spain.
Address correspondence to Dr. Figueiredo at Faculty of Medicine and
Health Sciences, Campus de Bellvitge, C/Feixa Llarga, s/n; Pavell Govern,
2a planta, Despatx 2.9, 08907 LHospitalet de Llobregat, University of Bar-
celona, Barcelona, Spain, e-mail rui@ruibf.com.
Dr. Valmaseda-Castelln is a professor, Oral Surgery, and a professor,
Master of Oral Surgery and Orofacial Implantology, Faculty of Medicine
and Health Sciences, University of Barcelona, Barcelona; and a researcher,
Bellvitge Biomedical Research Institute, Barcelona, Spain.
Disclosure. None of the authors reported any disclosures.
ORCID Number. Rui Figueiredo: http://orcid.org/0000-0002-2122-6530
The authors conducted this study in association with the Odontological
and Maxillofacial Pathology and Therapeutics Research Group of the
Bellvitge Biomedical Research Institute.
The authors thank Mary Georgina Hardinge for her assistance in
correcting the English in the manuscript.
1. Ghaeminia H, Gerlach NL, Hoppenreijs TJ, et al. Clinical relevance of
cone beam computed tomography in mandibular third molar removal: a
multicentre, randomised, controlled trial. J Craniomaxillofac Surg. 2015;
43(10):2158-2167.
2. Guerrero ME, Nackaerts O, Beinsberger J, Horner K, Schoenaers J,
Jacobs R; SEDENTEXCT Project Consortium. Inferior alveolar nerve
sensory disturbance after impacted mandibular third molar evaluation
using cone beam computed tomography and panoramic radiography: a
pilot study. J Oral Maxillofac Surg. 2012;70(10):2264-2270.
3. Guerrero ME, Botetano R, Beltran J, Horner K, Jacobs R. Can pre-
operative imaging help to predict postoperative outcome after wisdom
tooth removal? A randomized controlled trial using panoramic radiog-
raphy versus cone-beam CT. Clin Oral Investig. 2014;18(1):335-342.
4. Petersen LB, Vaeth M, Wenzel A. Neurosensoric disturbances after
surgical removal of the mandibular third molar based on either panoramic
8 JADA ( )
- - http://jada.ada.org - 2017
imaging or cone beam CT scanning: a randomized controlled trial (RCT).
Dentomaxillofac Radiol. 2016;45(2):20150224.
5. Gallesio C, Berrone M, Ruga E, Boffano P. Surgical extraction of
impacted inferior third molars at risk for inferior alveolar nerve injury.
J Craniofac Surg. 2010;21(6):2003-2007.
6. Sanmart-Garcia G, Valmaseda-Castelln E, Gay-Escoda C. Does
computed tomography prevent inferior alveolar nerve injuries caused by
lower third molar removal? J Oral Maxillofac Surg. 2012;70(1):5-11.
7. Nakamori K, Fujiwara K, Miyazaki A, et al. Clinical assessment of the
relationship between the third molar and the inferior alveolar canal using
panoramic images and computed tomography. J Oral Maxillofac Surg.
2008;66(11):2308-2313.
8. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group.
Preferred reporting items for systematic reviews and meta-analyses: the
PRISMA Statement. PLoS Med. 2009;6(7):e1000097.
9. Rood JP, Shehab BA. The radiological prediction of inferior alveolar
nerve injury during third molar surgery. Br J Oral Maxillofac Surg. 1990;
28(1):20-25.
10. Higgins JP, Green S, eds. Cochrane Handbook for Systematic Reviews
of Interventions, Version 5.1.0 (updated March 2011). The Cochrane
Collaboration; 2011. Available at: www.handbook.cochrane.org. Accessed
April 14, 2017.
11. Wells GA, Shea B, OConnell D, et al. The Newcastle-Ottawa Scale
(NOS) for assessing the quality of nonrandomised studies in meta-
analyses. The Ottawa Hospital Research Institute; 2014. Available at: www.
ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed April 14,
2017.
12. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-
analysis. Stat Med. 2002;21(11):1539-1558.
13. Korkmaz YT, Kaypmaz S, Senel FC, Atasoy KT, Gumrukcu Z. Does
additional cone beam computed tomography decrease the risk of inferior
alveolar nerve injury in high-risk cases undergoing third molar surgery?
Does CBCT decrease the risk of IAN injury? Int J Oral Maxillofac Surg.
2017;46(5):628-635.
14. Sigron GR, Pourmand PP, Mache B, Stadlinger B, Locher MC. The
most common complications after wisdom-tooth removal, part 1: a
retrospective study of 1,199 cases in the mandible. Swiss Dent J. 2014;124(10):
1042-1046, 1052-1056.
15. Roeder F, Wachtlin D, Schulze R. Necessity of 3D visualization for the
removal of lower wisdom teeth: required sample size to prove non-
inferiority of panoramic radiography compared to CBCT. Clin Oral
Investig. 2012;16(3):699-706.
