Epidemiology, Natural History, and Evaluation of Nonalcoholic Fatty Liver Disease

Prevalence

The true worldwide prevalence of NAFLD is not known as estimates differ between the populations studied because of different ethnicities, different dietary patterns, and the different sensitivities of the modalities used to detect the disease. One of the most accurate noninvasive techniques to quantify hepatic lipid accumulation is proton magnetic resonance spectroscopy ( ¹ H-MRS), with NAFLD being defined as an intrahepatic triglyceride content greater than 5.56%. Sampling 2349 U.S. adults from the North American general population using this ¹ H-MRS criterion, the Dallas Heart Study found NAFLD present in approximately 31% of individuals across ethnicities (45% of Hispanics, 33% of whites, 24% of blacks). This was slightly higher than the level found in a recent meta-analysis incorporating more than 8.5 million individuals from 45 different international studies, which estimated that NAFLD affects approximately 25% of the world's adult population. In the meta-analysis, prevalence differed by continent but was broadly similar in the United States and Europe (24.1% and 23.7%, respectively), with higher levels reported in the Middle East, South America, and Asia (31.8%, 31.5%, and 27.4%, respectively) and a lower level reported in Africa (13.5%). Studies also indicate that the prevalence of NAFLD differs significantly with sex (42% for white males vs. 24% for white females). Estimates of NASH prevalence are harder to establish as they require liver biopsy and so may be subject to greater selection and ascertainment bias. On the basis of histologic study of apparently healthy prospective living liver donors, the prevalence of NASH has been estimated to be 3% to 16% and 6% to 15% in the European and the U.S. populations, respectively. In the meta-analysis described above, 7% to 30% of histologically characterized NAFLD patients without a clear indication for liver biopsy were found to have NASH, equating to an overall global prevalence of NASH of between 1.5% and 6.5%.

The prevalence of NAFLD increases dramatically when groups with known metabolic syndrome risk factors are considered. As an illustration of this, 91% of obese people (body mass index [BMI] ≥30 kg/m ² ), 67% of overweight individuals (BMI 25 kg/m ² to 30 kg/m ² ), and 25% of normal weight individuals in an unselected European population sample had NAFLD. In bariatric surgery cohorts the prevalence of NAFLD has been reported to be approximately 73% to 97%, with NASH present in 25% to 33% of cases. Between 40% and 70% of patients with T2DM have NAFLD.

Whereas the prevalence of most liver diseases is stable, the prevalence of NAFLD is increasing, placing a greater burden on healthcare resources.

Causes of and Conditions Associated With Nonalcoholic Fatty Liver Disease

The overwhelming majority of NAFLD encountered in routine clinical practice occurs on a background of adiposity and is characterized by underlying IR as a pathophysiologic hallmark. Pooled data indicate that 43% of all patients with NAFLD (rising to 71% of all NASH patients) have multiple cardiovascular risk factors and so fulfill the diagnostic criteria for metabolic syndrome.

Among patients with NAFLD and NASH, the prevalence of T2DM is 23% and 44%, respectively, that of dyslipidemia is 69% and 72%, respectively, and that of hypertension is 39% and 68%, respectively. The relationship between NAFLD and T2DM is complex and bidirectional: NAFLD is an optimal biologic milieu in which T2DM can develop but, in addition, the presence of T2DM favors progression to NASH, cirrhosis, and HCC. Studies have consistently demonstrated that diabetes is associated with NASH and advanced fibrosis among patients with NAFLD. Because of the link between T2DM and progressive NAFLD, more careful surveillance of patients with T2DM may be indicated.

NAFLD is also associated with a number of other conditions that may contribute to more aggressive disease progression. These include:

• Polycystic ovary syndrome, a common endocrinopathy affecting 5% to 8% of premenopausal women and associated with increased IR. Several studies have shown an increased prevalence of NAFLD in polycystic ovary syndrome, with a North American study reporting NAFLD in 55% of women with polycystic ovary syndrome, almost 40% of whom were not overweight.

• Obstructive sleep apnea (OSA), which affects approximately 4% of the population; however, the prevalence rises to 35% to 45% among obese patients. The occurrence of chronic intermittent hypoxia in patients with OSA is thought to contribute to liver injury and inflammation. Several studies have demonstrated an association between OSA and NAFLD, with recent evidence also indicating that OSA is an independent risk factor for advanced fibrosis.

