ORIGINAL ARTICLE

Influence of common fixed retainers on

the diagnostic quality of cranial magnetic
resonance images
Miriam Shalish,a Nimrod Dykstein,b Silvina Friedlander-Barenboim,c Eliel Ben-David,d John Moshe Gomori,e
and Stella Chaushuf
Jerusalem, Israel

Introduction: Orthodontists are often asked to remove fixed retainers before magnetic resonance imaging
(MRI). This study was undertaken to assess the effects of 2 commonly used fixed retainers on MRI distortion
and whether they should be removed. Methods: MRI scans were performed on a dry skull with Twistflex (Den-
taurum, Ispringen, Germany) and Ortho Flex Tech (Reliance Orthodontic Products, Itasca, Ill) retainers. Two
neuroradiologists independently ranked the distortions. The influence of the fixed retainers' alloys, their distance
to the area of diagnosis, location, strength of the magnetic field, and the spin-echo sequence were examined.
Statistical analysis included kappa and Pearson chi-square tests. Results: Ortho Flex Tech retainers caused
no distortion. Twistflex retainers caused distortion in 46% of the tests in areas close to the retainer (tongue
and jaws). Maxillary fixed retainers and the combination of maxillary and mandibular fixed retainers further
increased the distortion. Greater distortion was observed with 3-T magnetic fields and T1-weighted spin-echo
sequences. Conclusions: Removal of the Ortho Flex Tech retainer is unnecessary before MRI. Removal of
the Twistflex should be considered if the MRI scans are performed to diagnose areas close to the fixed retainers,
when 3-T magnetic fields and T1-weighted sequences are used, and when both maxillary and mandibular fixed
retainers are present. (Am J Orthod Dentofacial Orthop 2015;147:604-9)

