DENTOALVEOLAR SURGERY

What Are the Risk Factors for External

Root Resorption of Second Molars
Associated With Impacted Third
Molars? A Cone-Beam Computed
Tomography Study
Melek Tassoker, DDS, PhD
Purpose: In relation to an impacted third molar, which is the most frequently impacted tooth, external
root resorption (ERR) or dental caries can be seen on the distal surface of the adjacent second molar. The
purpose of this study was to investigate the risk factors for ERR in second molars associated with impacted
third molars using cone-beam computed tomography (CBCT).
Materials and Methods: In this cross-sectional study, the presence of ERR on the adjacent second molar
was investigated retrospectively on sagittal and axial CBCT slices of patients who underwent CBCT exam-
ination for diagnostic reasons (surgical removal of third molars, orthodontic assessments, and so on) be-
tween 2013 and 2017 at the Necmettin Erbakan University Faculty of Dentistry, Konya, Turkey. The
primary outcome variable was ERR (yes or no). Predictor variables included demographic (age and gender)
and radiographic (impacted tooth angulation and tooth type) parameters. Given that there are multiple
observations per patient (1 to 4 impacted teeth), 1 third molar per patient was selected at random for study
inclusion. Statistical analysis was performed using descriptive statistics, the Mann-Whitney U test, and the
Pearson c2 test. Logistic regression analysis was used to determine the risk factors for ERR in second molars.
The Cohen k test was used to test intraobserver agreement. The significance level was set at P < .05.
Results: Two hundred patients were analyzed, and of the 200 impacted third molars evaluated, 42 (21%)
showed ERR. The severity of resorption increased with aging (P < .05). Logistic regression analysis indi-
cated that tooth type and inclination of the third molars could predict the occurrence of ERR associated
with an impacted third molar (P < .05).
Conclusions: Mesioangular and horizontal inclinations and impacted mandibular third molars posed a
greater risk of ERR lesions in adjacent second molars than other inclinations (distoangular and vertical)
and maxillary third molars.
Ó 2018 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 77:11-17, 2019

Third molars are the most frequently impacted teeth. If
they are not impacted, they generally erupt between 17
and 21 years of age.1 Removing or extracting a
third molar has been discussed for many years, but
prophylactically removing an asymptomatic third molar
might not be an easy decision.2-4 The harmful effect on
the adjacent second molar should be considered when
making such a decision.5,6 In relation to an impacted
third molar, external root resorption (ERR) or dental
caries can be seen on the distal surface of the
adjacent second molar.7-9 This type of resorption
emerges owing to the pressure from the impacted

Assistant Professor, Department of Oral and Maxillofacial Necmettin Erbakan University, 42050 Konya, Turkey; e-mail:
Radiology, Faculty of Dentistry, Necmettin Erbakan University, dishekmelek@gmail.com
Konya, Turkey. Received March 19 2018
Conflict of Interest Disclosures: None of the authors have any Accepted August 19 2018
relevant financial relationship(s) with a commercial interest. Ó 2018 American Association of Oral and Maxillofacial Surgeons
Address correspondence and reprint requests to Dr Tassoker: 0278-2391/18/30988-1
Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, https://doi.org/10.1016/j.joms.2018.08.023

