1. Non-Alcoholic Fatty Liver Disease ("NAFLD")
    1. Abnormal deposition of fat in the liver in the absence of alcohol abuse
    2. Natural history
      1. In a case series of 173 patients with biopsy-proven NAFLD, followed for minimum of 5y (42% had NASH; see below), there was no sig. diff. in overall mortality between NASH and non-NASH pts, but NASH pts had higher liver-related mortality (17.5% vs. 2.7%) (Clin. Gastroent. Hepatol. 7:234, 2009--JW)
    3. Treatment
      1. In a study in 173 pts 8-17yo with biopsy-confirmed NAFLD randomized to vitamine E 800IU QD, metformin 1000mg QD, or placebo, x 96wks, incidence of (sustained reduction in ALT by 50% or moreor 40 U/l or more) was not sig. diff. in either of the active-treatment groups vs. placebo, though among the pts who had NASH (see below), a sig. higher proportion resolved at 96wks with vitamin E vs. placebo (58% vs. 28%; no sig. diff. between metformin and placebo) (JAMA 305:1659, 2011-abst)
  1. Non-Alcoholic Steatohepatitis ("NASH")
    1. Definition = liver biopsy changes consistent with alcoholic hepatitis in absence of excessive alcohol intake (fatty changes and hepatocyte injury, with or w/o fibrosis; may see cirrhotic changes).
    2. A subset of Non-Alcoholic Fatty Liver Disease (NASH represents the more severe end of the spectrum of NAFLD)
    3. 9% of liver biopsies meet this definition
    4. Risk factors:
      1. Obesity
      2. Dyslipidemias
      3. Diabetes Mellitus
    5. Clinical Features and Natural History
      1. Usually asymptomatic though modest aminotransferase elevations and hepatomegaly are common
      2. Progression, when present, is usually indolent.
      3. In an observational study in 129 pts with biopsy-proven non-alcoholic fatty liver disease and elevated transaminase levels, compared with an age- and sex-matched reference population, over mean 13.7y f/u, the subgroup with nonalcoholic steatohepatitis had sig. lower survival rates than reference population (70% vs. 80%); no sig. dif. between those with simple steatosis vs. reference group. (Hepatol. 44:865, 2006--JW)
      4. cirrhosis occurs in a small number of patient--Risk factors:
        1. Type 2 DM
          Age > 45yo
        2. BMI > 30
        3. AST/ALT ratio > 1
    6. Diagnosis
      1. Ultrasound
      2. CT
      3. Liver biopsy (unlike other two, allows determination of severity)
      4. Elevated "caspase activity" as measured by plasma levels of cytokeratin-18 fragments, may be an independent predictor of the existence of NASH on liver biopsy (FP News, 5/15/06)
    7. Clinical features and Natural History
    8. Treatment
      1. Treatment of underlying risk factors (see above)
        1. In a study in 55pts with nonalcoholic steatohepatitis (configurmed histologically) and either impaired glucose tolerance or type 2 DM randomized to pioglitazone vs. placebo, at 6mos, pioglitazone recipients had sig. reductions in serum transaminase levels and sig. greater histologic improvements (NEJM 355:2297, 2006--JW)
        2. Gemfibrozil and metformin have been shown to reduce transaminase levels
        3. In a study in 110 nondiabetic patients with non-alcoholic fatty liver disease randomized to metformin 2g/d vs. vitamin E 800IU/d vs. weight-loss dietary counseling x 12mos, metforming group had sig. greater ALT reductions and incidence of ALT normalization (56% vs. 18% for the combination of the other 2 groups); this association persisted after adjustment for weight loss during the trial period (Am. J. Gastroent. 100:1082, 2005--abst)
      2. Liver transplantation for end-stage cirrhosis
      3. 49 pts with biopsy-proven NASH randomized to vitamin E 100IU QD + vit. C 1000mg QD vs. placebo x 6mos.  Improvement in disease activity scores on liver biopsies at 6mos were sig. greater in vitamin-treated group. (Am. J. Gastroent. 98:2485, 2003--abst)
      4. 166 pts with bx-proven NASH randomized to 13-15mg/kg/d of ursodeoxycholic acid vs. placebo; over 2y f/u, no sig. diff. in bx findings between the groups (Hepatology 39:770, 2004--abst)
      5. In a nonrandomized trial in 18 pts with histologically-proven NASH and persistently elevated ALT (> 1.5 x normal), treated with pentoxifylline 400mg TID x 6mos, mean serum ALT and AST were sig. reduced (Am. J. Gastroent. 99:1946, 2004--abst)
      6. Vitamin E
        1. In a study in 247 pts with biopsy-proven NASH randomized to vitamin E 800 mg/d vs. pioglitazone 30mg/d vs. placebo x 96wks, vitamin E (but not pioglizatone) was associated with lower nonalcoholic fatty liver disease activity scores c/w placebo (Study by Arun Sanyal presented at conference of American Association for the Study of Liver Diseases 2009-FP News 12/09)
        2. See also above under "NAFLD"
      7. Thiazolidinediones
        1. In a study in 63 pts with biopsy-proven NASH (32% with DM) randomized to (rosiglitazone 4mg/d x 1mo then 8mg/d x 11mos) vs. placebo, at 12mos, the rosiglitazone group ahd sig. higher likelihood of having normal hepatic transaminases (38% vs. 75%) and at least a 30% reduction in steatosis (47% vs. 16%) (Gastroent. 135:100, 2008-JW)
        2. In a study in 61 pts with biopsy-proven NASH (none with DM) randomized to pioglitazone 30mg/d vs. placebo, at 12mos, the pioglitazone group had no sig. diff. from the placebo group in improvement in steatosis, but had sig. greater improvements in hepatocyte injury and fibrosis (Gastroent. 135:1176, 2008-JW)

(Sources include Mayo Clin. Proc. 75:733, 2000--AFP; NEJM 346:1221, 2002--AFP, and others as cited)