Purpose The current gold standard for chronic endometritis (CE) diagnosis is immunohistochemistry (IHC) for CD-138. However, IHC for CD-138 is not exempt from diagnostic limitations. The aim of our study was to evaluate the reliability and accuracy of MUM-1 IHC, as compared with CD-138. Methods This is a multi-centre, retrospective, observational study, which included three tertiary hysteroscopic centres in university teaching hospitals. One hundred ninety-three consecutive women of reproductive age were referred to our hysteroscopy services due to infertility, recurrent miscarriage, abnormal uterine bleeding, endometrial polyps or myomas. All women underwent hysteroscopy plus endometrial biopsy. Endometrial samples were analysed through histology, CD138 and MUM-1 IHC. The primary outcome was to evaluate the diagnostic accuracy of MUM-1 IHC for CE, as compared with CD-138 IHC. Results Sensitivity and specificity of CD-138 and MUM-1 IHC were respectively 89.13%, 79.59% versus 93.48% and 85.03%. The overall diagnostic accuracy of MUM-1 and CD-138 IHC were similar (AUC = 0.893 vs AUC = 0.844). The intercorrelation coefficient for single measurements was high between the two techniques (ICC = 0.831, 0.761-0.881 95%CI). However, among CE positive women, MUM-1 allowed the identification of higher number of plasma cells/hpf than .80] vs 5.05 [SD 3.37]; p = 0.017). Additionally, MUM-1 showed a higher inter-observer agreement as compared to CD-138. Conclusion IHC for MUM-1 and CD-138 showed a similar accuracy for detecting endometrial stromal plasma cells. Notably, MUM-1 showed higher reliability in the paired comparison of the individual samples than CD-138. Thus, MUM-1 may represent a novel, promising add-on technique for the diagnosis of CE.