A comparison of the prevalence of chronic endometritis determined by the use of different diagnostic methods in women with and without reproductive failure

Based on new methods of plasma cell assessment, the prevalence of chronic endometritis in women with various categories of reproductive failure may have been overestimated in earlier studies.

Volume 109, Issue 5, Pages 832–839


Yingyu Liu, M.Phil., Xiaoyan Chen, Ph.D., Jin Huang, Ph.D., Chi-Chiu Wang, Ph.D., Mei-Yung Yu, M.B.Ch.B., Susan Laird, Ph.D., Tin-Chiu Li, M.D., Ph.D.



To compare the prevalence of chronic endometritis (CE) when different diagnostic methods are used.


Prospective observational study.


University-affiliated hospital.


Four groups of women were studied, including women with proven fertility (Fertile; n = 40), unexplained recurrent miscarriage (RM; n = 93), recurrent implantation failure (RIF; n = 39), and infertile subjects undergoing endometrial scratch in a natural cycle preceding frozen-thawed embryo transfer (Infertility; n = 48).


Endometrial biopsy was performed precisely 7 days after LH surge (LH+7). Plasma cells were identified by means of traditional hematoxylin and eosin (HE) staining and by means of immunohistochemistry (IHC) for Syndecan-1 (CD138).

Main Outcome Measure(s)

Prevalence of CE.


The use of CD138 epitope was more sensitive than HE staining in identifying plasma cells. The use of plasma cell count per unit area had the lowest observer variability compared with cell count per ten randomly chosen high-power fields and cell count per section. Using this method, the prevalence of CE in women with RM, RIF, and Infertility were 10.8%, 7.7%, and 10.4%, respectively, not significantly higher than that of Fertile subjects (5.0%).


Using what may be a new method of plasma cell assessment, it appears that the prevalence rates of CE reported in many earlier studies may have been overestimated.

Clinical Trial Registration Number


Read the full text here.