Background: Assisted reproduction technology (ART) has become a standard treatment for infertile couples. Increased success rates obtained over the years have resulted primarily from improved embryo quality, but implantation rates still remain lower than expected. The uterus, an important player in implantation, has been frequently neglected. While a number of uterine pathologies have been associated with decreased natural fertility, less information exists regarding the impact of these pathologies in ART. This report reviews the evidence to help clinicians advise ART patients.

Methods: An electronic search of PubMed and EMBASE was performed to identify articles in the English, French or Spanish language published until May 2014 which addressed uterine pathology and ART. Data from natural conception were used only in the absence of data from ART. Studies were classified in decreasing categories: RCTs, prospective controlled trials, prospective non-controlled trials, retrospective studies and experimental studies. Studies included in lower categories were only used if insufficient evidence was available. Pooled data were obtained from systematic reviews with meta-analyses when available. The summary of the evidence for the different outcomes and the degree of the recommendation for interventions were based on the GRADE (Grading of Recommendations Assessment, Development and Evaluation) statement recommendations.

Results: There is strong evidence that surrogacy is effective for uterine agenesia. For the remaining pathologies, however, there is very little evidence that the established treatments improve outcomes, or that these pathologies have a negative effect on ART. In the presence of an apparently normal uterus, assessing endometrial receptivity (ER) is the goal; however diagnostic tests are still under development.

Conclusions: The real effect of different uterine/endometrial integrity pathologies on ART is not known. Moreover, currently proposed treatments are not based on solid evidence, and little can be done to assess ER in normal or abnormal conditions. No strong recommendations can be given based on the published experience, bringing an urgent need for well-designed studies. In this context, we propose algorithms to study the uterus in ART.