Premium
Usefulness of a target sample check illuminator in the detection of target specimens in endoscopic ultrasound‐guided fine‐needle biopsy samples: Multicenter prospective study
Author(s) -
Matsumoto Kazuya,
Hara Kazuo,
Yasuda Ichiro,
Itoi Takao,
Kurumi Hiroki,
Matsumoto Shimpei,
Doi Shinpei,
Honjo Mitsuyoshi,
Takeda Yohei,
Shibuya Jin,
Noma Hisashi,
Isomoto Hajime
Publication year - 2021
Publication title -
digestive endoscopy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.5
H-Index - 56
eISSN - 1443-1661
pISSN - 0915-5635
DOI - 10.1111/den.13903
Subject(s) - medicine , endoscopic ultrasound , biopsy , prospective cohort study , predictive value , fine needle biopsy , nuclear medicine , ultrasound , radiology , pathological , fine needle aspiration
Objectives Endoscopic ultrasound‐guided fine‐needle biopsy (EUS‐FNB) uses a thin needle, rendering unclear whether the collected sample contains pathological evidence. We examined the usefulness of our target sample check illuminator (TSCI) through a multicenter prospective trial. Methods We included 52 consecutive patients. After assessing EUS‐FNB samples by conventional (visual observation) and TSCI methods, we evaluated consistency with the histopathological diagnosis. We compared the target sample confirmation rate between conventional and TSCI methods and evaluated the diagnostic ability separately. Results Comparison between the conventional and TSCI methods revealed the following: (i) for all cases: sensitivity, 51.0% (25/49) vs. 95.9% (47/49) ( P = 0.001); specificity, 100% (3/3) vs. 66.7% (2/3); positive predictive value (PPV), 100% (25/25) vs. 97.9% (47/48); and negative predictive value (NPV), 11.1% (3/27) vs. 50.0% (2/4) ( P = 0.002); (ii) for pancreatic masses: sensitivity, 28.0% (7/25) vs. 96.0% (24/25) ( P < 0.001); specificity, 100% (2/2) vs. 100% (2/2); PPV, 100% (7/7) vs. 100% (24/24); and NPV, 10.0% (2/20) vs. 66.7% (2/3) ( P < 0.001) (the TSCI method showed significantly better sensitivity and NPV than the conventional method); and (iii) for lymph node tumors: sensitivity, 75.0% (18/24) vs. 95.8% (23/24) ( P = 0.025); specificity, 100% (1/1) vs. 0% (0/1); PPV, 100% (18/18) vs. 95.8% (23/24); and NPV, 14.3% (1/7) vs. 0% (0/1). Conclusions The TSCI improved the sensitivity, NPV, and accuracy of target sample confirmation for pancreatic mass EUS‐FNB. Although the proportion of samples not including a target region was quite low, which could strongly influence our results, the TSCI method can be used in EUS‐FNB when rapid on‐site evaluation cannot be performed. (A multicenter prospective study for the utility of a target sample check illuminator, Clinical Trial ID: UMIN000023349).