Histologic scoring systems for chronic liver disease

INTRODUCTION — Histologic scoring systems for chronic liver disease are used to characterize and predict disease progression, to determine prognosis, to guide treatment strategies, and to provide standards in clinical trials.

This topic will review the major histologic scoring systems for chronic liver disease. The interpretation of liver biopsy specimens in general and methods to obtain liver biopsy specimens are discussed elsewhere. (See "Interpretation of nontargeted liver biopsy findings in adults" and "Approach to liver biopsy" and "Transjugular liver biopsy".)

LIMITATIONS OF LIVER BIOPSY — There are several limitations of liver biopsy for assessing liver disease, including:

●Variable quality of liver biopsy specimens – Specimens shorter than 2 cm in length may be difficult to interpret [1]. Larger caliber needles may yield better samples than fine needle biopsies [2,3]. As an example, small specimens may underestimate the degree of inflammatory activity and fibrosis in patients with viral hepatitis [4].

●Sampling variability – The assessment of liver histology is usually based on a percutaneous biopsy that samples a very small portion (1/50,000^th) of the liver [5]. Liver disease does not always affect the liver in a homogeneous pattern, leading to the possibility of sampling variability. The most common setting for sampling error is in a cirrhotic liver [5].

●Subjective assessment – This may lead to inter- and intraobserver variability or error [6-8].

●Fluctuating disease activity – Histologic changes obtained at a single point in time may not reflect overall disease activity, which may vary.

CHRONIC HEPATITIS — There are several scoring systems used to assess chronic hepatitis [9-14]. These systems help to assess prognosis and guide clinical management [15]. The scoring systems typically include descriptions of both necroinflammatory activity and the degree of fibrosis. An important difference among the scoring systems is in the staging of fibrosis. Scoring systems that include more stages for describing fibrosis are better able to document small changes in fibrosis over time. The Knodell score assigns patients to one of four stages, the METAVIR score to one of five, and the Ishak score to one of six.

Early scoring systems recognized two major histologic patterns, "chronic persistent hepatitis" and "chronic aggressive hepatitis" [16]. These patterns were associated with mild and more aggressive histologic categories of liver disease, respectively. Many other scoring systems were subsequently developed. For example, commonly used systems include:

●The Knodell score/histology activity index

●The METAVIR score

●The Ishak score (modified Knodell score)

Other scoring systems include:

●The Scheuer system

●The Batts-Ludwig system

●Laennec staging system

Most of the scoring systems for chronic hepatitis focus on viral etiologies, particularly hepatitis C virus (HCV) and hepatitis B virus infection. The same histologic scoring systems are used for autoimmune hepatitis. In rare instances, autoimmune hepatitis may be difficult to distinguish from chronic HCV [17].

The advent of highly effective, direct-acting antiviral medications for HCV has led to fewer liver biopsies being performed in order to grade and stage HCV because patients are often treated with anti-HCV medications regardless of the stage of fibrosis. (See "Patient evaluation and selection for antiviral therapy for chronic hepatitis C virus infection", section on 'Minimal role of liver biopsy'.)

Knodell score/histology activity index — The Knodell score (also known as the histology activity index) is a composite score that is based on histologic assessment of periportal and/or bridging necrosis, intralobular degeneration and focal necrosis, portal inflammation, and fibrosis (table 1 and picture 1A-E). The score ranges from 0 to 22, with higher scores representing more advanced disease [9]. A decrease in the Knodell score corresponds with histologic improvement.

The Knodell score is frequently used in trials of treatments for chronic hepatitis, particularly HCV. The score is used to assure that baseline histologic features in treatment groups are equally matched and to assess histologic changes after therapy.

A limitation of the Knodell score is that it combines inflammation and fibrosis to arrive at one composite score, so it is relatively insensitive to changes in fibrosis. This is important because it is fibrosis, and not inflammation per se, that leads to many of the sequelae of chronic liver disease. In addition, patients may have the same Knodell score despite having markedly different degrees of fibrosis.

The Knodell score is also associated with high inter- and intraobserver variability. In a study of 10 pathologists using a cohort of 30 liver biopsy specimens from patients who had documented HCV infection, interobserver correlation for the three inflammatory components of the Knodell score was relatively poor (with kappa coefficients ranging from 0.25 to 0.46 [a perfect coefficient being 1.00]) [10]. The interobserver correlation for the overall score was also relatively poor, ranging from 0.48 to 0.57. Only the fibrosis score had good reliability, with a kappa coefficient of approximately 0.80. Similar results were found for intraobserver reliability.