11
12
tooth.8,9 This pressure causes inflammation and triggers
the resorption via clastic cells.10
Two-dimensional radiographic studies conducted
with panoramic and periapical radiographs have re-
ported a low prevalence of ERR in second molars
(0.3-7%).8,9 Because of image distortion related to
their projective geometry, their diagnostic value is
quite low.11 On the other hand, cone-beam computed
tomography (CBCT) provides 4.3 times greater detec-
tion of ERR than panoramic radiographs.4 This system
is free of overlapping and provides 3-dimensional (3D)
multiplanar imaging with a low radiation dose and
high spatial resolution.11 Using conventional
computed tomography scanning instead of CBCT for
routine analysis of impacted teeth is not justifiable
given the higher radiation exposure.4
The position of the impacted tooth predisposes the
adjacent tooth to ERR or dental caries.12 In the litera-
ture, horizontal and mesioangular impactions were
found to be related to a higher frequency of ERR in
second molars because of larger contact surfaces or
the presence of a gap.4,9
It is difficult to diagnose ERR until periapical inflam-
mation or pulpitis occurs.4,8,11 Its progress is generally
asymptomatic.11 In a recent study, the impaction rate
of third molars was found to be 54.1% in the Turkish
population.13 Considering such a high prevalence of
third molar impaction, it is necessary to identify the
risk factors for ERR and to implement preventive mea-
sures.11 Very few studies have addressed these issues.
The purpose of this study was to assess the ERR of
second molars associated with impacted third molars
using CBCT. We hypothesized that impacted third
molars with different inclinations (mesioangular,
distoangular, horizontal, or vertical) and different loca-
tions (mandibular or maxillary) would have an impact
on the risk of ERR in adjacent second molars. The spe-
cific aim of this study was to investigate the risk factors
for ERR in second molars associated with impacted
third molars using CBCT.
Materials and Methods
STUDY DESIGN
To address the research purpose, we designed and
implemented a cross-sectional study. The study was
performed with the approval of the Necmettin Erba-
kan University Faculty of Dentistry Research Ethics
Committee (decision No. 2017/12). Furthermore, the
analysis was performed according to the stipulations
laid out by the Declaration of Helsinki.
STUDY SAMPLE
The study population was composed of all patients
who underwent CBCT examination for diagnostic rea-
EXTERNAL ROOT RESORPTION OF MOLARS
sons (surgical removal of third molars, orthodontic as-
sessments, and so on) presenting with at least 1
impacted third molar to the Department of Oral and
Maxillofacial Radiology, Faculty of Dentistry, Necmet-
tin Erbakan University, between 2013 and 2017.
The inclusion criteria were as follows: patients
aged 18 years or older, eruption of all teeth except
the third molar(s), and at least 1 impacted third
molar. The exclusion criteria were as follows:
completely erupted third molars with cystic or tu-
mor lesions associated with impacted teeth; third
molars with evidence of root development that is
less than two thirds; second molars with severe
carious lesions, crowns, or distal restorations; or
demonstration of any artifacts.
STUDY VARIABLES
The primary outcome variable was ERR (yes or no).
Predictor variables included demographic (age and
gender) and radiographic (impacted tooth angulation
and tooth type) parameters.
DATA COLLECTION METHODS
The patients were divided into 3 age groups (18 to
24, 25 to 34, and 35 to 50 years). The presence of
ERR on the adjacent second molar was investigated
on sagittal and axial CBCT slices by the same radiolo-
gist. The CBCT scans were acquired using a 3D Accui-
tomo 170 machine (Morita, Kyoto, Japan).
We used the diagnosis of ERR according to Al-
Khateeb and Bataineh.14 The observation of ‘‘a clear
loss of substance in the root of adjacent
second molar teeth due to direct contact between it
and impacted third molar’’ was diagnosed as ERR.
Once detected, ERR was categorized according to its
location (cervical, middle root third, and apical root
third) (Figs 1-3) and severity (slight, moderate, and
severe) (Figs 4-6).9 According to Ericson et al,15 ERR
involving less than half the dentin thickness was
scored as slight; ERR involving at least half the dentin
was scored as moderate; and ERR involving the pulp
cavity was scored as severe.
The differential diagnosis between dental caries and
ERR was based on radiologic appearance. If there
was a clear gap between the second molar and the
dental crown of the third molar and radiolucency
showed irregular morphology, the diagnosis was
dental caries.11
The inclination of the third molars was determined
using the classification of Winter16 (vertical, horizon-
tal, mesioangular, or distoangular). Two weeks later,
examinations were repeated, and these showed good
intraexaminer agreement.
MELEK TASSOKER
FIGURE 1. Sagittal cone-beam computed tomography section
