icon star paper   Articles  
Back grey_arrow_rt.gif
Fatty Liver Disease, ALT, & Overweight
  Determinants of the association of overweight with elevated serum alanine aminotransferase activity in the United States
Background & Aims: In the absence of other causes, overweight and obesity increase the risk of liver disease. We examined whether central adiposity and metabolic markers explain the association of body mass index (BMI as kg/m2) with abnormal serum alanine aminotransferase (ALT) activity in a national, population-based study.
Methods: Adult participants (5724) in the third U.S. National Health and Nutrition Examination Survey (1988-1994) underwent anthropometric measures and phlebotomy after an overnight fast. Participants with excessive alcohol consumption, hepatitis B, hepatitis C, iron overload, or known diabetes were excluded.
Results: Elevated ALT levels were found in 2.8% of the population. In univariate analysis, factors associated with elevated ALT levels (P < 0.05) included younger age, male sex, Mexican-American ethnicity, and higher BMI, waist-to-hip circumference ratio (WHR), and fasting serum leptin, triglyceride, insulin, and glucose concentrations. The proportion of elevated ALT activity due to overweight and obesity (BMI 25 kg/m2) was 65%. In multivariate logistic regression analysis, control for WHR, demographic factors, and glucose concentration diminished but did not eliminate the association of higher BMI with elevated ALT activity. After adding leptin and insulin concentrations, abnormal ALT activity was most strongly associated with higher WHR (odds ratio [OR], 1.32; 95% confidence interval [CI], 1.12-1.56) and leptin (OR, 1.12; 95% CI, 1.01-1.24) and insulin (OR, 1.27; 95% CI, 1.01-1.60) concentrations, whereas BMI was not independently related. Conclusions: In this large, national, population-based study, central adiposity, hyperleptinemia, and hyperinsulinemia were the major determinants of the association of overweight with elevated serum ALT activity.
Gastroenterology Jan 2003;124:71-79. Constance E. Ruhl, James E. Everhart
Defining nonalcoholic fatty liver disease: Implications for epidemiologic studies
Jeanne M. Clark, Anna Mae Diehl
In the United States, chronic liver disease and cirrhosis are the 10th leading cause of death overall, and among the top 7 leading causes of death in men and women ages 45-64. Despite recent advances in technology, physicians must still rely on the liver biopsy for diagnosing and particularly for staging liver disease. Liver function tests including serum transaminases and -glutamyltransferase are often used as screening tests among both low- and high-risk populations. However, there is ample evidence that these tests lack sensitivity and specificity for liver disease and cirrhosis. Thus, despite increased interest in a broad array of liver diseases, medical research has been hampered by a significant referral bias because of the need for liver biopsy for accurate diagnosis and staging.
In this issue of GASTROENTEROLOGY, Ruhl and Everhart6 attempt to expand our understanding of liver disease in the United States population, by presenting an analysis of data from the Third National Health and Nutrition Examination Survey (NHANES III). Using this data, they examine the prevalence of an abnormal alanine aminotransferase (ALT), defined as an ALT>43 U/L for men or women. By eliminating individuals with moderate-to-high alcohol consumption, hepatitis B or C, elevated transferrin saturation or a history of diabetes mellitus, the authors use the elevated ALT as a surrogate for nonalcoholic fatty liver disease (NAFLD).
After these exclusions, they found that 2.8% of the population had an elevated ALT. Elevated ALT was associated with younger age, male sex, Mexican-American ethnicity, impaired glucose metabolism and insulin resistance, obesity, and measures of central adiposity, as well as higher leptin, triglycerides, and C-peptide. In multivariate analyses central adiposity, insulin, and leptin concentrations were the most highly associated factors. They conclude that these factors are the major determinants of the association between elevated ALT concentration and higher body weight.
As with any good research, this article leaves the reader with more questions than answers. The first major question is how to study the epidemiology and risk factors of a disease for which the gold standard diagnostic test is an invasive procedure. Ruhl and other investigators conducting epidemiologic research have used serum aminotransferases. However, this has not been universally accepted, despite the lack of a superior alternative. Accepting that aminotransferases are the best, widely available serum markers of liver disease, it is unclear if ALT should be used alone, as in this article, or whether aspartate aminotransferase (AST) and ALT7,8 or other combinations of liver tests should be used. Furthermore, the level of elevation that is considered abnormal varies widely and has recently been brought into question.10 There is also debate as to whether or not different cutoffs are indicated for men and women. In fact, gender-specific norms for ALT were defined in the NHANES III manuals.11 Given all the controversy, it is clear that more acceptable and accurate means of noninvasively diagnosing liver disease in general, and NAFLD in particular, are needed.
The article by Ruhl and Everhart highlights the impact the current uncertainty in noninvasive diagnosis can have on epidemiological studies. Using ALT alone, they estimate the prevalence of NAFLD in the United States is 2.8%. This estimate is much lower than general population estimates of NAFLD in other countries, which range from 16%-20% based at least in part on ultrasound evidence for hepatic steatosis. It is also substantially lower than that predicted by a recent population-based study of seemingly healthy Italian blood donors.10 Moreover, using recently suggested lower normal values for ALT (>30 U/L for men and >19 U/L for women), Ruhl and Everhart found elevated levels in 12.4% of men and 13.9% of women, estimates significantly higher than their original 2.8%.
