Modification Patterns of Urinary Albumin Correlates With Serum Albumin and Outcome in Severe Alcoholic Hepatitis.

Abstract

Albumin modifications and deranged functions are well documented in serum of severe alcoholic hepatitis (SAH). We investigated whether urinary albumin (u-Alb) can serve as surrogate marker of circulatory albumin phenotype, functionality, and could predict outcome in SAH patients. Baseline serum and urine samples from 100 SAH, 20 alcoholic cirrhosis, and 20 healthy controls were subjected to u-Alb, ischemia modified albumin (IMA), IMA to albumin ratio (IMAr), advanced oxidation protein products, advanced glycation end-products, albumin-binding capacity determination. In addition, SAH urinary samples were also analyzed at day 4 and day 7 to predict nonresponse to corticosteroid therapy. Urine and serum levels of IMA, advanced oxidation protein products and advanced glycation end-products were higher (P<0.05) in SAH versus alcoholic cirrhosis and healthy controls. IMAr was low in urine but high in serum of SAH (P<0.05). Albumin-binding capacity was lower (P<0.05) in both urinary and serum albumin of SAH. Urinary and serum albumin parameters showed direct correlation, whereas IMAr showed inverse correlation (cc>0.2, P<0.05). Baseline u-Alb level was significantly higher in SAH, and was correlated directly with corticosteroid treatment outcome and 12-month mortality in SAH. Baseline u-Alb showed an area under the receivers operating curve analysis of 0.7 and a hazard ratio of 1.23 for prediction of 12-month mortality in SAH. Baseline u-Alb level >9.0 mg/dL was associated with reduced 12-month survival in SAH (log rank <0.01). u-Alb modifications are reflective of serum albumin modifications. Further baseline u-Alb levels could be exploited to predict steroid response and mortality in SAH patients.