Persistent viral hepatitis

COMBINED CLINICAL AND BASIC SCIENCE SEMINAR Selected and edited by Richard T. Silver, M.D. and Alexander Department of Medicine, The New York Hospital-Cornell Medical G. Bearn. M.D. Center, New York, New York Persistent Viral Hepatitis Moderator NORMAN B. JAVITT, M.D., Lecturers: ROGER HAND, M.D.* NIALL D. C. FINLAYSON, Ph.D.? Ph.D. M.B., CH.B., New York, New York From the Division of Gastroenterology, Department of Medicine, The New York Hospital-Cornell Medical Center, New York, New York. Requests for reprints should be addressed to Dr. Norman B. Javitt, Division of Gastroenterology, Department of Medicine, New York Hospital-Cornell Medical Center, 525 East 68th Street, New York, New York 10021. * Present address: Department of Medicine and Microbiology,. McGill University, Montreal, 112 P.Q., Canada. j’ Present address: Department of Therapeutics, Edinburgh University, Edinburgh, Scotland. Dr. Norman Javitt: Our topic for today dramatizes the progress that occurs when a causative agent of a disease is identified. Only a few years ago a relationship was recognized between the Australia antigen and serum hepatitis, for this antigen was associated with electron microscopic particles in the serum, some of which resembled viruses and caused transfusion hepatitis. It also became apparent that the virus could be transmitted by other than the parenteral route; thus, there is probably no essential difference in the possible modes of transmission of infectious and serum hepatitis. With the disappearance of this classic distinction, it became more appropriate to refer to the diseases as hepatitis A and hepatitis B virus disease, respectively. Hopefully, even this nomenclature is only a way-station on the path to more definitive terminology as our knowledge of these diseases expands. Although we still have no specific way of establishing hepatitis A virus disease, the capacity to detect the B virus has led to a most fascinating series of events, some of which have already had a major impact on human welfare. From a practical point of view, one of the major causes of transfusion hepatitis has been reduced. Development of even more sensitive methods for detection of the virus in donor blood may lead to a further reduction in this mode of hepatitis transmission. Although the virus has not been cultured in vitro, there is the possibility of prevention by immunization. More challenging is the guilt by association that is developing between the presence of hepatitis B virus and the presence of active chronic hepatitis, cirrhosis and hepatoma. Moreover, there is the association with polyarteritis nodosa lupus erythematosus and, occasionally, nephritis. Because the virus seems to affect man in a number of diverse ways, I have invited a virologist, Dr. Roger Hand, to review with us the various known models of persistent virus disease in man. Perhaps the hepatitis B virus interacts with man in a manner analagous to these models. Alternatively, a study of this interaction may lead to an entirely new mechanism that will be applicable to other illnesses. Since one of the end results of this interaction appears to be cirrhosis, a major cause of death, it behooves us to learn quickly about the mechanisms underlying chronic progressive liver disease. For the past sev- December 1973 The American Journal of Medicine Volume 55 799 COMBINED CLINICAL AND BASIC SCIENCE SEMINAR era1 years, Dr. Niall Finlayson has applied technics available. to both hurhoral and cellular immunologists to a study of chronic liver disease in man. He has ‘amassed valid statistical information that helps us’ to define the type of immunologic response. This information is a necessary prelude to the more specific type of immunologic characterization that will be forthcoming. This status report today represents the most recent thought in our continually evolving concepts. patients with either disease give no history of attack [2]. The relationship between chronic hepatitis and cryptogenic cirrhosis has not been defined. A significant portion of patients with either chronic persistent hepatitis [3] or chronic active hepatitis [4-61 have detectable serum levels of hepatitis B (HB) antigen. It is patients with HB antigenemia that I will discuss today. A segment of the patient population has no evidence of HB antigen. However, these patients do show a number of immunologic abnormalities such as antinuclear antibodies and antibodies against smooth muscle [6]. This group of patients will be discussed by Dr. Finlayson. First, I would like to review some of the data relating to the physical properties of the HB antigen. Electron microscopy of serum from patients with HB antigen discloses three different types of particles (Figure 1, [7-lo]). The first is a small round particle of 20 nm diameter, the second is a tubular form 20 nm in diameter and 30 to 700 nm in length. The third is the Dane particle which is 42 nm in diameter. The 20 nm particle is most common in serum from patients with HB antigen. The antigen is on the surface of the particle since these particles are clumped by appropriate antiserums [7]. It is predominantly lipid and protein, and contains little if any nucleic acid [lo-131. There are two main polypeptides in the particle, molecular weight zyxwvuts Dr. Roger Hand: Chronic hepatitis is a term that covers two largely distinct groups of patients, those with chronic persistent hepatitis, and those with chronic aggressive hepatitis [l1. The patients with chronic persistent hepatitis have relatively few symptoms, no abnormalities on physical examination and a mild derangement of liver function tests. Liver biopsy will show no inflammation of the liver parenchyma, mild inflammation of the portal triad and an intact portal limiting plate. The portal limiting plate is the sheet of hepatic parenchymal cells, one cell thick, surrounding and separating the portal canal from the liver parenchyma. Those patients with chronic aggressive hepatitis have persistent malaise, anorexia, hepatosplenomegaly and fairly marked derangements of liver function tests. Liver biopsy in this group reveals inflammation of the liver parenchyma as well as the portal triad, and the portal plate appears disrupted. Although both diseases appear after an acute attack of hepatitis B, a number of I micrograph of Figure 7. Electron p&tic/es purified from serum of a patient with HB antigenemia. The 20 , nm particles are most prominent at the upper left and lower right. Dane i particles are seen at upper center and lower center. Several tubular forms are interspersed throughout the collections of 20 nm particles. 