figshare
Browse
1/1
2 files

EUS-FNA WITH 19 OR 22 GAUGES NEEDLES FOR GASTRIC SUBEPITHELIAL LESIONS OF THE MUSCLE LAYER

dataset
posted on 2018-06-27, 02:43 authored by César Vivian LOPES, Antônio Atalíbio HARTMANN, Everson Luiz de Almeida ARTIFON

ABSTRACT Background: Tissue diagnosis is required for gastric subepithelial lesions for differential diagnosis of GISTs. However, there has not been consensus about the best needle for EUS-guided sampling of these lesions. Aim: To evaluate the diagnostic yield of EUS-FNA for gastric subepithelial lesions of the proper muscle layer with large-bore 19 gauge needles. Methods: A prospectively maintained database was retrospectively reviewed to identify consecutive patients who underwent EUS-FNA with 19 and 22 gauge needles for gastric subepithelial lesions of the fourth endosonographic layer in a tertiary care referral center. EUS-FNA was performed by the same endosonographer, using the fanning technique, without on-site cytopathologist. Specimens were analysed through cell blocks by the same pathologist. Procedure results were categorized into diagnostic, defined as enough material for histopathology and immunohistochemistry, or nondiagnostic. Results: Eighty-nine patients (mean age: 59 years, 77% women) underwent 92 EUS-FNA with 19 (75) or 22 (17) gauge needles. Mean lesion size was 22.6 mm. Overall diagnostic yield was 88%. The diagnostic yield of 19 gauge was higher than that of 22 gauge needle (92%x70.6%; p=0.0410), and similar for lesions >2 cm and <2 cm (93.7%x90.7%; p=0.9563). The best performance for 19 gauge needles was obtained performing <3 needle passes. Complication rate was 2.8%. Conclusions: Diagnostic yield of EUS-FNA with 19 gauge needles is 92% for gastric subepithelial lesions of the proper muscle layer. It is safe and highly valuable for differentiation between GIST and leiomyoma, no matter the size of the lesion.

History

Usage metrics

    ABCD. Arquivos Brasileiros de Cirurgia Digestiva (São Paulo)

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC