Document 93GpGqwQwNO9gOZa1ow9q6bkq

Gastroenterology 66:450-464, 1974 Copyright 1974 by The Williams & Wilkins Co. PROGRESS IN HEPATOLOGY ------------------------------------------------------------------------------------------------- Vol. 66, No. 3 Printed in U^A. N'inCfcTHlS MATERIAL MAY Bt. protected by copyright ) aw (TlTtJ 1? G-S. CODE! GGU-T CHRONIC ACTIVE LIVER DISEASE REEXAMINED: PROGNOSIS HOPEFUL W. H. J. SUMMERSKILL, M.D. Gastroenterology Unit, Mayo Clinic and Mayo Foundation, Rochester. Minnesota The concept of chronic active liver dis ease (CALD) was proposed 5 years ago to encompass several conditions of unknown etiology characterized by continuous or recurrent hepatic inflammation with cell necrosis.1 CALD was considered to be in curable and usually to progress to cirrho sis, liver failure, or both. Concern was ex pressed (a) that uniform diagnostic criteria and specific therapeutic regimens were neither applied nor available; and (b) that an insouciant approach to clinical mani festations, coupled with unvalidated pre sumptions of etiology, had endowed dis orders, not always objectively distinguish able from each other, with a plethora of de scriptive or eponymous labels. An orderly approach was urged, beginning with stan dardized criteria which would allow the natural history of CALD to be defined and examined from the practical aspects of chronicity, activity, and responses to treat ment. Since then, numerous clinical, biochemi cal, and histological features have been defined which usefully typify the inter related conditions comprising CALD and exclude other conditions earlier considered indistinguishable; associations between hepatitis B antigen (HB Ag) and all the component disorders of CALD have pro vided the first integrating concept with re gard to possible etiologies; and valuable information concerning treatment has at last been garnered from properly designed trails. These and other areas of progress will be reviewed, emphasizing (a) the characteristics and differential diagnosis of CALD; (b) current viewpoints concerning etiology; and (c) therapeutic indications and responses. What is CALD? CALD (table 1) comprises disorders with clinical, biochemical, immunochemical, and histological features which differ mainly in degree, often have similar re sponses to treatment, and are not etiologically distinct. The more extensive the he patic inflammation, the greater the changes in liver function or immunoserological tests; the worse the prognosis, the more urgent the need for treatment. The liver diseases include chronic active ("ag gressive") hepatitis; subacute hepatitis ("subacute hepatic necrosis") with bridg ing or multilobular necrosis; and an inac tive phase of hepatitis, which occurs spon taneously or is induced by therapy. Each may be associated with cirrhosis, and, as the typical histological patterns are not then always evident, cirrhosis with active or inactive hepatitis more conveniently describes such instances. This terminology approximates popular usage, preserves de scriptive morphological labels of demon strated clinical importance and retains from an extensive selection (a recent tally2 identified 47 options) only conditions with the scientific credentials--particularly dif ferences in prognosis and treatment--to be considered independent entities. Received August 6, 1973. Address requests for reprints to: Dr. W. H. J. Summerskill. Gastroenterology Unit, Mayo Clinic and Mayo Foundation. Rochester, Minnesota 55901. Diagnosis Differentiation between the conditions comprising CALD may be surmised from the clinical features, supported by changes 450 Vol.66, No. 3 Printed in USA. March 1974 IOSIS HOPEFUL zhester. Minnesota phasizing (a) the rential diagnosis of /points concerning peutic indications VLD? ises disorders with mmunochemical, res which differ have similar red are not etiologiextensive the hehe greater the or immunoseroorognosis, the eatment. The onic active ("agoacute hepatitis sis") with bridg es; and an inaciich occurs spony therapy. Each irrhosis, and, as attems are not osis with active re conveniently 'his terminology :e, preserves debels of demonre and retains (a recent tally* conditions with >articularly difsatment--to be ities. rhe conditions surmised from ed by changes PROGRESS IN HEPA TOLOGY 451 452 PROGRESS IN HEPATOLOGY Vol. 66, No. 3 in liver function tests, but reliably deter mined only by liver biopsy.''4 Expert inter pretation is often essential since the mor phological features exemplify interrela tionships which differ mainly with regard to the extent and degree of inflammation and cell necrosis. Chronic active (aggres sive) hepatitis5 features portal tract in flammation with moderate or severe piece meal necrosis of liver cells extending out wards from the limiting plate. More exten sive inflammatory changes characterize subacute hepatitis,4,6 and necrosis may span the lobule between portal or hepatic veins (bridging necrosis), or cause more diffuse destruction, with collapse of adja cent lobules (multilobular necrosis). Inac tive hepatitis is mainly limited to the portal areas and comprises round cell infil tration, Kupffer cell proliferation, and mild or absent piecemeal necrosis.4- 7 These findings are identical with those also typifying chronic persistent hepatitis.5 In all these conditions, patchy lobular changes with "acute viral" characteristics' may be identified in one-half the in stances,4 and cirrhosis, if present, is usu ally of the macro- or micromacronodular variety. Sampling and observer errors are trivial in detecting the presence, degree, and type of hepatitis in CALD; but blind biopsy may overlook the presence of cirrho sis, particularly when macronodular, in up to 50% of instances,9 and peritoneoscopy or other techniques may then be indicated to establish a complete diagnosis. Chronicity and Actiuity The poor prognosis earlier accorded CALD must be abandoned after the devel opment of better or successful treatments.7'l0-11 But variable diagnostic criteria and differing types or schedules of therapy, together with the uneven caliber and duration of follow-up, still preclude definitive revisions of natural