Prevalence and risk factors of nonalcoholic fatty liver disease in patients with inflammatory bowel diseases: A cross-sectional and longitudinal analysis

Table 1 Comparison of baseline characteristics between the subgroups of patients with nonalcoholic fatty liver disease versus those without nonalcoholic fatty liver disease in Crohn’s disease

Variable	NAFLD (n = 145)	No NAFLD (n = 157)	P value
Male, n (%)	75 (52)	74 (47)	0.495
Age at abdominal ultra sound (yr), median (IQR)	49.0 (20.0)	35.0 (19.0)	< 0.001c
BMI (kg/m²), median (IQR)	28.0 (7.6)	23.0 (4.4)	< 0.001c
Diabetes mellitus, n (%)	7 (5)	1 (1)	0.057
Dyslipidemia, n (%)	12 (8)	3 (2)	0.0131
Hypertension, n (%)	50 (34)	20 (13)	< 0.001c
Current smoking, n (%)	35 (24)	29 (18)	0.288
Smoking in the past, n (%)	28 (19)	31 (20)	1.000
Smoking in general, n (%)	63 (44)	60 (38)	0.420
HBI (IQR), n = 217	3.0 (5.0)	1.0 (3.0)	< 0.001c
HBI > 4, n (%), n = 217	46 (40)	15 (15)	< 0.001c
Inflammation in colonoscopy, n (%), n = 83	27 (61)	20 (51)	0.330
Inflammation in bowel ultrasound, n (%), n = 101	54 (82)	47 (85)	0.772
CRP, mg/L (IQR), n = 293	2.5 (7.0)	3.0 (7.6)	0.846
White blood cells, /nl (IQR), n = 293	7.8 (3.0)	7.8 (2.8)	0.737
Fecal calprotectin, µg/g (IQR), n = 172	111 (230)	112 (249)	0.638
γGT, U/l (IQR)	24.0 (25.0)	16 .0(16.5)	< 0.001c
Elevated transaminases, n (%), n = 260	18 (16)	9 (6)	0.069
Disease duration at abdominal ultrasound (yr), median (IQR)	16.0 (21.0)	11.0 (12.0)	0.002b
History of bowel resection(s), n (%)	72 (50)	54 (45)	0.010a
Biological treatment at the time of ultrasound, n (%)	104 (72)	103 (66)	0.308
Treatment with adalimumab at the time of ultrasound, n (%)	53 (37)	57 (36)	0.767
Treatment with infliximab at the time of ultrasound, n (%)	29 (20)	21 (13)	0.164
Treatment with vedolizumab at the time of ultrasound, n (%)	11 (8)	10 (6)	0.850
Treatment with ustekinumab at the time of ultrasound, n (%)	14 (10)	18 (11)	0.746
Treatment with azathioprine at the time of ultrasound, n (%)	19 (13)	18 (11)	0.796
Treatment with budesonide at the time of ultrasound, n (%)	7 (5)	12 (8)	0.576
Treatment with systemic steroids < 7.5 mg at the time of ultrasound, n (%)	6 (4)	8 (5)	0.903
Treatment with systemic steroids > 7.5 mg at the time of ultrasound, n (%)	10 (7)	15 (10)	0.530
Treatment with systemic steroids at the time of ultrasound, n (%)	14 (10)	21 (13)	0.407
Treatment with mesalazine at the time of ultrasound, n (%)	25 (17)	19 (12)	0.271
Treatment with sulfasalazine at the time of ultrasound, n (%)	13 (9)	3 (2)	0.013a

^aP < 0.05.

^bP < 0.01.

^cP < 0.001. BMI: Body mass index; CRP: C-reactive protein; γGT: Gamma-glutamyl transferase; HBI: Harvey bradshaw index; IQR: Interquartile range; NAFLD: Nonalcoholic fatty liver disease.

Table 2 Results from multivariable logistic regression model (backwards elimination) to identify risk factors for nonalcoholic fatty liver disease in Crohn’s disease patients

Parameter	Odds ratio	95%CI	P value
Intercept	< 0.01	< 0.01; < 0.01	< 0.001c
Elevated HBI	3.80	1.75-8.70	0.001b
Age at abdominal ultrasound (yr)	1.05	1.02-1.08	< 0.001c
BMI (kg/m²)	1.25	1.15-1.37	< 0.001c
History of bowel resection(s)	1.47	0.74-2.92	0.274
Hypertension	0.81	0.31-2.09	0.668
Current smoking	1.17	0.51-2.69	0.707
Dyslipidemia	3.48	0.52-39.13	0.245

^bP < 0.01.

