CLINICAL RESEARCH

Debora R. Dias, DDS,*

Accuracy of High-resolution Lilian C. V. Iwaki, PhD,*
Alexandre C. A. de Oliveira,
Small-volume Cone-Beam DDS,† Felipe S. Martinha ~o,
DDS, Robson M. Rossi, PhD,x
‡

Computed Tomography in the Maurício G. Arau
jo, PhD,* and

Roberto M. Hayacibara, PhD*
Diagnosis of Vertical Root
Fracture: An In Vivo Analysis

ABSTRACT
SIGNIFICANCE
Introduction: The purpose of this in vivo study was to evaluate the accuracy of small-volume
cone-beam computed tomographic (CBCT) imaging compared with conventional periapical VRFs have a challenging
radiography (CPR) in the diagnosis of vertical root fractures (VRFs) using exploratory surgery diagnosis. This in vivo study
as the reference standard. Methods: Eighty-two dental records of 85 teeth with suspected aimed to evaluate the
VRFs that underwent CPR, CBCT imaging, and exploratory surgery were included. Two diagnostic accuracy of small-
observers assessed CPR and CBCT images independently for the presence or absence of volume CBCT images
root fractures, and findings from the exploratory surgery were considered the reference compared with CPR for VRFs
standard. Diagnostic sensitivity, specificity, accuracy, and the receiver operating and to assess its limitations.
characteristic curve values were obtained. The effect of single- and multirooted teeth on
diagnostic accuracy as well as the association between clinical symptoms and the presence
of VRFs were also assessed. Results: VRFs were surgically detected in 64 of the 85 teeth
(75.3%), of which 62.5% were multirooted and 76.6% had intracanal posts. CBCT imaging
was more sensitive and accurate (65.6% and 64%) than CPR (27.3% and 40.5%). Both CPR
and CBCT diagnostic accuracies were higher in single- than multirooted teeth. Pain on
percussion, a localized periodontal pocket, and tooth mobility were associated with the
presence of VRFs (P , .05; odds ratio 5 4.15, 13.5 and 4.1, respectively). Conclusions: The
accuracy of CBCT imaging for the diagnosis of VRFs was poor, although it was higher than
with CPR. Multirooted teeth in the presence of intracanal posts may limit its diagnostic
value. (J Endod 2020;46:1059–1066.)

KEY WORDS
Cone-beam computed tomography; diagnostic accuracy; exploratory surgery; periapical
radiograph; vertical root fracture

Vertical root fractures (VRFs) generally present a challenge for clinicians because of their poor prognosis From the *Department of Dentistry and
and complex diagnosis. The final diagnosis is based on a combination of several signs and symptoms, x
Statistics Department, State University of
which most of the time lead to exploratory surgery1–3. Clinically speaking, fractures may resemble Maring
a, Maring
a, Parana
, Brazil; †Private

m, Par
a, Brazil; and ‡Private
Clinic, Bele
periodontal or endodontic lesions, and the most common symptoms are pain on percussion/mastication,

, Paran
a, Brazil
Clinic, Maringa
edema, tooth mobility, isolated and localized periodontal pockets, sinus tracts, and sensitivity on
palpation2,3. Address requests for reprints to Dr
Debora R. Dias, State University of
In conventional periapical radiography (CPR), VRFs are usually represented by a radiolucent line
Maring
a, Avenida Mandacaru nº 1550,
combined with a halo-shaped radiolucency around the root and angular bone loss1,3. Because of the Bloco S-08, 87080-000 Maringa
, PR,
overlap of adjacent structures, 2-dimensional radiographic images are limited, and fracture lines are only Brazil.
visible when the X-ray beam is parallel to the fracture plane or when root fragments are clearly E-mail address: debora.reis.dias@gmail.
separated4. com
0099-2399/$ - see front matter
Cone-beam computed tomographic (CBCT) imaging can provide more accurate 3-dimensional
images with real information concerning the size, shape, and texture of teeth plus the surrounding Copyright © 2020 American Association
of Endodontists.
structures5,6. Several ex vivo studies have shown that CBCT imaging is more accurate than CPR in https://doi.org/10.1016/
detecting root fractures7–9. However, studies with artificially created fractures do not take into account j.joen.2020.04.015