1 Original magnification X 760,000, reduced by 4 per ceht. Photo by Dr. P. T. Jokelainen. j 900 zyxwvutsrqpon De c e m b e r 1973 The Am e ric a n Jo urna l o f Me dic ine Vo lum e 55 PERSISTENT 26,000 and 32,000, and a third protein is present in trace amounts, molecular weight 40,000 [12]. The molecular weight of the whole particle is 3,600,OOO [14] or one-half that of polio virus, a picornavirus. The picornaviruses are a group of small RNA viruses, 27 to 30 nm in size. The tubular forms have antigenic properties [15] and staining characteristics [lo] similar to the small 20 nm particles. This suggests that they are either precursors or aggregates of the 20 nm particles [ 161. The Dane particle has been seen infrequently in serum from patients with acute hepatitis B but is more commonly seen in atypical presentations of hepatitis or chronic hepatitis. It consists of an envelope containing the hepatitis B antigen [a], and an inner core whose staining characteristics suggest it might contain nucleic acid [I 01. Which of these particles is the virus? Serum from patients with HB antigen contains the virus since the serum transmits the disease. The rod forms seem to be precursors or aggregates of the 20 nm particles. This leaves two candidates for the virus: the 20 nm particle and the Dane particle. There are cogent arguments against the 20 nm particle being virus. True, it resembles a virus superficially. Its diameter is only slightly less than that of a picornavirus such as polio, and the surface of the particle shows a substructure units similar to the capsomeres, the morphologic that make up the protein coat or capsid of a VIRAL HEPATITIS nonenveloped virus. However, the lack of nucleic acid is disturbing, and the physical properties of the particle are not characteristic of a nonenveloped virus. The Dane particle resembles an enveloped virus, such as a myxovirus or an arbovirus. Its core may contain some nucleic acid and its physical properties are those of an enveloped virus zyxwvutsr [141. In many virus infections, in addition to complete virus, cells produce empty capsids which have been defined as virus-like structures containing the protein coat of the virus without the nucleic acid of the core. Figure 2 is an electron micrograph of reovirus, a nonenveloped virus, and reovirus empty capsids. Under certain conditions of growth, more empty capsids are produced than complete virus [17]. Perhaps this is the situation with hepatitis B virus. The 20 nm particle contains at least two proteins of molecular weight 26,000 and 32,000. In order to code for these two proteins, the virus would have to contain nucleic acid of at least molecular weight 600,000, if the putative genome is single stranded and does nothing other than code these two proteins. Six hundred thousand is about 15 per cent of the molecular weight of the 20 nm particle. A particle the size and weight of the 20 nm particle containing 15 per cent nucleic acid would have many of the physical properties of a picornavirus [16]. Detergent treatment of the 20 nm particles yields material that has HB antigen Figure 2. Electron micrograph of reovirus (lower right) and reovirus empty capsids (upper left). Original magnification X 270,000, reduced by 4 per cent. Photo by Dr. Roger Hand. December 1973 The American Journal of Medicine Volume 55 801 COMBINED CLINICAL AND BASIC SCIENCE SEMINAR Figure 3. Left, electron micrograph ot Sindbiq virus. Photo by Dr. Richard Compans. Right, electron micrograph of 42 nm Dane particles. Photo by P. T. Jokelainen with the permission of Journal of Virology (6:685, 1970). American Societv for Microbioloav. Original magnifidation X 620,0&, reduced by 4 per cent. activity, contains nucleic acid and has the physical properties of a picornavirus [li]. Thus the untreated 20 rim particles may represent empty capsids, and the small number of complete virus particles of that size are revealed only after detergent treatment [16]. The alternate possibility is that the Dane particle represents the hepatitis B virus [8]. The Dane particle resembles an arbovirus. Its structure is similar to Sindbis virus, a group A arbovirus (Figure 3). The physical properties of the Dane particle are consistent with it being a small enveloped virus. Its core may be distinguished from the 20 nm particle, and it is possible that the antigen activity detected after detergent treatment represents the cores of Dane particles with some associated envelope material. Regardless of whether the Dane particle or the 20 nm particle is the virus, the antigen is associated with the virus. Recent studies have suggested there may be more than one strain of virus. Multiple specificities have been described for the antigen [16]. All serums positive for hepatitis B antigen share a common antigen determinant a, and have another antigenic determinant of either type d or type y [18]. Two recent studies have shown a preponderance of antigenic subtype a+d+yin serums of patients with chronic active hepatitis [19,20]. However, it is not certain whether the increase is significantly above the incidence of the same subtype in other forms of antigen-positive viral hepatitis. 802 December 1973 The American Journal of Medicine The next question is how does the hepatitis B virus damage the cell to produce the pathologic condition that we see in chronic hepatitis? In infections with nonenveloped viruses such as the picornaviruses [21], cellular macromolecular synthesis is shut off shortly after infection by some early product of the infecting viral genome, and the synthesis of new viral RNA and protein begins shortly thereafter. Some viral products are produced that alter cell membranes resulting in the leakage of cell contents, a loss of normal cell morphology and eventually cell lysis and release of new virus particles. In many virus infections excess amounts of viral proteins are made, and these are not incorporated into virus particles but are liberated along with the virus particles when the cell lyses. In infection with enveloped viruses, such as paramyxoviruses, the situation is more complex [22]. The paramyxoviruses are a group of enveloped, large RNA viruses 125 to 250 nm in diameter including mumps and measles viruses. New viral proteins and nucleic acid are synthesized and eventually collections of nucleocapsid are found in the cytoplasm. These nucleocapsids then associate with specific areas of the cell membrane, and the molecular constitution of these areas is altered by the incorporation of virus-specific proteins. The virus particles then bud from the cell surface, the altered areas of cell membrane becoming their envelopes. During the course of infection, destruction of the cell may re- Volume 55 PERSISTENT sult from either the production of toxic substances or alterations of the membrane leading to fusion with adjacent cells and their eventual death. Thus, viruses may damage cells directly, but they may also alter host membranes. There also are certain virus-cell systems in which the envelope viruses replicate, alter the host membranes to suit their own envelopes but, in the course of infection, the cells are not killed. The one obvious way that the hepatitis virus may damage the liver cell is by causing sufficient functional