Grading and staging systems for inflammation and fibrosis in chronic liver diseases☆
Article Outline
- Abstract
- 1. Introduction
- 2. Grading and scoring in medicine and in histopathology
- 3. Grading and staging of liver diseases other than chronic hepatitis
- References
- Copyright
Liver biopsy is an important part of the evaluation of patients with a variety of liver diseases. Besides establishing the diagnosis, the biopsy is often used to assess the severity of the disease in terms of both grade and stage. The stage in most chronic liver diseases relates to the degree of scarring with the end stage being cirrhosis with its clinical complications. The grade relates to the severity of the underlying disease process, with features that vary with the pathogenetic mechanisms. Chronic viral hepatitis has been the object of the most extensive efforts at grading and staging, stimulated by the advent of new forms of therapy. Systems have also been developed for fatty liver disease, allograft rejection and chronic cholestatic diseases, but these have not been as widely used. Simple grading and staging systems for chronic hepatitis, including the IASL, Batts–Ludwig, and Metavir systems, are most appropriate for management of individual patients, while more complex systems such as the Histology Activity Index (HAI) are appropriate for evaluation of large cohorts of patients when statistical analysis is required.
Abbreviations: IASL, International Association for the Study of the Liver, HAI, Histology Activity Index, NAFLD, nonalcoholic fatty liver disease, NAS, NAFLD activity score
Keywords: Liver biopsy, Grading, Staging, Chronic hepatitis, Steatohepatitis
1. Introduction
The liver biopsy serves two principal functions. First, it establishes or confirms the diagnosis of a particular type of liver disease. Then, it is frequently used to assess the severity of the disease; i.e. the grade and stage. In the hands of a skilled interpreter, a biopsy is quite good for the first function, which is a qualitative analysis. It is less reliable for the second, which can be considered a semiquantitative analysis, but it can still provide potentially useful information that can be used to assess prognosis and guide treatment.
The stage of any disease is a measure of how far it has progressed in its natural history, with the end stage resulting in death of the patient or failure of the organ. The grade of the disease is meant to reflect how quickly the disease is progressing to the end stage. In most forms of chronic liver disease, the end stage is cirrhosis with clinical decompensation, whereas earlier stages have lesser degrees of fibrosis or cirrhosis. The grade can be considered to relate to the severity of the underlying liver disease, with features that vary with the type and pattern of injury. Ideally, both grade and stage should predict prognosis and guide therapeutic intervention, although evidence for this is relatively meager in most chronic liver diseases, and therapeutic intervention may alter the natural history in ways that are not clearly defined. Nevertheless, pathologists are often asked to assign a grade and stage as part of the liver biopsy evaluation, with the intention that this will predict patient outcome.
2. Grading and scoring in medicine and in histopathology
There are many areas of medicine in which clinical, radiologic or laboratory findings can be ordered as to the degree of severity based on observation, even when they cannot be precisely quantified. This allows them to be graded or scored with the presumption that the higher grade or score implies a worse disease process or outcome. In clinical medicine grading is frequently used to indicate the severity of a diagnosis (e.g. esophageal varices), a physical finding (e.g. heart murmurs), or a degree of functional impairment (e.g. hepatic encephalopathy). In histopathology, grading is frequently used for prognosis and therapeutic decisions in tumors and to a lesser extent in inflammatory diseases in many tissues and organ systems.
Scores derived from these types of observations, no matter how precisely defined, invariably contain a subjective element related to the observer, and thus there is always some degree of observer variability. The kappa statistic, which can range from 0 (chance agreement) to 1.0 (perfect agreement) which is often used to assess inter- and intra-observer variability. The usual interpretation is 0.81–1.0 is almost perfect agreement, 0.61–0.8 substantial agreement, 0.41–0.6 moderate agreement, 0.21–0.4 fair agreement, and 0–0.2 slight agreement [1]. Agreement is always greater when there are fewer categories, so weighted kappa scores, accounting for how close or far the disagreements, are often used when there is a wide range of possible scores. Most publications that have used kappa statistics to evaluate scoring systems for inflammation and fibrosis in liver diseases have reported fair to excellent levels of agreement, usually better for fibrosis (kappa 0.5–0.9) than for inflammation (kappa 0.2–0.6) [2], [3], [4], [5], [6], [7], [8], [9]. This compares favorably to observer agreement in many other areas of medicine, such as cardiac auscultation (kappa 0.19) [10], endoscopic diagnosis of esophageal varices (kappa
=0.38) [11], interpretation of mammograms by radiologists (kappa=0.47) [12], and pathologic grading of breast cancers (kappa 0.43–0.74) [13].
