Objectives The aims of this study were to predict pelvic lymph node metastasis in uterine cervical cancer before surgery and to evaluate the potential efficacy of omitting pelvic lymphadenectomy.

Materials and Methods A total of 163 patients with invasive uterine cervical cancer in FIGO stage IA2 to IIB, all of whom underwent primary radical hysterectomy with pelvic lymphadenectomy, participated in this study.

Results The incidences of pelvic lymph node metastasis in stage IA2, stage IB1, stage IB2, stage IIA, and stage IIB cervical cancer were 0% (0/12), 17% (13/76), 22% (6/27), 33% (8/24), and 63% (15/24), respectively. A significant difference was observed in overall survival with nodal metastasis status (P < 0.0001). Univariate analysis revealed that parametrial invasion (P < 0.0001), tumor markers (P = 0.0006), tumor size greater than 2 cm (P < 0.0001), tumor size less than 3 cm (P = 0.0009), and tumor size greater than 4 cm (P = 0.0024) were correlated with pelvic lymph node metastasis. However, multivariate analysis revealed that parametrial invasion (P = 0.01; odds ratio, 3.37; 95% confidence interval, 1.31–9.0) and tumor size greater than 2 cm (P = 0.005; odds ratio, 4.93; 95% confidence interval, 1.54–22.01) were independently associated with nodal metastasis.

Conclusions Pelvic lymphadenectomy may be avoided in patients with negative parametrial invasion and a tumor size less than 2 cm, thereby minimizing postoperative complications.