disturbances to kill it. Many acute viral infections such as those of the upper respiratory tract kill cells in this fashion. Cell destruction probably plays a predominant role in two chronic viral infections. Subacute sclerosing panencephalitis [23] is a slowly progressive neurologic disease of man with an extremely long incubation period. The causative agent is measles virus, an enveloped virus classified as a paramyxovirus. Progressive multifocal leukoencephalopathy [24] is another neurologic illness with a protracted course occurring in immunosuppressed subjects. The causative agent is probably the nonenveloped, small DNA virus, simian virus 40, or a virus closely related to it. Alternatively, hepatitis B virus might produce chronic hepatitis through some sort of pathogenic immune mechanism. A virus infection might initiate a pathogenic immune response in several ways [25,26]. (1) Cellular material released as a result of virus infection may be immunogenic. (2) Viruses which form by budding at the cell surface may incorporate cellular components which may then become immunogenic, perhaps because of their association with viral components. (3) The cell membrane may be altered by incorporation of viral components and might, therefore, be recognized as foreign, and (4) the continued release of virus particles from the cell surface by cells not damaged by virus infection might lead to a cellmediated immune reaction or antigen antibody complexes at the cell surface. The best way to approach immunopathology is via the Coombs and Gel1 [27] classification. I will go through this classification to relate various aspects of chronic hepatitis to certain immunopathogenie mechanisms as they occur in defined animal virus infections. Type I is the anaphylactic type. There does not seem to be anything to relate this type of immune pathogenic mechanism to chronic active hepatitis and I include it only for completeness. It has been suggested that a type I reaction is involved in the pathogenesis of the bronchiolitis in infection with respiratory syncytial virus 1281. Type II is the De c e m b e r VIRAL HEPATITIS cytotoxic antimembrane type. In infection of mice with lymphocytic choriomeningitis virus (an enveloped virus), a focal hepatitis with varying degrees of cellular infiltration occurs [29]. Lymphocytic choriomeningitis antigens or neoantigens on the cell surface interact with complement and antibody to produce cell damage [30]. A similar effect can be produced by rabies virus in tissue culture systems [31]. HB antigen has been demonstrated in liver cells in patients with hepatitis [32-341, but it has never been demonstrated in liver cell membranes. The pathology of chronic hepatitis is different from that expected in the type II there being few polymorphonuclear reaction, leukocytes in the inflammatory infiltrate of chronic hepatitis. Type II I reactions are immune complex-serum sickness type. There are several chronic virus infections that are associated with this type of immune pathogenic mechanism. Aleutian disease of mink is a disease caused by a virus the size of a picornavirus. Antigen-antibody complexes are associated with the glomerulonephritis that results in the death of these animals 1351. In infections of mice with lymphocytic choriomeningitis virus the glomerulonephritis has also been shown to be caused by antigen-antibody complexes [29]. Is this mechanism operative in chronic hepatitis? A serum sickness-like illness with skin rash and arthritis has been described during the early stages of acute hepatitis [36]. HB antigen has been associated with polyarteritis nodosa [37]. Circulating antigen-antibody complexes have been demonstrated in patients with chronic active hepatitis [15]. However the amount of circulating complexes correlates poorly with the degree of liver damage [38,39]. The damage associated with this type of immune reaction usually calls forth a polymorphonuclear leukocyte response, yet chronic active hepatitis is associated with mononuclear cell response and chronic persistent hepatitis with almost no cell response. Type IV reactions are cell-mediated or delayed reactions. An example of chronic virus disease of animals mediated by this immune pathogenic mechanism is progressive pneumonia of sheep or maedi [40,41]. Proliferation of the virus in the lungs of infected sheep results in a gradual accumulation of lymphocytes in the interalveolar septurns. Nodular accumulations of mononuclear cells also appear in relation to the small air passages and the blood vessels. The virus structurally resembles murine leukemia virus, an enveloped oncornavirus, or one of a group of pleomorphic enveloped viruses which may induce tumors in susceptible animal hosts. The progressive pneu- zyxwvutsr 1973 The Am e ric a n Jo urna l o f Me dic ine Vo lum e 55 903 COMBINED TABLE I CLINICAL Features AND BASIC SCIENCE of Acute Infectious SEMINAR and Serum Hepatitis zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA lmmunitv zyxwvutsrqponmlkjihgfedcbaZ Incubation (days) Hepatitis Infectious (type A) Serum (type B) MS Infection Route Range Mean I 31-42 37 +++ ++ II 55-90 71 + ++ NOTE: SGOT = serum glutamic oxaloacetic Oral transaminase. HB = hepatitis monia agent may be the same virus that causes visna, a neurologic disease of sheep [41]. How might this pathogenic immune mechanism be associated with chronic hepatitis? Chronic hepatitis is associated primarily with a lymphocytic infiltration, and lymphocytes are mediators of this type of response. The outcome of infection with hepatitis B virus in patients may be partially determined by the vigor of their cell-mediated response to the infection [42]. Those with a vigorus cell-mediated response have acute hepatitis B. A few patients are overwhelmed by the infection, but the majority go on to recover. In some cases _there is little or no cell-mediated response due to either antigen overload or immunosuppression (as in patients with chronic renal disease undergoing dialysis). Here the virus will continue to proliferate, causing little or no damage, and circulating hepatitis B antigen will persist. In intermediate sita partial cell-mediated response will uations, ensue sufficient to cause some liver damage but insufficient to control the growth of the virus. Continuous cycles of limited growth and host response would result in chronic active hepatitis. There is evidence to support this hypothesis [43]. Possibly a combination of several of these immune pathogenic mechanisms are operative in chronic hepatitis. Certainly this occurs in the animal models described. Lymphocytic choriomeningitis virus produces both the type II and the type I I I reactions in the same animal [29]. Progressive pneumonia virus of sheep causes a type IV reaction and probably some cell damage as well [40]. Some aspects of acute hepatitis seem to involve cell destruction and the type III reaction [36,37]. A type II reaction might also be involved [32,34]. Chronic hepatitis seems to be best explained by type IV reaction along with some cell destruction. The hypothesis that the outcome of infection with hepatitis B virus is determined by the quality of the host’s cell-mediated response [42,43] is interesting and deserves further study. Dr. Niall Finlayson: Classically, acute viral hepatitis has been divided into two major types. (1) In- 804 December 1973 The American Journal of Medicine Parenteral Duration of SGOT Rise Homologous Cross HB Antigen Short Yes No - Long Yes No + B. fectious hepatitis (type A), which may occur either in epidemics or in the form of sporadic cases, and (2) serum hepatitis (type B), which results from the inoculation of the patient with infected blood or blood products. That these two major forms of acute viral hepafitis result from separate and distinct infectious agents has been best demonstrated by Krugman et al. [44]. These workers, whose MS I infection corresponds to infectious hepatitis and whose MS II infection corresponds to serum hepatitis, were able to delineate the main features of each infection and to show that there was no cross immunity between the two. These features are illustrated in Table I. Both the Australia antigen of Blumberg et al. [45] and the SH antigen of Prince [46] were shown to be associated with serum hepatitis. These two antigens were later found to be identical and Giles et al. [47] demonstrated that the antigen occurred only in Krugman’s MS II type of acute hepatitis. Electron microscopic studies [7,8,10] have disclosed that the Au-SH antigen is associated with virus-like particles which have been described by Dr. Hand. One or more of these may represent the infectious agent of serum hepatitis. In view of this association of the antigen with type B hepatitis, it has been suggested that the Au-Sh antigen should be called the hepatitis B (HB) antigen. In clinical practice, however, it has been found to be associated not only with serum hepatitis but also with sporadic cases of acute viral hepatitis not associated with parenteral inoculation. In general, the HB antigen has not been associated with the epidemic forms of acute viral hepatitis [48,49]. The virus can exist both in the blood and in the liver without clinical, biochemical or biopsy evidence of liver disease. It has been suggested that the acute hepatitis syndrome may depend more on the host’s own immune response than on the virus itself [42]. Certain symptoms associated with acute serum hepatitis may result from the formation of immune complexes such as arthralgia, arthritis and urticaria. Reduced levels of total Volume 55 PERSISTENT hemolytic complement and the third and fourth components of the complement system are associated with serum hepatitis at the time when this serum sickness-like syndrome is present [50]. Although these reduced levels may have been due to the presence of circulating immune complexes, it is important to note that certain of the components of the complement system, including the third and fourth, may be partly or wholly synthesized in the human liver [51,52]. Consequently, low levels of these components and of the total hemolytic complement activity, could result from liver damage. However, the Cl, subcomponent of complement, which is probably not made in the liver [54], is not particularly reduced in patients with the serum sickness-like syndrome [5O]. The HB antigen generally disappears from the blood within 12 weeks of the onset of clinical illness [55]. In some cases, though, the antigen may persist for years, even though the patient has made a clinical recovery from the acute illness [61]. The eventual fate of these patients in terms of chronic liver disease is not yet known [53]. Chronic Idiopathic Liver Disease. In the majority of patients with chronic liver disease the cause of the illness is unknown. These chronic idiopathic liver diseases have been divided into four main clinicopathologic syndromes: persistent hepatitis, chronic active hepatitis, cryptogenic cirrhosis and primary biliary cirrhosis. Evidence that the more serious forms of these illnesses result from acute hepatitis has been conflicting. The positive evidence consists mainly of clinical and autopsy reports in which hepatitis apparently led to the development of chronic liver disease with cirrhosis [57-601. On the other hand, follow-up studies of epidemics of infectious hepatitis [61] and serum hepatitis [62] do not suggest that clinical acute hepatitis results in a significant number of patients with serious chronic liver disease. There is evidence that suggests that subclinical cases of acute hepatitis may be more important as a source of chronic liver disease than clinical acute hepatitis [63-651. HB antigen is found in the serum of a significant proportion of patients with chronic idiopathic liver disease (Table I I; [66-691). The proportion of positive cases in each diagnostic category varies, but in general it has been found in a significant minority of persons with persistent hepatitis, chronic active hepatitis and cryptogenic cirrhosis. So far, using routine laboratory methods, the antigen has not been detected in a significant proportion of patients with primary biliary cirrhosis (Table I I). TABLE II VIRAL HEPATITIS Occurrence of the Hepatitis B Antigen in Chronic Liver Disease zyxwvutsrqponmlkjihgfedcbaZYXWV Per Cent of Cases with HB Antigenemia .___..__. Source ---.- Wright [66] Vischer (671 Krassnitsky et al. [68] Hadziyannis et al. [69] Author’s cases CPH CAH CC PBC AALD 50 - 25 22 60 40 16 27* 0 26 21 20 0 0 0 2t 0 14 47 NOTE: CPH = chronic persistent hepatitis; CAH = chronic active hepatitis; CC = cryptogenic cirrhosis; PBC = primary biliary cirrhosis; AALD = alcohol-associated liver disease. * Classified as posthepatic cirrhosis. t Previous blood transfusions. Circulating autoantibodies are commonly found in chronic idiopathic liver disease, the most important of these being antinuclear antibody, smooth muscle antibody and antimitochondrial antibody. All the autoantibodies are tissue and species nonspecific, and none is pathognomonic of any particular liver disease syndrome. For the most part, they are usually found in chronic active hepatitis, cryptogenic cirrhosis and primary biliary cirrhosis [70]. Recent studies have described the relationship of the autoantibodies to the HB antigen in chronic idiopathic liver disease [67,71-731. The results of these are summarized in Table II I. In general, autoantibodies and the HB antigen have not been found simultaneously in patients with chronic idiopathic liver disease. This separation has led to the suggestion that autoantibodies and the HB antigen are markers of liver diseases of different causes [74]. However, there are case reports in which autoantibodies and the HB antigen have occurred together [74-761. When autoantibodies and the HB antigen have been reTABLE Ill Number of Patients with Hepatitis B (HB) Antigen or Autoantibodies in Chronic Idiopathic Liver Disease Vischer 1671 .