Determinação de fatores prognósticos para tumores mamários caninos

Abstract

The mammary neoplasms are tumors that most affect bitches and represent a problem of great impact in veterinary medicine. Malignant tumors are associated to high mortality due to tumor recurrence or metastasis. This study aimed to evaluate prognostic factors of low cost and easy to perform diagnostic routines, capable of predicting the biological behavior of these tumors. The samples were collected from 114 bitches submitted surgical excision of mammary neoplasms from 2004 to 2008. The patients followed over a period of two years after therapy. This study did not include animals that had distant metastasis at diagnosis. Information regarding to ovariohysterectomy concurrent with the removal of the neoplasm, recurrences, time of observation of the tumor, age and weight of the animals were obtained through standard protocols for submission of tumors. The samples were fixed in 10% formalin, measured, the margins of interest identified with India ink and were systematically cleaved allowing full evaluation of the specimen. Histological sections were stained with hematoxylin-eosin and evaluated under light microscopy. The dependent variables evaluated were: histologic type, histologic grading, histological stage, invasion to adjacent tissues, neoplastic emboli in blood or lymphatic vessels, mitotic index and status of surgical margins. The variables were compared among themselves and with the survival time of postoperative animals. Survival evaluation was used log-rank test and evaluation of the variables between them was used Kruskall-Wallis. The results were presented by estimating survival curves using the Kaplan-Maier method. The values were statistically significant when p<0.05. The following variables were significantly associated with survival: age at diagnosis (p=0.004), size of tumor (p<0.000), histological type, histological grade (p=0.000), histological stage (p<0.000), invasion to adjacent tissues (p=0.033), presence of neoplastic emboli in blood or lymphatic vessels (p=0.000), mitotic index (p=0.006) and recurrence (p=0.034). Sized animals (p=0.591), ovariohysterectomy concurrent surgical excision (p=0.74), time observation of the tumor (p=0.3) and surgical margin status (p=0.369) showed no prognostic significance. Additionally, there was significant statistical correlation between the following variables: size of tumor with histological grade (p<0.000), histological stage (p=0.008) and mitotic index (p=0.002); histological grading with histological type and histological stage (both P<0.000); and histological stage with histological type and mitotic index (both p<0.000). The variables significantly associated with survival were helpful in determining the prognosis of the animals. There was an increase of characteristics that indicate malignancy with increasing the size of tumors. The evaluated factors are inexpensive and can be done relatively quickly. They should be used routinely by veterinary pathologists, can help clinicians and surgeons in therapeutic decisions for carrier animals.