Background: The cytological assessment of endometrium entails: (1) a collection device that is easy to use, collects material only from the endometrium, and obtains adequate samples; (2) a fixation process that ensures the preservation of individual cells, maintains cell polarity, and allows the recognition of three-dimensional tissue structures (microbiopsies); and, (3) interpretative algorithms that translate histopathologic to cytopathologic diagnoses. The purpose of this study is to show that it is feasible to achieve these ends when endometrial brush sampling is coupled with suspension fixation.
Methods: Obtain endometrium from 100 consecutive hysterectomy uteri using an Indiana University Medical Center sampler (Tao Brush), suspension-fix it in CytoRich fixative, prepare it with a cytocentrifuge using large diameter sample chambers, and compare the cytologic diagnosis to the histologic diagnosis of the hysterectomy specimen.
Results: There were no inadequate collections. Cytology regularly separated (1) benign endometrium, (2) low-grade (non-atypical) hyperplasia, (3) high-grade (atypical) hyperplasia/FIGO Grade I adenocarcinoma, and (4) higher-grade carcinomas from one another. Endometrial atrophy was diagnosed in three patients whose histology showed clinically asymptomatic, benign fibrous endometrial polyps. A low volume of abnormal cell aggregates interpreted as endometrial intraepithelial carcinoma was detected in one patient whose initial histology was reported as simple hyperplasia, but whose histology on review after p53 staining revealed intraepithelial surface cancer. In the remaining 96 cases, the cytologic diagnosis consistently represented the histologic diagnosis of the hysterectomy specimen. On a case-by-case basis, any one cytology slide accurately represented the diagnosis of the other cytology slides.
Conclusion: Endometrial brushing with suspension-fixation is advocated for the detection of endometrial lesions because (1) fixation is uniform, (2) there is substantial preservation of three-dimensional structures among cell aggregates, which allows pattern-based histologic diagnostic criteria to be applied to cytologic samples, and (3) only a limited number of slides need to be examined.