The Characteristics And Early Detection Of Sepsis In Chinese Intensive Care Units

Part ?: Epidemiology of sepsis in Chinese intensive care units: a national cross-sectional surveyBackground Sepsis is a common clinical syndrome which induced a significant mortality.Scarce of national epidemiology data of sepsis was reported.Purpose We performed this national cross-sectional survey to determine the epidemiologic characteristics of patients with sepsis in intensive care unit(ICU)in China.Methods A national survey in 44 randomized selection hospitals was performed.All sepsis patients in participating ICUs from December 1 2015 to January 31 2016 were included into analysis.We recorded demographic,physiological and bacteriological data and followed up for 90 days or until death after inclusion.The incidence at ICU,hospital,distict region and whole mainland China population was assessed.The 90 day,ICU and hospital mortality of sepsis in the ICU was calculated.The risk factors of 90 day mortality in septic patients were analyzed.Results(1)Patients characteristics: Of 2,322 patients with sepsis were included into the analysis,786(33.85%)patients were hospital acquired sepsis.Mean age of all included patients were 60.8,64.56% were male.The mean acute physiology and chronic health evaluation(APACHE)? and SOFA score were 18.00 and 7.78,respectively.The length of ICU and hospital stay were 8(4-15)and 20(10-40)days.(2)Infection site: Of all included patients,624(23.79%)patients had two or more infection site.The most common infection site was lung which accounted for 68.09%,followed by abdomen(26.57%),blood stream(7.80%),urinary(7.15%)and soft tissue(5.77%).(3)Pathogen: Pathogen samples were sent to test in 1919(82.64%)patients,among which 962(50.13%)were detected positive results.Gram-negative bacteria were the most common pathogen which account for 32.67%,followed by Gram-positive bacteria(13.76%)and fungi(10.47%).Acinetobacter Bauman was the most common detected pathogen(8.82%),followed by klebsiella pneumonia(7.76%)and Escherichia coli(7.45%).(4)Incidence: The incidence of sepsis in the ICU was 20.6 cases per 100 ICU admissions and 0.29 cases per 100 hospital admissions.For the distict population level,the incidence of sepsis in the ICU was 18.90,13.79,14.48,15.62,13.36 and 34.33 per 100,000 population in East,South,North,Northwest,Southwest and Northeast,respectively.For the whole mainland of China,the incidence of sepsis in the ICU was 16.78 per 100,000 population.There was a significant variation of the sepsis incidence in the ICU among different geography regions.(5)Outcome: 824(35.49%)patients died at day 90 after sepsis diagnosis.The mortality in the ICU and hospital were 29.63% and 32.13%,respectively.The survival decreased combine with the increased severity of the septic patients.(6)The risk factors of mortality: Compared with survival patients,the age was significant older in non-survival patients.APACHE ? and sequential organ failure assessment(SOFA)score were significant higher in non-survival patients.Patients in non-survival group had more proportion of coronary heart disease,heart failure,chronic obstructive pulmonary disease(COPD),diabetes mellitus,chronic renal insufficiency,maintenance hemodialysis,hematological system tumors,cirrhosis,connective tissue disease and immunosuppression increased significantly.Multivariate regression showed that body weight,SOFA score,heart failure,maintenance hemodialysis,connective tissue disease,mechanical ventilation,continuous renal replacement therapy(CRRT)and lactate level were risk factors associated with 90 day mortality.Conclusion About one fifth patients suffer sepsis in the ICU and more than one third sepsis died in 90 days after diagnosis in China.Part ?: Reassess the effect of Sepsis-3 in clinical settingBackground The new sepsis diagnosis criteria based on the Sequential Organ Failure Assessment(Sepsis-3)was found to have a better predictive value of sepsis mortality than the criteria based on the Systemic Inflammatory Response Syndrome(Sepsis-1).However,there are concerns that Sepsis-3 may delay the diagnosis of sepsis.Purpose The aim of our study was to compare the different definitions in detecting sepsis,especially in different settings of Intensive Care Units(ICUs).Methods Data from the Medical Information Mart for Intensive Care ?