JOE  Volume 46, Number 8, August 2020 CBCT and the Diagnosis of VRF 1059
factors such as the effect of the adjacent Strengthening the Reporting of following parameters: PreXion 3D Elite
tissues, minor movements of the patients, or Observational Studies in Epidemiology operating at 90 kVp and 4 mA with a small-
the possibility of a subtle fracture6,10. guidelines19. volume FOV of 56 ! 52 mm and an isotropic
In vivo studies comparing CBCT voxel size of 0.1 mm in the high-resolution
imaging and CPR for detecting root fractures Sample Selection mode (512 basis images).
have shown greater sensitivity and specificity Data were collected from the dental records of
in CBCT images11–15. However, the ability to patients referred for diagnosis and treatment to Surgical Intervention
visualize fractures is lower than in ex vivo a private practice specialist located in the city All surgical interventions were performed within
studies. CBCT imaging has also of Maring
a, Brazil, between April 2014 and 7 days after CBCT imaging by an experienced
demonstrated important limitations because August 2016. The inclusion criteria were (1) periodontist who was aware of the symptoms
of the presence of metallic artifacts caused by teeth with suspected VRFs, (2) dental records and had access to the radiographic and
metallic posts, restorative materials, and presenting a complete description of the tomographic images. Intrasulcular incisions
titanium implants that can negatively affect associated symptomatology, (3) CPR and were performed in the region with the
the overall quality of the image5,16,17. CBCT images, and (4) exploratory surgery for suspected root fracture, and a mucoperiosteal
Nonetheless, recent studies have shown that diagnostic confirmation (reference standard) flap was raised at the buccal and lingual
CBCT reconstructions assess not only the and treatment. A suspected VRF was defined aspects followed by visual inspection, probing,
fracture itself but also may help in analyzing as the presence of at least 1 clinical sign/ and curettage. Usually in cases of root fracture,
indirect signs of its presence, such as bone symptom, such as a localized periodontal because of the bone defect associated, the
resorption16,17. pocket, pain on percussion, mobility, and a visualization of the fracture line is made easier.
Moreover, limited voxel size and small- sinus tract combined or not with a radiographic Whenever required, high-resolution digital
volume field of view (FOV) selections are the sign, such as halo radiolucency around the photographs were taken during surgery,
imaging modality of choice for root fractures root. allowing the observation of all the faces of the
because they tend to offer the highest possible The following clinical parameters were tooth with the use of image magnification. If a
image resolution5,18. The 3D Accuitomo 170 (J collected from the records: root fracture was detected, the appropriate
Morita Mfg Corp, Kyoto, Japan) is 1 of the best
measures were taken, such as tooth
quality CBCT scans in the market, with a 1 Tooth characteristics: sound/restored, the
extraction, root resection, or crown
superfine voxel size of 0.08 mm. The PreXion presence of endodontic treatment, or
lengthening. In the case of no fractures, the site
3D Elite (TeraRecon, San Mateo, CA) is intracanal posts
was carefully assessed for the presence of any
another high-resolution small-volume CBCT 2 Clinical signs and symptoms: pain on
other pathology, treated accordingly, and
scan with a voxel size of 0.1 mm, and, although percussion, a localized periodontal pocket,
patients were instructed to return for regular
there are a few in vivo studies comparing the sinus tract, and tooth mobility
follow-ups.
3D Accuitomo 170 scan with CPR12,13, no 3 Parafunctional habits
studies using the PreXion 3D Elite have been 4 The presence or absence of root fracture as
conducted so far in the diagnosis of root assessed by exploratory surgery Image Analysis
fractures. CPR scans were photographed (EOS Rebel
Therefore, the aim of the present in vivo T5i; Canon, Tokyo, Japan) in high resolution,
study was to compare the accuracy of CPR Image Acquisition and the images were numbered and randomly
and small-volume CBCT images in the CPR (Timex 70E X-Rays; Gnatus, Ribeir~ao organized in a PDF file. CDs with randomly
diagnosis of VRFs using exploratory surgery as Preto, SP, Brazil [70 kVp, 7 mA]) were obtained organized Digital Imaging and
the reference standard. As secondary by the same operator using size 2 Carestream Communications in Medicine files containing
objectives, the effects of single- and E-Speed films (Eastman Carestream, the CBCT scans were also prepared. The PDF
multirooted teeth on the diagnostic accuracy Rochester, NY), an exposure time of 0.6 files and CDs were sent to 2 experienced
and the association between clinical signs and seconds, and an X-ray positioner (Indusbelo, dental radiologists previously calibrated and
symptoms with the presence of VRFs were Londrina, PR, Brazil) according to the bisecting working together on CBCT reconstructions for
evaluated. angle technique. All scans were manually more than 8 years. The examiners were blind
processed under standardized conditions. to the clinical signs and symptoms of the teeth
CBCT images were acquired in a to be assessed. CPR images were analyzed
MATERIALS AND METHODS maxillofacial radiology center located in the city using Adobe Reader software (Adobe
This diagnostic accuracy cross-sectional of Maringa
using the same CBCT unit with the Systems, San Jose, CA). CBCT axial, coronal,
study was designed to assess the
effectiveness of clinical, radiographic, and
TABLE 1 - The Absolute and Relative Distribution of Teeth with and without Fractures according to the Tooth Location
tomographic evidence in the diagnosis of
and Tooth Characteristics
VRFs. The present study followed the
Helsinki Declaration, and it was approved by Root fracture, n (%)
the Ethics Committee for Research Involving
Variables Present, n 5 64 (75.3) Absent, n 5 21 (24.7) Total, N 5 85 (100%)
Human Beings of the State University of
Maringa
, Maringa

