Clinical Study Of Procalcitonin In Differential Diagnosis And Illness Assessment Of Different Bacterial Infections

Objective To evaluate the accuracy of procalcitonin(PCT)serum concentration to diagnose sepsis with Gram-negative bacterium(GNB)infection and the association of PCT serum concentrations with more specific pathogens and the focus of infection.Method Patients with diagnosis of bloodstream infection to department of emergency and intensive care unit of China-Japan Friendship Hospital from January 2015 to June 2020 were enrolled.Sequential organ failure assessment(SOFA)score was calculated based on the worst parameters on the day of blood culture.Differences of various indicators between Gram-positive bacterium(GPB)and GNB infection were compared Differences in PCT serum concentrations were compared among specific pathogens and foci of infection.Test performance for the prediction of GNB infection was assessed by receiver operating characteristic(ROC)curve analysis.Results Among 956 septic patients with single bacterial infection,426(44.6%)patients with GPB infection and 530(55.4%)with GNB infection.PCT value was significantly higher in GNB infection compared to GPB infection(3.35μg/L vs 1.03μg/L,P<0.001).PCT values also varied among different species of bacteria.PCT values in Staphylococcus aureus and Enterococcus faecium infection were higher than those in Staphylococcus epidermis,Staphylococcus hominis and Staphylococcus hemolyticus infection(P<0.05),and Burkholderia onions were lower than those of Escherichia coli,Klebsiella pneumoniae and Enterobacter cloacae(P<0.05).PCT was highest in urogenital followed by abdominal infection.ROC curve analysis showed that:①The optimal cut-off value of PCT in the diagnosis of sepsis with GNB infection was 0.515μg/L,with the area under the ROC curve was 0.706[95%confidence interval(95%CI)was 0.665-0.747].The specificity was 80.1%with a cut-off value of 7.175 μg/L.②The optimal cut-off value of PCT in the diagnosis of pneumonia with GNB infection was 0.480 μg/L,with the area under the ROC curve was 0.730(95%CI 0.677-0.783).The specificity was 80.2%with a cut-off value of 4.535μg/L.③The optimal cut-off value of PCT in the diagnosis of GNB abdominal infection was 1.815μg/L,with the area under the ROC curve was 0.724(95%CI 0.628-0.820).The specificity was 76.6%with a cut-off value of 10.920 μg/L and the specificity was 80.9%with a cut-off value of 18.220μg/L.Conclusion The PCT value of GNB infection was significantly higher than that of GPB infection.Different cutoff values should be used to distinguish GNB from GPB sepsis for different infection sites.Objective To assess the accuracy of procalcitonin(PCT)in distinguishing the severity of different bacterial infections,and to explore the predictive value of PCT and critical illness scores(CISs)for death in sepsis.Method Patients with diagnosis of BSI to department of emergency and intensive care unit of China-Japan Friendship Hospital from January 2015 to June 2020 were enrolled.Sequential organ failure assessment(SOFA),mortality in emergency department sepsis(MEDS),logistic organ dysfunction system(LODS),acute physiology and chronic health evaluation Ⅱ(APACHE-Ⅱ)scores were calculated based on the worst parameters on the day of blood culture.Differences of various indicators between infection,sepsis,and septic shock group were compared.Receiver operating characteristic(ROC)curves were used to analyze the diagnostic accuracy for differentiating between infection,sepsis,and septic shock group,and the accuracy of PCT and CISs in predicting death within 28 and 60 days in BSI patients.Results Among 1270 patients with single bacterial infection,594(46.8%)patients with GPB infection and 676(53.2%)with GNB infection.The levels of PCT,SOFA,MEDS,LODS and APACHE-Ⅱ in the sepsis and septic shock group were significantly higher than those in the infection group(P<0.001).The fatality rates in 28 and 60 days were significantly higher in the septic shock group than in the sepsis group,and the sepsis group was significantly higher than that in the infection group(P<0.001).ROC curve analysis showed that:①The optimal cut-off values of PCT for distinguishing sepsis group+septic shock group and septic shock group were 1.645μg/L and 7.065 μg/L,with the areas under the ROC curve were 0.724(95%CI 0.678-0.770)and 0.682(95%CI 0.629-0.734),respectively.To achieve similar accuracy,the cut-off values were 1.020 μg/L and 3.530 μg/L,1.725 μg/L and 10.160μg/L for GPB and GNB infection.②The areas under the curve of SOFA,MEDS,LODS,APACHE-II scores were significantly larger than that of PCT in predicting death within 28 and 60 days in sepsis patients with single bacterial infection[28d:0.822(0.781-0.862),0.787(0.744-0.830),0,808(0.768-0.848),0.815(0.777-0.853)vs 0.622(0.573-0.671);60d:0.783(0.741-0.825),0.761(0.717-0.804),0.788(0.748-0.828),0.806(0.769-0.843)vs 0.604(0.557-0.651)].Compared with GNB infection,PCT had a larger area under the ROC curve for the prognosis of GPB induced sepsis[28d:0.712(0.648-0.776)vs 0.556(0.482-0.630);60d:0.687(0.622-0.753)vs 0.541(0.473-0.609)].Combining the PCT cut-off value with the SOFA score increased the sensitivity(93.0%,90.1%,90.1%)and specificity(84.0%,88.5%,86.5%)for predicting death within 28 days and the sensitivity(87.0%,86.2%,85.6%)and specificity(85.2%,88.1%,85.9%)for predicting death within 60 days caused by a single species of bacterial infection,GPB or GNB infection,respectively.Conclusion PCT may aid in the assessment of the severity of BSI,but for different bacterial infection,different cut-off values should be used.CISs are more advantageous in predicting death within 28 and 60 days in sepsis patients,and SOFA score combined with PCT can improve the accuracy of prediction.