Role of crush cytology in detecting gastrointestinal malignancies

Ayaskanta Singh; Rashmi Patnayak; Jimmy Narayan; Manoj Kumar Sahu; Manas Kumar Behera; Amitabh Jena

doi:10.4103/ghep.ghep_1_23

ORIGINAL ARTICLE

Year : 2023 | Volume : 3 | Issue : 2 | Page : 56-60

Role of crush cytology in detecting gastrointestinal malignancies

Ayaskanta Singh¹, Rashmi Patnayak², Jimmy Narayan¹, Manoj Kumar Sahu¹, Manas Kumar Behera¹, Amitabh Jena³
¹ Department of Gastroenterology and Biliary Sciences, Institute of Medical Sciences and SUM Hospital, Bhubaneswar, Odisha, India
² Department of Pathology, Institute of Medical Sciences and SUM Hospital, Bhubaneswar, Odisha, India
³ Department of Surgical Oncology, Institute of Medical Sciences and SUM Hospital, Bhubaneswar, Odisha, India

Date of Submission	01-Jan-2023
Date of Decision	14-Jan-2023
Date of Acceptance	22-Jan-2023
Date of Web Publication	09-Mar-2023

Correspondence Address:
Rashmi Patnayak
Department of Pathology, Institute of Medical Sciences and SUM Hospital, Bhubaneswar, Odisha
India

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/ghep.ghep_1_23

Abstract

Background: Adenocarcinomas are the most common malignancy of the gastrointestinal (GI) tract. Crush cytology is an effective method which can be used to detect neoplastic conditions of GI tract, especially in combination with biopsies. Materials and Methods: The objectives of the study were to study the efficacy of crush cytology as a convenient and near accurate method to evaluate endoscopic biopsy of GI neoplasms. Study Design: Retrospective and Prospective. The original cytopathology diagnoses were correlated with histology report on cases sent over 2-year period. In consecutive 89 patients attending the department of gastroenterology with clinical suspicion of malignancy, an endoscopy was performed. The material obtained was subjected to cytology as well as histopathology. The stained cytology slides along with corresponding histopathology slides were studied. Results: Out of 89 cases, both crush smear and histopathology sections were positive for malignant cells (65 cases), both negative for malignant cells (seven cases), crush smear positive and histopathology negative (three cases), and crush smear-negative and histopathology positive (14 cases). Out of the 14 cases which were negative in crush smears for malignancy, on histopathological examination, eight were signet-ring cell carcinoma, three were poorly differentiated adenocarcinoma, two turned out to be neuroendocrine carcinoma, and one non-Hodgkin lymphoma (NHL). Conclusion: Crush cytology smears can be used to diagnose malignant GI cases fairly accurately. Cases of signet-ring cell carcinoma, neuroendocrine tumor, poorly differentiated malignancy, and NHL may be missed by crush cytology technique alone. A special stain for mucin can be used to identify the signet-ring cells in signet-ring cell carcinoma cases.

Keywords: Crush cytology, endoscopic biopsy, gastrointestinal malignancy, signet-ring cell carcinoma

How to cite this article:
Singh A, Patnayak R, Narayan J, Sahu MK, Behera MK, Jena A. Role of crush cytology in detecting gastrointestinal malignancies. Gastroenterol Hepatol Endosc Pract 2023;3:56-60

How to cite this URL:
Singh A, Patnayak R, Narayan J, Sahu MK, Behera MK, Jena A. Role of crush cytology in detecting gastrointestinal malignancies. Gastroenterol Hepatol Endosc Pract [serial online] 2023 [cited 2023 Mar 27];3:56-60. Available from: http://www.ghepjournal.com/text.asp?2023/3/2/56/371274

Introduction

Worldwide gastrointestinal (GI) malignancies are a major cause of health problem.^[1] The most common malignancy of the GI tract is adenocarcinomas.^[2] In India, gastric cancer is the most common GI malignancy followed by colon cancer.^[3] Endoscopy with a proper biopsy follow-up remains the standard for early detection of GI cancer and related premalignant lesions.^[1] Crush cytology of GI tract lesions can be used to diagnose neoplastic and nonneoplastic conditions in a fairly accurate manner, especially in combination with biopsies. Cytology is a cost-effective method. It allows rapid interpretation and triaging of material.^[2],[4],[5] Furthermore, cytology has a shorter turnaround time and is potentially cheaper when compared to histopathology of biopsies.^[6]

In this study, the reliability of crush cytology is evaluated by correlating cytological diagnoses with that of histopathology.

