Chapter 18 Facial bones

The most frequent reason for radiological examination of the facial bones is trauma to the region; plain radiographic imaging of the area remains a popular and appropriate method of initial assessment in the acute setting, providing information relatively quickly and with a relatively low radiation dose compared to computed tomography (CT), the other imaging method best suited for providing information on bony injury to the area. Low-dose CT is considered a suitable method for demonstration and assessment of orbital fractures, as plain radiographic images can sometimes be inconclusive and may not give the 3D information needed before treatment of fractures. For many years it has been suggested that plain radiography may only be useful in cases showing clinical signs that clearly suggest surgical intervention,¹ but current guidelines still show that plain radiography has a place in the assessment of facial and orbital injury.² Magnetic resonance imaging (MRI) may also be considered, but as scans are undertaken supine, the teardrop effect of the herniating orbital tissue may not be as well demonstrated as in the prone CT scan with coronal sections. CT will provide better bony definition on the images.

CT may also be required to provide information in trauma cases when plain images in the general facial bones survey are inconclusive or are difficult to produce to a high enough standard; this is often due to difficulties associated with patient condition in severe trauma when excessive oedema may reduce image contrast.

The facial bones can be demonstrated by a general plain radiographic survey that includes the maxilla, mandible, orbits, nasal bones and zygomae. However, provision of specific information on some of these areas requires alternative or additional projections so that a diagnosis can be made. The mandible and zygomae both require individual examination in case of injury, and plain radiography is the initial examination method of choice for these areas. In non-trauma-related indications the mandible may require CT examination to assess the progress of dental implants.

The temporomandibular joints (TMJs) can also be imaged by plain radiography, which will provide information on condylar dislocation and loss of joint space. MRI will give more useful information regarding the joint itself and, since internal disruption is the most commonly encountered problem in the joint, MRI is most suitable. Arthrography will provide dynamic information regarding the joint.

Injury to the nasal bones is not considered a reason for routine radiographic examination, but clinical specialists (e.g. for ear, nose and throat, and maxillofacial follow-up) may consider special nasal bones projections to be useful.² This would be the case when assessing fragment displacement and septal deviation.

Although a significant number of patients presenting with facial trauma will attend on a trolley, patients also frequently arrive as a ‘walk-in’ case and can be examined erect at a skull unit or erect bucky. Erect examination with a horizontal beam is essential for some projections where it is necessary to demonstrate air–fluid levels, and must be attempted whenever possible. This is particularly relevant in the case of blow-out fractures of the orbital floor, where fluid level in the maxillary sinus is used as an indicator of this type of injury.

The choice of projections for facial bones appears to vary according to referring clinical or individual hospital protocol, but rarely includes the lateral facial bones projection. It is common to find that at least two OM projections are used, with tube angle or no angle, and there have been studies in the past to investigate whether a single projection can be used;^3,⁴ the most likely projection that can be suggested for this is referred to as the OM 30° in related articles, but it is necessary to ask whether this means that the orbitomeatal baseline (OMBL) lies at 30° to the IR and using a central ray perpendicular to the IR, or if a true OM with OMBL at 45° is used with a caudal tube angle of 30°. Fortunately, one article does include an image that shows the petrous ridge clearly level with the middle of the maxillary sinuses, indicating that the projection required an OMBL at 30° to the IR but with no tube angle.³ This position is familiar as the routine OM for orbits,⁵ which is collimated to include only the orbital outlines and maxillary sinuses for that area; clearly, if this projection is used for full facial bones assessment then all facial bones must be included in the primary beam. Investigation of the idea of one ‘ideal’ projection for facial bones assessment has involved consideration of articles and textbooks relating to radiographic positioning or recommendation of projections in facial trauma, and has yielded some additional interesting results that give rise to some very pertinent points when discussing imaging and referral.

