Rapid Onsite Evaluation in the Era of Endoscopic Ultrasound–Guided Fine-Needle Biopsy: Really Time to Say Goodbye?

Jun Li,Feng Liu

DOI: https://doi.org/10.1053/j.gastro.2021.06.043

IF: 29.4

2022-01-01

Gastroenterology

Abstract:We read with great interest the article by Crinò et al1Crinò S.F. et al.Gastroenterology. 2021; 161: 899-909.e5Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar regarding the role of rapid onsite evaluation (ROSE) in the era of endoscopic ultrasound (EUS)–guided fine-needle biopsy (FNB) for solid pancreatic lesions (SPLs). The researchers reported the largest multicenter randomized controlled trial (RCT) so far and concluded that ROSE should not be routinely recommended when new-generation FNB needles are used. We would like to congratulate them for the accomplishment of excellent work. However, we would also like to raise the following concerns. One of the key strengths of this study is its multicenter study design with a large sample size. However, this strength is weakened substantially because all involved centers were highly experienced, and all procedures were performed by expert endosonographers. ROSE was also reported previously to provide limited improvement in the adequacy of EUS-fine-needle aspiration (FNA) for SPLs,2Kong F. et al.PLoS One. 2016; 11e0163056Crossref PubMed Scopus (41) Google Scholar but it still possesses value in less-experienced centers; it is recommended by the European Society of Gastrointestinal Endoscopy (ESGE) and European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) that ROSE may improve adequacy in centers with an adequacy rate of <90%.3Polkowski M. et al.Endoscopy. 2017; 49: 989-1006Crossref PubMed Scopus (167) Google Scholar,4Jenssen C. et al.Ultraschall Med. 2016; 37: E33-E76PubMed Google Scholar In their study, Crinò et al1Crinò S.F. et al.Gastroenterology. 2021; 161: 899-909.e5Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar discussed that the availability of new-generation FNB needles may facilitate the widespread use of EUS-guided sampling without ROSE, especially in community hospitals where ROSE is restricted by a lack of cytopathologic expertise. This is a paradoxical situation because EUS-FNA/B expertise may also be limited in these hospitals, leading to even more demands on ROSE to help improve diagnostic yield. Efforts have been made to overcome this situation by training endosonographers to perform ROSE, and some inspiring results have been reported.5Hayashi T. et al.J Gastroenterol Hepatol. 2013; 28: 656-663Crossref PubMed Scopus (60) Google Scholar What’s more, the strength of the results in experienced centers may also be influenced by several drawbacks of the study protocol. First, there was quite a lot of variety among the needle types, suction methods, and fanning techniques. These factors have been reported to have a significant impact on the outcomes of EUS-guided sampling. Decisions at the discretion of the endosonographers not blinded to allocation made the study not a well-controlled randomized trial. Changes of suction methods and fanning techniques among passes in an individual patient are not uncommon during real-world procedures but were not mentioned in this study. Second, there were also discrepancies in specimen processing between the 2 groups. In the EUS-FNB–alone group, the specimens were actually evaluated by macroscopic onsite evaluation, which was reported in a recent multicenter RCT to allow fewer needle passes for similar diagnostic yield during EUS tissue acquisition.6Chong C.C.N. et al.Endoscopy. 2020; 52: 856-863Crossref PubMed Scopus (15) Google Scholar In this group, all specimens were placed into a single vial for histopathologic evaluation, but no cytologic smears were made. This would exert influences on the results of comparisons of the procedural time between the 2 groups. Third, the number of needle passes was another main concern. In a multicenter RCT investigating EUS-FNA for lymph nodes, the median number of passes needed for adequacy was 1 (interquartile range, 1–3).7Kappelle W.F.W. et al.Am J Gastroenterol. 2018; 113: 677-685Crossref PubMed Scopus (21) Google Scholar Another RCT investigating EUS-FNA for SPLs showed that the mean number of passes for diagnosis was 2.1 ± 2.2 (interquartile range, 1–2).8Bang J.Y. et al.Clin Gastroenterol Hepatol. 2018; 16: 1820-1828.e4Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar Even when ROSE was performed by endosonographers in EUS-FNA for SPLs, the median number of passes was also 2 (range, 1–5).5Hayashi T. et al.J Gastroenterol Hepatol. 2013; 28: 656-663Crossref PubMed Scopus (60) Google Scholar These studies indicated that usually only 1 or 2 passes were adequate for diagnosis. The advantage of ROSE is to provide a reliable immediate cytologic diagnosis and give a definite signal to cease additional needle passes. In this study, the procedure was repeated for up to 3 passes, even in the case of adequate ROSE. Therefore, the mean number of passes in the group of EUS-FNB with ROSE (2.88 ± 0.39) was significantly higher than reported. On the other hand, a fixed number of 3 needle passes performed in the EUS-FNB–alone group was unreasonable as well. Although the ESGE suggests the performance of 2 or 3 passes with an FNB needle, the strength of the recommendation was weak, with low-quality evidence.3Polkowski M. et al.Endoscopy. 2017; 49: 989-1006Crossref PubMed Scopus (167) Google Scholar Previous studies have reported a fixed number of 7 or 5 passes during EUS-FNA without ROSE.3Polkowski M. et al.Endoscopy. 2017; 49: 989-1006Crossref PubMed Scopus (167) Google Scholar The set of 3 passes was an improvement to that of previous studies but also does not reflect real-world situations. Because of uncertainty about specimen quantity and quality without ROSE, it is very likely that the endosonographers will perform more than 3 passes to ensure better diagnostic yield, which was also supported by several studies.9Zhou W. et al.Gastrointest Endosc. 2020; 91: 932-942Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar In summary, Crinò et al1Crinò S.F. et al.Gastroenterology. 2021; 161: 899-909.e5Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar have done great work, but the evidence is still not convincing enough to abandon ROSE in the era of EUS-FNB. Controversy remains, and further well-designed studies with uniformly controlled variables and per-pass analyses are warranted to fully elucidate this issue. Endoscopic Ultrasound–guided Fine-needle Biopsy With or Without Rapid On-site Evaluation for Diagnosis of Solid Pancreatic Lesions: A Randomized Controlled Non-Inferiority TrialGastroenterologyVol. 161Issue 3PreviewThe benefit of rapid on-site evaluation (ROSE) on the diagnostic accuracy of endoscopic ultrasound–guided fine-needle biopsy (EUS-FNB) has never been evaluated in a randomized study. This trial aimed to test the hypothesis that in solid pancreatic lesions (SPLs), diagnostic accuracy of EUS-FNB without ROSE was not inferior to that of EUS-FNB with ROSE. Full-Text PDF ReplyGastroenterologyVol. 162Issue 2PreviewWe thank Sun et al1 and Li et al2 for their comments on our article, “Endoscopic Ultrasound–Guided Fine-Needle Biopsy With or Without Rapid On-site Evaluation for Diagnosis of Solid Pancreatic Lesions: A Randomized Controlled Non-inferiority Trial.”3 Full-Text PDF