Gastroenterology Research, ISSN 1918-2805 print, 1918-2813 online, Open Access
Article copyright, the authors; Journal compilation copyright, Gastroenterol Res and Elmer Press Inc
Journal website https://www.gastrores.org

Original Article

Volume 16, Number 1, February 2023, pages 25-36


The Impact of Metabolic Syndrome on the Prognosis of High-Risk Alcoholic Hepatitis Patients: Redefining Alcoholic Hepatitis

Figures

Figure 1.
Figure 1. Data collection and analysis process.
Figure 2.
Figure 2. Study cohort categorization.
Figure 3.
Figure 3. Distribution of patients based on infection status.
Figure 4.
Figure 4. Forest plot of variables with significant P-value (< 0.05). For one unit increase of BMI, the mortality risk will decrease by 14%. BMI: body mass index.
Figure 5.
Figure 5. Kaplan-Meier mortality risk. (a) Stratified based upon metabolic syndrome and MELD status. (b) Stratified based upon metabolic syndrome and DF. MELD: model for end-stage liver disease; DF: discriminant function.

Tables

Table 1. Clinical Features Based Upon Metabolic Syndrome (t-Test, Chi-Square)
 
VariableOverall (N = 109)Alcohol hepatitis (N = 62)Alcohol hepatitis with metabolic syndrome (N = 47)P-value*
Data are presented as mean ± SD or n (%). *Derived from Wilcoxon rank-sum test for continuous variables and Fisher’s exact test for categorical variables. BMI: body mass index; WBC: white blood cell; Hb: hemoglobin; Hct: hematocrit; MCV: mean corpuscular volume; INR: international normalized ratio; PT: prothrombin time; AST: aspartate aminotransferase; ALT: alanine aminotransferase; AP: alkaline phosphatase; T. bili: total bilirubin; Fib-4: fibrosis-4; APRI: aspartate aminotransferase to platelet ratio index; CTP: Child-Turcotte-Pugh; MELD: model for end-stage liver disease; DF: discriminant function; ANI: alcoholic liver disease/non-alcoholic fatty liver disease index.
Age47.49 ± 12.7247.50 ± 11.9047.47 ± 13.870.67
Female43.00 (39.45%)29.00 (46.77%)14.00 (29.79%)0.08
Race0.38
  Caucasian72.00 (66.06%)43.00 (69.35%)29.00 (61.70%)
  Hispanic24.00 (22.02%)14.00 (22.58%)10.00 (21.28%)
  Others13.00 (11.93%)5.00 (8.06%)8.00 (17.02%)
Diabetic12.00 (11.11%)0.00 (0.00%)12.00 (25.53%)< 0.001
BMI26.85 ± 7.1223.57 ± 3.6430.97 ± 8.24< 0.001
WBC10.66 ± 5.9710.57 ± 6.0910.79 ± 5.860.67
Hb12.04 ± 2.8511.36 ± 2.4012.94 ± 3.16< 0.01
Hct34.96 ± 8.5932.84 ± 7.7837.77 ± 8.87< 0.01
MCV97.20 ± 8.5896.33 ± 9.1698.34 ± 7.670.37
Platelets185.06 ± 118.04188.48 ± 130.47180.53 ± 100.510.75
Creatinine1.35 ± 1.381.17 ± 0.961.59 ± 1.770.06
INR2.62 ± 4.632.26 ± 4.043.10 ± 5.320.76
PT27.28 ± 33.9823.87 ± 29.9231.70 ± 38.520.32
AST243.66 ± 552.55194.19 ± 243.90308.91 ± 793.910.46
ALT141.52 ± 357.81105.98 ± 133.78188.40 ± 522.400.46
AP158.62 ± 117.41165.90 ± 128.99148.63 ± 99.980.69
Albumin2.91 ± 0.893.02 ± 0.922.76 ± 0.850.14
< 3.564.00 (58.72%)34.00 (54.84%)30.00 (63.83%)0.43
T. bili6.23 ± 8.526.57 ± 8.995.78 ± 7.930.88
Na134.02 ± 6.87133.42 ± 6.87134.81 ± 6.880.58
Lactate3.48 ± 3.253.05 ± 3.114.20 ± 4.040.45
AST/ALT ratio2.25 ± 1.452.45 ± 1.592.00 ± 1.200.19
AST/AP ratio1.86 ± 3.421.60 ± 2.662.23 ± 4.260.73
ALT/AP ratio1.05 ± 2.640.86 ± 1.371.31 ± 3.760.47
ALT/AP index2.76 ± 6.942.25 ± 3.593.45 ± 9.860.47
AST/AP index4.88 ± 8.984.19 ± 6.995.84 ± 11.170.73
NR-ALT/AP index5.80 ± 14.624.75 ± 7.557.25 ± 20.770.47
NR-AST/AP index10.28 ± 18.918.82 ± 14.7212.30 ± 23.520.73
Fib-47.88 ± 12.937.70 ± 9.058.12 ± 16.850.46
APRI5.47 ± 18.394.37 ± 8.166.91 ± 26.490.67
HAI36.00 (33.03%)16.00 (25.81%)20.00 (42.55%)0.10
Antimicrobial54.00 (49.54%)28.00 (45.16%)26.00 (55.32%)0.34
CTP8.32 ± 2.658.29 ± 2.838.35 ± 2.460.80
MELD16.48 ± 14.2314.90 ± 13.8318.56 ± 14.640.14
≥ 2129.00 (26.61%)15.00 (24.19%)14.00 (29.79%)0.52
≥ 1837.00 (33.94%)19.00 (30.65%)18.00 (38.30%)0.42
DF75.34 ± 157.8759.39 ± 140.2896.39 ± 177.830.22
≥ 3241.00 (37.61%)22.00 (35.48%)19.00 (40.43%)0.69
ANI1.63 ± 9.163.34 ± 9.44-0.62 ± 8.340.02
≥ 2.273.00 (66.97%)43.00 (69.35%)30.00 (63.83%)0.68

