Head and neck cancer is the sixth most common cancer in the world, and in the United States it accounts for approximately 3% of all cancers in men and 2% of all cancers in women (1). Squamous cell carcinoma represents more than 90% of all head and neck cancers. The incidence rates are more than twice as high in men as in women and are greatest in men older than 50. The risk of malignancy is 6 times greater for smokers than nonsmokers. The synergistic effect of alcohol and smoking increases the risk of the disease 2.5 times more than the simple additive risk of either risk factor alone. Nasopharyngeal carcinoma is associated with Epstein-Barr virus. Human papillomavirus infection is another factor implicated in the development of oral cavity and oropharyngeal squamous cell carcinoma (2,3). Mutation of the tumor suppression gene p53 is one of the most common genetic alterations in squamous cell
carcinomas of the head and neck (4,5,6). Chewing betel nuts, particularly when mixed with tobacco, serves as a local irritant of oral mucosa and therefore a stimulant to the formation of malignancies. Wood dust and woodcarving have been implicated in the increased incidence of nasopharyngeal malignancies, as has the ingestion of smoked fish, particularly in parts of Asia.

Head and neck cancer is an important topic in oncologic imaging, because imaging findings can aid significantly in the detection, staging, and treatment evaluation of these tumors. Patients presenting with primary tumors that are confined at the time of initial diagnosis (T1-2NOMO) have an excellent cure rate. Unfortunately, at the time of initial diagnosis many patients already have regional nodal metastases (45%) or even distant metastases (10%) (7). For all stages combined, about 85% of persons with oral cavity and pharynx cancer survive 1 year after diagnosis. The 5-year and 10-year relative survival rates are 59% and 44%, respectively (1). Therefore, the major goals of the evaluation of patients with head and
neck cancer are locoregional staging, exclusion of distant metastases, and synchronous second primary tumors. Already in 1988, Minn et al. described increased uptake of fluorodeoxyglucose (FDG) in patients with tumors of the head and neck (8).

FDG is not specific for neoplastic processes in head and neck. It accumulates physiologically in various regions of head and neck (Table 34.1) (9). Low to strong FDG uptake occurs in the normal lymphoepithelial tissue of the nasopharynx and oropharynx (Fig. 34.1A,B), the palatine tonsils (Fig. 34.1H), and at the base of the tongue (Fig. 34.1I). Variable but usually low FDG uptake is visible in the salivary glands, which physiologically secrete low amounts of glucose (Figs. 33.21 and 34.1D,E). Muscular FDG uptake can be seen in the palate (Fig. 34.1C); in the floor of the mouth (Figs. 33.21G and 34.1E); in the tip of the tongue (Fig. 34.1F); in muscles of the face (Fig. 34.1H), including the masticatory muscles, neck (Fig. 34.1K), and larynx (Fig. 34.1J); in nervous patients; and in patients who speak during the FDG-uptake phase. However, physiological uptake in the laryngeal muscles is usually symmetric and can be discriminated from cancer. In patients with laryngeal nerve palsy, glottic muscle uptake on the healthy side can be observed (see Fig. 34.22). If patients do not close their eyes during the study, muscles of the eyes and eyelids will also show increased uptake.

Postsurgical changes that distort the normal anatomy after resection and surgical reconstruction can elicit atypical muscle activity (Fig. 34.2). Other reasons for FDG uptake not related to tumor are inflammation, infection, and radiation necrosis.

In patients with head and neck cancer, dental problems are common. If disease of the dental roots and/or periodontal space is present, the teeth will be surgically removed before starting radiation treatment. This will cause an inflammation reaction in the jaw, which can be visible in FDG-PET-CT for several weeks. Inflammation can mimic a malignant lesion, and it is important to interpret images in the light of adequate clinical information.

Dental metal work also causes artifacts. Nonremovable metal dentures, bridgework, and fillings will produce a defect in the emission image that can also be seen in the final attenuation-corrected image (9,10). However, most malignant lesions avidly take up FDG and therefore are usually identifiable without problems. In addition, attenuation correction by means of a CT scan can be deteriorated due to dental metalwork, which can mimic FDG
uptake in attenuation-corrected images (11). Measurement of SUV values can be influenced by these artifacts. Therefore, measurements of FDG concentrations in areas adjacent to nonremovable dentures may be of limited value. Based on our experience, we feel that additional viewing of the non-attenuation-corrected emission scans can render image interpretation of oral pathologies with PET more reliable.