___ Bulkley et al. [71] HB Antigen 24: + ? 3 a, 2” + - 2 17 H I3 Antigen 29 37 5 a Wright (721 + - $+ - 7 16 Finlayson et al. [73j HB Antigen + 1 30 + 0 7 11 17 HB Antigen L $J:z aza + - + 3 35 19 29 NOTE: ANA = antinuclear antibody; SMA = smooth-muscle antibody; AMA = antimitochondrial antibody. December 1973 The American Journal of Medicine Volume 55 805 COMBINED CLINICAL AND BASIC SCIENCE SEMINAR TABLE IV frequent [87], 40 to 60 per cent of cases. This considerably exceeds its ocPer Cent of currence in the general local population [76,87Cases with Hepatitis B Antigenemia 891. These results are given in Table IV. The mechanism whereby the HB virus could cause Source Hepatoma Controls hepatoma is quite unknown, although the recent Hadziyannis et al. (691 31 ... report of Primack et al. [93] gives no support to Welsh et al. [86J 0 11 the notion that it might be associated with general Tong et al. [87] 80 13.6 host immunodeficiency. Vogel et al. [88] 40 3.1 Chronic Asymptomatic HB Antigenemia. Routine * With and without cirrhosis. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA testing of blood donors has revealed that from 1 to 5/1,000 persons in western countries have ported together, the autoantibody has been either chronic asymptomatic HB antigenemia [90]. The antinuclear antibody or smooth muscle antibody incidence of chronic asymptomatic HB antigene[67]. When sequential tests have been carried mia is probably higher in tropical countries, [87]. out, the autoantibody has occurred particularly These asymptomatic carriers are generally during an exacerbation of the chronic liver dishealthy and have not had clinical acute hepatitis ease [73]. in the past. Some studies report normal or mild Host factors may be important for the producnonspecific abnormalities on liver biopsy of these tion of liver damage in patients with chronic HB patients [91], whereas others report aggressive antigenemia [77]. Low serum complement levels hepatitis and cirrhosis [92,93]. occur relatively frequently in chronic liver disease Our own studies [94], have shown that asymp[78], and this has suggested to some that circutomatic, apparently healthy persons with chronic lating immune complexes may be present which HB antigenemia do not generally give a history of could be important in the development of the liver undue susceptibility to infection. In few patients lesion. In addition, it has been reported that in was it possible to obtain a history suggesting a chronic liver disease the serum complement level possible parenteral route whereby the infection tends to be low more frequently in those with was acquired. Physical examination disclosed no chronic HB antigenemia 1791. However, as alabnormalities in any subject, and tests of liver ready mentioned, the liver produces a number of function demonstrated that a raised serum transthe components of the complement system, and aminase level was the only significant abnormalievidence linking low serum complement levels to ty. Liver biopsy findings ranged from a completely poor liver function [78,823] suggests that the usual normal liver to aggressive hepatitis, with or withcause of hypocomplementemia in chronic liver out cirrhosis. In general, a raised serum transamdisease is liver damage itself, whether or not inase was usually but not invariably accompanied chronic HB antigenemia is present. Abnormal celby an abnormal liver biopsy. Immunologic investilular immune responses have been put forward as gations showed that minor .abnormalities of serum being responsible for the development of chronic immunoglobulin levels were common but incon, sistent and unrelated to liver function tests or the liver disease following infection with the hepatitis findings on liver biopsy. There was no evidence of B virus. Moreover, in conditions such as Down’s any abnormality of the cellular immune responses syndrome [81], lepromatous leprosy [82] and lymas determined by skin testing and in vitro lymphophoma [83], in which chronic HB antigenemia occyte responsiveness to phytohemagglutinin. curs relatively frequently, abnormalities in cellular immunity are frequent. Patients recovering from It is not known why persons may become HB antigen;positive acute hepatitis showed lymchronic carriers of the HB antigen. Clearly the size of the serum hepatitis virus pool in any comphocyte transformation in response to serum containing the antigen; one patient who remained munity as well as the possibility of insect vectors for the virus are important in determining the antigenemic did not [84]. Comparable results were reported using a peripheral leukocyte migrachances of a person being infected. In addition, tion method in which migration-inhibition did not there is evidence to suggest that exposure to the occur in those with antigen-associated chronic antigen in neonatal life [95] may result in a high liver disease 1851. incidence of chronic antigenemia and that exposure during childhood [47] may result in a higher Although some have failed to find a Hepatoma. significant relationship between the HB antigen incidence of chronic antigenemia than exposure in adult life. The frequency of exposure to the HB and hepatoma [86], others, especially in countries 806 Occurrence of the Hepatitis B Antigen in in which chronic HB antigenemia is Hepatoma* zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA have found the HB antigen in from December 1973 The American Journal of Medicine Volume 55 PERSISTENT VIRAL. HEPAl Ills virus may also be important, as evidenced by the and guinea pig antiserums to this stool antigen, relatively high rate of chronic antigenemia in drug showed that these antiserums clumped the stool addicts [96]. It is not known why HB virus fails to particles but not the HB antigen-associated particlear from the body of a person in whom chronic cles. Shulman et al. [98] have reported finding a antigenemia develops; if this is the result of an serum antibody in acute hepatitis reacting with an immune deficiency, it is presumably an immune antigen in the patients’ own stool. This antigen deficiency of limited type. At the present time the was not HB antigen, and it was postulated that the prognosis in terms of liver disease for persons antibody arose as a result of failure of the diswho are chronic carriers of the HB antigen, and eased liver to remove bowel-derived antigen from therefore of the HB virus, is unknown. the portal blood. Over-all, there is no good confirmed evidence that the HB antigen reaches the Nonparenteral Spread of HB Virus. Clinical and epidemiologic evidence suggests that the HB virus stool in antigenemic persons. A report indicating can be spread by means other than the parenteral that HB antigen may be present in the bile but not route [97]. Direct evidence concerning alternate in the duodenal contents has led to the suggestion that inhibitors of the antigen may have resulted