Non-Alcoholic Fatty Liver Disease

A Practical Guide

Geoffrey C. Farrell, Arthur J. McCullough, Christopher P. Day, Geoffrey C. Farrell, Arthur J. McCullough, Christopher P. Day

• Publisher: Wiley-Blackwell
• Year: 2013
• ISBN:
  9781118556221
• Edition: 1
• Topic: Medicine
• Subtopic: Gastroenterology & Hepatology

The sharp rise in cases of Non-alcoholic fatty liver disease is fast becoming one of the major concerns for hepatologists worldwide.This comprehensive clinical guide explains how to diagnose NAFLD and manage patients according to the best standards of care. Contributors from the world'sleading institutions concentrate on patient care, drawing on their extensive experience.

CHAPTER 1

What is non-alcoholic fatty liver disease (NAFLD), and why is it important?

Geoffrey C. Farrell¹, Arthur J. McCullough² and Christopher P. Day³

¹The Canberra Hospital, Canberra, ACT, Australia

²Cleveland Clinic Lerner College of Medicine at Case Western Reserve University, Cleveland, OH, UK

³Newcastle University & Freeman Hospital Liver Unit, Newcastle upon Tyne, UK

---

Key Points

• Non-alcoholic fatty liver disease (NAFLD) is a highly prevalent form of fatty liver disease caused by over-nutrition; most patients show central obesity.
• NAFLD should be suspected in any overweight person with ultrasound evidence of fatty liver, particularly if metabolic complications such as fasting hyperglycemia, raised serum lipids, and high blood pressure are present.
• Diagnosis of NAFLD requires exclusion of alcoholic liver disease by a lifetime, quantitative history of alcohol intake: the limits of alcohol intake allowable for a diagnosis of NAFLD are 70 g/week (or one standard drink/day) in women and 140 g/week (two standard drinks/day) in men. Lower levels of alcohol intake may actually protect against liver complications of NAFLD.
• NAFLD comprises a pathological spectrum from simple steatosis, which rarely leads to liver fibrosis, through steatohepatitis (or NASH), which can lead to liver fibrosis, cirrhosis, and hepatocellular carcinoma (HCC).
• NAFLD is associated with a 1.7-fold increase in standardized mortality. Premature deaths are from common cancers and cardiovascular disease, with liver complications being third most common.
• While only liver biopsy reliably indicates NASH versus “not NASH” pathology in NAFLD, there have been recent advances in non-invasive approaches (clinicopathological scores, biomarkers, and transient elastography) to both disease activity and fibrotic severity.
• Lifestyle measures are the first approach to management of patients with NAFLD; weight loss of >7% appears to improve histology but is achieved in less than 50% of patients.
• Tight control of serum lipid abnormalities is vital for reducing cardiovascular risk in patients with NAFLD.

---

What is NAFLD?

Fatty liver is stainable fat in hepatocytes (steatosis). Among many causes, obesity and type 2 diabetes (T2D) have never been controversial. Despite this, there is no mention of non-alcoholic fatty liver disease (NAFLD) in the current iteration of the International Classification of Disease (ICD-10), developed in 1990. After early Japanese reports [1–3], American authors raised the possibility that obesity and T2D could also be associated with fatty liver disease complicated by liver cell injury and inflammation (“steatohepatitis”), as well as fibrosis or cirrhosis [4–6]. The pathological findings included Mallory hyaline (also termed Mallory–Denk bodies) [5–8], which until the mid-1970s had been regarded as a hallmark of alcoholic hepatitis. In light of this older concept, and to combat the skeptical view that these were likely instances of alcoholic liver disease in persons who had failed to disclose their alcohol dependence, Ludwig in 1980 coined the term “non-alcoholic steatohepatitis (NASH)” [5].

What is non-alcoholic?

While useful in its time, the term NASH has several disadvantages. First, it starts with a negative: “not alcohol.” This immediately raises the issue about what level of alcohol intake allows one to conceptualize liver disease as alcohol related or not, as discussed elsewhere [9]. A pragmatic definition of NAFLD stipulating no more than one standard drink per day (i.e. 70 g ethanol/week) for women and no more than two standard drinks per day (140 g ethanol/week) for men was proposed in the first edition of this book [10] and has been used by the National Institutes of Health (NIH) NASH clinical research network (CRN) [11]; this definition has been widely adopted for clinical studies, except in France where a slightly more liberal cut-off is favored [12, 13].