 

Table 2. Identification of Factors Associated With Combined Alcohol Hepatitis With Metabolic Syndrome (Logistic Regression, Stepwise Regression)
 
VariableUnadjustedaAdjustedb
OR (95% CI)aP valueOR (95% CI)P value
aDerived from logistic regression. bStepwise selection with an entry P-value of 0.05 and a removal P-value of 0.05 was performed to select variables, in which lactate was excluded due to an extremely high missing rate > 90%. BMI: body mass index; WBC: white blood cell; Hb: hemoglobin; Hct: hematocrit; MCV: mean corpuscular volume; Cr: creatinine; INR: international normalized ratio; PT: prothrombin time; AST: aspartate aminotransferase; ALT: alanine aminotransferase; AP: alkaline phosphatase; AAPI1: ALT/AP index; AAPI2: AST/AP index; Fib-4: fibrosis-4; APRI: aspartate aminotransferase to platelet ratio index; CTP: Child-Turcotte-Pugh; DF: discriminant function; ANI: alcoholic liver disease/non-alcoholic fatty liver disease index; OR: odds ratio; CI: confidence interval.
Age1.00 (0.97 - 1.03)0.99
Female0.48 (0.22 - 1.07)0.07
Race (vs. Caucasian)0.37
  Hispanic1.06 (0.41 - 2.71)0.90
  Others2.37 (0.71 - 7.98)0.16
BMI (n = 106)1.26 (1.14 - 1.39)< 0.00011.34 (1.17 - 1.54)< 0.0001
WBC1.01 (0.94 - 1.07)0.85
Hb1.23 (1.06 - 1.43)< 0.01
Hct1.08 (1.02 - 1.14)< 0.01
MCV1.03 (0.98 - 1.08)0.23
Platelets0.999 (0.996 - 1.003)0.73
Cr1.28 (0.92 - 1.77)0.14
INR1.04 (0.96 - 1.13)0.36
PT (n = 107)1.01 (0.995 - 1.019)0.25
AST1.000 (1.000 - 1.001)0.33
ALT1.001 (0.999 - 1.003)0.32
AP (n = 102)0.999 (0.995 - 1.002)0.46
Albumin0.71 (0.46 - 1.10)0.12
< 3.51.39 (0.62 - 3.13)0.43
Total bilirubin0.99 (0.98 - 1.04)0.63
Na1.03 (0.97 - 1.09)0.30
Lactate (n = 8)1.13 (0.71 - 1.80)0.61
AST/ALT ratio0.79 (0.59 - 1.06)0.11
ALT/AP ratio (n = 102)1.07 (0.90 - 1.28)0.43
AST/AP ratio (n = 102)1.06 (0.94 - 1.19)0.37
AAPI1 (n = 102)1.03 (0.96 - 1.10)0.43
AAPI2 (n = 102)1.02 (0.98 - 1.07)0.37
NR ALT/AP (n = 102)1.01 (0.98 - 1.05)0.43
NR AST/AP (n = 102)1.01 (0.99 - 1.03)0.37
Fib-41.00 (0.97 - 1.03)0.86
APRI1.01 (0.99 - 1.03)0.50
CTP (n = 88)1.01 (0.86 - 1.18)0.92
Infection2.13 (0.95 - 4.79)0.05
Antibiotics use1.50 (0.70 - 3.22)0.29
MELD1.02 (0.99 - 1.05)0.19
≥ 211.33 (0.57 - 3.12)0.51
≥ 181.41 (0.63 - 3.12)0.40
DF1.002 (0.999 - 1.004)0.24
≥ 321.23 (0.57 - 2.69)0.60
ANI0.95 (0.91 - 0.995)0.03
≥ 2.20.78 (0.35 - 1.74)0.54