^cP < 0.001. BMI: Body mass index; CRP: C-reactive protein; HBI: Harvey Bradshaw Index; OR: Odds ratio; CI: Confidence interval.

Table 3 Number of abdominal ultrasound examinations per patient

Ultrasound examinations (n)	CD patients	UC patients
1	119	83
2	62	50
3	57	31
4	53	21
5	48	14
6	37	9
7	24	14
8	17	17
9	14	7
10	4	13
11	2	3
12	2	1

CD: Crohn’s disease; UC: Ulcerative colitis.

Table 4 Results of mixed model analysis to identify factors associated with nonalcoholic fatty liver disease in Crohn’s disease patients

Confounder model
Variable	Estimate	95%CI	P value
Female sex	0.70	0.09-1.31	0.293
Age at abdominal ultrasound (in years)	1.07	1.04-1.09	< 0.001c
BMI at abdominal ultrasound (in years)	1.32	1.30-1.34	< 0.001c
Diabetes mellitus	2.36	0.11-4.61	0.509
Hypertension	0.78	-0.07-1.62	0.586
Dyslipidemia	10.82	8.85-12.80	0.049a
Current smoking	1.91	1.28-2.53	0.109
Steroid use	0.74	0.27-1.21	0.287
Parameters included in model with age and BMI
Elevated HBI	2.99	2.41-3.58	< 0.001c
Inflammation in colonoscopy	16.40	14.59-18.11	0.002b
Inflammation in bowel ultrasound	1.46	1.30-1.61	0.344
Elevated CRP concentration	1.17	1.01-1.33	0.502
Elevated white blood cell count	0.96	0.10-1.12	0.295
Elevated fecal calprotectin concentration	1.34	1.19-1.50	0.455
History of bowel resection(s)	2.03	1.42-2.64	0.023a
Biologic treatment	1.01	0.85-1.17	0.959
Adalimumab treatment	1.09	0.93-1.25	0.709
Infliximab treatment	1.64	1.48-1.79	0.135
Vedolizumab treatment	1.40	1.24-1.56	0.523
Ustekinumab treatment	1.13	0.01-2.25	0.833
Golimumab treatment	1.34	1.19-1.50	0.874
Azathioprine therapy	1.98	1.36-2.61	0.031a
Methotrexate treatment	1.15	0.99-1.30	0.843
6-Mercaptopurine treatment	1.08	0.92-1.24	0.940
Sulfasalazine treatment	0.53	0.38-0.69	0.310
Mesalazine treatment	1.14	0.98-1.29	0.688

^aP < 0.05.

^bP < 0.01.

^cP < 0.001. BMI: Body mass index; CRP: C-reactive protein; HBI: Harvey bradshaw index; CI: Confidence interval.

Table 5 Comparison of baseline characteristics between the subgroups of patients with nonalcoholic fatty liver disease versus those without nonalcoholic fatty liver disease in ulcerative colitis

Variable	NAFLD (n = 68)	No NAFLD (n = 85)	P value
Male, n (%)	40 (59)	37 (44)	0.086
Age at abdominal ultrasound (yr), median (IQR)	48.5 (18.5)	37.0 (21.0)	0.005b
BMI (kg/m²), median (IQR)	28.0 (5.7)	22.8 (5.3)	< 0.001c
Diabetes mellitus, n (%)	6 (9)	1 (1)	0.057
Dyslipidemia, n (%)	2 (3)	2 (2)	1.000
Hypertension, n (%)	22 (32)	6 (7)	< 0.001c
Current smoking, n (%)	4 (6)	1 (1)	0.242
Smoking in the past, n (%)	20 (30)	11 (13)	0.021a
Smoking in general, n (%)	24 (35)	12 (14)	0.004b
SCCAI (IQR), n = 124	2.0 (4.0)	1.0 (3.0)	0.253
SCCAI > 4, n (%), n = 124	14 (24)	9 (14)	0.156
Inflammation in colonoscopy, n (%), n = 45	15 (60)	9 (45)	0.123
Inflammation in bowel ultrasound, n (%), n = 28	10 (91)	15 (88)	0.812
CRP, mg/l (IQR), n = 146	2.9 (8.0)	1.0 (5.1)	0.114
White blood cells, /nl (IQR), n = 147	7.5 (2.9)	7.0 (3.2)	0.300
Fecal calprotectin, µg/g (IQR), n = 87	103 (390)	147 (359)	0.612
γGT, U/l (IQR)	23.0 (40.0)	18.0 (25.0)	0.007b
Elevated transaminases, n (%), n = 147	8 (13)	7 (8)	0.387
Disease duration at abdominal ultrasound (yr), median (IQR)	9.0 (12.5)	10.0 (30.5)	0.408
History of bowel resection(s), n (%)	6 (9)	12 (14)	0.449
Biological treatment at the time of ultrasound, n (%)	33 (49)	52 (61)	0.161
Treatment with adalimumab at the time of ultrasound, n (%)	8 (12)	13 (15)	0.694
Treatment with infliximab at the time of ultrasound, n (%)	7 (10)	20 (24)	0.055
Treatment with vedolizumab at the time of ultrasound, n (%)	19 (28)	19 (22)	0.544
Treatment with azathioprine at the time of ultrasound, n (%)	13 (19)	14 (16)	0.831
Treatment with budesonide at the time of ultrasound, n (%)	3 (4)	3 (4)	1.000
Treatment with systemic steroids < 7.5 mg at the time of ultrasound, n (%)	7 (10)	5 (6)	0.480
Treatment with systemic steroids > 7.5 mg at the time of ultrasound, n (%)	10 (15)	12 (14)	1.000
Treatment with systemic steroids at the time of ultrasound, n (%)	13 (19)	16 (19)	1.000
Treatment with mesalazine at the time of ultrasound, n (%)	42 (62)	42 (49)	0.173
Treatment with sulfasalazine at the time of ultrasound, n (%)	2 (3)	5 (6)	0.634

