Introduction/Background Though atypical endometriosis (AE) is hypothesized to be a precursor of endometriosis associated ovarian carcinoma (EAOC), its identification has no established clinical consequences and is therefore not structurally embedded in histopathological examination of endometriosis. The aim of this study was to determine whether AE is diagnosed more often when pathologists use a clear set of criteria for atypical features compared to common practice.

Methodology All cases of ovarian endometriosis, collected at the Radboud University Medical Centre Nijmegen between 1985 and 2017, were identified using the PALGA database. Pathology reports were searched for of atypical features and the notion of AE by the pathologist. Using a set list of criteria for AE, a total of 293 samples were revised by two independent pathologists to assess atypical features in the samples.

Results A total of 949 cases of ovarian endometriosis were included in the database. AE was documented in 105 cases . After revision of 266 endometriosis samples, the pathologist identified a total of 48 samples with AE. In contrast, in the original pathology report, only 31 cases of AE were identified, with 18 cases where AE was mentioned in the microscopy report, but not the concluding section. The agreement of the revising pathologist with the original report regarding AE occurrence was minimal.

Conclusion This study shows that AE is not properly identified in common practice. Furthermore, AE diagnosis may depend on the criteria used and the alertness of the pathologist and inter-observer variability may be a pitfall for proper identification. As AE has been identified in association with malignant transformation, it is of utmost importance to mark its occurrence. With this study we want to press the importance of adequate identification and documentation of atypical features in AE in order to move towards clinical implications for AE in order to aid EAOC prevention.