BACKGROUND AND AIM: Published studies have revealed the diagnostic yield of cytology obtained from endoscopic retrograde cholangiopancreatography (ERCP) in distinguishing malignant and benign intraductal papillary mucinous neoplasm (IPMN). However as a result of small sample sizes, the overall magnitude of benefit is unknown. Additionally, the optimal endoscopic procedure for cytology acquisition is also unclear. The aim of the present study was to evaluate the diagnostic yield of ERCP-based cytology in patients with IPMN and clarify the optimal sampling technique. METHODS: Relevant studies with a reference standard of definitive surgical histology were identified via MEDLINE and SCOPUS. Malignant IPMN included invasive adenocarcinoma, carcinoma in situ, and high-grade dysplasia. For ERCP, studies using aspiration, brush, and lavage cytology were included. The main objective was the diagnostic yield (pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio) of cytology obtained from ERCP to distinguish malignant and benign IPMN. RESULTS: Meta-analysis of 13 international studies with 483 IPMN patients was conducted. Pooled sensitivity was 35.1%, specificity 97.2%, and accuracy 92.9%. Among the three ERCP techniques, lavage cytology showed the best diagnostic ability (sensitivity 45.8%, specificity 97.9%). Malignant IPMN were observed in 45.1% (218/483) of patients in ERCP studies. CONCLUSIONS: Cytology from ERCP has good specificity but poor sensitivity in distinguishing benign from malignant IPMN. Newer techniques or markers are needed to improve diagnostic yield.
BACKGROUND AND AIM: Published studies have revealed the diagnostic yield of cytology obtained from endoscopic retrograde cholangiopancreatography (ERCP) in distinguishing malignant and benign intraductal papillary mucinous neoplasm (IPMN). However as a result of small sample sizes, the overall magnitude of benefit is unknown. Additionally, the optimal endoscopic procedure for cytology acquisition is also unclear. The aim of the present study was to evaluate the diagnostic yield of ERCP-based cytology in patients with IPMN and clarify the optimal sampling technique. METHODS: Relevant studies with a reference standard of definitive surgical histology were identified via MEDLINE and SCOPUS. Malignant IPMN included invasive adenocarcinoma, carcinoma in situ, and high-grade dysplasia. For ERCP, studies using aspiration, brush, and lavage cytology were included. The main objective was the diagnostic yield (pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio) of cytology obtained from ERCP to distinguish malignant and benign IPMN. RESULTS: Meta-analysis of 13 international studies with 483 IPMN patients was conducted. Pooled sensitivity was 35.1%, specificity 97.2%, and accuracy 92.9%. Among the three ERCP techniques, lavage cytology showed the best diagnostic ability (sensitivity 45.8%, specificity 97.9%). Malignant IPMN were observed in 45.1% (218/483) of patients in ERCP studies. CONCLUSIONS: Cytology from ERCP has good specificity but poor sensitivity in distinguishing benign from malignant IPMN. Newer techniques or markers are needed to improve diagnostic yield.