showing external root resorption (arrow) in cervical region.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.
DATA ANALYSIS
Given that there are multiple observations per
patient (1 to 4 impacted teeth), one third molar per
patient was selected at random for study inclusion,
and analysis of 200 patients (200 teeth) was per-
formed. All statistical analysis was performed using
SPSS software (version 21; IBM, Armonk, NY). Mean
FIGURE 2. Sagittal cone-beam computed tomography section
showing external root resorption (arrow) in middle root third region.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.
13
FIGURE 3. Sagittal cone-beam computed tomography section
showing external root resorption (arrow) in apical root third region.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.
and standard deviation values were calculated. Data
were tested for normality using the Shapiro-Wilk
test. Normality was violated, and differences in mean
age between the 2 genders were analyzed by the
Mann-Whitney U test. The Pearson c2 test was used
for categorical variables. In addition, logistic regres-
sion analysis was used to determine the risk factors
for ERR in second molars. Intraexaminer reliability
was tested with the Cohen k test. The k coefficient
was interpreted as being poor (0), slight (0.01 to
0.20), fair (0.21 to 0.40), moderate (0.41 to 0.60), sub-
stantial (0.61 to 0.80), or almost perfect (0.81 to 1.0)
according to Landis and Koch.17 Statistical significance
was assumed for P < .05.
FIGURE 4. Slight external root resorption lesion on axial cone-
beam computed tomography slice.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.
14
FIGURE 5. Moderate external root resorption lesion on sagittal
cone-beam computed tomography slice.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.
Results
We evaluated 200 impacted third molars in 200 pa-
tients. Of the 200 patients, with a mean age of
27.26
8.60 years (age range, 18 to 50 years), 118
were women and 82 were men. The mean age of
male patients (29.11
9.27 years) was higher than
that of female patients (25.97
7.90 years) (P = .018).
The k test showed excellent intraobserver values for
the repeated ERR observations (k = 0.96, P < .05). The
FIGURE 6. Severe external root resorption lesion on axial cone-
beam computed tomography slice.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.
EXTERNAL ROOT RESORPTION OF MOLARS
results of the initial examination were tabulated. The
overall frequency of ERR was 21% (42 teeth).
The severity of resorption increased with aging
(P < .05). Most slight ERR lesions were found in the
group aged 18 to 24 years. Severe ERR lesions were
only seen in the groups aged 25 to 34 years and 35
to 50 years (Table 1). Logistic regression analysis indi-
cated that age and gender could not predict the occur-
rence of ERR associated with an impacted third molar
(P > .05) (Table 1). There was no statistically signifi-
cant relationship between gender and the presence
and severity of ERR (P > .05). Of the 42 ERR lesions,
24 were seen in female patients and 18 in
male patients.
Logistic regression analysis indicated that tooth type
could predict the occurrence of ERR associated with
an impacted third molar (P < .05) (Table 1). ERR was
more prevalent in mandibular second molars than
maxillary second molars, and the risk of ERR was
about 4 times greater in mandibular molars than in
maxillary molars (P < .05). The distribution of ERR
regarding tooth type is presented in Table 1.
There was a statistically significant relationship be-
tween the impaction position of third molars and the
presence of ERR (P < .05). ERR was seen in an
extremely low number of cases of distoangular and
vertical molars. The mesioangular and horizontal posi-
tions of third molars were found to be more at risk of
ERR (Table 1). As indicated in Table 1, the distoangular
and vertical inclinations of impacted third molars pre-
sented less risk of ERR on second molars (P < .05).
The cross tabulation of severity and localization of
ERR is shown in Table 2. The cervical region in
mandibular second molars and the middle third region
in maxillary second molars were most affected by ERR.
Discussion
This study aimed to investigate the risk factors for
ERR in second molars associated with impacted third
molars. The impacted teeth may be related to some
pathologic conditions, such as pericoronitis, dental
caries, bone loss, odontogenic cysts or tumors,
swelling, cheek ulcerations, or ERR of adjacent teeth.4
ERR is not rare and may penetrate into the pulp, which
can lead to inflammation.8,11,18 The extraction of
impacted third molars removes the mechanical stress
from second molars, and the inflammatory process
ends. In cases of bacterial invasion into the pulp,
endodontic treatment is required.19 Our study indi-
cated that mesioangular and horizontal inclinations
of impacted third molars were more at risk of being
associated with ERR than other impaction positions.
This finding reinforces the theory of previous
studies.4,5,11,20-24 One of our patients who had ERR
MELEK TASSOKER 15

