Correlation between tumor size and surveillance of lymph node metastasis for IB and IIA cervical cancer by magnetic resonance images

Abstract

Purpose: To assess the feasibility of preoperative MRI based measurement of tumor size with regard to lymph node (LN) metastasis in early uterine cervical cancer. Material and Methods: A retrospective review of patients with FIGO stage IB–IIA cervical cancer who underwent lymphadenectomy was performed. Diagnostic accuracy of MRI in detecting LN metastasis and rate of LN recurrence in terms of tumor size (≤4 cmversus >4 cm) were analyzed. ROC curve analysis was used to determine LN size for differentiating LN metastasis in terms of tumor size. P < 0.05 was considered statistically significant. Results: Of the 200 patients, 45 (22.3%) had LN metastasis. There was no statistical difference between patients-based and region-specific analysis. The patients with tumor size with >4cmrevealed higher diagnostic accuracy of MRI in detecting LN metastasis (85.4% versus 50.6%, P = 0.023) and rate of LN recurrence (20.0% versus 6.4%, P = 0.031) in than those with size with ≤4 cm, the differences were statistically significant. Discriminant analysis of LN size for the differentiation of metastasis from non-metastasis resulted in cut-off values (11.8mm; size with >4cm versus 8.3mm; size with ≤4 cm) and diagnostic accuracy (84.0% of size with >4cm versus 72.0% of size with ≤4 cm). Conclusion: MRI has limited sensitivity, but high specificity in predicting surveillance of LN metastasis in the preoperative early cervical cancer, especially useful tool for patients with tumor size with >4 cm.