Preoperative Prediction of Cardiophrenic Lymph Node Metastasis in Advanced Ovarian Cancer Using Computed Tomography.

OBJECTIVE

The aim of this study was to determine whether the preoperative diameter of cardiophrenic lymph nodes (CPLNs) along the short and long axes, determined via computed tomography (CT), predicts CPLN metastasis in patients with advanced epithelial ovarian cancer.

METHODS

A total of 31 patients with primary advanced ovarian cancer who had CPLN dissection underwent preoperative CT that was reviewed twice by a radiologist blinded to the final pathology. An array of CT features and clinical factors [age, cancer antigen 125 (CA 125)] were compared between metastasis-positive and metastasis-negative CPLNs using the t test and Fisher's exact test. The optimal CPLN dimensions that best predicted metastasis were calculated using a receiver operating characteristic (ROC) curve, and were applied to find the correlation with other CT findings. Results were validated in an independent test set of nine patients by using the best cutoff value to predict metastasis.

RESULTS

Pathological evaluation revealed metastasis-positive CPLNs in 19 patients and metastasis-negative CPLNs in 12 patients. Metastasis-positive CPLNs had significantly reduced short axes than those of negative CPLNs (5.6 ± 1.7 and 7.8 ± 2.2, respectively; p < 0.001). ROC curve analysis showed that a cutoff value of 7 mm for the short axis demonstrated the largest area under the curve (0.789; p < 0.0007), with 63.16 % sensitivity and 83.33 % specificity. The presence of peritoneal metastasis and abdominopelvic adenopathy strongly related with CPLN adenopathy. Accuracies were 66.7 % (six of nine patients) based on the cutoff value in the independent test set.

CONCLUSIONS

The probability of detecting CPLN metastasis in patients with advanced ovarian cancer was approximately 85 % when the short axis of the CPLN was >7 mm in preoperative CT scans. Patients with CPLNs of this size may be candidates for CPLN dissection in order to confirm the pathological diagnosis.