Construction of the histologic scoring systems for use in liver disease has generally preceded demonstration of their clinical utility. Many grading and staging systems in liver disease, like those in other areas of medicine, have used three to four categories of increasing severity for the feature or disorder of interest. Thus, a simple grading system, such as the Metavir activity score for chronic hepatitis [14], can categorize a lesion or disease process as mild (A1), moderate (A2) or marked (A3). Within these categories, however, there is a range of severity, so that a system with more categories has the potential to provide more information. Many pathologists do this as a matter of course, inserting the categories of “mild-to-moderate” and “moderate-to-marked” into three-grade systems. It has been noted, however, that within a continuum, such as occurs in many natural processes, an observer tends to distinguish seven categories; for example the rainbow is considered to have seven colors, which is not a property of the electromagnetic radiation that comprises the visible spectrum but rather a reflection of the way the human mind organizes the visual information [15]. Reproducibility of a seven category system is lower than for a system with fewer categories, but the amount of information conveyed is greater. Thus, the Ishak fibrosis score [16], with 7 stages (0–6), provides more information than other fibrosis scores with only 5 stages (0–4) [14], [17], [18]. Furthermore, features that are graded separately can be added to give composite scores with an even greater range, as is typically done with the more complex scoring systems for chronic hepatitis [16], [19]. Adding categorical scores assigned to different histological features has been criticized as an inappropriate use of this type of data, but this approach has been used to generate clinically useful systems in other areas of medicine, for example the Child–Turcotte–Pugh score which adds grades for bilirubin, albumin, prothrombin time, hepatic encephalopathy and ascites to yield a score that relates to prognosis [20], so it does not seem unreasonable to add scores for histologic features to generate composite histologic scores for liver diseases.
2.1. Features for grading and staging
There are many histopathologic lesions that can be considered in formulating a grading and staging system for a liver disease (Table 1). To be useful, the lesions employed to formulate the score should have biological plausibility as to their role in the severity of the disease. Definition of the grades of severity of each lesion should be as precise as possible, and there should be at least moderate reproducibility (Table 2). Ideally, there should also be a demonstrable correlation with prognosis or clinical outcome.
Table 1. Histopathologic features of liver biopsies that can be considered for grading and staging systems in nonneoplastic liver diseases
| Hepatocellular changes |
| Ballooning degeneration |
| Acidophilic degeneration/apoptosis |
| Dropout/confluent necrosis |
| Mallory bodies |
| Steatosis |
| Iron storage |
| Bile stasis |
| Cholate stasis |
| Inflammation |
| Portal |
| Periportal (interface hepatitis) |
| Lobular |
| Intravascular (endotheliitis) |
| Kupffer cell hypertrophy |
| Granulomas |
| Biliary changes |
| Bile duct injury |
| Ductular reaction |
| Ductopenia |
| Fibrosis and architectural changes |
| Quantity of fibrous tissue |
| Location (portal, perisinusoidal, septa and nodule formation) |
Table 2. Histopathologic features of liver biopsies that can be graded and degree of reproducibly in chronic viral hepatitis [2], [3], [4], [5], [6], nonalcoholic fatty liver disease [7], [8], and liver allograft rejection [9]
| Interobserver agreement | Chronic hepatitis | Fatty liver disease | Rejection |
|---|---|---|---|
| Substantial to almost perfect (kappa⩾0.6) | Cirrhosis | Steatosis | Bile duct loss |
| Fibrosis | Fibrosis | Cholestasis | |
| Steatosis | Ballooning | Portal inflammation | |
| Mallory bodies | |||
| Moderate (kappa 0.4–0.6) | Portal inflammation | Portal inflammation | Arteriopathy |
| Interface hepatitis | Parenchymal inflammation | Parenchymal inflammation | |
| Parenchymal injury and inflammation | Centrilobular necrosis | ||
| Slight to fair (kappa<0.4) | Ballooning | Acidophilic bodies | Bile duct inflammation |
| Ductular proliferation | Microvesicular steatosis | Endotheliitis | |
| Bile duct lesion |
2.2. Grading and staging in chronic hepatitis
Grading and staging have been used most often in chronic viral hepatitis. In the chronic necroinflammatory diseases, including viral hepatitis (B, C and D) and autoimmune hepatitis, the grade of the disease is considered to be the degree of inflammation and hepatocellular injury, which is thought to lead to fibrosis. Several methods are currently used to express the grade and stage of chronic hepatitis. These can be grouped into those that are simple verbal descriptions; those that are relatively simple numerical grades and stages that correspond to the verbal descriptions; and those that use more complicated numerical scoring of histologic features to generate numbers that correspond to the grade and stage. Each has its advantages and disadvantages and the system used should be appropriate to its suitability for the task at hand. In general, the more complex systems have the capability to provide more information than simple ones but are less reproducible.
