Chronic hepatitis indicates a persistent, and often progressive, inflammatory process of the liver that is characterized histologically by lymphocytic infiltration of the portal tracts together with varying degrees of parenchymal inflammation, hepatocellular injury, and fibrosis.

Although this definition appears straightforward, there are certain difficulties in its practical application. The crux of the problem is in establishing that the process is indeed chronic and thus potentially progressive. Chronicity can be judged in several ways -- clinical findings, laboratory abnormalities, or morphologic assessment -- but, unfortunately, none of these is an absolutely reliable guide in separating acute hepatitis, which is usually self-limited, from chronic hepatitis. When all these evaluations are in agreement, the diagnosis poses few problems, but sometimes the criteria are discordant or ambiguous.² In those cases, the duration of the disease becomes an important consideration in designating chronicity. Although the choice is somewhat arbitrary, a minimum of six months duration is the commonly agreed-upon period. This temporal criterion is particularly useful in deciding when a patient with protracted acute viral hepatitis can be definitively said to have developed chronic hepatitis.

Chronic hepatitis encompasses several distinct disorders with differing etiologies and natural histories. The causes include several of the hepatitis viruses, presumed autoimmune reactions, and drug-induced injury. Certain inherited metabolic disturbances such as Wilson's disease and alpha-1-antitrypsin deficiency may also manifest with chronic hepatitis; briefly mentioned here, they are discussed in more detail in Chapter 9. As defined above, chronic hepatitis could also embrace such conditions as the chronic biliary disorders -- primary biliary cirrhosis, primary sclerosing cholangitis, and chronic biliary obstruction -- but these are generally considered to be separate conditions and are discussed in Chapters 4 and 5.

Chronic hepatitis has customarily been divided into two major nosologic groups, chronic persistent hepatitis and chronic active hepatitis. An additional group, chronic lobular hepatitis, was introduced later, but represents a minor and not uniformly accepted category.^1,3-5 These three groups were distinguished primarily on morphologic grounds (Table 3-1), but their discrimination was further justified by differences in prognosis. Thus, chronic persistent hepatitis and chronic lobular hepatitis, characterized by uncomplicated portal inflammation and pronounced lobular inflammation respectively, were considered benign, nonprogressive lesions, and chronic active hepatitis, with its aggressive hepatocellular necrosis and fibrosis, was regarded as a serious, progressive process leading to cirrhosis.

This widely-employed categorization, which was originally formulated 25 years ago, has contributed greatly to our understanding of chronic hepatitis and has provided a basis for patient evaluation and management, but it has certain drawbacks. Our current understanding of the biology and natural history of chronic hepatitis suggests that the three categories do not represent distinct disease entities; they instead define portions of a broad histologic and clinical spectrum. Some overlap among the groups is therefore inevitable. Moreover, although the prognostic differences continue to be generally accurate, they have become less distinct as further knowledge about the causes and evolution of chronic hepatitis accumulates. The classification nevertheless remains a practical means of designating the extent and severity of disease, expressing the range of histologic findings, and providing partial prognostic information, despite its weaknesses.

The primary causes of chronic hepatitis are the hepatitis viruses and autoimmune reactions, with drugs and metabolic disturbances being less frequent (Table 3-2). Clinical information and laboratory tests are the chief means of establishing an etiologic diagnosis, although histologic examination can yield useful information in some circumstances.^35,61 Three of the hepatitis viruses, the hepatitis B, C, and D viruses, have been associated with chronic hepatitis, and their overall biologic and epidemiologic aspects are discussed in Chapter 2.

Hepatitis B ^28,36,70

Between 1% and 10% of adults with acute hepatitis B will develop a chronic infection. The factors that control viral persistence are not well understood, but chronic infections are more apt to develop in infants and in individuals who are immunosuppressed, receiving cancer chemotherapy or hemodialysis, or have Down's syndrome.⁵²

The natural history of chronic hepatitis B infection, although incompletely defined, is separated into two general phases. The early phase is characterized by active, high-level viral replication with production of intact infectious virions. Serologic markers of replication, such as e antigen (HBeAg), HBV DNA, and DNA polymerase, are consequently detectable in the serum. This stage of the infection tends to correlate most strongly with hepatic injury, although there is a wide range in the severity of the damage: Severe chronic active hepatitis develops in some patients, but milder disease is more commonly the result and is particularly frequent among children, hemodialysis patients, and immunosuppressed individuals.^{13,24,30,31,57}

After an interval ranging from months to years, the infection evolves into a low-grade replicative phase and viral production declines. Serologically, this is marked by disappearance of HBeAg, HBV DNA, and DNA polymerase from the serum and the corresponding acquisition of antibody to HBe (anti-HBe). Seroconversion to anti-HBe occurs spontaneously in about 10% to 20% of infected patients per year.

