Utilization of Touch Imprint Cytology (TIC) of Liver Core Needle Biopsy Specimens

Abstracts cytology but not in other organs such as the pancreas. In this meta-analysis, we reviewed the frequency and cancer risk associated with atypical category trying to understand the heterogeneity among different studies. Materials and Methods: We searched PubMed and Scopus using the keyword: EUS-FNA and pancreas. We then selected studies focusing on solid pancreatic lesions. Studies that lacked data on frequency and outcomes of atypical category were excluded. The suspicious category was excluded also. The frequency and cancer risk associated with atypical category was calculated using Comprehensive Meta-Analysis program. Heterogeneity of studies was assessed using subgroup analysis and meta-regression under the mixed effect model. Factors included: onsite interpretation (ROSE), use of gold standard, the study type, size and site, and the inadequate and positive rates. Results: We identified 23 of 89 studies on solid pancreatic lesions with complete data on the frequency and 12 studies on outcomes of atypical category. The frequency of atypical category ranged from 0.8% to 14%, mean 7% (C.I., 6-8%) the risk of cancer associated with atypical diagnosis ranged 25 to 100%, mean 58% (C.I. 47-69%). There is significant heterogeneity among the studies (I-squared 75%, p <0.05). This heterogeneity is not explainable by ROSE, use of gold standard, the study type, size and site or the inadequate and positive rates. Conclusions: There is significant heterogeneity in studies reporting atypical category in the work-up of solid pancreatic lesions. This category rate in the pancreas is similar to that of the thyroid and breast but the risk of malignancy associated with it is higher. There is a high need for standardization of pancreatic EUS-FNA reporting. 161 Utilization of Touch Imprint Cytology (TIC) of Liver Core Needle Biopsy Specimens Ming Zhang, MD, PhD, Michael J. Carrozza, BS, Yajue Huang, MD, PhD Temple University Hospital, Philadelphia, Pennsylvania Introduction: Cytology plays an important role in the preoperative assessment of hepatic mass lesions by image guided fine needle aspiration (FNA). However, with the wide application of CT-guided core needle biopsy in the diagnosis of liver disease, more and more touch imprint cytology (TIC) of the core biopsy specimens are performed, for the purpose of on-site adequacy evaluation. The aim of the study is to evaluate the adequacy performance and the diagnostic value of TIC of liver core needle biopsy specimens. Materials and Methods: A retrospective study was performed to review all the CT-guided liver core needle biopsies (2010-2012) with on-site evaluation via TIC. The cytology and biopsy diagnoses were correlated and analyzed. The predictive values of TIC diagnosis (sensitivity and specificity) for malignancy including hepatocellular carcinoma (HCC) and adenocarcinoma were calculated. In addition, cases with discrepant cytology and surgical pathology diagnoses were further studied. Results: A total of 107 liver core needle biopsy specimens with TIC were studied, 95 were evaluated as adequate on-site (88.8%), 5 were sub-optimal (4.7%) and 7 were inadequate (6.5%). The sensitivity of TIC for the diagnosis of malignancy was 89.6% and the specificity was 91.3%. In addition, the sensitivity for the diagnosis of HCC or adenocarcinoma was 57.1% and 73.7%, respectively. The specificity reached 100% for HCC and 97.1% for adenocarcinoma. 17 cases were interpreted as “atypical cytology”, in which 7 cases were confirmed to be malignant by concurrent tissue biopsy. Although sampling error could lead to the false negative results, well differentiated HCC and neuroendocrine tumors most likely required imunohistochemical studies to reach the definite diagnosis. Conclusions: Our study showed that TIC of CT-guided liver core needle biopsy demonstrated excellent performance for the diagnosis of liver malignancy. The definite diagnosis of well differentiated HCC and neuroendocrine tumors could be challenging, if based on cytomorphology only. 162 PCR in Hepatic Space Occupying Lesions Reported as Granulomatous Inflammation/Tuberculosis Nalini Gupta, MD, DNB, Kusum Sharma, MD, Pranay Tanwar, MD, Adarsh Barwad, MD, Aman Sharma, MD, Arvind Rajwnashi, MD, FRCPath

