Skip to main content
. 2019 May 9;8(4):261–268. doi: 10.4103/eus.eus_14_19

Table 3.

20G ProCore™ performance on pancreatic lesions

20G PC performance Pancreatic lesions (n=236)
Age (years), mean±SD 68.3±11.8
Number of passes, mean±SD 2.8±1.0
Lesion site, n (%)
 Pancreatic head and uncinated process 150 (63.6)
 Pancreatic body 64 (27.1)
 Pancreatic tail 22 (9.3)
Possibility to obtain histological core, n (%) 170 (72.0)
Possibility to draw histological/cytological diagnosis from tissue sample (diagnostic yield), n (%) 218 (92.4)
Diagnosis at EUS vs. gold standard diagnosis type, n
 Malignant lesion 197/217
 Benign lesion 21/19*
 No sample or inadequate sample 18
20G PC diagnostic performance for malignant lesions, % (95% CI)
 Overall accuracy 91.5 (87.2–94.7)
 Sensitivity 90.8 (86.1–94.3)
 Specificity 100 (82.4–100)
 PPV 100 (98.1–100)
 NPV 48.7 (32.4–65.2)
NND 1.1 (1.1–1.5)
NNM 11.8 (7.8–18.9)
20G PC-related complications, n(%)
 Bleeding 1 (0.4)
Pain 1 (0.4)

*Two lesions were incorrectly diagnosed as benign at EUS-guided sampling, while were found to be malignant. PC: ProCore™, PPV: Positive predictive value, NPV: Negative predictive value, NNM: Number needed to misdiagnose, NND: Number needed to diagnose, CI: Confidence interval, EUS: Endoscopic ultrasound, SD: Standard deviation