Search

Nonepithelial tumors of the larynx are rare and represent a minority of all laryngeal neoplasms. Imaging has an important role in the diagnosis, treatment planning, and surveillance of these entities. However, unfamiliarity with these neoplasms can cause diagnostic difficulties for radiologists, especially because many of the imaging findings are nonspecific. By using a systematic approach based on clinical history, patient age and gender, lesion location, endoscopic results, and specific imaging findings, the differential diagnosis can often be narrowed. These tumors typically affect the submucosal layer, so if a tumor has an intact mucosa at endoscopy, a nonepithelial neoplasm is the most likely diagnosis. Nonepithelial tumors of the larynx can arise from the laryngeal cartilage or muscle or from the surrounding lymphoid tissue or blood vessels. Consequently, imaging findings typically correspond to the specific cell type from which it originated. Recognizing specific features (eg, metaplastic bone formation, macroscopic fat, or enhancement pattern) can often help narrow the differential diagnosis. In addition, identification of noncircumscribed borders of the lesion and invasion of the adjacent structures is key to diagnosis of a malignant process rather than a benign neoplasm. Understanding the pathologic correlation is fundamental to understanding the radiologic manifestations and is ultimately crucial for differentiation of nonepithelial laryngeal neoplasms. Online supplemental material is available for this article. © RSNA, 2020.

The goal of surgical treatment of laryngeal cancer is to achieve tumor control while preserving, whenever possible, the three primary functions of the larynx: breathing, swallowing, and phonation. The surgical procedure may consist of either a partial, conservative excision (eg, cordectomy, vertical partial laryngectomy, horizontal supraglottic laryngectomy, supracricoid laryngectomy with cricohyoidopexy or cricohyoidoepiglottopexy, or near total laryngectomy) or a radical excision (total laryngectomy). The procedure depends largely on the location and extension of the tumor, the stage of disease, and the patient's needs and preferences. Familiarity with the typical imaging appearance of the larynx after each procedure is crucial for differentiating normal postsurgical changes from persistent or recurrent disease as well as for diagnosing associated second primary malignancies. Since computed tomography (CT) is often used for follow-up evaluations, an ability to interpret the characteristic CT features is particularly important., (Copyright RSNA, 2008.)

Purpose: To evaluate different contrast material volumes, flow rates, and start delays for contrast material enhancement of neck structures and squamous cell carcinoma to determine the most effective examination protocol., Materials and Methods: Seventy patients with squamous cell carcinoma were prospectively randomized into four groups for examination with different protocols (125 mL of contrast material administered at a flow rate of 2.5 mL/sec, 100 mL at 2.0 mL/sec, 90 mL at 1.5 mL/sec, or 70 mL at 1.0 mL/sec). Dynamic series were performed on the tumors and relevant anatomic structures to obtain time-attenuation curves. The protocols were compared (analysis of variance and Tukey-Kramer tests) with regard to time and level of maximum tumor enhancement and carotid arterial enhancement of more than 150 HU. One selected protocol was tested in 30 additional routine examinations with start delays of 40 seconds (for laryngeal and/or hypopharyngeal tumors, 3-mm collimation) and 45 seconds (for oropharyngeal tumors, 5-mm collimation)., Results: Except for the 70-mL bolus administered at 1.0 mL/sec, the other protocols performed similarly well, yielding comparable maximum tumor enhancement at 52 seconds and later. In spite of a smaller volume of 90 mL, due to the prolonged flow time at 1.5 mL/sec, carotid arterial enhancement of more than 150 HU was prolonged (when compared with that in 100- or 125-mL protocols). As a result of these circumstances, injection of 90 mL at 1.5 mL/sec was considered more effective, providing no significant differences in tumor (P =.39) or carotid arterial (P =.52) enhancement between routine examinations and dynamic series., Conclusion: A single bolus of 90 mL administered at 1.5 mL/sec appears to be the most desirable protocol for contrast enhancement.

Purpose: To determine if follow-up computed tomography (CT) after definitive radiation therapy for laryngeal or hypopharyngeal (laryngopharyngeal) carcinoma allows the detection of local failure earlier than clinical examination alone., Materials and Methods: Pre- and post-radiation therapy follow-up CT scans in 66 patients were reviewed retrospectively. All patients underwent definitive hyperfractionated radiation therapy and were followed up clinically for at least 2 years after its completion. Post-radiation therapy CT scans (N = 153) were evaluated for posttreatment changes with a three-point score: A score of 1 represented expected posttreatment changes; 2, focal mass with a maximal diameter of less than 1 cm and/or asymmetric obliteration of laryngeal tissue planes; or 3, focal mass with a maximal diameter equal to or greater than 1 cm or estimated tumor volume reduction of less than 50%. All patients underwent the first posttreatment CT study 1-6 months after therapy. New or progressive laryngeal cartilage changes were noted. The clinical impression of the larynx at the time of each follow-up CT scan was also recorded., Results: In 12 of 29 (41%) patients with treatment failure at the primary site, follow-up CT scans were definite for local failure (score, 3) a mean of 5.5 months (median, 3.5 months; range, 1-17 months) before clinical examination results., Conclusion: In many patients, follow-up CT shows local failure earlier than does clinical examination alone.

Purpose: To evaluate the effectiveness of 2-[fluorine 18]fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) in the identification of early stage (T1-T2) primary and recurrent laryngeal cancer., Materials and Methods: Twelve patients with T1 or T2 laryngeal cancer underwent imaging prospectively with PET. Seven patients had new disease, and five had recurrent disease. All patients underwent imaging prior to planned therapy and tissue biopsy. PET images were evaluated by using standardized uptake ratios and visual analysis., Results: Histopathologic evidence of early stage cancer was documented in the 12 patients. One had a carcinoma in situ, nine had T1 tumors, and two had T2 tumors. Of the 12 patients, 10 had vocal cord tumors, one had a hypopharyngeal tumor, and one had a preepiglottic tumor. Eleven (92%) patients with early stage cancer had standardized uptake ratios indicative of malignancy (mean, 4.6; SD, 1.8; 95% CI, 1.2; range, 2.8-7.6). One had false-negative results (standardized uptake ratio = 2.3). Nine underwent CT, and results in the larynx were normal in seven and abnormal in two., Conclusion: FDG PET can be used to identify primary and recurrent early stage laryngeal cancer. It may be useful for follow-up after therapy.

A variety of benign and malignant non-squamous cell neoplasms may affect the larynx. Most of these uncommon laryngeal neoplasms are located beneath an intact mucosa, making diagnosis difficult with endoscopy alone, and sampling errors may occur if only traditional superficial biopsies are performed. In some laryngeal neoplasms, radiologic evaluation allows the correct diagnosis. Hemangiomas have very high signal intensity at T2-weighted magnetic resonance (MR) imaging and strong enhancement at both computed tomography (CT) and MR imaging after administration of contrast material. Phleboliths, which are pathognomonic for hemangiomas, are easily identified at CT. Chondrogenic tumors typically manifest with coarse or stippled calcifications at CT. Because of their high water content, chondrogenic tumors have very high signal intensity on T2-weighted MR images, whereas only moderate enhancement is observed after administration of contrast material. Lipomas typically manifest at both CT and MR imaging as homogeneous nonenhancing lesions. They are isoattenuating to subcutaneous fat at CT and isointense relative to subcutaneous fat with all MR pulse sequences. Metastases from renal adenocarcinoma typically demonstrate strong contrast enhancement and flow voids at MR imaging, and metastases from melanotic melanoma usually have high signal intensity on T1-weighted MR images and low signal intensity on T2-weighted images owing to the paramagnetic properties of melanin. Although radiologic findings are nonspecific in most other non-squamous cell neoplasms of the larynx (eg, Kaposi sarcoma, hematopoietic tumors, tumors of the minor salivary glands, metastases from amelanotic melanoma), cross-sectional imaging can play an important role in the diagnostic work-up of these unusual tumors by delineating the extent of submucosal tumor spread and directing the endoscopist to the appropriate site for the deep, transmucosal biopsies needed to establish the diagnosis. In addition, CT and MR imaging are crucial for posttherapeutic monitoring and early detection of local recurrence.