, Brazil (CAAE Tooth characteristics
52731515.0.0000.0104). Because of its Single rooted 24 (37.5) 9 (42.9) 33 (38.8)
retrospective nature, signed informed Multirooted 40 (62.5) 12 (57.1) 52 (61.2)
Sound or restored 7 (10.9) 2 (9.5) 9 (10.6)
consent was not required by the committee.
Root filling 8 (12.5) 8 (38.1) 16 (18.8)
This study was conducted in conformance
Intracanal post 49 (76.6) 11 (52.4) 60 (70.6)
with the recommendations of the

1060 Dias et al. JOE  Volume 46, Number 8, August 2020

and cross-sectional reconstructions were TABLE 2 - The Absolute and Relative Distribution of Clinical Signs and Symptoms and Their Association of Teeth with
evaluated using the PrexViewer program and without Fractures
(TeraRecon), with the possibility to adjust for
brightness, contrast, zoom, and rotation. Root fracture, n (%)
The 2 examiners were asked to classify Present, Absent, Total,
the teeth according to the following criteria n 5 64 n 5 21 N 5 85
using a 3-point scale for the diagnosis of VRFs: Variables (75.3) (24.7) (100) OR (95% CI) P value
0, absent; 1, uncertain; and 2, present. The Clinical signs and
images that received a score of 0 or 1 were symptoms
classified as not having any fractures, whereas Pain on percussion 46 (71.9) 8 (38.1) 56 (63.5) 4.15 (1.47–11.69) .01*
those that received a score of 2 did present a Periodontal pocket 54 (84.4) 6 (28.6) 60 (70.6) 13.50 (4.2–43.1) ,.0001*
fracture, which resulted in dichotomic data. Fistula 26 (34.3) 6 (28.6) 32 (37.6) — .46
Mobility 26 (34.3) 3 (14.3) 29 (34.1) 4.1 (1.09–15.37) .05*
Parafunctional habits 42 (65.6) 13 (61.9) 55 (64.7) — .96
Statistical Analysis
Possible associations between the presence *Statistically significant.
of root fracture and clinical signs and
symptoms were assessed using the
chi-square test and odds ratio with 95% (54/64). Pain on percussion, localized DISCUSSION
confidence intervals. Sensitivity, specificity, periodontal pocket, and tooth mobility were
accuracy, positive predictive value, and The objective of the present in vivo study was
statistically associated with the presence of
negative predictive value were calculated for to assess the accuracy of small-volume CBCT
VRFs (P , .05; OR 5 4.15, 13.5, and 4.1,
CPR and CBCT images in relation to the gold imaging compared with CPR in the diagnosis
respectively).
standard (exploratory surgery). Receiver of VRFs using exploratory surgery as the
The diagnostic values for the
operating characteristic (ROC) analysis was reference standard. ROC analysis showed that
identification of root fractures and the areas
used to determine the diagnostic accuracy (Az) CBCT imaging was more sensitive and
under the ROC curve obtained for each
of both modalities in detecting the presence/ accurate in detecting root fractures than CPR.
examiner and the mean CBCT and CPR
absence of VRFs. A separate analysis Sensitivity, sensibility, and accuracy of CPR
assessments are shown in Table 3 and
comparing the diagnostic results for CPR and and CBCT imaging in the analysis of
Figure 1A–C. The mean results demonstrate
CBCT imaging in single- and multirooted teeth multirooted teeth was lower than in single-
that CBCT imaging was more accurate
was also performed. Kappa analysis was rooted ones. Pain on percussion, a localized
(Az 5 0.625) than CPR (Az 5 0.541) in
