A: The T-STAT calculator estimates the probability of malignancy using a variety of data from the individual patient rather than broad probabilities that pertain to the general population of dogs with splenic masses. This is particularly valuable given that dogs with benign and malignant splenic masses may present with virtually indistinguishable histories and physical exam findings.

A: No - approximately 28% of the malignant masses included in the study were tumors other than hemangiosarcoma. Nevertheless, almost all of these tumor types were either multicentric in nature or highly metastatic, and like hemangiosarcoma are unlikely to be cured by splenectomy alone.

A: No - missing data invalidates the calculation.

A: While the accuracy of T-STAT is very good, it should not take the place of your own clinical judgment; at most, T-STAT should provide an objective adjunct to your decision- making process.

A: The T-STAT model was developed using data from a cohort of dogs undergoing splenectomy for a splenic mass at our institution. The vast majority of our population consisted of middle sized and larger breed dogs. Additionally there were relatively few dogs less than 5 or greater than 12 years of age. Consequently our model may have limited applicability to small breed dogs and to very young and very old dogs.

A: Severity of abdominal effusion and splenic mass inhomogeneity were initially assigned to one of four categories (none, mild, moderate and marked). However, our analysis indicated that for both variables, the selection of categories could be simplified by combining the none and mild categories and the moderate and marked categories, without losing predictive value. For both variables, if you are uncertain which category best fits your patient, selection of the none/mild category will result in a more conservative estimate of the probability of malignancy.

A: A limitation of the calculator is that quantification of these variables based on ultrasound is inherently subjective. Consequently, tutorials for making these assessments are available on the calculator page.

A: The predictive model is based on clinical tests that are commonly performed in general and referral practice. CT and MRI are rarely used in the evaluation of splenic masses, and while they may eventually prove less subjective and more informative than abdominal ultrasound, as of yet there is little data regarding their relative utility in this application.

A: It is likely that elevated nRBC levels in dogs with splenic masses is related to active erythropoiesis secondary to acute blood loss.

Although both acanthocytes and schistocytes can be observed in dogs with hemangiosarcoma, they are also frequently present in dogs with non-neoplastic diseases that may cause RBC fragmentation, and therefore have limited value in documenting the presence of hemangiosarcoma.

In the population of dogs used to develop the predictive model, dogs with multiple splenic masses or nodules had a higher probability of having malignant disease than dogs with solitary lesions. It can be intuited that many of these nodules represented intra- splenic metastatic disease; however, histopathology was generally confined to the largest mass.