British Journal of Radiology 75 (2002),552-562 © 2002 The British Institute of Radiology
MRI of the teeth
L M Tutton, BDS, MSc and
P R Goddard, MD, FRCR
Open Scanner, Bristol Oncology Centre and the University of the West of England, Bristol, UK
 |
Abstract
|
|---|
The teeth and periapical structures were demonstrated on MRI using an open MRI system. There was good visualization of normal structures including crowns of teeth, pulp chambers and the neurovascular bundle of the inferior dental nerve. Dental and periapical pathology was shown.
|
Introduction
|
|---|
The teeth and dento-alveolar area are always visible on MRI of the lower one third of the head but are usually ignored by radiologists and clinicians. There has been a paucity of literature studying dental pathology using MRI. Publications on the use of MRI in dentistry have included studies concerning the temporomandibular joint [1], nasopharyngeal tumours [2, 3] and assessment of mandibular and maxillary bone prior to dental implant placement [4]. This pictorial review describes and displays MRI appearances of the anatomy and pathology of the dento-alveolar region.
MRI was performed using a circular polarized receiver head coil on a Siemens 0.2T Open MRI system (Siemens, Erlangen, Germany), and 3 mm slices were undertaken in the transverse, coronal and oblique sagittal planes using T1 weighted spin echo, short tau inversion recovery (STIR) and fast low angle shot (FLASH) three-dimensional sequences. The best technique for demonstrating the crowns of the teeth was for the subject to keep the mouth closed and refrain from swallowing during scanning. Saliva retained in this way becomes a useful contrast medium in the oral cavity.
|
Normal anatomy
|
|---|
The normal volunteer (Figures 1–10
) had no dental pathology. With the help of the landmark key (Table 1), normal anatomy can be identified on all the scan sequences. The enamel and dentine of the teeth appeared black owing to a lack of unbound protons. The dental pulp chamber, containing nerves, blood vessels and connective tissue within the teeth, appeared white or grey on T1 weighted and STIR imaging. Size and shape varied from tooth to tooth, and at what level the scan was taken, but closely followed the expected shape that would be visualized if the tooth were physically sectioned. The anatomy and morphology of the teeth and their interrelationship with surrounding structures were well visualized. Cortical bone was seen as a black zone outlined by moderate signal from external soft tissues and high signal internal fatty marrow. On fat annulled scans (STIR), fatty marrow had low signal and appeared dark grey.
View larger version (169K):
[in this window]
[in a new window]
|
Figure 1. Scout view for the oblique sagittal T1 weighted scans. Slices were taken at 3 mm intervals with Figure 2 being the buccal slice (labelled as 95).
|
|
View larger version (112K):
[in this window]
[in a new window]
|
Figure 2. Right oblique sagittal T1 weighted spin echo scans. The crowns and roots of the premolar and molar teeth are clearly seen. The two buccal roots (B) of the upper first molar are visible on (a) and disappear as the sections move medially at 3 mm intervals to (c), where only the longer and larger palatal root apex (I) is seen. Note the proximity of the root apices of the first and second molars to the floor of the maxillary antrum on which there are two large mucous retention cysts. The root apex of the lower third molar (H) is in close proximity to the inferior dental (ID) canal (arrowed A). The ID canal containing the ID nerve is seen as two parallel grey lines running in the ramus and body of the mandible, posterosuperior to the anterior inferior aspects. Between the two lines the neurovascular bundle and surrounding tissue and fat are seen as higher signal (whiter). Extensions of the ID neurovascular bundle contents are seen as thin grey lines passing into the pulp chambers of the mandibular teeth via their respective apical foramina (R). On the more medial scan (c), the ID canal is seen to connect to the mental foramen (C) and through this the mental branches of the ID nerve innervate the lower lip and skin of the chin. The unerupted upper right third molar (wisdom tooth) (G) and its root apex is seen. White (high signal) pulp chambers (F) within the structure of most of the teeth are shown well on (b), as this image is passing through the midline of most of the teeth from canine to molars. The temporomandibular joint including the condylar head (E) glenoid fossa, articular eminence and the articular disc (D) are clearly visible anterior to the external auditory canal. On (c), the cortex of the mylohyoid ridge or line (J) (into which the mylohyoid muscle is inserted) is seen as a black (low signal) line. In (c), the soft tissues of the chin, lips and nose are clearly seen and the angulation and relationship of the upper and lower central incisors to one another. The upper incisors appear to be attached to the premaxilla by the palatal root surface only, but there is an overlying very thin black line of cortical bone over the labial root surface.
