Accuracy of MRI for diagnosing pelvic and para‑aortic lymph node metastasis in cervical cancer.

The current study aimed to evaluate the accuracy of diffusion‑weighted imaging and morphological aspects at 3 Tesla (T) and 1.5T MRI for diagnosing metastatic lymph nodes (LN) in cervical cancer. A retrospective study was conducted at the Barretos Cancer Hospital. A total of 45 patients with cervical cancer who underwent MRI examination and pelvic and/or para‑aortic lymphadenectomy as part of surgical procedure were included. Data regarding LN images included size (short‑axis diameters), morphology (usual, rounded or amorphous), appearance (homogeneous or heterogeneous), limits (regular, irregular or imprecise), presence or absence of necrosis, diffusion (normal or greater restriction than expected for normal tissue) and aspect (suspected, undetermined or normal). These findings were compared with histopathological results. According to histology results, among the 45 patients, 14 (31.1%) LNs were tested positive for metastasis and 31 (68.9%) LNs were tested negative. A total of 41 metastatic positive LNs were detected from a total of 976 resected nodes. Twelve patients from the 45 (26.7%) had LN classified as metastatic by histology and suspected by MRI, 26 (57.8%) as negative in both evaluations, 2 (4.4%) as positive by histology and negative by MRI and five (11.1%) as negative by histology and positive by MRI. Based on these results, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy were 85.7, 83.9, 70.6, 92.9 and 84.4%, respectively. The Cohen's κ test exposed a general outcome of 0.657 (P<0.05), demonstrating that the two variables (histology and MRI) have substantial concordance. The κ test results between histological and MRI data for paraaortic and pelvic LNs were found to be 1 and 0.657, respectively. Finally, short axis >10 mm, T2 hypointensity, rounded morphology and greater restriction than expected for normal tissues are the four most common MRI findings associated with metastatic LN. The concordance between MRI and histology was substantial, indicating that this method using MRI for diagnosing suspected LN metastasis is reliable. The results of the current study revealed that the most important aspects to be evaluated in MRI include: Short axis >10 mm, T2 hypointensity, rounded morphology and greater restriction than expected for normal tissues. If these four characteristics are present in MRI, histological evaluation is likely to reveal positive lymph node metastasis.