RESUMEN
OBJECTIVE@#To compare the ability of the step-by-step approach and the lab-score method in early identification of non-bacterial infection in febrile infants with less than 90 days old.@*METHODS@#A prospective study was conducted. The febrile infants with less than 90 days old hospitalized in the department of pediatrics of Xuzhou Central Hospital from August 2019 to November 2021 were enrolled. The basic data of the infants were recorded. The infants with high risk or low risk of bacterial infection was evaluated by the step-by-step approach and the lab-score method, respectively. The step-by-step approach was based on clinical manifestations, age, blood neutrophil absolute value or C-reactive protein (CRP), urine white blood cells, blood venous blood procalcitonin (PCT) or interleukin-6 (IL-6) to gradually assess the high risk or low risk of bacterial infection in infants with fever. The lab-score method was based on the levels of laboratory indicators such as blood PCT, CRP and urine white blood cells, which were assigned different scores to evaluate the high risk or low risk of bacterial infection in febrile infants according to the total score. Using clinical bacterial culture results as the "gold standard", the negative predictive value (NPV), positive predictive value (PPV), negative likelihood ratio, positive likelihood ratio, sensitivity, specificity, and accuracy of the two methods were calculated. The consistency of the two evaluation methods was tested by Kappa.@*RESULTS@#A total of 246 patients were enrolled in the analysis, and ultimately confirmed by bacterial culture as non-bacterial infections in 173 cases (70.3%), bacterial infection in 72 cases (29.3%), and unclear in 1 case (0.4%). There were 105 cases with low risk evaluated by the step-by-step approach, and 98 cases (93.3%) were ultimately confirmed as non-bacterial infection; 181 cases with low risk evaluated by the lab-score method, and 140 cases (77.4%) were ultimately confirmed as non-bacterial infection. The consistency of the two evaluation methods was poor (Kappa value = 0.253, P < 0.001). The ability of the step-by-step approach in early identification of non-bacterial infection in febrile infants with less than 90 days old was superior to the lab-score method (NPV: 0.933 vs. 0.773, negative likelihood ratio: 5.835 vs. 1.421), but the sensitivity of the former was lower than that of the latter (0.566 vs. 0.809). The ability of the step-by-step approach in early identification of bacterial infection in febrile infants with less than 90 days old was similar to the lab-score method (PPV: 0.464 vs. 0.484, positive likelihood ratio: 0.481 vs. 0.443), but the specificity of the former was higher than that of the latter (0.903 vs. 0.431). The overall accuracy of the step-by-step approach and the lab-score method was similar (66.5% vs. 69.8%).@*CONCLUSIONS@#The ability of the step-by-step approach in early identification of non-bacterial infections in febrile infants with less than 90 days old is superior to the lab-score method.