L26/P-473 Lactobacillus-containing probiotics may enhance pregnancy rates in infertile women undergoing assisted reproductive technology: systematic review and meta-analysis

Abstract

Study question In infertile women undergoing assisted reproductive technology (ART), does Lactobacillus-containing probiotic supplementation improve clinical pregnancy rates versus placebo or no treatment? Summary answer Lactobacillus-containing probiotics were associated with higher clinical pregnancy rates (RR 1.49), with stronger effects when administered for >14 days (RR 1.99). What is known already Despite advances in ART, a subset of patients continues to experience implantation failure, and uterine/endometrial factors remain difficult to modify once embryo quality is optimised. Emerging evidence links endometrial microbiota composition to implantation, with non-Lactobacillus-dominant profiles more frequently observed in infertile women and associated with poorer reproductive outcomes; however, trials of Lactobacillus-containing probiotics in ART have shown inconsistent results, potentially due to heterogeneity in patient selection, probiotic strains, administration routes, and treatment duration. Study design, size, duration Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-compliant systematic review and random-effects meta-analysis registered in PROSPERO (CRD420251149619). We searched PubMed, Embase, Web of Science, Cochrane CENTRAL, and ClinicalTrials.gov (January 2005–October 2025; English). The primary analysis synthesized randomised controlled trials (RCTs) in infertile women undergoing ART comparing Lactobacillus-containing probiotics with placebo/no treatment for clinical pregnancy. A secondary exploratory analysis pooled observational 16S rRNA studies assessing conversion to Lactobacillus-dominant endometrial microbiota. Participants/materials, setting, methods Infertile women undergoing ART were included. Lactobacillus-containing probiotics (vaginal/oral) were compared with placebo or no treatment. Two reviewers independently screened studies, extracted data, and assessed risk of bias (RoB 2; ROBINS-I). Random-effects models pooled risk ratios for clinical pregnancy, with a prespecified subgroup by treatment duration (>14 vs ≤ 14 days). Exploratory observational data were synthesized by pooling the proportion of women converting from non-Lactobacillus-dominant to Lactobacillus-dominant endometrial microbiota (defined as ≥ 90% Lactobacillus spp.). Main results and the role of chance Of 517 records screened, seven RCTs (n = 796; probiotics 399, control 397; published 2023–2025) were included. Lactobacillus-containing probiotics were associated with higher clinical pregnancy rates (41.98% vs 28.25%; RR 1.49, 95% CI 1.07–2.07; I²=29%). Effects were stronger with >14 days of treatment (RR 1.99, 95% CI 1.44–2.75; I²=0%) than with ≤14 days (RR 1.16, 95% CI 0.67–2.00). No adverse events were reported in the included trials, although adverse-event reporting was limited. RoB 2 judged no trial at high overall risk, but several had some concerns, mainly due to unclear blinding/allocation reporting; funnel plot inspection did not suggest marked asymmetry (Egger’s test not feasible). Exploratory synthesis of three observational studies (n = 67) found that 79% of infertile women with non-Lactobacillus-dominant endometrial microbiota converted to Lactobacillus-dominant status after probiotic treatment (pooled proportion 0.79, 95% CI 0.67–0.87; I²=0%), although ROBINS-I indicated moderate/high risk of bias. Limitations, reasons for caution Eligible RCTs were few, and live birth was rarely reported. Definitions of clinical pregnancy may vary across studies. Several trials had blinding/reporting limitations, and the microbiota endpoint relied on observational designs vulnerable to confounding and selection bias; therefore, mechanistic inferences should be cautious. Wider implications of the findings Lactobacillus-containing probiotics may be a noninvasive adjunct to improve ART outcomes, particularly with treatment beyond 2 weeks and in women with non-Lactobacillus-dominant endometrial microbiota. Future multicentre RCTs should standardise strains/doses/routes, incorporate baseline microbiota stratification, and evaluate live birth and safety with adequate power. Trial registration number Yes