16. Ghaeminia H, Meijer GJ, Soehardi A, et al. The use of cone beam CT
for the removal of wisdom teeth changes the surgical approach compared
with panoramic radiography: a pilot study. Int J Oral Maxillofac Surg. 2011;
40(8):834-839.
17. Selvi F, Dodson TB, Nattestad A, Robertson K, Tolstunov L. Factors
that are associated with injury to the inferior alveolar nerve in high-risk
patients after removal of third molars. Br J Oral Maxillofac Surg. 2013;51(8):
868-873.
18. Hasegawa T, Ri S, Shigeta T, et al. Risk factors associated with inferior
alveolar nerve injury after extraction of the mandibular third molar: a
comparative study of preoperative images by panoramic radiography and
computed tomography. Int J Oral Maxillofac Surg. 2013;42(7):843-851.
19. Xu GZ, Yang C, Fan XD, et al. Anatomic relationship between
impacted third mandibular molar and the mandibular canal as the risk
factor of inferior alveolar nerve injury. Br J Oral Maxillofac Surg. 2013;51(8):
e215-e219.
20. Jun SH, Kim CH, Ahn JS, Padwa BL, Kwon JJ. Anatomical differ-
ences in lower third molars visualized by 2D and 3D X-ray imaging: clinical
outcomes after extraction. Int J Oral Maxillofac Surg. 2013;42(4):489-496.
21. Umar G, Obisesan O, Bryant C, Rood JP. Elimination of permanent
injuries to the inferior alveolar nerve following surgical intervention of the
high risk third molar. Br J Oral Maxillofac Surg. 2013;51(4):353-357.
22. Neves FS, de Almeida SM, Bscolo FN, et al. Risk assessment of
inferior alveolar neurovascular bundle by multidetector computed to-
mography in extractions of third molars. Surg Radiol Anat. 2012;34(7):
619-624.
23. Ueda M, Nakamori K, Shiratori K, et al. Clinical signicance of
computed tomographic assessment and anatomic features of the inferior
alveolar canal as risk factors for injury of the inferior alveolar nerve at third
molar surgery. J Oral Maxillofac Surg. 2012;70(3):514-520.
 ORIGINAL CONTRIBUTIONS
24. Nakayama K, Nonoyama M, Takaki Y, et al. Assessment of the
relationship between impacted mandibular third molars and inferior
alveolar nerve with dental 3-dimensional computed tomography. J Oral
Maxillofac Surg. 2009;67(12):2587-2591.
25. Jhamb A, Dolas RS, Pandilwar PK, Mohanty S. Comparative efcacy
of spiral computed tomography and orthopantomography in preoperative
detection of relation of inferior alveolar neurovascular bundle to the
impacted mandibular third molar. J Oral Maxillofac Surg. 2009;67(1):
58-66.
26. Tantanapornkul W, Okouchi K, Fujiwara Y, et al. A comparative
study of cone-beam computed tomography and conventional panoramic
radiography in assessing the topographic relationship between the
mandibular canal and impacted third molars. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod. 2007;103(2):253-259.
27. Maegawa H, Sano K, Kitagawa Y, et al. Preoperative assessment of
the relationship between the mandibular third molar and the mandibular
canal by axial computed tomography with coronal and sagittal recon-
struction. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2003;96(5):
639-646.
28. Shiratori K, Nakamori K, Ueda M, Sonoda T, Dehari H. Assessment
of the shape of the inferior alveolar canal as a marker for increased risk of
injury to the inferior alveolar nerve at third molar surgery: a prospective
study. J Oral Maxillofac Surg. 2013;71(12):2012-2019.
29. Ghaeminia H, Meijer GJ, Soehardi A, Borstlap WA, Mulder J,
Berge SJ. Position of the impacted third molar in relation to the
mandibular canal: diagnostic accuracy of cone beam computed tomogra-
phy compared with panoramic radiography. Int J Oral Maxillofac Surg.
2009;38(9):964-971.
30. Suomalainen A, Apajalahti S, Vehmas T, Venta I. Availability of
CBCT and iatrogenic alveolar nerve injuries. Acta Odontol Scand. 2013;
71(1):151-156.
31. Susarla SM, Dodson TB. Preoperative computed tomography imaging
in the management of impacted mandibular third molars. J Oral Max-
illofac Surg. 2007;65(1):83-88.
32. Cheung LK, Leung YY, Chow LK, Wong MC, Chan EK, Fok YH.
Incidence of neurosensory decits and recovery after lower third molar
surgery: a prospective clinical study of 4338 cases. Int J Oral Maxillofac
Surg. 2010;39(4):320-326.
33. Valmaseda-Castelln E, Berini-Ayts L, Gay-Escoda C. Inferior
alveolar nerve damage after lower third molar surgical extraction: a pro-
spective study of 1117 surgical extractions. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod. 2001;92(4):377-383.
JADA ( )
- - http://jada.ada.org - 2017 9