METAVIR score — The METAVIR score is a semiquantitative classification system that consists of an activity score and a fibrosis score (represented by a two letter and two number coding system) (table 2 and picture 2A-D) [10,11]:

●The activity score is graded according to the intensity of necroinflammatory lesions (A0 = no activity, A1 = mild activity, A2 = moderate activity, A3 = severe activity).

●The fibrosis score is assessed on a five-point scale (F0 = no fibrosis, F1 = portal fibrosis without septa, F2 = few septa, F3 = numerous septa without cirrhosis, F4 = cirrhosis).

In contrast to the Knodell score, which was designed as a generic scoring system for chronic hepatitis, the METAVIR score was specifically designed and validated for patients with HCV [10]. The inter- and intraobserver reliability of the activity and fibrosis scores of the METAVIR system are similar to the Knodell score. A study found that the interobserver agreement of the METAVIR score depends highly upon the experience of the hepatopathologist [6]. Agreement was influenced more heavily by the interpreter's experience than by features of the specimen itself, such as its length. However, a separate study evaluating the fibrosis and activity scores in specimens of various lengths suggested that the length of the biopsy is also important [3].

Ishak score (modified Knodell score) — The Ishak score is a modification of the Knodell score that includes six stages of fibrosis (table 3 and picture 3A-G) [12]. This permits documentation of small changes in fibrosis compared with the standard Knodell fibrosis score, which has only four stages. Like the METAVIR score, the grades for necroinflammation (picture 4) are categorized separately from the stage of fibrosis (table 4).

This staging system has become widely used in clinical trials because of its ability to detect mild changes in fibrosis.

Scheuer system — The Scheuer system is a simple scoring system that separates necroinflammation from fibrosis [13]. Histologic findings of portal inflammation, interface hepatitis, and lobular inflammation are each assigned a score of 0 to 4 (table 5). A separate score (0 to 4) is assigned to the stage of fibrosis (picture 5A-D).

Batts-Ludwig system (modified Scheuer system) — This system is also known as the modified Scheuer system [14]. In the original description, diagrams of the different grades and stages of chronic hepatitis were depicted. Either a semiquantitative numeric value or a descriptive term can be used, depending on local preferences (table 6).

●Necroinflammatory activity (grade) is based on lymphocytic piecemeal necrosis (interface hepatitis) and lobular necroinflammatory activity (picture 6A-E).

●The stage refers to the degree of fibrosis (picture 7).

This grading system is applicable to both chronic viral hepatitis (picture 8) and autoimmune hepatitis. This system is more useful for assessing an individual patient's liver biopsy for clinical care than it is for therapeutic trials.

Scoring systems for cirrhosis — More advanced stages of fibrosis, specifically cirrhosis, are not captured by the traditional staging systems described above. Two classifications, the Laennec staging system and the Beijing classification, have been proposed to address this issue [18,19].

Laennec staging system — The Laennec staging system is a modification of the METAVIR F4 (cirrhosis) fibrosis category and was developed based on the observation that cirrhosis may have different degrees of severity that influence the disease behavior and prognosis [19]. (See 'METAVIR score' above.)

The system subdivides F4 cirrhosis into three categories based on the width and number of fibrous septa, as well as the size of the nodules, with more emphasis given to micronodules (picture 9A-C and table 7).

A positive correlation between the Laennec stage and hepatic venous pressure gradient was demonstrated by one study [20]. (See "Portal hypertension in adults", section on 'Hepatic venous pressure gradient'.)

Additional studies have shown that Laennec cirrhosis stage has prognostic value. For example, more severe cirrhosis has been associated with higher risk for a liver-related event (eg, decompensation, hepatocellular carcinoma [HCC]) or a worse outcome after surgical resection for HCC compared with less severe cirrhosis [21,22].

Beijing classification — The Beijing system subclassifies cirrhosis into three categories: predominantly progressive, indeterminate, or predominantly regressive (the P-I-R score) [18,23].The progressive category (P score) shows broad fibrous septa while the regressive category (R score) shows thin septa with foci of disintegration. The indeterminate category (I score) is a balance of regression and progression. This classification is particularly useful for patients with chronic viral hepatitis because it facilitates comparing the degree of cirrhosis before and after antiviral therapy (image 1).  

DISEASE-SPECIFIC SCORING SYSTEMS — Disease-specific scoring systems are also available, including scoring systems for nonalcoholic fatty liver disease (NAFLD), alcoholic hepatitis, primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC). However, it is not uncommon to encounter two or more concurrent diseases in a liver biopsy specimen, and no scoring systems are available that specifically address these situations.