conducted to assess interexaminer agreement periodontal pocket, and tooth mobility were
detecting VRFs in the suspected teeth. Table 4
considering the following levels: ,0.2 5 poor, shows CBCT and CPR diagnostic values for
statistically associated with the presence of
0.2–0.4 5 moderate, 0.4–0.6 5 good, 0.6– single- and multirooted teeth. Both imaging
root fractures.
0.8 5 very good, and 0.8–1.0 5 excellent20. The first in vivo studies evaluating the
methods clearly presented higher accuracy
The significance level was set at P , .05. All results for single-rooted teeth (Az 5 0.770 and
presence of VRFs in the literature presented
statistical analyses were computed using high sensitivity values (.85%) for CBCT
0.575, respectively) compared with
Bioestat 5.0 software (Sociedade Civil imaging13,21. Similar to the present study,
multirooted teeth (Az 5 0.560 and 0.480,
Mamirau
a; MCT-CNPq, Bel
em, Brazil). Wang et al13 assessed 135 teeth with
respectively). Figure 2A–G illustrates an
suspected root fractures and found that
example of true-positive imaging results
CBCT imaging (3DX Accuitomo, 0.125 voxel
RESULTS compared with exploratory surgery. Examples
size) was much more sensitive (89.5%) than
of different tooth characteristics and the
The records of 85 teeth belonging to 82 CPR (26.3%). Although CPR sensitivity was
visualization of VRFs are illustrated in
individuals, 56 women (68%) and 26 men similar to that found in the present study
Figure 3A–C. The kappa coefficient was 0.50
(32%), with a median age of 55 years (range, (27.3%), our CBCT results were
for CPR and 0.52 for CBCT imaging, indicating
19–79 years) were included in the present considerably lower (65.6%). This difference
good interobserver agreement.
study. VRF was intraoperatively detected in may be associated with the presence of a
64 of the 85 teeth (75.3%), whereas 21
(24,7%) had mimicking conditions
(periodontitis, strip perforations, root TABLE 3 - The Overall Individual and Mean Diagnostic Results Based on Conventional Periapical Radiography (CPR)
resorption, and endodontic complications). and Cone-beam Computed Tomographic (CBCT) Scan Analysis by Examiner A and B in Relation to the Gold Standard
Table 1 shows the absolute and relative (Exploratory Surgery)
distribution of the studied teeth with and
without fractures according to tooth location Diagnostic
and tooth characteristics. Multirooted teeth imaging Examiner SN SP ACC PPV NPV Az (95% CI)
(52/85) and the presence of intracanal posts CPR A 0.359 0.666 0.435 0.766 0.255 0.512 (0.370–0.655)
supporting a crown (60/85) were the most B 0.187 0.952 0.376 0.922 0.376 0.569 (0.431–0.707)
common characteristics. Mean 0.273 0.809 0.405 0.812 0.268 0.541 (0.400–0.681)
Clinical signs and symptoms and their CBCT A 0.531 0.714 0.576 0.849 0.334 0.622 (0.490–0.754)
association with root fractures (odds ratio [OR]; B 0.781 0.476 0.705 0.818 0.417 0.628 (0.497–0.759)
Mean 0.656 0.595 0.640 0.830 0.363 0.625 (0.493–0.757)
95% confidence interval) are shown in Table 2.
Localized periodontal pockets were found in ACC, accuracy; Az, area under the receiver operating characteristic curve; CI, confidence interval; NPV, negative predictive
approximately 85% of cases of fracture value; PPV, positive predictive value; SN, sensitivity; SP, specificity.