2.3. Simple grading and staging systems
For routine diagnosis and patient management a simple system of grading and staging is preferred. Perhaps the best is that proposed by a panel of experts convened by the International Association for the Study of the Liver (IASL) in 1994 [17]. They recommended grading of chronic hepatitis by the degree of necroinflammatory activity as mild, moderate or severe (marked), based on correlation with the Knodell Histology Activity Index [19]. Batts and Ludwig [18] adapted this into a four category scale with simple definitions based on the degrees of interface hepatitis and parenchymal injury and inflammation, and the Metavir group [14] developed a similar algorithm to arrive at a three category (mild, moderate, severe) activity score. The principal features of activity in these systems are the degree of periportal interface hepatitis (“piecemeal necrosis”) and parenchymal injury (apoptosis and spotty necrosis), each of which can be graded. Brief definitions of what constitutes mild, moderate and severe interface hepatitis and parenchymal injury are given in the publications, some more easily applied than others. In practice, these are subjective assessments based on the experience of the pathologist. For example, interface hepatitis is usually recognized as extension of the portal lymphoplasmacytic inflammation into the surrounding parenchyma to replace hepatocytes lost by apoptosis (Fig. 1). It can be considered mild when one must search to find any foci in the biopsy; moderate (as defined for the Knodell score) when the majority of portal areas have some, but it extends around less than 50% of the circumference of most portal areas; and marked when the majority of portal areas have interface hepatitis extending around more than 50% of the circumference. Since the lesions can vary considerably from one portal tract to the next, the pathologist must do a mental average of all the portal tracts in the biopsy, which readily explains why reproducibility is only moderate. Parenchymal injury is most easily graded using the 10× (medium power) objective of the microscope with the usual 10× ocular lens. At this magnification one can detect acidophilic bodies, ballooned hepatocytes and clusters of inflammatory cells at sites of focal necrosis, and it is relatively easy to estimate the amount of injury and form an overall impression of the degree of injury (Fig. 2). Parenchymal injury can be considered mild when on average one sees only a few injured cells or clusters of inflammatory cells; moderate when there are more than five foci of inflammation or apoptotic bodies per 10× field; and marked when there are numerous necroinflammatory foci. In the most severe cases there may be confluent necrosis and stromal collapse, usually involving acinar zone 3, producing centrilobular or bridging necrosis depending on the plane of section, or even multiacinar necrosis. This degree of severity is rare in chronic viral hepatitis but not uncommon in autoimmune hepatitis. In grading the overall activity, one can follow the guidelines of Batts and Ludwig (Fig. 3) and consider the chronic hepatitis minimal when both interface hepatitis and parenchymal injury are minimal or absent; mild if either or both interface hepatitis and parenchymal injury are mild; moderate if both interface hepatitis and parenchymal necrosis are moderate or if one is moderate but the other is mild; and marked if either or both of interface hepatitis and parenchymal injury are marked. Bedossa et al. [14], on behalf of the Metavir group, proposed a similar algorithm to grade activity into three categories (Fig. 4).
Fig. 1.
Schematic diagram of examples of mild, moderate and marked interface hepatitis. The portal tract on the left has only a focal area of interface hepatitis, indicated by the blue line. The portal tract in the center has a moderate amount of interface hepatitis with several foci around the circumference. The portal area on the right has marked interface hepatitis, nearly surrounding the entire circumference.
Fig. 2.
Schematic diagram of examples of mild, moderate and marked parenchymal injury. The field on the left has only a few foci of inflammatory cells and rare acidophilic bodies. The field in the center has more numerous foci of inflammation and hepatocellular injury. That on the right has many inflammatory foci and injured hepatocytes.
Fig. 3.