history. The decisive potential of treatment is exempli fied by a series in which more than onethird of the patients assigned to less effec tive therapies died within 6 months, whereas progressive remission of all clini cal, biochemical, and morphological fea tures of the disease were the rule in those allocated to the better regimens.7 However, spontaneous remission, even of severe dis ease, occurs in 20% of instances. Prognosis appears worse with regard to early death and later complications with subacute hep atitis, *7 especially when multilobular ne crosis is present,7 or with cirrhosis and active hepatitis.12 By contrast, it is uncer tain if chronic active hepatitis is inevitably aggressive and associated with an ominous outlook.4- 12 Cook et al.10 found that 18 of 49 patients died during a period of 6 years, and that cirrhosis had developed in all 10 available for autopsy; but- whether this outcome is attributable to chronic active hepatitis rather than fluctuations in dis ease activity,3 with episodes of subacute hepatitis,4 is uncertain. These and other factors determining the course of CALD. including remission, relapse, and treat ment failure, are more fully considered later in this paper in relation to therapy. Chronicity of active liver disease can now be predicted from the morphological features on biopsy. The Ranges of chronicactive or subacute hepatitis signify the probability of a prolonged course and chro nicity is presumed irreversible when cirrho sis is demonstrated. The features of chronic active (aggressive) hepatitis described by the European group5 have been substanti ated as reliable indicators of continuing disease by the behavior of patients selected by these criteria.7-10-11 The great majority had continuing disease activity throughout the period of observation, sometimes de spite treatment and with progress to cir rhosis and a fatal outcome. The finding of subacute hepatic necrosis (comprising both bridging and multilobular necrosis associ ated with subacute hepatitis) as described by Boyer and Klatskin* also portends con tinuing disease, with or without cirrhosis, in those who survive.4,7 Of 42 such patients in their study, which was confined to those developing subacute hepatitis as an early sequel to acute viral hepatitis, 50% devel oped cirrhosis, chronic active hepatitis, or both during a follow-up period of 4 years. The Mayo study, comprising patients who developed subacute hepatitis at different stages of the course of CALD, reported these lesions in 16 of 35 survivors12; fur- Vol. 66, No. 3 ns.7. However, 3v _,i of severe di stances. Prognosis ird to early death vith subacute hepn multilobular neuth cirrhosis and itrast. it is unceratitis is inevitably i with an ominous 0 found that 18 of 1 period of 6 years, eveloped in all 10 but whether this to chronic active actuations in di odes of subacute These and other course of CALD, apse, and treat- fully considered ition to therapy, liver disease can he morphological hanges of chronic ititis signify the l course and chrosible when cirrho- ires of chronic , described by e been substanti>rs of continuing patients selected he great majority tivity throughout i, sometimes de- i progress to cire. The finding of (comprising both r necrosis associitis) as described Iso portends condthout cirrhosis, `42 such patients :onfined to those titis as an early tit is, 50% devel;ive hepatitis, or eriod of 4 years, ng patients who ;itis at different 3ALD, reported survivors1*; fur- March 1974 PROGRESS IN HEPATOLOGY 453 thermore, cirrhosis had already developed in the majority who died with multilobular necrosis.7 Biochemical criteria can also gauge the likelihood of disease chronicity. The 10-week elevations of serum glutamic oxaloacetic transaminase (SGOT) and yglobulin continuing without improvement, and cited by Geall and his associates,1 proved effective when applied in practice.7 Also, in evaluating large series of patients with acute hepatitis reported in the litera ture, all but five of 370 instances of self limited disease would have been excluded by using these criteria.13 Activity of CALD can also be assessed by histological and biochemical methods. Morphological features of activity are re flected by the extent of hepatic cell necro sis, together with the degree of associated round cell infiltration and Kupffer cell proliferation. The most sensitive biochemi cal indicators of these changes are the degrees of elevation of serum transami nases and -y-globulin.7- 11 Preliminary evidence suggests that serum bile acid concentrations are also valuable, and, in addition, discern instances of activity not revealed by standard histological and bio chemical techniques.14 Geall and associates1 predicted that 50% of patients with x 10 elevation of SGOT, or x 5 of SGOT and x 2 of-/-globulin, who failed to improve after 10 weeks or more, would have disease of such activity that death would occur within 3 years. When applied prospectively, more than one-third of such patients died within 6 months unless adequately treated.7 Less extreme abnor malities of liver function tests are consist ent with the presence of CALD, 10- 11 and sometimes little or no abnormality may be detected, especially when the disease is approaching remission,7-17 or is difficult to differentiate from persistent hepatitis. Newer Clinical Associations of CALD Findings from two large series of patients with CALD have been published in the last 5 years.