Table 1. RESULTS OF REGRESSION ANALYSIS OF ALL STUDIED VARIABLES (200 PATIENTS, 200 TEETH)

ERR Severity, n

Variable No. of Patients OR (95% CI) P Value ERR Presence, n Slight Moderate Severe

Age
18-24 yr 100 1 (reference) 17 16 1 0
25-34 yr 57 1.70 (0.83-3.47) .143 15 10 2 3
35-50 yr 43 1.44 (0.65-3.22) .366 10 4 1 5
Gender
Male 82 1 (reference) 18 11 1 6
Female 118 0.92 (0.52-1.65) .902 24 19 3 2
Tooth type
Upper right second molar 40 1 (reference) 3 2 0 1
Upper left second molar 38 0.78 (0.26-2.36) .670 2 1 0 1
Lower left second molar 62 4.35 (1.93-9.80) <.001* 19 15 1 3
Lower right second molar 60 4.09 (1.80-9.28) .001* 18 12 3 3
Angulation of third molar
Mesioangular 124 1 (reference) 27 20 3 4
Distoangular 20 0.17 (0.03-0.78) .023y 2 1 0 1
Horizontal 35 1.65 (0.78-3.49) .190 12 8 1 3
Vertical 21 0.03 (0.01-0.71) .023y 1 1 0 0
Total 200 42 30 4 8
Abbreviations: CI, confidential interval; ERR, external root resorption; OR, odds ratio.
* The significance level is P < .01.
y The significance level is P < .05.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxillofac Surg 2019.

on a maxillary second molar is described in Figures 7
and 8 before and after tooth removal.
Examination of ERR on the root surface, especially
in the early stages, requires 3D imaging. In a recent
study, CBCT and panoramic radiography were
compared and the 2 methods corresponded in only
4.3% of cases.4 The overall prevalence of ERR was
21% in our study, which was higher than that in previ-
ous reports by Nitzan et al10 and Enabulele and Ob-
uekwe,21 both of whom used periapical radiographs
to find ERR prevalence rates of 8% and 4.7%, respec-
tively. These 2-dimensional methods can cause misin-
terpretation of the images because of overlapping
and distortion.11 Al-Khateeb and Bataineh14 found
that ERR lesions on second molars were present in
only 0.3% of cases when using panoramic radiography.
Oenning et al4 reported that the prevalence of ERR
was 22.88% when using CBCT images, and this was
corroborated by the findings of our study. It should
be noted that the bias of sample selection (unerupted
or partially erupted third molars)5 or a different defini-
tion of ERR14 could yield different results.
The findings of this study, which evaluated both
partially erupted and unerupted third molars, showed
that the cervical region in mandibular second molars
and the cervical and middle third regions in maxillary
second molars were most affected by ERR. In contrast
to the results of this study, Nemcovsky et al,5 who eval-
uated only completely unerupted third molars, re-
ported that the apical third region of the adjacent
second molar was most affected by ERR. The ERR le-
sions most commonly detected in the cervical region
of the second molars were associated with partially
erupted third molars because of their susceptibility
to inflammation.9,10
Our study showed that the risk of ERR was about 4
times greater in mandibular molars than in maxillary
molars and a greater number of ERR lesions (37 teeth)
were found in the mandible than in the maxilla (5
teeth). This result was in accordance with other
studies conducted using panoramic radiographs,18
periapical radiographs,10 and CBCT scans.4 This can
be explained by the complex eruption process of
mandibular third molars.
One limitation of this study was the correct distinc-
tion of ERR lesions from dental caries, as other
researchers previously reported.4 The presence of a
gap between the second and third molars that leads
to food retention may actually suggest dental caries,
especially in cases of partially erupted third mo-
lars.4,7,25 Enabulele and Obuekwe21 indicated that
second molars with cervical resorption had direct
tooth-to-tooth contact with the impacted third molar
whereas most second molars with distal caries did
not have tooth-to-tooth contact. Another limitation
of our study was using only radiologic diagnosis for
16 EXTERNAL ROOT RESORPTION OF MOLARS