in modes of infections has been sought either by failure to find it in the stool [98]. feeding various secretions or excretions to volunteers [98] or by seeking the HB antigen in these Work on saliva has given both positive and negmaterials. In general, with one exception in which ative results [99]. Recently Ward et al. [106], it was clearly shown that orally administered using the sensitive radioimmunoassay method for serum containing the HB antigen could result in detecting HB antigen, found that the antigen could HB antigen-positive acute hepatitis [44], feeding frequently be detected and that its presence was experiments with volunteers have given equivocal not always coincident with detectable blood in the results in that there was always the possibility that specimens. Our own results using radioimmupositive transmission was of the hepatitis A virus noassay support these findings [102]. only [99]. Over-all, with the possible exception of saliva, Negative results have been reported [99] in atevidence for the presence of HB antigen in the tempts to detect HB antigen in the urine. On the secretions and excretions tested is not very other hand, Blainey et al. [IOO] found positive restrong. Further studies are needed, and attempts sults in all eight subjects using a complement should be made to determine whether positive fixation method; immunodiffusion and counter findings could be due to the presence of blood in electrophoresis tests were positive in only one of the specimens; this might be expected in the case these. All eight subjects had been recipients of of stool, as some 1 to 5 ml of blood is lost daily renal transplants, however, and immunosuppresfrom the bowel even in normal persons. In addisants could have resulted in unusually high blood tion, although material containing the HB antigen antigen levels with spillage into the urine. In addiis considered to be infectious, actual transmission tion, hematuria is common in such patients, and of type B hepatitis by secretions or excretions has positive results may have been due to this, as not been shown. tests for blood in the urine were not reported. Neonatal hepatitis associated with the HB antiApostolov et al. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA [lo1 ] have also reported finding gen has been described in the offspring of antiHB antigen in the urine, but a number of the pagenemic women. Recent evidence [95] suggests tients who had antigen in their urine did not have that spread may be by more than one route: antigen in their serum. Our own studies [102] on across the placenta, during birth as a result of healthy subjects with chronic HB antigenemia and contamination of the baby with maternal blood or without hematuria have not given positive results. in the period of close maternal-child relations imGrob and Jamelka (1031 have reported finding mediately thereafter. Routes may differ in different HB antigen and antibody in the stool in all their cases. Chronic antigenemia is very common in patients with HB antigen-positive hepatitis. The neonatal infection with the virus [95]. complexity of investigation with stool is shown by Finally, insects have been considered as vecFerris et al. [104], who reported that specimens tors of transmission of the virus. Prince et al. from 90 to 220 patients with acute hepatitis gave [107] have detected the HB antigen in 28 of 187 a precipitation in agar with an antibody present in pools of mosquitoes caught in the wild in Kenya the serum hemophiliacs who had received multiand Uganda. These represented eight different ple transfusions. Although this antigen was assospecies. As yet, however, actual spread by insect ciated with the presence of 200 to 400 A diameter vectors has not been demonstrated, and it is not particles in the stool, it was not identical with the known whether the HB virus is simply carried by HB antigen. Subsequent work [105], using rabbit or also proliferates in the mosquito. December 1973 The American Journal of Medicine Volume 55 007 COMBINED CLINICAL AND BASIC SCIENCE SEMINAR SUMMARY The discovery of the HB antigen has resulted in a relatively simple way to detect the HB virus. This has allowed not only the separate nature of the two main forms of acute viral hepatitis to be confirmed, but has also allowed an assessment of the role of the HB virus in chronic liver disease. Investigations on the latter have indicated that the HB virus may be responsible for a very significant proportion of cases of serious chronic liver disease. As yet it is not known how the HB virus causes acute or chronic liver damage, although immunologic mechanisms are suggested. The HB virus can spread by other than parenteral routes, and investigations for the HB antigen indicate that spread by at least the saliva and across the placenta may be among these. Insect vectors may be important in certain areas. REFERENCES 2. 3. 4. 5. 6. 7. a. 9. 10. 11. 12. 13. 14. 15. 16. 909 deGroote J, Desmet VJ, Gedigk P, Korb G, Popper H, Poulsen H, Sgheuer PJ, Schmid M, Thaler H, Uehlinger E, Wepler W: A classification of chronic hepatitis. Lancet 2: 626, 1966. Nielsen JO, Dietrichson 0, Eking P, Christofferson P: Incidence and meaning of persistence of Australia antigen in patients with acute viral hepatitis: development of chronic hepatitis. N Engl J Med 265: 1157,1971. Sherlock S, Fox RA, Niazi SP, Scheuer PJ: Chronic liver disease and primary liver cell cancer with hepatitis associated (Australia) antigen in serum. Lancet 1: 1243, 1970. Wright R, McCollum RW, Klatskin G: Australia antigen in acute and chronic liver disease. Lancet 2: 117, 1969. Gitnick G, Gleich GJ, Schoenfield W, Baggenstoss, AH, Sutnick Al, Blumberg BS, London WT, Summerskill WHJ: Australia antigen in chronic active liver disease with cirrhosis. Lancet 2: 265, 1969. Buckley BH, Heizer WD, Goldfinger SE, lsselbacher KJ, Schulman NR: Distinctions in chronic active hepatitis based on hepatitis-associated antigen. Lancet 2: 1323,197O. Bayer ME, Blumberg BS, Werner B: Particles associated with Australia antigen in the sera of patients with leukemia, Down’s syndrome and hepatitis. Nature (Lond) 216: 1057, 1966. Dane DS, Cameron CH, Briggs M: Virus-like particles in serum of patients with Australia-antigen-associated hepatitis. Lancet 1: 695, 1970. Gust ID, Cross G, Kaldor J, Ferris AA: Virus-like particles in Australia-antigen associated-hepatitis. Lancet 1: 953. 1970. Jokelainen PT, Krohn K, Prince AM, Finlayson NDC: Electron microscopic observations on virus-like particles associated with SH antigen. J Virol 6: 665, 1970. Gerin JL, Purcell RH, Hoggan MD, Holland PV, Channeck RM: Biophysical properties of Australia antigen. J Virol 4: 763, 1969. Gerin JL, Holland PV, Purcell RH: Australia antigen: large scale purification from human serum and biochemical studies of its proteins. J Virol 7: 569, 1971. Jozwiak W, Koscielak J, Madalinski K, Brzosko WJ, Nowoslawski A, Kloczewiak M: RNA of Australia antigen. Nature (New Biol) 229: 92. 