• Small bowel bacterial overgrowth, which frequently occurs in patients with diabetes and is thought to increase the risk of disease progression because of increased hepatic exposure to bacterial endotoxin via the portal vein.

As defined in Table 26-2 , secondary hepatic steatosis and steatohepatitis may be attributable to numerous acquired exposures as well as numerous rare monogenic inherited disorders; these are summarized in Table 26-3 .

Progression to Advanced Liver Fibrosis and Cirrhosis in Nonalcoholic Fatty Liver Disease

There remains much uncertainty about the natural history and prognosis of NAFLD. However, in recent years data have emerged that provide new insights into disease progression and challenge the long-standing dogma that simple steatosis (NAFL) is a benign condition with little risk of clinical sequelae and that NASH is the only prognostically relevant form of disease. These data come from two types of study:

• Serial liver biopsy studies , which examine histologic disease progression in patients who have undergone multiple liver biopsies over intervals of several years. These provide detailed information on how NAFL and NASH evolve and the rate at which fibrosis progresses.

• Longitudinal follow-up studies , which report the clinical course of the disease by following well-defined patient cohorts and determine prognosis and risk of specific disease end points or clinically relevant outcomes such as transplant.

With regard to the serial biopsy studies, a systematic review and meta-analysis of 11 small studies, together comprising 411 patients with histologically characterized NAFLD (150 with steatosis, 261 with NASH) provided more than 2,145 person-years of follow-up. Demonstrating the dynamic nature of hepatic fibrosis in NAFLD, fibrosis during this time progressed in 34% of patients and remained stable in 43% of patients, and some regression of liver fibrosis was observed in the remaining 23% of patients. In that meta-analysis, NAFL patients without evidence of fibrosis at the baseline exhibited a mean fibrosis progression rate of 0.07 stages per annum. NASH patients exhibited more rapid fibrosis, progressing at 0.14 fibrosis stages per annum. This corresponds to an approximately one-stage increase in fibrosis every 14 years for NAFL and every 7 years for NASH. Importantly, irrespective of whether the index biopsy showed NAFL or NASH, approximately 20% of patients exhibited rapid fibrosis progression, progressing from Stage 0 to Stage 3/4 ( rapid progressors ), whereas the remaining 80% exhibited little or no fibrosis progression ( slow progressors ). A separate large single-center study corroborated these results. In a cohort of 108 NAFLD patients undergoing repeat liver biopsy at a median interval of 6.6 years, 42% of patients had fibrosis progression, 40% had stable fibrosis, and 18% had fibrosis regression. All those patients who exhibited progressive fibrosis had also developed NASH by the time of the follow-up biopsy, underlining the biological importance of steatohepatitis for disease progression. Importantly, when those patients with NAFL on baseline biopsy were compared with those with NASH, no significant difference in the proportion exhibiting fibrosis progression was observed (37% vs. 43%), once again demonstrating that the distinction between NAFL and NASH is of limited prognostic value. Providing a clinical indictor of more aggressive disease, the development of T2DM in the interval between biopsies was associated with progressive fibrosis: 80% of NAFL patients who exhibited fibrosis progression had developed T2DM by the time of the follow-up liver biopsy compared with 25% of nonprogressors.

Several large longitudinal cohort studies have monitored patients representing the full spectrum of NAFLD severity (including histologically confirmed NAFL, NASH, and all stages of fibrosis) at enrolment to determine disease outcomes, including liver transplant and death. In comparison with a reference population sample, a cohort of 229 histologically characterized European NAFLD patients had a 29% increase in all-cause mortality over a mean 26-years follow-up equating to 5,400 person-years. When specific end points were considered, the presence of NAFLD was associated with a modest 1.29-fold increased risk of cardiovascular disease, a 3.2-fold increased risk of cirrhosis, and a 6.5-fold increased risk of HCC. However, on multivariate analysis the only histologic feature significantly associated with overall long-term mortality, cardiovascular disease, or cirrhosis was the presence of fibrosis. In the second study, based on a cohort of 619 patients (7799 person-years), an increased long-term risk of transplant or liver-related death was evident even at the earliest stages of fibrosis and increased stepwise, with greater degrees of fibrosis portending a worse prognosis. Compared with individuals without any fibrosis (Stage F0), patients exhibiting mild fibrosis (Stages F1 and F2) at enrolment had an 11.2-fold increased risk of transplant or death, and the presence of advanced fibrosis/cirrhosis (Stages F3 and F4) conferred an 85.8-fold increased risk over a median follow-up of 12.6 years. These results again show that the long-term prognosis of patients with NAFLD is best guided by fibrosis stage (i.e., the presence of liver fibrosis) rather than whether the index biopsy showed NAFL or NASH.