M
agnetic resonance imaging (MRI) has become
a common diagnostic tool for various medical
and dental conditions (migraine and cluster
headaches, epilepsy and other seizure disorders, multiple
sclerosis, head and neck tumors, temporomandibular
joint disorder, and many others).1 The advantages of
MRI over other imaging techniques include excellent
soft-tissue quality and absence of ionizing radiation.
a
Clinical lecturer and director, Postgraduate Program, Department of Orthodon-
tics, Hebrew University-Hadassah School of Dental Medicine, Jerusalem, Israel.
b
Clinical instructor, Department of Orthodontics, Hebrew University-Hadassah
School of Dental Medicine, Jerusalem, Israel.
c
Head, Oral & Maxillofacial Radiology Unit, Oral Medicine Department, Hebrew
University-Hadassah School of Dental Medicine, Jerusalem, Israel.
d
Clinical instructor, Department of Radiology, Hadassah Medical Center,
Jerusalem, Israel.
e
Professor and head, Neuroradiology Unit, Department of Radiology, Hadassah
Medical Center, Jerusalem, Israel.
f
Associate professor and chair, Department of Orthodontics, Hebrew University-
Hadassah School of Dental Medicine, Jerusalem, Israel.
All authors have completed and submitted the ICMJE Form for Disclosure of
Potential Conflicts of Interest, and none were reported.
Address correspondence to: Miriam Shalish, Department of Orthodontics,
Hebrew University-Hadassah School of Dental Medicine, POB 12272, Jerusalem,
Israel; e-mail, mshalish@hadassah.org.il.
Submitted, July 2014; revised and accepted, November 2014.
0889-5406/$36.00
Copyright Ó 2015 by the American Association of Orthodontists.
http://dx.doi.org/10.1016/j.ajodo.2014.11.022
604
The common MRI apparatus, often designated by their
magnetic fields, are of 1.5 and 3 teslas (T), which repre-
sent the magnetic intensity of the machine.1 Ferromag-
netic materials (several common metals) distort the
magnetic field and result in distortion of the magnetic
resonance image, affecting its diagnostic value.1
Recently, there has been an increase in the number
of orthodontic patients worldwide. According to Proffit
et al2 in 1998, over 30% of white youths, 11% of
Mexican-Americans, and 8% of African Americans
reported receiving orthodontic treatment. At the end
of active orthodontic treatment, the results are usually
retained for many years using fixed retainers made of a
metal wire bonded to the lingual side of the anterior
teeth.
Owing to the increased number of MRI referrals,3
orthodontists are frequently requested to remove metal
orthodontic appliances before the MRI procedures.
Removal and reinstallation of an orthodontic appliance
is laborious and bears a risk of relapse of teeth to unde-
sired positions. As a result, orthodontists are interested
in understanding the effects of orthodontic appliances
on MRI imaging to minimize the need for removal.
The presence of orthodontic appliances has 3 impor-
tant implications for MRI diagnosis. First, magnetic field
Shalish et al
Fig 1. Mandibular Ortho Flex Tech fixed retainer
embedded into a polyvinylsiloxane key.
interactions may pose a risk of detachment to the
patient. Loose orthodontic devices pose a significant
danger for the patient.4 Second, metals in the appliances
may cause heating.5 Third, they may produce image dis-
tortions that could affect the diagnostic quality of
the MRI.6
Only a few studies have been conducted to evaluate
the influence of different fixed orthodontic appliances
on the diagnostic quality of magnetic resonance im-
ages.4-9 One study showed that stainless steel brackets
cause significant distortion, rendering several cranial
regions nondiagnostic, whereas plastic, ceramic, and
titanium brackets cause minimal interference.6
The effect of different types of fixed retainers
has scarcely been addressed so far.7,8 When bonded
firmly, fixed retainers pose no risk to the patient in the
magnetic field of a clinical MRI scanner.9 However,
image quality may be significantly affected.8
The purpose of this study was to assess the in-vitro
effects of the 2 most common types of fixed retainers
on MRI distortion and answer the dilemma of whether
they should be removed before MRI.
MATERIAL AND METHODS
A dry skull of a female was used for this study. Algi-
nate impressions of the maxillary and mandibular dental
arches were taken, and plaster models were poured. The
2 types of fixed retainers were fabricated directly on the
lingual aspects of the maxillary and mandibular incisors
and canines of the plaster models. The fixed retainer wire
was attached to the lingual surface of each tooth using
dental floss. Polyvinylsiloxane key (Elite HD 1 type
0 putty consistency; Zhermack, Badia Polesine, Italy)
was prepared for each jaw to hold the retainer in place.
The retainers were embedded into this material (Fig 1).
For each retainer key, the skull was immersed in a
container including a diluted 2% solution of gadolinium
contrast agent (Dotarem; Guerbet, Villepinte, France)
American Journal of Orthodontics and Dentofacial Orthopedics
605
Fig 2. The skull with the maxillary and mandibular polyvi-
nylsiloxane keys.
and placed in the scanner. The MRI sequences were per-
formed for that particular set; then a different tray was
inserted, and subsequent magnetic resonance scans
were completed (Fig 2). The diluted gadolinium solution
created a background signal to the bone that simulated
that of soft tissues. It did not affect the metal distortion
artifacts but allowed them to be seen using MRI
sequences identical to those in clinical imaging.10,11
To determine the effect of the polyvinylsiloxane key
on the MRI scans, 2 scans were carried out on the key
without the retainers, which served as the controls.
Two experienced neuroradiologists (J.M.G. and
E.B.-D.) independently evaluated the images to deter-
mine the amount of distortion. All images were pre-
sented at once on the neuroradiologists' computer
screens. The images were randomly labeled. The neuro-
radiologists were asked to rank the images according to
the distortions in these regions using a modified receiver
operating characteristic method of distortion classifica-
tion (Table I).
The following 5 variables were examined.
1. The type of alloy of the fixed retainer. Two
commonly used fixed retainers were chosen for
this study: Twistflex, 0.18 in (Dentaflex, triple
strand twisted; Dentaurum, Ispringen, Germany),
an alloy of stainless steel (chromium, 17%-20%;
nickel, 8%-12%; carbon, 0.08%-0.15%; the rest is
May 2015  Vol 147  Issue 5
606 Shalish et al

