Aims Diagnostic yield of EUS-FNA for pancreatic lesions varies around 70-90%, affected
by many factors including tissue processing.In order to increase the diagnostic accuracy,
we evaluated the effect of histological examination of core tissues acquired from
EUS-FNA.
Methods A single-center observational study was conducted at a tertiary university hospital.
The patients who underwent EUS-FNA for pancreatic lesion using 22G EUS-FNA needle
were included only if a core was observed from the retrieved specimen. A total of
63 patients was enrolled. For liquid-based cytology, the specimen was directly expressed
into a liquid-based fixation medium for processing, and cellblock was also made for
the evaluation. The whitish core was withdrawn before any process for cytology was
made and was placed in a container of 10% neutral-buffered formalin fixative for the
histologic evaluation.
Results In total 63 patients, 61 patients were finally diagnosed as pancreatic cancer and
2 patients had benign lesions. Cytologic diagnosis showed sensitivity of 73.8 %, specificity
of 100.00%, accuracy of 74.6 %, positive predictive value (PPV) of 100.0 % and negative
predictive value (NPV) of 11.1 %. Histologic diagnosis showed sensitivity of 68.9
%, specificity of 100.0 %, accuracy of 69.8 %, PPV of 100 % and NPV of 9.5 %. Overall
diagnostic yield of EUS-FNA showed increased sensitivity of 86.9 %, specificity of
100.0 %, accuracy of 87.3 %, PPV of 100.0 % and NPV of 20.0 %. We also compared the
weight of the core and needle depth from the surface of the lesion between the histology
positive and negative group, but there were no significance.
Conclusions Histologic evaluation of core material obtained from EUS-FNA improved diagnostic
sensitivity, accuracy, and NPV, and histologic results had higher ratio of atypical
cell over negative result, providing more clinical information. Further studies with
prospective randomized trial is recommended to support our data.