Algorithm for grading hepatitis activity according to Batts and Ludwig [18].
Fig. 4.
Algorithm for grading hepatitis activity according to Metavir activity score of Bedossa et al. [14].
Staging of chronic hepatitis requires assessment of the degree of fibrosis, which requires a Masson trichrome stain for proper evaluation. There is progression in stage of disease as the fibrosis advances from none to fibrous portal expansion to bridging fibrosis to incomplete cirrhosis and finally to established cirrhosis. The advantage of verbal descriptive systems is that intermediate stages can be assigned, such as early bridging fibrosis, marked bridging fibrosis, early cirrhosis and advanced cirrhosis, thus conveying additional information. Numbers can be used to correspond to these stages (Table 3) if desired.
Table 3. Simple grading and staging systems for chronic viral or autoimmune hepatitis
| IASL | Batts–Ludwig | Metavir |
|---|---|---|
| Grade | ||
| Chronic hepatitis with minimal activity | Grade 1 | A1 |
| Chronic hepatitis with mild activity | Grade 2 | A1 |
| Chronic hepatitis with moderate activity | Grade 3 | A2 |
| Chronic hepatitis with marked activity | Grade 4 | A3 |
| Chronic hepatitis with marked activity and bridging or multiacinar necrosis | Grade 4 | A3 |
| Definition | IASL | Batts–Ludwig | Metavir |
| Stage | |||
| No fibrosis | No fibrosis | Stage 0 | F0 |
| Fibrous portal expansion | Mild fibrosis | Stage 1 | F1 |
| Few bridges or septa | Moderate fibrosis | Stage 2 | F2 |
| Numerous bridges or septa | Severe fibrosis | Stage 3 | F3 |
| Cirrhosis | Cirrhosis | Stage 4 | F4 |
2.4. Complex systems
Complex numerical systems have also been proposed for grading and staging of chronic hepatitis, providing a wider range of values with the intent that these could be used in statistical analysis of liver biopsies obtained in the course of clinical trials. Of these complex systems, the most widely used is the Histology Activity Index (HAI), first described by Knodell et al. in 1981 [19], and commonly called the Knodell score. Some technical criticisms of the original score resulted in a revised form that was described by an international group led by Ishak (the pathologist who worked with Knodell to devise the original score) and subsequently called the Ishak score [16]. Features used to calculate the scores are listed in Table 4. For grading, these assign numbers to the severity of the necroinflammatory features (interface hepatitis, confluent necrosis, parenchymal injury and portal inflammation) and add the numbers to arrive at a grade that can range from 0 to 18. The stage, ranging from 0 to 4, may or may not be added into the Knodell score; while in the Ishak score, the stage, ranging from 0 to 6, is reported separately.
Table 4. Histology Activity Index in its original form, the Knodell score [19] and modified form, the Ishak score [21]
| Knodell Score | Score | Ishak Grade | Score |
|---|---|---|---|
| Periportal±bridging necrosis (piecemeal necrosis) | Periportal or periseptal interface hepatitis (piecemeal necrosis) | ||
| None | 0 | None | 0 |
| Mild piecemeal necrosis | 1 | Mild (focal, few portal areas) | 1 |
| Mild/moderate (focal, most portal areas | 2 | ||
| Moderate piecemeal necrosis (involves less than 50% of circumference of most portal tracts) | 3 | Moderate (continuous around<50% of tracts or septa) | 3 |
| Marked piecemeal necrosis (involves more than 50% of circumference of most portal tracts) | 4 | Severe (continuous around>50% of tracts or septa) | 4 |
| Confluent necrosis | |||
| None | 0 | ||
| Focal confluent necrosis | 1 | ||
| Zone 3 necrosis in some areas | 2 | ||
| Zone 3 necrosis in most areas | 3 | ||
| Moderate piecemeal necrosis plus bridging necrosis | 5 | Zone 3 necrosis+occasional portal–central bridging | 4 |
| Marked piecemeal necrosis plus bridging necrosis | 6 | Zone 3 necrosis+multiple portal–central bridging | 5 |
| Multilobular necrosis | 10 | Panacinar or multiacinar necrosis | 6 |
| Intralobular degeneration and focal necrosis | Focal (spotty) lytic necrosis, apoptosis and focal inflammation | ||
