Pathology of NASH and fibrosis

Introduction 

Non-alcoholic steatohepatitis (NASH) was the term coined by Ludwig in 1980 for a set of histological features similar to those of alcoholic hepatitis, which he noted in liver biopsies of patients without a significant history of alcohol intake or any clinical evidence of alcohol abuse [14]. The disease was more common in women, obese, and those with obesity-associated diseases, such as diabetes mellitus and cholelithiasis.

Non-alcoholic fatty liver disease (NAFLD) represents a spectrum of liver lesions that occur in the absence of significant alcohol consumption. At one end of the spectrum is the possibly non-progressive non-alcoholic fatty liver (NAFL) or isolated hepatic macrovesicular steatosis, while at the other end is a potentially progressive NASH. NAFLD is possibly the most common liver disease in the West and is rapidly gaining footage in the East as well owing to the changes in lifestyle trends. NASH has been defined pathologically as significant steatohepatitis that is not the result of alcohol, drugs or any other single identifiable cause [4]. NAFLD is the manifestation of ‘metabolic syndrome’, which includes central obesity, type-II diabetes, insulin resistance, hypertriglyceridemia, and hypertension. However, NASH has also been described in men, in the absence of obesity, diabetes and hyperlipidemia [1].

Diagnosis 

NASH is diagnosed essentially by clinicopathological correlation, wherein clinically, besides ruling out significant alcohol consumption, other causes such as viral hepatitis, autoimmune, A1AT and cholestatic aetiologies also need to be excluded.

The summary reports of the American Association for the Study of Liver Diseases (AASLD) Single Topic Conference on NASH held in September 2002 suggested criteria for a diagnosis of NASH as steatosis, ballooning degeneration and lobular inflammation [18]. The four most important features are steatosis, ballooning, lobular inflammation and perisinusoidal fibrosis. This has recently been validated in a study published by the NASH – Clinical Research Network [12].

Patterns of liver fibrosis in NASH 

The onset of fibrosis in NASH is often in the perisinusoidal, pericellular location. The characteristic pattern of fibrosis in NASH is due to deposition of collagenous matrix along the hepatocytes in the space of Disse, giving rise to a ‘chickenwire’ appearance. This is usually more prominent, or begins in zone III. This pattern of perisinusoidal fibrosis is also seen in alcoholic hepatitis and may be seen in cases of chronic cholestasis [5].

As the fibrosis progresses septum formation, bridging fibrosis and cirrhosis (micronodular, macronodular or mixed) can occur. As cirrhosis sets in the perisinusoidal fibrosis may no longer be detectable. Furthermore, steatosis often decreases and may even be completely absent. The zone 3 pattern of injury, cytoplasmic ballooning and Mallory hyaline may also no longer be detectable, thus leading to a diagnosis of cryptogenic cirrhosis. The characteristic zonal distribution of steatosis in NASH may not be seen due to the architectural distortion of cirrhosis [26].

For the assessment of fibrosis the quality of Trichrome staining is important as an overstained section will mask the perisinusoidal fibrosis and an understained one will not detect it.

Prevalence of mild to moderate fibrosis in NASH varies from 76% to 100%, severe fibrosis may be seen in 15–50% and cirrhosis in 7–16% [22]. Activated hepatic stellate cells in the space of Disse may have a role to play and a high degree of stellate cell activation may indicate a risk for disease progression [6].

Grading and staging 

Brunt et al. [3] studied 10 histologic variables in cases of NASH. An overall impression of mild, moderate and severe was made and then the histological variables considered the most significant were used to propose a grading and staging system. These were steatosis, ballooning, lobular and portal inflammation.