(MIMIC-?)were used to identify adult(?18 years)admissions with suspected infections.Infection was defined based on the ICD-9 code.Two patient groups meeting either Sepsis-1(group A)or Sepsis-3(group B)and three mutually exclusive groups were defined:(C)meeting Sepsis-1 but not Sepsis-3;(D)meeting Sepsis-3 but not Sepsis-1;and(E)meeting both definitions.Patient characteristics,clinical outcomes,and resource use were compared among the 5 groups.Logistic regressions were conducted to identify predictors of meeting Sepsis-1 and/or Sepsis-3 criteria.Results(1)Patients characteristics: Of a total of 61,532 ICU admissions,17,848 patients with infection or suspected infection were included into analysis.There were no significant differences of characteristics,outcome between group A and B.Sepsis-1 and Sepsis-3 agreed on 85.6% of suspected infections.(2)The distribution of patients from different ICU who met sepsis-1 and sepsis-3 criteria: Among different ICUs,TSICU met the sepsis-1 criteria with highest proportion of 93.4% while CCU with the lowest of 90.8%.A significant different of proportion fulfill sepsis-1 was found among these ICUs(p=0.001).By contrast,the proportion of patients met sepsis-3 in different ICUs was from 84.5% to 91.3% with a significant difference among these ICUs.Paitents from SICU and TSICU had a significant higher proportion of group C(11.3% and 13.7% vs.8.7%,p<0.001)while CCU and CSRU had a a significant higher proportion of group D(6.1% and 7.5% vs.5.8%,P=0.003).(3)The mortality of patients who missed by sepsis-1 or sepsis-3: Of all included paitents,ICU mortality was 14.6%.The mortality of group E was significant higher than group C and D(5.9% and 7.8% vs.16.2%,both p<0.001).No difference of mortality was found between group C and D.The mortality of patients from different ICUs who met group C was from 5.0% to 8.1% while who met group D was from 4.7% to 11.3%,indicating that about 5%-11% patients who missed by etheir sepsis-1 or sepsis-3 will die in 90 days.(4)Risk factors associated with Sepsis-1 and/or Sepsis-3: Male was an independent risk factor that met both sepsis-1 and sepsis-3 criteria [OR=1.25(1.14-1.37),P<0.001] while female was nor prone to sepsis-3.In different ICUs,MICU patietns were more likely to meet both sepsis-1 and sepsis-3.Patients from SICU [1.18(1.04-1.33),p=0.009] and TSICU[1.28(1.11-1.47),p=0.001] were more likely missed by sepsis-3 while Patients from CCU[1.19(1.02-1.38),p=0.024] were more likely missed by sepsis-1.Patients with higher SAPS score were more likely fulfill both sepsis-1 and sepsis-3.Conclusion For ICU patients,both sepsis-1 and sepsis-3 may miss part sepsis patients which indicate combine using two criteria for sepsis diagnosis.Part ? SIRS is still important for sepsis diagnosis in patients with sepsis in Chinese intensive care unitsBackground: A big controversy between sepsis-1 and sepsis-3 still remain nowadays.Purpose: We performed this study to identify whether Sepsis-3 replace sepsis-1 in sepsis diagnosis in ICU patients in China.Methods: This was a secondary analysis the data from CHESS study which performed in 44 hospitals from every province/municipality/autonomous region in the mainland of China.All the patients who met sepsis-1 enrolled in Part I study were included this analysis.We extracted the data APACHE ?,SOFA and q SOFA scores.Physiological and laboratory variables were collected for calculation of modified early warning score(MEWS).The performance of 90 day mortality prediction of APACHE ?,SOFA,q SOFA and MEWS scores were calculated by a receiver operating characteristic(ROC)curve and assessed by the area under the ROC curve(AUC).Results:(1)Patients characteristics: Of a total 2322 patient who met sepsis-1,2254(97.07%)patients met sepsis-3 criteria.No significant difference of characteristics was found between patients met sepsis-1 and patients met both sepsis-1 and sepsis-3.