Nonalcoholic fatty liver disease — NAFLD is subdivided into nonalcoholic fatty liver (NAFL) and nonalcoholic steatohepatitis (NASH) although these are considered to be part of a spectrum. In NAFL, hepatic steatosis is present without evidence of hepatocyte ballooning degeneration or significant inflammation, whereas in NASH, hepatic steatosis is associated with hepatocyte ballooning degeneration and hepatic inflammation that may be histologically indistinguishable from alcohol-associated steatohepatitis [24,25]. The clinical features and diagnosis of NAFLD are discussed separately (See "Epidemiology, clinical features, and diagnosis of nonalcoholic fatty liver disease in adults".)

Although liver biopsy can confirm the diagnosis of NASH, grade disease severity, and accurately stage fibrosis, noninvasive methods for evaluating NASH are being studied.

Brunt scheme for NASH — The Brunt scheme includes grading and staging histologic features of NASH. This system is easily reproducible and continues to be used in most medical centers in the United States. Grade, which ranges from mild, moderate, to severe (corresponding to grades 1,2,3) is based on the combined features of steatosis, ballooning, and lobular and portal inflammation. Stage refers to the degree of fibrosis. The stage of fibrosis also takes into consideration the location of fibrosis, particularly in the early stage of disease [26]: Stage 1, zone 3 perisinusoidal fibrosis; Stage 2, both zone 3 perisinusoidal fibrosis and portal fibrosis; Stage 3, bridging fibrosis; and Stage 4, cirrhosis.

Nonalcohol-associated fatty liver activity score — The nonalcohol-associated fatty liver activity score (NAS) (picture 10A-C and table 8) is a histologic evaluation system that includes the full spectrum of nonalcohol-associated fatty liver disease histologic changes and can assess changes following therapy in both adult and pediatric patients [24]. The NAS is the unweighted sum of scores for:

●Steatosis

●Lobular inflammation

●Ballooning

The NAS ranges from 0 to 8 [24]. Fibrosis is not included in the NAS. In the study that derived the NAS, scores of 0 to 2 occurred in cases largely considered not diagnostic of NASH; scores of 3 to 4 were evenly divided among those considered not diagnostic, borderline, or positive for NASH; and scores of 5 to 8 occurred in cases that were largely considered diagnostic of NASH [24]. In practice, the NAS is used to grade the activity of NASH, rather than to make a histologic diagnosis of NASH [27]. The original role for the NAS was to assess liver biopsies for patients who were enrolling in clinical trials. However, threshold values for NAS do not always correlate with the diagnosis of NASH and therefore may not be used solely for establishing diagnosis.

Steatosis, activity and fibrosis system — With the steatosis, activity, and fibrosis (SAF) system, a diagnosis of NASH can be established [28]. Steatosis, activity, and fibrosis are separately assessed and the sum of the scores for lobular inflammation and ballooning define activity (range, 0 to 4). Patients with activity score of 2 or greater are diagnosed with NASH.

Fatty liver inhibition and progression — The fatty liver inhibition of progression (FLIP) pathology consortium created a histologic algorithm (FLIP algorithm) based on the SAF scoring system to limit interobserver variation among pathologists. [29].

Alcoholic liver disease and alcoholic hepatitis — Histologic grading and staging systems for alcoholic liver disease have not been validated, and scoring systems developed for NAFLD and NASH are being used for alcohol-related liver disease. Histologically, alcoholic steatohepatitis (ASH) is similar to NASH, and often these two entities are indistinguishable from each other on biopsy. Some features may favor a diagnosis of alcoholic hepatitis, though they are not specific to alcoholic hepatitis. These features include cholestasis, numerous well-formed Mallory-Denk bodies, neutrophilic satellitosis, and central hyaline sclerosis [30]. (See "Interpretation of nontargeted liver biopsy findings in adults", section on 'Diseases associated with fatty liver'.)

One study proposed a semiquantitative histologic scoring system (the Alcoholic Hepatitis Histologic Score [AHHS]) that relates to the prognosis in patients with alcoholic hepatitis (table 9) [31]. Based on multivariate regression analysis, the group came up with a scoring system that utilizes four histologic features, scoring each feature individually (degree of fibrosis, degree of neutrophil infiltration, type of bilirubinostasis, and presence of megamitochondria) (picture 11A-D). The risk of death is positively associated with the stage of fibrosis and degree of bilirubinostasis and is negatively correlated with neutrophil infiltration and the presence of megamitochondria (ie, mild neutrophil infiltration or the absence of megamitochondria is associated with a worse prognosis). The individual scores are combined to reach a final score (maximum score of 9). The investigators found that a low score (0 to 3) was associated with a 90-day mortality of 3 percent, whereas moderate (4 to 5 points) and high (6 to 9 points) scores were associated with higher mortality rates (19 and 51 percent, respectively). However, a subsequent validation study showed only a fair level of agreement in assigning the different categories, and this may limit the use of AHHS in clinical practice [32].