|
|
View larger version (163K):
[in this window]
[in a new window]
|
Figure 3. Transverse Transaxial T1 weighted spin echo image. The teeth are shown in their relationship to the dental arch in cross-section from superior (maxilla) to inferior (mandible). The crowns of the unerupted upper wisdom teeth (G) are seen at the most posterior aspect of the dentoalveolar process with the small black circles of the apices of the second molars just visible anterior to them. The root apices of these and other erupted teeth in the dental arch are seen as black circles, and the pulp chamber (F) can be seen centrally as a white dot within a few of them. The ID nerve (N) enters theinferior dental canal of the mandible via the mandibular foramen. Immediately anterior and medial to this is the prominent extension of bone known as the lingula, to which the sphenomandibular ligament is attached. The crowns of the upper teeth are well shown with much larger central high signal (white) pulp chambers (F).
|
|
View larger version (161K):
[in this window]
[in a new window]
|
Figure 4. This shows the pulp chamber of the crown of the second molars ellipsoidally shaped and flattened buccopalatally (F). The roots of the first molars are closer together than on previous scans and the buccal roots are more ellipsoidal than circular in shape.
|
|
View larger version (166K):
[in this window]
[in a new window]
|
Figure 5. This shows the unerupted lower wisdom teeth (H) lying with their incompletely formed roots angled buccally.
|
|
View larger version (161K):
[in this window]
[in a new window]
|
Figure 6. Bilateral thickening of cortical bone on the lingual surface of the mandible, each known as the torus mandibularis (O).
|
|
View larger version (158K):
[in this window]
[in a new window]
|
Figure 7. The transverse axial short tau inversion recovery sequence (at 3 mm above Figure 5) shows the pulp chambers (F) of the teeth to have high signal (white and light grey) owing to a relatively high water content. There is also high signal around the unerupted wisdom teeth owing to high water content in the dental follicles (P), formed during development of the tooth crown. There is high signal (white) at the back of the tongue (T) owing to saliva being retained in the oropharynx. High signal is also seen in the cerebrospinal fluid, in slow flowing vessels in the neck and in the parotid glands.
|
|
View larger version (117K):
[in this window]
[in a new window]
|
Figure 8. Coronal T1 weighted scans show the roots of the molar teeth are well visualized with their pulp chambers in vertical cross-section. The most posterior scan slices through the third molar region. The unerupted lower wisdom teeth (H) are seen with their crowns angled lingually and roots angled buccally in close approximation with the buccal cortex of the mandible. There is high signal from their pulp chambers and there is a band of high signal immediately surrounding the crown, demarcated by a lower signal edge of bone. This band or "halo" enveloping the crown is an eruption follicle (Q) developed from the remnants of cells left over from the formation of the enamel and crown of the tooth, and usually disappears on eruption of the tooth into the mouth. It is sometimes referred to as an eruption cyst and occasionally enlarges and is termed a dentigerous cyst, which has to be surgically marsupialized.
|
|
View larger version (90K):
[in this window]
[in a new window]
|
Figure 9. Image 6 mm forward of Figure 8 shows therelationship of the roots of the molar teeth to the floor of the maxillary antrum and the mucous retention cysts (benign mucosal antral cysts) on the floor of the antrum. The low signal from the cortical bone forming the floor of the maxillary antrum is seen sinuously passing over the contours of the apices of the molar teeth.
|
|
View larger version (65K):
[in this window]
[in a new window]
|
Figure 10. (a, b) Coronal T1 weighted images through the canine and incisor teeth showing the nasopalatine suture (S) and the pulp chambers (F) of the upper central incisors.
|
|
|
Dental pathology
|
|---|
A series of patients with known periapical pathology were also imaged. The MRI studies (Figures 11–29) show a variety of dental problems.