 

Table 3. Identification of Factors Associated With Survival (Cox Regression)
 
VariableUnadjusted (n = 108)Adjusteda (n = 105)Adjustedb (n = 105)
HR (95% CI)aP valueHR (95% CI)P valueHR (95% CI)P value
Derived from Cox regression. Stepwise selection with an entry P-value of 0.05 and a removal P-value of 0.05 was performed to select variables, in which lactate was not included due to an extremely high missing rate > 90%. aAdjusted for age, BMI, WBC, albumin, Na, MELD, and IR. bAdjusted for age, BMI, WBC, albumin, Na, DF, and IR. Note: One patient had a survival time < 0 so this patient was excluded from survival analysis. BMI: body mass index; WBC: white blood cell; Hb: hemoglobin; Hct: hematocrit; MCV: mean corpuscular volume; Cr: creatinine; INR: international normalized ratio; PT: prothrombin time; AST: aspartate aminotransferase; ALT: alanine aminotransferase; AP: alkaline phosphatase; AAPI1: ALT/AP index; AAPI2: AST/AP index; Fib-4: fibrosis-4; APRI: aspartate aminotransferase to platelet ratio index; CTP: Child-Turcotte-Pugh; DF: discriminant function; ANI: alcoholic liver disease/non-alcoholic fatty liver disease index; HR: hazard ratio; CI: confidence interval.
Age1.05 (1.02 - 1.08)< 0.0011.05 (1.01 - 1.08)< 0.011.05 (1.02 - 1.09)< 0.01
Female1.21 (0.58 - 2.54)0.62
Race (vs. Caucasian)0.26
  Hispanic1.67 (0.75 - 3.73)0.21
  Others0.54 (0.13 - 2.35)0.41
BMI (n = 105)0.98 (0.93 - 1.04)0.470.91 (0.84 - 0.98)0.020.92 (0.86 - 0.98)0.01
WBC1.09 (1.03 - 1.14)< 0.011.04 (0.97 - 1.12)0.281.07 (0.99 - 1.16)0.08
Hb0.89 (0.77 - 1.02)0.08
Hct0.98 (0.94 - 1.02)0.25
MCV1.01 (0.96 - 1.06)0.72
Platelet1.001 (0.998 - 1.004)0.41
Cr1.57 (1.32 -1.87)< 0.0001
INR1.07 (1.01 - 1.13)0.02
PT (n = 107)1.013 (1.005 - 1.020)< 0.01
AST1.000 (0.999 - 1.001)0.63
ALT1.000 (0.999 - 1.001)0.92
AP (n = 102)1.001 (0.998 - 1.004)0.56
Albumin0.38 (0.24 - 0.62)< 0.00010.63 (0.34 - 1.15)0.130.53 (0.30 - 0.96)0.03
< 3.59.47 (2.24 - 39.96)< 0.01
Total bilirubin1.04 (1.01 - 1.08)< 0.01
Na0.92 (0.88 - 0.97)< 0.010.95 (0.89 - 1.02)0.140.94 (0.88 - 0.996)0.04
Lactate (n = 8)1.19 (0.79 - 1.78)0.41
AST/ALT1.08 (0.83 - 1.42)0.56
ALT/AP (n = 102)0.98 (0.80 - 1.21)0.85
AST/AP (n = 102)1.01 (0.89 - 1.13)0.94
AAPI1 (n = 102)0.99 (0.92 - 1.07)0.85
AAPI2 (n = 102)1.00 (0.96 - 1.05)0.94
NR ALT/AP (n = 102)1.00 (0.96 - 1.04)0.85
NR AST/AP (n = 102)1.00 (0.98 - 1.02)0.94
FIB 41.01 (0.98 - 1.03)0.59
APRI1.00 (0.98 - 1.02)0.89
CTP (n = 87)1.44 (1.21 - 1.71)< 0.0001
Infection1.95 (0.93 - 4.07)0.08
Antibiotics use (Y/N)2.05 (0.96 - 4.38)0.06
MELD1.06 (1.03 - 1.08)< 0.00011.04 (1.01 - 1.08)< 0.01
≥ 215.81 (2.74 - 12.30)< 0.0001
≥185.30 (2.46 - 11.40)< 0.0001
DF1.003 (1.001 - 1.004)< 0.0011.003 (1.001 - 1.005)< 0.01
≥ 324.62 (2.14 - 9.99)0.0001
ANI1.02 (0.98 - 1.07)0.24
≥ 2.21.17 (0.52 - 2.65)0.71
IR1.37 (0.66 - 2.84)0.402.63 (1.08 - 6.40)0.032.84 (1.13 - 7.13)0.03