Table I. Distortion classification Table II. Effect of the type of fixed retainer alloy:
distortion scores of Twistflex (TF) vs Ortho Flex Tech
Diagnostic or
Score Image appearance nondiagnostic (OF) retainers
1 No distortion or artifact Diagnostic Distortion
2 Minimal distortion or artifact Diagnostic
3 Moderate distortion or artifact Moderately diagnostic Wires None Minimal Moderate Severe Total
4 Severe distortion Nondiagnostic TF
Count 54 20 7 19 100
% 54.0 20.0 7.0 19.0 100.0
mainly iron); and Ortho Flex Tech (Reliance Ortho- OF
dontic Products, Itasca, Ill), an alloy composed of Count 80 0 0 0 80
gold, 58.33%-58.70%; silver, 1.5%-6.0%; copper, % 100.0 0.0 0.0 0.0 100.0
28.0%-31.0%; nickel, 0.3%-6.5%; and zinc, Control
Count 20 0 0 0 20
5.0%-6.3%.
% 100.0 0.0 0.0 0.0 100.0
2. The distance between the region of interest and the
retainer. Ten regions of the head were assessed:
tongue, mandible, anterior maxilla, posterior distortion in 46% of the tests. In 20% of the cases, the
maxilla, orbits, nasopharynx, pituitary gland, frontal distortion was minimal; in 7%, it was moderate; and in
lobe, temporal lobes, and brain stem. 19%, it was severe (Table II).
3. The location of the fixed retainer. Six trays were The Twistflex retainers mainly affected anatomic re-
fabricated: maxillary and mandibular Twistflex, gions in their vicinity, such as tongue, mandible, and
maxillary and mandibular Ortho Flex Tech, and maxilla. Other anatomic regions, not near the Twistflex
maxillary and mandibular control polyvinylsiloxane retainer, were unaffected (Table III). The differences be-
keys without retainers. tween the distortion scores in the different anatomic re-
4. The strength of the magnetic field. The scans were gions were highly statistically significant (P \0.001).
performed in the MRI unit, Hadassah Medical Cen- Significant differences related to the location of the
ter, Jerusalem, Israel, on both 1.5-T (Avanto) and Twistflex fixed retainer were found. Maxillary retainers
3-T (Trio) systems (both, Siemens Medical Solu- caused more severe distortion than did mandibular re-
tions, Erlangen, Germany) using standard multi- tainers, and both maxillary and mandibular retainers
element head coils. further increased the severity of the distortion
5. The spin-echo sequence. Imaging sequences (P 5 0.05) (Table IV).
included axial fast-spin echo T2-weighted images The intensity of the magnetic field (3 T vs 1.5 T) had a
(TR/TE 3500/90 ms), axial and sagittal conven- marked impact on the distortion scores. Greater distor-
tional spin-echo T1-weighted images (TR/TE tion was observed for the stronger magnetic field (3 T)
500/14 ms), with a slice thickness of 5 mm and (P 5 0.005) (Table V).
a 20% gap. The spin-echo sequence also significantly affected the
degree of distortion, with T1-weighted images causing
Statistical analysis more severe distortion than T2-weighted images
Statistical analysis included descriptive statistics, (P 5 0.03) (Table V).
kappa tests to evaluate interexaminer calibrations, and The polyvinylsyloxane tray (control) caused no
Pearson chi-square tests. The Pearson chi-square test distortion of the magnetic resonance images.
was used to compare the distortions caused by the The evaluations of the 2 neuroradiologists were
different retainers and different MRI settings. The signif- compared using kappa tests. Satisfactory calibration
icance level for the statistical testing was set at P \0.05. between the 2 examiners was obtained (kappa values,
The analysis was carried out using SPSS software 0.6-0.8).
(version 21.0; IBM, Armonk, NY).
DISCUSSION
RESULTS Fixed lingual retainers, bonded to mandibular ante-
Statistically significant differences were found rior teeth, were introduced in the 1970s12 and recog-
between the distortion scores assigned to the Twistflex nized as an important part of orthodontic treatment to
and Ortho Flex Tech retainers (P \0.001). No distortion prevent relapse or secondary crowding of the mandib-
was observed with the Ortho Flex Tech retainers, similar ular incisors.13 The main advantages of mandibular fixed
to the control, whereas the Twistflex retainers caused retainers compared with removable retainers are that

