AbstractAimThis retrospective cohort study aimed to assess the utility of maternal C‐reactive protein (CRP) and leukocyte levels in predicting neonatal sepsis after preterm premature rupture of membranes (pPROM).MethodsWe conducted a retrospective cohort study (2009–2021), encompassing preterm infants born ≤29 + 6 weeks of gestation following pPROM. The primary outcome was early‐onset neonatal sepsis within the initial 72 h of life.ResultsWe analysed data from 706 patients with a median gestational age at pPROM of 25.1 weeks and a median gestational age at birth of 26.4 weeks. Overall survival rate was 86.1%, with 65.7% survival without severe morbidities. These rates were significantly worse in preterm infants with sepsis. Maternal CRP and leukocyte levels correlated significantly with neonatal infection markers and sepsis. However, their predictive values, correlation coefficients, and area under the curve values were generally low. Using maternal CRP ≥2 mg/dL to predict neonatal sepsis yielded a positive predictive value of 18.5%, negative predictive value of 91.5%, AUC of 0.589, 45.5% sensitivity, and 74.5% specificity.ConclusionMaternal CRP and leukocyte levels were ineffective as a tool for predicting early‐onset neonatal sepsis following early pPROM. Consequently, these biomarkers lack the reliability required for clinical decision‐making in this context.