A meta-analysis of slow pull versus suction for endoscopic ultrasound-guided tissue acquisition

Background/Aims: Endoscopic ultrasound (EUS)-guided tissue acquisition is widely utilized as a diagnostic modality for intra-abdominal masses, but there remains debate regarding which suction technique, slow pull (SP) or conventional suction (CS), is better. A meta-analysis of reported studies was conducted to compare the diagnostic yields of SP and CS during EUS-guided tissue acquisition. Methods: We conducted a systematic electronic search using MEDLINE/PubMed, Web of Science, and the Cochrane Central Register of Controlled Trials to identify clinical studies comparing SP and CS. We meta-analyzed accuracy, sensitivity, blood contamination and cellularity using the random-effects model. Results: A total of 17 studies (seven randomized controlled trials, four prospective studies, and six retrospective studies) with 1,616 cases were included in the analysis. Compared to CS, there was a trend toward better accuracy (odds ratio [OR], 1.48; 95% confidence interval [CI], 0.97 to 2.27; p=0.07) and sensitivity (OR, 1.67; 95% CI, 0.95 to 2.93; p=0.08) with SP and a significantly lower rate of blood contamination (OR, 0.48; 95% CI, 0.33 to 0.69; p<0.01). However, there was no significant difference in cellularity between SP and CS, with an OR of 1.28 (95% CI, 0.68 to 2.40; p=0.45). When the use of a 25-gauge needle was analyzed, the accuracy and sensitivity of SP were significantly better than those of CS, with ORs of 4.81 (95% CI, 1.99 to 11.62; p<0.01) and 4.69 (95% CI, 1.93 to 11.40; p<0.01), respectively. Conclusions: Compared to CS, SP appears to provide better accuracy and sensitivity in EUS-guided tissue acquisition, especially when a 25-gauge needle is used.