External quality assurance in nongynecologic cytology: The Australasian experience

Our plain-English summary of External quality assurance in nongynecologic cytology: The Australasian experience, by P.W. Shield, F. Frost, J.L. Finnimore, et al (Cancer Cytopathology, 2017).

The study

• Authors: P.W. Shield, F. Frost, J.L. Finnimore, R.G. Wright, M.C. Cummings
• Journal: Cancer Cytopathology
• Published: 2017;125(5):349-361
• Cited by: 5 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Background: The Royal College of Pathologists of Australasia Cytopathology Quality Assurance Program has operated an external quality assurance program in nongynecologic cytopathology since 1993. Glass slide preparations of a wide range of nongynecologic cases were circulated to approximately 200 cytopathology laboratories in 16 countries.

Methods: General nongynecologic cytology cases were manufactured from residual specimens after routine diagnosis. Fine-needle aspiration (FNA) cases were made by sampling fresh tissue and making direct specimens. The majority of cases consisted of both air-dried and fixed preparations. Results returned to laboratories included illustrated case discussions highlighting diagnostic features, key differential diagnoses, and useful adjunctive tests.

Results: The current study reviewed >22,000 results for 123 nongynecologic cases. Cases found to cause the most diagnostic difficulties included serous effusion cases with metastatic carcinoma in a dispersed pattern, well-differentiated carcinoma, and cellular reactive cases; urine specimens with sparse malignant cells; reactive pneumocytes in a bronchoalveolar lavage; breast FNA cases with papillary lesions; gestational specimens; and fibroadenoma. FNA specimens from the lung and thyroid, particularly papillary thyroid carcinoma, generally were well reported.

Conclusions: The use of multiple preparations of the same specimen has allowed interlaboratory comparison, and the quality assurance program has played an educational role as well as informing the laboratory accreditation process. Cancer Cytopathol 2017;125:349-361. © 2017 American Cancer Society.

© 2017 American Cancer Society.

In plain English

This article reviewed an external quality assurance program run by the Royal College of Pathologists of Australasia. The program sent identical glass slide test cases of nongynecologic cytology to about 200 laboratories in 16 countries. The review covered 1993 to 2010: 123 cases and more than 22,000 laboratory responses. It did not study asbestos exposure. It measured how often laboratory reports were consistent with or suspicious for the expected diagnosis, or listed it as a possibility. For 27 serous effusion (body cavity fluid) cases, this expected-response rate ranged from 49% to 98%. The 4 mesothelioma cases had expected-response rates of 54% to 92%, and in every one some laboratories read the tumor as metastatic adenocarcinoma. In one mesothelioma case made up mainly of single cells, 7% of laboratories called the tumor cells benign reactive cells and 31% reported adenocarcinoma. Other difficult cases included a reactive pleural fluid, where 34% of responses were false positives, and a bronchoalveolar lavage, which 59% of laboratories reported as malignant or suspicious. The authors concluded that multiple preparations allowed interlaboratory comparison and that the program was educational and informed laboratory accreditation.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Descriptive review of an external quality assurance (proficiency testing) program's glass slide survey results from 1993 to 2010. The main measure was the expected-response rate. It counted reports that were consistent with or suspicious for the expected response, or that included it in a differential diagnosis.
• Who was studied: Cytopathology laboratories enrolled in the RCPA Cytopathology Quality Assurance Programme between 1993 and 2010. There were about 200 laboratories in 16 countries, mostly in Australia and New Zealand. General cytology enrollments were 175 to 199 laboratories per year, and FNA enrollments were 180 to 205. In total, 123 test cases generated more than 22,000 laboratory responses.
• Exposure: Not applicable. The article does not study asbestos or any other exposure. Laboratories reported on circulated test slides, which included 4 mesothelioma cases among the serous effusion specimens.
• Main results:
  • For the 27 serous effusion cases, expected-response rates ranged from 49% to 98% (median 77%).
  • The 4 mesothelioma cases ranged from classic to single cell presentations and had expected-response rates of 54% to 92%.
  • In every mesothelioma case, some laboratories interpreted the tumor as metastatic adenocarcinoma. This was not counted as a false negative, but it was not the expected response.
  • In one mesothelioma case made up mainly of single large malignant mesothelial cells, 7% of laboratories called the tumor cells benign reactive cells and 31% reported adenocarcinoma.
  • A cellular reactive pleural fluid specimen from a patient with a history of breast carcinoma drew 34% false-positive responses.
  • In a peritoneal fluid case of well-differentiated metastatic colonic adenocarcinoma, 32% of laboratories reported the case as benign.
  • Two urothelial carcinoma urine cases with sparse malignant cells had false-negative rates of 8% and 16%.
  • For a bronchoalveolar lavage with type 2 pneumocyte hyperplasia (a benign case), about 59% of laboratories reported the finding as malignant or suspicious.
  • A breast FNA of high-grade ductal carcinoma from a pregnant woman was circulated twice, 6 years apart. Expected-response rates were 80% and 81%, and false-negative rates were 16% and 16.5%.
  • For the 3 classic papillary thyroid carcinoma cases, there was 1 false-negative result among 558 reports (0.2%).
  • A benign fluid case shown by virtual microscopy had a 73% false-positive rate, compared with 33% when the same case was circulated on glass slides.
• The authors' conclusion: The authors concluded that using multiple preparations of the same specimen allowed comparison between laboratories. They also concluded that the program played an educational role and informed the laboratory accreditation process.
• Limitations the authors state:
  • Manufacturing cases was resource-intensive and depended on donor laboratories and individuals.
  • Some specimen types, such as cerebrospinal fluid, never have enough volume for this approach.
  • Frequent false-positive results may partly reflect an examination artifact of the testing procedure.
  • Clinical notes and the testing environment appeared to influence results in some cases.
  • The poor results on benign urinary cases may reflect case selection.
  • The system could not monitor trends in reporting accuracy over time.

Topics

Medical terms in this study, explained in our Lexicon: Cytology, Cytopathology, Fine-Needle Aspiration, Carcinoma, Well-Differentiated, Bronchoalveolar Lavage (BAL).

Related studies

• Categorical systems for reporting of cytology specimens: Following the footsteps of Bethesda-like reporting systems — L.J. Layfield, Z. Baloch (2020)
• Ultrastructural studies of malignant cells in fluids — G.A. Herrera, J.A. Wilkerson (1985)
• The role of liquid-based cytology and ancillary techniques in pleural and pericardic effusions: an institutional experience — E.D. Rossi, T. Bizzarro, et al. (2015)
• Investigating unilateral pleural effusions: the role of cytology — D.T. Arnold, D. De Fonseka, et al. (2018)
• The marriage of cytology and cytogenetics — P. Dal Cin, X. Qian, et al. (2013)
• Electron microscopy in the diagnosis of percutaneous fine needle aspiration specimens — E.J. Wills, S. Carr, et al. (1987)

Browse more research in Cancer Cytopathology, or return to medical research.