There are several reasons why Ruhl and Everhart may have arrived at a lower estimate of NAFLD than other population studies. First is their decision to use ALT alone as a marker for occult liver disease. Although many have suggested that ALT is higher that AST in NAFLD, an elevated ALT/AST ratio may be true only of early disease. Furthermore, in NAFLD, as in most other types of chronic hepatitis, ALT values fluctuate over time. Thus, diagnostic criteria that rely solely on a single elevated ALT value could underestimate disease prevalence. In addition, the authors used a higher cutoff value for normal ALT than many suggest (and one that is higher than published in the National Center for Health Statistics NHANES III manuals). Naturally, this would lower the estimated prevalence of occult liver disease. Furthermore, as the authors themselves acknowledge, ALT testing in NHANES III was conducted on serum that had been frozen. Freeze-thawing of samples may affect enzyme activity and artifactually lower ALT values, leading to an underestimation of disease burden in the population. Finally, the authors excluded people with known diabetes, a group who are at high risk for NAFLD. This eliminated a relatively large subpopulation that is likely to be enriched with fatty liver disease, and may have biased their conclusions. However, although several common causes of liver disease were excluded from this study, it is likely that not all of the remaining elevated ALT values were due to NAFLD. Some were probably caused by other conditions, such as autoimmune hepatitis. This possibility is supported by a recent biopsy study that demonstrated histologic proof of NAFLD or NASH in only about 65%-90% of patients with "cryptogenic" hepatitis. Nevertheless, Drs. Ruhl and Everhart's estimate of NAFLD prevalence and the associations they found are likely to be more indicative of the burden of NAFLD in the general population than information obtained from patients who are selected for further study based on referral to gastroenterology clinics. Even if the true prevalence of NAFLD in the United States population is merely 2.8%, fatty liver disease would be twice a common as chronic hepatitis C (1.3%).
This leads to the second question: what does a diagnosis of NAFLD really mean? Stated another way, is this condition merely NAFL, or is it truly NAFLD? Fatty liver is often considered to be an incidental condition with a benign prognosis. However, a growing body of literature contests this idea, indicating that the natural history of NAFLD, at least for some, includes progression to cirrhosis and to hepatocellular carcinoma. The latter data defy conventional wisdom, which suggests that most people with this condition remain asymptomatic for decades. Indeed, the preponderance of asymptomatic individuals with fatty liver has tempted some to conclude that fatty liver alone may be a variant of normalcy, rather than a true disease. However, applying similar logic would force one to conclude that asymptomatic individuals with a blood pressure of 180/100 mm Hg, a fasting blood sugar of 180 mg/dL, or low density lipoproteins of 200 mg/dL, while abnormal, do not have a disease state. Fortunately, epidemiological studies of hypertension, diabetes, and dyslipidemia have taught us that each of these diseases is indolently progressive, leading to clinically significantly morbidity and mortality in a subset of afflicted individuals decades after disease onset. The knowledge of subsequent serious outcomes drives the current practice of early therapeutic intervention in individuals with those diseases.
Certainly, before NAFLD is dismissed as quirk of nature, more studies are required to clarify its genuine, health-related outcomes. In addition to elucidating the natural history of portal hypertension-related morbidity and mortality, these should assess more subtle symptoms that may affect an individual's quality of life. Our own analysis of NHANES III data, as well as a recent prospective survey of NAFLD patients who were referred to the Mayo Clinic, suggest that individuals with NAFLD may not be truly asymptomatic because they describe their health status as significantly below average. In addition, if NAFLD is part of the metabolic syndrome, as these authors and others suggest, then it will be important to determine what pathogenic role fatty liver plays, relative to the other components of the syndrome. It will also be important to determine whether or not treatment of NAFLD improves the other aspects of the metabolic syndrome, and vice versa. Finally, more work is needed to identify which patients with fatty livers are most likely to progress from steatosis to steatohepatitis and cirrhosis. A better understanding of the factors that modulate liver disease progression is critical, so that we can target selected patients for more aggressive monitoring, lifestyle interventions, and pharmacotherapy.
In summary, it is important to emphasize that epidemiological studies are crucial to further our understanding of many diseases, particularly those that are indolently progressive over decades, like NAFLD. Ruhl and Everhart6 have made a major advance toward clarifying the burden of liver disease in the United States population. Their findings suggest that occult liver disease may be much more common in the general adult population than we previously suspected, affecting 4 to 5 million people. Moreover, evidence that such ALT abnormalities associate with other components of the metabolic syndrome supports the authors' conclusion that most of the abnormalities are probably due to NAFLD. Given the increasing prevalence of obesity in children, as well as adults, this has potentially important public health implications. The cynics among us will argue that these interpretations are premature without definitive histologic proof of fatty liver. Others remain skeptical that, even if valid, the diagnosis of NAFLD rarely portends a real risk for clinically significant adverse health outcomes. The debate is likely to smolder until better, noninvasive tests to diagnose and stage NAFLD are developed and used to identify and track afflicted individuals over time. Until then, it seems prudent to place NAFLD in the same health risk category as other components of the metabolic syndrome, and to consider it no more or less worthy of medical attention than its "bed fellows," i.e., obesity, diabetes, hypertension, and dyslipidemia. Sometimes, "still waters run deep."
  icon paper stack View Older Articles   Back to Top   www.natap.org