The proposed levels of alcohol intake are based on evidence about daily alcohol intake and risk of cirrhosis [9, 14, 15], and the “cut-off” values are set lower than the apparent “threshold levels” so as to avoid the issue of overlap between alcoholic liver disease and obesity, T2D, and metabolic syndrome in progression to cirrhosis. In clinical practice, however, such overlap often exists. Managing safe levels of both alcohol intake and overweight, obesity, or T2D is likely to be critical to obtain optimal outcomes in these cases. Further, patients who may be drinking at safe levels at the time of presentation with liver disease may have a past history of chronic excessive alcohol intake for a prolonged period of time, and may therefore have cirrhosis. Lifetime alcohol intake is therefore important [16] and needs to be incorporated into history taking. However, recent evidence is mounting that levels of alcohol intake between zero (abstinence) and one standard drink per day may be beneficial for both cardiovascular health and the liver, potentially ameliorating or preventing the progression of more banal forms of NAFLD to NASH and fibrosis. These apparently conflicting issues are canvassed more fully by one of us (AMcC) in Chapter 21.

Steatosis and NASH

A second issue is that NASH is a pathological diagnosis (see the “Pathological Definition of NASH” section), not one that can be made clinically or by hepatic imaging (which can show evidence of steatosis; see Chapter 9) or laboratory tests (such as raised serum alanine aminotransferase [ALT]) [17, 18]. Hence, if a person has fatty liver related to over-nutrition, it is not possible to “label” them as having NASH or simple steatosis (“not NASH”) without recourse to a liver biopsy. Recent evidence indicates that between 10 and 25% of NAFLD patients have NASH at any one time [9, 19]. In this book, we will use NAFLD when referring to the full spectrum of non-alcoholic fatty liver disease or if the pathology is unknown, and NASH only when referring to steatohepatitis (which requires pathological definition). Another term that has been used is non-alcoholic fatty liver (NAFL) for the “not NASH” cases of NAFLD, but NAFLD has gained widespread acceptance and will be the preferred nomenclature for ICD-11 (scheduled for release, if approved by the World Health Organization, in 2015).

Pathological definition of NASH

In Chapter 3, Brunt and Kleiner discuss the patholo­gical assessment of fatty liver disease. By an increas­ing consensus, the diagnosis of NASH requires recognition by an experienced liver pathologist [20–27]; the elements are steatosis complicated by liver cell injury (evident as ballooned hepatocytes, or Mallory hyaline) with substantial lobular (and occasionally portal) [23, 26] inflammation. Inflammation is of mixed cellularity: polymorphonuclear leukocytes, lymphocytes, and macrophages. There is also often a characteristic pattern of pericellular fibrosis with centrilobular accentuation, and an alternative pattern of predominantly portal fibrosis is also recognized (particularly but not exclusively in children) [20, 23]. In the presence of fibrosis, the diagnosis is clearer (referred to as fibrotic NASH), and the probability of progression of liver disease to cirrhosis is higher [20–22, 27–30].

Words like “recognition by an experienced liver pathologist” and “substantial” reflect the relative lack of absolute (reproducible) criteria to define NASH pathology. In addition, there is relatively poor interobserver correlation for recognizing ballooning [31], underscoring the problem of pathological definition, even among experts. Special stains can partly overcome this challenge, such as ubiquitin stain, which enhances recognition of Mallory hyaline, and cytokeratin (CK)8/18 immunohistochemistry, which identifies hepatocytes in which this intermediate filament protein has been destroyed [32–34]. These aspects are considered in Chapter 3. In addition, scoring for fibrosis severity is subject to a sampling error [27]. Finally, the relatively high rate (∼15%) of improved liver histology (from NASH to not NASH) in placebo arms of clinical randomized controlled trials [RCTs]) [35, 36] suggests either temporal lability or between-sample variability of liver biopsies in NAFLD, factors that “take the gloss off” biopsy as the “gold standard” for NASH diagnosis.

These issues notwithstanding, clinical outcome data (see Chapter 4) do support the dichotomous classification into a NASH versus “not NASH” pathology of NAFLD [22, 27, 37]. Even more reproducibly, they emphasize the critical predictive value of fibrosis [22, 27–30]. Older categorizations, such as the Matteoni et al. types 1–4 [28] that were mentioned in the first edition of this book, were very important in advancing our understanding of the particular significance of ballooning and Mallory bodies in NAFLD [32–34], but simpler discriminations (e.g., NASH vs. not NASH, and fibrosis vs. no fibrosis) are now supported by a stronger evidence base for prognostic purposes.

Another semantic issue is what to call “not NASH” NAFLD pathology. It has been referred to in those terms, but others have noted that characterizing a disorder only by two things that it isn’t [38, 39], rather than a clear statement about what it is, increases its vagueness. The problem is con­siderable given that the majority (75% or more) of NAFLD cases that are biopsied fall into this category [9, 19], and arguably a higher proportion among those not biopsied because of perceived lesser severity. When there is unambiguously no inflammation, no liver cell injury, and no fibrosis, one can use the term simple steatosis [28, 31, 37], but many biopsied cases show minor inflammation as well as steatosis (and these are “not NASH” cases), while o...