REVIEW ARTICLE

Amanda Pelegrin Candemil,

DDS, MSc, PhD,*†
Accuracy of Magnetic
Guilherme Nilson Alves dos
Santos, DDS, MSc,*
Resonance Imaging in Clinical
Matheus L. Oliveira, DDS, MSc,
PhD,‡ Filippo Gatti, DDS, MSc,
Endodontic Applications:
PhD,† Yara T. C. Silva-Sousa,
DDS, MSc, PhD,§ Alice C. Silva-
A Systematic Review
Sousa, DDS,*
Jardel Francisco Mazzi-Chaves,
DDS, MSc, PhD,* and
Manoel Damia ~o Sousa-Neto,
DDS, MSc, PhD*
ABSTRACT
SIGNIFICANCE
Introduction: The development of dedicated coils and new magnetic resonance imaging
MRI has high diagnostic value (MRI) sequences has led to an increase in image resolution and a reduction in artifacts.
in the evaluation of tooth and Consequently, numerous studies have demonstrated the utility of MRI as a nonionizing
root canal anatomy, pulp alternative to cone-beam computed tomographic imaging. The aim of this systematic review
vitality, and regeneration, as was to evaluate the accuracy of MRI in clinical applications in endodontics. Methods: A
well as in assessing the effect literature search was conducted using PubMed, Embase, Scopus, and Web of Science. The
of caries on dental pulp, inclusion criteria encompassed studies evaluating MRI applications in endodontics, covering
periapical lesions, and root tooth and root canal anatomy, root canal working length, pulp vitality and regeneration, the
cracks or fracture. The effect of caries on dental pulp, guided endodontics, periapical lesions, and root cracks/
accuracy of MRI is comparable fractures. The selected studies examined both ex vivo and in vivo human teeth using clinical
to or even higher than that of MRI units. Two researchers independently screened the studies, applied the eligibility criteria,
reference standards such as and assessed the potential risk of bias using the revised QUADAS-2 tool (Bristol Medical
CBCT imaging, micro-CT School, University of Bristol, UK). Results: A total of 18 studies were included in this
imaging, and histology. systematic review, demonstrating that the use of MRI has a high diagnostic value in
endodontics. In the evaluation of tooth and root canal anatomy, pulp vitality and regeneration,
the effect of caries on dental pulp, periapical lesions, and root cracks/fractures, MRI’s
accuracy is comparable to or even higher than reference standards such as cone-beam
computed tomographic imaging, micro–computed tomographic imaging, and histology.
Conclusions: MRI has high potential accuracy for diagnosing various clinical endodontic
From the *Department of Restorative tasks, except for root canal length, size of caries, and periapical lesion dimensions, which are
Dentistry, School of Dentistry of Ribeira~o
~o Paulo, Sa~o Paulo,
overestimated in MRI. (J Endod 2024;50:434–449.)
Preto, University of Sa
Brazil; †CentraleSup elec, Ecole Normale
Supe rieure Paris-Saclay, Centre national KEY WORDS
de la recherche scientifique, Laboratoire
de Me canique Paris-Saclay, Universite Dentistry; diagnosis; endodontics; magnetic resonance imaging
Paris-Saclay, Gif-sur-Yvette, France;
‡
Division of Oral Radiology, Department of
Oral Diagnosis, Piracicaba Dental School, During the pre-, intra-, and postoperative stages of endodontic treatment, professionals require imaging
University of Campinas, Piracicaba, Brazil; examinations as a complementary method to aid in diagnosis and treatment planning. Generally, the first-
and §School of Dentistry, University of choice examination is periapical radiography; however, because of its limitation regarding overlapping of
Ribeir~ao Preto, Ribeira~o Preto, Brazil
structures1,2, cone-beam computed tomographic (CBCT) imaging has been frequently indicated despite
Address requests for reprints to Dr its relatively higher radiation dose3,4.
Amanda Pelegrin Candemil, Department
The dose of ionizing radiation delivered to the patient should be considered when requesting an
of Restorative Dentistry, School of
Dentistry of Ribeir~
ao Preto, University of imaging examination, taking into account the purpose of the examination and the individual needs and
~o Paulo, Av do Cafe
Sa -Subsetor Oeste- physical aspects of the patient5. Accordingly, a higher frequency of CBCT indication reflects greater
~o Preto, Sa
11 (N-11), Ribeira ~o Paulo radiation exposure to the population, particularly in young and child populations, who are more sensitive
14040-904, Brazil. to the deleterious effects of X-rays6,7. Thus, the use of nonionizing imaging examinations, especially in
E-mail address: [email protected]
0099-2399/$ - see front matter
young patients and children, is fundamental.
Magnetic resonance imaging (MRI) is formed by the interaction between atomic nuclei, commonly
Copyright © 2024 American Association
of Endodontists.
those of hydrogen, and a powerful static magnetic field. The nuclei capture energy from an oscillating
https://doi.org/10.1016/ magnetic field at a specific resonant frequency. The use of spatially varying magnetic fields enables spatial
j.joen.2024.01.014 encoding of nuclei and the generation of cross-sectional images8. Although there is hydrogen in the water

