Nathan C. Nelson Department of Molecular Biomedical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA More than any other body region, when contemplating an imaging study of the head, the clinician should ask, “What question needs to be answered?” The choice of imaging modality depends on whether the suspected disease process involves osseous or soft tissue structures, the location of the affected structures in the head, and the availability and cost of different imaging modalities. Radiographs are a useful initial diagnostic test for suspected osseous pathology, as radiography is more widely available in general veterinary practice and has a lower financial cost. Radiographs provide a rapid global survey of the skull in cases of head trauma and identify major fractures or dislocations. They can serve as an initial imaging test for intranasal pathology, but superimposition of the osseous structures of the nasal cavity complicates interpretation. Skull morphology is complex, and requires careful positioning for radiographs, as even slight obliquity of the patient will significantly limit diagnostic interpretation of osseous structures. In comparison, computed tomography (CT) and magnetic resonance imaging (MRI) are more sensitive and specific tests for imaging of the head. Generally, CT is the preferred imaging test for intranasal pathology, as it is rapid and provides good contrast resolution between gas, osseous structures, and soft tissue pathology. MRI is the preferred test for intracalvarial disease and disease affecting neurologic structures, as it has superior soft tissue contrast resolution compared to CT. The continued increase in availability of CT and MRI, along with their superior imaging abilities, explains the decrease in skull radiography in recent years at major referral centers though radiography of the head remains common in general practice. Ultrasound is the preferred first‐line imaging test for superficial soft tissue pathology in the head. The eye and structures in the ventral neck (such as the thyroid lobes, lymph nodes, and salivary glands) are readily accessible to ultrasound imaging. Additionally, only mild sedation or manual restraint alone is typically sufficient to complete an ultrasound examination of these areas, providing a more reasonable cost to the client compared to CT or MRI. Surrounding osseous anatomy prevents ultrasound evaluation of intranasal or intracalvarial soft tissue diseases. Dental imaging requires high resolution given the small size of teeth in canine/feline patients. Dental pathology may be relatively subtle and affects small anatomic structures. Standard digital radiography detectors, such as the type used for abdominal/thoracic radiographs, may not provide the high‐resolution imaging necessary to detect subtle dental disease. Additionally, the large size of standard digital radiography detectors limits detector positioning around the head. Ideally, dental radiography is performed with dedicated dental digital imaging plates, which are much smaller than standard digital plates, allowing intraoral positioning of the plate. Additionally, dental digital plates are also higher resolution than standard digital imaging systems, allowing optimal diagnostic ability of dental pathology. Cone beam CT units are becoming more common in veterinary practice. Cone beam CT units produce cross‐sectional images similar to standard CT units, but have higher spatial resolution, making them ideal for dental imaging (Figure 11.1). Cone beam CT units have limited contrast resolution, providing inadequate imaging of soft tissue structures of the head. They scan more slowly than traditional CT scanners, making them especially sensitive to patient motion artifacts (Figure 11.2). For these reasons, their use should be limited to evaluating dental or other osseous structure of the head and not for evaluation of soft tissue pathology. High‐quality, diagnostic radiographs of the head require careful patient positioning. Ventrodorsal projections are particularly important for intranasal pathology. Comparison between the left and right nasal cavities on the VD projection facilitates identification of subtle, unilateral abnormalities which disrupt left/right nasal symmetry. Left/right comparison is only possible when the patient is straight and the left/right nasal cavities are symmetrically positioned, which can be challenging or impossible in the awake patient. Some areas of the skull require specialized oblique projections for complete evaluation (as discussed below), and may result in the patient discomfort if performed when the patient is awake. Strong sedation or anesthesia allows careful patient positioning, prevents patient discomfort, and is required for skull radiographic procedures. Due to the complex skull anatomy, there are many potential