I. Analysis

Lindberg analyzed patients with "clinical evidence" of cervical node metastases(19, 25). A "clinically positive" neck was defined by Lindberg as a node "usually greater than 1cm, spherical rather than flat ovoid, and harder than the nonmetastatic lymph node"(19, 25). The percent involvement of the nodal subsites used in this manuscript was derived from Lindberg's data using the following equation for each anatomic subsite:

# of patients with clinically positive nodes within each nodal group
total # of patients with clinically positive necks for each anatomic subsite

Lindberg did not report the distribution of cervical lymph nodes metastases from true vocal cord carcinoma. Byers, et al., reported the relative percentage of node involvement by nodal subsite in the N0 neck following elective node dissection for true vocal cord carcinoma. Thus, the information regarding the relative percentages of lymph node involvement from true vocal cord carcinoma presented in this manuscript was directly obtained by Byers et al(24, 25). Only the ipsilateral neck was evaluated since this was the only side reported(24, 25).