Comparison of plasma-thrombin, HistoGel, and CellGel cell block preparation methods with paired ThinPrep slides in the setting of mediastinal granulomatous disease

Abstract

Various cell block (CB) preparation methods are utilized by different laboratories, and not all laboratories perform CBs in tandem with ThinPreps (TPs). To compare the performance of different CB methods and their diagnostic value when used in conjunction with TP, we assessed the quantity and size of granulomas obtained from endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) of lymph nodes in the evaluation of granulomatous mediastinal disease. A retrospective analysis of mediastinal lymph node EBUS-TBNA specimens that detected granulomas at our institution was performed. A total of 264 specimens from 124 patients had a TP followed by a CB (either plasma-thrombin, HistoGel, or CellGel) prepared from the residual material in the PreservCyt vial. The number and size of granulomas on each preparation was assessed using digital software. Granulomas were detected only on the CB in 18.9% of cases and only on the TP in 5.3%. All 3 CB preparation methods showed significantly more and larger granulomas compared with the paired TP, with the plasma-thrombin and CellGel methods yielding more diagnostic material than the HistoGel method. In addition, the average number of granulomas (4.0 ± 0.4 versus 15.3 ± 1.1) and granuloma size (119.2 ± 3.2 μm versus 271.8 ± 7.3 μm) were significantly lower on TP compared with CB, respectively. Plasma-thrombin and CellGel CB preparation methods had a higher granuloma yield compared with the HistoGel method. Additionally, significantly more numerous and larger granulomas were present on CBs compared with TP slides. Therefore, solely relying on TP slide evaluation may unintentionally overlook larger tissue fragments obtained during needle aspirations.