Diagnostic efficacy of liquid-based cytology for solid pancreatic lesion samples obtained with endoscopic ultrasound-guided fine-needle aspiration: Propensity score-matched analysis

Abstract

Background and aim: There is a paucity of data on the diagnostic efficacy of liquid-based cytology (LBC) for pancreatic samples obtained by endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA). Using propensity score matching, we retrospectively analyzed the additional diagnostic value of LBC compared to a conventional Papanicolaou smear (CPS) for samples of solid pancreatic lesions obtained by EUS-FNA.

Methods: This cohort study included 126 matched patients who underwent initial EUS-FNA for solid pancreatic lesions between January 2009 and August 2014. CPS was used for cytology of EUS-FNA samples obtained until May 2012 (63 patients). Subsequently, LBC was used for cytological analysis (63 patients). Diagnostic yields of CPS and LBC for malignancy were compared. Risk factors for cytological misdiagnosis with LBC were investigated.

Results: Overall rate of malignancy was 86% after matching. LBC had higher diagnostic sensitivity and accuracy than CPS (96.6% vs 84.0%, P = 0.03; and 96.8% vs 87.3%, P = 0.05). LBC was significantly more sensitive for diagnosing pancreatic head lesions (96.4% vs 78.1%, P = 0.04). The sensitivity for pancreatic ductal adenocarcinoma (PDAC) with LBC was higher (98.1% vs 83.0%, P = 0.009). Multivariate analysis revealed that malignant tumors other than PDAC (P = 0.004) and lesion size ≤20 mm (P = 0.046) were risk factors for LBC misdiagnosis in all participants.

Conclusions: For solid pancreatic lesions, LBC of EUS-FNA samples contributes to the diagnosis of malignancy. Malignant tumors other than PDAC and small tumors are difficult to diagnose using EUS-FNA and LBC.

Keywords: BD SurePath technique; conventional Papanicolaou smear; endoscopic ultrasonography-guided fine-needle aspiration; liquid-based cytology; solid pancreatic lesion.