The transition from the high to the low replication phase may be preceded by an episode of hepatocellular injury and inflammation, which may even result in massive hepatic necrosis and death.^45,73 However, once seroconversion to anti-HBe has occurred, the liver damage tends to resolve and become inactive. The liver may completely return to normal, a circumstance often referred to as the "healthy" viral carrier state, but any fibrosis that has supplanted earlier hepatic necrosis may still be evident.^13,27,36 Moreover, portions of the HBV genome may later become integrated into the host DNA. This can enable continued synthesis of HBsAg despite the lack of complete virus production. However, approximately 1% of patients annually also lose detectable HBsAg.^31,36 Incorporation of the viral genome appears to be an important step in the development of hepatocellular carcinoma, a major long-term complication of chronic hepatitis B infections.⁶⁴

Although the later stages of the infection are therefore generally accompanied by quiescent disease, there are exceptions. Some patients experience an acute exacerbation with a reactivation of active viral replication and hepatic injury.^22,46,63 This reactivation, which may be transient, recurring, or persistent, can occur spontaneously or follow immunosuppressive or antineoplastic therapy, and it may lead to severe or progressive disease. Another exception is noted in other patients who, despite being anti-HBe positive, have ongoing viral replication and suffer continuing liver disease. In some instances, viral mutants appear to be responsible for these anomalous findings.^27,31,62

The serologic diagnosis of chronic hepatitis B is established by detecting HBsAg in the serum for more than six months. Usually no (or only low-titer) IgM anti-HBc is present, and, depending on the phase of the disease, HBeAg, HBV DNA, or DNA polymerase activity may be noted.³⁵

Although many aspects remain unclear, the pathogenesis of chronic hepatitis B is generally considered to involve cell-mediated immunologic mechanisms. The general idea is that the infected hepatocyte, perhaps through the expression of HBcAg on its surface, becomes a target for attack by cytotoxic T lymphocytes. Although this is a process similar to that occurring in acute hepatitis, it differs because the immune response is deficient in some way and fails to eliminate the virus; the destruction of liver cells therefore continues apace. The reasons for this immunologic impotence are not known, but several possibilities with variable experimental support have been advanced.^6,58,59,70 For example, the virus itself may disrupt the production or response to the endogenous anti-viral agent interferon, thereby curbing viral eradication. Deficient interferon activity may also affect the expression of the class I histocompatibility products that are required for appropriate recognition of the viral antigens. Another suggestion is that high-titers of anti-HBc, by binding up the HBcAg on the hepatocyte surface, might inhibit the T lymphocyte response. Spontaneous mutation of the viral genome during the course of the disease could potentially alter viral antigen expression, disrupt the immune response, and modify the activity of the disease. All these mechanisms reflect alterations in host-viral interactions, and further insights in their nature should be forthcoming.

Hepatitis C ^23,74

Our understanding of hepatitis C infections are at a preliminary stage since the responsible virus has only recently been discovered. Nonetheless, since most cases of so-called non-A, non-B hepatitis are now known to be caused by the hepatitis C virus, the conclusions drawn about that condition remain generally pertinent.

Hepatitis C infection frequently results in chronic hepatitis; in several studies, some 50% to 70% of patients with acute hepatitis C have persistent liver function abnormalities.^8,43,54,67 Moreover, more than half of otherwise unexplained cases of chronic hepatitis can be attributed to hepatitis C.³⁸ These observations indicate that hepatitis C is a major and probably underrecognized cause of chronic liver disease.

The natural history of chronic hepatitis C infection remains to be fully delineated, and there are few prospective studies that utilize serial liver biopsies. One difficulty is that chronic hepatitis C is often clinically silent: Even in the face of advanced disease, patients may be only mildly symptomatic. The typical course is indolent, often subclinical, and characterized by fluctuating degrees of hepatic damage. Some cases evolve only slowly or not at all, but others covertly progress; some 20% to 25% of patients will have cirrhosis at the time of initial liver biopsy.^9,74 A more rapid progression is occasionally noted, and, in some instances, this has been associated with coinfection by other agents such as the human immunodeficiency virus.^53,67

Biochemical abnormalities may disappear in the later stages of the disease, but the virus can persist in the circulation for many years; whether the infection may eventually resolve is yet unanswered.²⁵ Chronic hepatitis C infection may also be complicated by hepatocellular carcinoma.^32,42

The serologic diagnosis of chronic hepatitis C is limited by the relative insensitivity and nonspecificity of the early tests, but better assays are becoming more widely available. The presence of anti-HCV in the proper setting provides presumptive diagnostic evidence.³⁵

Hepatitis D ^11,16,34

Because the hepatitis D virus (delta agent) requires HBsAg for its replication, chronic hepatitis D infections can arise only in patients with concurrent chronic hepatitis B. Consequently, chronic hepatitis D is most apt to develop in hepatitis B carriers who are acutely superinfected by hepatitis D, although it may occasionally follow an acute coinfection with both viruses.¹²

As a general rule, chronic hepatitis D tends to be more severe, with greater activity and risk of progression, than does uncomplicated chronic hepatitis B.^26,40,47 Rapid progression to cirrhosis is the result in a minority of patients, particularly those with active hepatitis B replication. More commonly, the evolution is slow and insidious, leading to the development of cirrhosis only after many years.^10,68 The factors that control the outcome are not well understood, but the status of the concurrent hepatitis B infection, the presence of coinfections, and the genetic background of the infected patient have all been suggested to play a role.

