The association between development of liver disease and obesity in low-alcohol-consuming individuals was described a few decades ago. However, only in the last 10 years have the importance and the scale of the problem been realized and linked to the epidemic of obesity currently sweeping the globe.
Non-alcoholic fatty liver disease (NAFLD), represents a spectrum of illness. The earliest stage is steatosis, which represents fat accumulation in liver tissue without inflammation. Non-alcoholic steatohepatitis (NASH), the next stage, is characterized by steatosis and inflammation with or without fibrosis. The former stage is usually benign while the latter stage has higher risk of progression to cirrhosis. Currently, only histological examination is able to differentiate between the two stages.
One of the major risk factors for the accumulation of excess liver fat is obesity leading to insulin resistance. Therefore, the prevalence of NAFLD increases in parallel with weight or BMI (body mass index). The prevalence of NAFLD in unselected populations from developed countries is estimated as 20 to 30%. About 2–3% of the same population will have NASH. In obese individuals (BMI > 30 kg/m2) the prevalence of steatosis is about 65-75% and the prevalence of NASH increases to 15-20%.1
Most individuals with NAFLD are asymptomatic. The diagnosis is often made following abnormal findings on routine biochemistry or following abdominal ultrasound performed for another reason. Symptoms, when present, may include fatigue and right upper quadrant pain. The most commonly reported clinical finding is hepatomegaly.
If cirrhosis eventually develops prior to diagnosis of NASH, presentation is similar to other causes. NAFLD is now recognized as the most common cause of cryptogenic cirrhosis and has been the underlying diagnosis in about 10% of liver transplant cases.
Predictors for NAFLD are listed in Table 1. Fatty liver is associated with elevated serum ALT and GGT. But neither is sufficiently sensitive nor specific for a diagnosis of NAFLD. It’s important to emphasise that AST can be higher than ALT in cirrhosis. Therefore, in a patient with known NAFLD, a rising AST can be a bad prognostic sign. ALT and GGT appear to correlate with the amount of liver fat present as measured by MRI or ultrasound (US).
A combination of serum adiponectin, HOMA-IR (homoeostasis model assessment of insulin resistance) and type IV collagen 7S was shown to have a sensitivity of 94% and specificity of 74% for identifying early NASH. In another study, hyaluronic acid levels were significantly different between steatosis and NASH. Moreover, HA was the strongest independent predictor of severe fibrosis.1,2
Steatosis may be diagnosed by US, CT or MRI scan. Ultrasound is the cheapest option and has been reported to have a sensitivity of 89% and specificity of 93% for the identification of fatty liver. Drawbacks to US, however, include the requirement that at least 30% of the hepatocytes are fat-filled, and in the morbidly obese the performance of US is considerably weaker. MRI allows quantification of liver fat, which is useful in clinical studies. But, currently, imaging has no role to distinguish between the various stages of NAFLD. The additional benefit of imaging investigations to diagnose NAFLD in an individual with clinical features of insulin resistance and mild ALT elevation (ALT < x3 upper limit of normal) is questionable. Ultrasound should be performed in those with moderate or marked elevations in serum liver enzymes and those with persistent liver enzyme elevations despite adequate weight loss, and in cases of diagnostic doubt.2
Histology continues to provide the only proven method to diagnose NAFLD, to distinguish between NASH and steatosis and to provide prognostic information. At present, there’s no clear guidance for biopsy in NAFLD. Liver biopsy has clinical risk and has the potential for false negatives. Moreover, there’s still no consensus regarding histological criteria for classification of NAFLD.
In order to reduce the need for liver biopsy a few algorithms have been published that can detect individuals with a higher likelihood of fibrosis in NASH. For example, significant fibrosis was found in patients above age 45 years and who have abnormal liver enzymes with no clear explanation, BMI > 28 kg/m2, hypertriglyceridemia and type 2 diabetes.
Transient elastography (Fibroscan®), a non-invasive technique used to measure liver tissue stiffness, can provide information on the severity of fibrosis. A few studies have shown a good correlation between the histological staging of fibrosis and Fibroscan® results. This led to a suggestion that the number of liver biopsies might be reduced. However, significant intraobserver variability has been reported for this device and therefore, at present, Fibroscan® isn’t reliable for diagnosis of fibrosis in patients with fatty liver. Biopsy should be limited to those more likely to have high fibrosis stage. These are people of age > 45 years, AST/ALT ratio > 1, with type 2 diabetes and BMI > 30 kg/m2; and those with persistent elevations of serum liver enzymes despite weight loss; and in case of diagnostic doubt.2
Iron overload is sometimes seen in fatty liver but usually it’s not severe enough to warrant therapy. Hepatitis C and in some cases the HFE gene for hemochromatosis should be ruled out.
Pathogenesis of NAFLD/NASH
Causes of fatty liver are listed in Table 2. But development of NAFLD, progression to NASH, cirrhosis, and in some cases, hepatocellular carcinoma can take 20-40 years. The advanced part of this process may be influenced by genes. That suspicion is based on the observation that only some 10% of patients with NAFLD progress to NASH. Genetics applies to the inheritance of type 2 diabetes, dyslipidemia and other specific aspects of inflammatory cytokines and receptors.3,4
The NAFLD stage depends on a number of interrelated metabolic events. Accumulation of triglycerides into histological macrovesicles depends on impaired free fatty acid (FFA) oxidation, enhanced FFA delivery and extraction, increased de-novo synthesis and limited very low density lipoprotein release. New production is based on glucose, fructose or protein availability. About 15% of the fat is derived from diet, about 25% from new synthesis and the rest from circulating nonesterified FFA.5 Insulin is involved through regulation of FFA oxidation and glucose mediated lipogenesis. In cases of insulin resistance lipogenesis and failure of β-oxidation go unabated. Interventions aimed at increasing insulin sensitivity could reverse this first stage.