JOE  Volume 46, Number 8, August 2020 CBCT and the Diagnosis of VRF 1061
FIGURE 1 – ROC curves showing the performance of the radiographic and CBCT assessments of root fractures in vivo. (A ) Examiner A, (B ) examiner B, and (C ) examiner A and
examiner B mean results.

higher number of teeth with root fillings presented higher sensitivity and accuracy findings, whereas our examiners were blind to
(endodontic treatment and intracanal posts) values of 65.6% and 64%, respectively. this. After 2 weeks, they were asked to detect
in the present sample (90% vs 36%). Chavda Although adjacent bone loss has not been a fracture line only on the axial slides;
et al12 evaluated the presence of root assessed, as examiners evaluated the sensitivity, accuracy, and area under the ROC
fractures in 21 teeth using digital diagnosis of VRFs, adjacent bone loss is a curve values for the detection of VRFs were
radiography and CBCT imaging (3D common sign of fracture and may have significantly (P , .05) lower (17%, 54% and
Accuitomo 170, 0.125 voxel size) and found helped to locate the fractured tooth face. 52%, respectively).
poor sensitivity with both imaging modalities Byakova et al16 also investigated the The majority (62%) of teeth affected with
(16% and 27%, respectively). presence of VRFs on 3D Accuitomo 170 root fractures in the present study were
Recent studies have also associated scans using a 0.16 voxel size and a sample multirooted teeth. The literature shows that
indirect CBCT findings (bone loss patterns) very similar to the present study (88 root filled maxillary and mandibular molar teeth are more
with the presence of VRFs. Zhang et al17 teeth, 65 positive and 23 negative for VRFs). commonly affected by root fractures than
investigated the presence of subtle VRFs in Five blinded observers assessed the CBCT anterior teeth2,3. PradeepKumar et al3
29 endodontically treated teeth (27 positive images in 2 sessions. In the first one, they suggested that posterior teeth are 5 times more
and 2 negative) using the NewTom VG unit assessed CBCT images associated to clinical likely to present root fractures within the first 5
(QR srl, Verona, Italy; voxel size 5 0.125 mm). data and found higher sensitivity, accuracy, years after endodontic treatment compared
Sensitivity and accuracy were low (33.3% and and area under the ROC curve values for a with anterior teeth. These differences may be
37.9%, respectively) based on the direct VRF diagnosis (84%, 81%, and 0.84, attributed to the root anatomy of posterior teeth
finding of a fracture line on CBCT images. respectively) than in the present study and the greater masticatory forces to which
However, vertical buccopalatal bone loss was (65.6%, 64%, and 0.625, respectively), which they are submitted. Being aware that those
found in 25 of 27 teeth with VRFs. Our study may be attributable to the associated clinical teeth are the most affected, a further analysis
was conducted to assess the influence of the
number of roots in the identification of root
TABLE 4 - The Overall Individual and Mean Diagnostic Results Based on Conventional Periapical Radiography (CPR)
fractures. To the best of our knowledge, this is
and Cone-beam Computed Tomographic (CBCT) Scan Analysis by Examiner A and B of Single- and Multirooted Teeth in
the first time that such an analysis has been
Relation to the Gold Standard (Exploratory Surgery)
conducted. ROC analysis showed poorer
accuracy for multirooted teeth compared with
Diagnostic imaging Examiner SN SP ACC PPV NPV Az (95% CI)
single-rooted ones using both methods. CBCT
Single-rooted teeth, n 5 33 imaging was again superior to CPR in the
CPR A 0.42 0.89 0.55 0.90 0.37 0.655 (0.527–0.782)
diagnosis of root fractures in both groups.
B 0.33 0.67 0.42 0.72 0.28 0.5 (0.356–0.643)
These clear differences between single- and
Mean 0.37 0.78 0.48 0.81 0.32 0.575 (0.437–0.712)
CBCT A 0.50 0.78 0.57 0.85 0.37 0.64 (0.510–0.769) multirooted teeth might be associated with
B 0.92 0.89 0.92 0.95 0.81 0.905 (0.842–0.967) metal artifacts and the overlapping of multiple
Mean 0.71 0.83 0.74 0.91 0.52 0.77 (0.665–0.874) roots and neighboring skeletal structures in the
Multirooted teeth, n 5 52 upper molars8,13,15.
CPR A 0.32 0.50 0.36 0.68 0.18 0.41 (0.265–0.554) The PreXion 3D Elite used in the present
B 0.10 1.0 0.30 1.0 0.25 0.55 (0.410–0.689) study has a small FOV and a superfine voxel of
Mean 0.21 0.75 0.33 0.73 0.22 0.48 (0.336–0.623) 0.1 mm. This combination has been shown to
CBCT A 0.55 0.67 0.57 0.84 0.30 0.61 (0.476 – 0.743) positively influence the diagnostic accuracy of
B 0.70 0.33 0.61 0.77 0.24 0.515 (0.372 – 0.657)
root fractures22. Nevertheless, other
Mean 0.62 0.50 0.59 0.80 0.28 0.56 (0.421 – 0.699)
studies12,13,23 using high-resolution units (3D
ACC, accuracy; Az, area under the receiver operating characteristic curve; CI, confidence interval; NPV, negative predictive Accuitomo 170, voxel size 5 0.08 mm) and
value; PPV, positive predictive value; SN, sensitivity; SP, specificity. ours suggest that although small voxel sizes

