P335 Prevalence and risk factors for non alcoholic fatty liver disease in inflammatory bowel disease
Iannone A.*1, Losurdo G.2, Shahini E.1, Albano F.1, La Fortezza R.F.1, Rizzi S.F.1, Contaldo A.1, Barone M.1, Ierardi E.1, Principi M.1, Di Leo A.1
1University of Bari, DETO, Bari, Italy 2University of Bari, Department of Emergency and Organ Transplantation, Bari, Italy
Non alcoholic fatty liver disease (NAFLD) is responsible for up to 40% of hepatic alterations diagnosed in inflammatory bowel diseases (IBD). We aimed to evaluate the prevalence and risk factors for NAFLD in a large cohort of IBD patients and the stage of liver fibrosis by transient elastography (TE).
This study included consecutive patients affected by Crohn's disease (CD) or ulcerative colitis (UC), referred to our tertiary centre for IBD from December 2015 to July 2016. Patients with previous diagnosis of chronic liver disease with different etiology from NAFLD, daily alcohol consumption higher than 20 g in women and 30 g in men or conditions affecting the execution of TE were excluded. Data regarding IBD characteristics and liver laboratory tests were collected. The markers of metabolic syndrome were analyzed. All subjects underwent an abdominal ultrasonography to estimate presence/degree of steatosis and TE to measure liver stiffness (LS). An age and sex-matched group of controls was recruited. Student's t or Chi-squared tests were used where indicated. For multivariate analysis, multinomial logistic regression was used to determine which factors could have influenced the presence of NAFLD.
NAFLD was observed in 106 out of 378 (28%) patients with IBD and in 33 (20.1%) of 162 controls (p=0.04). The prevalence of diabetes and abdominal excessive adiposity was higher for IBD than controls. Patients with NAFLD were more frequently male, young and affected by diabetes, hypertension and insulin resistance. Their mean waist circumference and BMI were higher in NAFLD compared to non-NAFLD patients. Additionally, NAFLD subjects showed higher levels of transaminases and gamma-glutamyltranspeptidase, HDL cholesterol, triglycerides and fasting blood glucose. Finally, their mean LS was higher than in non-NAFLD patients. At multivariate analysis the risk of NAFLD in IBD was directly correlated to insulin resistance (odd ratio OR=14.73 p<0.0001), high waist circumference (OR=4.85 p=0.04), high BMI (OR=1.6 p=0.01), high gamma-glutamyltranspeptidase level (OR=3.9 p=0.04) and high fasting blood glucose (OR=1.3 p=0.04). Detailed aspects of our analysis are reported in the table.
Variable IBD (n=378) Controls (n=162) p value OR (95% CI) Age 46.3±15.6 43.8±17.8 0.10 NE Alcohol intake <10 g/day 10.8%, >10 g/day 0.7% <10 g/day 28.0%, >10 g/day 0% <0.0001 NE Diabetes 7.4% 3.7% 0.01 2.91 (1.24–6.81) Hypertension 17.7% 21.7 0.26 0.78 (0.50–1.20) Insulin resistance 1.6% 0% 0.08 6.61 (0.37–118.1) Abdominal circumference (>102 for male, >88 for female) 15.6% 6.8% 0.003 2.50 (1.34–4.69) Liver steatosis 28.0% 20.1% 0.04 1.55 (1.02–2.36) Liver stiffness 5.1±1.7 4.9±1.7 0.42 NE BMI 24.7±5.3 24.4±4.3 0.44 NE
Despite NAFLD is an increasing problem in IBD, it seems to be correlated to the presence of metabolic syndrome rather than to IBD characteristics.