Materials and Methods

Patients included in the study group with clinical suspicion of malignancy underwent endoscopic biopsy procedures. These patients attended the outpatient department of gastroenterology. In 89 consecutive patients, both crush smear and histopathological evaluation of the biopsy were done.

All the patients, suspected to have GI malignancies, on upper or lower endoscopies were included in the study. The clinical profile and investigations of each patient were noted. Olympus CV-150 series were used to do the upper GI endoscopies and colonoscopies. At endoscopy, the site, extent, appearance, and mucosal friability were noted. The mucosal lesions were characterized either as ulcerative or polypoidal growths. Biopsy specimens were taken using disposable forceps (Olympus EndoJaw) that had a swing jaw, which was oval and fenestrated with needle. The diameter of the forceps was 2.45 mm. Biopsy samples were taken either, from the surface, especially in the polypoidal masses or from the margin, especially in the ulcerative lesions. Six biopsy specimens were taken, two of which were randomly selected for crush cytology preparation. The rest four samples were taken in 10% neutral buffer formalin for histopathological examination.

The biopsy material obtained was analyzed by both crush smear technique and histopathology. The crush smear was made by gentle crushing the tissue between the two glass slides to generate a uniformly spread smear. The smears were fixed in 95% alcohol and stained by hematoxylin and eosin stain. The air-dried smears were stained by Diff-Quik stain. Simultaneously, endoscopic biopsy tissues fixed in 10% neutral buffer formalin were subjected for histopathological examination. Hematoxylin and eosin-stained slides cut from the paraffin block were analyzed. Alcian blue-Periodic acid–Schiff special stain highlighted the signet-ring cells in signet-ring cell carcinomas. In poorly differentiated carcinomas, immunohistochemical analysis with CK7, CK 20, and leukocyte common antigen was done. In case of non-Hodgkin lymphoma (NHL), lymphoma panel was used to establish the diagnosis. For neuroendocrine tumors, immunohistochemical markers such as chromogranin, synaptophysin, neuron-specific enolase, and Ki-67 were done.

The results of both crush cytology and histopathology were compared. The crush smears and histopathology slides were evaluated by experienced pathologists. The terminologies used for the reporting of crush cytology smears were positive for malignancy, suspicious for malignancy, possibly reactive atypia of unknown significance, and negative for malignancy. For statistical analysis, both positive for malignancy and suspicious for malignancy were clubbed under the category of positive for malignancy.^[2],[4] Smears showing possible reactive atypia and negative for malignancy were grouped under negative for malignancy cases. Smears showing numerous singly scattered malignant cells in the background with nuclear features such as round to irregular nuclear contours, irregularly thickened nuclear borders, and at least some hyperchromatic nuclei were considered positive for malignancy.^[2],[4] The World Health Organization (WHO) classification was followed for histopathological diagnoses.

Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated by comparing the histopathology report. Descriptive statistics were used to analyze the data.

Results

Altogether consecutive 89 cases were included in the study period. For all these cases, in addition to the crush smear simultaneous biopsy specimens were obtained. They were divided into four categories such as both crush smears and histopathology sections positive for malignant cells (65 cases), both negative for malignant cells (seven cases), crush smear positive and histopathology negative (three cases), and crush smear negative and histopathology positive (14 cases) [Table 1]. Out of the 14 cases which were negative for malignancy in crush smears and positive for malignancy on histopathology examination, eight were signet-ring cell carcinoma, two turned out to be neuroendocrine carcinoma, three were poorly differentiated adenocarcinoma, and one NHL [Table 2] and [Figure 1].

