Endoscopic ultrasound fine-needle biopsy vs fine-needle aspiration for lymph nodes tissue acquisition: a systematic review and meta-analysis
Antonio Facciorusso, Stefano Francesco Crinò, Paraskevas Gkolfakis, Daryl Ramai, Andrea Lisotti, Ioannis S. Papanikolaou, Benedetto Mangiavillano, Ilaria Tarantino, Andrea Anderloni, Carlo Fabbri, Κonstantinos Τriantafyllou, Pietro Fusaroli
Abstract
Abstract Background Endoscopic ultrasound (EUS)-guided tissue acquisition represents the choice of methods for suspected lymph nodes (LNs) located next to the gastrointestinal tract. This study aimed to compare the pooled diagnostic performance of EUS-guided fine-needle biopsy (EUS-FNB) and fine-needle aspiration (EUS-FNA) for LNs sampling. Methods We searched PubMed/MedLine and Embase databases through August 2021. Primary outcome was diagnostic accuracy; secondary outcomes were sensitivity, specificity, sample adequacy, optimal histological core procurement, number of passes, and adverse events. We performed a pairwise meta-analysis using a random-effects model. The results are presented as odds ratio (OR) or mean difference along with 95% confidence interval (CI). Results We identified nine studies (1,276 patients) in this meta-analysis. Among these patients, 66.4% were male; the median age was 67 years. Diagnostic accuracy was not significantly different between the two approaches (OR, 1.31; 95% CI, 0.81–2.10; P = 0.270). The accuracy of EUS-FNB was significantly higher when being performed with newer end-cutting needles (OR, 1.87; 95% CI, 1.17–3.00; P = 0.009) and in abdominal LNs (OR, 2.48; 95% CI, 1.52–4.05; P < 0.001) than that of EUS-FNA. No difference in terms of sample adequacy was observed between the two approaches (OR, 1.40; 95% CI, 0.46–4.26; P = 0.550); however, histological core procurement and diagnostic sensitivity with EUS-FNB were significantly higher than those with EUS-FNA (OR, 6.15; 95% CI, 1.51–25.07; P = 0.010 and OR, 1.87; 95% CI, 1.27–2.74, P = 0.001). The number of needle passes needed was significantly lower in the EUS-FNB group than in the EUS-FNA group (mean difference, −0.54; 95% CI, −0.97 to −0.12; P = 0.010). Conclusions EUS-FNA and EUS-FNB perform similarly in LN sampling; however, FNB performed with end-cutting needles outperformed FNA in terms of diagnostic accuracy.