Autologous platelet-rich plasma for assisted reproduction

Background Autologous platelet‐rich plasma (PRP) consists of plasma and a concentrate of platelets extracted from fresh whole blood of the person being treated. Research has suggested that intrauterine or intraovarian infusion/injection of PRP before embryo transfer may improve endometrial receptivity and response to ovarian stimulation in women undergoing assisted reproduction. We compared these interventions to standard treatment, placebo, or other interventions (mechanical or pharmacological). Objectives To assess the effectiveness and safety of intrauterine and intraovarian infusion/injection of platelet‐rich plasma in infertile women undergoing assisted reproductive technology cycles. Search methods We searched the Cochrane Gynaecology and Fertility Group's Specialised Register, CENTRAL, MEDLINE, Embase, and the Epistemonikos database in January 2023. We also searched the reference lists of relevant articles and contacted the trial authors and experts in the field for any additional trials. Selection criteria We included randomized controlled trials (RCTs) that evaluated the application of PRP in the uterine cavity, ovaries, or both versus no intervention, placebo, or any other intervention (either mechanical or pharmacological) in women undergoing in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) cycles. Data collection and analysis We followed standard methodological procedures recommended by Cochrane, including use of the updated risk of bias tool (RoB 2). The primary outcomes were live birth (or ongoing pregnancy) and miscarriage. The secondary outcomes were clinical pregnancy, complications of the procedure, multiple pregnancy, ectopic pregnancy, fetal growth restriction, preterm delivery, and fetal abnormality. We estimated the average effect of the interventions by fitting a Der Simonian‐Laird's random‐effects meta‐analysis model. We reported pooled odds ratios (ORs) with 95% confidence intervals (CIs). We restricted the primary analyses to trials at low risk of bias for the outcomes and performed sensitivity analyses that included all studies. Main results We included 12 parallel‐group RCTs that recruited a total of 1069 women. We identified three different comparison groups. Using GRADE, we assessed the certainty of evidence as very low for almost all outcomes. Intrauterine injection/infusion of platelet‐rich plasma versus no intervention or placebo Nine studies evaluated intrauterine PRP versus no intervention or placebo. Eight included women with at least two or three previous implantation failures. Only one was assessed at low risk of bias for each outcome. This study provided very low‐certainty evidence about the effect of intrauterine PRP injection versus no intervention on live birth (OR 1.10, 95% CI 0.38 to 3.14; 94 women) and miscarriage (OR 0.96, 95% CI 0.13 to 7.09; 94 women). If the likelihood of live birth following no intervention is assumed to be 17%, then the likelihood following intrauterine PRP would be 7% to 40%; and if the risk of miscarriage following no intervention is 4%, then the risk following intrauterine PRP would be 1% to 24%. When we analyzed all studies (regardless of risk of bias), we found very low‐certainty evidence about the effect of intrauterine PRP compared with placebo or no intervention on live birth or ongoing pregnancy (OR 2.38, 95% CI 1.16 to 4.86; I² = 54%; 6 studies, 564 women) and miscarriage (OR 1.54, 95% CI 0.59 to 4.01; I² = 0%; 5 studies, 504 women). The study at low risk of bias provided very low‐certainty evidence about the effect of intrauterine PRP compared with no intervention on clinical pregnancy (OR 1.55, 95% CI 0.64 to 3.76; 94 women) and ectopic pregnancy (OR 2.94, 95% CI 0.12 to 73.95; 94 women). The synthesis of all studies provided very low‐certainty evidence about the effect of intrauterine PRP compared with placebo or no intervention on clinical pregnancy (OR 2.22, 95% CI 1.50 to 3.27; I² = 24%; 9 studies, 824 women), multiple pregnancy (OR 2.68, 95% CI 0.81 to 8.88; I² = 0%; 2 studies, 240 women), and ectopic pregnancy (OR 2.94, 95% CI 0.12 to 73.95; 1 study, 94 women; very low‐certainty evidence). Intrauterine infusion of PRP may increase the risk of preterm delivery compared with no intervention (OR 8.02, 95% CI 1.72 to 37.33; 1 study, 120 women; low‐certainty evidence). No studies reported pain, infection, allergic reaction, fetal growth restriction, or fetal abnormality. Intrauterine infusion of platelet‐rich plasma versus intrauterine infusion of granulocyte colony‐stimulating factor Two RCTs evaluated intrauterine PRP versus intrauterine granulocyte colony‐stimulating factor (G‐CSF); both included women with thin endometrium, and neither was judged at low risk of bias for any outcome. We are uncertain about the effect of intrauterine PRP compared with intrauterine G‐CSF on live birth (OR 0.88, 95% CI 0.43 to 1.81; 1 study, 132 women; very low‐certainty evidence), miscarriage (OR 1.94, 95% CI 0.63 to 5.96; 1 study, 132 women; very low‐certainty evidence), and clinical pregnancy (OR 1.24, 95% CI 0.66 to 2.35; 2 studies, 172 women; very low‐certainty evidence). Neither study reported adverse outcomes other than miscarriage. Intraovarian injection of platelet‐rich plasma versus no intervention One RCT evaluated PRP injection into both ovaries versus no intervention; it was judged at high risk of bias for the two outcomes it reported. We are uncertain about the effect of intraovarian PRP injection compared with no intervention on ongoing pregnancy (OR 1.09, 95% CI 0.33 to 3.63; 73 women; very low‐certainty evidence) and clinical pregnancy (OR 0.90, 95% CI 0.31 to 2.60; 73 women; very low‐certainty evidence). The study examined no safety outcomes. Authors' conclusions We are uncertain about the effect of intrauterine or intraovarian administration of PRP on outcomes of assisted reproduction technology in infertile women. The pooled results should be interpreted with caution. Only one of the 12 included studies was judged at low risk of bias. Other limitations of the included trials were failure to report live birth, poor reporting of methods, lack of prospective protocol registration, low precision due to the small number of enrolled participants, indirectness due to the specific subpopulations and settings studied, and insufficient or absent safety data.