radiographic projections that may be employed to “pick off” or accentuate a particular area of interest. A standard skull examination may include some or all of these projections. The following skull projections are available. As the head is symmetric, images must be labeled with indicators for sidedness. On ventrodorsal and rostrocaudal projections, a single L or R indicator is sufficient to unambiguously label the side of the head. On oblique images, double simultaneous marking with both L and R indicators leaves no question as to which is left and which is right anatomy (Figure 11.6). Typically, closed mouth ventrodorsal and lateral projections are the minimum requirements for a study of the skull, but additional projections are recommended depending on the clinical question to answer (Figure 11.9). Below are recommended imaging protocols for each area of interest. Where available, CT has largely replaced skull radiography as the preferred modality for imaging the skull or nasal cavity. Indications for head CT are broad, with scanning commonly performed in patients with epistaxis or chronic/recurrent intranasal symptoms, masses anywhere in the head (outside the brain), oral tumors, swelling in the laryngeal/pharyngeal reason, or retrobulbar disease. Imaging before and after the administration of intravenous iodinated contrast is standard, except in cases of trauma when surveying for osseous abnormalities. Scanning after the administration of intravenous contrast improves the visibility of most soft tissue pathology, and is particularly important when evaluating intracranial lesions if MRI is not available. When scanning for disease within the oral cavity or pharyngeal region, the mouth should be opened, which helps separate soft tissues of the oral and laryngeal regions. Compression on the ventral head/neck should be avoided, as it can distort regional anatomy. Magnetic resonance imaging provides superior contrast resolution of soft tissue structures, and is preferred over CT when imaging the brain. Though MRI can image other structures of the head, such as the nasal or oral cavity, the increased time and cost of MRI compared to CT mean that it is rarely used for other purposes (Figure 11.10). Additionally, MRI artifacts at gas–soft tissue interfaces interfere with evaluation of structures in the nasal cavity and airway. Ultrasound is infrequently used in head imaging compared to other modalities, and is primarily used to investigate the ventral soft tissues of the head or in cases of ocular pathology. The osseous structures of the skull block ultrasound beam transmission, so it can only be used for intracranial evaluation in neonatal patients (where the open sutures are used as acoustic windows) or in patients with persistent open fontanelles (Figure 11.11). Ultrasound is ideally suited for imaging of the eye and readily identifies structures such as the cornea, lens, iris, and posterior chamber. The cornea is anesthetized through topical medications. The ultrasound transducer is placed directly on the cornea (coupled with sterile ultrasound gel) or on the closed eyelids. Light sedation limits patient movement, though many patients will allow imaging without sedation and only manual restraint. Indications for ocular ultrasound are many. Patients with cataracts are imaged to ensure lack of retinal separation prior to phacoemulsification (Figure 11.12). Ocular trauma or acute onset of hyphema are indications for ultrasound to evaluate for ocular damage, lens luxation, or masses that are otherwise obscured from direct examination by the presence of intraocular blood (Figure 11.13). Intraocular tumors are readily identified on ocular ultrasound and the tissue of origin may be identified (Figure 11.14). Retrobulbar ultrasound is performed in cases of exophthalmos or when a retrobulbar mass or inflammation is suspected. The globe may be used as an acoustic window to investigate the retrobulbar space, though approaches dorsal or ventral to the zygomatic arch allow imaging posterior to the globe as well. Other areas of the head and neck, primarily around the pharyngeal/laryngeal region, are accessible to ultrasound imaging. When examining this region, a routine methodology is followed to identify the following structures: medial retropharyngeal and mandibular lymph nodes, carotid artery, thyroid lobes, mandibular salivary glands, esophagus, trachea, laryngeal tissues. The goal of imaging with radiographs, CT, or MRI is to determine whether nasal disease is more likely to be inflammatory (some form of rhinitis) or neoplastic in origin, and the extent of disease [2]. While radiographs may provide this information, CT or MRI better define the boundary of nasal pathology, and allow a more definitive or specific prioritized differential diagnosis list. There are many causes of nasal inflammation, including