1971. Dreesman GR, Hollinger FB, Suriano JR, Fujioka RS, Brunschwig JP, Melnick JL: Biophysical and biochemical heterogeneity of purified hepatitis B antigen. J Virol 10: 469, 1972. Almeida JD, Waterson AP: Immune complexes in hepatitis. Lancet 2: 963, 1969. LeBouvier GL. McCollum RW: Australia (hepatitis as- December 1973 The American Journal of Medicine 17. ia. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30, sociated) antigen: physiochemical and immunological characteristics. Advs Virus Res 16: 357, 1970. Loh PC, Oie HK: Role of lysine in the replication of reovirus. I. Synthesis of complete and empty virions. J Virol 4: 690, 1969. LeBouvier GL: The heterogeneity of Australia antigen. J Infect Dis 123: 671, 1971. Holland PV, Purcell RH, Smith H, Alter HJ: Subtyping of hepatitis associated antigen (HB-Ag): simplified technique with counter electrophoresis. J lmmunol 109: 420,1972. Gordon I, Berberian M, Stevenson D, Redeker AG: Distribution of hepatitis B antigenic determinants in different forms of viral hepatitis. J Infect Dis 126: 569, 1972. Bablanian R: Mechanisms of virus cytopathic effects. Symp Sot Gen Microbial 22: 359, 1972. Choppin PW, Klenk HD, Compans RW, Caliquiri LA: The parainfluenza virus SV5 and its relationship to the cell membrane. Perspect Virol 7: 127, 1971. Payne FE, Baublis JV: Measles virus and subacute sclerosing panencephalitis. Perspect Virol 7: 179, 1971. (Editorial) Polyoma-like viruses and progressive multifocal leucoencephalopathy. Lancet 1: 626, 1972. Harter DH, Choppin PW: Possible mechanisms in the pathogenesis of “post-infectious” encephalomyelitis, Immunological Disorders of the Nervous System, Research Publication ARNMD 49, Baltimore, Williams & Wilkins, 1971, p 342. Compans RW, Mountcastle WE, Choppin PW: Parainfluenza viruses and their interactions with cell membranes, Multiple Sclerosis, Immunology, Virology and Ultrastructure, UCLA Forum in Medical Sciences 16. (Wolfgram F, Ellison GW, Stevens JG, Andrews JM, eds), New York, Academic Press, 1972, p 191. Coombs RRA, Gell PGH: The classification of allergic reactions underlying disease, Clinical Aspects of Immunology (Gel1 PGH, Coombs RRA, eds), Philadelphia, F. A. Davis Co., 1962, p 317. Gardner PS, McQuillen J, Court SDM: Speculation on pathogenesis in death from respiratory syncytial virus infection. Br Med J 1: 327, 1970. Oldstone MBA, Dixon FJ: Pathogenesis of chronic dislymphocytic with persistent associate ease choriomeningitis viral infection. I. Relationship of antibody production to disease in neonatally infected mice. J Exp Med 129: 463,1969. Oldstone MBA, Dixon FJ: Pathogenesis of chronic diswith persistent lymphocytic associated ease choriomeningitis viral infection. I I. Relationship of the antilymphocytic choriomeningitis immune response to tissue injury in chronic lymphocytic choriomeningitis disease. J Exp,Med 131: 1, 1970. zyxwvutsrqponm Volume 55 PERSISTENT VIRAL HEPATITIS 31. Fernandes MV, Wiktor TJ, Koprowski H: Endosymbiotic relationship between animal viruses and host cells: a study of rabies virus in tissue culture. J Exp Med 120: 1099,1964. 32. Coyne (Zavatone) VE, Millman I, Cerda J, Gerstley BJS. London T, Sutnick A, Blumberg BS: The localization of Australia antigen by immunofluorescence. J Exp Med 131: 307,197O. Nowoslawski A, Brzosko WJ, Madelinski K, Krawczyn33. ski K: Cellular localization of Australia antigen in the liver of patients with lymphoproliferative disorders. Lancet 1: 494, 1970. 34. Edington TS, Ritt DJ: Intrahepatic expression of serum hepatitis-virus-associated antigens. J Exp Med 134: 871,197l. 35. Porter D, Larsen A, Porter H: The pathogenesis of Aleutian disease of mink. 1. In vivo viral replication and the host antibody response to viral antigen. J Exp Med 130: 575.1969. 36. Alpert E, lsselbacher KJ, Schur PH: The pathogenesis of arthritis associated with viral hepatitis: complement-component studies. N Engl J Med 285: 185, 1971. 37. Gocke DJ. Hsu C, Morgan C, Bombardieri S, Lockshin M, Christian CL: Vasculitis in association with Australia antigen. J Exp Med 134 (suppl): 3305, 1971. 38. Dudley FJ, Fox RA, Sherlock S: Relationship of hepatitis-associated-antigen to acute and chronic liver injury. Lancet 2: 1, 1971. 39. Prince AM, Trepo C: Role of immune complexes involving SH Antigen in pathogenesis of chronic active hepatitis and polyarteritis nodosa. Lancet 1: 1309,197l. 40. Duran-Reynolds F, Jungherr E, Cuba-Caper0 A, Rafferty KA, Helmboldt, C: The pulmonary adenomatosis complex in sheep. Ann NY Acad Sci 70: 726, 1958. 41. Takemoto KK, Matter CFT, Stone LB, Coe JE, Lavelle G: Antigenic and morphological similarities of progressive pneumonia virus, a recently isolated “slow virus” of sheep, to visna and maedi viruses. J Virol 7: 301,197l. 42. Dudley FJ, FOX RA, Sherlock S: Cellular immunity and hepatitis-associated Australia antigen liver disease. Lancet 1: 723,1972. Guistino V. Dudley FJ, Sherlock S: Thymus dependent 43. lymphocyte function in patients with hepatitis-associated antigen. Lancet 2: 850, 1972. 44. Krugman S, Giles JP, Hammond J: Infectious hepatitis: evidence for two distinctive clinical, epidemiological. and immunological types of injection. JAMA 200: 365, 1967. 45. Blumberg BS, Alter HJ, Visnich S: A “new” antigen in leukemia sera. JAMA 191: 541, 1965. 46. Prince AM: An antigen detected in the blood during the incubation period of serum hepatitis. Proc Natl Acad Sci USA. 60: 814, 1968. 47. Giles JP, McCollum RW, Berndtson LW Jr, Krugman S: Relation of Australia/SH antigen to the Willowbrook MS-2 strain. N Engl J Med 281: 119, 1969. 48. Mosley JW, Barker LF, Shulman NR, Hatch MH: Failure to detect an antigen associated with hepatitis in a community epidemic. Nature (Lond) 225: 953, 1970. 49. Chang LW, O’Brien TF: Australia antigen serology in the Holy Cross football team hepatitis outbreak. Lancet 2: 59, 1970. 50. Alpert E, lsselbacher KJ, Schur P: The pathogenesis of arthritis associated with viral hepatitis. N Engl J Med 285: 185, 1971. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. December 1973 Thorbecke GJ, Hochwald GM, Furth R Van, MullerEberhard HJ, Jacobson EB: Problems in determining the site of synthesis of complement components, Ciba Foundation Symposium-Complement (Wostenholme GEW, Knight E. eds). London. J & A Churchill, Ltd., 1965, p 99. Alper CA, Johnson AM, Bertch AG, Moore FD: Human C’3: evidence for the liver as the primary site of synthesis. Science 163: 286, 1969. Boyer JL, Klatskin G: Pattern of necrosis in acute viral hepatitis. N Engl J Med 283: 1063, 1970. Colton HR, Gordon JM, Borsos T. Rapp HJ: Synthesis of the first component of human complement in vitro. J Exp Med 128: 595, 1968. Hirschman RJ, Shulman NR, Barker LF, Smith KO: Virus-like particles in sera of patients with infectious and serum hepatitis. JAMA 208: 1667, 1969. Nielsen JO, Dietrichson 0, Elling P, Christoffersen P: Incidence and meaning of persistence of Australia antigen in patients with acute viral hepatitis: development of chronic hepatitis. N Engl J Med 285: 1157.1972. Krarup NB, Roholm K: The development of cirrhosis of the liver after acute hepatitis, elucidated by aspiration biopsy. Acta Med Stand 108: 306. 