Table 1: Correlation between diagnosis of gastrointestinal lesions on crush cytology and histopathology examination (n=89)

Click here to view

Table 2: Details of crush smear negative (false negative) cases

Click here to view

Figure 1: (a) Crush smear of signet-ring cell carcinoma showing individually scattered cells, few with eccentric nuclei (Diff-Quik ×200). (b) Crush smear of poorly differentiated carcinoma showing high cellularity, neoplastic cells present in necrotic background (Diff-Quik, ×100). (c) Crush smear of poorly differentiated carcinoma showing pleomorphic neoplastic cells with high N:C ratio (H and E, ×400). (d) Crush smear of neuroendocrine tumor showing acinar pattern of arrangement of tumor cells (H and E, ×200)

Click here to view

Considering histopathology examination as gold standard, the sensitivity, specificity, PPV, NPV, and accuracy of crush smear test (reference test in this study) can be calculated as follows: sensitivity –82% (69%–90%), specificity –70% (30.4%–94%), PPV –96%, NPV –33%, and accuracy –65%.

Discussion

Adenocarcinoma is the most common malignant lesion in the GI tract.^[2] In GI malignancy clinical, imageological and serologic correlation is required for proper diagnosis. However, for definitive diagnosis, cytology and histology sampling with morphologic evaluation of lesional tissue is important before initiation of treatment.^[4]

Gastric adenocarcinomas account for 90%–95% of gastric malignancies. In our study, also majority were adenocarcinomas. They are classified according to the WHO classification.^[4] In well-differentiated adenocarcinoma, usually, loosely cohesive three-dimensional groups of neoplastic cells with occasional single scattered cell are seen in the background.^[4] The majority of our cases were adenocarcinomas with typical cytology findings.

Cytology is a well-known screening as well as a diagnostic procedure. Cytology of GI tract lesions can be used to diagnose neoplastic and nonneoplastic conditions fairly accurately. In combination with biopsies, the diagnostic accuracy can increase manifold.^[6] Evaluation by cytology is a helpful method as it is cost-effective and it helps in rapid interpretation. With the advancement of technology, simultaneous visualization of abnormal tissue and procurement of tissue from mucosal and deeper-seated lesions is possible. For successful cytology examination of the GI tract, a good combination of the skill of the gastroendoscopist, specimen preparation, and the expertise of the pathologist is essential.^[4],[5],[7]

However, there are limitations of cytology technique. The absence of positivity for a malignant cell does not necessarily exclude malignancy.^[2] A similar observation is noted in our study. There were 14 cases, which were negative for malignancy in crush cytology but turned out to be malignant in histopathology examination of the endoscopic biopsy. Of these 14 cases, eight were signet-ring cell carcinoma. Hence, distinguishing signet-ring cells in cytology are important. In signet-ring cell carcinoma, typical signet-ring cells with hyperchromatic, eccentric nuclei, and large cytoplasmic mucin vacuoles are usually present. Some signet-ring cells may have bland nuclei and can be confused with histiocytes. Signet-ring carcinoma is difficult to detect on both cytologic and histologic specimens. Careful examination and a high degree of suspicion are essential to detect this carcinoma. Immunohistochemistry with keratin, epithelial membrane antigen, and mucin stains is helpful to differentiate the single tumor cells from histiocytes. Histiocytes express markers such as CD68 and KP-1.^[4]

The diffuse type of malignancy usually tends to be more infiltrative with less mucosal involvement. In these cases, there is a chance of a higher rate of false-negative diagnosis. The background is usually clean without apparent tumor diathesis. These specimens are less cellular, with a majority of single cells. In neuroendocrine tumors, the tumor cells are noncohesive and monomorphic, with stippled “salt and pepper” nuclear chromatin. These tumor cells have a moderate amount of granular cytoplasm and may have a spindle cell appearance. Many stripped and bare nuclei also are frequently present.^[2]

In our study, there were cases where crush smear was negative for malignancy, but on histopathology examination, the diagnosis was poorly differentiated adenocarcinoma and signet-ring cell carcinoma. In these cases, histopathology revealed that the tumor was present more in the subepithelial location. The neuroendocrine tumors which were missed in cytology were mainly of well-differentiated type.