infectious rhinitis, immune‐mediated rhinitis, and foreign body rhinitis, though many cases are idiopathic in nature. Rhinitis typically results in bilateral increase in soft tissue within the nasal cavity, without destruction of the turbinates or surrounding bones. Fine turbinate detail is retained but with chronic rhinitis, mild turbinate atrophy may occur. Concurrent sinusitis causes mild fluid accumulation within the sinuses and/or thin soft tissue thickening along the luminal margin of the sinus due to inflamed or hyperplastic sinus mucosa. The appearance of Aspergillus rhinitis in dogs is unique, and does not conform to the description of other causes of rhinitis described above. While Aspergillus does cause increased opacity in the nasal cavity, it also causes extensive turbinate lysis (Figure 11.15) and tends to be unilateral. Unlike neoplasia (which also causes turbinate lysis), the turbinate loss due to Aspergillus is more multifocal, resulting in larger gas‐filled voids throughout the nasal cavity. A solitary, large mass is not present but instead, the soft tissue thickening is more peripheral among the margins of the nasal cavity and along the residual turbinates. Fungal plaques in the frontal sinuses are common, and appear as heterogeneous mixed gas and fluid (unlike other causes of sinusitis where the thickening is homogeneous). The frontal sinus skyline projection in particular is useful to identify these changes. Other less common fungal agents may also cause rhinitis, and can appear mass‐like or more diffuse [3]. Fungal rhinitis in cats is rare, but Aspergillus and Cryptococcus are most common and appear as a solitary mass or more diffuse disease (Figure 11.16) [4]. Foreign body rhinitis typically localizes around the foreign body, resulting in a more focal and unilateral abnormality compared to other causes of rhinitis (Figure 11.17). The rostral nasal cavity is typically affected, as the nasal aperture is the point of entry for most foreign bodies (Figure 11.18). The foreign body establishes conditions ideal for secondary Aspergillus infection, which can result in more extensive turbinate lysis in the area of the foreign body. Unless mineral or metallic, the foreign body is not often easily identified, particularly for small foreign bodies such as plant material (grass) or other soft tissue material (such as nasal parasites) (Figure 11.19). Neoplasms arise from the soft tissues or osseous structures of the nasal cavity. There is a wide variety of tumor types reported, with epithelial origin (carcinoma) being the most common tumor type in dogs (up to 75% of cases), but tumors arising from connective tissue (sarcomas) are also common. Lymphoma, osseous tumors, and other uncommon tumor types (such as leiomyoma and transmissible veneral tumor) also occur [5–7]. Most nasal neoplasms result in a solid, solitary mass that causes local turbinate destruction (Figures 11.20 and 11.9). If the mass become very large, it causes lysis of the larger osseous structures surrounding the nasal cavity, such as the nasal or maxillary bones, but this tends to be reserved for only the largest, most advanced masses. Commonly, there will be a large amount of fluid among the remaining turbinates due to secondary hemorrhage or rhinitis. Large masses trap fluid caudally in the nasal cavity or frontal sinus due to obstruction of nasal drainage. Most nasal tumors (carcinomas, sarcomas) do not have unique features that allow further differentiation of tumor type, though some have unique characteristics that can be identified on CT. For instance, nasal chondrosarcoma typically has patchy mineral regions within it, a feature not expected in carcinomas [8]. Nasal lymphoma may be a solitary mass but tends to be more widespread than other nasal tumors and can cause mild diffuse thickening that mimics rhinitis (Figure 11.21). Tumors (e.g., osteosarcoma) arising from the larger bones surrounding the nasal cavity are centered on the osseous structures rather than within the nasal cavity. Olfactory neuroblastomas are uncommon tumors that arise from the sensory neuroendocrine olfactory cells in the upper part of the nasal cavity. These center on the cribriform plate with invasion into the cranial vault and caudal nasal cavity (Figure 11.22) [9]. Most of these features are not distinguishable on radiography, but readily are identified on CT, allowing more specific prioritization of neoplastic differential diagnoses [10]. CT allows assessment of draining lymph nodes in cases with nasal neoplasia, which can guide the decision on whether to sample them for complete staging purposes. CT lacks specificity to distinguish inflammatory/reactive lymph nodes from neoplastic lymph nodes, so ultimately image‐guided aspiration and cytologic/histopathologic evaluation may be necessary to make the final