*1S These reemphasize the fre quency of systemic features, associations between CALD and diseases of other or gans considered to have an immunopathological basis, the varying patterns of abnor mality of liver function and immunoserological tests, and the diverse morphological features of the condition. The similarity of chronic active hepatitis in the pediatric age group to the adult form of the disease has also been depicted,1' and occasional associ ations between CALD and fibrosing al veolitis,17 thrvrotoxicosis,1* or eosinophilia with Coombs positive hemolytic anemia1* have been recorded. Resemblances be tween CALD and primary biliary cirrhosis or cryoglobulinemia have practical impor tance. Cooksley and his colleagues20 de scribed clinical, biochemical, or immuno chemical features of cholestasis consistent with the diagnosis of primary biliary cir rhosis in 14 of 30 patients whose course and biopsy features were those of CALD. The correct interpretation of dysglobulinemias in CALD can also be obscure. Jori and Buonanno21 describe relationships between chronic (active or persistent) hepatitis, or cirrhosis and cryoglobulinemia. In patients whose purpura, arthralgia, and hepatosplenomegaly were attributed to cryo globulinemia, overt signs of liver disease were usually inapparent, whereas, in others, evidence of liver disease was more prominent. The authors suggest that the cryoglobulin-associated syndrome may lead to liver involvement, but the alterna tive that CALD is associated with synthe sis of abnormal plasma immunoglobulins, some of which reversibly precipitate on cooling, also merits consideration. Zawadzki and Edwards22 in fact document dysim- munoglobulinemia in 20 cases of chronic hepatobiliary disease, find transformation of polyclonal to monoclonal gammopathy in two instances, and cite evidence that prolonged paraproteinemia associated with liver disease may ultimately lead to overt myeloma. Abnormal protein metabolism in CALD also accounts for the other numer ous immunoserological phenomena which are discussed later. Some recent additions to the list include false-positive tests for serum carcinoembryonic antigen,23'25 Sal monella agglutinins,25 and antimitochon- drial antibody. **27 Patients with chronic hepatitis (includ ing CALD) and HB Ag may present newly recognized syndromes. The development of !? 454 PROGRESS IN HEPATOLOGY Vol. 66, No. 3 periarteritis nodosa under such circum stances was considered by Gocke et al." to represent tissue damage caused by cir culating immune complexes of antigen, homologous IgM antibody, and comple ment components. This explanation was not supported by the density gradient ex periments of others using sera from pa tients with periarteritis or HB Ag.29 The pathogenesis of the association between membranous glomerulonephritis and chronic persistent hepatitis with HB Ag, described by Combes and his colleagues,30 also invokes immune complex deposition. Immunofluorescent staining of kidney tis sues suggested that deposits of IgG, C-3, and antigen may cause the glomerular damage. What Isn't CALD? Differential Diagnosis The majority or all of the findings char acterizing CALD may be mimicked by related conditions (unresolved or relapsing acute hepatitis and chronic persistent hep atitis), as well as by certain unrelated conditions (primary biliary cirrhosis, Wil son's disease, or hepatitis due to certain drugs). Difficult clinical distinctions between CALD and unresolved or relapsing acute hepatitis usually require tissue examina tion. Scrutiny of the limiting plate area is essential, since portal and lobular changes may be similar in all these conditions. Acute viral hepatitis rarely continues for 10 weeks without improvement,13 whereas re lapse, accompanied by deterioration in clinical and biochemical findings, is char acterized by a limited course and followed by improvement. Should this not occur, the possibility of subacute hepatitis* is pertinent. Differentiating chronic persistent hepa titis from CALD can be particularly diffi cult. The natural history of chronic persist ent hepatitis is not fully documented--no information is available to determine whether the disease continues through life, is self-limited, or is subject to fluctuations in activity. Furthermore, the clinical, bio chemical, and morphological features of chronic persistent and mild chronic active hepatitis resemble each other, and sharp criteria separating the conditions are lack ing. Chronic persistent hepatitis is typi cally benign and nonprogressive and re quires no treatment. Symptoms are non specific or absent. Objective abnormalities comprise occasional enlargment or tender ness of the liver, associated with elevation of serum transaminase.31 Becker et al.31 will accept concentrations of serum aspar tate transaminase of up to x 20 as consist ent with the diagnosis, provided that other tests are normal and that the liver biopsy shows only inflammatory round cell infil trate in the portal tract ("triaditis") with minimal or absent piecemeal necrosis. Pa tients with mild chronic active hepatitis have very similar features, and, indeed, findings may be identical when they are in remission.7 One well documented study showed that 5 of 30 patients originally classified as having chronic persistent hep atitis progressed to chronic active hepatitis during a follow-up which exceeded 2 years.32 By contrast, another study of 63 patients with severe CALD contained 2 patients with florid clinical manifestations and widespread abnormalities of liver function tests whose initial biopsies were interpreted as chronic persistent hepatitis--although each later developed character istic morphological features of chronic ac tive hepatitis.7 Given these uncertainties, and the slight sampling error of liver bi opsy,9 all patients considered to have chronic persistent hepatitis should be fol lowed for 1 year.31 Longer supervision is recommended for those who continue to have prominent symptoms. SGOT exceed ing x 5 normal, abnormalities of other liver function tests, positive immunoserological tests (including elevated serum y-globulin concentrations), or the presence of identi fiable piecemeal necrosis on biopsy.12 Certain conditions now accepted as hav ing different etiologies may nevertheless feature elevations of serum transaminases and 7-globulin, positive immunoserology. and patterns of necrosis on biopsy which are identical with those found in CALD. The diagnosis of primary biliary cirrhosis March 1974 under such circum: age and sex (espec third decade), prc hyperlipidemia. Di; not necessarily fat antimitochondrial ' this may occur in confirmation depe: characteristic bile purative cholangiti they have develop* only in rare instanc conventional doses without azathioprin much more consists cirrhosis than CALI Similarities betv and CALD have rec Sternlieb and Schei patients, ages 12 to clinical, biochemica