Embryo implantation following transfer is considered a critical success-determining step of in vitro fertilisation (IVF); however, the probability that an embryo will successfully implant into the uterus after transfer is surprisingly low, standing at approximately 30%.1 Implantation failure is devastating for the couple; it can have significant financial implications as well as the associated physical and psychological burden. In cases of recurrent implantation failure (RIF), these consequences are amplified. RIF arises after embryo transfer when implantation repeatedly fails to reach a stage identifiable by ultrasonography.2
At present, the management of RIF remains ambiguous among clinicians and investigations and treatments are often prescribed depending on the clinician. In response to this, we conducted a systematic review and meta-analysis aiming firstly to compare the prevalence of positive immunological test outcomes between women with a history of RIF and both subfertile and fertile women with no history of the condition; and secondly to evaluate the efficacy of a number of empirical treatments to improve IVF outcome in women experiencing unexplained RIF. We searched two scientific databases, MEDLINE and the Cochrane Library, from commencement to June 2017. To be eligible for inclusion, our first objective studies had to be observational in design, comparing the prevalence of positive immunological test outcomes between women with a history of RIF and fertile and subfertile women with no history of the condition. For the second objective, the studies had to be either observational or randomised controlled trials comparing IVF outcome between women with a history of unexplained RIF who were administered an empirical treatment prior to the cycle, and women with RIF who were administered a placebo or no treatment prior to the cycle.
Our search yielded a total of 10,034 citations and, after careful scrutiny, we identified 30 relevant citations for the first objective, and 17 relevant citations for the second. The quality of observational studies was assessed using the Newcastle–Ottawa scale, whereas the quality of randomised controlled trials was evaluated using The Cochrane Risk of Bias Tool.
The prevalence of positive antiphospholipid antibody test outcomes was significantly higher in women with RIF than in controls (odds ratio [OR]: 3.35; 95% confidence interval [CI]: 2.05–6.05; p<0.001), more specifically lupus anticoagulant (OR: 5.03; 95% CI: 1.81–13.99; p=0.002). Similar results were seen for inherited thrombophilias, specifically Factor V Leiden (OR: 2.69; 95% CI: 1.28–5.63; p=0.009). While meta-analyses for the prevalence of high natural killer cells and Type 1 T helper:Type 2 T helper cell ratio were not possible, no difference was identified in the prevalence of anti-thyroid antibodies. The prevalence of certain human leukocyte antigen-G variants was found to be significantly higher in women with RIF; -14 base pair (bp) (bp/-14 bp and -14 bp/+14 bp genotype) as well as the 010101, 0106, 010106, and 0105 N alleles. Endometrial injury in the previous menstrual cycle was shown to benefit clinical pregnancy rate in women with RIF (OR: 3.38; 95% CI: 2.26–5.06; p<0.001). Similar results were seen for peripheral blood mononuclear cell administration to the uterus (OR: 3.68; 95% CI: 2.01–6.64; p<0.001).
Additionally, we showed a beneficial effect of intravenous immunoglobulin therapy on live birth rate in this cohort (OR: 10.51; 95% CI: 1.52–76.66; p=0.02). Whilst our results support testing for certain congenital and inherited thrombophilias in women with RIF, evidence to suggest a role of other immunological factors in the pathogenesis of this condition is limited. Furthermore, our results suggest that endometrial injury, intravenous immunoglobulin therapy, and peripheral blood mononuclear cell administration prior to embryo transfer show promise for the treatment of unexplained RIF in the future; however, this is not before optimisation studies are performed to establish appropriate clinical guidelines for their prescription.
