#344 : Intrauterine Platelet-Rich Plasma and Peripheral Blood Mononuclear Cells Doubles Clinical Pregnancy Rate in Recurrent Implantation Failure: Efficacies Obscured by Non-Randomized Controlled Trials - Two Meta-Analyses

Abstract

Background and Aims: Recurrent implantation failure (RIF) is a frustrating problem confronting prospective mothers following embryo transfer. Two promising intrauterine immunotherapies, plasma-rich plasma (PRP) and peripheral blood mononuclear cells (PBMC) infusions, have been investigated with mixed results, leaving experts querying if these are “add-on treatments getting out of hand?”. The current meta-analyses aim to identify the source of the conflicting evidence and determine the true efficacy of IU-PRP and IU-PBMC. Method: A systematic search was conducted on April 30, 2023, using OVID Medline and Embase. Eligible articles pertaining to the efficacy of IU-PRP and IU-PBMC upon clinical pregnancy in fertility research were retrieved. Risk ratios and confidence intervals were calculated via the random effects model (Cochrane Review Manager). Results: IU-PRP Pooled data: 26 studies, 3016 participants, RR=1.64(1.41-1.90), X2=0.009, I2=44%, showing significant heterogeneity. RCTs: 15 studies, 1557 participants with RR=2.02(1.71-2.39), X2p=0.43, I2=2%. Non-RCTs: 11 studies, 1459 participants with RR=1.34(1.19-1.52), X2p=0.39, I2=5%. Subgroup analysis: homogenous effect sizes with non-overlapping 95%CI, X2p=0.0001, I2=93.4%. (Figure 1) IU-PBMC RCTs: 6 studies, 652 participants, RR=2.17(1.71-2.74), X2p=0.92, I2=0%. Non-RCTs: 6 studies, 866 participants, RR=1.43(1.20-1.70), X2p=0.69, I2=0%. Subgroup analysis: homogenous effect sizes with non-overlapping 95%CI, X2p=0.005, I2=87.4%. (Figure 2) Conclusion: Non-RCTs provide savings in cost and time that can lead to early awareness of innovative therapies. This may be counterproductive in RIF research due to flawed non-randomized controls: retrospective cohorts, women refusing consent to therapy, women without RIF or having first IVF attempt. Intervention groups therefore may have a higher baseline risk of failure, resulting in apples-to-oranges comparisons that obscure the actual efficacy. Randomization is fundamental in reducing patient selection bias and the RCTs in these analyses demonstrated that IU-PRP and IU-PBMC doubles the clinical pregnancy rate in RIF patients. A debate into the role of non-RCTs in RIF research is essential to avoid a potential systemic problem.