(2)The effect of SIRS on severity and mortality: Of all included patients who met sepsis-1,711(30.62%),1010(43.50%)and 601(25.88%)patients met 2,3,and 4 systemic inflammatory response syndrome(SIRS)criteria.The APACHE ?,SOFA score and lactate level increased with who met the increased number of SIRS criteria(all p<0.001).The mortality was 27.85%,36.34 and 43.09% in patients who met number of 2,3 and 4 SIRS criteria,respectively.(3)The effect of q SOFA score on severity and mortality: Of all included patients who met sepsis-1patients,there were 265(11.41%),911(39.23%),853(36.74%)and 94(12.66%)patients fulfilled q SOFA score of 1,2,3 and 4,respectively.Similarly,the APACHE ?,SOFA score and lactate level increased with the increased q SOFA score(all p<0.001).The mortality of patients with q SOFA score of 0,1,2 and 3 was 15.53%,29.31%,40.09% and 59.18%,respectively.(4)The predictive value of various scores: The 90 day mortality in patients met both sepsis-1 and sepsis-3 and met sepsis-1only was 36.49% and 8.82%,respectively.The AUCs of APACHE ?,SOFA,SIRS,q SOFA and MEWS in 90 day mortality prediction were 0.720,0.729,0.568,0.635 and 0.656,respectively.SOFA and APACHE ? socres had a significant higher predicted value for 90-day mortality prediction than MEWS,q SOFA and SIRS(all p<0.001).Sensitivity analysis showed that similar results in patients with Sepsis-3 only.Conclusion: SIRS still play in important role in sepsis diagnosis in the ICU in China.SOFA and APACHE ? but not q SOFA and MEWS had a better accurate in predicting 90 day mortality in patients with sepsis.Part ?: Combination of C-reactive protein,procalcitonin and sepsis-related organ failure score for the diagnosis of sepsis in critical patientsBackground: Early appropriate management is crucial to improve the outcome of patients with sepsis.However,how to diagnose sepsis early is difficult in clinic setting.Purpose: To measure the accuracy of a new bioscore to diagnose sepsis in a general critical care population.Methods: The study was undertaken at the intensive care unit of hospital from April to December 2012.Demographic,clinical patient information were recorded at admission while blood samples were taken to assess C-reactive protein(CRP),procalcitonin(PCT)and interleukin 6(IL-6).Receiver operating characteristic(ROC)curves and the area under the curve(AUC)were evaluated the diagnostic accuracy of CRP,PCT and IL-6.Extended bioscores(e-BS)were computed by CRP,PCT,IL-6,body temperature,white blood cell(WBC)count and sepsis-related organ failure(SOFA)with reference to cut-off values from the ROC curves and multivariate analysis to provide a score calculation for sepsis diagnosis and prognosis,respectively.Results:(1)Patients characteristics: Three hundred patients were enrolled,which 107 of patients were sepsis and 193 of patients were non-sepsis.Compare to the non-septic patients,patients with sepsis had older age,higher proportion of male,higher APACHE ? and SOFA score.(2)The diagnostic value of biomarkers and SOFA score: Compare to non-septic patients,the CRP and PCT level were significant higher in patients with sepsis [CRP,63.8(32.3–114)ng/ml vs.21.8(5.20–60.55)ng/ml,p<0.001;PCT,1.79(0.57–8.95)pg/ml vs.0.38(0.11–1.54),p<0.001].No difference of IL-6 between septic and non-septic patients.For sepsis diagnosis significant AUC were for CRP0.729(p<0.001,95%CI 0.671–0.787),PCT 0.711(p<0.001,95%CI 0.652–0.770),and SOFA 0.670(p<0.001,95%CI 0.607–0.733).(3)The diagnostic value of combine score: Multivariate logistic regression showed that age,gender,CRP,PCT and SOFA were risk factor for occurrence of sepsis.The AUC of combine score in sepsis diagnosis was 0.781(p<0.001,95%CI 0.727– 0.835)based on the PCT,CRP and SOFA score.(4)The risk factors of 28 day mortality in patients with sepsis: Multivariate logistic regression analysis showed that age and SOFA were independent associated with 28-day mortality(OR=1.022,95%CI: 1.003–1.041,p=0.024;OR=1.263,95%CI: 1.148–1.389,p<0.001)Conclusions: Combined bioscore methods with CRP,PCT and SOFA score improved the value in sepsis diagnosis.