Skip to main content

Table 3 Area under the ROC curves of biomarkers at baseline for prediction of overt DIC

From: Combination of thrombin-antithrombin complex, plasminogen activator inhibitor-1, and protein C activity for early identification of severe coagulopathy in initial phase of sepsis: a prospective observational study

Biomarkers (Day 0)

AUC (95% CI)

Cutoff values*

Sensitivity

Specificity

PPV

NPV

Global markers

      

  Platelet

0.65 (0.51 to 0.76)

158 (× 103/μL)

0.62

0.65

0.62

0.65

  PT-INR

0.61 (0.48 to 0.73)

1.3

0.62

0.63

0.61

0.64

  APTT

0.61 (0.48 to 0.73)

42 (sec)

0.54

0.75

0.67

0.64

  Fibrinogen

0.64 (0.51 to 0.76)

310 (mg/dL)

0.54

0.78

0.69

0.65

  FDP

0.67 (0.54 to 0.78)

28 (μg/mL)

0.43

0.88

0.76

0.63

Thrombin generation

      

  TAT

0.77 (0.64 to 0.86)

15 (ng/mL)

0.67

0.85

0.81

0.72

  SF

0.67 (0.54 to 0.78)

7.9 (μg/mL)

0.77

0.54

0.61

0.72

Anticoagulant activity

      

  PC

0.85 (0.76 to 0.91)

46 (%)

0.81

0.79

0.79

0.82

  AT

0.76 (0.63 to 0.85)

46 (%)

0.60

0.85

0.78

0.69

Fibrinolytic activity

      

  Plasminogen

0.76 (0.63 to 0.85)

52 (%)

0.60

0.79

0.73

0.67

  α2-PI

0.79 (0.67 to 0.88)

70 (%)

0.81

0.67

0.70

0.79

  PAI-1

0.87 (0.78 to 0.92)

269 (ng/mL)

0.72

0.92

0.89

0.78

  PIC

0.49 (0.36 to 0.63)

1.9 (μg/mL)

0.35

0.89

0.76

0.59

Endothelial activation

      

  sES

0.59 (0.45 to 0.72)

67 (ng/mL)

0.51

0.72

0.62

0.61

  1. α2-PI, α2-plasmin inhibitor activity; APTT, activated partial thromboplastin time; AT, antithrombin activity; AUC, area under the curve; CI, confidence interval; FDP, fibrin degradation products; NPV, negative predictive value; PAI-1, plasminogen activator inhibitor-1; PC, protein C activity; PIC, plasmin-α2-plasmin inhibitor complex; PPV, positive predictive value; PT-INR, prothrombin time-international normalized ratio; ROC, receiver operating characteristic; sES, soluble E selectin; SF, soluble fibrin; TAT, thrombin- antithrombin complex. *Cutoff values were calculated to maximize the sum of sensitivity and specificity.