The frequency and cancer risk associated with the atypical cytologic diagnostic category in endoscopic ultrasound-guided fine-needle aspiration specimens of solid pancreatic lesions: a meta-analysis and argument for a Bethesda System for Reporting Cytopathology of the Pancreas.

BACKGROUND: The atypical cytologic diagnostic category is ambiguous and presents a management problem for pathologists and clinicians. This meta-analysis reviewed the frequency and cancer risk associated with atypical diagnoses in endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) specimens of solid pancreatic lesions.

METHODS: PubMed and Scopus were searched using the keywords "EUS-FNA" and "pancreas." Articles were screened focusing on studies of solid lesions. Studies with information regarding the frequency and outcomes of atypical diagnoses were included; the "suspicious" category was excluded from the analysis. The frequency of atypical diagnoses and the associated risk were calculated using the Comprehensive Meta-Analysis software. The authors assessed whether the following factors explained the heterogeneity of the studies: rapid on-site interpretation; type of reference standard; the study type, size, and site; and the frequency of inadequate, atypical, and positive categories.

RESULTS: A total of 23 studies with complete data regarding atypical diagnoses were identified, 12 of which had complete data available regarding outcomes. The frequency of the atypical category ranged from 1% to 14% (mean, 5.3%; 95% confidence interval, 4.1%-6.9%). The risk of malignancy associated with an atypical diagnosis ranged from 25% to 100% (mean, 58%; 95% confidence interval, 47%-69%). There was significant heterogeneity noted among the studies (I-squared, 62%; P = .0004). The frequency of the atypical category and its associated risk were found to be correlated only with the frequency of the specimens being positive for malignancy.

CONCLUSIONS: The rate of atypical diagnoses of the pancreas is similar to that of the thyroid but the risk of malignancy is higher. Significant heterogeneity exists among the studies reporting atypical diagnoses. There is a need for standardization of the reporting and management of atypical diagnoses in EUS-FNA specimens from the pancreas.