| None | 0 | None | 0 |
| Mild (acidophilic bodies, ballooning degeneration, and/or scattered foci of necrosis in <1/3 of lobules or nodules) | 1 | One focus or less per 10× objective | 1 |
| Two to four foci per 10× objective | 2 | ||
| Moderate (involvement of 1/3–2/3 of lobules or nodules) | 3 | Five to ten foci per 10× objective | 3 |
| Marked (involvement of >2/3 of lobules or nodules) | 4 | More than 10 foci per 10× objective | 4 |
| Portal inflammation | Portal inflammation | ||
| None | 0 | None | 0 |
| Mild (sprinkling of inflammatory cells in <1/3 of portal tracts) | 1 | Mild, some or all portal areas | 1 |
| Moderate, some or all portal areas | 2 | ||
| Moderate (increased inflammatory cells in 1/3–2/3 of portal tracts) | 3 | Moderate/marked, all portal areas | 3 |
| Marked (dense packing of inflammatory cells in >2/3 of portal areas) | 4 | Marked, all portal areas | 4 |
| Fibrosis | Ishak Stage | ||
| No fibrosis | 0 | No fibrosis | 0 |
| Fibrous portal expansion | 1 | Fibrous expansion of some portal areas, with or without short fibrous septa | 1 |
| Fibrous expansion of most portal areas, with or without short fibrous septa | 2 | ||
| Bridging fibrosis | 3 | Fibrous expansion of most portal areas with occasional portal to portal bridging | 3 |
| Fibrous expansion of portal areas with marked bridging (portal to portal as well as portal to central) | 4 | ||
| Marked bridging (portal–portal and/or portal–central) with occasional nodules (incomplete cirrhosis) | 5 | ||
| Cirrhosis | 4 | Cirrhosis, probable or definite | 6 |
Numerical scores generated by these systems are not sufficiently reproducible to be useful in the management of individual patients, but they are very useful for investigational studies that involve large numbers of biopsies requiring statistical analysis. For example, the scores are a very good way to show differences in histologic response between cohorts of patients receiving different forms of therapy, and they have been used successfully in many large clinical trials. This can be most readily demonstrated in placebo controlled trials where some placebo-treated patients show improvement in HAI due to natural variations in disease activity, sampling variability of the biopsy, or variability in interpretation by the pathologist, but the numbers that improve is balanced by the numbers that worsen, so that the net effect is close to zero. By contrast, an effective therapy results in a significant decrease in mean HAI of the cohort; for example, in a trial of adefovir for chronic hepatitis B [21], the mean of the necroinflammatory components of the HAI of the placebo-treated patients was nearly unchanged (−0.16) after 48 weeks of therapy, while the patients receiving the active drug showed a highly significant improvement in mean score (−2.58).
It has been repeatedly noted that the numbers in the scores represent ordered categories rather than values, and that adding and averaging these numbers is not a scientifically valid process. A biopsy with a score of 12 is not necessarily twice as bad as one with a score of 6 or three times as bad as one with a score of 4. While this may be strictly true, the original version of the HAI score [19] was conceived as a weighted score, which is the explanation for missing numbers in the scores for the various components (Table 4). Thus, in the expert opinion of the score’s authors, moderate piecemeal necrosis (score=3) was much worse than mild piecemeal necrosis (score=1), while marked piecemeal necrosis (score=4) was only a little worse than moderate. Although it cannot be independently validated, the overall score was intended to be used as a global assessment of histologic severity that could be used in statistical analysis. The modified HAI [16] has no missing numbers but maintains the same range of scores. Consequently, there is some justification for comparing mean scores and changes in the mean between cohorts of patients as a measure of changes in disease severity.