The pathology subcommittee of the Clinical Research Network for NASH has recently designed and validated a histologic feature scoring system for the full spectrum of lesions of NAFLD [12]. They found inter-rater agreement of 0.84 for fibrosis using weighted kappa statistics. This group proposed a NAFLD activity score (NAS). Only routine H&E and Masson Trichrome stains are recommended for evaluation. It is defined as the unweighted sum of scores for steatosis (0–3), lobular inflammation (0–3) and ballooning (0–2). The sum is thus 0–8. The NAS has been defined for the purpose of evaluating histologic changes after therapeutic trials and to assess overall histologic change and not to replace diagnostic criteria or assess severity of disease in relation to disease progression. Nonetheless the authors have suggested that a NAS score of ⩾5 correlated with a diagnosis of NASH and <3 with a diagnosis of ‘not NASH’.

In the original grading and staging system proposed by Brunt et al. fibrosis was staged from 0 to 4 (Table 1), where F 1 is focal or extensive zone 3 perisinusoidal or pericellular fibrosis. The NAS score, has further elaborated the staging system in leiu of the portal-based pattern of inflammation and fibrosis in children and rarely in adults. Thus presence of perisinusoidal or periportal fibrosis is stage 1, which is further subcategorised (Table 2).

Table 1. Staging of NASH.

	Stage 1	Zone 3 perisinusoidal/pericellular fibrosis; focal or extensive
	Stage 2	Zone 3 perisinusoidal/pericellular fibrosis with focal or extensive periportal fibrosis
	Stage 3	Zone 3 perisinusoidal/pericellular fibrosis and portal fibrosis with focal or extensive bridging fibrosis
	Stage 4	Cirrhosis with or without residual perisinusoidal fibrosis

Modified from Brunt et al. [3].

Table 2. Staging of fibrosis in the NAFLD activity score as defined by NASH clinical research network [12].

	Item	Definition	Score
	Fibrosis stage
		None	0
		Perisinusoidal or periportal	1
		Mild, zone 3, perisinusoidal	1A
		Moderate, zone 3, perisinusoidal	1B
		Portal/periportal	1C
		Perisinusoidal and portal/periportal	2
		Bridging fibrosis	3
		Cirrhosis	4

Pathogenesis 

The understanding of the pathogenesis of NASH has evolved in the last two decades from a two-step to a four step model [24]. The earlier hypotheses explained the mechanism by which fat accumulates in hepatocytes and causes the latter to undergo ballooning and necrosis. Triglycerides can accumulate in the liver because of decreased synthesis of lipoproteins and decreased export of lipids from the liver. Any mechanism leading to hepatic steatosis, is presumed to be the ‘first hit’ in the pathogenesis of NASH. This may be related to obesity, insulin resistance or hypertriglyceridemia. Increased concentrations of intracellular fatty acids may be directly toxic to hepatocytes or lead to oxidative stress. Lipid peroxidation and oxidative stress with formation of reactive oxygen species (ROS) form the ‘second hit’ in the pathogenesis of NASH. In a recent ‘four-step’ model proposed by Wanless et al., the 3rd step is the release of lipid to the extracellular space causing inflammation leading to venular obstruction which is the fourth step culminating in fibrous septation and cirrhosis.

Progression of fibrosis 

There are extensive data to suggest that progression of fibrosis occurs in 40% of cases with NASH over a period of 4years and that cirrhosis develops in about 10% [5].

Role of liver biopsy 

The role of liver biopsy in NASH has been the subject of debate in the recent past. However, it remains the gold standard in the diagnosis of NASH. The diagnosis of steatosis can be made by radiological methods [21]; however, the estimation of the extent of liver injury and the associated fibrosis can only be made by histological evaluation of liver biopsy. Although certain serum markers [17] and radiological techniques like fibroscan are promising, they need validation but may still be helpful to use in conjunction. Furthermore, a liver biopsy can exclude other causes of liver disease and has a role in determining the efficacy of therapeutic trials.

Limitations of liver biopsy 

The limitations of liver biopsy in NASH are the same as for chronic hepatitis, the main factor being the sample size. Even an adequate biopsy of >1.5cm represents a very minute fraction of the whole liver. Goldstein et al. [8] have demonstrated the heterogeneity in fibrosis observed in liver biopsy in NASH and recommend that a 1.6cm length of liver biopsy is required for accurate fibrosis staging.