All radiographers use eponymous terms for a few projections, for example Towne’s projection of the skull, Judet’s views of the acetabulum and Garth’s projection of the shoulder. Unfortunately, this makes the actual technique used less memorable than the name. In the last 30 years UK textbooks have aimed to use nomenclature that indicates the actual position for the projection, rather than the name of the projection’s designer, with addition of the eponymous title next to the descriptive title. Unfortunately this is not necessarily the case internationally, and eponymous titles are frequently used, leading to a varying range of projection names which are then incorporated into journal articles, potentially creating confusion or even misinterpretation. A search for a list of all eponymously named projections showed that there are approximately 200 in existence,⁶ although many are supplementary specialist projections that have been largely superseded by additional imaging modalities. Of this long list, only 17 appeared to be familiar in the UK.

An example of variation in nomenclature when discussing radiography of the facial bones can be centred around the OM projection and therefore has particular relevance to this chapter. In the UK OM tends to refer to a position with the OMBL at 45°, to ensure that the petrous ridge is cleared from the bases of the maxillary sinuses;^5,^7,⁸ in the US the same projection is named PA axial, transoral, Waters’ or even parietocanthal projection.^9,¹⁰ Position descriptors for this same projection also vary, with UK texts indicating an OMBL angle of 45°^5,^6,⁸ and US texts stating 37°,^9,¹⁰ yet all who provide image evaluation criteria insist that their position will see the petrous ridge in the same place, just clear of the lower borders of the maxillary sinuses. One point to raise is that, although it is fairly easy to judge a 45° OMBL to IR angle, can anyone actually claim to accurately judge 37°? US authors do use an alternative way to ensure their positioning is accurate, by referring to alignment of the meatomental line (MML) at 90° to the IR.⁹ The MML is the line joining the external auditory meatus and the chin, and it is not clear whether it can be relied on as accurate in patients with developmental deformities of the mandible, such as mandibular prognathism.

It is also noted that the way the relationship of OMBL to IR is described can also vary, with texts giving the suggested OMBL angle either related to the IR ^5,⁷ or related to the perpendicular.⁹ This is very confusing, even for experienced authors in radiography, but probably almost impossible for students to understand.

Even articles written in the UK cause confusion: another article exploring the concept of a single view assessment in trauma investigates the potential of either ‘the OM 15° and OM 30° view’ but does not make it clear what the actual positioning for the projections entails (again, is the OMBL at 45° for each, with caudal angle, or does the angle refer to the OMBL position?). Study of the article reveals that the OM 15° is referred to thus:

The occipitomental film with a 15° tilt (OM 15°) was considered to be the most useful view because it projects the orbital floor separate from the petrous ridge and displays the zygomatic arches.¹¹

Unfortunately this means little in the search for an explanation of the actual projection details, since the tilt referred to is not explained as either tube angle or OMBL angle. Clearance of the petrous ridge to below the inferior orbital margins is seen in any OM projection with more than 20° chin lift (i.e. when the OMBL starts at 90° to the IR and the chin is raised so that the OMBL is moved through 20°), and so information on the petrous ridge in the quote above does not help clarify the situation. There is no positional information provided for the OM 30° projection in the article.

A word of warning: ensure you know the correct relationship of baselines and IR before proceeding. In addition, much work has been written by maxillofacial surgeons on appropriate projections in facial trauma; in the absence of extensive radiographic experience on their part, how can we expect these articles to be consistent in their meaning for everyone?

General survey of facial bones

Requests that define the desired examination as ‘facial bones’ require a general OM and (sometimes) lateral survey of the area. OM projections are based on a position with the OMBL at 45°, using a range of caudal beam angles. More than one OM projection may be included in the survey, and two examples are shown of the 45° OM: without angulation in Figure 18.1B and with 30° caudal angulation in Figure 18.1C. Although discussion in the previous section shows that a 30° elevation of the OMBL from a perpendicular relationship to the IR has been suggested as a standalone projection for survey of facial bones,³ it does not appear to be universally adopted as such at present.

The IR is vertical for all projections of facial bones, orbits and nose unless the patient presents supine on a trolley; antiscatter grid is required, with the exception of lateral nasal bones.