The SALVE Grading and Staging System was developed and validated by a group of expert liver pathologists under the auspices of the European Association for the Study of the Liver [33]. Grade is based on the degree of steatosis, activity, and cholestasis. Alcohol-associated steatohepatitis requires both ballooning of hepatocytes and neutrophils in the lobule. The stage, which is adapted from the NASH Clinical Research Network staging for NAFLD and the Laennec system, refers to the degree of fibrosis and the subtypes of cirrhosis, respectively. There are seven possible fibrosis stages in this scheme. Both the grade and the stage in this system provide prognostic value.

Primary biliary cholangitis — Primary biliary cholangitis (PBC) is classically staged using either the Scheuer system for PBC (table 10 and picture 12A-D) or a system proposed by Ludwig (table 11) that differs from the Batts-Ludwig system described above [34-36]. (See "Clinical manifestations, diagnosis, and prognosis of primary biliary cholangitis", section on 'Liver biopsy'.)

Because these scoring systems do not include the degree of portal lymphoplasmacytic inflammation that may be found in association with PBC, a histologic staging and grading system specifically for PBC has been proposed. In the Nakanuma system, histologic stage is based on three features: fibrosis, bile duct loss, and deposition of copper-associated protein on orcein stain.

The final Nakanuma stage is obtained from the total score of the three features of histologic stage [37,38]:

●Stage I = no or minimal progression (score of 0)

●Stage II = mild progression (score of 1 to 3)

●Stage III = moderate progression (score of 4 to 6)

●Stage IV = advanced progression (score of 7 to 9)

The Nakanuma scoring system for histologic grade includes degree of cholangitis activity (score 0 to 3), and hepatitis activity (score 0 to 3) [37]. A validation study using this system showed that the cholangitis and hepatitis activity scores did not correlate well with each other in terms of the degree of inflammation, while the staging system correlated with the patient outcome [39]. Another study demonstrated that this staging system correlated with laboratory values, progression of the disease, and the development of cirrhosis [40].

Primary sclerosing cholangitis — It is uncommon to perform a liver biopsy to make a diagnosis of PSC, but if a biopsy is performed, the Ludwig histologic staging system used for PSC is as follows [41-44] (picture 13A-D and table 12) (see "Primary sclerosing cholangitis in adults: Clinical manifestations and diagnosis", section on 'Liver biopsy'):

●Stage 1: The portal tracts are expanded by edema and fibrosis, and there is a mononuclear cell infiltration with some interface hepatitis and damage to isolated bile ducts. Proliferation of bile ductules with mononuclear and polymorphonuclear cells may also be present, although the inflammation is usually less dense than in PBC.

●Stage 2: Expansion of portal triads with fibrosis extending into the surrounding parenchyma.

●Stage 3: Bridging fibrosis.

●Stage 4: Cirrhosis.

In a study that evaluated three scoring systems (Ishak, Ludwig and Nakanuma) to determine the applicability and prognostic value of scores in a group of PSC patients, interobserver agreement was moderate to substantial in all three systems [12,35,37,38]. All three staging systems were also shown to be independent predictors of both liver transplantation and complications related to cirrhosis, but they found that the Nakanuma staging system (while initially intended for staging PBC) was independently associated with the endpoint of PSC-related death and liver transplantation (HR 2.14, 95% CI 1.22-3.77).

Autoimmune hepatitis — The histologic grading systems for chronic hepatitis are also applied to autoimmune hepatitis (AIH) because, no formal histologic classification system specifically for AIH has been established. (See 'Chronic hepatitis' above.)

For the last decade, a diagnostic scoring system developed by the International Autoimmune Hepatitis Group (IAIHG) has used simplified criteria that are based upon titers of autoantibodies, immunoglobulin levels, liver histology, and the exclusion of viral hepatitis [45]. (See "Overview of autoimmune hepatitis", section on 'Diagnostic scoring systems'.) 