The serologic diagnosis is established by detecting, together with HBsAg, high titers of antibody to hepatitis D antigen (anti-HDV). Markers of active replication, such as IgM anti-HDV, HDV RNA, and HDV antigen, can help identify ongoing hepatitis D disease, but they are not widely available.^11,35

Autoimmune-Associated ^19,39,69

The cause of this important and classic type of chronic hepatitis is not known, but many of its characteristic features point to an autoimmune origin. These include a female predominance, the frequent presence of circulating autoantibodies, an association with other presumed autoimmune disorders (for example, thyroiditis, vasculitis, uveitis, and arthritis), and a dramatic response to corticosteroid therapy. Several names have been applied to this entity since its original description in the 1950's -- including lupoid hepatitis, plasma cell hepatitis, and active juvenile cirrhosis -- but the term autoimmune chronic hepatitis is commonly employed today. Some patients with idiopathic, nonviral ("cryptogenic") chronic hepatitis lack the typical immunoserologic abnormalities, but they are otherwise comparable to those with autoimmune features.^20,21,41

The typical patient is a woman of either 15 to 35 years or perimenopausal age, but examples are reported at all ages and in both sexes. A genetic predisposition is suggested by the presence of serum autoantibodies and other autoimmune manifestations in family members and by the increased prevalence, in common with other autoimmune diseases, of HLA haplotypes A1, B8, and DR3. Serum IgG levels are elevated, often to a striking degree, and a variety of non-organ-specific autoantibodies can be detected. These variably include antinuclear antibodies, antibodies to double-stranded DNA, positive tests for lupus erythematosus cells, antismooth muscle antibodies directed against actin, and, uncommonly, antimitochondrial antibodies at low titer.⁵¹ Some investigators have identified an anti-liver/kidney microsomal antibody (anti-LKM-1) that reacts to a cytochrome P450 isozyme and have further suggested that its presence defines a distinct subgroup.³³ Although in most respects this uncommon variant conforms to the usual picture of autoimmune hepatitis, it tends to occur in younger patients, often in the absence of antinuclear and antismooth muscle antibodies, and is associated with more serious disease and a worse prognosis.^48,50

The natural history of autoimmune chronic hepatitis is not completely characterized, and, in particular, the course of the less severe forms is poorly defined. In general, the disease evolves with recurring bouts of hepatic injury that variably, and often insidiously, terminate in cirrhosis. An acute onset has sometimes been recognized, but it is unclear whether this is a true inaugural phase of the disease or simply represents an episode of active exacerbation complicating an ongoing chronic process.^18,44 Progression can be swift in patients presenting with markedly active disease, but the evolution in milder cases appears less aggressive and the exact risk of cirrhosis unsettled.^14,15 Immunosuppressive therapy is effective in repressing hepatic damage, retarding disease progression, and prolonging survival in severe cases, but the development of cirrhosis is not prevented.

The pathogenesis of this condition is a matter of considerable conjecture, but it is generally thought to involve a immunologically-mediated attack on hepatocytes. There is indirect and circumstantial support for this concept, but the initiation, target, and nature of the attack remain unknown.^56,76 Numerous experimental studies using in vitro assays have demonstrated evidence of immune sensitization to the liver cell membrane and, in particular, to an antigen complex referred to as liver specific protein. Almost all patients with autoimmune chronic hepatitis have autoantibodies directed against liver specific protein; recent evidence indicates that the asialoglycoprotein receptor is the major target.^65,77 By binding to the hepatocyte surface, these antibodies presumably provoke an assault through antibody-dependent cellular cytotoxicity. A decrease in suppressor T cell activity is also commonly noted and may act to facilitate the synthesis of the autoantibodies. These possibilities are intriguing, and their role in inducing or perpetuating chronic hepatitis is under intense investigation.

Drug reactions are an unusual -- but perhaps underrated -- cause of chronic hepatitis. Several agents have been implicated: the now-discontinued laxative oxyphenisatin, methyldopa, and nitrofurantoin are most often cited, and sporadic reports incriminate such drugs as isoniazid, dantrolene, halothane, ticrynafen, and aspirin.^37,49 Prolonged exposure to these drugs is usually a prerequisite for the development of chronic hepatitis; presumably the persistence of an initial low-grade or unrecognized acute hepatitis is the responsible process.

The clinical and histologic features of drug-induced chronic hepatitis may be indistinguishable from any of the other forms, and a careful drug history prompted by a proper degree of suspicion is the sole means of establishing the correct diagnosis. The importance of recognizing a drug etiology is that, once the offending agent is discontinued, the disease resolves.

Two inherited metabolic disorders, Wilson's disease and alpha-1-antitrypsin deficiency, can give rise to a clinical and pathologic syndrome indistinguishable from chronic hepatitis.^7,66,71,72 These uncommon causes, described in detail in Chapter 9, are important to recognize because of the implications for treatment and family counseling. The clinical setting and histologic features may be suggestive, but appropriate laboratory studies (such as hepatic copper determinations or serum phenotyping) are needed to establish the diagnosis. With both conditions, advanced disease and cirrhosis are commonly present at the time of presentation.