View larger version (64K):
[in this window]
[in a new window]
|
Figure 11. (a, b) Fast low angle shot T1 Transverse images at the level of mandibular teeth apices and 3 mm inferiorly, showing an area in the midline of reduced medullary (marrow) bone signal. Medullary fatty marrow is normally seen as high signal (white) and the abnormal area is seen as moderate signal (grey). This abnormal area surrounds the roots of lower central incisors, with evidence of considerable thinning of the cortical bone both lingually and buccally, which is seen as a thin black line. Lateral to the abnormal area on both sides there is an area of very low signal (black) in the medullary bone similar to cortical bone, and this indicates sclerosis.
|
|
View larger version (71K):
[in this window]
[in a new window]
|
Figure 12. (a, b) T2 weighted images at the same level as Figure 11 show the moderate signal bone lesion seen on the T1 weighted scans is shown as high signal (white) that does not easily distinguish it from marrow. It does, however, on close inspection, appear to have slightly higher signal than marrow.
|
|
View larger version (119K):
[in this window]
[in a new window]
|
Figure 13. Periapical radiographs of lower central incisors showing a large radiolucency at the apices of both central incisors and extending to the margins of the mesial root surface of the lateral incisors, with loss of the lamina dura of all teeth involved. The radiolucency extends inferiorly into the mandible and has a ragged and uneven inferior margin.
|
|
View larger version (74K):
[in this window]
[in a new window]
|
Figure 14. (a) T1 weighted coronal image through midline of teeth shows a diamond shaped area of moderate signal (light grey) around the roots of the lower central incisors, extending from the mesial aspect of the lower lateral root apices to a point 1 cm lower in the midline. This corresponds to the abnormal area previously demonstrated on transverse MR images. It is demarcated around its sides and base by a thin dark grey/black line of low signal due to sclerotic bone. The lower right central incisor root apex is displaced laterally away from the midline by about 0.75 cm. (b) Coronal short tau inversion recovery (STIR) (fat suppressed) image comparable to (a). It shows the area described above as bright white, which indicates that it has a high water content and may be oedematous in nature. The corresponding moderate signal on T1 weighted scans shows that it is not a solitary bone cyst. It must either contain a high protein or high cellular content. The appearances are not typical of blood and were considered likely to be owing to infection, which, based on the history, was chronic. The diagnosis of infected periapical cyst was confirmed by drainage of pus and bloodstained exudate via the root canals of the lower central incisors, and was successfully followed by root filling of these teeth.
|
|
View larger version (128K):
[in this window]
[in a new window]
|
Figure 15. Transverse T1 weighted spin echo image at the level of the maxillary incisor root apices shows a grey mottled area to the left of the midline.
|
|
View larger version (133K):
[in this window]
[in a new window]
|
Figure 16. A scan taken 3 mm inferior to Figure 15 shows a distinct circular area of slightly lower signal than surrounding marrow around the root apex of the upper left central incisor. There is no cortical bone on the labial aspect of the root surface, which is at the level of the discharging sinus as ascertained clinically. There is a thick black line of low signal on the palatal aspect that indicates sclerosis and thickening of the palatal cortex of the dentoalveolar ridge, probably in response to the chronic infection.
|
|
View larger version (130K):
[in this window]
[in a new window]
|
Figure 17. Descending 3 mm inferiorly to Figure 16, into the crown area of the incisor teeth, there is no signal from the root canal of the upper left central incisor that can be compared with the normal signal from the pulp chambers of adjacent teeth. This is owing to the root filling and restoration, which has obliterated the pulp chamber. There is a slight haziness to the margins of this crown which may be due to the metallic nature of the restoration interfering as an artefact.
|
|
View larger version (136K):
[in this window]
[in a new window]
|
Figure 18. Short tau inversion recovery (STIR) sequence at comparative levels to Figure 15 shows a mottled grey appearance of mixed signal from the same area, indicating high water content in the lesion. This heterogeneity, and the moderate signal on T1 weighted imaging, is not typical of a simple cyst but of chronic infection.
|
|
View larger version (125K):
[in this window]
[in a new window]
|
Figure 19. Periapical radiograph of upper left central incisor showing an area of radiolucency around the root with well defined margins. There is an inadequate root filling in the same tooth.
