Splenic metastases in gynaecologic cancer: clinical considerations, US, and CT diagnostic results.

Purpose: Splenic metastases from lesions in the genital system are an uncommon finding, both at staging and at follow-up. Most metastases are not an isolated finding but are quite frequently associated with metastases to lymph node or parenchymal sites. The aim of this paper is to describe the patterns of splenic metastasis at US and CT, the diagnostic imaging techniques most often employed in primary evaluation and follow-up of gynaecologic cancer. We consider both the issues of differential diagnosis and clinical decision making.

Materials And Methods: From January, 1996, to May, 2001, we retrospectively reviewed the CT and US findings of 16 (2.2%) patients with splenic metastases out of 724 patients examined for malignant gynaecologic lesions. US was performed on 479 patients with standard equipment, applying colour and power-Doppler. CT was performed with helical scan, at baseline and after the i.v. administration of a nonionic iodinated contrast agent (120-140 mL at 300-350 mgI/mL concentration, 2 mL/s flow, 70 s administration delay). We used pitch 1.2 with 5-7 mm collimation. We considered metastases lesions that appeared as roundish or irregular, avascular parenchymal nodules at US, and lesions which were hypodense both at baseline and after i.v. administration of iodinated contrast agent at CT. At baseline examination of serous ovarian cancer, metastatic calcifications due to the presence of psammomatous bodies were considered. The presence or absence of splenic metastasis was assessed for all patients by pathology of the surgical specimen, or based on the clinical evolution of the disease.

Results: Epithelial ovarian cancer was the most frequent tumour causing splenic metastases (14 cases), while the other two cases were due to advanced endometrial adenocarcinoma and squamous cell vaginal carcinoma. Splenic metastases were detected most frequently during the follow-up and were associated with other sites of recurrence. In two cases US gave false-positive results (multiple micronodular lesions, missed at CT). Negative follow-up at 12 months confirmed absence of recurrence. CT provided only one false-negative result (clinical and US-CT evidence of recurrent disease two months later). We recorded the diagnostic accuracy (99.9% vs 99.9%), sensitivity (100% vs 93.6%), specificity (99.9% vs 100%), positive (87.5% vs 100%) and negative (100% vs 99.9%) predictive value, for US and CT, respectively.

Conclusions: Splenic metastases from female genital system cancers are an uncommon finding. These lesions must be considered with other recurrences, especially in follow-up patients. US and CT provided high accuracy and both can be used for lesion assessment. The micronodular splenic pattern may cause false-positive findings at US.