Diagnostic Accuracy Of Quick Sequential Organ Failure Assessment Score For Sepsis Among Patients With Infection In General Wards

Background:The sepsis is the leading pathway to death of hospitalized patients with infection.The third international consensus meeting modified the definition of sepsis to a life-threatening organ dysfunction caused by a dysregulated host response to infection,i.e.,an infection associated an increased in the Sequential Organ Failure Assessment(SOFA)score of 2 points or more.For patients outside ICU,a quick-SOFA(qSOFA)score was recommended as a screen tool before systemic assessment of SOFA.However,the diagnostic accuracy of qSOFA has not been well estimated among patients in general wards until now.Purpose:To estimate the diagnostic accuracy of qSOFA score for sepsis,and compare the merits and demerits with the Systemic Inflammatory Response Syndrome(SIRS)score.Methods:We prospectively rated qSOFA daily and SOFA,SIRS when laboratory examination available from hospital admission to 28 day or discharge among patients in ten general wards with high prevalence of infection.The baseline characteristics,scores and clinical outcomes were compared between sepsis and non-sepsis patients.The diagnostic accuracy for sepsis was compared by using the receiver operating curve(ROC)between qSOFA and SIRS scores.The 28-day mortality and unfavorable outcome(28-day mortality or intensive care unit length of stay?3days)were also compared among false negative,false positive and true positive groups,and further adjusted by a logistic regression.The relations between scores and clinical outcomes were examined by logistic regression and ROC methods.Results:1.Characteristics:During the study period,203 infected patients were included and 103 were diagnosed with sepsis.The patients with sepsis were older than non-sepsis patients,and have lower platelet counts but higher heart rates,respiratory rates on the first day of infection,meanwhile suffered more organ dysfunction required more supportive therapies.The mortality and unfavorable outcome were also more common among sepsis patients.More patients in sepsis group met the qSOFA and SIRS criteria,however the qSOFA positive was little delayed.2.The diagnostic accuracy for sepsis:The qSOFA score possessed a larger area under the ROC than the SIRS score(0.74[0.68-0.80]vs.0.66[0.59-0.73],P=0.043)with different cutoff values of 2 and 3 score.The qSOFA score had much higher specificity(87%[84%-90%]vs.46%[40%-50%],P<0.001)but lower sensitivity(82%[78%-85%]vs.50%[45%-54%],P<0.001)than SIRS score.The negative predictive values of qSOFA was 63%[58%-67%]and the false negative group accounted for 27%of study population.Fortunately,the patients in false negative group of qSOFA enjoyed much better clinical outcomes,with a OR of 0.12[0.02-0.48]for mortality and 0.10[0.03-0.28]for unfavorable outcome referring to true positive group.Similarly,the patients in false positive group of SIRS also enjoyed much better clinical outcomes,with an OR of 0.03[0.00-0.15]for unfavorable outcome referring to true positive group.3.The predictive accuracy for clinical outcome:The ORs of qSOFA,SIRS,SOFA and?SOFA scores for mortality were 8.3[3.6-21],2.5[1.3-5.7],1.9[1.5-2.5]and 1.8[1.5-2.3],moreover,the ORs unfavorable outcome were 12.2[5.7-31.9],2.7[1.6-4.8],1.9[1.6-2.5],2.3[1.8-3.1],respectively.The AUROC of qSOFA,SIRS,SOFA andASOFA scores were 0.85[0.77-0.94],0.68[0.59-0.77],0.95[0.91-0.99],0.95[0.92-0.99]for mortality and 0.86[0.8-0.92],0.70[0.62-0.77],0.90[0.85-0.96],0.94[0.91-0.98]for unfavorable outcome.Conclusion:These finding suggest the qSOFA score can make specific,but un-sensitive and un-timely detection of sepsis.However,the SIRS score is a sensitive but un-specific tool for screening sepsis.The patients in qSOFA false negative group and SIRS false positive group enjoyed fairly good clinical outcomes when compared with patients in true positive group.