Prognostic and diagnostic scoring models in acute alcohol-associated hepatitis: A review comparing the performance of different scoring systems

Table 1 Alcohol-associated hepatitis diagnosis probability in clinically suspected Alcohol-associated hepatitis[11]

Category	Potential confounding factors1	Biopsy indication
Definite AAH	N/A	AAH clinically diagnosed and biopsy proven. Biopsy may inform of the mechanism of injury
Probable AAH	No confounding factors	AAH clinically diagnosed, biopsy not indicated
Possible AAH	Potential confounding factor present	AAH clinically diagnosed but biopsy is indicated for confirmation

¹Potential confounding factors: (1) Possible ischemic hepatitis (hypotension, severe upper gastrointestinal bleed, cocaine use within 7 d); (2) Possible metabolic liver disease (Wilson disease, alpha 1 antitrypsin deficiency); (3) Hepatotoxic medication within 30 d of onset of jaundice (drug-induced liver injury); (4) Uncertain alcohol intake (if the patient denies excessive alcohol use); (5) Atypical liver function test pattern (aspartate aminotransferase < 50 or > 400 IU/L, aspartate aminotransferase/alanine aminotransferase < 1.5); and (6) positive antinuclear antibody ANA > 1:160 or smooth muscle antibody SMA > 1:80; AAH: Alcohol-associated hepatitis.

Table 2 Alcohol-associated hepatitis prognostic scores: Components, purpose, clinical application and interpretation

Clinical score	Components	Purpose	Clinical application	Interpretation
MELD	INR, bilirubin (total), Creatinine, Sodium	Assess severity of liver disease and predict short-term mortality	Calculate on initial presentation	MELD ≥ 20 = severe AAH
Maddrey Discriminant function	PT (measured and control), bilirubin (total)	Assess severity and prognosis of alcoholic hepatitis	Calculate on initial presentation	MDF ≥ 32 = severe AAH
GAHS	Age, WBC, BUN, Bilirubin, PT (measured and control)	Assess severity and prognosis of alcoholic hepatitis	Calculate on initial presentation	GAHS ≥ 9 = severe AAH
ABIC	Age, bilirubin, INR, PT (measured and control)	Assess prognosis of alcoholic hepatitis in patients on steroid therapy	Use in patients on steroid therapy	< 6.71 low mortality risk; 6.71-8.99 intermediate mortality risk; ≥ 9.00 high mortality risk
Lille score	Age, bilirubin (initial, and day 4 OR day 7), albumin, creatinine, PT	Assess response to corticosteroid therapy in patients with alcoholic hepatitis	Use in patients on steroid therapy, at day 4 and/or day 7 to assess response and indication to continue steroids	< 0.45 at day 4-7 = favorable response to steroid therapy; > 0.45 at day 4-7 = little/no response to steroid therapy
Alcoholic hepatitis histological score	Histologic features of liver injury	Assess severity and prognosis of alcoholic hepatitis	Calculate on biopsy based on: Fibrosis stage, bilirubinostasis, polymorphonuclear infiltration, and megamitochondria	0-3: Mild AAH; 4-5: Moderate AAH; 6-9: Severe AAH

MELD: Model for end-stage liver disease; INR: International normalized ratio; BUN: Blood urea nitrogen; WBC: White blood cells; MDF: Maddrey discriminant function; AAH: Alcohol-associated hepatitis; PT: Prothrombin time; GAHS: Glasgow alcoholic hepatitis score; ABIC: Age-bilirubin-INR-creatinine score.

Table 3 Alcohol-associated hepatitis prognostic scores advantages and limitations

Clinical score	Components	Advantages	Limitations
MELD	INR, bilirubin (total), creatinine, sodium	MELD or MELD-Na ≥ 20 predicts high mortality at 30 d, consider corticosteroid therapy	(1) Mortality overestimation with elevated creatinine levels; (2) interpersonal variability of creatinine levels; (3) extrahepatic causes of sodium fluctuations; and (4) does not account for markers of AAH complications other than kidney and liver failure
Maddrey discriminant function	PT (measured and control), bilirubin (total)	MDF ≥ 32 predicts high mortality at 30 d, consider corticosteroid therapy. Oldest, most commonly used score	(1) AKI and other AAH complications not reflected in MDF; (2) PT use instead of INR; and (3) low specificity
GAHS	Age, WBC, BUN, bilirubin, PT (measured and control)	GAHS ≥ 9 is in favor of high mortality, helpful for selecting candidates for steroid treatment	(1) Only studied on the British population; and (2) lower sensitivity for short-term mortality compared to MELD/MDF
ABIC	Age, Bilirubin, INR, PT (measured and control)	Score < 6.71 has high negative predictive value to detect patients with low risk	(1) Not used for deciding on steroid initiation; and (2) low accuracy for predicting mortality in severe group
Lille score	Age, bilirubin (initial, and day 4 OR day 7), albumin, creatinine, and PT	Lille score ≤ 0.45 at day 7 (or 4) implies good response to corticosteroids	(1) Complex to calculate; (2) uses PT instead of INR; and (3) bias secondary to elevated creatinine levels and interpersonal variability of creatinine
Alcoholic hepatitis histological score	Histologic features of liver injury	Can be combined with clinical prognostic scores for more accurate mortality risk stratification	(1) Requires liver biopsy (invasive); and (2) static

MELD: Model for end-stage liver disease; INR: International normalized ratio; BUN: Blood urea nitrogen; AKI: Acute kidney injury; WBC: White blood cells; MDF: Maddrey discriminant function; AAH: Alcohol-associated hepatitis; PT: Prothrombin time; GAHS: Glasgow alcoholic hepatitis score; ABIC: Age-bilirubin-INR-creatinine score.