Three hundred of the estimated 800 lymph nodes in the body are situated in the neck(1, 2). For nearly four decades, the most commonly used classification of the cervical lymph nodes was that of Rouvière developed in 1938(1). His work, and that prior to his, was based on data derived from detailed anatomic studies which defined both the location and drainage patterns of the cervical lymph nodes(1-3).
A drawback of Rouvière's classification was that the anatomic descriptions were not easily correlated with surgical landmarks during a neck dissection. In an attempt to overcome these difficulties and establish reproducible criteria, a group of surgeons suggested that the anatomically based terminology be replaced with a simpler classification based on levels(4). Since then, a number of classifications have been proposed that categorize the lymph nodes into "levels", "groups", or "zones"(5-21). The purpose of these newer classifications was not to change terminology, but to assemble the cervical nodes on the basis of the clinical and pathophysiologic information obtained in the intervening 60 years.
Pre-treatment imaging has gained acceptance for evaluating the cervical lymph nodes for metastases in patients with various head and neck neoplasms. At least 80% of patients with head and neck cancer now undergo CT or MR imaging before treatment. The anatomic criteria described by Rouvière and those who followed are often difficult to directly translate to cross-sectional images. As a result, this has resulted in a lack of consistency when assessing the nodal groups(14).