 

Table 4. Comparison of Various Models to Determine Important Factors to Predict Mortality
 
ModelFormulaUno’s C-statistic
Based on the comparisons in the difference of Uno’s C-statistic between all of the models, MELD ≥ 21 and DF ≥ 32 alone was significantly worse than model 1 - 14 with a P-value < 0.05. In contrast, the differences in Uno’s C-statistic between all the 14 models were not significantly different with a P-value > 0.05. MELD: model for end-stage liver disease; DF: discriminant function; IR: insulin resistance.
MELD ≥ 21NA0.65
DF ≥ 32NA0.64
Model 1h(t) = h0(t)exp(0.034 × Age + 1.474 × MELD ≥ 21)0.78
Model 2h(t) = h0(t)exp(0.037 × Age + 1.243 × DF ≥ 32)0.75
Model 3h(t) = h0(t)exp(1.337 × MELD ≥ 21 + 2.025 × Albumin < 3.5)0.76
Model 4h(t) = h0(t)exp(1.082 × DF ≥ 32 + 1.857 × Albumin < 3.5)0.74
Model 5h(t) = h0(t)exp(0.028 × Age + 1.154 × MELD ≥ 21 + 1.704 × Albumin < 3.5)0.82
Model 6h(t) = h0(t)exp(0.032 × Age + 0.907 × DF ≥ 32 + 1.737 × Albumin < 3.5)0.79
Model 7h(t) = h0(t)exp(0.034 × Age + 1.506 × MELD ≥ 21 + 0.412 × IR)0.73
Model 8h(t) = h0(t)exp(0.037 × Age + 1.223 × DF ≥ 32 + 0.210 × IR)0.73
Model 9h(t) = h0(t)exp(0.033 × Age + 1.445 × MELD ≥ 21 + 0.147 × Infection)0.78
Model 10h(t) = h0(t)exp(0.036 × Age + 1.208 × DF ≥ 32 + 0.204 × Infection)0.75
Model 11h(t) = h0(t)exp(0.034 × Age + 1.491 × MELD ≥ 21 + 0.401 × IR + 0.084 × Infection)0.73
Model 12h(t) = h0(t)exp(0.036 × Age + 1.197 × DF ≥ 32 + 0.186 × IR + 0.175 × Infection)0.73
Model 13h(t) = h0(t)exp(0.027 × Age + 1.191 × MELD ≥ 21 + 1.665 × Albumin < 3.5 + 0.315 × IR + 0.087 × Infection)0.79
Model 14h(t) = h0(t)exp(0.031 × Age + 0.885 × DF ≥ 32 + 1.718 × Albumin < 3.5 + 0.118 × IR + 0.099 × Infection)0.80