Evaluating predelivery platelet and coagulation indices as predictors of immediate postpartum haemorrhage in low-risk women undergoing vaginal delivery
摘要
Although monitoring systems for high-risk postpartum haemorrhage (PPH) are well-established, predicting immediate PPH—defined as blood loss ≥ 500 mL within 2 h postpartum, distinct from general PPH (≥ 500 mL within 24 h)—remains challenging in low-risk vaginal deliveries. This case-control study aimed to explore the association between predelivery coagulation profiles and the occurrence of immediate PPH in low-risk parturients, specifically those without severe pregnancy complications and with singleton vertex presentations.
MethodsA retrospective analysis was conducted on 409 vaginal deliveries at a tertiary hospital from 2014 to 2019. Of these, 179 cases met the WHO criteria for immediate PPH, while 230 served as controls (blood loss < 500 mL). Thirty clinical and laboratory variables were extracted, including predelivery coagulation parameters—platelet count (PLT), prothrombin time (PT), activated partial thromboplastin time (APTT), and thrombin time (TT)—as well as delivery characteristics such as forceps-assisted delivery and placental retention. Logistic regression was used to identify independent risk factors, and a multivariable prediction model was subsequently developed.
ResultsMultivariate analysis identified several independent predictors of immediate PPH: Rural residence, Forceps deliveries, Retained placenta and membrane, Newborn birth weight ≥ 3500 g, PLT ≤ 212 × 10⁹/L, PT > 11 s, APTT > 28.8 s, and TT > 13.8 s. (all P < 0.05). The combined prediction model demonstrated excellent predictive performance, with an area under the receiver operating characteristic curve (AUC) of 0.854, achieving 82.58% sensitivity and 74.78% specificity.
ConclusionsThis multidimensional predictive model effectively identifies parturients at elevated risk for immediate PPH in low-risk deliveries, enabling more targeted preventive interventions. Prospective studies are warranted to validate and refine this model in broader clinical settings.