We read with great interest the study by Akahoshi et al., 1 highlighting the diagnostic outcome of endoscopic ultrasonography (EUS)-guided fine-needle aspiration (FNA) for gastric subepithelial hypoechoic masses. The authors enrolled 291 cases prospectively and reported a diagnosis rate of 80.1%, with most cases being gastrointestinal stromal tumours. The diagnostic yield was lower for lesions located in the antrum and those smaller than 2 cm. However, there are a few concerns that need to be addressed.A meta-analysis by Facciorusso et al., 2 analysing 10 studies, reported a greater proportion of sample adequacy (94.9% vs. 80.6%), a higher rate of optimal histologic core procurement (89.7% vs. 65%) and a higher rate of diagnostic accuracy (odds ratio, 4.10; 95% CI, 2.48-6.79) with a lower number of passes in cases using fine-needle biopsy (FNB) needle. Another metaanalysis comparing EUS-guided tissue acquisition (TA) with mucosal incision-assisted biopsy (MIAB) from subepithelial lesions (SEL) reported a lower positive diagnostic yield with EUS-TA than with MIAB (risk ratio, 0.83; 95% CI, 0.71-0.98). However, on subgroup analysis, the diagnostic yield was comparable when using a FNB needle (RR, 0.93; 95% CI, 0.83-1.04). 3 Thus, using a FNB needle could have helped improve the diagnostic yield, especially in lesions smaller than 2 cm. 4 The choice of optimal suction technique for SELs remains a subject of debate, and Akahoshi et al. 1 used dry suction in all cases. A randomised trial comparing standard suction and wet suction during EUS-FNA did not show any significant difference in the cellularity of the specimen. 5 Another retrospective study reported comparable diagnostic accuracies with standard suction and slow stylet pull techniques during EUS-FNB. 6 However, both these studies were underpowered. A previous network meta-analysis reported slightly better outcomes with wet suction in terms of lower odds of bloodiness. 7 Thus, further studies are required to determine the optimal suction technique for EUS-TA from SELs.Akahoshi et al. 1 used rapid on-site evaluation (ROSE) in all cases. However, ROSE may not be available at all centres, and in such cases, macroscopic on-site evaluation (MOSE) may
Authorship statementBoth SG and SS contributed to the conception and preparation of this letter, to the analysis of the results and to the writing of the manuscript.