Purpose: To evaluate eight different diagnostic criteria to help detect neoplastic invasion of laryngeal cartilage at computed tomography (CT)., Materials and Methods: In a prospective series, 111 patients with carcinoma of the larynx or hypopharynx underwent thin-section, contrast material-enhanced CT before total or partial laryngectomy. The following CT criteria were evaluated: extralaryngeal tumor, sclerosis, tumor adjacent to nonossified cartilage, serpiginous contour, erosion or lysis, obliteration of marrow space, cartilaginous blowout, and bowing. Histologic findings were correlated with findings on CT scans obtained at each level., Results: Histologically, 122 of 412 cartilages were invaded. Depending on the diagnostic criteria and each specific cartilage, there was great variation in sensitivity (7%-83%) and specificity (40%-100%). Sclerosis was the most sensitive criteria in all cartilages but often corresponded to reactive inflammation in the thyroid cartilage. Extralaryngeal tumor and erosion or lysis yielded the highest specificity. Tumor adjacent to nonossified cartilage, serpiginous contour, and obliteration of marrow space were specific although not sensitive signs of invasion in the arytenoid and cricoid cartilage and were nonspecific in the thyroid cartilage. Blowout and bowing were not useful. Selection of the appropriate combination of criteria enabled an overall sensitivity of 91% (associated specificity, 68%) or an overall specificity of 79% (associated sensitivity, 82%)., Conclusion: Detection of neoplastic cartilage invasion with CT greatly depended on the appropriate use of individual and combined CT criteria.

Purpose: To compare the usefulness of computed tomography (CT) and gadolinium-enhanced magnetic resonance (MR) imaging in the detection of neoplastic invasion of laryngeal cartilage., Materials and Methods: In a prospective study, 53 patients with carcinoma of the larynx or piriform sinus underwent CT and MR imaging before total or partial laryngectomy. The findings at imaging and pathologic examination were compared., Results: At histologic examination, neoplastic invasion of cartilage was present in 34 patients and absent in 19. MR imaging was more sensitive than CT (89% vs 66%; P = .001). Inflammatory changes and fibrosis, however, were indistinguishable from tumor on MR images, resulting in overestimation of neoplastic invasion in a large number of patients. Therefore, MR imaging was less specific than CT (84% vs 94%; P = .004)., Conclusion: MR imaging is more sensitive than CT in detecting neoplastic invasion of cartilage, but the inability to differentiate between nonneoplastic inflammatory changes and tumor with MR imaging leads to overestimation of neoplastic invasion.

Purpose: To evaluate the computed tomographic (CT) appearance of laryngeal tumors treated with radiation therapy and the ability of CT to depict persistent or residual tumor., Materials and Methods: Sixty-one patients with primary squamous cell carcinoma of the larynx or hypopharynx were treated with definitive radiation therapy. CT was performed in all patients before and after treatment., Results: In 32 of 41 patients with cancer controlled at the primary site, CT showed complete resolution of tumor, whereas in 10 of 14 patients in whom radiation therapy failed, there was minimal or no reduction in tumor. In a subpopulation of patients who underwent repeat imaging, 18 of 19 with tumor controlled at the primary site had complete resolution of tumor. Overall, in four of 13 patients with 50%-75% reduction in tumor size or persistent substantial asymmetry at CT, therapy eventually failed at the primary site., Conclusion: Lesions that are reduced by 50% or less at 4-month follow-up CT are highly suspicious for treatment failure. Repeat CT studies every 4 months is recommended in addition to careful clinical follow-up.

