RSS-Feed abonnieren
DOI: 10.1055/s-0031-1292091
Diagnostic yield of EUS-FNA in the diagnosis of subepithelial lesions of the gastrointestinal tract
Introduction: Endoscopic Ultrasound-Fine Needle Aspiration (EUS-FNA) has become standard practice in the evaluation of subepithelial masses of the gastrointestinal (GI) tract, since it allows a clear identification of the layer of origin of the lesion and its sampling. This kind of lesions needs a definite diagnosis because some of them, mainly the Gastrointestinal Stromal Tumors (GISTs), bear a malignant potential. However a tissue diagnosis is seldom necessary because usually treatment decisions can be planned mainly on clinical bases.
Aims: We describe our 7 years experience in EUS-FNA of subepithelial lesions and all the techniques we were able to apply in order to reach this difficult differenzial diagnosis even on a very limited amount of material.
Methods: From 2003 to June 2010 we performed EUS-FNA on 36 subepithelial masses of the GI tract. All the procedures were accomplished with an on-site cytopathologist to assess the adequacy of the specimens. According to the cytomorphological features of the lesions and the cytopathologist's judgment, immunocytochemical analysis for CD117, CD34, S100, SMA, AE1/AE3 and Ki67 was employed. In 7/36 cases it has also been possible to apply genomic analyses to the retrieved specimen in order to check on possible mutations in c-KIT gene and PDGFRA gene or other mutations, such as translocations t(11;22) and t(X;18).
Results: For each lesion a mean of 5.05 +/- 2.28 needle passes (range 2–10) was performed; aspiration needles from 25 to 19 gauge or trucut-needles (in 10 patients) were used according to the operator's preferences and mass location. A single case of minor bleeding occurred: it was endoscopically treated and no transfusion was required. An adequate specimen could be collected in all 36 cases. Based on immunocytochemical results, a definite diagnosis was reached in 30/36 cases (83.33%): we observed 20 (55.55%) GISTs, 6 (16.66%) leyomiomas, 2 (5.55%) synovial sarcomas, 1 (2.77%) peripheral nerve ectodermal tumour (PNET) and 1 (2.77%) fibrovascular tumor. In the remaining 6/36 cases (16.66%) we could not go beyond a generic diagnosis of “stromal spindle-cell lesion”.
Conclusions: Although subepithelial lesions still represent a diagnostic challenge, an extensive immuno-cytochemical evaluation, performed on specimens obtained by EUS-FNA, can reach a definite diagnosis in a significant number of cases.