1062 Dias et al. JOE  Volume 46, Number 8, August 2020

FIGURE 2 – A 40-year-old woman with a single crown in the right maxillary lateral incisor with the presence of endodontic treatment and an intracanal post. The patient presented with
pain on percussion, an isolated periodontal pocket, and a fistula. Exploratory surgery confirmed the presence of a VRF. Both examiners reported the presence of fracture with CBCT
imaging (true positive), but both of them reported the absence of fracture with CPR (false negative). (A ) A clinical view of the affected tooth, (B ) the CPR image, (C ) the CBCT sagittal
view, (D ) the CBCT coronal view, (E ) the CBCT axial view, (F ) the PrexViewer interface, and (G ) exploratory surgery.

may generally provide higher sensitivity and metallic artifacts mimicking the fracture line can to create artifacts, particularly in multirooted teeth,
accuracy in the detection of VRFs, the also result in CBCT false positives and impair the can compromise image quality and impair image
characteristics of the teeth studied can also visualization of actual ones8,11,13,15,24,25. The interpretation3,8,13,21,22,25.
significantly impact the accuracy of the results of the present study attest to the Thus, the clinician should always
diagnosis of root fractures, limiting the superiority of CBCT imaging over CPR, which has approach the diagnosis of root fractures with
generalizability of results. already been demonstrated by previous in vivo caution, carefully taking into account the clinical
The presence of gutta-percha or intracanal studies. However, they also indicate that CBCT signs and symptoms in association with
metallic posts have been shown to negatively scans, regardless of their isotropic voxel size, imaging data in an attempt to confirm the
affect accurate identification of root fractures both have limitations because of the inherent nature of diagnosis of VRFs2,12,13,16. In this study, the
with CPR and CBCT imaging7–9,21,23,24. When teeth with suspected root fractures. The high presence of an isolated and localized
the fracture line is superimposed by the prevalence of radiopaque materials such as periodontal pocket was the strongest indicator
radiopaque material, CPR results can generate gutta-percha, intracanal posts, artificial crowns, of root fractures. Approximately 85% of all
false negatives. Moreover, the formation of and other metallic restorations, which are bound fractured teeth (54/64) presented an isolated

JOE  Volume 46, Number 8, August 2020 CBCT and the Diagnosis of VRF 1063
FIGURE 3 – The effect of different tooth characteristics and the number of roots in the diagnosis of VRFs using CPR and CBCT images (axial, cross-sectional, and coronal re-
constructions). Note that all teeth had associated bone loss. (A ) The right maxillary first premolar, a multirooted tooth without endodontic filling; note the fracture line can be observed in
all reconstructions. (B ) The left maxillary lateral incisor, a single-rooted tooth with endodontic treatment and an intracanal post; note the influence of metal artifacts in the image
visualization, although in the cross-sectional reconstruction, the fracture line in the buccal region might be observed. (C ) The right mandibular first molar, a multirooted tooth with
endodontic treatment and an intracanal post; a direct visualization of the fracture line is difficult.