Table 2. CROSS TABULATION OF SEVERITY AND

LOCALIZATION OF 42 TEETH WITH ERR

Severity of Cervical, Middle Root Apical Root

ERR n Third, n Third, n

Upper right second molar

Slight 1 1 0
Moderate 0 0 0
Severe 0 1 0
Maxilla 1 2 0
total
Upper left second molar
Slight 1 0 0
Moderate 0 0 0
Severe 0 0 1
Maxilla 1 0 1
total
Lower left second molar
Slight 10 5 0
Moderate 1 0 0
Severe 2 1 0 FIGURE 7. Cone-beam computed tomography examination of
Mandible 13 6 0 external root resorption in apical root third of second molar before
total extraction.
Lower right second molar Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
Slight 9 3 0 lofac Surg 2019.
Moderate 2 1 0
Severe 2 0 1
Mandible 13 4 1
total
Abbreviation: ERR, external root resorption.
Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
lofac Surg 2019.

ERR and not correlating radiology with clinical find-

ings. To provide better scientific evidence for this
study, direct visualization of ERR after tooth extraction
and histopathologic analysis could be useful.
Age was not found to be a risk factor for ERR in our
research. This was inconsistent with the findings of
Wang et al,11 who reported that being over the age
of 35 years was an independent risk factor for ERR. A
cross-sectional study cannot provide full insight into
age as a risk factor, as older individuals may have had
their third molar teeth with ERR extracted earlier
in life.26
Some controversial studies have suggested that men
are more susceptible to the development of ERR than
women, which researchers explained as a difference
in sex hormones.8,10 However, gender was not found
to be a predictive parameter for ERR in our study.
This study showed that there was a tendency to-
ward moderate and severe resorption in older patients
whereas slight ERR lesions tended to be seen in FIGURE 8. Second molar resorption in same patient shown in
younger individuals. This finding is not unexpected Figure 7 after removal of teeth.
because root resorption progresses over time.11,27,28 Melek Tassoker. External Root Resorption of Molars. J Oral Maxil-
The eruptive tooth movement continues, and lofac Surg 2019.
MELEK TASSOKER
mechanical pressure on second molars stimulates the
progression of ERR formation.11
Within the limitations of this study, age and gender
were not found to be independent risk factors for
ERR in second molars. Mesioangular and horizontal
inclinations and impacted mandibular third molars
posed a greater risk of ERR lesions in adjacent
second molars than other inclinations (distoangular
and vertical) and maxillary third molars. Whether
there are symptoms or not, the impacted mandibular
third molars with mesioangular and horizontal inclina-
tions should be examined carefully when making deci-
sions regarding extraction. This was a cross-sectional
retrospective study based on CBCT images, and future
research studies need clinical and histologic validation
to confirm our findings.
References
1. Breik O, Grubor D: The incidence of mandibular third molar im-
pactions in different skeletal face types. Aust Dent J 53:320,
2008
2. Dodson TB: Surveillance as a management strategy for retained
third molars: Is it desirable? J Oral Maxillofac Surg 70:20, 2012
3. Marciani RD: Is there pathology associated with asymptomatic
third molars? J Oral Maxillofac Surg 70:15, 2012
4. Oenning ACC, Neves FS, Alencar PNB, et al: External root resorp-
tion of the second molar associated with third molar impaction:
Comparison of panoramic radiography and cone beam
computed tomography. J Oral Maxillofac Surg 72:1444, 2014
5. Nemcovsky CE, Libfeld H, Zubery Y: Effect of non-erupted 3rd
molars on distal roots and supporting structures of approximal
teeth. A radiographic survey of 202 cases. J Clin Periodontol
23:810, 1996
6. Nunn ME, Fish MD, Garcia RI, et al: Retained asymptomatic third