tures consistent wit! a careful history ; lamp examination rings, serum ceruloj and urinary' copper < datory screening test Changes resembl subacute hepatitis i lately been shown drugs oxyphenisatir More recently, iso criminated as a caus with bridging or ir Evidence of possible identified in 120 the United States.1 appears frequently i ample, of 21 consec * Goldstein and his c that 9 were associa laxatives containing cases could have oc< with methyl dopa. ' subacute hepatitis 1 several investigator 5 toxic etiology, presui sitivity reaction, by tests. Improvement i Srawal of the drugs raft whether this is aXft additional thera ub Voi.ee. No. 3 active , sharp >ns are lackitis is typiive and Te ns are non-normalities t or tenderh elevation ker et ai.31 rum aspar) as consisti that other iver biopsy i cell infillitis") with ;crosis. Pae hepatitis d, indeed, they are in ited study originally istent hepe hepatitis :ceeded 2 ;udy of 63 -ntained 2 ifestations c liver .s were hepatitis:haracterhronic acirtainties, f liver bito have Id be folrvision is ntinue to T exceed>ther liver erological '-globulin of identi- y.12 d as havertheless iminases serology. >y which 3ALD. cirrhosis March 1974 PROGRESS IN HEPATOLOGY under such circumstances is suggested by age and sex (especially females after the third decade), prominent pruritus, and hyperlipidemia. Distinction from CALD is not necessarily facilitated by a positive antimitochondrial antibody test,26- 27 since this may occur in both conditions, and confirmation depends on demonstrating characteristic bile duct lesions (nonsup purative cholangitis) on liver biopsy once they have developed.53 Prior to this, and only in rare instances, failure to respond to conventional doses of prednisone, with or without azathioprine, can be interpreted as much more consistent with primary biliarycirrhosis than CALD. Similarities between Wilson's disease and CALD have recently been depicted by Sternlieb and Scheinberg34 who reviewed 7 patients, ages 12 to 28 years, all presenting clinical, biochemical, and histological fea tures consistent with CALD. In addition to a careful history and examination, slit lamp examination for Kayser-Fleischer rings, serum ceruloplasmin concentration, and urinary' copper excretion are now man datory screening tests for suspected CALD. Changes resembling chronic active or subacute hepatitis in all respects have lately been shown to be induced by the drugs oxyphenisatin33 and methyl dopa. More recently, isoniazid has been in criminated as a cause of subacute hepatitis with bridging or multilobular necrosis.36 Evidence of possible drug toxicity was not identified in 120 consecutive patients in the United States,15 but the association appears frequently in Australasia. For ex ample, of 21 consecutive cases of CALD, Goldstein and his colleagues37 considered that 9 were associated with ingestion of laxatives containing oxyphenisatin, and 5 cases could have occurred after treatment with methyl dopa. The high incidence of subacute hepatitis has been stressed, and several investigators have confirmed the toxic etiology, presumed due to a hypersen sitivity reaction, by recourse to challenge tests. Improvement of the disease on with drawal of the drugs has been documented, but whether this is always complete with out additional therapy in severely affected cases is uncertain. Careful clinical evalua tion will probably incriminate other com pounds capable of causing hepatic damage similar to that found in CALD. Etiology of CALD Examination of pathogenesis involves two areas. The first concerns external fac tors initiating chronic inflammatory dis ease of the liver (how many are there and are they similar or dissimilar?); and the second covers those host mechanisms which may determine a chronic reaction to factors causing self-limited disease or no recognizable reaction at all in others. The subject is conveniently approached by considering the relationships evolving between HB Ag, diseases of the liver, and host responses. Evidence that HB Ag is in separably linked with the virus of type B hepatitis is persuasive, and persistence of the antigen in chronic liver disease pre sumes that hepatic inflammation was initi ated by the virus, perpetuated by chronic viral infection or both.3* HB Ag is key to etiology since it alone has been identified in all types of CALD--inactive (or chronic persistent) hepatitis, chronic active hepati tis, subacute hepatitis, and cirrhosis.39-42 These associations are documented in ap proximately 10 to 30% of instances of CALD in the Western world. The preva lence is lower in Australasia and much higher in some other areas. The incidence of chronic hepatitis developing as a sequel to acute type B hepatitis is about 10%. Redeker43 found that 9.7% of 134 cases progressed to chronic disease (all with chronic persistent hepatitis), and that antigenemia continued in half of them. In Copenhagen, Nielsen and his colleagues44 reported that the antigen remained in the serum of 11 of 112 patients with acute virus B hepatitis for more than 13 weeks. Eight of these patients developed chronic active hepatitis, and 2 had chronic persistent hepatitis. A chronic course may be predictable from the appearances and characteristics of "viral-like" material in the sera. Large numbers of "Dane" particles are usually present in patients with chronic hepatitis /K 456 PROGRESS IN HEPATOLOGY Vol.66, No.3 and/or cirrhosis, whereas they are seldom identified in acute uncomplicated hepatitis and absent in healthy "carriers" of HB Ag. Differences in ratios between the small spherical and rod-shaped bodies also asso ciated with HB Ag are less striking.45 Once CALD with HB Ag is established, the dis ease runs a more serious course in un treated patients with anticomplementary assays,4* although no strong evidence ex ists that deposition of immune complexes causes the hepatic (as opposed to extrahepatic) damage in CALD. Both the spectrum of CALD and the proportion of cases attributable to vriral infection would be greatly expanded if the Milan antigen47 was confirmed as a valid marker for the virus of type A hepatitis, since Doniach4* identified this