1064 Dias et al. JOE  Volume 46, Number 8, August 2020

periodontal pocket, showing a high association The sample selected may be CONCLUSION
(OR 5 13.5) and thus corroborating previous considered as being composed by cases of
The accuracy of high-resolution CBCT imaging
studies that found such pockets in 81% of difficult identification. Sound or restored teeth
for the diagnosis of VRFs was poor, although it
cases3. Apart from that, pain on percussion and represented only about 10% of the present
was higher than CPR. CPR and CBCT imaging
tooth mobility also showed a significant sample, whereas around 90% had been
are useful tools in the diagnosis of root
correlation with the presence of root fractures, endodontically treated with or without
fractures; nevertheless, their accuracy was
which is also in agreement with previous intracanal posts and 61.2% were multirooted,
limited by the high incidence of intracanal
studies3,14,15. Cohen et al26 found pain on which seem to have had an important influence
metallic posts and multirooted teeth in the
percussion was associated with the presence of on the results. However, the identification of
sample. Moreover, localized periodontal
root fracture in 70% of the cases, similar to the root fractures is no simple matter, and the
pockets were highly associated with root
present study (72%). sample perhaps reflects the general clinical
fractures.
Because of their poor prognosis, reality of this condition. One of the limitations of
extraction is the usual course of treatment for the sample concerns the difference between
root fractures1,13,26. As a result, clinicians have the number of teeth with and without fractures, CREDIT AUTHORSHIP
to be extremely cautious while interpreting which may have had a negative effect on the CONTRIBUTION STATEMENT
diagnostic imaging because a false-positive specificity and negative predictive value of the
Debora R. Dias: Conceptualization,
diagnosis could condemn a treatable tooth. In imaging diagnosis16. Furthermore,
Methodology, Investigation, Project
case a definitive diagnosis may not be interexaminer agreement was around 0.5 for
administration, Writing - original draft, Writing -
reached, despite the clinical, radiographic, and both imaging methods, which may be
review & editing. Lilian C.V. Iwaki:
tomographic evidence, exploratory surgery considered less than ideal. Nonetheless, it
Methodology, Validation, Writing - review &
may be required5,13,14. As part of the shows that the identification of root fractures
editing. Alexandre C.A. de Oliveira:
treatment, curettage followed by visual and can be difficult even for experienced
Investigation, Data curation. Felipe S.
clinical inspection with a clinical probe and radiologists and is indicative of the subjective
Martinha ~ o: Investigation, Data curation.
digital photography were conducted in all teeth nature of interpreting both CPR and CBCT
Robson M. Rossi: Investigation, Formal
in the present study. Of the 85 teeth with signs imaging. Similar agreements have also been
analysis, Writing - review & editing. Maurício
and symptoms consistent with root fracture observed in previous in vivo studies12,16,17,
G. Arau
jo: Methodology, Writing - review &
initially included in the study, 64 were especially in teeth with endodontic fillings.
editing. Roberto M. Hayacibara:
confirmed as fractured during the surgical CPR, CBCT imaging, and exploratory surgery
Conceptualization, Resources, Methodology,
intervention. Despite not being conclusive, were performed within a short period of time,
Writing - original draft, Supervision.
CBCT images were used to assist the surgeon particularly the surgical intervention, which was
in choosing the appropriate surgical approach always conducted shortly after CBCT imaging
by identifying the root that should be inspected in a maximum period of 7 days in order to avoid
ACKNOWLEDGMENTS
and reducing the amount of surgical trauma on possible alterations in the outcome of the The authors deny any conflicts of interest
the adjacent tissues. disease27. related to this study.

REFERENCES
1. Tsesis I, Rosen E, Tamse A, et al. Diagnosis of vertical root fractures in endodontically treated
teeth based on clinical and radiographic indices: a systematic review. J Endod 2010;36:1455–8.

2. Cohen S, Berman LH, Blanco L, et al. A demographic analysis of vertical root fractures. J Endod
2006;32:1160–3.
3. PradeepKumar AR, Shemesh H, Jothilatha S, et al. Diagnosis of vertical root fractures in restored
endodontically treated teeth: a time-dependent retrospective cohort study. J Endod
2016;42:1175–80.
4. Tsesis I, Kamburog lu K, Katz A, et al. Comparison of digital with conventional radiography in
detection of vertical root fractures in endodontically treated maxillary premolars: an ex vivo study.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;106:124–8.

5. Salineiro FCS, Kobayashi-Velasco S, Braga MM, Cavalcanti MG. Radiographic diagnosis of root
fractures: a systematic review, meta-analyses and sources of heterogeneity. Dentomaxillofac
Radiol 2017;46:20170400.

