Skip to main content

Table 2 The curve of receiver operating characteristic (ROC) of the studied indices for optimum cutoff points in differentiating the study hemoglobinopathy groups from each other

From: Significance of the mathematically calculated red cell indices in patients with qualitative and quantitative hemoglobinopathies

Indices

Cutoff

AUC

95% CI

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Youden index (%)

p-value

Quantitative hemoglobinopathy (β-thalassemia)

 RDWI

≤ 241.58

0.533

(0.464, 0.601)

47.86

66.67

62.9

52

14.53

0.405

 ML_I

≤ 14.62

0.654

(0.586, 0.717)

53.85

81.82

77.8

60

35.66

< 0.001*

 GK

> 73.24

0.561

(0.492, 0.628)

66.67

51.52

61.9

56.7

18.18

0.122

 EF

> 14.2

0.586

(0.517, 0.652)

60.68

59.60

64

56.2

20.28

0.029*

 SL

≤ 1142.64

0.738

(0.675, 0.796)

70.09

69.70

73.2

66.3

39.78

< 0.001*

 SI

≤ 5.25

0.553

(0.484, 0.620)

53.85

62.63

63

53.4

16.47

0.181

 RI

> 4.11

0.589

(0.521, 0.656)

51.28

67.68

65.2

54

18.96

0.022*

 EI

≤ 20

0.687

(0.621, 0.748)

52.99

88.89

84.9

61.5

41.88

< 0.001*

 Sir_I

≤ 28.5

0.549

(0.480, 0.617)

31.62

82.83

68.5

50.6

14.45

0.209

 HH

> 25.84

0.606

(0.538, 0.672)

59.83

63.64

66

57.3

23.47

0.007*

Qualitative hemoglobinopathy (sickle cell anemia)

 RDWI

≤ 525

0.5

(0.432–0.569)

72

6.63

18.8

44

21.37

0.994

 ML_I

> 28.67

0.605

(0.536–0.670)

26

94.58

59.1

80.9

20.58

0.027*

 GK

> 103.6

0.516

(0.447–0.584)

44

81.33

41.5

82.8

25.33

0.783

 EF

> 23.7

0.518

(0.449–0.586)

46

84.94

47.9

83.9

30.94

0.757

 SL

> 1243.8

0.760

(0.698–0.816)

76

71.08

44.2

90.8

47.08

< 0.001*

 SI

> 6.41

0.603

(0.534–0.669)

42

81.93

41.2

82.4

23.93

0.039*

 RI

≤ 2.88

0.558

(0.489–0.625)

38

89.76

52.8

82.8

27.76

0.299

 EI

> 34.9

0.647

(0.579–0.711)

50

74.10

36.8

83.1

24.1

0.001*

 Sir_I

> 43.6

0.586

(0.517–0.652)

44

86.75

50

83.7

30.75

0.111

 HH

≤ 22.37

0.528

(0.459–0.596)

44

81.93

42.3

82.9

25.93

0.611

  1. *Statistically significant at 5%