OM facial bones – basic projection (Fig. 18.1A,B,C)

Positioning

• The patient is seated, facing the IR

• The chin is placed in contact with the midline of the IR and the chin position is adjusted until the OMBL has been raised 45° from the horizontal

• The median sagittal plane (MSP) is perpendicular to the IR, which is assessed by checking that the external auditory meati (EAMs) or lateral orbital margins are equidistant from it

Figure 18.1 (A) OM facial bones. (B) OM facial bones; (C) OM facial bones with 30° caudal angulation.

Beam direction and focus receptor distance (FRD)

1 Horizontal, at 90° to the IR and making an angle of 45° with the OMBL or

2 Initially horizontal, with caudal angulation applied according to requirements of the examination

100 cm FRD

Common errors	Possible reasons
Asymmetry of facial structures	Rotation about MSP
Position of petrous ridge too high	Chin not raised enough. It has been noted that radiographers frequently ask patients to put their nose and chin onto the erect IR for this projection; this will only serve to raise the chin approximately 30°. It has also been noted that some imaging departments use this method with a 15° caudal angle, which only serves to clear the petrous ridge to the inferior margins of the antra; an almost identical image to the true OM 45° with horizontal beam will result, but there will be some distortion caused by application of the angle

Common errors

Possible reasons

Asymmetry of facial structures

Rotation about MSP

Position of petrous ridge too high

Chin not raised enough. It has been noted that radiographers frequently ask patients to put their nose and chin onto the erect IR for this projection; this will only serve to raise the chin approximately 30°. It has also been noted that some imaging departments use this method with a 15° caudal angle, which only serves to clear the petrous ridge to the inferior margins of the antra; an almost identical image to the true OM 45° with horizontal beam will result, but there will be some distortion caused by application of the angle

Lateral facial bones (Fig. 18.2A,B,C)

This projection is largely considered of little or no value ³ but may still be used in some centres.

Figure 18.2 (A) Lateral facial bones; (B) centring for lateral facial bones; (C) lateral facial bones.

(C) Reproduced with permission from Bryan GJ. Skeletal anatomy. 3rd ed. Edinburgh: Churchill Livingstone; 1996 and Gunn C. Bones and joints. 4th ed. Edinburgh: Churchill Livingstone; 2002.

Positioning

• The patient is initially seated, facing the IR

• The trunk is brought as close as possible to the receptor unit and the patient is asked to sit with their spine as erect as possible. This helps the patient turn their head more easily into the required lateral position

• The head is turned through 90° to bring the affected side in contact with the IR

• The MSP is parallel to the IR; there should be no tilt or rotation of the head. This can be assessed by checking the midline of the cranium over the top and symmetry of the frontal bone and orbits

• Asking the patient to gently close their eyes will assist in maintenance of the position; as the radiographer leaves the receptor unit the patient will often follow this movement with their eyes and potentially affect the position of the head

Beam direction and FRD

Horizontal, at 90° to the IR

100 cm FRD

Centring

To the inferior border of the zygoma

Collimation

Superior orbital margins, symphysis menti, TMJs, nasal bones

Criteria for assessing image quality

• Superior orbital margins, symphysis menti, TMJs and nasal bones are demonstrated

• Superimposition of the malar processes of maxilla, orbital outlines and TMJs

• Sharp image demonstrating the malar processes of maxilla in contrast to the air-filled maxillary sinuses, and the orbits in contrast to other bones of the face. The mandible is seen in contrast to the soft tissues of the face. Nasal bones are over-penetrated

Common errors	Possible reasons
Non-superimposition of the floor of the anterior cranial fossa; malar processes seen as one above the other; orbital outlines seen as one above the other	MSP tilted, usually with the upper part of the head tilted towards the IR. If a patient cannot comply with the required position, a compensating caudal angle can be used to reduce the effects of the tilt
Lateral orbital margins seen side by side and not superimposed; malar processes seen displaced in a horizontal direction	Head is rotated; this is often encouraged when the patient ‘slumps’ in their chair rather than sitting with their spine erect, as described in ‘positioning’ for this projection