on rapid on-site evaluation (ROSE) for adequacy of FNA from solid pancreatic lesions. The role of ROSE in FNA of pancreatic lesions is discussed, as is the triage of material for making smears and cell block preparation. Different techniques of cell block preparation are briefly mentioned. Pancreatic cystic fluid obtained from pancreatic cysts is triaged differently as compared to specimens obtained from solid pancreatic lesions. An algorithmic approach to the processing of pancreatic cystic fluid for molecular and biochemical assays and cytology is discussed. Proper specimen handling is crucial to the accurate interpretation of pancreatic FNA specimens. The methods used to process a sample depend on whether the aspirated sample is solid or cystic and the type of device used for sampling. ROSE has been shown to reduce the number of inadequate specimens and to improve specimen preparation. The details of the various cytological preparation methods available are described in numerous texts. Here we focus on providing a broad overview of specimen collection and processing as it relates to pancreatic FNA, with guidance to the reader based on published and personal experiences.
© 2020 S. Karger AG, Basel
Pancreatic lesions undergoing fine-needle aspiration (FNA) can either be solid or cystic, and this determines the specimen handling. Material aspirated from solid lesions is smeared onto the slides and any extra material aspirated goes for cell block preparation. In contrast, fluid aspirated from cystic lesions is spun down onto the slides and sent for biochemical analysis and molecular testing, details of which are discussed below.
Processing of Cytology Samples from Pancreatic Solid Masses
FNA samples from a solid mass can be prepared by using direct smears, or the sample may be collected directly into a preservative, such as a CytoLyt®, for ThinPrep® processing, or into the SurePath preservative.
Smears
Adequate smear preparation is critical to accurate interpretation. Techniques for adequate smear preparation have been well described in the cytology literature [1]. A good-quality smear shows the cellular material evenly and thinly spread out on the slide (Fig. 1). Factors causing difficulty in smear interpretation include a bloody smear, a thick smear, and crush artifact (Fig. 2). It is recommended that large clots or tissue fragments be lifted from the smear and submitted for cell block analysis. Thick tissue fragments or clots will not be well visualized on smears.
Fig. 1. A well-spread smear – the aspirated material is evenly smeared making it easy to assess the cytological features of the cells. Diff-Quik stain, ×10.
Fig. 2. A thick smear – the aspirated material is entangled in a blood clot making it difficult to assess the cytological features of the cells. Diff-Quik stain, ×10.
Rapid On-Site Evaluation
For FNA of solid pancreatic lesions, rapid on-site evaluation (ROSE) has become an integral part of the procedure in many hospitals. Some studies recommend that a cytopathologist/cytotechnologist be present during the procedure for ROSE [2]. There have been several studies suggesting that there is an increase in diagnostic yield and a decreased need for repeat endoscopic ultrasound-guided (EUS)-FNA with ROSE [3–8]. ROSE is said to reduce the probability of false negative and unsatisfactory aspirations. ROSE with EUS-FNA has been reported to provide more accurate diagnoses than EUS-FNA alone [9–11]. ROSE is useful for three different reasons: (i) it helps in the assessment of the adequacy of aspirates to ensure that the lesion is appropriately sampled, (ii) it helps triage the specimen acquired appropriately, based on the initial impression after examining the slides, for additional studies such as flow cytometry, microbiology, or cell block for immunohistochemical stains, and (iii) it provides a provisional diagnosis to the echoendoscopists [11]. Based on this information, a decision is made as to whether additional passes are required or the material collected is sufficient for diagnosis. However, in recent years there have been many studies which have questioned the added benefit of ROSE. Few studies have concluded that the diagnostic yield in unassisted EUS-FNA is not inferior to ROSE-assisted EUS-FNA if at least seven passes are made from the pancreatic lesion [12–15]. A meta-analysis review concluded that ROSE does not improve EUS-FNA adequacy in pancreatic masses [16]. Hence, routine use of ROSE in EUS-FNA at tertiary cancer centers may not change clinical outcomes. This study also concluded that, since ROSE is a time-consuming service with poor reimbursement and is not available in many centers, it should not be strongly recommended for all EUS-FNA procedures for pancreatic lesions [16]. Non-assisted EUS-FNA procedures incur lower costs and a comparable rate of complications.
Performing ROSE
A staff member from the clinical team/cytotechnologist/cytopathologist assists in specimen collection during the procedure and with smear preparation in the procedure room or suite. At our institution two slides are prepared from each pass. One slide is air dried for Diff-Quik staining and the second slide is fixed in 95% alcohol for later Papanicolaou (PAP) staining in the laboratory. The needle is rinsed in balanced salt solution for cell block processing. The Diff-Quik-stained slide is used for on-site morphological evaluation of adequacy. Additional needle passes are performed based on the assessment of the Diff-Quik-stained smears, and material collected for additional ancillary studies based on the on-site adequacy.
Stains
The stains used for on-site adequacy depend on the institution. Some centers use an alcohol-fixed slide and stain it with rapid hematoxylin and eosin (HE). Other centers may use a rapid PAP or a different type of Romanowsky stain [17]. PAP staining is the standard at most institutions for the alcohol-fixed slides and is used for the final interpretation.
Collection Media
If ancillary studies are needed, additional dedicated passes for cell block material are made. Different types of collection mediums, such as saline, Roswell Park Memorial Institute (RPMI), neutral-buffered formalin, a mixture of formalin and 95% ethanol, and methanol, are available and can be utilized for needle rinses. Specimens collected in saline and RPMI can be used for flow cytometry if needed. Needle rinses can be processed into cell blocks or cytospin slides if the material collected is scant. Needle rinses collected in preservatives containing alcohol (methanol in PreservCyt and ethanol in SurePath Preservative Fluid) can be processed using ThinPrep® and SurePathTM respectively. Although ThinPrep preparation is being used for pancreatic FNAs, there are certain pitfalls and limitations to it. It is difficult to recognize mucin on ThinPrep slides and mucinous neoplasms may be incorrectly diagnosed. A study by de Luna et al. [18] concluded that the diagnostic accuracy of pancreatic