material in the sera of 16 to 36% of patients with chronic active hepati tis, primary biliary cirrhosis, or crypto genic cirrhosis. The proposal that host rather than exter nal factors determine the course (chronicity or otherwise) of hepatitis with HB Ag originates from clinical observations. These, summarized by Sherlock,35 com prise inability to correlate the variable severity and course of hepatitis with'differ ences in virulence of the virus or quantities of the antigen in the serum, and the fact that infants or patients with impaired cellular immunity (for example, chronic renal disease, leprosy, lymphoma) more often have subclinical or mild chronic hep atitis than severe acute or subacute dis ease. In preliminary studies, hypotheses are being tested that chronicity is specifi cally determined by abnormalities of host immunity mediated by T cells. Dudley and his colleagues49 suggest that T lympho cytes normally induce proliferation of spe cifically sensitized cells which recognize the antigen when it is extruded to the surface of infected hepatocytes, and then, perhaps through mediators, destroy both the liver cell and the infective agents (thus causing acute self-limited hepatitis). Im pairments of T cell function might lead to variable amounts of hepatic damage and permit antigen continuously to proliferate within liver cells. Preliminary and circum stantial evidence of reduced cell-mediated immunity in patients who carry HB Ag (with or without chronic hepatitis) includes the failure of HB Ag sera to induce blast transformation of lymphocytes,49 inhibit leukocyte migration,50 or impair lympho cyte transformation induced by phyto hemagglutinin51 in such instances. Normal responses are preserved in individuals with acute B-type hepatitis. On the other hand, Bolin and others 51 employed dinitrochlorobenzene skin sensitization as an indicator of cell-mediated immunity, and found in tact responses in carriers of HB Ag. The belief that CALD sometimes repre sents an "autoimmune" process invokes a different mechanism and is perpetuated by titles such as "autoallergic," "autoim mune," or "lupoid" hepatitis, although it has not gained universal acceptance. The postulate demands that primary host mechanisms be directed against tissue con stituents of the host in the absence of any triggering external agent,53 as may occur in Addisonian pernicious anemia or Ha- shimoto's thyroiditis. The process arises de novo, perhaps as a forbidden clone. Several variably nonspecific but unexplained simi larities between CALD and conditions, considered to have an immunopathic basis, support the hypothesis. Such common "immune markers" include the frequent coexistence with CALD of inflammatory diseases of unknown etiology involving other organ systems (ulcerative colitis, thy roiditis, sclerodermatomyositis, etc.): ele vations of serum y-globulin concentra tions, together with all immunoglobulin fractions; positive serological tests for nu merous non-organ specific antibodies; the presence of plasma cells on liver biopsy: and the possibly immunosuppressive role of adrenocorticotropin and corticosteroids, with or without azathioprine, in the ther apy of CALD. The major objections to an autoimmune basis for CALD comprise the failure to demonstrate organ- (liver) spe cific antigens and antibodies; the fact that the majority of the immune markers cited do not specify the type of liver disease, but may be present in a variety of hepatobili ary disorders; and evidence that such markers can also occur in CALD associated with HB Ag, or in chronic hepatitis induced March 1974 J sponse to external 5*. 55 p0pper an(j y inconsistencies be viral etiologies in identical features, tions. The virion complexed with m customary for oth< may include pre-e liver cell; or speci nents may be in antigen particle, events is envisage sive immunologice against the virus c cytes damaged b search for autoan liminary work of scribes isolation fr of an IgG type experimental cone only by liver speci An extensive 1 inconclusive sean etiological signtfic serum non-organ timitochondrial, a cle), immunoglobt thematosus cell eases or HB Ag. I suggested that HI antibody are mutt a stance later mo believed that pat lower titers of s than others. Bulk studying a small s lack of overlap ot lupus erythematoe might account for ior of CALD.59 Lthat lupoid hepa with HB Ag.54-55 prospective studyidentical pre-estal showed no major and sex), biochei vqr histological dif Sents with, or 58 erythematosus tes lion of a positive t active and seven V0I.66.N0.3 : ` HB Ag t includes induce blast tes,49 inhibit pair lvmphod by phyto- ,nces. Normal lividuals with ie other hand, dinitrochloro; an indicator md found in- HB Ag. etimes repreess invokes a srpetuated by "autoim5, although it eptance. The rimary host 1st tissue conbsence of any may occur in mia or Hatcess arises de done. Several plained simii conditions, - -'thic basis, common the frequent inflammatory igy involving re colitis, thy:is, etc.); elen concentrauunoglobulin tests for nultibodies; the liver biopsy; Dpressive role irticosteroids, , in the therjections to an comprise the 1- (liver) spe- the fact that markers cited r disease, but of hepatobilie that such JD associated atitis induced March 1974 PROGRESS IN HEPATOLOGY 457 by drugs, and therefore can arise in re sponse to external (viral or toxic) agents.35 si. 5s p0pper and McKay.5* acknowledging inconsistencies between autoimmune and viral etiologies in patients with otherwise identical features, offer compromise solu tions. The virion in the liver cell may be complexed with more host protein than is customary for other viruses; the viral coat may include pre-existing structures of the liver cell; or specific host protein compo nents may be incorporated into the B antigen particle. One or more of these events is envisaged as exciting an aggres sive immunological reaction directed both against the virus and against host hepatocytes damaged by it. Nevertheless, the search for autoantibodies continues. Pre liminary work of potential pertinence de scribes isolation from patients with CALD of an IgG type globulin, which, under experimental conditions, is