Search for a command to run...
Abstract Study question Is intrauterine platelet-rich plasma infusion effective in the treatment of recurrent implantation failure in terms of biochemical pregnancy, clinical pregnancy, miscarriage and live birth/ongoing pregnancy? Summary answer Intrauterine platelet-rich plasma infusion increases biochemical and clinical pregnancy whilst decreasing miscarriage, resulting in increased live birth/ongoing pregnancy in patients suffering from recurrent implantation failure. What is known already It is emerging that one in ten patients undergoing embryo transfers repeatedly fail to conceive, commonly defined as recurrent implantation failure (RIF). Intrauterine autologous platelet-rich plasma infusion (PRP) promotes endometrial and embryonic interaction which is essential for embryo implantation. After the initial promise, subsequent publication of conflicting results led the ESHRE RIF Working Group to not recommend its use. We believe this conclusion was based upon out-of-date meta-analyses with limited numbers of studies and small sample size, including a mixture of controlled clinical trials (i.e. randomized controlled trials and non-randomized controlled trials) that have obscured the true efficacy of PRP. Study design, size, duration We conducted a systematic review and meta-analysis, after a search of OVID MEDLINE and Embase from inception to January 31st 2025. The primary outcomes were biochemical pregnancy, clinical pregnancy, miscarriage and live birth/ongoing pregnancy. All controlled clinical trials comparing intrauterine PRP to control (another standard stimulation cycle) were included. Pooled analyses and subgroup analyses (when appropriate) were conducted, the latter was with differing trial designs of randomized controlled trials versus non-randomized controlled trials in mind. Participants/materials, setting, methods This systematic review was prospectively registered in PROSPERO: CRD42024421275. We followed the Cochrane Handbook recommendations and PRISMA guidelines. The number of events of primary outcomes was entered to calculate the risk ratio and 95%CI with the Mantel-Hansel method and the random-effects model (Cochrane RevMan 5.4). Heterogeneity was assessed via visual inspection of the alignment of 95%CIs in forest plots, Chi-square test and Higgins I2 statistic. Subgroup analysis differences were considered significant at p < 0.05 and I2>80%. Main results and the role of chance We analysed 31 controlled clinical trials (n = 3,813) consisting of 21 randomized controlled trials (n = 2,107) and 11 non-randomized controlled trials (n = 1,706) comparing intrauterine PRP to control (embryo transfer under another standard stimulation cycle). Biochemical pregnancy: A pooled analysis of 20 controlled trials (n = 2,924) showed a modest increase, RR = 1.56(1.38-1.77) with moderate heterogeneity in the forest plot with malalignment of confidence intervals (p = 0.03 and I2=41%). Subgroup analysis with “trial design” in mind demonstrated two distinct homogenous effect sizes with non-overlapping 95%CI in 13 randomized trials (n = 1,489), RR = 1.80(1.56-2.07) compared to 7 non-randomized trials (n = 1,345) RR = 1.35(1.20-1.52); subgroup differences significant at p = 0.003 and I2=88.7%, strongly suggestive as the sole source of heterogeneity/conflict. Clinical pregnancy: A pooled analysis of 30 trials (n = 3,663) showed a modest increase, RR = 1.67(1.47-1.90) with moderate heterogeneity (p = 0.01 and I2=40%). Subgroup analysis demonstrated two distinct homogenous effect sizes with non-overlapping 95%CI in 19 randomized trials (n = 1,957), RR = 1.93(1.69-2.20) compared to 11 non-randomized trials (n = 1,706) RR = 1.40(1.21-1.62); subgroup differences p = 0.002 and I2=90.0%, strongly suggestive as source of heterogeneity/conflict. Miscarriage: Loss of biochemical pregnancy in 9 randomized trials (n = 387), RR = 0.51(0.33-0.78, p = 0.002) and loss of clinical pregnancy in 10 randomized trials (n = 367) RR = 0.44(0.23-0.85, p = 0.01), both significantly in favour of PRP. Live birth/ongoing pregnancy: 10 randomized trials (n = 1,130), RR = 2.36(1.50-3.71, p = 0.0002). Limitations, reasons for caution A systemic flaw exists in the non-randomized controls (retrospective cohorts, patients refusing consent to therapy, patients without RIF, or having their first in-vitro fertilization attempt). This has led to apples-to-oranges comparisons that have obscured the true efficacy of intrauterine PRP. Interpretation of the latter should be conducted within randomized trials. Wider implications of the findings The current review is an up-to-date analysis of the largest sample of clinical trials regarding RIF therapy. This review allows clinicians to confidently inform patients that intrauterine PRP can increase the chance of achieving pregnancy, with less likelihood of miscarriage and hence can double the chance of a live birth. Trial registration number No