Among our cases, there are three cases in which a diagnosis of suspicious for malignancy was given on crush cytology, whereas in histopathology these cases were negative. On review of both crush smear cytology slides and corresponding histopathology cases, it was noted that these cases were clinically suspected as carcinoma stomach and were ulcerative in nature. Studies have described conditions like ulcers, reparative change, inflammatory bowel disease, or other inflammatory conditions and infections mimic features of malignancy . They may show changes such as noncohesive cells with prominent nucleoli, increased mitotic figures which bias many a pathologist to give a diagnosis of malignancy.^[2],[8],[9] These are potential diagnostic pitfalls. Proper diagnosis is made by awareness of relevant clinical history, consultation with another GI pathologist, and examining the biopsy sections.^[2],[10]

In a similar study by Chaitra et al., the authors have evaluated the role of crush cytology in the diagnosis of large intestinal lesions correlating with histopathology. They have noted that inadequate report is the result of the sampling of nonviable tissue.^[2],[11] Sampling difficulty is also commonly encountered in those lesions which are located in the subepithelial region.^[12] The reported sensitivity and specificity in their study were 96% and 63.2%, respectively. The PPV was 91.1% and NPV 80%. In our study, sensitivity was 82% and specificity was 70%. The PPV and NPV were 96% and 33%, respectively. Another study by Batra et al. analyzed the efficacy of various cytology techniques including crush cytology. The diagnostic yield for crush cytology in their study was 81.5%. They opined that a combination of cytology and biopsy has a higher diagnostic yield.^[13] In another study by Tyagi et al. compared GI brushing samples with histopathology. They encountered discordance in 21.8% of cases. The explanation for this was inadequacy of the cytology sample and overlapping of nuclear atypia caused by regenerative changes and malignancy.^[14] These problems were also encountered in our study.

Rout et al., in their evaluation of cytology technique in case of gastroesophageal malignancy, obtained 98.7% positive yield in combining histology and cytology which was more than evaluation by cytology alone. They opined that squash cytology is a useful adjunct to conventional endoscopic biopsy for gastroesophageal malignancy. The advantage is requirement of no additional effort or equipment to add to the diagnostic yield besides helping the workup.^[6]

Saha et al. analyzed 60 cases of gastric lesions and 57 cases of colonic lesions. By cytology method alone accuracy was 73.3% for gastric lesions and 89.5% for colonic lesions. By combination of two methods, the accuracy was 95% and 96.5% in detecting gastric and colonic malignancies, respectively. They also were of the opinion that crush smear cytology is a cheap, easy technique with the rapid result. The diagnostic yield improved significantly when the technique is combined with histopathology. It may be used as a useful adjunct to conventional histopathology.^[15] Our study also highlights the above-mentioned facts.

Newer techniques such as liquid biopsies, mainly represented by circulating tumor cells, circulating tumor DNA, tumor exosomes, and microRNAs, have the potential to assess various biomarkers for early detection of cancer, carrying out genomic/immune profiling for not only selection of appropriate therapy but also to monitor effect of therapy.^[16]

In our subcontinent setup, crush cytology can be used fairly accurately in a cost-effective method for the diagnosis of GI malignancies.

As always clinical, serologic, and radiologic correlation is essential. Multidisciplinary discussions should be performed before institution of definitive treatment.^[4]

Limitation of the study

The cases included in our study were clinically suspected as GI neoplasms. As reflected by our results, the majority turned out to be malignant by both crush and histopathology examination. Crush cytology will add to the cost of the diagnosis.

This study has several beneficial points like crush smear diagnosis can be obtained on the same day as the endoscopy procedure. It will save time as an early consultation with surgical or medical oncologists and appropriate treatment plan can be availed by the patient.