fully absorbed only by liver specific protein.57 An extensive literature attests to the inconclusive searches for correlations of etiological significance between CALD. and serum non-organ specific antibodies (antimitochondrial, antinuclear, smooth mus cle), immunoglobulin fractions, lupus ery thematosus cell tests, autoimmune dis eases or HB Ag. For example, it was once suggested that HB Ag and smooth muscle antibody are mutually exclusive in CALD, a stance later modified by Wright,5* who believed that patients with HB Ag had lower titers of smooth muscle antibody than others. Bulkley and her colleagues,59 studying a small series, speculated that the lack of overlap observed between positive lupus erythematosus cell and HB Ag tests might account for differences in the behav ior of CALD.59 Later studies have shown that lupoid hepatitis can be associated with HB Ag.54- 55 Indeed, a comprehensive prospective study of patients selected by identical pre-established criteria for CALD showed no major clinical (including age and sex), biochemical, immunochemical, or histological differences between 30 pa tients with, or 58 without, a positive lupus erythematosus test, other than the associa tion of a positive test with evidence of more active and severe disease. Responses to treatment were identical in the two groups.55 In studying 94 patients with "chronic idiopathic liver disease," Finlayson et al.* reported significantly more autoantibodies to nuclei, smooth muscle, and mitochon dria in patients without HB Ag. Similar findings in CALD51 included greater eleva tions of IgG in HB Ag-negative individuals. In both series, the overlap of results was too great to support contentions of entirely different etiologies. The possibility that clinical as well as serological findings may yield clues to etiology' deserves emphasis. In evaluating extensive immunological data from 128 consecutive patients with severe CALD chosen by pre-established criteria,7 the only significant difference involving HB Ag was absence of this marker in all of 26 patients who had as sociated inflammatory diseases of an auto immune type and the corresponding ab sence of associated inflammatory disorders in all of 20 patients with HB Ag.12 Abnormalities of the complement sys tem were originally implicated in patho genesis of the arthritis with acute hepatitis by Alpert and his colleagues,52 and later evaluated as causes of other extrahepatic manifestations in chronic hepatitis. Mea surements of total hemolytic complement, C-3 and C-4, in 110 patients which chronic liver diseases indicated that low comple ment levels are associated with impaired liver synthetic function rather than related to the activity of the liver disease*3 or to the postulated etiology, as reflected by HB Ag or circulating non-organ specific antibod ies. On the other hand, Grob and associ ates*4 found decreased concentrations of B,A (a constituent of C-3) in more than one-half of 31 patients with chronic active or chronic persistent hepatitis and variable results in acute hepatitis. Patients with HB Ag more often had decreased levels of B,A that others. No correlations were dis covered between immunoglobulins G and M and B,A. The reduction of B,A in liver disease might be due to the formation of immune complexes, decreased synthesis, increased catabolism, or the formation of anticomplementary serum complements. Since B ,A concentrations are also reduced 600696&2 'it:-: 458 PROGRESS IN HEPATOLOGY Vol. 66. No. 3 in some autoimmune diseases, ihe authors speculate that low levels after acute hepa titis may herald the intervention of au toimmune processes and a chronic course. Another determinant of chronic disease, in this instance of postulated genetic origin, could be the twice normal incidence of histocompatibility antigens (HL-A1 and A8) in CALD. These have also been pre sented as immunopathic markers, signify ing an abnormal response to autoantigens and a predisposition to autoimmune hepatitis.65 A critical summary of this literature leads to the conclusion that, with the important exception of HB Ag. neither in dividual immunological tests, nor their permutations, yet reliably indicate etiology', diagnosis, or prognosis in CALD.21-68 Rather, such tests mainly reflect disease activity, since all revert to normal when CALD is effectively treated.7-!i Increased production of normal or abnormal proteins is probably responsible and immunoglobu lin G synthesis has been measured in CALD. found to be increased, and shown to return toward normal during treatment with prednisone.67 What "turns on" the production of various globulins in chronic liver disease is unknown. A theory recentlyadvanced by Thomas and others68 holds that the magnitude of hyperglobulinemia and immune response may be determined by the relative distribution of antigens between the liver and spleen. They demon strated that, in experimental cirrhosis, gutderived antigens contribute to globulin production since they are not efficientlytrapped in the liver and are consequentlydispersed to antibody-producing tissues, such as the spleen. Further clarification of etiologies of CALD is most likely to follow investiga tions directed (a) toward identification of additional external mechanisms (whether viral or otherwise) initiating the process; (b) toward host factors determining im munological responsiveness and chronicity; and (c) characterization of mechanisms whereby viral or toxic agents produce ne crosis and inflammation of liver cells and then cause diverse clinical and serological abnormalities. To these ends, the develop ment of suitable animal models for CALD is desirable, and the selection of clinical material more precisely defined than has usually been the case will be essential. Treatment of CALD For more than a decade, corticosteroids, adrenocorticotropin, and azathioprine, or its metabolic product, 6-mercaptopurine. were employed empirically in numerous, nonstandardized and uncontrolled treat ment schedules, propagating findings which could not be interpreted in relation to therapeutic indications, benefit-risk ratio, and effects on long term prognosis in CALD.1 Similar descriptions have