INTER–OBSERVER CONCURRENCE (OR LACK THEREOF) ON CONVENTIONALLY – STAINED SMEARS ASPIRATED FROM PATIENTS OF ROSAI – DORFMAN DISEASE

Abstract

Objective: To determine the degree of inter-observer reproducibility of results in fine-needle aspirated material in patients with Rosai-Dorfman disease (RDD), also known as sinus histiocytosis with massive lymphadenopathy (SHML).

Study design: Case Series.

Place & duration of study: Departments of Pathology of eleven Armed Forces medical institutions, inland and abroad, April 25, 2004 through April 30, 2016.

Materials and Methods: Three qualified cytopathologists independently opined on all aspirates from all RDD patients in fine-needle aspirate (FNA) clinics of their respective canters. All smears examined by any one worker were circulated amongst both the other. These were all followed by confirmation on biopsy (the “gold standardâ€). Those that were diagnosed SHML, on examination of either aspiration smears or on histopathology sections, were selected for this project. As a routine, all our fine-needle aspirates are: (1) subjected to fixation in absolute alcohol, to be stained with Papanicolaou method as well as haematoxylin-and-eosin (H&E) and (2) dried in air, to be dyed with Romanovsky stains.

After the diagnosis was histologically confirmed, comparisons were drawn between the cytology verdicts of all three assessors

Results: A total number of 13 patients formed the size of our sample. Out of these, 11 (85%) were diagnosed correctly by all three workers. Of the remainder, 1 (8%) case had been missed by all three examiners, while two of them overlooked another separate case each. Hence the overall agreement in diagnosis turned out to be 92% between the first author and either of the second and third, whereas it was 85% between the second and third as well as between all three. The dearth or absolute lack of emperipolesis (earlier thought of as lymphophagocytosis) was found to be the most important “confounding†factor.

Conclusion: For a large examiner-to-sample ratio (3:13) as this, the agreement in cytological diagnosis of RDD is strong. With a greater number of cases and the application of immunohistochemical tools, FNA diagnosis is expected to bring in a higher degree of agreement. With fully-developed emperipolesis, FNA can be relied on for an accurate diagnosis of SHML.

Key words: Rosai-Dorfman disease, Inter-observer concurrence, Conventionally-stained smears, Cytological examination, Emperipolesis.