The best way to use histologic data such as HAI has never been established. In the registration trials for a number of products for treatment of hepatitis B and hepatitis C, histologic improvement was used as the primary or secondary endpoint [21], [22], [23], [24], [25], [26], [27]. The criterion for improvement, defined by the products’ sponsors and the regulatory agencies, was a 2-point or greater decrease in the inflammatory components of the Knodell HAI, later expanded to a 2-point or greater decrease in inflammation with no worsening of the Knodell fibrosis score. The 2-point improvement was apparently chosen as a clinically meaningful change, and the analysis was confined to the proportion of patients in each cohort (placebo, comparator or study drug) achieving this change. Since the studies had adequate statistical power, the drugs in the various studies all produced 2-point improvement in significantly greater proportion of patients than did the placebo or comparator. This type of analysis, however, negates the value of using a complex score with its wealth of data, and fails to present the magnitude of actual change in histology. For example, hepatitis C patients with sustained virologic response, 80–90%, show a 2-point improvement, but they actually improve by a mean of 4–6 points; by contrast, 30–50% of nonresponders show a 2-point histologic improvement, but the mean improvement is less than one point [22], [24], [25], [26]. The differences are highly significant with both methods, but the degree of improvement is vastly understated when the 2-point criterion is used. In actual practice, if a therapy is truly effective in improving histology, then almost any method or scoring system will show statistically significance differences between the therapy and placebo and equivalence or superiority to a comparator. Whether using Knodell, Ishak or Metavir, the scores all improve, as do the means and medians of HAI as well as proportions with 1-point, 2-point, or 3-point improvement, 50% improvement, or proportion of patients with nearly normal biopsies (HAI⩽3) [unpublished data].
2.5. Fibrosis scores
A number of investigators have shown an association between activity grade and progression of fibrosis to cirrhosis [28], [29], [30], [31]. Although staging of liver disease by assessment of fibrous scarring has always been a major function of liver biopsy interpretation, until recently it has been assumed that fibrosis is an irreversible process and that changes over time are relatively slow, so most grading and staging systems have relatively few stages (Table 3, Table 4), and consequently they are very insensitive for describing changes in fibrosis. Of the commonly used systems, the Ishak fibrosis score with six stages is the most sensitive, and its stages can be readily translated into the other scores, as shown diagrammatically in Fig. 5. A complex fibrosis scoring system with a theoretical range of 0–37 has also been described [32] but has not been widely used. In all systems, the stages are determined by both the quantity and location of the fibrosis, with the formation of septa and nodules as major factors in the transition from one stage to the next. Even more than with grading of inflammation, it is important to remember that the stages, although indicated by a number, are ordered categories, not equal units. The absolute amount of fibrous tissue, measured quantitatively by computer-assisted image analysis, is extremely variable within each stage, and there is considerable overlap between stages in the amount of fibrous tissue [33], [34]. A number of publications have calculated fibrosis progression rates in terms of “units per year” in the various scoring systems, but this is subject to the valid criticism that since the stages represented by numbers do not accurately reflect equal units in terms of severity, amount of fibrous tissue or time of development, calculations based on those numbers, often carried to several decimal places, are meaningless [35], [36].
Fig. 5.
Diagrammatic representation of the seven stages (0–6) of the Ishak fibrosis score compared to the five stages (0–4) of the Metavir fibrosis score and three stages (1, 3 and 4) of the Knodell score. Normal liver (stage 0) has very little fibrous tissue in the portal areas (dark circles) and the walls of central veins (open circles). Chronic hepatitis results in fibrous expansion of portal tracts, which may maintain a rounded contour or develop short spike-like septa, first involving only a few portal tracts (stage 1) and eventually of all portal tracts (stage 2). With progression of disease, fibrous septa extend to form bridges between adjacent vascular structures, both portal to portal and portal to central, occasional bridges (stage 3), progressing to numerous bridges or septa (stage 4). Eventually, while some areas maintain lobular architecture, parenchymal nodules completely surrounded by fibrosis may form, indicating an early or incomplete cirrhosis (stage 5), and when the tissue is entirely composed of nodules it can be considered an established cirrhosis (stage 6).
3. Grading and staging of liver diseases other than chronic hepatitis
3.1. Nonalcoholic Fatty Liver Disease and Steatohepatitis
Grading and staging of fatty liver disease and steatohepatitis, whether due to alcohol or obesity/diabetes, is supported by less data than chronic hepatitis. The mechanisms of disease progression and the relationship of histological features to progression are less well understood, and consequently the criteria, especially for grading, are less well established. Brunt et al. [37] proposed a three grade system (mild, moderate marked) based on a combination of features including fat, hepatocyte ballooning and inflammation with four stages. Kleiner et al. [8] studied the reproducibility of histologic features of nonalcoholic fatty liver disease and (NAFLD) and devised a NAFLD Activity Score (NAS) to be used in natural history studies and treatment trials (Table 5) in the ways that the HAI has been used in studies of chronic hepatitis.