Chronic hepatitis is also occasionally noted in conjunction with chronic alcohol abuse, and an etiologic relationship between the two has been proposed.^17,29,60 The issue is complicated, however, because alcoholic patients have a greater exposure to hepatitis viruses, particularly the hepatitis C virus, than the general population. The suspicion is therefore that a superimposed chronic viral infection is actually responsible for the observed chronic hepatitis in some patients. In other patients though, viral markers are absent and no other cause of chronic hepatitis can be implicated: these may represent examples of alcohol-induced chronic hepatitis, presumably arising on an immunologic basis, although such cases are rare.^55,75

CLINICAL FEATURES ^61,79,87

As befits a heterogeneous disorder, chronic hepatitis encompasses a wide range of clinical manifestations. Many patients are either asymptomatic or have only minor complaints while others exhibit features of chronic liver disease or hepatic failure.^86,89 The disease may have an insidious onset or may present abruptly; occasionally it follows an episode of acute hepatitis that fails to remit. In many patients, the sum of the clinical, biochemical, and histologic findings clearly indicate that the patient is suffering chronic disease, but this is not always the case, and documenting a six-month or longer duration of disease may then be necessary.

The most common symptoms include fatigue or malaise, with mild abdominal discomfort, jaundice, anorexia, fever, nausea, or arthralgias less often noted. Extrahepatic manifestations such as amenorrhea, acne, gynecomastia, or Cushinoid changes can be prominent, particularly with autoimmune chronic hepatitis, and may represent the presenting complaint. Physical examination findings are few: hepatomegaly, mild splenomegaly, and, occasionally, spider angiomas or palmar erythema may be discovered. In advanced disease, hepatic decompensation may supervene, as manifest by ascites, varices, or encephalopathy. Unfortunately, the clinical findings are not specific and correlate poorly with the severity of the disease.¹⁰¹

The characteristic laboratory abnormalities include an increase, usually of two- to 10-fold, in serum transaminase levels. There is no good correlation between the degree of elevation and the state of progression, although, as a general rule, substantial increases signify markedly active hepatocellular damage. Other liver tests are generally normal or only mildly abnormal; increasing serum bilirubin, falling albumin levels, and a prolonged prothrombin time signal advanced, progressive disease.

The etiology is typically established through serologic studies. Chronic viral hepatitis is suggested by the appropriate antibodies and confirmed by markers of ongoing viral replication.³⁵ Autoimmune chronic hepatitis is largely a diagnosis of exclusion; the finding of elevated serum immunoglobulins and antinuclear or antismooth muscle antibodies, particularly at high titers, points to this condition, but these alterations are not specific and may be missing in some cases.²⁰

The overall survival of chronic hepatitis is approximately 80% at five years, but the figure varies from 40% to almost 100% depending on the cause, the severity of the disease, and extent of progression. In general, patients who have severe parenchymal injury, developed advanced disease, or ongoing viral replication suffer the poorest outcome, but the variables are many and confounding.^{41,84,91,92,105} The long-term outcome is also compromised by the development of hepatocellular carcinoma, a clearly established complication of chronic hepatitis B and increasingly recognized with chronic hepatitis C as well.^32,42 Autoimmune chronic hepatitis has less often been associated with carcinoma, but the risk may be greater than usually appreciated, particularly in patients with long-standing cirrhosis.¹⁰⁴

Immunosuppressive therapy represents the principal treatment for chronic hepatitis, and the greatest benefit is found in symptomatic patients with severe chronic autoimmune hepatitis. In these patients, corticosteroids (with or without added azathioprine) diminishes disease activity, eases symptoms, and prolongs five-year survival from approximately 40% to over 80%.^20,41,88 However, most patients with autoimmune hepatitis have disease of lesser severity, posing a therapeutic dilemma for the clinician since corticosteroids are not of proven benefit in this setting.⁹³

In viral-induced chronic hepatitis, immunosuppressive drugs have limited value, except perhaps in patients in severe and rapidly progressing disease. Antiviral and immunomodulating agents would be logical therapeutic choices, but most studies have yielded disappointing results. Recently, however, interferon alfa has shown promise: A biochemical and histologic remission is induced in approximately 30% to 50% of patients with chronic hepatitis B or C.^81,97 Whether this alters the natural history of the disease is yet uncertain, and further studies are necessary to define the optimal treatment regime and the cost-benefit ratio of its use.

