Comparative Assessment of Endoscopic Ultrasound-Guided Biopsies vs. Percutaneous Biopsies of Pancreatic Lesions: A Systematic Review and Meta-Analysis of Diagnostic Performance.

Introduction: Pancreatic cancer ranks as the fourth deadliest form of cancer. However, it is essential to note that not all pancreatic masses signal primary malignancy. Therefore, it is imperative to establish the correct differential diagnosis, a process further supported by pre-operative biopsy procedures. This meta-analysis aims to compare the diagnostic performance of two minimally invasive biopsy approaches for pancreatic tissue sampling: percutaneous biopsies guided by computed tomography or ultrasound, and transduodenal biopsies guided by endoscopic ultrasound (EUS). Methods: A systematic literature search was conducted in the MEDLINE and Scopus databases. The included studies analyzed the diagnostic performance of the two biopsy methods, and they were assessed for risk of bias using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Statistical analysis was carried out using the RevMan and MetaDisc software packages. Results: The statistical analysis of the results demonstrated the superiority of the percutaneous approach. Specifically, the pooled sensitivity, specificity, LR+, LR-and DOR for the percutaneous approach were 0.896 [95% CI: 0.878-0.913], 0.949 [95% CI: 0.892-0.981], 9.70 [95% CI: 5.20-18.09], 0.20 [95% CI: 0.12-0.32] and 68.55 [95% CI: 32.63-143.98], respectively. The corresponding values for EUS-guided biopsies were 0.806 [95% CI: 0.775-0.834], 0.955 [95% CI: 0.926-0.974], 12.04 [95% CI: 2.67-54.17], 0.24 [95% CI: 0.15-0.39] and 52.56 [95% CI: 13.81-200.09], respectively. Nevertheless, it appears that this statistical superiority is also linked to the selection bias favoring larger and hence more readily accessible tumors during percutaneous biopsy procedures. Conclusions: Concisely, our meta-analysis indicates the statistical superiority of the percutaneous approach. However, selecting the optimal biopsy method is complex, influenced by factors like patient and tumor characteristics, clinical resources, and other relevant considerations.