Early diagnosis of nodal metastasis has been shown to impact prognosis and treatment options of dogs with mast cell tumors (MCT). Lymph node (LN) characteristics on computed tomography (CT) have been used for detection of MCT metastasis but no consensus exists with regards to the criteria of malignancy. The objective of this study was to determine if a correlation exists between CT characteristics of LNs and histologic nodal metastasis using the HN classification system, in dogs with cutaneous or subcutaneous MCT and locoregional LN(s) removal. Dogs that had removal of cutaneous or subcutaneous MCT and regional lymphadenectomy within 31 days of the initial staging CT were enrolled. CT characteristics LN used include margin, shape, presence of fat at hilus, perinodal fat pattern, increasing in number of nodes, short-long axis ratios, enhancement, and heterogeneity. Prediction of metastasis on CT were compared with histopathology. Seventy-one LNs from 37 dogs were included. Metastasis was present in 6/7 LNs with ill-defined margins, in 5/8 LNs with irregular margin shape, in 4/7 LNs with loss of fat at hilus, in 6/7 LNs with nonuniform perinodal fat, and in 6/6 LNs with increased number of nodes. No strong correlation was found between nodal metastasis and short-long axis, enhancement, and heterogeneity, respectively. Limitations include its retrospective nature, small sample size, and lack of standardized lymphadenectomy protocol. Study results suggest that use of multiple CT criteria may increase accuracy of nodal metastasis detection.