434 Candemil et al. JOE  Volume 50, Number 4, April 2024

and adipose tissue that make up bones, 2023. To assess all applications of clinical MRI in vivo human teeth using clinical MRI units and
morphologic studies with MRI are still in endodontics, a guiding question and an were indexed in the search.
considered challenging. The mineralized eligibility strategy were used in the review The exclusion criteria covered studies
composition of bone causes the signal emitted process. The guiding question of the study using animal studies, small sample sizes (,5
by it to quickly disappear, requiring specific was the following: Among ex vivo and in vivo teeth), review studies, systematic reviews,
pulse sequences that are not always available teeth, how does MRI compare with 2- editorials, and book chapters. Studies
in devices used daily in diagnostic centers9. dimensional and 3-dimensional images (CBCT unavailable for full-text assessment were also
Furthermore, although dental pulp produces a and micro–computed tomographic [micro-CT] excluded.
hyperintense signal (bright) in MRI images, imaging) and other images such as histology in
enamel, dentin, and cementum reveal a terms of applicability and diagnostic accuracy Data Extraction
hypointense signal (black) because of their low in endodontics? The PICO (Patient, The data extraction considered general
water concentration and very short relaxation Intervention, Comparison, and Outcome) information that identified the study (author
constants of the hydrogen atoms. Thus, these strategy was used to define the key elements and year), study topic, MRI system, MRI unit,
structures can be indirectly depicted by of the guiding question. The study population MRI coil, MRI pulse sequence, MRI acquisition
contrasting them with adjacent signal-emitting (P) consisted of ex vivo and in vivo teeth parameters, presence of contrast, objective,
structures9. samples, the intervention (I) involved studies design, sample, and presence of control. The
Recently, the development of dedicated using MRI in endodontics, and the comparison summary of this extraction is presented in
coils and new MRI sequence techniques has (C) comprised studies that used 2-dimensional Tables 1 and 2. The weighted kappa test was
led to an increase in image resolution and a and 3-dimensional image (CBCT and micro- used to measure intrareviewer agreements,
decrease in artifacts originating from CT imaging) evaluation and histology. The and the results were interpreted according to
metals10,11. Thus, the scientific literature has outcome (O) evaluated the applicability and Landis and Koch13 as follows: 0.00–0.20,
been showing different studies evaluating the diagnostic accuracy in endodontics. poor; 0.21–0.40, reasonable; 0.41–0.60,
use of MRI in endodontics as a nonionizing moderate; 0.61–0.80, good; and 0.81–1.00,
alternative to CBCT imaging or other imaging Literature Search Strategy excellent.
modalities on diagnosis tasks such as tooth The search strategy, as described in Table 1,
and root canal anatomy, root canal working included only studies indexed in the following Risk of Bias Assessment
length, pulp vitality and regeneration, dental electronic databases: PubMed, Scopus, Two reviewers (A.P.C. and G.N.A.S.)
pulp and caries, guided endodontics, Embase, and Web of Science. No search was independently evaluated the potential risk of
periapical lesion, and root cracks or fractures. conducted in the gray literature. All included bias using the revised QUADAS-230. In the
However, there is still a need for further studies were written in English. The search event of disagreements, a third reviewer was
elucidation regarding the applicability of MRI in was performed using terms from Medical consulted to reach a consensus. This
endodontics. Subject Headings, Emtree (Embase), or Text assessment tool comprises 4 key domains:
Considering the extensive use of X-ray Word combined with the Boolean connectors patient selection, index test, reference
imaging in endodontics and the potential “OR” and “AND” to encompass the key standard, and flow and timing. Each item was
biological harm by X-rays to patients, elements from the guiding question. appraised by addressing specific guiding
nonionizing imaging examinations should be questions, resulting in a classification of "low
preferred. However, with the increasing risk of bias" (for negative responses), "high risk
adoption of MRI for different diagnostic tasks in Study Selection and Screening
of bias" (for positive responses), or "unclear"
endodontics, it is essential for professionals to Process
(for missing information).
understand when and how to recommend it as Two researchers (A.P.C. and G.N.A.S.)
well as to recognize its limitations and independently screened titles and abstracts of
advantages. Therefore, the aim of this the identified studies and selected those that RESULTS
systematic review was to evaluate the met the eligibility criteria described later. Selected Studies
accuracy of MRI in clinical applications in Disagreements were resolved through A total of 5658 studies were preselected; 5611
endodontics in comparison with other imaging consensus. The full text of the selected studies were excluded because of duplication or a lack
methods and/or clinical examination. was then assessed to determine if they aligned of relevance to the study topic. The titles and
with the present guiding question. Both A.P.C. abstracts of the 47 remaining studies were
and G.N.A.S. independently reviewed these carefully reviewed, leading to the exclusion of
MATERIALS AND METHODS studies to decide which ones to include in this 25 studies that used in vitro MRI units.
Protocol and Study Design systematic review, with any disagreements Additionally, 2 studies were excluded because
This systematic review followed the Preferred resolved through consensus. Additionally, 1 of the use of animal samples, and another 2
Reporting Items for Systematic Reviews and examiner (A.C.S.S.) assessed the selected studies were excluded because of a low
Meta-Analysis12 checklist, and the protocol studies to identify additional ones that could sample number (,5 teeth). Therefore, a total of
was registered in the International Prospective address the guiding question. Language was 18 studies were included in this systematic
Register of Systematic Reviews (PROSPERO- not a restriction. Microsoft Excel (Microsoft review (Fig. 1). The level of agreement between
2022 CRD42022382668). Based on this Corporation, Redmond, WA) was used to the reviewers was excellent (0.98).
protocol, the review explored aspects related register the decisions. The extracted data from the studies are
to the application of clinical MRI in the field of summarized based on the MRI acquisition
endodontics. The sample for this review Inclusion and Exclusion Criteria parameters and the studies’ characteristics in
comprised all relevant in vivo and ex vivo The inclusion criteria encompassed all studies accordance with 8 diagnosis tasks: tooth and
studies concerning the applicability of MRI in evaluating MRI applications in endodontics. root canal anatomy (5%), root canal working
endodontics, covering the period from 2004– The selected studies examined ex vivo and length (10%), pulp vitality (16%), pulp

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 435
TABLE 1 - Search Strategy According to the Database

PubMed: 517 studies found

(Tooth[MeSH Terms]) OR (Teeth[Title/Abstract])
((((((((((((((((((((((((((((((((((((((((Magnetic Resonance Imaging[MeSH Terms]) OR (Imaging, Magnetic Resonance[Title/Abstract])) OR (NMR Imaging[Title/
Abstract])) OR (Imaging, NMR[Title/Abstract])) OR (Zeugmatography[Title/Abstract])) OR (MR Tomography[Title/Abstract])) OR (NMR
Tomography[Title/Abstract])) OR (Tomography, NMR[Title/Abstract])) OR (Steady-State Free Precession MRI[Title/Abstract])) OR (Steady State
Free Precession MRI[Title/Abstract])) OR (Tomography, MR[Title/Abstract])) OR (Imaging, Chemical Shift[Title/Abstract])) OR (Chemical Shift
Imagings[Title/Abstract])) OR (Imagings, Chemical Shift[Title/Abstract])) OR (Shift Imaging, Chemical[Title/Abstract])) OR (Shift Imagings,
Chemical[Title/Abstract])) OR (Chemical Shift Imaging[Title/Abstract])) OR (Magnetic Resonance Image[Title/Abstract])) OR (Image, Magnetic
Resonance[Title/Abstract])) OR (Magnetic Resonance Images[Title/Abstract])) OR (Resonance Image, Magnetic[Title/Abstract])) OR
(Magnetization Transfer Contrast Imaging[Title/Abstract])) OR (MRI Scans[Title/Abstract])) OR (MRI Scan[Title/Abstract])) OR (Scan, MRI[Title/
Abstract])) OR (Scans, MRI[Title/Abstract])) OR (Tomography, Proton Spin[Title/Abstract])) OR (Proton Spin Tomography[Title/Abstract])) OR
(fMRI[Title/Abstract])) OR (MRI, Functional[Title/Abstract])) OR (Functional MRI[Title/Abstract])) OR (Functional MRIs[Title/Abstract])) OR (MRIs,
Functional[Title/Abstract])) OR (Functional Magnetic Resonance Imaging[Title/Abstract])) OR (Magnetic Resonance Imaging, Functional[Title/
Abstract])) OR (Spin Echo Imaging[Title/Abstract])) OR (Echo Imaging, Spin[Title/Abstract])) OR (Echo Imagings, Spin[Title/Abstract])) OR
(Imaging, Spin Echo[Title/Abstract])) OR (Imagings, Spin Echo[Title/Abstract])) OR (Spin Echo Imagings[Title/Abstract])
Scopus: 3.836 studies found
(TITLE-ABS-KEY (magnetic AND resonance AND imaging)) AND (TITLE-ABS-KEY (tooth))
Embase: 512 studies found
tooth:ti,ab,kw AND ’magnetic resonance imaging’:ti,ab,kw
Web of Science: 793 studies found
tooth (All Fields) AND magnetic resonance imaging (All Fields)

