Although tumor size is a prognostic factor in cervical cancer patients, its role in the diagnosis of lymph node metastasis is unclear. We therefore evaluated the diagnostic value of tumor and lymph node size compared with lymph node size alone in the detection of metastatic lymph nodes in patients with early-stage cervical cancer.
We retrospectively evaluated 699 patients with International Federation of Obstetrics and Gynecology stage IB1-IIA cervical carcinoma who underwent magnetic resonance imaging before lymphadenectomy involving all visible lymph nodes in the surgical fields. Seven nodal groups were evaluated: para-aortic, both common iliac, both external iliac, and both internal/obturator areas. Pathologic evaluation was the diagnostic standard. The largest short-axis diameter of lymph nodes in each region and the largest tumor diameters were measured in magnetic resonance images. The value of additional information from magnetic resonance images was evaluated by receiver operating characteristic curve analysis.
Of the 699 patients, 108 (15.8%) had lymph node metastases. The areas under the curve for measurements of lymph node size, tumor size, and both were (A) 0.635, (B) 0.706, and (C) 0.742, respectively (A vs B, P = 0.006; A vs C P < 0.001; B vs C, P = 0.002).
This study illustrates that magnetic resonance imaging measurements of tumor size and tumor size plus lymph node size showed a higher diagnostic performance than lymph node size alone in predicting lymph node metastasis in patients with early-stage cervical cancer.
From the Departments of *Radiology and †Radiation Oncology, Asan Medical Center, University of Ulsan, Seoul, Korea.
Received for publication August 10, 2012; accepted February 5, 2013.
Reprints: Hyuck Jae Choi, MD, Department of Radiology, Asan Medical Center, University of Ulsan, 388-1, Poongnap-dong, Songpa-gu, Seoul, 138-736, Korea (e-mail: email@example.com).
S.K. and Y.S.K. contributed equally to this article.
This study was supported by a grant (2010-056) from the Asan Institute of Life Sciences and a National Research Foundation of Korea (NRF) grant funded by the Korea Government (MEST) (2011-0001302).
The authors have no conflict of interest to declare.