Liver transplantation remains a therapeutic option for decompensated, end-stage chronic hepatitis. The results are generally favorable, but chronic hepatitis B is the disappointing exception. The hepatitis B infection recurs in most patients, leading to a return of hepatic injury with consequently greater risk of graft loss and death.¹⁰²

The indications for transplantation in this population thus remain unclear, although various strategies to prevent reinfection are being actively investigated.⁹⁰ In contrast, recurrence of hepatitis C appears in preliminary studies to occur in only a minority of patient; the reappearance of autoimmune chronic hepatitis is an exceptional event.^94,95,99

Chronic hepatitis encompasses a broad range of histologic appearances that center around three primary alterations: portal inflammation, lobular inflammation and damage, and piecemeal necrosis. Fibrosis may also develop as a complication of these lesions, and, in advanced cases, cirrhosis may result. The alterations vary considerably in degree and are often distributed irregularly across the liver; sampling error therefore becomes an important consideration when assessing severity and progression in needle biopsy specimens.^106,148,151

The central feature of most cases of chronic hepatitis is a prominent inflammatory infiltration of the portal tracts (Figure 3-1). This infiltrate consists primarily of lymphocytes with variable plasma cells and macrophages and occasionally a few neutrophils or eosinophils. The lymphocytes may be organized into lymphoid follicles, sometimes with germinal centers (Figure 3-2). Most portal tracts are involved to some extent, but the intensity of involvement is ranges from a sparse collection to a dense crowding of the portal tract; some portal tracts may be spared. Bile ductular proliferation is occasionally noted but is generally of mild degree, and interlobular bile ducts may be damaged, particularly in chronic hepatitis C, as manifest by epithelial swelling, vacuolation, and inflammatory infiltration (Chapter 5).

Lobular inflammation and damage is an inconsistent feature of chronic hepatitis (Figure 3-3). It is characterized by changes similar to those seen in acute hepatitis, including varying degrees of ballooning or acidophilic degeneration, focal hepatocellular necrosis, acidophilic bodies, and an accompanying mononuclear inflammatory component. In severe cases, confluent necrosis including bridging or multilobular necrosis may be noted. The magnitude of these lobular changes generally mirrors the clinical and biochemical activity of the disease: minor alterations are noted in quiescent phases and more conspicuous findings seen with exacerbations of active hepatic injury or with superinfections by other hepatitis viruses.¹⁵⁵ In chronic hepatitis, this lobular reaction tends to be milder and less regularly distributed than in acute hepatitis, but this is not always the case.^143,147

Piecemeal necrosis designates the presence of periportal injury and inflammation. It is characterized by extension of the portal inflammatory infiltrate into the adjacent parenchyma together with destruction of individual hepatocytes along the edges of the portal tract (Figure 3-4) . The interface between liver cells and connective tissue consequently becomes ragged and irregular, sometimes with trapping of small clusters of liver cells within the infiltrate.^109,111 Periportal fibrosis is an almost universal accompaniment and imparts a stellate profile to the edge of the portal tract best appreciated on connective tissue stains (Figure 3-5). The extent and distribution of piecemeal necrosis is variable and often uneven. In mild disease, only occasional foci of periportal necrosis may be seen, whereas with more vigorous involvement, the entire portal tract circumference may be affected.

The nature of piecemeal necrosis is not well understood, but, from the time the term was introduced, it has been considered to be an immunologically-mediated phenomenon.¹⁴⁴ That concept derives indirect support from several investigative studies. The lymphocytes that accumulate in piecemeal necrosis are largely suppressor/cytotoxic (CD8+) T lymphocytes together with a variable number of CD4+ cells. They express various cellular adhesion molecules and receptors, indicating that they are biologically active and presumably engaged in liver cell injury. This is further implied by the intimate contact between the membranes of the lymphocytes and damaged hepatocytes.^107,120 Moreover, antigen-presenting cells such as interdigitating and dendritic reticulum cells are also found in the inflammatory infiltrate.¹¹⁰ These observations suggest that piecemeal necrosis represents a liver-based immune response fundamentally similar to that occurring in lymphoid organs. The reasons for the periportal localization of such processes are unknown, but this may relate to the differential distribution of the target antigens.¹³⁶

Based on the relative predominance of these alterations, the three customary types of chronic hepatitis -- chronic persistent hepatitis, chronic active hepatitis, and chronic lobular hepatitis -- can be defined (Table 3-1). These should be considered not as discrete diagnostic entities but rather as histologic patterns that denote the extent and severity of disease within a morphologic continuum. Consequently, no absolute lines of demarcation among the three types exist, and their distinction blurs at the borders of their definitions. In addition, transitions from one type to another occur spontaneous and following treatment.¹¹⁸ Fortunately, once clinical information is taken into account, these pathologic problems do not translate into great differences in prognosis or treatment. As noted before, there are drawbacks to classifying chronic hepatitis into the these traditional groups, but the scheme serves as a practical strategy for characterizing the histologic spectrum of chronic hepatitis.

A full histologic description of a liver biopsy specimen of chronic hepatitis involves the assessment of each of the independent morphologic features, including the degree of portal inflammation and piecemeal necrosis, the severity of associated lobular activity, and the extent and distribution of fibrosis. Chronic active hepatitis, because of its wide range of appearance, has often been separated into mild, moderate, and severe grades, based usually on the degree of piecemeal necrosis and the presence of bridging necrosis.^112,119,142 Knodell et al. have developed a numerical scoring system that semiquantitatively evaluates the major histologic features, but its utility in routine biopsy interpretation needs to be established.^130,135

Chronic persistent hepatitis designates a basic pattern of uncomplicated portal inflammation. The inflammatory infiltrate is largely confined to the portal tract with little extension into the lobules (Figure 3-6). Occasional lymphocytes may spill into the adjacent parenchyma, but the periportal hepatocytes do not suffer destruction and piecemeal necrosis is therefore absent.

