Send to

Choose Destination
Diabetes Metab Res Rev. 2017 May;33(4). doi: 10.1002/dmrr.2878. Epub 2017 Jan 27.

Occurrence over time and regression of nonalcoholic fatty liver disease in type 2 diabetes.

Author information

Diabetes and Metabolism Unit ASL Torino 5, Chieri, Italy.
Unit of Metabolic Diseases and Clinical Dietetics, "Alma Mater Studiorum" University of Bologna, Bologna, Italy.
IRCCS Casa Sollievo della Sofferenza, Unit of Internal Medicine, San Giovanni Rotondo, Italy.
Center for Outcomes Research and clinical Epidemiology (CORE), Pescara, Italy.
Department of Clinical and Experimental Medicine, Policlinico Universitario "G. Martino", Messina, Italy.
AUSL Diabetes Unit Romagna, Ravenna, Italy.
Institut d'Investigacions Biomèdiques August PiiSunyer (IDIBAPS) and Centro de Investigación Biomédicaen Red de Diabetes y Enfermeda des Metabólicas Asociadas (CIBERDEM), Barcelona, Spain.
IRCCS MultiMedica, Milano, Italy.
Statistical Consultant for AMD c/o Associazione Medici Diabetologi, Rome, Italy.
AMD (Italian Association of Clinical Diabetologists), Rome, Italy.



This analysis was aimed to assess the incidence, regression, and correlated factors of nonalcoholic fatty liver disease (NAFLD) in type 2 diabetes, which are poorly known.


Nonalcoholic fatty liver disease (defined as fatty liver index [FLI] score ≥ 60) in patients with type 2 diabetes, and related factors was investigated in a nationwide database containing information from the Italian network of diabetes clinics. A 10% variation of FLI was the cut-off considered in the analyses of a cohort of 5030 patients, which was separately conducted for those who developed, maintained, or recovered from FLI-assessed NAFLD (FLI-NAFLD) over a 3-year period.


At baseline, FLI-NAFLD was diagnosed in 61.3% of patients. Within the 3-year study period, FLI-NAFLD occurred in 313 patients and remitted in 410. The FLI score remained unchanged in 4307. Body-mass index (odds ratio, 1.45 95%; confidence interval, 1.35-1.55), abdominal obesity (2.11; 1.64-2.72), low HDL cholesterol levels (1.38; 1.02-1.87), and triglycerides (1.20; 1.12-1.28) all emerged as notable negative prognostic factors for the development or maintenance of FLI-NAFLD. The regression rate of FLI-NAFLD was higher among patients who managed to partially control these factors. Male sex and established organ damage, especially kidney function (1.64; 1.12-2.42), were independent risk predictors. Unlike other diabetes complications, FLI-NAFLD was more frequent among younger patients or those with a shorter duration of diabetes.


FLI-assessed NAFLD is a dynamic condition, with about 5% of diabetic patients entering or leaving the status every year. Younger male patients with insulin resistance or organ damage have a higher risk of presenting with FLI-NAFLD at baseline, developing FLI-NAFLD within 3 years, and a lower probability of regression.


associated factors; hepatic steatosis; incidence; longitudinal analysis; recovery; type 2 diabetes

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center