|
|
View larger version (134K):
[in this window]
[in a new window]
|
Figure 20. (a) Spin echo T1 weighted coronal scan through the upper central incisors shows deviation of the midline to the left of the nasopalatine suture and the conchae of the nose. There is an arc shaped black line (arrowed) of cortical bone above the root apex of the upper left central incisor with a dark grey vertical line to it from its apex. (b) Short tau inversion recovery (STIR) sequence in the coronal plane at a comparative level to (a) shows a light grey spherical area (arrowed) of high signal at the apex of the upper left central incisor. The appearances of high signal on T1 weighted and STIR images are typical of an inflammatory exudate indicating a chronic infection related to the upper left central incisor.
|
|
View larger version (149K):
[in this window]
[in a new window]
|
Figure 21. Periapical radiograph of lower right premolars and molar. There is radiolucency around the partially resorbed root of the first premolar. There is also radiolucency through the crown of this tooth.
|
|
View larger version (87K):
[in this window]
[in a new window]
|
Figure 22. (a) Oblique sagittal T1weighted spin echo image ofthe right side of the face, hemisecting the teeth of both jawsthrough their anteroposterior midline. There is loss of signal around the roots of the lower premolars (arrowed) compared with the surrounding medullary bone signal. There is a large overjet (arrowed) of the upper anteriors in relation to the lower anterior teeth, typical of a Class 2 Division 1 (Angles classification) jaw relationship for this subject. (b) Oblique short tau inversion recovery (STIR) sequence image at a position equivalent to (a) shows an area of high signal (arrowed) at the same site as the area of signal loss in (a). The high signal around the crowns (V) of the lower molars is saliva retained in the mouth.
|
|
View larger version (125K):
[in this window]
[in a new window]
|
Figure 23. Transverse T1 weighted spin echo image at the level of the root apices in the mandible. There is an area of reduced signal from the medullary bone inthe region of the right canine and premolar root apices, suggesting a more cystic nature. This is lower signal than muscle. The short tau inversion recovery (STIR) fat annulling sequence relating to this same area showed high signal around and below the apex of the lower first premolar, indicative of a lesion with high water and protein content. This agreed with the radiological and clinical findings that this was a periapical granuloma or cyst relating to the lower right first premolar with localized osteitic reaction in the surrounding medullary bone. The infected cyst later discharged through the buccal plate and the tooth was successfully treated endodontically.
|
|
View larger version (168K):
[in this window]
[in a new window]
|
Figure 24. Right oblique sagittal T1 weighted spin echo image. The crown of the lower right first molar is missing and there is an area ofreduced signal at its apex (arrowed), probably owing to replacement of fatty marrow in the bone by inflammatory exudate. Similar areas of abnormality were seen in the left side of the mandible. The abnormality on MRI was considerably more extensive than on orthopantomogram and intraoral radiographs, and indicated the areas that required treatment.
|
|
View larger version (131K):
[in this window]
[in a new window]
|
Figure 25. Transverse T1 weighted spin echo image shows the broken down crown of the lower right molar with pooling of saliva.
|
|
View larger version (185K):
[in this window]
[in a new window]
|
Figure 26. Oblique sagittal T1 weighted spin echo image through the buccal roots of the upper molar teeth. A low signal area posterior to the lower right second molar, just below the superior cortex of the mandible (arrowed), is shown. This is the lower third molar (wisdom tooth) retained root fragment from the surgical removal of the tooth.
|
|
View larger version (149K):
[in this window]
[in a new window]
|
Figure 27. Transverse T1 weighted spin echo image at a level 3 mm below the incisal surface of the upper molars shows a low signal inclusion in the medullary bone of the mandible in the same area (arrowed), which confirms the finding. There is also an artefact in the leftpremolar/molar region owing to a ferromagnetic inclusion in a dental restoration [5].
|
|
View larger version (170K):
[in this window]
[in a new window]
|
Figure 28. Transverse T1 weighted image at the level of mandibular root apices shows a further finding where a round area (arrowed) of 1 cm diameter, medium signal (dark grey), is seen on the mesiolingual aspect of the left ramus of the mandible, near the angle where it joins the body of the mandible.
