Preparation technique for cytologic study of EUS-guided FNA samples when onsite cytopathologist is absent: Cell block, smear, or both?

Abstract

176 Background: Smearing EUS-guided fine-needle aspiration (FNA) samples, with or without cell block, is most commonly performed for cytologic study in centers without onsite cytology. The diagnostic yield of such technique is often low (35%-50%) and may be related to loss of diagnostic material during the smearing process. Cell block technique captures not only aspirated cells but also small tissue fragments, improving diagnostic yield and enabling additional ancillary studies. This study aimed to compare the diagnostic yield of cell block alone against conventional smear (±cell block) technique in the EUS-guided FNA evaluation of pancreatic solid mass without onsite cytological assessment. Methods: Data on the diagnostic yield of cell block alone technique performed over the last 12 months were prospectively collected and were compared to the yield of conventional smear ± cell block technique. A positive diagnostic yield was defined by the presence of adenocarcinoma and/or FNA findings consistent with the final surgical diagnosis. Results: Data were available for 96 patients, of which 66 had cell block alone and 30 had smear +/-cell block preparation. The diagnostic yield of cell block alone was significantly higher that that of smear +/- cell block (53/66 vs 13/30; P=0.003). The addition of cell block after smearing did not improve the diagnostic yield (smear: 42% vs. both: 45%, p=0.88). Of patients who had smear +/-cell block, the diagnostic yield was significantly higher when the smearing was performed by cytopathologist as compared to that by nurse (10/16 vs. 3/14, p=0.02). There was a direct relationship between the rate of diagnostic yield and the number of FNA passes performed. The yield was highest when 4 or more FNA passes was performed (85%) as compared to that of 3 passes (63%, p=0.049) and 2 or less passes (38%, p<0.0001). Conclusions: In the absence of onsite cytology services, the diagnostic yield from cell block alone was high (80%) and was superior to smear technique whether cell block was added to the evaluation of smears. In order to obtain adequate tissue material to maximise the diagnostic yield EUS-guided FNA of pancreatic mass, at least 4 FNA passes should be performed. No significant financial relationships to disclose.