regeneration (16%), dental pulp and caries [Siemens Healthcare GmbH], 32-channel Germany) enterally, 0.1 mmoL/kg gadoterate
(10%), guided endodontics (10%), periapical head, and 64-channel head) (Table 2). meglumine, 469 mg/mL Magnevist, and
lesion (22%), and root cracks or fractures Eighteen pulse sequences were used in gadolinium. One study25 did not specify which
(11%) in accordance with the study aims the studies, including turbo spin echo, multiple contrast agent was used (Table 2).
(Tables 2 and 3). slab acquisition with view angle tilting gradient
with sampling perfection with application-
Study Characteristics
optimized contrasts using different flip angle
The publication years of the selected studies
MRI Acquisition Parameters evolution, fast field echo, volumetric
ranged from 2004–2023, with study designs
The selected studies used 12 different MRI interpolated breath-hold examination, fast low
distributed as 80% in vivo and 20% ex vivo. In
units (Magnetom Avanto [Siemens angle, constructive interference in steady state,
the in vivo studies, sample sizes ranged from
Healthineers, Erlangen, Germany], fat saturated, sampling perfection with
5–92 patients, evaluating 5–584 teeth. In the
Magnetom Trio [Siemens Healthineers], application-optimized contrasts using different
ex vivo studies, sample sizes ranged from 58–
Siemens Skyra [Siemens Healthineers], flip angle evolution, fat suppression with turbo
100 teeth. Further examining sample sizes in
Siemens Sinfonie [Siemens Healthineers], rapid acquisition with refocusing echoes,
relation to diagnostic tasks, studies on pulp
Magnetom Prisma [Siemens Healthineers], cartesian turbo spin echo, fat suppression,
regeneration used the lowest (5–18 teeth),
Sonata Vision [Siemens Healthineers], balanced steady, fast recovery fast spin echo,
whereas those on pulp vitality used the highest
Achieva [Philips Healthcare, Best, fast spoiled gradient echo, spin echo,
(162–1685 teeth), both having a 100% in vivo
Netherlands], Achieva Gyroscan [Philips multiband sweep imaging with Fourier
design. Among all of the studies, 61% included
Healthcare], TX Achieva [Philips Healthcare], transformation, sweep imaging with Fourier
a control examination, which could be
Signa HDxt [GE Medical System, Milwaukee, transformation, and gradient echo (Table 2).
panoramic radiography (9%), CBCT imaging
WI], OxfordMagnet [Oxford Instruments, The acquisition parameters used varied
(46%), digital periapical radiography (9%),
Oxford, UK], and Magnex Scientific across a wide range: field of view from 8–
histology examination (18%), micro-CT
[Magnex, Oxford, UK]) with 1.5, 3.0, and 220 mm2, matrix size from 64 ! 64–
imaging (9%), or micro-CT and CBCT imaging
4.0 Tesla systems and 15 coils (custom 328 ! 512 mm, voxel size from 0.02–
(9%) (Table 3).
intraoral, 50-mm diameter intraoral, 125 mm2, acquisition time from 2:19–
radiofrequency receiver, intraoral 10:25 minutes, echo time from 1.5–
radiofrequency transceiver, 8-inch diameter 400 milliseconds, repetition time from 4– Risk of Bias
temporomandibular joint surface, head and 4580 milliseconds, number of slices from 10– Figure 2A and B displays the risk of bias of the
neck [Achieva Gyroscan], 4-channel special 80, averages between 1 and 2, echo train included studies. Overall, the methods used
purpose, 4-channel multifunctional length from 1–73, flip angle from 10 –90 , were homogeneous and showed a low risk of
radiofrequency [Noras MRI Products, factor from 2–40, and bandwidth from 90– bias, except for 2 studies11,29 that exhibited a
Ho€chberg, Germany], 12-channel head and 501 Hz/pixel. In 9 studies in which contrast high risk. The highest bias risk was associated
matrix neck [Siemens Healthcare GmbH, was considered, 7 contrast agents were used. with domain 1, “patient selection.” In in vitro
Erlangen, Germany], 15-channel dental These included 0.1 mmol/kg gadolinium- studies, the issue of patient selection was a
surface [Mandibula, Noras MRI Products], diethylenetriamine pentaacetic acid, 0.1 mmol/ concern in the bias analysis because random
channel multipurpose [Variety, Noras MRI kg gadolinium, ultrasmall superparamagnetic patient selection is not advocated. However,
Products], 16-element neurovascular iron oxide, 1.5% solution of agar with 0.2% most studies answered yes to the second
receive, 20-channel head and neck Magnevist (Bayer Schering Pharma, Berlin, question in domain 1, which is related to the

436 Candemil et al. JOE  Volume 50, Number 4, April 2024

TABLE 2 - Magnetic Resonance Imaging Acquisition Parameters

System
Study Study topic (Tesla) Unit Coil Pulse sequence Acquisition parameters Contrast
Tymofiyeva Tooth and root canal 1.5 Magnetom 4-channel multifunctional 3D turbo spin echo FOV of 100 ! 60 ! 50 mm, matrix *
et al9 anatomy Avanto radiofrequency (Noras of 128 ! 76 ! 44 mm, voxel
MRI Products) size of 0.8 ! 0.8 ! 1 mm, echo
time of 100 ms and repetition
time of 1000 ms, turbo factor of
17
Zidan et al14 Root canal working 3.0 Magnetom 15-channel dental surface 3D MSVAT-SPACE FOV of 168 ! 131, matrix of *
length Trio coil (Mandibula) 384 ! 300 mm, voxel size of
0.44 mm3, acquisition time of
7:45 min, echo time of 6.4 ms,
repetition time of 1170 ms, and
80 slices
Zidan et al15 3.0 Magnetom 15-channel dental surface 3D MSVAT-SPACE FOV of 168 ! 131 mm, matrix of Noncontrast
Trio (Mandibula) 384 ! 300 mm, voxel size of
0.44 mm3, acquisition time of
7:45 min, echo time of 6.4 ms,
repetition time of 1170 ms, and
80 slices
Kress et al16 Pulp vitality 1.5 Siemens 8-inch diameter 1. T2 weighted 1. T2w: FOV of 230 mm3, 3. Gadolinium-
Sinfonie temporomandibular turbo spin matrix of 328 ! 512 mm, DTPA
joint surface 0.1 mmol/kg
echo (axial) acquisition time of
2. T1 weighted 3:46 min, echo time of
fast field echo 100 ms, repetition time of
(sagittal) 4523 ms, and slice
3. T1 weighted thickness of 3 mm
contrast 2. T1w: FOV of 270 mm3,
enhanced with matrix of 328 ! 512 mm,
fast field echo acquisition time 3.50 min,
(sagittal) echo time of 6.1 ms,
repetition time of 15 ms,
slice thickness of 1.5 mm,
and fat suppression with
proset water select
3. T1wgd: FOV of 270 mm3,
matrix of 328 ! 512 mm,
acquisition time 3:50 min,
echo time of 6.1,
repetition time of 15 ms,
slice thickness of 1.5 mm,
and fat suppression with
proset water select
Kress et al17 * * * 1. T2 weighted 1. T2w: FOV of 230, matrix 3. Gadolinium-
with turbo spin of 328 ! 512 mm, DTPA
0.1 mmol/kg
echo (axial) acquisition time of
2. T1 weighted 3:46 min, echo time of
with fast field 100 ms, repetition time of
echo (sagittal) 4523 ms, and slice
3. T1 weighted thickness of 3 mm
with contrast 2. T1w: FOV of 270, matrix
enhanced with of 328 ! 512 mm,
fast field echo acquisition time 3.50 min,
(sagittal) echo time of 6.1 ms,
repetition time of 15 ms,
slice thickness of 1.5 mm,
and fat suppression with
proset water select
3. T1wgd: FOV of 270,
matrix of 328 ! 512 mm,
acquisition time 3:50 min,
echo time of 6.1,
repetition time of 15 ms,
slice thickness of 1.5 mm,
and fat suppression with
proset water select
3.0

(continued on next page )

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 437
TABLE 2 - Continued

System
Study Study topic (Tesla) Unit Coil Pulse sequence Acquisition parameters Contrast
Juerchott Magnetom 16-channel multipurpose 2 Gadolinium
1. T1 weighted 1. T1w: FOV of 140 mm ,
et al18 Trio coil (Variety) turbo spin matrix of 256 ! 256 mm, 0.1 mmol/
kg
echo voxel size of
2. T1 weighted 0.55 ! 0.55 ! 1.5 mm,
VIBE acquisition time of 6 min,
echo time of 15 ms,
repetition time of 779 ms,
2 averages, 3 echo train
length, 40 slices, and
bandwidth of 501 Hz/
pixel
2. T1w: FOV of 180 mm2,
matrix of 192 mm2, voxel
size of 0.9 mm3,
acquisition time of
5:46 min, echo time of
2.46 ms, repetition time of
4.47 ms, 1 average, 1
echo train length, flip
angle of 15 , and
bandwidth of 501 Hz/
pixel
Assaf et al19 Pulp regeneration 3.0 Siemens 20-channel head and 1. T1 weighted FOV of 180 ! 180 ! 96 mm, Noncontrast
Skyra neck (Siemens Medical (FLASH) matrix of 256 ! 256 ! 120 mm
Solutions, Erlangen, and voxel size of
Germany)
2. Fat saturated 0.7 ! 0.7 ! 0.8 mm
T1 weighted 1. T1w: acquisition time of 4:14 min,
(CISS); echo time of 3.26 ms, repetition
3. Fat saturated time of 21 ms, flip angle of 35 ,
T2 weighted bandwidth of 500 Hz/pixel
2. fsT1w: acquisition time of
4. CISS (turbo
5:18 min and repetition time of
spin echo with 34 ms
SPACE) 3. fsT2w: acquisition time of
5:14 min, echo time of 113 ms,
repetition time of 2000 ms, echo
train length of 72, echo distance
of 4.35 ms, and bandwidth of
575 Hz/pixel
4. CISS: acquisition time of
3:43 min, echo time of 3.43 ms,
repetition time of 6.85 ms, flip
angle of 49 , and bandwidth of
300 Hz/pixel
El-Kateb 3.0 Achieva Head and neck (Achieva 1. T2 weighted 1. T2 axial: voxel size of *
et al20 Gyroscan Gyroscan) turbo spin 1.3 ! 31.3 mm, echo
echo(axial) time of 100 ms, repetition
2. T2 weighted time of 54.5 ms, flip angle
3D of 90 , and 3-mm slice
thickness
2. T2 3D: voxel size of
1.3 ! 31.3 mm,
acquisition time of
3.50 min, echo time of
160 ms, repetition time of
1500 ms, 2.5-mm slice
thickness, and turbo
factor of 40
Nakashima 1.5 (Philips * Fat suppression T2- FOV of 220 ! 316 mm, matrix of *
et al21 Electronics weighted with 128 mm2, voxel size of
Japan, Turbo RARE T2 0.02 mm2, echo time 70–80 ms,
Tokyo, (axial) repetition time of 2500 ms, 4
Japan) average, slice thickness of 3 mm,
10–20 slices FA90, number of
excitations 3
Dental pulp and caries 1.5 Radiofrequency receiver