Purpose: To present the expected appearance of the irradiated larynx and neck as seen at computed tomography (CT)., Materials and Methods: Sixty-one patients with primary squamous cell carcinoma of the larynx or hypopharynx were treated with radiation therapy. All patients underwent CT before and after treatment., Results: Expected changes include symmetric thickening of the epiglottis, aryepiglottic folds, and false cords and increased attenuation of the paralaryngeal fat. The posterior pharyngeal wall tends to thicken and its mucosa enhances; retropharyngeal space edema is common. Glottic changes include increased attenuation of the paraglottic fat planes and thickening of the anterior and posterior commissures. Subglottic changes include thickening of the mucosa and submucosa. Soft-tissue changes include skin and platysmal thickening, as well as reticulation and increased attenuation of the subcutaneous and deeper fat., Conclusion: Familiarization with expected radiologic changes is essential for interpretation of CT images of the irradiated larynx so that such changes are not mistaken for signs of persistent or recurrent tumor.

Purpose: To determine whether sclerosis of the laryngeal cartilages was a predictor of a poor outcome in patients with stage T3 glottic cancer treated with radiation therapy., Materials and Methods: Thirty-three patients with stage T3 glottic cancer underwent computed tomography (CT) before radiation therapy. Twenty-two patients underwent posttreatment CT. The presence of cartilage sclerosis, cartilage erosion, marrow invasion, and cartilage necrosis was determined., Results: Nineteen of the 33 patients had cartilage sclerosis at CT. Seventeen patients had sclerosis of a single laryngeal cartilage (14 arytenoid, two cricoid, and one thyroid), and two had sclerosis of adjacent laryngeal cartilages (arytenoid and cricoid in both cases). Of the 17 patients with isolated laryngeal cartilage sclerosis, disease was controlled with radiation therapy alone in 15 and with salvage laryngectomy in two. Both patients with cricoid and arytenoid sclerosis died of their original cancer despite undergoing early salvage laryngectomy. Of the 14 patients without sclerosis, eight had no evidence of disease, two died of their disease, and four died of intercurrent disease., Conclusion: T3 glottic cancer with isolated laryngeal cartilage sclerosis can be cured with radiation therapy.

Purpose: To assess the significance of sclerotic-appearing cricoid and arytenoid cartilage with computed tomography (CT) in patients with laryngeal carcinoma., Materials and Methods: CT scans obtained in 75 patients with laryngeal carcinoma were prospectively studied; laryngeal CT studies obtained in 50 patients without laryngeal carcinoma were retrospectively reviewed., Results: Twenty-four of the 75 patients (32%) with laryngeal carcinoma had sclerotic cartilage. Evaluation of pathologic specimens obtained in 12 of these 24 patients showed 11 cases of sclerotic arytenoid cartilage and two cases of sclerotic cricoid cartilage. Tumor infiltration was demonstrated in six of these cases but not in the seven others. In 11 of the 12 cases with pathologic proof, however, tumor was adjacent to the perichondrium. In the 12 cases without pathologic proof, the proportion was similar. The positive predictive value of this sign for cartilaginous invasion was 46%., Conclusion: Although it is not a reliable sign of cartilaginous invasion, sclerotic-appearing cricoid and arytenoid cartilage in patients with laryngeal carcinoma is predictive of the tumor to this cartilage.

The aim of this study was to define and quantitate the normal anatomy of the extracranial head and neck with 2-[fluorine-18]fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET). This information was used to study 12 patients with primary squamous cell carcinomas. In all cases, the lymphoid tissue of the Waldeyer ring and the palatine and lingual tonsils could be differentiated from the airway, striated muscle, osseous structures, and salivary glands. Striated muscle had markedly less activity than lymphoid or salivary gland tissue. In the 12 patients with primary tumors, FDG PET depicted the tumor as an area of increased activity significantly higher than that of normal tissue. In one instance, FDG PET allowed detection of a tumor not seen at magnetic resonance (MR) imaging or computed tomography. Of the 34 lymph nodes positive for carcinoma, 24 were positive according to MR size criteria and 25 were detected with FDG PET. FDG PET allowed detection of three nonenlarged metastatic nodes that were negative at MR imaging.

