Background: Postpartum hemorrhage remains the leading cause of maternal mortality. However, there is an insufficient understanding of atonic postpartum hemorrhage. Uterine atony is the most common cause of postpartum hemorrhage. Although an association between myometrium inflammatory cytokines and atonic postpartum hemorrhage has been demonstrated preliminarily, it is not clinically useful in predicting postpartum hemorrhage. Plasma is more readily available, and the assessment of its inflammatory status is more relevant to biological markers of postpartum hemorrhage and might explain the pathophysiology of atonic postpartum hemorrhage.Objective: Our objective was to examine changes in maternal plasma cytokines in women with atonic postpartum hemorrhage.Study design: This was a retrospective longitudinal case-control study of pregnant women with singleton gestations at term undergoing vaginal delivery. Cases were women with atonic postpartum hemorrhage, and 1:1 propensity-score matching was used to match the control group. Maternal plasma was collected in the first trimester, early third trimester, and late third trimester, and multiplex Luminex assay was used to determine the cytokine concentrations. Multivariate logistic regressions were used to determine the association between maternal cytokines at different stages of pregnancy and atonic postpartum hemorrhage.Results: A total of 36 pregnant women met the clinical diagnostic criteria for atonic postpartum hemorrhage, and 36 patients without postpartum hemorrhage were matched as the control group. Concentrations were lower for most cytokines in the atonic postpartum hemorrhage group in the first and early third trimester. However, in the late third trimester, higher plasma concentrations of basic fibroblast growth factor, interleukin-1 alpha, interleukin-1 beta, interleukin-1 receptor antagonist, interleukin-2 receptor alpha, interleukin-16, interleukin-18, macrophage colony stimulating factor, macrophage inflammatory protein-1 alpha, beta-nerve growth factor, tumor necrosis factor-related apoptosis-induced ligand, and stem cell factor were significantly associated with increased risk of atonic postpartum hemorrhage. Multiple testing correction showed that basic fibroblast growth factor (P<.001; fold change [FC]=1.16), macrophage inflammatory protein-1 alpha (P<.001; FC=1.15), and stem cell factor (P=.001; FC=1.25) had the most significant difference (P<.001). The prediction model of atonic postpartum hemorrhage constructed by these significantly changed cytokines had a high predictive efficiency (area under the curve, 0.84; sensitivity, 0.78; specificity, 0.83; +likelihood ratio, 4.66; -likelihood ratio, 0.27).Conclusion: Higher concentrations of maternal plasma cytokines in the late third trimester are associated with high risk of subsequent atonic postpartum hemorrhage. These indicators may be potential biomarkers for predicting atonic postpartum hemorrhage.