May 2015  Vol 147  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Shalish et al 607

Table III. Effect of the distance from the Twistflex Table V. Effect of magnetic field strength and spin-
retainer: distortion scores in different anatomic echo sequence: distortion scores of Twistflex at 1.5 T
regions vs 3 T, and T1-weighted vs T2-weighted sequence
Distortion Distortion

None Minimal Moderate Severe MRI None Minimal Moderate Severe

Tongue 1.5 T
Count 26 6 3 5 Count 37 15 2 6
% in anatomic site 65.0 15.0 7.5 12.5 % in MRI 61.7 25.0 3.3 10.0
Mandible 3T
Count 27 11 0 2 Count 17 5 5 13
% in anatomic site 67.5 27.5 0.0 5.0 % in MRI 42.5 12.5 12.5 32.5
Maxilla, anterior T1
Count 23 1 4 12 Count 28 15 2 15
% in anatomic site 57.5 2.5 10.0 30.0 % in T 46.7 25.0 3.3 25.0
Maxilla, posterior T2
Count 38 2 0 0 Count 26 5 5 4
% in anatomic site 95.0 5.0 0.0 0 % in T 65.0 12.5 12.5 10.0
Orbit and other areas
T, Teslas; T1, T1-weighted; T2, T2-weighted.
Count 40 0 0 0
% in anatomic site 100.0 0.0 0.0 0

The use of MRI has recently increased,20 as well as

3-T MRI for pediatric patients.21 A recent study, based
Table IV. Effect of location of fixed retainers: distor- on electronic records of members of 6 large integrated
tion scores of Twistflex in the jaws health systems from different regions of the United
Distortion States, reported a nearly quadruple increase in MRI
between 1996 and 2010.3
Jaw None Minimal Moderate Severe Total
The influence of orthodontic appliances on MRI has
Both jaws
Count 28 10 6 16 60
already been studied in several articles. Elison et al6
% in jaw 46.7 16.7 10.0 26.7 100.0 showed that stainless steel brackets cause significant
Maxilla distortions. Another study showed that orthodontic
Count 13 3 1 3 20 appliances have a significant influence on imaging of
% in jaw 65.0 15.0 5.0 15.0 100.0 the frontal and temporal lobes of the brain and should
Mandible
Count 13 7 0 0 20
be removed for accurate MRI diagnosis of epilepsy.22
% in jaw 65.0 35.0 0.0 0.0 100.0 In contrast, and despite the fact that there is a defin-
itive trend to use fixed retention for long-term stabil-
ity,17 there is limited information regarding its effects
on MRI scans.7,8 The constant rise in the frequency
they are invisible and compliance free.14 The disadvan- of fixed retainers use, the trend to extend the periods
tages include the demanding technique,15 the potential of fixed retention, and the increase in the frequency of
for distortion of the wire, and increased plaque and MRI referrals increase the chances of finding a fixed
calculus accumulation compared with removable retainer in a patient referred for MRI. Since the
retainers.16 A survey showed that nearly a third of the magnetic cross-section of the fixed retainer is small rela-
practitioners in the United States use fixed retainers in tive to common orthodontic appliances, it should be
the mandibular arch.17 Another more recent study expected to cause a smaller perturbation of the field.
showed that 40% of the members of the American Asso- Therefore, a careful study of the effects could minimize
ciation of Orthodontists use fixed retainers and do not considerably the need for removal of fixed retainers
instruct patients to have them removed at a specific before MRI.
time.18 In Norway, a fixed retainer bonded to all anterior The most commonly used fixed retainers are Twist-
teeth in the mandible was most common (66.4%). flex and Ortho Flex Tech. There are additional less
Although retention lasted 2 to 3 years (34.7%) in the commonly used fixed retainers based on stainless steel,
maxilla, or 3 to 5 years (23.8%) in the mandible, nickel-titanium alloy, or fiberglass.
41.5% of the orthodontists left the retainers in place Our study showed that in most cases, the observed
for more than 5 years.19 distortion was minimal for both types of retainers; for