molars and risk for second molar pathology. J Dent Res 92:1095,
2013
7. Falci SGM, de Castro CR, Santos RC, et al: Association between
the presence of a partially erupted mandibular third molar and
the existence of caries in the distal of the second molars. Int J
Oral Maxillofac Surg 41:1270, 2012
8. Yamaoka M, Furusawa K, Ikeda M, et al: Root resorption of
mandibular second molar teeth associated with the presence
of the third molars. Aust Dent J 44:112, 1999
9. Oenning AC, Melo SL, Groppo FC, et al: Mesial inclination of
impacted third molars and its propensity to stimulate external
root resorption in second molars—A cone-beam computed
tomographic evaluation. J Oral Maxillofac Surg 73:379, 2015
10. Nitzan D, Keren T, Marmary Y: Does an impacted tooth cause
root resorption of the adjacent one? Oral Surg Oral Med Oral
Pathol 51:221, 1981
17
11. Wang D, He X, Wang Y, et al: External root resorption of the sec-
ond molar associated with mesially and horizontally impacted
mandibular third molar: Evidence from cone beam computed to-
mography. Clin Oral Invest 21:1335, 2017
12. Akarslan ZZ, Kocabay C: Assessment of the associated symp-
toms, pathologies, positions and angulations of bilateral occur-
ring mandibular third molars: Is there any similarity? Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 108:e26, 2009
13. Topkara A, Sari Z: Investigation of third molar impaction in
Turkish orthodontic patients: Prevalence, depth and angular
positions. Eur J Dent 7:94, 2013
14. Al-Khateeb TH, Bataineh AB: Pathology associated with
impacted mandibular third molars in a group of Jordanians. J
Oral Maxillofac Surg 64:1598, 2006
15. Ericson S, Bjerklin K, Falahat B: Does the canine dental follicle
cause resorption of permanent incisor roots? A computed tomo-
graphic study of erupting maxillary canines. Angle Orthod 72:
95, 2002
16. Winter GB: Impacted Mandibular Third Molars. St Louis, Amer-
ican Medical Book, 1926
17. Landis JR, Koch GG: The measurement of observer agreement
for categorical data. Biometrics 33:159, 1977
18. van der Linden W, Cleaton-Jones P, Lownie M: Diseases and
lesions associated with third molars. Review of 1001 cases.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 79:142,
1995
19. Mayrink G, Ballista PR, Kinderlly L, et al: External root resorption
associated with impacted third molars: A case report. J Oral
Health Craniofac Sci 2:43, 2017
20. Knutsson K, Brehmer B, Lysell L, et al: Pathoses associated with
mandibular third molars subjected to removal. Oral Surg Oral
Med Oral Pathol Oral Radiol Endod 82:10, 1996
21. Enabulele JE, Obuekwe ON: Prevalence of caries and cervical
resorption on adjacent second molar associated with impacted
third molar. J Oral Maxillofac Surg Med Pathol 29:301, 2017
22. Sewerin I, von Wowern N: A radiographic four-year follow-up
study of asymptomatic mandibular third molars in young adults.
Int Dent J 40:24, 1990
23. Matzen LH, Schropp L, Spin-Neto R, et al: Radiographic signs of
pathology determining removal of an impacted mandibular third
molar assessed in a panoramic image or CBCT. Dentomaxillofac
Radiol 46:20160330, 2017
24. Camargo IB, Sobrinho JB, Andrade ES, et al: Correlational study
of impacted and non-functional lower third molar position with
occurrence of pathologies. Prog Orthod 17:26, 2016
€
25. Ozeç I_, Herg€
uner Siso Ş, Taşdemir U, et al: Prevalence and
factors affecting the formation of second molar distal caries
in a Turkish population. Int J Oral Maxillofac Surg 38:1279,
2009
26. Matzen LH, Schropp L, Spin-Neto R, et al: Use of cone beam
computed tomography to assess significant imaging findings
related to mandibular third molar impaction. Oral Surg Oral
Med Oral Pathol Oral Radiol 124:506, 2017
27. Benenati FW: Root resorption: Types and treatment. Gen Dent
45:42, 1997
28. Larson TD: Causes and treatment of root resorption. Northwest
Dent 89:45, 2010