When ROSE is not possible: Evaluating the diagnostic yield of>4mm tissue specimen with macroscopi c onsite evaluation. 22G EUS-FNB Acquire needle experience

Aims Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) is a technique used for tissue acquisition providing the higher diagnostic accuracy for solid pancreatic and gastrointestinal lesions. Rapid onsite evaluation (ROSE) is the us eful firsthand diagnostic technique. Because of its expensive technical requirement, macroscopic onsite evaluation (MOSE) is a desired method used to assess the visual characteristics/features of a sample/specimen for positive hist opathological output. However, adequate size of the tissue is the requirement. What tissue size should we have for histopathological diagnosis, answer is debatable. Literature showed variability in tissue size with the favor towards>4mm tissue size. Therefore, the aim of the study is to evaluate the>4mm of tissue size for the adequacy of cellbl ock and histopathological diagnostic output.

Methods This is a retrospective study, conducted at the gastroenterology-department of Liaquat National Hospital, Karachi, P akistan. Data collection duration was Jan-2019 to July-2023 after asking institutional permission. It included all cons ecutive data of patient’s tissue size, ROSE, and histopathological output of cellblock, obtained from patient’s gastroi ntestinal/pancreatic lesions, and recorded in departmental Electronic Medical Record (EMR). Abandoned procedures data were excluded. Data was entered and analyzed using SPSS version 25. ROC curve was plotted to determine th e performance of tissue size in prediction of positive histopathologic/cell block and AUC was calculated. Sensitivity,s pecificity,positive and negative predictive values were computed at threshold of tissue size of>4mm.

Results Total 122 EUS-FNB data (69.9%, 84/122 malignant) was collected from EMR. Median age 60 years (IQR=48-67), m ales (59.8%) dominance, pancreas common biopsy site (73.8%). ROSE was 94% (79/84) malignant, Histopathologi cal/Cellblocks were Malignant in 93% (78/84) lesion. Tissue size of>4mm was malignant in 82% (64/78) cellblocks. ROC curv, for>4mm tissue size and taking malignant histopathology/cellblock as gold standard, showed an AUC of 0.85 with statistical significance. At threshold of 4mm tissue size and above, sensitivity, specificity were 82%, 82%, while positive predictive and negative predictive value were 89% and 72% respectively. [1] [2] [3] [4]

Conclusions 4mm and above tissue size can be useful for the predictability of higher histopathological (Cellblock) diagnostic outp ut in gastrointestinal and pancreatic solid lesions.

Publication History

Article published online:
15 April 2024

• References

• 1 E. L. Leung Ki, A. I. Lemaistre, F. Fumex, R. Gincul, C. Lefort, V. Lepilliez, et al. Endosc Int Open 2019 Vol. 7 Issue 2 Pages E189-E194 .Accession Number: 30705952 PMCID: PMC6338542 DOI: 10 .1055/a-0770-2726
• 2 Crino S.F., Manfrin E., Scarpa A., Baldaque-Silva F., Carrara S., De Nucci G.. et al. Dig Liver Dis 2019; Vol. 51 Issue 6: 901-906 Accession Number: 30975612
• 3 Chung MJ, Park SW, Kim SH, Cho CM, Choi JH, Choi EK. Clin Endosc 2021; 54: 161-181
• 4 Sundaram S., Chhanchure U., Patil P., Seth V., Mahajan A., Bal M.. et al. Ann Gastroenterol 2023; Vol. 36 Issue 3: 340-346 Accession Number: 37144017 PMCID: PMC10152805