American Journal of Orthodontics and Dentofacial Orthopedics May 2015  Vol 147  Issue 5
608
most MRI screenings, there is actually no indication for
removal of the fixed retainer. Distortion was observed
only with Twistflex retainers. The physics underlying
these results appears to be rather simple. Ortho Flex
Tech retainers contain only minor amounts (0.3%-
6.5%) of the ferromagnetic metal nickel, and all the
other metals in the alloy are nonferromagnetic. The
ferromagnetic cross-section is therefore small and so is
the perturbation it can cause to the flow of the magnetic
field. In contrast, the Twistflex retainer is a stainless steel
alloy with a high percentage of ferromagnetic metals,
such as iron. Our results support clinical findings on
MRI scans of 3 patients wearing Orthoflex retainers7
and in-vitro findings on MRI scans of stainless steel vs
nickel-titanium retainers.8However, in this last study,
the authors used a simplified in-vitro model, in which
stainless steel and nickel-titanium wires were embedded
in a cylinder placed in the MRI field.8 To the best of our
knowledge, our study is the first in-vitro study that
mimics the actual position of retainers in a patient's
mouth and compares the widely used Ortho Flex Tech
and Twistflex retainers in a controlled manner.
Even though a distortion was observed with Twist-
flex, it was rather limited, affecting only the closest
anatomic regions, such as tongue, mandible, or maxilla.
As already mentioned, the reason for this rather small
effect was most probably the small mechanical cross-
section of the fixed retainer. Our results support previous
studies on the effects of stainless steel brackets, which
also showed that the distortions diminish with the dis-
tance from the appliance.6
Importantly, the amount and location of the material
influenced the scores of distortion, meaning that lon-
ger maxillary retainers, or maxillary and mandibular
retainers together, pose higher risks than a mandibular
retainer alone. Fortunately, maxillary fixed retainers
are less commonly used than mandibular ones and are
left in situ for shorter periods after treatment.19
The distortion was directly proportional to the strength
of the magnetic field—ie, a greater distortion was observed
for the stronger intensity field (3 T)—and was also influ-
enced by the spin-echo sequence, with T1-weighted
causing much more distortion than T2-weighted.
Based on our results, we can therefore conclude that
fixed retainers in which the metal alloy contains greater
amounts of ferromagnetic metals, such as Twistflex,
may distort MRI images of the maxilla, mandible,
or tongue, and their removal should be limited to
MRI scans performed to analyze pathology in these
anatomic regions. However, the use of fixed retainers
made of nonferromagnetic metals, or small percentages
thereof, such as Ortho Flex Tech, avoids removal in
all cases.
May 2015  Vol 147  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Shalish et al
CONCLUSIONS
1. According to our findings, removal of Ortho Flex
Tech fixed retainers is not necessary.
2. Removal of Twistflex fixed retainers is not required
before most MRI scans. Removal should be consid-
ered only if the MRI scan is aimed at diagnosing pa-
thologies in the jaws or tongue, especially when using
the 3-T, T1-weighted, spin-echo sequence, and when
both maxillary and mandibular retainers are present.
3. According to our findings, if the region of interest of
the MRI is not near the fixed retainer, there is no
need to remove it.
REFERENCES
1. Yousem DM, Grossman RI. Neuroradiology: the requisites. 3rd ed.
Philadelphia: Mosby/Elsevier; 2010. p. 4-15.
2. Proffit WR, Fields HW Jr, Moray LJ. Prevalence of malocclusion
