La mia nuova descrizione qui!
Price of a print issue €25.00
BACKGROUND: Gastrointestinal Stromal Tumours (GIST) are the most frequent mesenchymal tumour of the alimentary
tract. Their prognosis is largely variable as are their size, mitotic rate and site, the stomach being mostly affected. Several
risk classifications have been proposed: two developed by the NIH, one proposed by the AFIP and one presented by the
AJCC in 2010. The objective of this study is to compare the accuracy of the three prognostic models (AJCC, NIH and
AFIP) with regard to survival after surgery, also based on the different surgical approaches.
METHODS: A retrospective review of all cases of gastric GIST’s performed at the General and Breast Surgery Unit of
the Department of General Surgery the University of Catania and at the “Gemelli” General Surgery Unit of Taormina
Hospital, Italy between 2001 and 2016 was conducted. The cases were reviewed and re- classified according to the three
prognostic models. Analysis of data, including Kaplan-Meyer survival curves, was performed using SPSS version 21.0.
RESULTS: Among a total of 1,625 gastrectomies and gastric resections were found 25 primary GIST’s patients, 13 females,
and 12 males, with a mean age 63 years. Cancer size varied between 1.5 cm and 37 cm and number of mitosis
between 2 and 50/50 HPF. A total of 12 (48%) underwent sub-total gastrectomy (STG), seven (28%) underwent a
wedge resection (WR), and 6 (12%) total Gastrectomy (TG). Twenty-three patients (92%) are currently alive at a follow
up of 18 months to 17 years, and only two patients died during the long term follow-up. Both patients were AFIP
high risk (6b), AJCC stage IV, already metastatic at the time of surgery. Both patients underwent total extended gastrectomy
and therapy with imatinib, but died 8 and 9 years after surgery. Recurrences have been observed in 2 patients
(8%), with high risk according to AFIP (6a) with AJCC stage IIIa disease.
CONCLUSIONS: In localized GISTs R0 surgical resection is the standard therapy as it leads to excellent outcomes. Our
findings suggest that all the three classifications considered are adequate to achieve a correct prognostic evaluation.