Progression to the more severe histologic stage of NASH with inflammation, ballooning hepatocytes, Malory bodies and early fibrosis is generally thought to follow a second hit, which may be oxidative stress. A new development in our thinking about pathogenesis is the contribution of intestinal microflora. First, microflora is shown to contribute to obesity.6,7 Second, bacterial translocations from the gut stimulate inflammatory processes which activate a number of biochemical steps.8 Ongoing insulin resistance also leads to activation of a number of key kinases that affect insulin receptor signalling. Some of the activated kinases lead to oxidative stress or direct activation of TNFα and inflammation. Concomitant enhanced expression of P450(CYP)2E1 (similar to that activated by chronic alcohol consumption) aggravates oxidative stress. These processes are accompanied by reduction of adiponectin, which counteracts insulin effects. Progressive accumulation of fat leads to lipotoxicity and cell death. This subsequently releases oxidized lipids and further increases oxidative stress. Progression of NASH aggravates mitochondrial injury, oxidative stress and fibrogenesis. Some authors have suggested that once this process is initiated without therapy, liver-related complications begin in about 7 years and may affect 30-60% of patients.9 As for cancer, while several reports have appeared on the association of hepatocellular carcinoma and end stage NASH, the relationship is still debated. Certainly this tumour is more common with viral diseases like hep C and B.
Treatment of fatty liver
Therapy of fatty liver begins with recognition of the condition. Since NAFLD/NASH is often accompanied by obesity, type 2 diabetes and metabolic syndrome, and other cardiovascular risk factors, treatment may also target these as well. Treatment is divided into diet and lifestyle changes, medications or surgery.
Intuitively, diet is the cornerstone of treatment. There’s evidence that a modest weight loss of 7-12% over a period of 6-12 months is associated with improved liver enzymes, insulin sensitivity and possible reduction of liver fat.1,5
Rather than “fad” diets, the goal should be total caloric restriction. Cutting total fat and carbohydrates may also be beneficial. Because saturated fatty acids increase oxidative stress these likely should also be curtailed. In addition, simple sugars and high fructose containing foods should be specifically avoided, as these are thought to aggravate obesity. Finally, alcohol consumption should be restricted in individuals with risk factors.10 But rapid weight loss should be avoided, since rarely this may precipitate subfulminant hepatic failure. The recommended weekly loss should not exceed 1 kilo (2 pounds).
In conjunction with diet, increased physical activity independently helps to improve insulin sensitivity.1,5 The recommended time is about 30-45 minutes a day and could include a brisk walk or aerobic exercises. Unfortunately, there’s very limited proof in controlled trials on the efficacy of these lifestyle interventions.
Medical therapy is divided into agents designed to help with weight loss, improve insulin sensitivity, interfere with hepatic fatty acid synthesis or gluconeogenesis or antioxidants.
Trials of various agents are listed in Table 3. Drugs which may aid in weight loss include orlistat, which inhibits gastric and pancreatic lipase, interfering with fat absorption; and rimonabant (Endocannabinoid receptor 1 antagonist). Side effects make both unsuitable for regular use.10 The drug metformin, which inhibits hepatic gluconeogenesis and increases insulin sensitivity, has shown a modest benefit in improving liver enzymes and features of liver histology.10 Thiazolidinediones, which improve insulin sensitivity, have met with some limited success, although troglitazone was withdrawn because of hepatotoxicity. Statins may help by reducing dyslipidemia, and their use is safe even with mild liver test abnormalities.11
The antioxidant vitamin E used in different doses up to 1,000 IU has had varying success.11 Most recently, 800 IU showed significant benefit on transaminases and histology for NASH.12 Overall the benefits for insulin sensitizing drugs and vitamin E are modestly supported and side effects are minimal.11,12
For patients failing diet and medical therapy, bariatric surgery holds promise. Here Roux- en –Y bypass or laparoscopic gastric banding has shown benefit over 18-24 months. Obesity, insulin sensitivity and fatty liver all improve, but this option may be the last resort.1,5
The last word
While there’s no proven treatment for NASH, weight reduction and select medical treatments hold promise. Surgery is a last resort for those in whom other efforts fail. Ultimately, NAFLD is a problem paralleling the epidemic of obesity and its prevalence could be reduced by efforts to reduce weight. Perhaps we could learn from the preparation of that delicacy, pâté de foie gras, and realize that what’s bad for the goose is also bad for the gander.
Nir Hilzenrat, MD, CSPQ is a hepatologist at the Jewish General Hospital in Montreal. His research interests are mainly related to clinical hepatology.
Andrew Szilagyi, MD, FRCPC is a gastroenterologist, also at the Jewish General Hospital. He is also a Staff Teacher at McGill University. He has research interests in both gastroenterology and hepatology.