To estimate the accuracy of different radiologic criteria used to detect cervical lymph node metastasis in patients with head and neck carcinoma, seven different characteristics of 2,719 lymph nodes in 71 neck dissection specimens from 55 patients were assessed. Three lymph node diameters, their location, their number, the presence of a tumor, and the amount of necrosis and fatty metaplasia were recorded. The minimal diameter in the axial plane was found to be the most accurate size criterion for predicting lymph node metastasis. A minimal axial diameter of 10 mm was determined to be the most effective size criterion. The size criterion for lymph nodes in the subdigastric region was 1 mm larger (11 mm). Groups of three or more borderline nodes were proved to increase the sensitivity but did not significantly decrease the specificity. Radiologically detectable necrosis (3 mm or larger) was found only in tumorous nodes and was present in 74% of the positive neck dissection specimens. Shape was not a valuable criterion for the radiologic assessment of the cervical lymph node status.

Computed tomographic scans in 23 patients who had undergone total laryngectomy were analyzed retrospectively to determine normal postoperative appearance and to evaluate the role of CT in assessing recurrent neoplasm. Nine patients without clinical evidence of recurrence illustrated the normal postoperative changes: a round or ovoid neopharynx connecting the base of the tongue with the cervical esophagus and intact fat planes surrounding the neopharynx, neurovascular bundles, and sternocleidomastoid muscles. In the 12 patients with recurrent neoplasm, the CT manifestations included masses involving the internal jugular lymph node chain (adjacent to the neopharynx, neurovascular bundles, or sternocleidomastoid muscles), tracheostomy site, or paratracheal region. Recurrence was mimicked on CT in two patients, one with an abscess and one with metastases from an adenocarcinoma of unknown primary site. In eight patients, a distended neopharyngeal lumen correlated with benign or malignant stricture. CT supplemented physical examination and indirect mirror examination, providing data regarding presence and extent of recurrent tumor and aiding in planning the mode and scope of therapy.

Twenty consecutive patients underwent computed tomography (CT) prior to total laryngectomy in order to assess the accuracy of CT scanning in the evaluation of carcinoma of the larynx and hypopharynx. CT scans were prospectively interpreted and the extent of tumor was recorded graphically at five major levels of the larynx. The extent of tumor was similarly evaluated on whole-mount histologic sections of the laryngeal specimens prepared in the horizontal plane similar to the CT scans. The results were compared to assess the ability of CT to identify tumor extent accurately. Despite good correlation of gross tumor extent between histologic specimens and CT scanning, specific pitfalls in CT diagnosis were identified. Overestimation of tumor extent was caused by edematous changes in six patients and tumor-associated inflammatory changes in three patients. In seven patients, mass effect from adjacent bulky tumor significantly distorted normal structures, mimicking tumor involvement. Small foci of mucosal tumor in three patients and microscopic cartilage involvement in two patients were not identified on CT scanning, causing underestimation of tumor extent.

A prospective study of 66 patients was conducted to evaluate the relative merits of computed tomography (CT) and laryngography in the assessment of laryngeal disorders. CT complemented direct laryngoscopy and biopsy for treatment planning and was superior to both modalities for showing deep infiltration, invasion of cartilage, and extension to the soft tissues of the neck. The subglottic space can be studied adequately by CT, which could also be useful in diagnosis of lymph node metastases. Distortion of the thyroid cartilage can be shown by CT. CT was equal to or better than laryngography in approximately 90% of cases.

Nineteen patients with laryngeal lesions each underwent diagnostic studies consisting of laryngoscopy and computed tomography (CT). Sixteen patients had linear tomography and coronal and sagittal reconstruction of their transaxial CT images. The results of each examination were compared with the surgical and pathological findings to assess their relative value. The results confirmed previous reports of the accuracy and value of transaxial CT and direct laryngoscopy, but indicated that coronal and sagittal reconstruction as well as linear tomography added no significant information.

