### Description

Sloan et al developed a model for predicting 30 day mortality in patients who develop intracranial hemorrhage after thrombolytic therapy for acute myocardial infarction (AMI). This can help identify patients who may benefit from more aggressive management and monitoring. The authors are investigators from the GUSTO-I trial.

Parameters:

(1) age

(2) Glasgow Coma Score

(3) time interval between thrombolysis and onset of stroke in hours

(4) volume of intracranial hemorrhage in mL

Age

Points

<= 40

0

45

2

50

4

55

5

60

7

65

9

70

11

75

13

80

14

85

16

>=90

18

When modeled in JMP, this can be represented as:

points =

=  (0.3326 * (age)) - 12.985

GCS

Points

15

0

14

2

13

3

12

5

11

7

10

8

9

10

8

12

7

13

6

15

5

16

4

18

3

20

Hours after Thrombolysis

Points

0

100

20

93

40

87

60

80

80

73

100

67

120

60

140

53

160

47

180

40

200

33

220

27

240

20

260

13

280

7

>= 300

0

When modeled in JMP, this can be represented as:

points =

= (-0.333 * (hours)) + 99.963

Volume in mL

Points

0 – 50

0

60

2

80

5

100

9

120

13

140

16

160

20

180

24

200

27

220

31

240

35

260

38

280

42

300

46

When modeled in JMP, this can be represented as:

points =

=  (0.1833 * (volume in mL)) - 9.2905

total score =

= SUM(points for all 4 parameters)

Interpretation:

• minimum score:0

• maximum score: 256

• The higher the score the greater the risk of 30 day mortality.

Total Score

30 Day Mortality Rate

0 – 86

< 1%

87

1%

97

5%

101

10%

109

30%

115

50%

120

70%

128

90%

132

95%

142

99%

143 – 256

> 99%

When modeled in JMP, this can be represented as:

Score

Mortality Rate in Percent

87 - 109

(0.001975 * ((score)^3)) - (0.5022 * ((score)^2)) + (42.602 * (score)) - 1204.894

109 - 120

(0.0606 * ((score)^2)) - (10.2324 * (score)) + 426.36

120 - 142

(0.001975 * ((score)^3)) - (0.8547 * ((score)^2)) + (123.328 * (score)) - 5834.571

Alternatively, analyzing the data in JMP:

X = (0.166 * (score)) - 19.06

probability =

= 1 / (1 + EXP((-1) * X))