latelyrecorded experiences with penicillamine, cyclophosphamide, or chloroquin. By con trast, the results of acceptably designed therapeutic trials in CALD, published in the past 2 years, dramatize improvements in management of the disease and promise a potential for more advances. Remission can be secured in a large proportion of instances, and there are even indications that the ultimate goal of cure may some times be attainable. Three controlled studies of CALD de serve scrutiny. In London, Cook and his colleagues10 reduced mortality with varia ble doses of prednisolone (usually approx imating 10 mg daily of prednisone). Treat ment did not uniformly improve liver func tion tests and effects on hepatic histology were not sought. The Mayo study em ployed higher doses of medications, differ ent schedules, and additional methods. These included double blind evaluation of responses, strict criteria for beginning and discontinuing therapy, and regular record ing of data which included hepatic histology.7 Maintenance schedules with prednisone (20 mg daily), or a combination of smaller daily doses of prednisone (10 mg) with azathioprine (50 mg), were signif icantly more effective than azathioprine (100 mg daily) alone or placebo. Predni sone, or the combination, not only in creased survival, but resulted in remission of jaundice, ascites, and other clinical features. Biochemical, immunochemical, and histological indicators of activity usu- March 1974 ally later underwent r trast, most patients al prine or placebo deteric died from hepatic failu of the start of therapy.' Murray-Lyon and his c that the probability of with CALD receiving a of prednisone, 15 mg than for those randomi (75 mg daily). Impro' liver function tests wen nisone, but alterations tures were not investi were initially given in these studies. Careful effects was undertaken attributable to drugs v the benefits of effectiv selection differed. C ( leagues10 enrolled 44 basis of histological e active hepatitis. The 47 Murray-Lyon and hi.chronic active hepatit cirrhosis, and at least ties of serum asparti y-globulin concentratic acceding to the Maysevere disease, since gr in liver function tests admission, and indivic hepatitis or cirrhosis co mated. Selection of Do all patients with apy? Risks to be balan include the likelihood c in 10% of patients after treatment,7- 10 and the ease subject to sponta those patients who wil available, cannot yet t fied. There is no evide lated etiology of CA sponses,7-12-55 althouj tures definitely affect rently, and providing confirmation of CALD biopsy is feasible, the for therapy are: (a) t tions of SGOT and y-g. um*t*uau3: ft li >1 s m a 460 PROGRESS IN HEPATOLOGY Vol. 66, No. 3 icity than the combination, but addition of azathioprine requires frequent monitoring of the hemogram, and the side effects of this drug are less well understood." Inci dental factors may force a decision. Pa tients with diabetes, peptic ulcer, cataract, or other partial contraindications to steroid therapy may be candidates for the combi nation, whereas those with cvtopenia asso ciated with liver disease or hypersplenism need prednisone. Rarely, contraindications to effective doses of either drug may coex ist. Splenectomy can then sometimes be successfully performed, improving cytopenia and permitting addition of azathio prine to therapeutic regimens.74 Alternative approaches to treatment with steroids, variably established in other conditions, but not shown to secure remis sion in CALD, include upward or down ward "titration" of the dose of steroids in response to changes in liver function tests.3,10 This, although minimizing side effects, may suppress biochemical indica tors of disease without causing subsidence of inflammatory activity in the liver. Alter nate day treatment with "double doses" of steroids may also reduce steroid toxicity75 but has not been fully tested in CALD. End Points of Treatment The major therapeutic end points in CALD should comprise full remission or cure of the disease on the one hand, and failure of treatment with ultimate death at the other extreme. Remission of all features of the disease in response to treatment has thus far been demonstrated in only one study7,73 and was characterized as (a) absence of symp toms and return of the patient to all customary activities; (b) disappearance of biochemical and immunochemical abnor malities associated with active hepatitis; and (c) resolution of the histological hall marks of active hepatitis. Trivial (< x 2 normal) elevation of SGOT is consistent with remission, as is the presence of inac tive hepatitis on biopsy, since these changes are chemically and histologically indistinguishable from benign chronic per sistent hepatitis. The presence of cirrhosis does not preclude remission of the hepati tis. The sequence of remission7,7* in 43 patients" receiving conventional treat ments and followed for 2 years or longer, showed that two-thirds were free from symptoms or clinical features of disease by 6 months; previously abnormal liver func tion and immunological tests have reverted to normal in three-fourths after 1 year; and, by 2 years, when 90% of patients have no clinical or biochemical evidence of ac tive disease, the liver biopsy shows either inactive hepatitis or normal appearances in 70%. Full remission occurred in less than 1 year in a minority, whereas others required treatment for 3 years or more. When remis sion is established, treatment is slowly discontinued over a 6-week period13,76 and patients are instructed to report unusual symptoms and arrange two weekly liver function tests during the first 6 months. In the absence of relapse, liver function tests and biopsy appearances revert t,o norma! in 30% of patients followed for 2 years or more.12 Relapse of CALD follows discontinua tion of therapy in one-half the patients and is usually evident within 6 months. "76 Fa tigue and arthralgias are associated with elevations of SGOT, which usually confirm deterioration before other liver function tests become abnormal. Liver biopsy veri fies relapse--always showing features ei ther of chronic active or subacute hepati tis. In 10% of