The simplified criteria have two categories: "definite autoimmune hepatitis" (score of 7 or more) or "probable autoimmune hepatitis" (score of 6 or less). On biopsy, 2 points are given to the typical AIH features of interface hepatitis with lymphocytic or lymphoplasmacytic infiltrate, rosettes, and emperipolesis. If one of these features is absent, only 1 point (compatible) is given. Although the simplified criteria have a high specificity for the diagnosis of AIH, their sensitivity is quite low. Histologic underscoring may be a reason for the low sensitivity of the simplified criteria. Furthermore, rosettes are a nonspecific feature and may be seen in any type of regenerative condition while emperipolesis is very difficult to identify on routine histology. Thus, several investigators have attempted to improve the sensitivity of diagnosing AIH on liver biopsy.  

Modified histologic criteria for AIH have been proposed and have used the Ishak grading scheme to determine the degree of inflammation and the Batts-Ludwig scheme for the stage of fibrosis. (See 'Ishak score (modified Knodell score)' above and 'Batts-Ludwig system (modified Scheuer system)' above.)

In addition to the Ishak necro-inflammatory activity score, the extent of plasma cell infiltrates and copper and CK/7 staining were incorporated to result in a composite histology score (score 0, 1, and 2). These modified criteria have increased the number of diagnostic histologic AIH cases in the probable or definite category, whereas such cases would have previously been classified as non-AIH using the simplified criteria [46].  

Other groups have included additional features such as increased lymphocyte apoptotic body count in portal tracts. Adding such features has resulted in more patients with AIH who meet criteria for the "definite" category using the simplified AIH scoring system [47]. 

Congestive hepatopathy — The congestive hepatic fibrosis score (CHFS) is a grading system for congestive hepatopathy, and it is a useful indicator of disease severity. In a study including 42 patients with congestive hepatopathy who had liver biopsies and cardiac testing (ie, echocardiography or right heart catheterization), histologic features of sinusoidal dilatation, centrilobular hepatocyte atrophy, centrilobular fibrosis, and portal fibrosis correlated with right atrial pressures. Higher CHFS scores also correlated with right atrial and right ventricular dilation [48]. In addition, a validation study showed that the CHFS correlated with clinical markers of congestive hepatopathy (eg, laboratory studies, cardiac imaging) [49] (image 1). However, limitations of liver biopsy include the heterogeneous distribution of fibrosis in patients with congestive hepatopathy [50]. Accurate staging of liver fibrosis helps to assess hepatic function in patients with congestive hepatopathy, and for some patients, it is an important part of the evaluation for heart transplantation. (See "Heart transplantation in adults: Indications and contraindications".)

Chronic elevation of hepatic venous pressure secondary to right-sided heart failure can lead to liver fibrosis. The etiology and management of congestive hepatopathy is discussed separately. (See "Congestive hepatopathy".)

A growing number of patients with congestive hepatopathy are those who had congenital single ventricle systems and underwent the Fontan procedure, and the risk of liver disease after the Fontan procedure is discussed separately. (See "Management of complications in patients with Fontan circulation", section on 'Liver disease'.)

SUMMARY

●Background – Histologic scoring systems for chronic liver disease are used to characterize and predict disease progression, to determine prognosis, to guide treatment strategies, and to provide standards in clinical trials. (See 'Introduction' above.)

●Scoring systems for chronic hepatitis – Commonly used scoring systems for chronic hepatitis include:

•The Knodell score/histology activity index (table 1) (see 'Knodell score/histology activity index' above).

•The METAVIR score (table 2) (see 'METAVIR score' above).

•The Ishak score (modified Knodell score) (table 3 and table 4) (see 'Ishak score (modified Knodell score)' above).

These scoring systems include descriptions of both necroinflammatory activity and the degree of fibrosis. An important difference among the scoring systems is in the staging of fibrosis. Scoring systems that include more stages for describing fibrosis are better able to document small changes in fibrosis over time. The Knodell score assigns patients to one of four stages, the METAVIR score to one of five, and the Ishak score to one of six.

Other scoring systems include:

•The Scheuer system (table 5) (see 'Scheuer system' above).

•The Batts-Ludwig system (table 6) (see 'Batts-Ludwig system (modified Scheuer system)' above).

•Laennec staging system (table 7) (see 'Laennec staging system' above).

●Other disease-specific scoring systems – Scoring systems also exist for specific liver diseases, including:

•Nonalcohol-associated fatty liver disease (table 8 and table 13) (see 'Nonalcoholic fatty liver disease' above).

•Alcoholic hepatitis (table 9) (see 'Alcoholic liver disease and alcoholic hepatitis' above).

•Primary biliary cholangitis (table 10 and table 11) (see 'Primary biliary cholangitis' above).

•Primary sclerosing cholangitis (table 12) (see 'Primary sclerosing cholangitis' above).

•Congestive hepatopathy (see 'Congestive hepatopathy' above).