Skip to main content

Table 3 ROC curve parameters for predictors of unfavorable outcomes at follow-up and early neurological deterioration

From: Clinical and imaging risk factors for early neurological deterioration and long-term neurological disability in patients with single subcortical small infarction

outcome

Variables

AUC

95%CI

P value

cut-off value

Sensitivity

Specificity

DeLong test

mRS>2

Age

0.746

0.654–0.838

< 0.001

67.5

0.655

0.792

0.0001*

 

NIHSS

0.804

0.723–0.885

< 0.001

3.5

0.862

0.65

0.0018*

 

Lacune number

0.675

0.564–0.786

0.002

1.5

0.552

0.75

< 0.0001*

 

Cerebral atrophy

0.747

0.643–0.85

< 0.001

0.5

0.724

0.708

< 0.0001*

 

Predictive mode 1

0.926

0.889–0.964

< 0.001

-

0.897

0.815

-

END

NIHSS

0.833

0.754–0.911

< 0.001

4.5

0.833

0.76

0.9094^

 

NHR

0.664

0.524–0.804

0.02

5.28

0.667

0.657

0.0026^

 

Predictive mode 2

0.836

0.744–0.927

< 0.001

-

0.722

0.875

-

  1. *Compared with prediction model 1; ^compared with prediction model 2
  2. AUC Area Under the Curve, NIHSS national institutes of health stroke scale, MCHC Mean Hemoglobin Concentration, NHR neutrophil to high-density lipoprotein cholesterol ratio