Risk factors for prognosis of premature infants with septic shock: analysis of 114 cases

Abstract

Objective To investigate the risk factors affecting the prognosis of preterm infants with septic shock. Methods A retrospective study was conducted to analyze the clinical data of 114 preterm children with septic shock admitted to the Neonatal Intensive Care Unit (NICU) of the PLA Army General Hospital from February 2014 to January 2017. According to the outcomes, these cases were divided into two groups, the cured group and the death group (including those died after ineffective treatment and withdrawal of treatment). Clinical data including the general clinical data, perinatal risk factors, clinical features and prognosis of the two groups of children, as well as the occurrence of related complications were statistically analyzed by t-test, Wilcoxon nonparametric test, Chi-square test or Fisher's exact probability method. Logistic regression was used to analyze the factors influencing the prognosis. Predictive values of the indicators were evaluated using receiver operating characteristic (ROC) curve. Results (1) Among the 114 patients, 87(76.3%) were cured and 27(23.7%) were dead. (2) In the death group, there were more infants complicated with amniotic fluid pollution, anemia and thrombocytopenia (platelet count <100×109/L) and the C-reactive protein (CRP) levels were higher than those in the cured group [29.6% (8/27) vs 8.1% (7/87), χ2=6.618; 22.2% (6/27) vs 5.9% (5/87), χ2=4.665; 59.3% (16/27) vs 23.3% (20/87), χ2=12.546; 36.0 (1.0-80.0) mg/L vs 7.5 (1.0-25.0) mg/L, Z=2.400], while the hemoglobin level was lower [(122.2±43.3) g/L vs (140.5±34.4) g/L, t=2.260] (all P<0.05). (3) The percentages of infants with patent ductus arteriosus, pulmonary hemorrhage and coagulopathy in the death group were higher than those in the cured group [81.5% (22/27) vs 60.9% (53/87), χ2=3.871; 37.0% (10/27) vs 12.6% (11/87), χ2=6.616; 48.2% (13/27) vs 20.7% (18/87), χ2=7.847; all P<0.05]. (4) Multivariate logistic regression analysis showed that amniotic fluid contamination, coagulopathy, patent ductus arteriosus and CRP level were risk factors for poor prognosis in neonates (all P<0.05). (5) A total of 77 pathogens were isolated from the 114 infants with 66 in the cured group and 11 in the death group. Pathogens of Gram-positive and Gram-negative bacteria and fungi in the cured and death groups accounted for 37.9% (25/66) vs 3/11, 37.9% (25/66) vs 6/11, and 24.2% (16/66) vs 2/11, respectively. No significant difference in pathogen distribution was observed between the two groups. (6) The area under the ROC curve of CRP was 0.649 (P=0.024). When the cut-off value of CRP was set at 31 mg/L, the sensitivity and specificity for predicting adverse outcomes in preterm infants with septic shock were 0.802 and 0.556, respectively, and the Yoden index was 0.358. The area under the ROC curve of thrombocytopenia was 0.708 (P<0.001). When the platelet level was set at 94×109/L, its sensitivity and specificity were 0.767 and 0.593, respectively, and the Yoden index was 0.360. Conclusions Amniotic fluid contamination, patent ductus arteriosus, coagulopathy and elevated CRP are important risk factors for death in preterm infants with septic shock. Thrombocytopenia and persistently elevated CRP has predictive values for the prognosis of preterm infants with sepsis shock. Key words: Shock, septic; Infant, premature; C-reactive protein; Infant death; Risk factors