A number of other studies also indicate that the presence and severity of fibrosis on a liver biopsy is of greater value as a histologic determinant of long-term prognosis than the prognostic difference between NAFL and NASH, which is largely due to the greater likelihood of fibrosis being present in patients with NASH compared with patients with NAFL rather than any additional adverse effect due to steatohepatitis per se. In a study of 118 patients with biopsy-confirmed NAFLD followed up for a median of 21 years there was no difference in overall or liver-related mortality between those with NAFL and those with NASH (classified with the NASH Clinical Research Network [CRN] scoring system). In contrast, patients who died were more likely that survivors to exhibit any stage of fibrosis, and nonsurvivors in particular more frequently exhibited a fibrosis stage of F2 or greater. In a study of 209 NAFLD patients with a median follow-up of 12 years, the presence of NASH correlated only with liver-related mortality when fibrosis was included in its definition. When the individual histologic features were analyzed, only Stage F3/F4 was independently associated with liver-related mortality, with a 5.68-fold hazard ratio. Studies showing that noninvasive scoring systems for the degree of fibrosis in NAFLD, including the NAFLD fibrosis score, are capable of predicting liver-related events, transplant, and death provide further evidence of the prognostic significance of fibrosis.

Taking these studies as a whole (see Fig. 26-2 ), we see there appears to be a highly dynamic bidirectional transit between the states of NAFL and NASH within the liver. At the time of the snapshot captured by liver biopsy, some of those patients who have NAFL will go on to develop NASH and, if they are susceptible , advanced fibrosis and adverse liver outcomes; others who have NASH may regress to NAFL. Therefore the presence or absence of NASH on baseline histologic examination provides little overall prognostic information, although patients with mild/moderate steatosis in the absence of any inflammation may be at the lowest risk of progression. In addition, fibrosis progression in most NAFLD patients is generally slow, with fibrosis taking approximately 8 years to progress from Stage F0 to Stage F1. There is, however, a subgroup of NAFLD patients who are rapid progressors whose fibrosis can progress to Stages F3 or F4 within between approximately 2 to 6 years. Given the evidence that similar rates of fibrosis progression occur in NAFL and NASH, it seems likely that the higher stages of fibrosis seen in patients with NASH simply reflect a longer duration of disease and that NASH usually develops after steatosis. Supporting this assertion, patients with NASH were 9 years older than those with NAFL in the most recent study, and 44% of the NAFL patients had developed NASH after a median follow-up of 8 years.

Nonalcoholic Fatty Liver Disease and the Risk of Hepatocellular Carcinoma

Although most HCC cases worldwide are related to chronic viral hepatitis, more than 50% of cases in developed countries occur in nonvirally infected patients. The prevalence of HCC is estimated to be approximately 0.5% in NAFLD and approximately 2.8% in NASH. A North American population study concluded that NAFLD was the most frequent cause, present in 58.5% of 4406 HCC patients surveyed, followed by diabetes, which was present in 35.8% of the patients. This association persisted even in the subset of patients who possessed only a single risk factor for HCC, suggesting that the association between NAFLD and HCC was not simply through potentiation of another liver disease. Few publications provide robust information on the progression of NAFLD or NASH to HCC; however, a meta-analysis of the available data indicates that the annual incidence of HCC in patients with NAFLD is 0.44 per 1,000 person-years, rising to 5.29 per 1000 person-years in NASH. Although these are lower incidence rates than observed in hepatitis B or hepatitis C, the prevalence of NAFLD in the population means that the number of patients with NAFLD- or NASH-related HCC will continue to rise.

Cirrhosis is present in approximately 80% of patients with HCC ; however, several small series and numerous case reports have described HCC in noncirrhotic NASH. The extent to which this occurs is not well defined, and further prospective data will be needed to determine the incidence of HCC in the absence of advanced fibrosis in NAFLD.