(continued on next page )

438 Candemil et al. JOE  Volume 50, Number 4, April 2024

TABLE 2 - Continued

System
Study Study topic (Tesla) Unit Coil Pulse sequence Acquisition parameters Contrast
Tymofiyeva Magnetom 3D Cartesian Turbo FOV of 60 ! 30 ! 17 mm, voxel 1.5% solution
et al22 Avanto spin echo size of 0.3 ! 0.3 ! 0.4 mm, of agar with
time of acquisition of 8 min, echo 0.2%
time of 400 ms, repetition time of Magnevist
12 ms, and turbo factor of 5 enterally
Cankar 3.0 TX Achieva 32-channel head 1. T2-weighted 1. T1-weighted: FOV of *
et al23 turbo spin 140 mm2, matrix of
echo (axial) 156 ! 124 mm,
2. T1-weighted acquisition time of
fat 3.25 min, voxel size of
suppressed 0.9 ! 1.12 ! 2 mm,
proton density echo time of 400 ms,
(coronal and repetition time of 100 ms,
sagittal) single echo, thickness of
2 mm, 5 slices, and
bandwidth of 253 Hz/
pixel
2. T2 weighted: FOV of
160 mm2, matrix
380 ! 311 mm,
acquisition time of
10:24 min, voxel size
0.42 ! 0.51 ! 2.5 mm,
echo time of 15, 30, 45,
60, 75, 90 ms, repetition
time of 2000 ms, single
slice, slice thickness of
2.5 mm, and bandwidth
of 290 Hz/pixel
Leontiev Guided endodontics 3.0 Magnetom 64-channel head Balanced steady- Matrix of 384 ! 384 ! 120 mm, *
et al24 Prisma state free voxel size of
precession 0.4 ! 0.4 ! 0.8 mm, echo time
of 4.45 ms, repetition time
8.89 ms, acquisition time of
2:19 min, factor of 2, slice
oversampling of 20%, flip angle
of 40 , and bandwidth of
213 Hz/pixel
Geibel et al25 Periapical lesion 3.0 Achieva 16-channel neurovascular T1- (T1w) and T2- Acquisition time of 9:06 min (T1w) Used, however,
receive weighted (T2w) and 5:43 min (T2w) not specified
contrast
Lizio et al26 1.5 Signa HDxt Head-neck coil 1. T1-weighted 1. T1-weighted: matrix, Gadolinium-
fast spin echo 320 ! 224 mm, echo DTPA

(axial) time of 9–12 ms,

2. T2-weighted repetition time of 400–
fast recovery 580 ms, echo train length
fast spin echo of 3, slice thickness of
(axial or 3 mm, interslice gap of
sagittal) 0.3 mm, and number of
3. T1-weighted excitations of 3
fast spoiled 2. T2-weighted: matrix of
gradient echo 320 ! 224 mm, echo
(axial or time of 120 ms, repetition
sagittal) time of 3440–3680 ms,
echo train length of 17,
slice thickness of 3 mm,
interslice gap of 0.3 mm,
and number of excitations
of 6
3. T1 weighted: matrix of
256 ! 192 mm, echo
time of 1.5–5 ms,
repetition time of 6–
10 ms, flip angle of 10–
(continued on next page )

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 439
TABLE 2 - Continued

System
Study Study topic (Tesla) Unit Coil Pulse sequence Acquisition parameters Contrast

15 , echo train length of
1, slice thickness of
2 mm, interslice gap of
0 mm, and number of
excitations of 2
Juerchott 3 Magnetom 12-channel head coil and 1. T2 weighted 1. T2wfs: FOV of 150 mm2, 0.1 mmoL/kg
et al27 Trio matrix neck coil with fat matrix of 240 ! 320 mm, gadoterate
(Siemens Healthcare) meglumine
suppression acquisition time of
(T2wfs turbo 4:26 min, echo time of
spin echo) 87 ms, repetition time of
2. T1 weighted 4580 ms, 2 averages
(T1w spin number, echo train length
echo) of 9, flip angle of 150 ,
3. T1 weighted slice thickness of 2 mm,
with fat 40 slices, and bandwidth
suppression of 90 Hz/pixel
(T1wfs spin 2. T1w: FOV of
echo) 168 ! 210 mm, matrix of
192 ! 320 mm,
acquisition time of
4:20 min, echo time of
6.4 ms, repetition time of
680 ms, 1 average
number, 1 echo train
length, flip angle: 90 ,
slice thickness of 2 mm,
40 slices, and bandwidth
of 248 Hz/pixel
3. T1wfd: FOV of
168 ! 210 mm, matrix of
218 ! 320 mm,
acquisition time of
3:18 min, echo time of
8.1 ms, repetition time of
700 ms, 2 average
number, 1 echo train
length, excitation flip
angle of 58 , slice
thickness of 1.8 mm, 19
slices, and bandwidth of
159 Hz/pixel
Pigg et al28 1.5 Sonata Vision S/I 1. Axial T1  T1, T2 SIR, T1 gd and T1 Magnevist
weighted fs gd (1, 2, 4, 5) with voxel (469 mg/mL,
Schering
2. Axial T2 SIR size of 0.9 ! Nordiska)
3. Axial 3D CISS 0.8 ! 3.0 mm and
4. Axial T1  3D CISS (3) with voxel size gadolinium
weighted with of 0.6 mm3 contrast
gadolinium
contrast (T1
gd)
5. Axial T1
weighted with
fat
suppression
and
gadolinium
contrast
administration
(T1 fs gd)
Groenke Root cracks or 4.0 OxfordMagnet Custom intraoral (Agilent MB-SWIFT FOV of 120 mm3, acquisition time of *
et al11 fractures DirectDrive console, 5:39 min, voxel size of
Palo Alto, CA) 0.25 mm3, repetition time of

(continued on next page )

440 Candemil et al. JOE  Volume 50, Number 4, April 2024

TABLE 2 - Continued

System
Study Study topic (Tesla) Unit Coil Pulse sequence Acquisition parameters Contrast

4 ms, flip angle of 12.0 , pulse
length of 4 ms, delay after pulse
of 3.7 ms, and bandwidth of
128 kHz
Schuurmans 4.0 Magnex Intraoral radiofrequency SWIFT FOV of 120 mm3, acquisition time of *
et al29 transceiver 3.5 min, voxel size of 0.27mm3,
and excitation bandwidth of
100 kHz

