Abstract Study question Is there any relation between vaginal fluid cytokine levels and the composition of the vaginal microbiome in recurrent implantation failure (RIF) patients? Summary answer IL-1β and IL-2 levels are higher in patients with greater alpha diversity in vaginal microbiome, which associates dysbiotic bacteria with cytokine imbalance towards inflammation. What is known already In recent years, the development of sequencing-based technologies has enabled the evaluation of the vaginal microbiome. Studies have shown that a vaginal microbiota with less than 90% Lactobacillus is predictive of a poor IVF outcome, suggesting that dysbiosis in the reproductive tract negatively affects reproductive outcomes compared to women with a Lactobacillus-dominated microbiome. Furthermore, it is known that pregnancy is characterized by an anti-inflammatory cytokine pattern. In contrast, an altered Th1/Th2 ratio, and altered NK cell and macrophage numbers are more prevalent in women with RIF. Also, interleukin-1 system and interleukin-2 are involved in human implantation and pregnancy. Study design, size, duration 35 patients with RIF were included in a prospective study between November 2022 and July 2023. 21 had normal microbiome and 14 had dysbiotic microbiome. In all of them, two vaginal swabs were obtained to analyze microbiome composition and the levels of cytokines IL-2, IL-4, IL- 6, IL-8, IL-10, VEGF, IFNγ, TNFα, IL-1α, IL-1β, MCP-1 and EGF, respectively. Patients with uterine malformations, uterine surgery or other known factors which may influence RIF were excluded. Participants/materials, setting, methods To measure cytokine levels, one vaginal swab was resuspended in 500 μl PBS and a sandwich immunoassay with specific antibodies was performed for the cytokines IL-2, IL-4, IL-6, IL-8, IL-10, VEGF, IFNγ, TNFα, IL -1α, IL-1β, MCP-1 and EGF. Vaginal microbiome was analyzed by mass sequencing of the V3V4 region of 16S rRNA. Statistical analysis was performed with R Statistical Software and bioinformatic analysis of vaginal microbiome was performed using QIIME2, Phyloseq and MicrobiomeAnalyst packages. Main results and the role of chance For the cytokines measured in vaginal fluid, we observe that IL-1β levels (p = 0.040) are higher in patients with dysbiotic microbiome, which have an imbalance towards inflammation. Comparing levels of cytokines with respect to alpha diversity, this is greater in patients with higher levels of IL-1β (p = 0.005) and IL-2 (p = 0.032). Furthermore, the multivariate model that predicts alpha diversity does consider IL-1β and IL-2 levels (p = 0.036; AIC=45.123). Alpha diversity, moreover, is higher in vaginal samples from patients with dysbiotic microbiome (p < 0.001). Regarding beta diversity, it is also higher in patients with altered microbiome (p = 0.001). Considering the relative abundance of each bacteria, we can predict the levels of the aforementioned cytokines: for IL-2, the multivariate model reaches 88.1% of prediction including the relative abundance of Aerococcus, Burkholderia, Dialister, Escherichia, Lactobacillus, Megasphaera, Porphyromonas, Streptococcus and Ralstonia. For IL-1β, the multivariate model reaches 76.0% of prediction including Aerococcus, Dialister, Enterobacter, Lactobacillus, Megasphaera, Porphyromonas, Streptococcus, Veillonella and Ureaplasma. Some of these genera cause dysbiotic microbiome, which suggests that the relative abundance of some pathogenic bacteria is what alters the production of cytokines. Limitations, reasons for caution The dysbiotic microbiome group includes patients with various vaginal microbiome compositions, which can make it difficult to observe the specific effects of certain bacteria. Additionally, the sample size in the dysbiotic microbiome group is limited. Further studies, incorporating a metagenomic approach, might be able to identify functional groups of bacteria. Wider implications of the findings These results indicate that dysbiotic bacteria, such as Aerococcus or Streptococcus, induce an imbalance in cytokine levels, resulting in inflammation associated with lower implantation rates. When the vaginal microbiome is altered, the administration of antibiotics or probiotics may improve the immunologic environment, thus, avoiding the use of immunosuppressants. Trial registration number -