S69 Postgraduate Institute of Medical Education and Research, Chandigarh, India Introduction: FNAC is frequently performed for evaluation of space occupying lesions (SOL) of the liver. Tubercular involvement of the liver is uncommon, but is a serious consideration in differential diagnosis of granulomatous conditions especially in endemic regions like India. Objective: To assess the role of PCR done on archival cytological material in diagnosing tuberculosis (TB) in cases reported as granulomatous inflammation/TB in liver SOLs. Material and Methods: A total of 17 cases of liver SOLs reported as granulomatous inflammation [nZ12] and TB [nZ5] were retrieved from the departmental records over a period of six years. Ultrasound-guided FNAC was performed and the smears were reviewed for the cytomorphological features. The air-dried smears stained with MGG from the archival material were assessed for the representative material in the form of epithelioid granulomas and giant cells. One/two MGG smears from each case were de-stained and the material was scraped to extract DNA. PCR for Mycobacterium tuberculosis was performed for amplification of 123 bp fragment of the IS6110 insertion element. Results: The age of the patients ranged from 3 to 61 years (median age- 41 years). There were 12 females and 5 males. The patients presented with solitary/ multiple liver SOLs. DNA could be extracted from 10/17 cases from archival MGG smears. PCR positivity was noted in 8/10 cases [including 4 AFB smear positive cases], confirming a diagnosis of tuberculosis. DNA could not be extracted from one smear positive case possibly due to predominance of necrosis. Conclusions: Cytomorphology alone may not be sufficient for differentiating various granulomatous lesions reported in Liver SOLs. DNA can be extracted from archival cytological MGG stained smears. PCR should be carried out if ZN-staining is negative in granulomatous lesions, especially when material has not been submitted for culture. 163 A Retrospective Review of 86 Pancreatobiliary Brush Samples with Routine Cytology and Fluorescence In-Situ Hybridization (FISH) Assessing Optimal Referral for FISH Testing Mark Wittchow1, Barbara D. Lietzau, BA, SCT(ASCP)2, Linda C. Trugillo, CT, HT(ASCP)2, Samantha J. Bigalke, BS, CT(ASCP)2, Richard Wittchow, MD2 1 Minnesota State University e Mankato, Minnesota; 2Gundersen Lutheran Medical Center, La Crosse Wisconsin Introduction: FISH is a useful adjunctive test to routine cytology in the evaluation of pancreatobiliary strictures. Detection of chromosomal aneuploidy by FISH improves sensitivity over routine cytology in the detection of malignancy in pancreatobiliary brush (PBB) cytology specimens. In the interest of cost-effective use of FISH, this retrospective study reviewed results of cytology and FISH performed on split PBB samples to assess optimal referral for FISH testing. Materials and Methods: 86 PBB samples were reviewed. 1/3 of each sample was processed for cytology and the remaining 2/3 referred for FISH utilizing DNA probes to chromosomes 3, 7 and 17 (Urovysion). A cytology diagnosis of malignant, atypical/suspicious or negative was rendered independent of FISH results. Medical records were reviewed to determine if a stricture was benign or malignant based on clinicopathologic and radiographic findings. For statistical analysis, atypical/suspicious diagnoses by cytology, or trisomy 7 by FISH are considered negative; polysomy (gain of two or more chromosomes 3, 7 or 17) by FISH is equivalent to malignancy. Trisomy 7 is nonspecific, occuring in benign and malignant conditions. Results: Clinicopathologic diagnoses and findings are tabulated in Table 1. Sensitivity/specificity for cytology and FISH were 16%/100% and 32%/ 100%, respectively. Polysomy was detected only in cases with atypical/ suspicious or malignant cytology. Polysomy was not observed when routine cytology was negative, in both benign and malignant strictures (Figures 1 and 2). FISH impacted the final diagnosis only in cases with atypical/suspicious cytology (Table 2). Conclusions: Detection of polysomy by FISH appears dependent on the presence of morphologically abnormal cells in the sample. FISH is most beneficial in cases with atypical/suspicious cytology, assisting differentiation