*Not informed.
3D, 3-dimensional; CISS, constructive interference in steady state; FLASH, fast low angle; FOV, field of view; gd, gadolinium; GRE, gradient echo; MB, multiband; MSVAT, multiple slab
acquisition with view angle tilting gradient; RARE, rapid acquisition with refocusing echoes; SIR, short tau inversion recovery; SPACE, sampling perfection with application optimized
contrasts using different flip angle evolution; SWIFT, sweep imaging with Fourier transformation; VIBE, volumetric interpolated breath-hold examination.

control group (Was a case-control design observed between vital and nonvital teeth. Dental Pulp and Caries
avoided?). Regarding domain 2 (index test), Conversely, Kress et al17 measured age- Tymofiyeya et al22 assessed carious lesions
which is related to the blinding of the examiner related changes in pulp cavity signal intensity and their proximity to the dental pulp using
during sample analysis, 7 with MRI and revealed that older teeth 1.5 Tesla MRI. Although the evaluations were
studies11,16,17,19,25,29,31 did not comply with or exhibited a lower signal intensity difference feasible, the images showed an overestimation
specify the criteria. In domain 3, which pertains between the non–contrast- and contrast- of the cavity size compared with digital
to the reference standard, conduct, and enhanced images due to dental pulp volume periapical radiography. The authors of the
interpretation of results to reduce bias risk, 4 decreasing with age as a result of additional aforementioned study hypothesized that the
studies11,19,25,29 did not report or specify this dentin deposition. More recently, Juerchott MRI signal was obtained from the porous,
domain. In domain 4, which is related to et al18 assessed pulpal contrast enhancement demineralized tooth substance or from an
sample standardization, only 1 study29 did not using 3 Tesla MRI by measuring post- and occult carious lesion. The results obtained
report or specify this domain. precontrast pulp signal intensity. They found demonstrated that the potential of MRI for
no effects of age, sex, or jaw type. Although caries diagnosis extends beyond open cavities
there were minor differences observed among and includes measuring the minimum distance
Application in Endodontics tooth types, statistical significance was to the dental pulp. Conversely, Cankar et al23
Tooth and Root Canal Anatomy reached. According to the authors, blood examined the dental pulp’s response to caries
Tymofiyeya et al9 compared the diagnostic intensity can influence the increase in T1 values progression using 3 Tesla MRI. The results
accuracy of 1.5 Tesla MRI images and on postcontrast images and/or the circulation showed differences in signal intensity T2
panoramic images in pediatric patients. The of the contrast agent in the extravascular area. between intact and affected dental pulp
results showed that MRI is well suited for Because dental pulp does not have a tight regions, with these values showing a strong
evaluating the anatomy of teeth, dental pulp, blood-tissue barrier, the lack of contrast correlation with the severity of caries according
mandibular canal, and cortical bone. enhancement may be considered as an to the International Caries Detection and
absence of pulp vascular supply. Assessment System (ICDAS) score.
Root Canal Working Length Regardless of dental pulp depth or tooth type,
Recent studies evaluated the applicability of Pulp Regeneration intact teeth (ICDAS 0) exhibited approximately
3 Tesla MRI in determining the working length Assaf et al19 evaluated revascularization of 40-millisecond higher signal intensity T2 values
of premolars and molars14 as well as incisors teeth after traumatic dental injuries in children than severely caries-affected teeth (ICDAS
and canines15. Although the measurement and adolescents. Using 3 Tesla MRI, they 4–6).
reliability was excellent, MRI had a tendency to demonstrated a feasible means of visualizing Although the influence of carious lesions
underestimate the working length in 67.6%– changes in in vivo dental pulp to assess tooth on dental pulp is well-documented in the
95.5% of the measurements compared with vitality without the need for a contrast agent. scientific literature, the present systematic
CBCT imaging, restricting its clinical The authors believe that 3 Tesla MRI can be review focused solely on studies using clinical
application. The authors believe that the lower indicated for children to prevent unnecessary MRI units. Excluded studies used a
accuracy of MRI may be attributed to its lower root canal fillings in affected teeth because the microscopic MRI unit with Tesla values
spatial resolution compared with CBCT pulp signal alterations of the affected tooth exceeding 4 and longer exposure times, which
imaging, primarily associated with voxel size compared with a vital one can be explained by are not used in routine clinical patient care.
and acquisition time. early degeneration of the tissue. Similarly, but
without the use of contrast, El-Kateb et al20 Guided Endodontics
Pulp Vitality evaluated tissue regeneration after a Recently, Leontiev et al24 evaluated the
According to Kress et al16, it is possible to regenerative endodontic procedure using accuracy of 3 Tesla MRI for detecting root
produce 1.5 Tesla MRI images that show the 3 Tesla MRI. The signal intensity in treated canals during guided endodontics. The mean
perfusion of dental pulp when a contrast agent teeth was higher after 3 months of accuracy in the buccolingual and mesiodistal
is present. However, when comparing images regeneration, decreased significantly during dimensions, as well as angle deviation,
before and after contrast agent administration, follow-up, and became equivalent to that of generally exhibited similarities to the results of
significant differences in signal intensities were normal contralateral teeth after 12 months. Zehnder et al32, who considered only CBCT

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 441
FIGURE 1 – A Flow diagram of the study selection following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement.

images. However, there was a notable because the resolution of measurements was origin and compared the results with
difference in the overall success rate, with MRI lower in MRI compared with CBCT imaging, histopathological information. The observers
achieving 91% accuracy compared with a the MRI-based method exhibited lower overall exhibited the highest consistency when
100% rate in the CBCT study. The authors of accuracy. identifying characteristics such as pattern
this study believe that this discrepancy may be (homogeneous in granulomas and
attributed to lower precise planning or Periapical Lesion nonhomogenous in cysts), the presence of a
accurate access cavity preparation because Geibel et al25 evaluated the applicability of low-intensity outline (fibrous capsule), and the
teeth are solid structures, and direct imaging is 3 Tesla MRI for evaluating apical periodontitis contrast-enhanced distribution pattern on T1-
challenging with MRI33. To ensure adequate in direct comparison with CBCT imaging. In all weighted images (epithelial wall). Interrater
alignment between MRI and stereolithography of the patients analyzed, MRI showed similar agreement between radiologists and
data, an indirect visualization of the tooth sensitivity in diagnosing periapical lesions pathologists was excellent. However, the
crown was performed. Unlike CBCT planning compared with CBCT imaging. However, authors of this study believe that making a
software, which can overlay Digital Imaging significant differences were observed in lesion specific diagnosis based solely on MRI can be
and Communications in Medicine and dimensions between the 2 imaging modalities. challenging, especially when distinguishing
stereolithography data automatically, manual This difference may be attributed to the lower between cysts and granulomas, which may
positioning was necessary for MRI scans. spatial resolution of MRI compared with CBCT pose difficulties for less experienced
According to the authors of the mentioned imaging or the distinct physical principles radiologists. Similarly, Juerchott et al27
study, manual positioning may have underlying these 2 imaging modalities. evaluated the differentiation of periapical
introduced errors that can account for the Lizio et al26 investigated the diagnostic granulomas from periapical cysts using 3 Tesla
observed variation, outliers, and angle reliability and accuracy of 1.5 Tesla MRI in MRI. The results showed 6 lesion
deviation between the models. Furthermore, distinguishing periapical lesions of endodontic characteristics that could clearly distinguish

442 Candemil et al. JOE  Volume 50, Number 4, April 2024

 TABLE 3 - Data Extraction from the Included Studies
JOE  Volume 50, Number 4, April 2024