Supraglottic subtotal laryngectomy (SSL) is a radical, yet voice-conserving, surgical procedure commonly performed for carcinoma of the supraglottic larynx. The pharyngograms and computed tomographic (CT) scans of 35 patients obtained after SSL were evaluated retrospectively. These examinations reliably demonstrated the changes in anatomy caused by removal of the epiglottis, aryepiglottic folds, and false vocal cords. Fourteen patients had documented recurrence of cancer; five mucosal, nine extramucosal. Three of five macroscopic mucosal recurrences in the larynx/pharynx were detected on the barium pharyngograms; the two mucosal lesions not seen were in the base of the tongue and tonsillar fossa. CT enabled detection of five of five recurrences and was superior to pharyngography in demonstrating the soft-tissue extent of disease. CT findings mimicking recurrence were seen in two patients: one with diffuse histiocytic lymphoma; the second, with benign hyperkeratosis. Barium and CT examinations are useful adjuncts to the clinical examination in detecting recurrent squamous cell carcinoma in patients following SSL.

The authors studied 162 patients with advanced cancer of the laryngopharynx who were treated over the same period of time by standard techniques. High dose preoperative supervoltage irradiation to 5000 rads/5 weeks through large opposed lateral fields and a lower anterior neck field is not accompanied by a prohibitive increase in mortality or morbidity for patients with stage III-IV largynogopharyngeal cancer. Rescue surgery for recurrence after definitive irradiation can be accomplished.

Nine pyriform sinus tumors and 16 supraglottic laryngeal lesions were studied by computed tomography. The pyriform sinus tumors had a much higher incidence of thyroid cartilage invasion (five of nine patients) and involved the posterolateral margins of the cartilage. Only two of 16 supraglottic tumors reached the thyroid cartilage, involving the midline or more inferior margins. The supraglottic tumors grew in a circumferential pattern, and when the pre-epiglottic space was involved, extension was bilateral. Lesions of the pyriform sinus more frequently showed unilateral involvement. Lymph node metastases were seen in an approximately equal proportion of patients in the two groups, and they correlated more with the size of the primary tumor than with its site of origin. Posteroinferior invasion of the space between the thyroid and cricoid cartilage was seen only with pyriform sinus lesions.

Seven patients with known carcinoma of the larynx underwent computed tomography (CT) of the larynx prior to surgery. Whole-mount sections of the extirpated larynx cut in the horizontal plane were compared with the corresponding level of the preoperative CT sections to demonstrate the validity of CT scanning in the evaluation of tumors of the larynx. The results indicate that CT scanning accurately demonstrates the anatomic location and gross size of laryngeal tumor, although early invasion of the laryngeal cartilages may be difficult to diagnose with CT. It is concluded that preoperative CT scanning of the larynx is the radiologic procedure of choice for evaluating carcinoma of the larynx.

High-resolution 2-mm thick sequential CT scans of 23 patients with primary laryngeal carcinoma were reviewed to judge the value of CT in early tumor detection and the ability of CT to establish tumor extent reliably, particularly early cartilage invasion. False-negative results were likely when superficial lesions were confined to the true vocal cord mucosa. False-positive examinations occurred with vocal cord polyps and with acute and chronic granulomatous reaction of the vocal cords. In tumors large enough to be imaged, there were no false negatives. False-positive results for clearly identified tumors were related to the interpretation of tumor on CT where only edema was present at surgery. Both early and gross cartilage invasion were well correlated with surgically resected specimen sections. Early cartilage involvement by tumor can be detected by the development of a fenestrated chondral margin. Other benign expansive cartilage processes, such as occur with chondromas, have calcified cartilage fragment distributions that are out of proportion to any soft-tissue mass present.

Computed tomography (CT) offers a unique method of displaying the anatomy of laryngeal cancer. Asymmetrical contours of the thyroid cartilages have been found as a normal variant. Marked asymmetry may indicate buckling of the cartilage which could contribute to acute airway obstruction following radiation therapy. A low-density, soft-tissue plane medial to the thyroid cartilages may prove valuable in evaluating cartilaginous involvement with tumor. Computed tomography should make a significant contribution to more accurate planning of radiation therapy portals by showing the relationship of lymphatic spread to the primary tumors.