1941. Sherlock S: Post-hepatitis cirrhosis. Lancet 1: 817, 1948. Kunkel HG, Labby DH: Chronic liver disease following infectious hepatitis. I. Cirrhosis of the liver. Ann Intern Med 32: 433, 1950. MacDonald RA, Mallory GK: The natural history of post-necrotic cirrhosis: a study of 221 autopsy cases. Am J Med 24: 334,1958. Barile AW, Taguchi JT, Maimon SN: Epidemic hepatitis: prolonged observation on the clinical course with liver biopsy studies. Gastroenterology 19: 755, 1951. Beebe GW, Simon AH: Cirrhosis of the liver following viral hepatitis: a twenty-year mortality follow-up Am J Epidemio192: 279, 1970. Klatskin G: Subacute necrosis and post-necrotic cirrhosis due to anicteric infections with the hepatitis virus. Am J Med 25: 333, 1958. Chung WK, Moon SK, Gershon RK, Prince AM, Popper H: Anicteric hepatitis in Korea. Arch Intern Med 113: 535,1964. Cooper WC, Gershon RK, Sun SC, Fresh JW: Anicteric viral hepatitis: a clinicopathological follow-up study in Taiwan. N Engl J Med 274: 585, 1966. Wright R: The Australia antigen in auto-immune liver disease, Immunology of the Liver (Williams R, Smith M. eds). London, William Heinemann Medical Books, 1971, p 28. Vischer TL: Australia antigen and autoantibodies in chronic hepatitis. Br Med J 2: 695, 1970. Krassnitsky 0, Pesendorfer F, Wewalka F: AustraliaSH-antigen und Lebererkrankungen. Dtsch Med Wochenschr 95: 249,197O. Hadziyannis SJ, Merikas GE, Afroudakis AP: Hepatitisassociated antigen in chronic liver disease. Lancet 2: 100,197o. Doniach D, Roitt IM, Walker JG, Sherlock S: Tissue autoantibodies in primary biliary cirrhosis and other liver diseases and their clinical implications. Clin Exp lmmunol 1: 237, 1966. Bulkley B, Heizer WD. Goldfinger SE, lsselbacher KJ, Schulman NR: Distinctions in chronic active hepatitis based on circulating hepatitis-associated antigen. Lancet 2: 1323,197O. Wright R: Australia antigen and smooth-muscle anti- The American Journal of Medicine Volume 55 809 COMBINED 73. 74. 75. 76. 77. 78. 79. 80. ai. 82. 83. 84. 85. 86. 87. 88. 89. 810 CLINICAL AND BASIC SCIENCE SEMINAR body in acute and chronic hepatitis. Lancet 1: 521, 1970. Finlayson NDC, Krohn K, Anderson KE, Jokelainen PT, Prince AM: Interrelations of hepatitis B antigen and autoantibodies in chronic idiopathic liver disease. Gastroenterology 63: 646, 1972. Kater L, Van Den Tweel JG, Lourens J, Borst-Eilers E, Mul NAJ: HAA and ANA in chronic active hepatitis. Lancet 1: 598,197l. Urowitz MB, Koehler BE: Lupoid hepatitis and HAA. Lancet 1: 1014, 1971. Chaudhuri AKR, McKenzie P: Australia antigen and smooth-muscle antibody in chronic active hepatitis. Lancet 1: lia4,1971. MacKay IR, Morris PJ: Association of autoimmune active chronic hepatitis with HL-AL, 6. Lancet 2: 793, 1972. Fintayson NDC, Krohn K, Fauconnet MH, Anderson KE: Significance of serum complement levels in chronic liver disease. Gastroenterology 63: 653, 1972. Grob PJ, Jemelka HJ, Muller JW: BAand SH antigen in chronic hepatitis. Gastroenterology 61: 91, 1971. Kaboth U, Arnold R: B,a globulin in liver disease. Verh Dtsch Ges Inn Med 75: 339,1969. Agarwaal SS, Blumberg BS, Gerstley BJS, London WT, Sutnick Al, Loeb LA: DNA polymerase activity as an index of lymphocyte stimulation: studies in Down’s syndrome. J Clin Invest 49: 161, 1970. Sutnick Al, London WT, Blumberg BS: Effect of host and environment on immunoglobulins in Down’s syndrome. Arch Intern Med 124: 722. 1969. Sutnick Al, London WT, Blumberg BS, Yankee RA, Gerstley BJS, Millman I: Australia antigen (a hepatitis-associated antigen) in leukemia. J Natl Cancer lnst 44: 1241, 1970. Yeung Laiwah AAC: Lymphocyte transformation by Australia antigen. Lancet 2: 470, 1971. Ito K, Nakagawa J, Okimoto Y, Nakano H: Chronic hepatitis-migration inhibition of leucocytes in the presence of Australia antigen. N Engl J Med 266: 1005.1972. Welsh JD, Brown JD, Arnold K, Chandler AM, Mau HM, Thuc TK: Hepatitis-association with antigen in hepatoma in South Vietnam. Lancet 1: 592, 1972. Tong MJ, Sun SC, Schaeffer BT, Chang NK. Lo KJ, Peters RL: Hepatitis-associated antigen and hepatocellular carcinoma in Taiwan. Ann Intern Med 75: 687.1971. Vogel CL, Anthony PP, Sadikali F, Barker LF, Peterson MR: Hepatitis-associated antigen and antibody in hepatocellular carcinoma: result of a continuing study. J Natl Cancer lnst 46: 1563, 1972. Primack A, Vogel CL, Barker LF: Immunological stud- December 1973 The American Journal of Medicine 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. Volume 55 ies in Ugandan patients with hepatocellular carcinoma. Br Med J 1: 16, 1973. Chalmers TC, Alter HJ: Management of the asymptomatic carrier of the hepatitis-associated (Australia) antigen. N Engl J Med 265: 613, 1971. Reinicke V, Dybkjaer E, Poulsen H, Banke 0, Lyloff K, Nordenfelt E: A study of Australia-antigen-positive blood donors and their recipients, with special reference to liver histology. N Engl J Med 266: 667, 1972. Prince AM, Hargrove RL, Jeffries GH: The role of serum hepatitis virus in chronic liver disease. Trans Assoc Am Physicians 62: 265,1969. Singleton JW, Merrill DA, Fitch RA, Kohler PF, Rettberg WAH: Liver disease in Australia-antigenpositive blood donors. Lancet 2: 765, 1971. Finlayson NDC, Fauconnet MH, Prince AM: In preparation. Schweitzer IL, Wing A, McPeak C, Spears RL: Hepatitis and HAA in 56 mother-infant pairs. JAMA 220: 1092,1972. Cherubin CE, Hargrove RL, Prince AM: The serum hepatitis related antigen (SH) in illicit drug users. Am J Epidemiol zyxwvutsrqponmlkjihgfedcbaZYXWVU 61: 510, 1970. Garibaldi RA, Hatch FE, Bisno AL, Hatch MH, Gregg MB: Nonparenteral serum hepatitis. JAMA 220: 963, 1972. Shulman NR, Lange RF, Knepp CS, et al.: Hepatitis and Blood Transfusion (Vyas GE, Perkins HA, Schmid R, eds), New York, Grune B Stratton, 1972, p 257. Shulman NR: Hepatitis-associated antigen. Am J Med 49: 669, 1970. Blainey JD, Earle A, Flewett TH, Williams LKL: Is the urine infective in serum hepatitis? Lancet 1: 797, 1971. Apostolov K, Bauer DJ, Selway JWT, Fox RA, Dudley FJ, Sherlock S: Australia antigen in urine. Lancet 1: 1274, 1971. Finlayson NDC. Prince AM: In preparation. Grob PJ, Jemelka H: Faecal SH (Australia) antigen in acute hepatitis. Lancet 1: 206, 1971. Ferris AA, Kaldor J, Gust ID, Cross G: Faecal antigen in viral hepatitis. Lancet 2: 243. 1970. Cross GF, Waugh M, Ferris AA: Virus-like particles associated with a faecal antigen from hepatitis patients and with Australia antigen, Aust J Exp Biol Med Sci 49: 1, 1071. Ward R, Borchert P, Wright A, Kline E: Hepatitis B antigen in saliva and mouth washings. Lancet 2: 726, 1972. Prince AM, Metselaar D, Kafuke GW, Kukwaya LG, Ling CM, Overby LR: Hepatitis B antigen in wildcaught mosquitoes in Africa. Lancet 2: 247, 1972.