The degree of lobular activity is usually minor with scattered focal necroses and inflammatory infiltration. A mild but diffuse swelling of hepatocytes in some cases can obscure the hepatic plates and impart a cobblestone appearance to the lobule.¹⁴¹ Bridging or multilobular necrosis are exceptional.

The hepatic architecture is intact, and fibrosis other than mild portal expansion is not a feature. Liver biopsies showing a nonaggressive, portal-based infiltrate of mononuclear cells in a background of periportal or bridging fibrosis should not be diagnosed as chronic persistent hepatitis without qualification. Such an appearance generally represents regression of a more substantial form of liver injury, such as chronic active hepatitis, and clinical correlation is therefore necessary. The term chronic septal hepatitis has sometimes been applied to these cases, but it has not gained great currency.¹²⁵

An additional problem is distinguishing chronic persistent hepatitis from mild examples of chronic active hepatitis. Drawing a fine distinction between the two is somewhat of an arbitrary exercise, particularly since the original definition of chronic persistent hepatitis permitted a "slight" degree of piecemeal necrosis.³ When surveyed at medium power, the outlines of the portal tracts in chronic persistent hepatitis are generally smooth and unbroken, whereas in chronic active hepatitis, they are irregular and spiculated with interruptions by inflammatory and fibroblastic cells. Biopsies that show only small or inconspicuous interruptions and lack periportal fibrotic spurs can be considered examples of chronic persistent hepatitis. Examination of multiple levels is indicated, however, to insure that more striking portal tract changes are not present deeper in the paraffin block. Nevertheless, there are indeterminate cases where the changes are equivocal, and a diagnosis of borderline (or minimal) chronic active hepatitis may be appropriate.¹¹²

The hallmark of chronic active hepatitis is piecemeal necrosis with progressive destruction of the periportal parenchyma. However, depending on the variety and extent of additional histologic features, the pattern comprises a broad range of morphologic appearances that vary from a mild lesion, differing only slightly from chronic persistent hepatitis, to a severe process marked by substantial hepatocellular destruction and cirrhosis (Figure 3-7).

Both piecemeal necrosis and its associated portal inflammation vary considerably in degree and are often focally distributed. Within the periportal zone, surviving hepatocytes can display a variety of alterations, especially with more active piecemeal necrosis; these may include ballooning degeneration, oncocytic cytoplasm (Figure 3-8), multinucleation (Figure 3-9), or regenerative changes as evinced by a gland-like arrangement (hepatitic rosettes) or thickened hepatic plates.^{132,137,140,146,153}

Lobular inflammation and injury are commonly present in chronic active hepatitis and become more conspicuous during acute exacerbations of disease activity. At its most striking, lobular activity takes the form of confluent necrosis with destruction and loss of groups of contiguous hepatocytes. Confluent necrosis is recognized as zones of loosely gathered reticulin fibers devoid of hepatocytes and containing various inflammation cells. These zones may link adjacent portal tracts and central veins -- so-called bridging necrosis (Figure 3-10) -- or, in extreme cases, may affect entire lobules to yield multilobular necrosis.^108,114 Bridging necrosis joining portal tracts and central veins (portal-central bridging) has been considered to portend a worse prognosis than simple portal-portal bridging, but the nature of the bridge may be difficult to establish in biopsy specimens.^1,111

With confluent necrosis, the reticulin framework collapses and the hepatic architecture is disrupted. Piecemeal necrosis can spread from affected portal tracts to involve hepatocytes bordering the necrotic bridges. These alterations all lead to fibrosis, serving to transform bridging necrosis into fibrous septa (fibrotic bridges) that dissect the lobule and set the stage for the development of cirrhosis. Severe lobular necrosis is therefore an important stage in the progression of chronic active hepatitis.^143,147

As an indicator of prior hepatocellular necrosis, fibrosis can complicate both piecemeal and confluent necrosis and is therefore an almost invariable component of chronic active hepatitis. The degree varies from minor, patchy periportal fibrosis to fibrous septa of differing extent and cirrhosis (Figure 3-11). Since fibrosis is often unevenly distributed across the liver, its extent may be underestimated in biopsy specimens, which commonly fragment at sites of scarring. The histologic activity tends to fade as the disease becomes quiescent, and identifying residual fibrosis can then be helpful in recognizing the prior extent of active disease. However, even advanced fibrosis can be accompanied by ongoing piecemeal and lobular necrosis: When cirrhosis supervenes, the picture can be described either as active cirrhosis or as cirrhosis with chronic active hepatitis. Biopsy specimens often fragment at sites of fibrosis, so the extent may be underestimated.

As the name suggests, this pattern is distinguished by a predominance of lobular inflammation and injury (Figure 3-12).^4,5 The appearance is therefore one of acute hepatitis occurring over a chronic time span, and biologically it is perhaps best regarded as an unresolved or persistent viral hepatitis.^141,156 Chronic lobular hepatitis is an uncommon pattern that has been most often recognized in chronic hepatitis B carriers from the Far East.^133,156

The severity of the lobular activity varies, and, at the lesser extreme, only minimal changes indistinguishable from nonspecific reactive hepatitis may be found.^45,134 Similarly, the accompanying portal inflammation may range from a minimal infiltrate to a dense and more obvious accumulation; the distinction of the latter pattern from chronic persistent hepatitis becomes a debatable issue. The liver architecture is preserved, and fibrosis not present.