A review of 100 consecutive laryngograms showed visualization of the appendix of the laryngeal ventricle in 40% of cases. This high incidence of visulization is related to thorough anesthesia, repeated performance of phonation and the reverse "E" maneuver as well as awareness of the anatomical structure. The appendix of the laryngeal ventricle courses superiorly between the laryngeal vestibule and the thyroid cartilage which differentiates this normal structure from ulcerations and fistulous tracts of laryngeal tumors.

Not every laryngeal abnormality requires imaging. If a lesion is small and there is no evidence of deep extension, the clinician can derive all visual examination. Similarly, when a lesion is large and obviously involves both true and false cords (transglottic), the clinician already knows that a total laryngectomy is necessary and imaging is of little value unless there is a question about nodal involvement. The real value of imaging is in the questionable cases in which a decision must be made about the feasibility of speech conservation therapy. The radiologist must understand the various conservation techniques and the key information needed to determine the feasibility of each. In cases in which patients are able to cooperate with the examination, MR with multiplanar imaging capabilities and increased tissue differentiation has, in my view, an edge over CT. However, many patients cannot cooperate, and in these cases CT provides a more consistently good examination. Some centers rely totally on CT and some rely completely on MR imaging units with low or middle field strength. Either can give excellent laryngeal imaging. At my institution, with a high-field-strength unit, MR imaging is the first choice if the patient is fairly cooperative and is thought able to undergo the examination. Even if the patient cannot tolerate the entire protocol, the examination usually provides enough necessary information. If the patient has major problems with secretions or has difficulty cooperating, we do not try MR imaging but use CT; the examination is almost always adequate even though restricted to one plane. The imaging modality used is less important than the radiologist's knowledge of the key anatomic landmarks. The most important landmarks, from a surgeon's perspective, are the ventricle, anterior commissure, and the cricoid cartilage. The structures most helpful in identifying the position of the ventricle are the paraglottic fat, the thyroarytenoid muscle, and the arytenoid cartilage. The clinician remains responsible for evaluating the mucosal surface. The goal of the radiologic examination is to find deep tumoral extension that the clinician cannot see.

Vocal cord paralysis is a relatively common entity, usually resulting from a pathologic process of the vagus nerve or its recurrent laryngeal branch. It is rarely caused by intralaryngeal lesions. Fourteen patients with recurrent laryngeal nerve paralysis (RLNP) were evaluated by laryngography, computed tomography (CT), or both. In the evaluation of the paramedian cord, CT was limited in its ability to differentiate between tumor or RLNP as the cause of the fixed cord, but it yielded more information than laryngography on the structural abnormalities of the larynx and pre-epiglottic and paralaryngeal spaces. Laryngography revealed distinct features of RLNP and is the procedure of choice for evaluation of functional abnormalities of the larynx until further experience with faster CT scanners and dynamic scanning of the larynx is gained.

Computed tomographic scans in 22 postoperative vertical hemilaryngectomy patients were analyzed retrospectively to determine the normal postoperative appearance and to evaluate the role of CT in assessing recurrent neoplasm. Twelve patients without clinical evidence of recurrence illustrated the normal postoperative changes: loss of the preepiglottic space fat and the aryepiglottic fold on the operated side, convexity of the residual true vocal cord, and regeneration of the resected thyroid cartilage ala. In the six patients with recurrent neoplasm, the CT manifestations included increased width of the remaining true vocal cord, convexity of the surgically formed pseudocord at glottic level, subglottic tumor, and extralaryngeal neck masses. Recurrence was mimicked in four patients by bulky soft tissue at the endolaryngeal operative site at both CT and laryngoscopy; this was due to hyperkeratosis or granulation tissue in three cases and an inclusion cyst in one case. CT supplemented the physical examination and indirect laryngoscopy, providing information regarding the presence and extent of tumor that was useful in planning the mode or scope of subsequent therapy.