The role of ultrasonography in the management of tumors of the neck.
By themselves, clinical examination and palpation do not provide an accurate assessment of metastatic neck nodes. They do not yield sufficient information to ascertain the benign or malignant nature of nodes or to determine the presence of extracapsular spread and vascular invasion. The use of real-time ultrasonography with high-frequency transducers can significantly improve the evaluation of patients with neck masses. We studied the use of ultrasound in evaluating metastatic neck disease in 25 patients. We found that it is useful not only in detecting neck nodes, but in assessing their characteristics and the degree of vascular invasion. We recommend that ultrasonography be routinely performed as part of the evaluation of all patients with head and neck masses. It is also valuable in the postoperative or postirradiation followup of patients whose necks are otherwise difficult to evaluate.
Head and neck cancers constitute approximately 10% of all malignancies.  Depending on their site and histology, 20 to 90% of these tumors will progress to nodal metastasis.  Palpation, which has long been the mainstay of the workup of patients with metastatic neck nodes, is an inaccurate and inadequate method of evaluating nodal status. Palpation can underestimate the size of a tumor and provide insufficient information regarding extracapsular spread and vascular involvement.
Cross-sectional imaging has improved our ability to detect abnormal nodes.  With a sensitivity of 90% or more, ultrasonography is an excellent method of assessing the nodal status of the neck.  Bruneton et al reported that the sensitivity of sonography was 93%, compared with only 78% for palpation.  Hajek et al described its utility in differentiating single from multiple lymph nodes and in detecting nodes smaller than 12 and 15 mm in superficial and deep locations, respectively.  Vassallo et al reported that ultrasound is useful in differentiating malignant from benign lymph nodes.  They found that a ratio of longitudinal diameter to transverse diameter (LT ratio) of less than 2-to-1 in the absence of a hilum is an indication of malignant lymphadenopathy. Its ability to detect unsuspected metastases has been reported to result in a change in clinical management for as many as 29% of patients. 
Gluckman et al suggested that sonography is an important adjunct in the evaluation of difficult-to-examine necks, such as those in patients who are obese and in those who have radiation-induced fibrosis or pedicle grafts with edema.  Others have documented that (1) vascular involvement by metastatic cervical nodes has a strong bearing on treatment and prognosis, (2) clinical examination is inadequate to assess vascular invasion, and (3) sonography is useful in assessing it. [5-8]
The objective of our study was to evaluate the role of ultrasound in metastatic neck disease and to compare it with clinical examination in terms of relative accuracy in determining the nodal status of the neck, while keeping histopathologic examination as the gold standard.
Materials and methods
We studied 25 patients with neck malignancy who were scheduled for surgery at our institution. We performed a routine general examination and a detailed otolaryngologic examination on each patient. As part of the latter, we examined the entire neck, particularly the primary tumor sites, and we noted the number, size, and mobility of the nodes. We classified nodal levels according to the system suggested by Som. 
Sonography of the neck was performed with a B-mode scanner and 5-MHz probe. A 7.5-MHz probe was used to enhance any images that lacked clarity. The radiologist was blinded to the results of the clinical findings. The neck was scanned along its longitudinal axis from the midline anteriorly to the anterior border of the trapezius posteriorly. A transverse scan was obtained from the ramus of the mandible to the clavicle.
Once the images were obtained, the radiologist noted the size of the lymph nodes, their longitudinal and transverse diameters, the presence or absence of a hilum, and any extracapsular spread. The radiologist assessed the nodes for vascular invasion by examining the wall of the vessel along its longitudinal and transverse axis. Breakage of the wall in two planes and the presence of a mass clearly invading the lumen were taken as signs of vascular invasion.
Following the excision of the primary tumor, the masses were transported to the pathology laboratory in formalin. The pathologist sectioned the masses into 5-[micro]m slices, stained them with hematoxylin and eosin, and studied them in detail.
We then compared the findings of our clinical examination with those of ultrasound to determine their relative accuracy in detecting neck nodes and their characteristics. We used the histopathology results as a gold standard.
The two most common primary carcinomas in this series were those of the larynx/laryngopharynx (68%) and the tongue (16%).
Tumor staging. Analysis of ultrasound images led to a change in nodal staging for 22 of the 25 patients (88%) (table). Twenty-one patients had their nodal stage revised upward; the other was downgraded from N1 to N0 when the nodal involvement was found to be a direct extension of the primary tumor rather than a metastasis.
Vascular invasion. Sonography predicted vascular invasion in seven patients who did not have any clinical features suggestive of such, and histopathology confirmed that it was accurate in four of these cases. Two patients had internal jugular vein involvement and two had carotid invasion (figures 1, 2). There were two false positives, as sonography wrongly predicted internal jugular vein invasion in two patients. The remaining patient was not operated on because the tumor was so extensive.
Postradiotherapy assessment. Nine patients in this series received preoperative radiotherapy. Following irradiation, the cervical nodes of four of these patients were not palpable. Ultrasonography was able to detect the cervical nodes in these four patients, and these findings were corroborated during surgery.
Treatment protocol. Sonographic findings led to a change in the treatment protocol for 14 of the 25 patients (56%). Eleven of these patients were treated with more radical surgery than they otherwise would have been, and five of them benefitted from it (the other six patients did not benefit). Three patients were administered radiation therapy after sonography detected extensive tumor.
We based our assessment of the accuracy of the clinical examination and ultrasonography on the results of the histopathologic examination. Analysis of data subjected to Wilcoxon's signed rank test confirmed the superiority of ultrasonography over clinical examination (p = 0.0005). As expected, we also found that ultrasonography is not superior to histopathology.
