Prognostic value of blood inflammatory parameters for predicting mortality in neonates with sepsis

Background: Neonatal sepsis is a major cause of morbidity and mortality in neonates. The diagnosis of neonatal sepsis has been widely explored using blood inflammatory parameters. However, few researches have focused on the predictive significance of blood inflammation parameters for predicting mortality. This study aimed to evaluate the prognostic value of blood inflammatory parameters, including white blood cell (WBC), neutrophil, lymphocyte, monocyte, platelet and C-reactive protein (CRP) for predicting mortality in neonates with sepsis.Methods: Neonates with culture-proven sepsis were enrolled in this study. The clinical characteristics and levels of white blood cell, neutrophil, lymphocyte, monocyte, platelet and CRP were recorded. The receiver-operating characteristic (ROC) curve was applied to calculate the area under the curve (AUC) and determine the optimal cutoff values. Multivariable Cox regression model was used to evaluate the independent prognostic significance of variables. Kaplan-Meier curve was used to assess survival.Results: A total of 188 neonates with culture-proven sepsis were included for analysis. The 7-day mortality rate was 11.2 % (21/188) and the 28-day mortality rate was 13.8 % (26/188). The levels of white blood cell, neutrophil, monocyte and platelet in non-survivors were lower than those in survivors (P < 0.05). Platelet yielded higher AUC values than other parameters for predicting mortality with the best cutoff value of 132 x 10(9)/L, followed by WBC with the optimal cutoff value of 6.15 x 10(9)/L. Multivariable Cox regression analysis showed platelet and WBC were independent prognostic factors for predicting mortality. Low platelet group showed lower survival according to Kaplan-Meier method.Conclusions: In conclusion, the levels of platelet and WBC on the day of sepsis onset are valuable indicators for predicting mortality in neonates with sepsis.