Skip to main content

Table 3 Assessment by ROC curve analyses of the individual prediction performance of each of the eight plasma proteins to differentiate between clinical malaria conditions

From: Plasma levels of eight different mediators and their potential as biomarkers of various clinical malaria conditions in African children

Test result variables

AUROCC ± Std error

Asymptomatic 95 % lower bound

Asymptomatic 95 % upper bound

A

 Fractalkine

0.985 ± 0.015

0.956

1.000

 sCD163

0.959 ± 0.019

0.922

0.996

 Neopterin

0.857 ± 0.050

0.760

0.954

 sCD14

0.842 ± 0.049

0.745

0.939

 MIG

0.788 ± 0.062

0.666

0.910

 sTREM-1

0.605 ± 0.081

0.446

0.764

 suPAR

0.485 ± 0.075

0.337

0.632

 Pentraxin

0.233 ± 0.05

0.135

0.331

B

 Pentraxin

0.979 ± 0.010

0.958

0.999

 suPAR

0.958 ± 0.018

0.922

0.994

 sCD14

0.876 ± 0.030

0.817

0.936

 sTREM-1

0.832 ± 0.037

0.759

0.905

 sCD163

0.499 ± 0.053

0.395

0.602

 MIG

0.490 ± 0.053

0.387

0.593

 Fractalkine

0.314 ± 0.049

0.218

0.410

 Neopterin

0.246 ± 0.049

0.149

0.343

C

 Neopterin

1.000 ± 0.000

1.000

0.000

 sTREM-1

0.957 ± 0.019

0.919

0.994

 suPAR

0.678 ± 0.053

0.575

0.782

 MIG

0.566 ± 0.069

0.431

0.702

 Pentraxin

0.401 ± 0.053

0.297

0.504

 sCD163

0.308 ± 0.049

0.213

0.404

 sCD14

0.100 ± 0.027

0.047

0.154

 Fractalkine

0.000 ± 0.000

0.000

0.000

  1. Plasma levels of eight proteins were quantified in 80 AM children, 69 UM patients and 66 SM or CM patients. Receiver operator characteristics (ROC) curves were used to estimate the power of each of the eight mediators to individually distinguish between two distinct clinical malaria conditions. Results of ROC curve analyses are indicated for AM children versus Controls (A), UM patients versus AM children (B) and SM–CM versus UM patients (C)