6. Patel S, Durack C, Abella F, et al. Cone beam computed tomography in Endodontics - a review.
Int Endod J 2015;48:3–15.

7. Mora MA, Mol A, Tyndall DA, Rivera EM. In vitro assessment of local computed tomography for
the detection of longitudinal tooth fractures. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
2007;103:825–9.

8. Hassan B, Metska ME, Ozok AR, et al. Detection of vertical root fractures in endodontically treated
teeth by a cone beam computed tomography scan. J Endod 2009;35:719–22.

JOE  Volume 46, Number 8, August 2020 CBCT and the Diagnosis of VRF 1065
 9. €
Ozer SY. Detection of vertical root fractures of different thicknesses in endodontically enlarged teeth
by cone beam computed tomography versus digital radiography. J Endod 2010;36:1245–9.

10. Makeeva IM, Byakova SF, Novozhilova NE, et al. Detection of artificially induced vertical root
fractures of different widths by CBCT in vitro and in vivo. Int Endod J 2016;49:980–9.
11. €
Metska ME, Aartman IH, Wesselink PR, Ozok AR. Detection of vertical root fractures in vivo in
endodontically treated teeth by cone-beam computed tomography scans. J Endod
2012;38:1344–7.
12. Chavda R, Mannocci F, Andiappan M, Patel S. Comparing the in vivo diagnostic accuracy of
digital periapical radiography with cone-beam computed tomography for the detection of vertical
root fracture. J Endod 2014;40:1524–9.
13. Wang P, Yan XB, Lui DG, et al. Detection of dental root fractures by using cone-beam computed
tomography. Dentomaxillofacial Radiol 2011;40:290–8.

14. Kajan ZD, Taromsari M. Value of cone beam CT in detection of dental root fractures.
Dentomaxillofac Radiol 2012;41:3–10.

15. Bernardes RA, de Moraes IG, Hu
ngaro Duarte MA, et al. Use of cone-beam volumetric
tomography in the diagnosis of root fractures. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
2009;108:270–7.

16. Byakova SF, Novozhilova NE, Makeeva IM, et al. The accuracy of CBCT for the detection and
diagnosis of vertical root fractures in vivo. Int Endod J 2019;52:1255–63.
17. Zhang L, Wang T, Cao Y, et al. In vivo detection of subtle vertical root fracture in endodontically
treated teeth by cone-beam computed tomography. J Endod 2019;45:856–62.

18. Tolentino ES, Amoroso-Silva PA, Alcalde MP, et al. Accuracy of high-resolution small-volume
cone-beam computed tomography in detecting complex anatomy of the apical isthmi: ex vivo
analysis. J Endod 2018;44:1862–6.

19. von Elm E, Altman DG, Egger M, et al. The strengthening the reporting of observational studies in
epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin
Epidemiol 2008;61:344–9.

20. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics
1977;33:159–74.

21. Dutra KL, Pache^co-Pereira C, Bortoluzzi EA, et al. Influence of intracanal materials in vertical root
fracture pathway detection with cone-beam computed tomography. J Endod 2017;43:1170–5.
22. Wanderley VA, Neves FS, Nascimento MC, et al. Detection of incomplete root fractures in
endodontically treated teeth using different high-resolution cone-beam computed tomographic
imaging protocols. J Endod 2017;43:1720–4.
23. Neves FS, Freitas DQ, Flores Campos PS, et al. Evaluation of cone-beam computed tomography
in the diagnosis of vertical root fractures: the influence of imaging modes and root canal materials.
J Endod 2014;40:1530–6.

24. Edlund M, Nair MK, Nair UP. Detection of vertical root fractures by using cone-beam computed
tomography: a clinical study. J Endod 2011;37:768–72.

25. Schulze R, Heil U, Gross D, et al. Artefacts in CBCT: a review. Dentomaxillofac Radiol
2011;40:265–73.
26. Cohen S, Blanco L, Berman L. Vertical root fractures: clinical and radiographic diagnosis. J Am
Dent Assoc 2003;134:434–41.
27. Chang E, Lam E, Shah P, Azarpazhooh A. Cone-beam computed tomography for detecting
vertical root fractures in endodontically treated teeth: a systematic review. J Endod
2016;42:177–85.

1066 Dias et al. JOE  Volume 46, Number 8, August 2020