and orthodontic treatment need in the United States: estimates
from the NHANES III survey. Int J Adult Orthodon Orthognath
Surg 1998;13:97-106.
3. Smith-Bindman R, Miglioretti DL, Johnson E, Lee C, Feigelson HS,
Flynn M, et al. Use of diagnostic imaging studies and associated
radiation exposure for patients enrolled in large integrated health
care systems, 1996-2010. JAMA 2012;307:2400-9.
4. Kemper J, Priest AN, Schulze D, Kahl-Nieke B, Adam G, Klocke A.
Orthodontic springs and auxiliary appliances: assessment of mag-
netic field interactions associated with 1.5 T and 3 T magnetic
resonance systems. Eur Radiol 2007;17:533-40.
5. G€org€ul€
u S, Ayyildiz S, Kamburoglu K, G€okçe S, Ozen T. Effect of
orthodontic brackets and different wires on radiofrequency heat-
ing and magnetic field interactions during 3-T MRI. Dentomaxil-
lofac Radiol 2014;43:20130356.
6. Elison JM, Leggitt VL, Thomson M, Oyoyo U, Wycliffe ND. Influ-
ence of common orthodontic appliances on the diagnostic quality
of cranial magnetic resonance images. Am J Orthod Dentofacial
Orthop 2008;134:563-72.
7. Aizenbud D, Hazan-Molina H, Einy S, Goldsher D. Craniofacial
magnetic resonance imaging with a gold solder-filled chain-like
wire fixed orthodontic retainer. J Craniofac Surg 2012;23:e654-7.
8. Tymofiyeva O, Vaegler S, Rottner K, Boldt J, Hopfgartner AJ,
Proff PC, et al. Influence of dental materials on dental MRI. Den-
tomaxillofac Radiol 2013;42:20120271.
9. Klocke A, Kahl-Nieke B, Adam G, Kemper J. Magnetic forces on
orthodontic wires in high field magnetic resonance imaging
(MRI) at 3 tesla. J Orofac Orthop 2006;67:424-9.
10. Yoshioka H, Schlechtweg PM, Kose K. Magnetic resonance imag-
ing. In: Weissman B, editor. Imaging of arthritis and metabolic
bone disease. Philadelphia: Mosby/Elsevier; 2009. p. 34-48.
11. Hornak JP. The basics of MRI. 1996. Available at: http://www.cis.
rit.edu/htbooks/mri/. Accessed October 27, 2014.
12. Knelrim RW. Invisible mandibular cuspid to cuspid retainer. Angle
Orthod 1973;43:218-20.
13. Little RM, Riedel RA, 
Artun J. An evaluation of changes in mandib-
ular anterior alignment from 10 to 20 years postretention. Am J
Orthod Dentofacial Orthop 1988;93:423-8.
14. Zachrisson BU. Clinical experience with direct-bonded orthodontic
retainers. Am J Orthod 1977;71:440-8.
15. Zachrisson BU. The bonded lingual retainer and multiple spacing
of anterior teeth. J Clin Orthod 1983;17:838-44.
Shalish et al
16. Pandis N, Vlahopoulos K, Madianos P, Eliades T. Long-term peri-
odontal status of patients with mandibular lingual fixed retention.
Eur J Orthod 2007;29:471-6.
17. Keim RG, Gottlieb EL, Nelson AH, Vogels DS 3rd. 2002 JCO study
of orthodontic diagnosis and treatment procedures. Part 1. Results
and trends. J Clin Orthod 2002;36:553-68.
18. Valiathan M, Hughes E. Results of a survey-based study to identify
common retention practices in the United States. Am J Orthod
Dentofacial Orthop 2010;137:170-7.
19. Vandevska-Radunovic V, Espeland L, Stenvik A. Retention:
type, duration and need for common guidelines. A survey
Nostalgia Advertisement from a 1935 issue of the Journal
American Journal of Orthodontics and Dentofacial Orthopedics
609
of Norwegian orthodontists. Orthodontics (Chic.) 2013;14:
e110-7.
20. Dang CM, Zaghiyan K, Karlan SR, Phillips EH. Increased use of MRI
for breast cancer surveillance and staging is not associated with
increased rate of mastectomy. Am Surg 2009;75:937-40.
21. Dagia C, Ditchfield M. 3T MRI in paediatrics: challenges and clin-
ical applications. Eur J Radiol 2008;68:3009-19.
22. Costa AL, Appenzeller S, Yasuda CL, Pereira FR, Zanardi VA,
Cendes F. Artifacts in brain magnetic resonance imaging due to
metallic dental objects. Med Oral Patol Oral Cir Bucal 2009;1:
E278-82.
May 2015  Vol 147  Issue 5