As a general rule, the various causes of chronic hepatitis cannot be conclusively distinguished on histologic grounds.

The one exception is chronic hepatitis B, which can sometimes be identified by the presence of "ground-glass" hepatocytes.¹²⁷ The cytoplasm of these cells has a uniform, finely granular, and faintly eosinophilic appearance that is likened to frosted glass (Figure 3-13). This change derives from a marked proliferation of endoplasmic reticulum containing accumulated particles of HBsAg. All or part of the cytoplasm may be affected, and an unstained halo typically separates the involved from the uninvolved portion. Ground-glass cells are scattered throughout the lobule in variable numbers, but there is a rough inverse correlation with the activity of the disease: Many ground-glass cells are seen in "healthy" viral carriers with minimal hepatocellular injury, but they are sparse or absent in severe chronic active hepatitis. Note that they are not seen in uncomplicated acute hepatitis B.

Similar ground-glass cells are seen in other settings, including Lafora's disease (myoclonic epilepsy), type IV glycogenosis, therapy with cyanamide, and exposure to drugs that induce proliferation of the endoplasmic reticulum, such as phenobarbital or phenytoin (Figure 3-14).^115,129 Other cases of unknown significance reflect the intracellular accumulation of fibrinogen or other proteins.^116,138 The hepatitis B ground-glass cells are distinguished by their positive staining with orcein, Victoria blue, or aldehyde fuschin stains, or by specific immunostaining for HBsAg (Figure 3-15).^149,150

In addition, hepatocyte nuclei in chronic hepatitis B may develop a finely granular "sanded" look that reflects the accumulation of HBcAg.¹¹³ This feature can be difficult to identify with certainty, and the presence of HBcAg is better secured by immunohistochemistry.

Chronic hepatitis C can demonstrate the same inconsistent and nonspecific histologic changes seen in acute disease. Although not definitive, these features are frequently suggestive and include fatty change, prominent lobular activity that is often characterized by a sinusoidal arrangement of inflammatory cells, dense portal inflammation with lymphoid aggregates or germinal centers, and bile duct damage (Figure 3-16).^{107a,121,131,147a,154} The conspicuous lobular activity often results in difficult in separating acute from chronic hepatitis C on morphologic appearances alone.

Autoimmune chronic hepatitis is often characterized by severe hepatocellular injury and broad areas of reticulin collapse (Figure 3-17). Plasma cells may be abundant, as emphasized in early reports, but this is neither an absolute nor required feature.^{18,44,122,147a}

Wilson's disease is suggested by such features as copper accumulation (as detected with copper stains), Mallory bodies, fatty change, and glycogenated nuclei (Chapter 9). These are not specific and, moreover, are not invariably present; appropriate suspicion when chronic hepatitis is encountered in a patient less than 30 years is the key to considering the diagnosis.^72,152

The histologic marker for alpha-1-antitrypsin deficiency is the PAS-positive, diastase resistant cytoplasmic globule (Chapter 9). Typically located in periportal (or periseptal) hepatocytes, these globules can be specifically stained by immunohistochemistry.^139,145

Immunohistochemistry is a sensitive and specific method for detecting hepatitis B antigens in liver tissue, but it seldom plays an important diagnostic role when serologic tests are available.¹²⁴ HBsAg is generally identified within the cytoplasm of the hepatocyte -- particularly in ground-glass cells -- or along the cell membrane; HBcAg is typically located in the nucleus but cytoplasmic and membranous staining are also seen. Several patterns of viral antigen expression have been described and correlations with the severity and course of disease have been proposed, but the results have been variable.^123,126,128 In particular, the presence of HBcAg signifies active viral replication, and its cytoplasmic expression appears to be related to more active disease (Figure 3-18).^117,157

Hepatitis C viral antigens have been demonstrated in liver tissue by immunoperoxidase techniques, but the procedure is not widely available and awaits further confirmation of its utility.

The hepatitis D viral antigen is detected by immunohistochemistry in both acute and chronic hepatitis D. The antigen is expressed primarily in nuclei and occasionally in cytoplasm, and its presence is noted in almost all patients with chronic hepatitis D. Positive staining, which correlates with active viral replication, is therefore frequently useful in confirming the diagnosis.^35,68,124

The course of a patient with chronic hepatitis depends on many factors, but the issue is clouded by differences among various studies in patient selection, etiology, histologic criteria, and length and rigor of follow-up. Firm conclusions are therefore difficult to reach. Histologic features are likely important indicators of outcome, but these must be tempered by the clinical circumstances, the exact cause, and, in viral-associated cases, the presence of continuing viral replication.