^aP < 0.05.

^bP < 0.01.

^cP < 0.001. BMI: Body mass index; CRP: C-reactive protein; γGT: Gamma-glutamyl transferase; IQR: Interquartile range; NAFLD: Nonalcoholic fatty liver disease; SCCAI: Simple Clinical Colitis Activity Index.

Table 6 Results from multivariable logistic regression model (backwards elimination) to identify risk factors for nonalcoholic fatty liver disease in ulcerative colitis patients

Parameter	Odds ratio	95%CI	P value
Intercept	< 0.01	< 0.01; < 0.01	< 0.001c
Age at abdominal ultrasound (yr)	1.01	0.98-1.04	0.689
BMI (kg/m²)	1.25	1.14-1.38	< 0.001c
Hypertension	3.01	0.99-10.04	0.058
Adalimumab therapy	2.10	0.68-7.20	0.213

^cP < 0.001. BMI: Body mass index; OR: Odds ratio; CI: Confidence interval.

Table 7 Results from multivariable logistic regression model (backwards elimination) to identify risk factors for nonalcoholic fatty liver disease in ulcerative colitis patients

Confounder model
Variable	Estimate	95%CI	P value
Female sex	0.62	0.50-0.65	0.159
Age at ultrasound (in years)	0.99	0.98-1.03	0.637
BMI at ultrasound (in years)	1.28	0.71-1.85	< 0.001c
Diabetes mellitus	2.14	0.95-4.79	0.441
Hypertension	2.89	2.51-4.10	0.037a
Dyslipidemia	1.46	0.98-2.84	0.695
Current smoking	1.77	1.26-2.40	0.525
Steroid use	0.88	0.20-1.54	0.652
Parameters included in model with age and BMI
Elevated SCCAI	1.58	1.42-1.74	0.145
Inflammation in colonoscopy	3.60	2.69-4.50	0.005b
Inflammation in bowel ultrasound	2.31	2.16-2.47	0.162
Elevated CRP concentration	0.99	0.83-1.14	0.959
Elevated white blood cell count	1.01	0.85-1.17	0.782
Elevated fecal calprotectin concentration	1.06	0.90-1.22	0.923
History of bowel resection(s)	1.66	1.50-1.82	0.332
Biologic treatment	0.69	0.53-0.85	0.167
Adalimumab treatment	0.74	0.58-0.90	0.426
Infliximab treatment	0.77	0.62-0.93	0.484
Vedolizumab treatment	1.14	0.98-1.30	0.739
Ustekinumab treatment	0.19	0.03-0.35	0.407
Golimumab treatment	1.14	0.98-1.30	0.892
Azathioprine therapy	1.19	1.03-1.35	0.597
Methotrexate treatment	1.50	1.34-1.66	0.734
6-Mercaptopurine treatment	0.94	0.79-1.10	0.947
Sulfasalazine treatment	1.43	1.27-1.58	0.491
Mesalazine treatment	0.88	-0.25-2.00	0.635

^aP < 0.05.

^bP < 0.01.

^cP < 0.001. BMI: Body mass index; CRP: C-reactive protein; SCCAI: Simple Clinical Colitis Activity Index; CI: Confidence interval.