Table 5. NAFLD activity score and staging system devised by the Pathology Committee of the NASH Clinical Research Network [8]
| Item | Definition | Score |
|---|---|---|
| Steatosis | Low- to medium-power evaluation of parenchymal involvement | |
| <5% | 0 | |
| 5–33% | 1 | |
| >33–66% | 2 | |
| >66% | 3 | |
| Lobular inflammation | Overall assessment of all inflammatory foci | |
| No foci | 0 | |
| <2 foci per 200× field | 1 | |
| 2–4 foci per 200× field | 2 | |
| >4 foci per 200× field | 3 | |
| Hepatocyte ballooning | None | 0 |
| Few ballooned cells | 1 | |
| Many cells/prominent ballooning | 2 | |
| NAFLD Activity Score | Sum of steatosis, lobular inflammation and ballooning scores | 0–8 |
| Fibrosis Stage | None | 0 |
| Perisinusoidal or periportal | 1 | |
| Mild, zone 3, perisinusoidal | 1A | |
| Moderate, zone 3, perisinusoidal | 1B | |
| Portal/periportal only | 1C | |
| Both perisinusoidal and portal/periportal | 2 | |
| Bridging fibrosis | 3 | |
| Cirrhosis | 4 |
3.2. Chronic cholestatic diseases
The chronic cholestatic diseases, primary biliary cirrhosis and primary sclerosing cholangitis, have some of the best data correlating stage of disease with prognosis. However, even though there is abundant literature on histopathologic features of these diseases, no grading system that incorporates the severity of the various lesions (cholatestasis, ductopenia, inflammation) has been proposed.
Fibrosis follows loss of bile ducts in both sclerosing cholangitis and primary biliary cirrhosis. Both ductular proliferation and interface hepatitis, accompanied by collagen deposition, appear to be important, but the possible roles of other factors related to chronic cholestasis have not been studied in detail. The fibrosis extends progressively with portal–portal bridging and septum formation, eventually with nodule formation and development of a micronodular biliary cirrhosis indistinguishable from that caused by chronic mechanical obstruction. Ludwig suggested four stages, similar to those in his system for chronic hepatitis, to describe the fibrosis stages for both primary biliary cirrhosis [38] and sclerosing cholangitis [39] – stage 1 (portal), stage 2 (periportal), stage 3 (septal) and stage 4 (cirrhosis), with progressively worse prognosis at later stages.
3.3. Transplant rejection
An international panel devised the widely used Banff schema for grading and staging of liver allograft rejection, including a Rejection Activity Index (0–9) based the degrees of portal/periportal inflammation, bile duct inflammation and damage, and venous endothelial inflammation, which are the major histologic features associated with rejection [40]. Although useful in describing histologic severity, correlation with response to therapy and outcome has been limited [41].
3.4. Future directions
As long as liver biopsies continue to be useful in diagnosing and assessing the severity of liver diseases, there will be a need for methods to express the histologic information in ways that permit that information to be applied most effectively. In studies of natural history of diseases, it will be useful to use semiquantitative scoring systems as described in this review to test hypotheses about the roles of various factors in clinicopathologic outcomes and to compare these to other sources of information, such as biochemical, genomic and proteomic markers. As the natural history studies identify features relevant to clinical outcomes, the scoring systems can be modified to refine the scores to match the natural history more closely, and ancillary information, such as image analysis and molecular studies, can be incorporated. The same approach can be used in clinical trials of therapeutic interventions. If a therapy is curative, such as one that produces a sustained virologic response in chronic hepatitis C, then sequential biopsies are not needed to assess therapeutic response. However, if the therapy is intended to ameliorate but not to cure the liver disease, then it will be reassuring to know that histologic improvement has occurred, and scoring to assess the degree of improvement in that particular disease will be appropriate. As the natural history of different forms of liver disease and the results of therapy become better understood, appropriate simple grading and staging systems for use in the management of individual patients can be validated against clinical outcomes and therapeutic responses and become incorporated into routine practice.