Conclusion

Crush smear cytology can help to arrive at a diagnosis fairly accurately and in less time compared to histopathology. The lesions which are usually missed in crush cytology are signet-ring cell carcinoma, neuroendocrine tumors, poorly differentiated malignancy, and NHL cases. Pathologists should keep the possibility of these lesions in mind before commenting a crush cytology smear as negative for malignancy.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Pasechnikov V, Chukov S, Fedorov E, Kikuste I, Leja M. Gastric cancer: Prevention, screening and early diagnosis. World J Gastroenterol 2014;20:13842-62.
2.	Chaitra GV, Saha D, Yadav R, Adiga DS, Lobo FD, Ghosh A, et al. The role of crush cytology in the diagnosis of large-intestine lesions with correlation on histopathology. Acta Cytol 2018;62:215-22.
3.	Rana S. Prevalence of gastrointestinal cancers in India. In: Gandhi V, Mehta K, Grover R, Pathak S, Aggarwal B, editors. Multi-Targeted Approach to Treatment of Cancer. Basel: Springer; 2015. p. 217-31.
4.	Conrad R, Castelino-Prabhu S, Cobb C, Raza A. Role of cytopathology in the diagnosis and management of gastrointestinal tract cancers. J Gastrointest Oncol 2012;3:285-98.
5.	Buchireddy D, Chakraborti S, Subba SH. Crush cytology of gastrointestinal malignancy: A cytohistologic comparison. Am J Clin Pathol 2012;138:A274.
6.	Rout N, Singh SP, Satpathy BK, Nanda BK. Rapid cytodiagnosis of endoscopic biopsy specimens in gastrooesophageal malignancy. Trop Gastroenterol 1993;14:99-103.
7.	Jhala N, Jhala D. Gastrointestinal tract cytology: Advancing horizons. Adv Anat Pathol 2003;10:261-77.
8.	Edwards BK, Ward E, Kohler BA, Eheman C, Zauber AG, Anderson RN, et al. Annual Report to the Nation on the Status of Cancer, 1975-2006, featuring colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. Cancer 2010;116:544-73.
9.	Yu GH, Nayar R, Furth EE. Adenocarcinoma in colonic brushing cytology: High-grade dysplasia as a diagnostic pitfall. Diagn Cytopathol 2001;24:364-8.
10.	Greenson JK, Odze RD. Inflammatory disorders of the large intestine. In: Odze RD, Goldblum JR, editors. Surgical Pathology of GI Tract, Liver, Billiary Tract and Pancreas. Philadelphia: Saunders; 2009. p. 355-94.
11.	Brouwer R, MacDonald A, Matthews R, Gunn J, Monson JR, Hartley JE. Brush cytology for the diagnosis of colorectal cancer. Dis Colon Rectum 2009;52:598-601.
12.	Bibbo M, Keebler CM. Alimentary tract. In: Bibbo M, Wilbur D, editors. Comprehensive Cytopathology. Philadelphia: Elsevier; 2015. p. 59-64.
13.	Batra M, Handa U, Mohan H, Sachdev A. Comparison of cytohistologic techniques in diagnosis of gastroesophageal malignancy. Acta Cytol 2008;52:77-82.
14.	Tyagi R, Kaur J, Kaur G, Selhi PK, Puri HK, Sood N. Cytohistological discordance on gastrointestinal brushings: Facts unfolded. Niger Med J 2016;57:299-302. [PUBMED] [Full text]
15.	Saha M, Hossain A, Bhuiyan SH, Islam MN, Chowdhury MS, Kumar SU. Role of crush smear cytology in the diagnosis of gastrointestinal malignancy. Mymensingh Med J 2014;23:496-502.
16.	Lopez A, Harada K, Mizrak Kaya D, Dong X, Song S, Ajani JA: Liquid biopsies in gastrointestinal malignancies: When is the big day? Expert Rev Anticancer Ther 2018;18:19-38.