Secondary findings. We also observed noteworthy findings regarding the relationship between malignancy and the size of a node, the LT ratio, and the nodal hilum. Nodes that were smaller than 10 mm were more likely to be benign on histology than larger nodes; the chance that a node was malignant increased in proportion to its size (p[less than]0.05). The likelihood of malignancy also increased as the LT ratio decreased (p[less than]0.05). Finally, the demonstration of an intact hilum on sonography was seen infrequently in malignant nodes (22%). However, the absence of a hilum was seen in both malignant and benign nodes in nearly equal proportions--52 and 48%, respectively (p = NS).
Ultrasonography is a simple, readily available, and inexpensive investigation that carries no risk of complications. With a 5- or 7.5-MHz transducer, it has the ability to detect pathologic nodes larger than 5 mm with a sensitivity of 90 to 95%. 
The dynamic relationship of lymph node metastasis to surrounding vascular structures is important in planning surgery. [6,11] In our study, preoperative sonography led us to suspect that seven patients had vascular involvement, and postsurgical histopathology confirmed that four actually did. We found that restricted mobility on sonopalpation and a loss of normal echogenic vascular interface were important signs of malignant infiltration.
Before the advent of imaging, extracapsular spread was thought to occur only in nodes larger than 3 cm.  We detected extracapsular spread in the nodes of seven patients, and six of the nodes were smaller than 3 cm.
Four of our patients underwent ultrasound-guided fineneedle aspiration cytology. This procedure has a high degree of sensitivity and specificity, and it often leads to a change in staging and management. The major advantage of this procedure is its ability to biopsy surgically inaccessible nodes.
Secondary findings revealed that the size of a node is directly proportional to its likelihood of being malignant. However, other studies have not found such a clear-cut association.  Also, the LT ratio is known to be a reliable indicator in differentiating benign and malignant nodes. In our study, approximately 80% of nodes that had an LT ratio greater than 2-to- 1 were benign. Conversely, an LT ratio less than 1.4-to-1 indicated a strong possibility of malignancy. Various studies have reported similar findings. [5,13,14] Finally, it has been suggested that the presence of a well-defined hilum is an indication of a benign lymph node, whereas a thin or absent hilum with cortical widening suggests malignancy.  Our findings in this regard were equivocal.
It is still not possible to rule out malignant transformation in a lymph node with certainty. The presence of echogenic foci, an LT ratio of less than 1.4, and the absence of a hilum provide evidence of nodal involvement by a tumor. This is of critical clinical importance because a properly planned surgical excision can provide a lasting cure.
Sonography has an additional advantage in its ability to assess postradiotherapy neck nodes that are not palpable on clinical examination. We did not find in the literature a similar observation because of the paucity of studies on the postradiotherapy evaluation of neck nodes by ultrasonography.
From the Department of Otolaryngology (Dr. Sajeeda, Dr. Panda, and Dr. Mann) and the Department of Radiology (Dr. Katariya and Dr. Kalagara), Postgraduate Institute of Medical Education and Research, Chandigarh, India.
(1.) Baatenburg de Jong RJ, Rongen RJ, Verwoerd CD, et al. Ultrasound-guided fine-needle aspiration biopsy of neck nodes. Arch Otolaryngol Head Neck Surg 1991;117:402-4.
(2.) Bruneton JN, Roux P, Caramella E, et al. Ear, nose, and throat cancer: Ultrasound diagnosis of metastasis to cervical lymph nodes. Radiology 1984;152:771-3.
(3.) Gluckman JL, Mann W, Gapany M, Welkoborsky HJ. The role of realtime ultrasonography in an otolaryngology office setting. Laryngoscope 1993;103:570-4.
(4.) Hajek PC, Salomnowitz E, Turk R, et al. Lymph nodes of the neck: Evaluation with US. Radiology 1986;158:739-42.
(5.) Vassallo P, Wemecke K, Roos N, Peters PE. Differentiation of benign from malignant superficial lymphadenopashy: The role of high-resolution ultrasonography. Radiology 1992;183:215-20.
(6.) Kennedy JT, Krause CJ, Loevy S. The importance of tumor attachment to the carotid artery. Arch Otolaryngol 1977;103:70-3.
(7.) Mann WJ, Beck A, Schreiber J, et al. Ultrasonography for evaluation of the carotid artery in head and neck cancer. Laryngoscope 1994;104:885-8.
(8.) Rothstein SG, Persky MS, Horii S. Evaluation of malignant invasion of the carotid artery by CT scan and ultrasound. Laryngoscope 1988;98:321-4.
(9.) Som PM. Lymph nodes of the neck. Radiology 1987;165:593-600.
(10.) Friedman M, Mafee MF, Pacella BL Jr., et al. Rationale for elective neck dissection in 1990. Laryngoscope 1990;100:54-9.
(11.) Yonetsu K, Ikemura K. Ultrasonographic study of the relation of meta static nodes to the carotid artery. Head Neck Surg 1987;9:279-83.
(12.) Savoury WL, Gluckman JL. Cervical metastasis. In: Paparella MM, ed. Otolaryngology. Vol. 3. Philadelphia: W.B. Saunders, 1991:2570.
(13.) Bruneton JN, Balu-Maestro C, Marcy PY, et al. Very high frequency (13 MHz) ultrasonographic examination of the normal neck: Detection of normal lymph nodes and thyroid nodules. J Ultrasound Med 1994;13:87-90.
(14.) Eichhorn T, Schroeder HG. Ultrasound in metastatic neck disease. ORL J Otorhinolaryngol Relat Spec 1993;55:258-62.