ABSTRACT
Objective:To explore the predictive value of relevant hemodynemics indexes on cardiac shock prognosis by PiCCO technology. Methods:A total of 54 consecutive patients with cardiac shock treated in our hospital from 2012-01 to 2013-05 were studied. The patients’ general information with hempdynemics indexes as cardiac index (CI), extra vascular lung water index (EVWI), general ejection fraction (GEF), general end diastolic index (GEDI), systolic vascular resistance index (SVRI) were monitored by PiCCO technology at before and 48 hours after treatment. Based on 28 days surviving condition, the patients were divided into 2 groups:Survival group, n=28 and Death group, n=26. The levels of above indexes were compared between 2 groups and their predictive values on cardiac shock prognosis were calculated by ROC curve analysis. Results:Compared with Death group, Survival group presented much higher CI, GEF and much lower EVWI, SVRI at before and 48 hours after treatment, P0.05. With 48 hours treatment, the ROC analysis showed AUCEVWI=0.846 with cut-off point at 7.5ml/kg, AUCCI=0.884 with cut-off point at 3.46 L/(min·m2) and AUCGEF=0.853 with cut-off point at 16.5%;the sensitivity and speciifcity of EVWI, CI, GEF were 79.2%and 81.0%, 82.8%and 83.0%, 74.1%and 88.7%respectively. Multivariate Logistic regression analysis indicated that CI and EVWI were the independent predictors for cardiac shock prognosis. Conclusion:PiCCO monitoring technology may guide the balance of relevant hempdynemics indexes in patients with cardiac shock. With 48 hours treatment, the levels of CI and EVWI had predictive value for cardiac shock prognosis.
ABSTRACT
Objective To study the dynamic change in plasma N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and its correlation with extravascular lung water index (EVLWI) in patients with septic shock. Methods Sixty-two patients with septic shock admitted to Department of Critical Care Medicine of Beijing Shijitan Hospital were enrolled. The patients were divided into survival group(39 cases)and non-survivors group (23 cases)according to 28-day prognosis. Venous blood was collected after intensive care unit(ICU)admission. The changes in plasma NT-proBNP and hemodynamics indexes levels were analyzed to evaluate their predictive value for clinical outcomes. Results Acute physiology and chronic health evaluation Ⅱ(APACHEⅡ)score(23.2±2.5 vs. 28.1±2.6),sequential organ failure assessment(SOFA score:7.74±2.80 vs. 12.43±3.00)and hemodynamics indexes including EVLWI〔ml/kg:7.0(6.0,8.0)vs. 9.0(7.0,12.0)〕,blood lactate(mmol/L):3.60±2.30 vs. 10.40±2.70)and NT-proBNP〔ng/L:945.0(228.0,1 246.0)vs. 5 471.0(3 308.0,11 174.0)〕in survivors were significantly lower than those in non-survivors,and cardiac index〔CI(L?min-1?m-2):4.23±0.85 vs. 3.31±0.74〕, global ejection fraction(GEF:0.205±0.054 vs. 0.149±0.054)were significantly higher than those in non-survivors (P<0.05 or P<0.01). Correlation analysis showed a positive correlation was found between NT-proBNP and EVLWI (r=0.277,P=0.010),and negative correlations were found between NT-proBNP and CI(r=-0.367,P=0.001), GEF(r=-0.259,P=0.017). No correlation was found between NT-proBNP and GEDVI,SVRI. Receiver operating characteristic curve(ROC curve)analysis showed that the area under the ROC curve(AUC)for plasma NT-proBNP predicting the outcome of septic shock patients was 0.869±0.042,95% confidence interval(95%CI)was 0.786-0.952,with the maximum sum of sensitivity and specificity 1.695 to determine NT-proBNP predicting septic shock patient's death cut-off value was 2 071 ng/L,under this cut-off value,the sensitivity was 81.4% and specificity,88.1%. The maximum AUC for EVLWI predicting the outcome of patients with septic shock was 0.690,cut-off value was 7.5 mL/kg,under this cut-off value,the sensitivity was 69.8% and specificity,66.7%. Maximum AUC for CI predicting the outcome of patients with septic shock was 0.785,cut-off value was 3.48 L?min-1?m-2,under this cut-off value,the sensitivity was 69.8%and specificity,66.7%. Maximum AUC for GEF predicting the outcome of septic shock patients was 0.794,cut-off value 0.175,under this cut-off value,the sensitivity was 76.2% and specificity, 81.4%. Multivariate analyses showed CI and NT-proBNP levels were independent predictors of the prognosis〔CI:P=0.001,odds ratio(OR)=9.183,95%CI 2.362-35.694;NT-proBNP:P=0.024,OR=1.000,95%CI 0.999-1.000〕. Conclusion The plasma NT-proBNP level which is correlated significantly to EVLWI can evaluate the severity of septic shock and can predict the prognosis of such patients.
ABSTRACT
Objective To investigate the value of inflammatory biomarkers such as procalcitonin (PCT),C-reactive protein (CRP),and endotoxin in early diagnosis of bacteriemia patients infected with gram-negative bacteria.Methods A cohort of 79 bacteriemia patients infected with gram-negative bacteria admitted from February 2011 to May 2013 were enrolled for retrospective study.Collected data for analysis included gender,age,disease severity (APACHE Ⅱ score),bacterial isolates from blood culture and other general information.The inflammatory biomarkers such as white blood cell (WBC),neutrophils (NEU),Creactive protein (CRP),procalcitonin (PCT),and endotoxin were assayed within 6 hours after admission.SPSS version 16.0 software was used for statistical analysis.The test of normality was used for analysis of continuous variables,t-test for inter-group comparison and non-parametric statistics for non-normal distribution variables.The AUC of ROC was calculated for determining the sensitivity and specificity of biomarkers for diagnosis of bacteriemia.Results (1) Statistically positive correlations were found among serum PCT,CRP,and endotoxin levels (PCT/CRP =0.916,PCT/endotoxin =0.496,Endotoxin/CRP =0.387),and between those and APACHE Ⅱ score were (PCT/APACHE Ⅱ =0.505,Endotoxin/APACHE Ⅱ =0.467,CRP/APACHE Ⅱ =0.278),respectively,in bacteriemia patients infected with gram-negative bacteria.(2) The receiver operating characteristic (ROC) curve indicated that AUC PCT =0.715 (sen 64.6％,spe 80.7％),AUC CRP =0.666 (sen 67.7％,spe 78.6％),AUC endotoxin =0.771 (sen 78.8％,spe 81.8％) in gram-negative bacteria bloodstream infection patients.(3) The AUC PCT =0.865 (sen 86.2％,spe 77.5％),AUC CRP =0.733 (sen 72.4％,spe 75.0％),AUCendotoxin =0.618 (sen 70.7％,spe 67.5％) in bacteriemia patients infected with gram-negative bacteria in severe sepsis and septic shock group.Conclusions The plasma PCT,CRP,and endotoxin have early predictive value in bacteriemia patients infected with Gram-negative bacteria.In sepsis stage,the level of serum endotoxin has the most significant value for diagnosis.In severe sepsis and septic shock stages,the PCT is the most value for diagnosis of bacteriemia.All biomarkers are positively correlated with severity of the disease.