|
|
View larger version (157K):
[in this window]
[in a new window]
|
Figure 29. Short tau inversion recovery (STIR) sequence at a corresponding level to Figure 28, shows a high signal lesion (white), 1 cm in diameter, in the same area (arrowed). The patient had been to the dentist the previous day and recollected an abnormally unpleasant sensation in the area when a second inferior dental nerve block injection was given. The position of the lesion was near the site where the neurovascular bundle of the mandibular branch of the trigeminal nerve splits into the lingual nerve and the inferior dental nerve and associated vessels. The injection had perforated one of these vessels, resulting in a small haematoma visible on MRI.
|
|
|
Case history 1 (Figures 11–14)
|
|---|
A 40-year-old Caucasian male with a history of trauma to his upper and lower incisors, sustained as a young rugby player, had suffered an acute flare-up relating to the lower central incisor region 2 years previously. Although root canal therapy and apical surgery were recommended at that time, the patient had eschewed conventional antibiotic therapy and endodontic therapy and taken the homeopathic remedy "Silica" for approximately 2 years. Intraroral radiographs (Figure 13) demonstrated a large diamond shaped radiolucency, 2 cm in size, extending half way up the roots of the central incisors and inferiorly towards the lower border of the mandible with the apex of the diamond in the midline. The lower central incisor roots were displaced laterally away from the midline.
|
Case history 2 (Figures 15–20)
|
|---|
A 34-year-old male with well maintained dentition. The upper left central incisor was fractured in a childhood accident and was inadequately root filled and restored with too short a post and crown by a previous dentist. This tooth had recently developed a sinus discharging pus in the buccal sulcus. The radiographic appearances (Figure 19) were typical of chronic periapical pathology such as abscess or granuloma. The tooth was firm on palpation and caused no problems to the patient. The tooth was subsequently successfully root filled and restored with resolution of pathology.
|
Case history 3 (Figures 21–23)
|
|---|
A 52-year-old male presented with an acute infection relating to a longstanding chronic periapical translucency on the lower right first premolar, with evidence of root resorption, as shown radiographically (Figure 21).
|
Case history 4 (Figures 24–25)
|
|---|
A 55-year-old male with neglected dentition. Areas of chronically infected bone were shown on MRI. The patient's general and dental health markedly improved after removal of the offending teeth and infected bone, as indicated by MRI.
|
Case history 5 (Figures 26–29)
|
|---|
A 46-year-old female with advanced conservation on all but the canines and incisors in both jaws, with a history of a retained root following surgical extraction of the lower right wisdom tooth.
MRI of the teeth and the periapical region provides useful information about the pulp chambers of the teeth, the relationship of teeth to adjacent structures and the localization of impacted teeth. In cases of suspected occult periapical infection, MRI may demonstrate abnormality that is not apparent on radiographs. In cases where the radiograph is known to be abnormal, MRI may demonstrate the extent of disease and the relationship to other critical structures. MRI has proven as useful as CT in assessment before implantation and has become established as the optimal method of assessing the temporo-mandibular joints.
|
Acknowledgments
|
|---|
This paper was based on posters presented at the Radiology '99 Conference, Birmingham. The project was sponsored by The Friends of the Bristol Oncology Centre and Bristol Royal Infirmary Magnetic Scanner Appeal. Ethical approval was given by the Ethical Committee of UBHT. Acknowledgments are due to Mrs Sadie Dunne, Dr Mike Keen and Mr Ron Hartley-Davies.
Received for publication December 18, 2000.
Revision received July 16, 2001.
Accepted for publication August 23, 2001.
|
References
|
|---|
- Uberoi R, Goddard P, Ward-Booth P, Kabala J. TMJ function and dysfunction. Developments in Magnetic Resonance 1996;2:9–15.
- Kabala J, Goddard P, Cook P. Magnetic resonance imaging of extracranial head and neck tumours. Br J Radiol 1992;65:375–83.[Abstract]
- Leslie A, Fyfe E, Guest P, Goddard P, Kabala JE. Staging of squamous cell carcinoma of the oral cavity and oropharynx: a comparison of MRI and CT in T- and N-staging. J Comput Assist Tomogr 1999;23:43–9.[Medline]
- Gray CF, Redpath TW, Smith FW. Pre-surgical dental implant assessment by magnetic resonance imaging. J Oral Implantol 1996;22:147–53.[Medline]
- Tutton LM, Dunne S, Goddard PR. MRI of the teeth, part 2: artefacts caused by dental materials. Clinical MRI 1999;9:31–4.
Top
Abstract
Introduction
