web analytics

Description

Guha et al developed several discriminant functions for evaluating patients with nonalcoholic steatohepatitis and chronic liver disease. The authors are from the University of Southampton, Newcastle University and University Hospital Nottingham in England.


Patient selection: chronic liver disease, nonalcoholic liver disease

 

Outcome: significant hepatic fibrosis

 

Difference from simple panels: ELF discriminant score replaces age in years

 

Parameters:

(1) ELF discriminant score

(2) body mass index in kg per square meter

(3) diabetes or impaired fasting glucose (no = 0, yes = 1)

(4) AST to ALT ratio

(5) platelet count in 10^9/L

(6) serum albumin in g/L

 

score for any fibrosis =

= (0.062 * (BMI)) + (1.241 * (points for diabetes)) - (0.59 * (AST to ALT ratio)) + (1.482* (ELF)) - (0.002 * (platelet count)) - (0.043 * (albumin)) - 2.722

 

score for moderate fibrosis =

= (0.084 * (BMI)) + (1.848 * (points for diabetes)) - (1.839 * (AST to ALT ratio)) + (2.408 * (ELF)) - (0.012 * (platelet count)) + (0.141 * (albumin)) - 5.257

 

score for severe fibrosis =

= (-0.058 * (BMI)) + (4.513 * (points for diabetes)) - (3.144 * (AST to ALT ratio)) + (5.506 * (ELF)) - (0.026 * (platelet count)) + (0.639 * (albumin)) - 20.87

 

 

Score

Cutoff

Negative PV

Positive PV

any

-5.002

0.86

0.66

any

-3.346

0.69

0.87

moderate

-0.995

0.94

0.75

moderate

-0.016

0.90

0.81

severe

-0.2826

0.77

0.99

severe

0.0033

0.70

1.00

 


To read more or access our algorithms and calculators, please log in or register.