Study Study topic Objective Design Sample Control Conclusion

9
Tymofiyeva et al Tooth and root canal Assess the feasibility of MRI of In vivo 16 patients (3 with Panoramic MRI is a safe, well-tolerated
anatomy dental abnormalities in mesiodens, 9 with radiography imaging method that can be
children supernumerary teeth used for the diagnosis of dental
other than a mesiodens, 1 abnormalities in children.
with gemination, 1 with
dilaceration, 1 with
transmigration, and 1 with
transposition)
Zidan et al14 Root canal working Assess the reliability and In vivo 9 patients (34 root canals CBCT imaging MRI root canal lengths were
length accuracy of high-resolution from 12 molars [4/8, systematically underestimated
MRI for endodontic working upper/lower jaw], 22 root compared to CBCT.
length measurements canals and 11 premolars
[2/9 upper/lower jaw], 12
root canals)
Zidan et al15 Evaluate whether high- In vivo 31 patients (67 teeth [25 CBCT imaging MRI measurements were
resolution, non–contrast- central incisors, 22 lateral underestimated compared with
enhanced MRI can reliably incisors, and 20 canines], CBCT imaging.
and accurately measure the 28 from the upper jaw and
canal length 39 from the lower jaw)
Kress et al16 Pulp vitality Assess whether it is possible to In vivo 35 patients (clinical vitality None Non-contrast sequences cannot
measure tooth vitality using evaluated on 162 teeth) provide sufficient information
MRI about pulpal perfusion.
However, a comparison of
sequences before and after
contrast injection shows a
significant difference between
vital and avital teeth.
Endodontic Applications of Magnetic Resonance Imaging

Kress et al17 Determine if it is possible to In vivo 92 patients born between None MRI can be used to determine pulp
measure age-related pulp 1930 and after 1980 (302 cavity signal intensity changes
cavity signal intensity changes inferior premolars and 282 with age.
by using MRI inferior molars)
Juerchott et al18 Optimize the assessment of In vivo 70 patients (1585 teeth–33 None Pulpal contrast enhancement in
pulpal contrast enhancement females, 37 males; 50 dental MRI was a stable
on MRI and investigate patients, 348 incisors, 189 intraindividual marker with only
physiological pulpal contrast canines, 351 premolars minor differences between
enhancement patterns and 396 molars; 638 different tooth types.
superior 646 inferior)
Assaf et al19 Pulp regeneration Evaluate whether root canal In vivo 7 patients (13 upper anterior None MRI could demonstrate that root
treatment is necessary and teeth) canal treatment is not necessary
whether revitalization of in every case.
affected teeth is possible and
measurable
El-Kateb et al20 Quantify assessed tissue In vivo 18 necrotic mature maxillary None Vital pulplike tissue could be
regeneration after anterior teeth (17 maxillary successfully regenerated in
regenerative endodontic mature teeth using regenerative
443

(continued on next page )

 TABLE 3 - Continued
444
Candemil et al.

Study Study topic Objective Design Sample Control Conclusion

procedures with 2 different central incisors and 1 endodontic procedures, which
apical preparation sizes in lateral incisor) was not significantly affected by
mature teeth the size of the apical diameter.
Nakashima et al21 Assess the safety, potential In vivo 5 patients (5 teeth [1 upper CBCT imaging The signal intensity of MRI of the
efficacy, and feasibility of premolar, 2 upper regenerated tissue in the root
autologous transplantation of incisors, and 2 lower canal after 24 weeks was similar
mobilized dental pulp stem premolars]) to the normal dental pulp in the
cells in pulpectomized teeth untreated teeth.
Tymofiyeva et al22 Dental pulp and Assess the feasibility of MRI of 3- In vivo 5 patients (7 teeth with Digital periapical MRI demonstrated high feasibility
caries dimensional visualization and carious lesions – 1 upper radiography to 3-dimensionally visualize and
quantification of carious incisor, 1 upper premolar, quantify carious lesions,
lesions, as well as 4 upper molars and 1 including approximal lesions and
measurement of the distance lower premolar) occult dentin lesions, and
between the lesion and dental measure the minimum distance
pulp in vivo to the dental pulp in vivo.
Cankar et al23 Quantify dental pulp response to In vivo 12 patients (74 teeth, 34 None T2 mapping of dental pulp can be
caries progression using the single rooted and 40 used to reliably quantify its
T2 mapping method multirooted) response to caries progression.
Leontiev et al24 Guided endodontics Demonstrate that MRI is Ex vivo One hundred extracted teeth None MRI is suitable for guided
sufficiently accurate for the (anterior and premolar) endodontic access cavity
detection of root canals using preparation.
guided endodontics
Geibel et al25 Periapical lesion Evaluate the applicability of MRI In vivo 19 patients (34 teeth with CBCT imaging No difference on lesion visibility
for the assessment of apical periapical lesions was observed.
periodontitis in direct [3 molars, 14 premolars,
comparison with CBCT and 7 front teeth])
imaging
Lizio et al26 Investigate the diagnostic In vivo 34 patients (34 teeth with Histopathological The reliability and accuracy of MRI
reliability and accuracy of MRI periapical lesion) analysis were high and comparable to
to differentiate periapical histopathological reliability.
lesions of endodontic origin
Juerchott et al27 Evaluate whether periapical In vivo 11 patients (11 teeth with Histopathological MRI enables a reliable
granulomas can be periapical lesion) analysis differentiation between
differentiated from periapical periapical cysts and
JOE  Volume 50, Number 4, April 2024

cysts granulomas.
Pigg et al28 Examine the occurrence of MRI In vivo 20 patients (30 teeth, 1-3 per CBCT imaging MRI examination revealed no
signal changes in the painful patient) changes in the painful region in a
regions of patients with majority of patients with atypical
atypical odontalgia odontalgia, suggesting that
inflammation was not present.
Groenke et al11 Root cracks or Describe the narrowest VRF Ex vivo 62 teeth Micro-CT MRI showed the ability to detect
fractures detectable using micro-CT VRFs with median widths of
imaging as reference 45 mm with an interquartile range
of 26–64 mm.
(continued on next page )
 between the 2 lesion types in line with Although MRI shows high accuracy in
histopathological findings. The study

cracks/fractures., in relation to
different diagnostic tasks, this imaging

specificity for the detection of

MRI demonstrated comparable
measures of sensitivity and
underscores the limited diagnostic value of modality is not reliable in measuring root canal
Conclusion non–contrast-enhanced T1-weighted images length14,15, cavity sizes of caries22, and
because lesion margins may not be distinctly periapical lesion dimensions25. It also
delineated in many cases. demonstrated less accurate planning and

CBCT imaging.
Pigg et al28 evaluated the occurrence of cavity preparation for guided endodontics with
1.5 Tesla MRI signal changes in painful regions a 91% success rate24. The studies indicate
of patients with atypical odontalgia, correlating that the results of these objective evaluations
them with periapical bone defects detected on have a consensus of overestimation compared
radiographic examinations. The main finding with measurements performed using ionizing
indicated that radiographic and MRI evaluation imaging modalities14,15,22,25. Conversely,
had a similarity of 75% of patients with atypical regarding the evaluation of root cracks or
Micro-CT and CBCT

odontalgia. The authors of the aforementioned fractures, although the results show similar
study hypothesized that the results may be diagnostic accuracy between MRI and CBCT
Control

attributed to the MRI signal alterations when a images, the sensitivity of both images is
imaging

defect in periapical bone was detected, and poor11,29. Overall, the reliability of CBCT
probably inflammatory changes were present. diagnosis of root cracks and fractures is
On the contrary, in case of the absence of questionable, mainly because of artifacts
signal alteration, despite the presence of a arising from high-density materials in the tooth
periapical bone defect, bone healing may be in of interest, such as filling material34,35. MRI
progress. However, because this study did not may be an alternative diagnostic imaging
consider a reference standard image, the true method because of the absence of ionizing
CBCT, cone-beam computed tomographic; MRI, magnetic resonance imaging; SWIFT, sweep imaging with Fourier transformation; VRF, vertical root fracture.

significance of these results remains uncertain. radiation, better contrast (depending on the
Sample

