Engh Marie Anne, Teutsch Brigitta, Schulze Wenning Alexander, Kói Tamás, Hegyi Péter, Erőss Bálint
Centre for Translational Medicine, Semmelweis University, 1085 Budapest, Hungary.
Institute for Translational Medicine, Medical School, University of Pécs, 7624 Pécs, Hungary.
J Clin Med. 2025 Sep 22;14(18):6685. doi: 10.3390/jcm14186685.
Endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) or biopsy (FNB) is the standard method for diagnosing abdominal masses, but sample inadequacy and diagnostic accuracy remain challenges. Conventional smear (CS) and liquid-based cytology (LBC) are standard processing methods, yet their comparative effectiveness and potential combined benefit remain unclear. We performed a systematic review and meta-analysis to evaluate and compare the diagnostic performance and adequacy of CS, LBC, and their combination. A systematic search was conducted in Medline, Embase, and CENTRAL on 17 November 2024. Studies comparing CS, LBC, or their combination following EUS-FNA/FNB for abdominal masses were included. Diagnostic parameters, including sensitivity, specificity, accuracy, and inadequacy rates, were extracted and analyzed. Methodological quality was assessed using QUADAS-2. : 16 studies (2128 patients) were included. Sensitivity for pancreatic masses was 71.4% (CI: 62.9-78.7) for CS, 74.7% (CI: 64.3-82.8) for LBC, and 86.2% (CI: 82.4-89.3) for combined methods ( = 0.001). For all abdominal masses, sensitivity was 76.3% (CI: 67.9-83.0) for CS, 73.6% (CI: 65.6-80.2) for LBC, and 88.0% (CI: 84.0-91.2) for combined methods ( ≤ 0.006). Specificity was nearly 100%. Inadequacy rates were lowest for combined methods (1.5%, CI: 0-36.2), when compared to LBC (7.7%, CI: 2.7-20.4) and CS (4.4%, CI: 2.4-7.9). Moderate bias risk was noted, primarily due to incorporation bias. Domain 3 (reference standard) of QUADAS was uniformly moderate-risk across studies. : Combining CS and LBC methods improves diagnostic sensitivity and reduces sample inadequacy after EUS-guided tissue acquisition for abdominal masses, particularly pancreatic lesions. Clinical guidelines should consider recommending the combined approach to enhance diagnostic yield and clinical outcomes.
内镜超声（EUS）引导下细针穿刺抽吸（FNA）或活检（FNB）是诊断腹部肿块的标准方法，但样本不足和诊断准确性仍是挑战。传统涂片（CS）和液基细胞学检查（LBC）是标准的处理方法，但其相对有效性和潜在的联合益处仍不明确。我们进行了一项系统评价和荟萃分析，以评估和比较CS、LBC及其联合方法的诊断性能和充分性。2024年11月17日在Medline、Embase和CENTRAL进行了系统检索。纳入了比较EUS-FNA/FNB后CS、LBC或其联合方法用于腹部肿块的研究。提取并分析了包括敏感性、特异性、准确性和不足率在内的诊断参数。使用QUADAS-2评估方法学质量。共纳入16项研究（2128例患者）。CS对胰腺肿块的敏感性为71.4%（CI：62.9-78.7），LBC为74.7%（CI：64.3-82.8），联合方法为86.2%（P = 0.001）。对于所有腹部肿块，CS的敏感性为76.3%（CI：67.9-83.0），LBC为73.6%（CI：65.6-80.2），联合方法为88.0%（P≤0.006）。特异性接近100%。与LBC（7.7%，CI：2.7-20.4）和CS（4.4%，CI：2.4-7.9）相比，联合方法的不足率最低（1.5%，CI：0-36.2）。注意到存在中度偏倚风险，主要是由于纳入偏倚。QUADAS的第3领域（参考标准）在各研究中均为中度风险。结论：在EUS引导下获取腹部肿块组织后，联合CS和LBC方法可提高诊断敏感性并减少样本不足，尤其是胰腺病变。临床指南应考虑推荐联合方法以提高诊断率和临床结局。