References
- . The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–174
- . Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. Hepatology. 1994;20:15–20
- Intra-observer and inter-observer variation in the histopathological assessment of chronic viral hepatitis. J Hepatol. 1996;25:649–654
- . Interobserver study of liver histopathology using the Ishak score in patients with chronic hepatitis C virus infection. Liver. 1999;19:183–187
- Interobserver variation in interpretation of serial liver biopsies from patients with chronic hepatitis C. J Viral Hepat. 2002;9:443–449
- Sources of variability in histological scoring of chronic viral hepatitis. Hepatology. 2005;41:257–264
- Nonalcoholic fatty liver disease: assessment of variability in pathologic interpretations. Mod Pathol. 1998;11:560–565
- Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005;41:1313–1321
- Chronic liver allograft rejection: a National Institute of Diabetes and Digestive and Kidney Diseases interinstitutional study analyzing the reliability of current criteria and proposal of an expanded definition. Am J Surg Pathol. 1998;22:28–39
- Effect of teaching and type of stethoscope on cardiac auscultatory performance. Am Heart J. 2006;152:85–87
- . Agreement among multiple observers on endoscopic diagnosis of esophageal varices before bleeding. Hepatology. 1990;11:341–347
- . Variability in radiologists’ interpretations of mammograms. N Engl J Med. 1994;331:1493–1499
- Interobserver reproducibility of the Nottingham modification of the Bloom and Richardson histologic grading scheme for infiltrating ductal carcinoma. Am J Clin Pathol. 1995;103:195–198
- . An algorithm for grading activity in chronic hepatitis C. Hepatology. 1996;24:289–293
- . Information and observer disagreement in histopathology. Histopathology. 1994;25:123–128
- Histological grading and staging of chronic hepatitis. J Hepatol. 1995;22:696–699
- . Classification of chronic hepatitis: diagnosis, grading and staging. Hepatology. 1994;19:1513–1520
- . Chronic hepatitis: an update on terminology and reporting. Am J Surg Pathol. 1995;19:1409–1417
- Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology. 1981;1:431–435
- . Natural history and prognostic indicators of survival in cirrhosis: A systematic review of 118 studies. J Hepatol. 2006;44:217–231
- Adefovir dipivoxil for the treatment of hepatitis B e antigen-positive chronic hepatitis B. New Engl J Med. 2003;348:808–816
- Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. New Engl J Med. 1998;339:1485–1492
- Lamivudine as initial treatment for chronic hepatitis B in the United States. New Engl J Med. 1999;341:1256–1263
- Peginterferon alfa-2a in patients with chronic hepatitis C. N Engl J Med. 2000;343:1666–1672
- A randomized, double-blind trial comparing pegylated interferon alfa-2b to interferon alfa-2b as initial treatment for chronic hepatitis C. Hepatology. 2001;34:395–403
- Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet. 2001;358:958–965
- A comparison of entecavir and lamivudine for HBeAg-positive chronic hepatitis B. N Engl J Med. 2006;354:1001–1010
- The long-term pathological evolution of chronic hepatitis C. Hepatology. 1996;23:1334–1340
- Hepatitis activity index is a key factor in determining the natural history of chronic hepatitis C. Hum Pathol. 2001;32:904–909
- Progression of fibrosis in chronic hepatitis C. Gastroenterology. 2003;124:97–104
- . Predicting progression to cirrhosis in chronic hepatitis C virus infection. J Viral Hepat. 2003;10:285–293
- . A histological semiquantitative scoring system for evaluation of hepatic fibrosis in needle liver biopsy specimens: comparison with morphometric studies. Hepatology. 1994;20:349–355
- An assessment of digital image analysis to measure fibrosis in liver biopsy specimens of patients with chronic hepatitis C. Am J Clin Pathol. 2000;114:712–718
- . Progression of fibrosis in chronic hepatitis C: Evaluation by morphometric image analysis. Hepatology. 2007;45:886–894
- . Rating fibrosis progression in chronic liver diseases. J Hepatol. 2003;38:357–360
- . An appraisal of the histopathological assessment of liver fibrosis. Gut. 2006;55:569–578
- . Nonalcoholic steatohepatitis: A proposal for grading and staging the histological lesions. Am J Gastroenterol. 1999;94:2467–2474
- . Staging of chronic non-suppurative destructive cholangitis (syndrome of primary biliary cirrhosis). Virchows Arch A. 1978;379:103–112
- . Morphologic features of chronic hepatitis associated with primary sclerosing cholangitis and chronic ulcerative colitis. Hepatology. 1981;1:632–640
- Banff schema for grading liver allograft rejection: an international consensus document. Hepatology. 1997;25:658–663
- Does the Banff rejection activity index predict outcome in patients with early acute cellular rejection following liver transplantation?. Liver Transplant. 2006;12:1144–1151
☆ The author who has taken part in this study declared that he does not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.
PII: S0168-8278(07)00402-3
doi:10.1016/j.jhep.2007.07.006
Published by Elsevier Inc.