The histologic patterns continue to be useful overall guides to prognosis, despite their weaknesses as nosologic categories.^91,105 Chronic persistent hepatitis and chronic lobular hepatitis generally tend to be benign conditions, typically with stationary courses.^{80,103,133,156} On the other hand, chronic active hepatitis can be a serious and progressive disorder leading to cirrhosis and death; in severe cases, the five-year survival may be as low as 40% to 50%.^15,108

Nevertheless, the histologic pattern only denotes the disease severity at a particular time, and the general prognostic precepts therefore have exceptions that limit their unconditional applicability. Chronic persistent hepatitis B, for example, may progress if viral replication is sustained.^78,85,98 Immunosuppressive therapy of chronic active hepatitis can reduce the periportal inflammation and piecemeal necrosis, thereby converting it into chronic persistent hepatitis, but this does not necessarily imply a favorable long-term outcome.^96,118

Moreover, chronic active hepatitis is associated with a range of outcomes that, in part, depend the severity of additional histologic changes. Most notably, the presence of bridging (or multilobular) necrosis indicates a group with a particularly poor prognosis, and this feature should be specifically evaluated in biopsy specimens.^14,15 Chronic active hepatitis that displays only piecemeal necrosis and lacks bridging necrosis appears to have a better prognosis, although the exact outcome is incompletely defined. At the minimal end of the spectrum, the disease may be nonprogressive and the course indistinguishable from that of chronic persistent hepatitis.^82,83,91,100

The degree of fibrosis, as a measure of the severity of previous hepatic injury, can be used to stage chronic hepatitis. Although no formal staging schemes have been proposed, fibrosis can be assessed in the same fashion as other types of chronic liver disease, and, based on the low-power architectural appearance, three stages can be distinguished: portal and periportal fibrosis, bridging fibrosis, and cirrhosis. No study has examined the prognostic implications of fibrosis independently of other features, and, in most investigations, bridging fibrosis has not been clearly distinguished from bridging necrosis. Nevertheless, more advanced fibrosis and, in particular, cirrhosis portend greater mortality.^41,91,105 A practical difficulty in evaluating the stage is the uneven distribution of fibrosis in chronic hepatitis; although therefore subject to sampling error, a rough estimation is generally possible.

DIFFERENTIAL DIAGNOSIS

Although the diagnosis of chronic hepatitis can often be suggested by the morphologic picture, none of the various histologic features is entirely specific, and a final diagnosis therefore generally requires clinical correlation.

The mild portal and lobular inflammation seen in the lesser forms of chronic hepatitis can be noted in such disorders as resolving acute hepatitis and nonspecific reactive hepatitis. The histologic findings overlap greatly, and the distinction depends largely on the duration of the disease and the presence of appropriate laboratory abnormalities. Unless this ancillary evidence indicates that a low-grade but nonetheless chronic inflammatory disease of the liver is present, a diagnosis of chronic hepatitis should not be entertained.

The lobular activity of chronic hepatitis represents the primary feature of acute hepatitis, and distinguishing the two can be difficult. In chronic hepatitis, the lobular inflammation is typically distributed in a more focal manner than in acute disease, but this is often difficult to appreciate. In addition, bridging necrosis can be seen in either acute or chronic hepatitis, further complicating the distinction. The presence of underlying fibrosis is the best clue to chronicity, but attention to the clinical setting, serologic data, and time course of the disease are often required. In equivocal cases, a six-month duration may be the only indicator of chronic disease.

Portal inflammation -- with or without accompanying periportal inflammation or piecemeal necrosis -- occurs in wide range of conditions. The principal considerations are primary biliary cirrhosis and primary sclerosing cholangitis, both of which share characteristics with chronic hepatitis. Their main distinguishing features are chronic cholestasis and loss of the interlobular bile ducts (Chapter 5). Chronic cholestasis, although accompanied by periportal inflammation and hepatocyte injury, is characterized by a heterogeneous inflammatory population that includes neutrophils and macrophages, the occasional presence of Mallory bodies, and the accumulation of copper. Stains for copper and copper-associated protein are therefore useful: Typically positive in either of the biliary conditions, they are negative in precirrhotic chronic hepatitis. Bile duct loss is identified by finding hepatic arteries that are missing their accompanying bile duct; when more than half of the arteries lack ducts, bile duct loss is clearly present. In addition, granulomas may be seen in primary biliary cirrhosis but not the usual case of chronic hepatitis. Before chronic hepatitis is diagnosed in a woman over 40 years, serious consideration should be given to primary biliary cirrhosis. Although patients with ulcerative colitis can certainly develop chronic hepatitis, the possibility of primary sclerosing cholangitis should be entertained when a histologic pattern suggesting chronic hepatitis is encountered; the diagnosis of primary sclerosing cholangitis can be established by cholangiography.

Chronic biliary obstruction, especially when intermittent or low-grade, can produce a picture similar to that of chronic hepatitis. Chronic cholestasis may be present, but the clinical and radiographic data generally permit the distinction to be made.

Malignant lymphomas, primarily the low-grade lymphocytic varieties, mimic chronic hepatitis by filling the portal tracts and sometimes spilling out into the adjacent parenchyma. The malignant infiltrates tend to be cytologically atypical and more dense and monomorphous in character than inflammatory infiltrates; in questionable cases, lymphocyte markers may be helpful in establishing the nature of the cells.