Diagnostic yield of EUS-guided through-the-needle microforceps biopsy versus EUS-FNA of pancreatic cystic lesions: a systematic review and meta-analysis

Abstract

Background and study aims Accurate diagnosis and risk stratification of pancreatic cysts (PCs) is challenging. The aim of this study was to perform a systematic review and meta-analysis to assess the feasibility, safety, and diagnostic yield of endoscopic ultrasound-guided through-the-needle biopsy (TTNB) versus fine-needle aspiration (FNA) in PCs. Methods Comprehensive search of databases (PubMed, EMBASE, Cochrane, Web of Science) for relevant studies on TTNB of PCs (from inception to June 2019). The primary outcome was to compare the pooled diagnostic yield and concordance rate with surgical pathology of TTNB histology and FNA cytology of PCs. The secondary outcome was to estimate the safety profile of TTNB. Results: Eight studies (426 patients) were included. The diagnostic yield was significantly higher with TTNB over FNA for a specific cyst type (OR: 9.4; 95 % CI: [5.7-15.4]; I ² = 48) or a mucinous cyst (MC) (OR: 3.9; 95 % CI: [2.0-7.4], I ² = 72 %). The concordance rate with surgical pathology was significantly higher with TTNB over FNA for a specific cyst type (OR: 13.5; 95 % CI: [3.5-52.3]; I ² = 48), for a MC (OR: 8.9; 95 % [CI: 1.9-40.8]; I ² = 29), and for MC histologic severity (OR: 10.4; 95 % CI: [2.9-36.9]; I ² = 0). The pooled sensitivity and specificity of TTNB for MCs were 90.1 % (95 % CI: [78.4-97.6]; I ² = 36.5 %) and 94 % (95 % CI: [81.5-99.7]; I ² = 0), respectively. The pooled adverse event rate was 7.0 % (95 % CI: [2.3-14.1]; I ² = 82.9). Conclusions TTNB is safe, has a high sensitivity and specificity for MCs and may be superior to FNA cytology in risk-stratifying MCs and providing a specific cyst diagnosis.

Fig. 1

PRISMA Flowchart of included studies. From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(7): e1000097. doi:10.1371/journal.pmed1000097

Fig. 2

Pooled technical success of EUS-TTNB in the evaluation of pancreatic cysts (Fixed-effect Model).

Fig. 3

Pooled diagnostic yield of TTNB histology vs FNA cytology for a specific cyst type diagnosis (Random-Effect Model).

Fig. 4 a

Pooled diagnostic yield of TTNB histology vs FNA cytology for mucinous cysts (Random-Effect Model). b Pooled diagnostic yield of TTNB histology vs. cyst fluid CEA ≥ 192 ng/mL for mucinous cysts (Random-Effect Model). c Pooled diagnostic yield of TTNB histology vs FNA cytology for serous cystadenoma (Fixed-Effect Model).

Fig. 5

Pooled concordance of TTNB histology vs FNA cytology with surgical pathology for a specific cyst type (Random-Effect Model).

Fig. 6 a

Pooled concordance of TTNB histology vs FNA cytology with surgical pathology for a mucinous cyst (Random-Effect Model). b Pooled sensitivity of TTNB histology for diagnosing a mucinous cyst using surgical pathology as reference standard (Random-Effect Model). c Pooled specificity of TTNB histology for diagnosing a mucinous cyst using surgical pathology as reference standard (Fixed-Effect Model).

Fig. 7

Pooled concordance of TTNB histology vs FNA cytology with histological grading of mucinous cysts on surgical pathology (Fixed-Effect Model)

Fig. 8 a

Pooled adverse events with TTNB of pancreatic cysts. b Pooled occurrence of intracystic hemorrhage with TTNB of pancreatic cysts. c Pooled occurrence of acute pancreatitis with TTNB of pancreatic cysts.