determination [11]. Though MRI can be used to image nasal neoplastic disease, it is less commonly employed compared to CT, due to the necessity of general anesthesia, greater cost, and longer scan times. MRI provides additional information about structures that are difficult to assess on CT, such as reactive meningeal changes surrounding a tumor or bone marrow involvement of the calvarium [12] (Figure 11.23). Intranasal epidermoid cysts are rare causes of mass effects within the nasal cavity of brachycephalic dogs. They have a characteristic appearance on CT and MRI, being entirely fluid attenuating on CT or homogenously fluid hyperintense on MRI (Figure 11.24). They cause resorption of adjacent osseous structures via pressure necrosis [13]. Patients with retrobulbar disease often present with a complaint of exophthalmos or loss of vision. Ultrasound is a useful first‐line imaging test in these patients given its accessibility, speed, and ease of use. Because deeper tissues are imaged, a microconvex transducer is more useful than a linear transducer. Multiple imaging approaches are available to examine the retrobulbar tissues. The globe can be used as a standoff, imaging the deeper retrobulbar tissues through the globe. Approaches dorsal to the globe (just above the upper eyelid) or just dorsal or ventral to the zygomatic arch provide a more complete retrobulbar examination. The osseous orbit limits ultrasound examination by preventing evaluation of the most medial aspect of the retrobulbar space, so ultimately CT or MRI may be necessary for further examination if ultrasound is unsuccessful in identifying the cause of clinical signs. External puncture wounds or wounds extending from the oral cavity into the retrobulbar region from oral or external foreign body puncture cause retrobulbar abscesses (Figure 11.25). Foreign material often causes distal acoustic shadowing on ultrasound, though very small or nondense/nonattenuating material (such as chronic, water‐logged plant material) lacks a shadow. Regional fat appears reactive, being hyperechoic/hyperattenuating on ultrasound and causing stranding within retrobulbar fat on CT. Distinct fluid pockets may be seen. Regional contrast enhancement is expected on CT and MRI (Figure 11.26). Retrobulbar neoplasia may arise from the osseous structures of the skull or the retrobulbar soft tissues (Figure 11.27). CT and MRI are more useful than ultrasound to determine tissue of origin, particularly if the tumor is deep in the retrobulbar space where shadowing from osseous structures prevents definitive determination with sonography [14]. Myxosarcomas often extend into the retrobulbar space [15]. Uncommonly, retrobulbar lipomas and rhabdomyosarcomas occur [14, 16]. Primary bone tumors such as osteosarcoma, chondrosarcoma, or multilobular tumor of bone may become large enough to extend into the retrobulbar space; osseous involvement makes neoplasia much more likely than inflammatory causes of retrobulbar disease [14]. Zygomatic sialadenitis is a rare cause of retrobulbar mass effect and is discussed below. Traumatic injuries unpredictably damage osseous and soft tissue structures of the head. Survey radiography provides a broad overview of affected structures, but is insensitive to fissures and small fractures in some areas. Fractures of the mandibular ramus are difficult to identify as they superimpose with the zygomatic arch. In comparison, mandibular fractures are readily identified. Calvarial or maxillary fractures may require multiple oblique projections as the fractures are seen only in a tangential plane. CT is preferred over radiography when surveying for head trauma due to improved performance in identifying small fractures and areas of soft tissue pathology. Intravenous contrast is typically not necessary, but improves the ability to screen for some types of soft tissue pathology or intranasal trauma (Figure 11.28). If acute brain trauma is suspected, then MRI is preferred as it better identifies brain contusions or direct brain injury, though CT is relatively sensitive for acute hemorrhage so may be a useful screening test in patients with subdural or subarachnoid hemorrhage (Figure 11.29) [17].
CHAPTER 11
Imagingof the Head
Choice of Modality
Imaging Protocols
Radiography
Lateral Projection
Closed Mouth Ventrodorsal Projection
Open Mouth Ventrodorsal Projection
Frontal Sinus Skyline Projection
Left 45° Ventral‐Right Dorsal Oblique and Right 45° Ventral‐Left Dorsal Oblique
Left 10° Rostral‐Right Caudal Oblique and Right 10° Rostral‐Left Caudal Oblique [1]
Bulla Projections
Nasal Disease
Middle Ear Disease
TMJs
Routine Trauma Survey
Computed Tomography
Magnetic Resonance Imaging
Ultrasound
Specific Diseases of the Head/Skull
Nasal Diseases
Rhinitis
Nasal Foreign Bodies
Nasal Neoplasia
Intranasal Epidermoid Cysts
Retrobulbar Diseases
Trauma