Furthermore, 70% of the painful area had a unit and acquisition parameters), and inclusion
history of endodontic treatment, rendering of useful information about soft tissues8.
tooth vitality evaluation impossible and Because MRI has a high ability to
58 teeth

compromising the identification of periapical characterize soft tissues, such as in the

bone defect. differentiation of chronic and acute lesions and
the detection of early inflammation, pulp
Root Cracks or Fractures vitality, pulp regeneration, and periapical
Schuurmans et al29 evaluated the diagnostic lesions have shown to be well evaluated19,36. It
Design

accuracy of 4 Tesla MRI for identifying root enables precise differentiation between blood-
Ex vivo

cracks and fractures. When comparing filled structures within dental pulp and
diagnoses between MRI and CBCT imaging, neighboring tooth components19,37, while also
both modalities exhibited high specificity (0.83 aiding in the detection of soft tissue anomalies
and to establish reliability and

and 0.90, respectively) and poor sensitivity resulting from inflammation because of water
crack/fracture identification

increase38,39. MRI can reveal tooth vitality by

accuracy in their detection

(0.59 and 0.59, respectively). Similarly,

Develop MRI criteria for root

Groenke et al11 detected 55% of the vertical the increase of signal intensity within the pulpal
root fracture in root canal–treated teeth, with cavity after the administration of contrast,
Objective

the smallest observed width falling within the facilitating the distinction between vital and
calculated theoretical limit of detection for MRI, nonvital teeth16-18,21.
ranging from 10–20 mm, demonstrating the The outcome of conservative pulp-
capability of 4 Tesla MRI for this diagnosis. saving treatment is significantly influenced by
the extent of dental pulp inflammation and the
remaining dentin thickness treatment40,41.
Consequently, an early and precise evaluation
DISCUSSION
of dental pulp tissue response to caries
The present systematic review showed that progression is of paramount importance
Study topic

the use of MRI has a high diagnostic value in before making treatment decisions42. The MRI
endodontics in most of the tasks discussed signal intensity in carious lesions increases due
earlier, including the evaluation of tooth and to an increase in relaxation time caused by the
root canal anatomy, pulp vitality and increase of liquid within the dentin tubules,
regeneration, the effect of caries on dental stemming from the porous structure of the
pulp, periapical lesions, and root cracks or affected dentin43. Conversely, Weglarz et al44
fractures. MRI’s accuracy is similar to or even explained this signal enhancement as a
29
TABLE 3 - Continued

Schuurmans et al

higher than imaging references such as CBCT consequence of contrast medium penetration
imaging, micro-CT imaging, and histology. into the porous demineralized tooth
Furthermore, MRI offers significant advantages substance, resulting in an elevated proton
because of its high soft tissue contrast and the density.
Study

ability to vary the contrast by altering the Compared with CBCT imaging, MRI has
sequence design27. the advantages of the absence of harmful

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 445
FIGURE 2 – Risk of bias charts results, encompassing the assessment of 4 key domains, as well as the overall classification of the sample review (n 5 20). (A ) The traffic plot
illustrates the classification of each domain, and (B ) the summary plot shows the distribution of classifications across these domains.

446 Candemil et al. JOE  Volume 50, Number 4, April 2024

radiation and artifact from filling materials resolution requires higher magnetic field and a higher sample size are highly
(endodontic sealer and gutta-percha) and strengths and the use of dedicated intraoral/ encouraged.
increased contrast25,29,45. However, it is dental coils22,23,46. For instance, using a Although MRI can be considered an
important to highlight that the presence of 3 Tesla magnet increases the signal-to-noise important alternative diagnostic tool in
metallic materials, patient movement during ratio by 70% compared with a 1.5 Tesla endodontics, further developments are
the examination or even irregular illumination, system47, and this increase may be even needed to enhance image acquisition speed
reading errors and cross talk with adjacent greater with higher magnetic field strengths and resolution, reduce cost, and improve
slices, eddy currents, and B0 inhomogeneities (greater than 4 Tesla). However, such high- accessibility14. In addition, there is a great
can reduce the image quality19,39. On the other field systems are not suitable for human in vivo need for further research on this topic to
hand, specific optimized sequences with imaging. Additionally, dedicated coils can provide professionals with the confidence to
reduced acquisition time have been shown to reduce the field of view, acquisition time, and recommend MRI in view of the great benefit for
minimize the risk of motion and metal generate thinner slices compared with patients.
artifacts10,27,29. standard head coils, achieving a resolution of
Despite MRI not being a new modality up to 300 mm23,25. These coils have been
on the market, its application in endodontics is successfully used in combination with different CONCLUSION
relatively recent, and limited information is acquisition sequences, including turbo spin
Based on the studies published in the literature
available in the literature. Thus, there is no echo22 and sweep imaging with Fourier
regarding the use of MRI in endodontics, the
consensus or guideline regarding its transformation11,29.
present systematic review can conclude that
indications or even the type of acquisition The limitations of the present
MRI holds high accuracy potential for
protocol. Many factors can influence MRI systematic review include the high
diagnostic application in evaluating various
image quality such as Tesla strength, methodological heterogeneity of the selected
clinical endodontic tasks, with the exception of
channels, coil design, sequence, field of view, studies with different study topics, objectives,
root canal length, cavity size of caries, and
imaging/reconstruction matrix, and voxel size. designs, sample types and sizes, control
periapical lesion dimension measurements,
Among the selected studies, there was images, MRI units, and acquisition
which tend to be overestimated in MRI. It is of
significant variability in terms of device parameters. It should be borne in mind that
high importance that further studies and MRI
characteristics, coil types, and acquisition the sample factor was responsible for the item
development be performed to provide more
parameters used. The measurement values with the highest risk of bias, which was
information to the scientific community.
may be affected by the specific MRI analyzed based on prior testing, presentation,
techniques, including hardware and software intended use of the index test, and setting of
features, as well as the type of contrast agent. patient selection. Furthermore, it is important
Therefore, different studies are often not to highlight the limitations of the studies, such
ACKNOWLEDGMENTS
directly comparable18. MRI resolution is as low sample size (,15 teeth)9,19,21,27 and Supported by Sa ~o Paulo Research Foundation
directly associated with sensitivity and the absence of a gold standard for the (grant nos. 2018/14450-1, 2021/01623-8,
acquisition time. Given the need to limit diagnosis16-20,23,24, which may indicate poor and 2022/07081-5).
acquisition time for clinical use (typically scientific evidence. Therefore, further studies The authors deny any conflicts of
10 minutes per acquisition), increasing with a lower risk of bias in the patient selection interest related to this study.

REFERENCES
1. Estrela C, Couto GS, Bueno MR, et al. Apical foramen position in relation to proximal root surfaces
of human permanent teeth determined by using a new cone-beam computed tomographic
software. J Endod 2018;44:1741–8.
2. Lo Giudice R, Nicita F, Puleio F, et al. Accuracy of periapical radiography and CBCT in endodontic
evaluation. Int J Dent 2018;16:2514243.
3. Special Committee to Revise the Joint AAE/AAOMR Position Statement on use of CBCT in
Endodontics. AAE and AAOMR Joint Position Statement: use of cone beam computed
tomography in endodontics 2015 update. Oral Surg Oral Med Oral Pathol Oral Radiol
2015;120:508–12.
4. Patel S, Brown J, Semper M, et al. European Society of Endodontology position statement: use of
cone beam computed tomography in Endodontics: European Society of Endodontology (ESE)
developed by Int Endod J 2019;52:1675–8.
5. Oenning AC, Jacobs R, Pauwels R, et al. Cone-beam CT in paediatric dentistry: DIMITRA project
position statement. Pediatr Radiol 2018;48:308–16.

6. Pauwels R, Silkosessak O, Jacobs R, et al. A pragmatic approach to determine the optimal kVp in
cone beam CT: balancing contrast-to-noise ratio and radiation dose. Dentomaxillofac Radiol
2014;43:20140059.

7. Yeh JK, Chen CH. Estimated radiation risk of cancer from dental cone-beam computed
tomography imaging in orthodontics patients. BMC Oral Health 2018;18:131.

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 447
 8. Di Nardo D, Gambarini G, Capuani S, Testarelli L. Nuclear magnetic resonance imaging in
endodontics: a review. J Endod 2018;44:536–42.

