Distribution of metastasis in the brain in relation to the hippocampus: a retrospective single-center analysis of 565 metastases in 116 patients

OBJECTIVE: To examine the distribution of brain metastases (BM) in relation to the hippocampus, so as to determine the risk of metastasis in the hippocampus, and thus provide experimental evidence for the hippocampal avoidance (HA) in patients with BM during radiotherapy. METHODS: (1) For the retrospective analysis of 116 patients diagnosed with malignancies, confirmed as BM, from December 2014 to December 2016 at the First Affiliated Hospital of Bengbu Medical College. We obtained the T1-weighted, postcontrast axial, sagittal, and coronal Magnetic Resonance imaging (MRI) images f the patients, in supine position, using the head restraints and head thermoplastic masks to adjust the positioning, with computed tomography (CT) positioning scan ranging from the head to the mandible (layer thickness: 3 mm). CT and MRI images were fused on a Philips Pinnacle v9.8 treatment planning system;(2) Every metastasis of the 565 metastases was contoured;(3) hippocampus were contoured, and hippocampus with 5 mm expansion envelopes were analyzed;(4) Using the SPSS 16.0 software, we analyzed the relation between the distribution and age, sex, Karnofsky performance status (KPS), primary site, aggregate volume of intracranial metastases and the whole brain. The data were analyzed using a binary logistic regression analysis method, with two-sided P < 0.05 for statistical significance. RESULTS: In this study, we recruited 116 patients with 565 metastases. Among them, 1.7% (n = 2) had metastases in the hippocampus, and 11.2% (n = 13) had metastases within 5 mm of the hippocampus, of which more than half were patients with non-small cell lung cancer (n = 7). Using a binary logistic regression to analyze the relation between the metastases located within 5 mm of the hippocampus and age (P = 0.395), sex (P = 0.139), KPS (P = 0.724), primary site (P = 0.894), aggregate volume of intracranial metastases (p = 0.093) and the whole brain (p = 0.998), and none of them showed statistically significant difference between them and the metastases location (P>0.05). CONCLUSION: This study showed a low risk for the perihippocampal metastases (PHM) and no significant correlation between PHM and age, sex, KPS, primary site, aggregate volume of intracranial metastases and the whole brain. Accordingly, it is may be acceptable to avoid the perihippocampal region during whole brain radiotherapy.