9. Tymofiyeva O, Proff PC, Rottner K, et al. Diagnosis of dental abnormalities in children using 3-
dimensional magnetic resonance imaging. J Oral Maxillofac Surg 2013;71:1159–69.
10. Hilgenfeld T, Prager M, Heil A, et al. PETRA, MSVAT-SPACE and SEMAC sequences for metal
artefact reduction in dental MR imaging. Eur Radiol 2017;27:5104–12.
11. Groenke BR, Idiyatullin D, Gaalaas L, et al. Minimal detectable width of tooth fractures using
magnetic resonance imaging and method to measure. J Endod 2022;48:1414–1420.e1.

12. 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:e1000097.
13. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of
majority agreement among multiple observers. Biometrics 1977;33:363–74.
14. Zidan M, Schwindling FS, Juerchott A, et al. Endodontic working length measurements of
premolars and molars in high-resolution dental MRI: a clinical pilot study for assessment of
reliability and accuracy. Clin Oral Investig 2022;26:6765–72.

15. Zidan M, Schwindling FS, Juerchott A, et al. Reliability and accuracy of dental MRI for measuring
root canal length of incisors and canines: a clinical pilot study. Sci Rep 2022;12:14068.

16. Kress B, Buhl Y, Anders L, et al. Quantitative analysis of MRI signal intensity as a tool for
evaluating tooth pulp vitality. Dentomaxillofac Radiol 2004;33:241–4.
17. Kress B, Buhl Y, H€ahnel S, et al. Age- and tooth-related pulp cavity signal intensity changes in
healthy teeth: a comparative magnetic resonance imaging analysis. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 2007;103:134–7.
18. Juerchott A, Jelinek C, Kronsteiner D, et al. Quantitative assessment of contrast-enhancement
patterns of the healthy dental pulp by magnetic resonance imaging: A prospective in vivo study.
Int Endod J 2022;55:252–62.
19. Assaf AT, Zrnc TA, Remus CC, et al. Early detection of pulp necrosis and dental vitality after
traumatic dental injuries in children and adolescents by 3-Tesla magnetic resonance imaging. J
Craniomaxillofac Surg 2015;43:1088–93.
20. El-Kateb NM, El-Backly RN, Amin WM, Abdalla AM. Quantitative assessment of intracanal
regenerated tissues after regenerative endodontic procedures in mature teeth using magnetic
resonance imaging: a randomized controlled clinical trial. J Endod 2020;46:563–74.
21. Nakashima M, Iohara K, Murakami M, et al. Pulp regeneration by transplantation of dental pulp
stem cells in pulpitis: a pilot clinical study. Stem Cell Res Ther 2017;8:61.

22. Tymofiyeva O, Rottner K, Jakob PM, et al. Three-dimensional localization of impacted teeth using
magnetic resonance imaging. Clin Oral Investig 2010;14:169–76.

23. Cankar K, Vidmar J, Nemeth L, Sersa I. T2 mapping as a tool for assessment of dental pulp
response to caries progression: an in vivo MRI study. Caries Res 2020;54:24–35.
24. Leontiev W, Bieri O, Mado€rin P, et al. Suitability of magnetic resonance imaging for guided
endodontics: proof of principle. J Endod 2021;47:954–60.

25. Geibel MA, Schreiber ES, Bracher AK, et al. Assessment of apical periodontitis by MRI: a
€fo 2015;187:269–75.
feasibility study. Ro

26. Lizio G, Salizzoni E, Coe M, et al. Differential diagnosis between a granuloma and radicular cyst:
effectiveness of magnetic resonance imaging. Int Endod J 2018;51:1077–87.
27. Juerchott A, Pfefferle T, Flechtenmacher C, et al. Differentiation of periapical granulomas and
cysts by using dental MRI: a pilot study. Int J Oral Sci 2017;10:17.

28. Pigg M, List T, Abul-Kasim K, et al. A comparative analysis of magnetic resonance imaging and
radiographic examinations of patients with atypical odontalgia. J Oral Facial Pain Headache
2014;28:233–42.

29. Schuurmans TJ, Nixdorf DR, Idiyatullin DS, et al. Accuracy and reliability of root crack and
fracture detection in teeth using magnetic resonance imaging. J Endod 2019;45:750–755.e2.

30. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2: a revised tool for the quality
assessment of diagnostic accuracy studies. Ann Intern Med 2011;155:529–36.
31. Zhang W, Zheng Y, Liu H, et al. A non-invasive monitoring of USPIO labeled silk fibroin/
hydroxyapatite scaffold loaded DPSCs for dental pulp regeneration. Mater Sci Eng C Mater Biol
Appl 2019;103:109736.

448 Candemil et al. JOE  Volume 50, Number 4, April 2024

 32. Zehnder MS, Connert T, Weiger R, et al. Guided endodontics: accuracy of a novel method for
guided access cavity preparation and root canal location. Int Endod J 2016;49:966–72.

33. Ho€vener JB, Zwick S, Leupold J, et al. Dental MRI: imaging of soft and solid components without
ionizing radiation. J Magn Reson Imaging 2012;36:841–6.
34. Brady E, Mannocci F, Brown J, et al. A comparison of cone beam computed tomography and
periapical radiography for the detection of vertical root fractures in nonendodontically treated
teeth. Int Endod J 2014;47:735–46.
35. Setzer FC, Hinckley N, Kohli MR, Karabucak B. A survey of cone-beam computed tomographic
use among endodontic practitioners in the United States. J Endod 2017;43:699–704.

36. Gaudino C, Cosgarea R, Heiland S, et al. MR-Imaging of teeth and periodontal apparatus: an
experimental study comparing high-resolution MRI with MDCT and CBCT. Eur Radiol
2011;21:2575–83.

37. Gahleitner A, Solar P, Nasel C, et al. Die Magnetresonanztomographie in der Dentalradiologie

(Dental-MRT). Radiologe 1999;39:1044–50.

38. Schara R, Sersa I, Skaleric U. T1 relaxation time and magnetic resonance imaging of inflamed
gingival tissue. Dentomaxillofac Radiol 2009;38:216–23.
39. Assaf AT, Zrnc TA, Remus CC, et al. Evaluation of four different optimized magnetic-resonance-
imaging sequences for visualization of dental and maxillo-mandibular structures at 3 T. J Cranio-
Maxillo-Fac Surg 2014;42:1356–63.
40. €r IA. Pulp-dentin biology in restorative dentistry. Part 7: The exposed pulp. Quintessence Int
Mjo
2002;33:113–35.

41. Murray PE, Hafez AA, Smith AJ, et al. Bacterial microleakage and pulp inflammation associated
with various restorative materials. Dent Mater 2002;18:470–8.

42. Zero DT, Zandona AF, Vail MM, Spolnik KJ. Dental caries and pulpal disease. Dent Clin North Am
2011;55:29–46.
43. Lloyd CH, Scrimgeour SN, Chudek JA, et al. Application of magnetic resonance microimaging to
the study of dental caries. Caries Res 2000;34:53–8.

44. Weglarz WP, Tanasiewicz M, Kupka T, et al. 3D MR imaging of dental cavities-an in vitro study.
Solid State Nucl Magn Reson 2004;25:84–7.

45. Idiyatullin D, Garwood M, Gaalaas L, Nixdorf DR. Role of MRI for detecting micro cracks in teeth.
Dentomaxillofac Radiol 2016;45:20160150.
46. Choe BY, Woo DC, Ahn HS. Dental rf coil, head-fixing unit, and magnetic resonance imaging
system including same. US Patent US 20120288820 A1, Alexandria, VA: United States Patent
Application Publication; 2012.
47. Soher BJ, Dale BM, Merkle EM. A review of MR physics: 3T versus 1.5T. Magn Reson Imaging.
Clin N Am 2007;15:277–90.

JOE  Volume 50, Number 4, April 2024 Endodontic Applications of Magnetic Resonance Imaging 449