patients, histological relapse occurs without characteristic symptoms or changes in liver function tests. Responses to second or subsequent courses of treat ment are identical with those of the first course. Each results in a 50% remission rate, but patients with subacute hepatitis and multilobular necrosis or with cirrhosis relapse more frequently.12,39 Treatment failure can be anticipated in approximately 20% of patients with severe CALD, since more than twO-thirds enter remission with effective therapy7* and few remain unchanged.7 Evidence of deteriora tion may be obvious within 3 months after the start of therapy and death then usually occurs from fulminating disease associated March 1974 with subacute I progressed to ci failure occurs lat ously adequate end-stage liver f; vanced cirrhosis The most commc ment failure prol of the therapeuti incorrect diagnos diagnostic criteri treatments have respond indicate cirrhosis with a< quires empirical roids with or wit basis of the relai gree or irreversit and treatment fa verity of the dise, but differing etic may determine n hepatic destructic cause circulatory which impair the Additional Ther Patients with ( libitum nutritious tivities that can t fatigue.7 Complice i managed by stanc potentially hepato be avoided.7 Won advised against pi ease has been in i ment for 6 month genic potential o unexcluded deletei on active liver dis tions should inclu liver function, to investigations nece disease or complic; incidence and nat gether with the ne> k, fully described els opsy is particular! remission or relaps- determining causes Patients with Ci % 462 PROGRESS IN HEPATOLOGY Vol. 66, No. 3 March 1974 mine chronicity of disease are reviewed, 12. Summerskill WHJ, Ammon HV, Baggenstoss active hepat and the importance of autoimmune mech anisms as causes of CALD is questioned. Recent advances in treatment show that effective therapy can induce remission in some instances of CALD and that ultimate cure may be feasible. By contrast, other findings in CALD are cited as indicating a poorer prognosis with the development of cirrhosis, liver failure, or both, despite AH: Treatment of chronic hepatitis. Hepatology 74 (Festschrift for Hans Popper) (in press) 13. Soloway RD, Summerskill WHJ: Chronic active liver disease: classification and treatment. Post grad Med 53:88-94, 1973 14. Korman MG, Summerskill WHJ, Go VLW et al: Sensitivity and predictive value of serum bile acid concentrations (SBA) in patients with chronic active liver disease (CALD) (abstr). Gastroenterology 65:554, 1973 1:1309-1312. 30. Combes B. S nephritis wii antibody co membrane. I 31. Becker MD. Prognosis of 1:53-57, I97C 32. Vido I. Selnu der chronisc treatment. It is to be concluded that en 15 Summerskill WHJ: Chronic active liver disease. chenschr 94:: couraging theoretical and practical ad vances have occurred, since until recently, CALD was considered to be of unknown etiology, and the majority of patients were believed unlikely to survive. Evaluation and treatment. Viewpoints on Diges tive Disease Voi 5, No 3, May, 1973 16. Dubois RS, Silverman A, Slovis TL: Chronic active hepatitis in children. Am J Dig Dis 17:575-582, 1972 17. Tumer-VVarwick M: Fibrosing alveolitis and 33. Christoffersei mal bile due hepatitis an Pathol 3:227 34. Sternlieb I. 5 first manifes REFERENCES chronic liver disease. Q J Med 37:133-149, 1968 18. Thompson WG, Hart IR: Chronic active hepatitis tern Med 76: 35. Reynolds TE 1. Geatl MG, Schoenfield LJ, Summerskill WHJ: and Graves' disease. Am J Dig Dis 18:111-119. active and lu Classification and treatment of chronic active 1973 oxvphenisatii liver disease. Gastroenterology 55:724-729, 1968 19. Panush RS, Wilkinson LS. Fagin RR: Chronic 36. Maddrey \VC 2. Desmet V: Chronic hepatitis (including primary active hepatitis associated with eosinophilia and Ann Intern \ biliary cirrhosis). The Liver. Baltimore, Williams Coombs -positive hemolyt ic anemia. Gastroenter 37. Goldstein Gi & Wilkins, 1972, p '286-341 ology 64:1015-1019, 1973 induced acti' 3. Mistilis SP. Blackburn CRB: Active chronic hep 20. Cooksley WG, Powell LIV. Kerr JF, et al: Choles 18:177-184. 1 atitis. Am J Med 48:484-495, 1970 tasis in active chronic hepatitis. Am J Dig Dis 38. Sherlock S: 4. Baggenstoss AH. Soloway RD, Summerskill 17:495. 1972 associated! h WHJ, et al: Chronic active liver disease. The 21. Jori GP. Buonanno G: Chronic hepatitis and 39. Gitnick GL. a range of histologic lesions, their responses to treatment, and evolution. Hum Pathol 3:183-198, 1072 cirrhosis of the liver in crvoglobulinaemia. Gut 13:610-613. 1972 22. Zawadzki ZA. Edwards GA: Dysimmuno- Australia ant with cirrhosis 40. Wright R. Mi n 5. De Groote J, Desmet VJ, Gedigk P, et al: A globulinemia associated with hepatobiliary disor antigen in aci classification of chronic hepatitis. Lancet 2:626- ders. Am J Med 48:196-202. 1970 2:117-121. 19 628. 1968 23. Go VLW, Schutt AJ. Moertell CG, et al: Radi 41. Summerskill 6. Boyer JL, Klatskin G: Pattern of necrosis in acute oimmunoassay of carcinoembryonic antigen AH, et al: T1 viral hepatitis. Prognostic value of bridging (sub (CEA).'A modified method and a clinical evalua chronic post-1 acute hepatic necrosis). N Engl J Med tion (abstr). Gastroenterology 63:95-101. 1972 antigenemia '283:1063-1071, 1970 24. Kupchik HZ, Zamcheck N: Carcinoembryonic 60:755. 1971 7. Soloway RD, Summerskill WHJ, Baggenstoss antigen(s) in liver disease. II. Isolation from 42. Dudley FJ. AH. et al: Clinical, biochemical and histological human cirrhotic liver and serum and from normal history of hi remission of severe chronic active liver disease: a liver. Gastroenterology 63:95-101, 1972 chronic liver i controlled study of treatments and early progno 25. Protell RL, Soloway RD, Martin WJ, et al: 43. Redeker AG: sis. Gastroenterology 63:820-833, 1972 Anti-Salmonella agglutinins in chronic active and Blood Tr 8. Poulsen H, Scheuer PJ. Schmid M, et al: Mor liver disease. Lancet 2:330-331, 1971 sium held at phological criteria in viral hepatitis. Review by an 26. Klatskin G: Mitochondrial antibody in primary cisco, March International Group. Lancet 1:333-337, 1971 9. Soloway RD, Baggenstoss AH, Schoenfield LJ, et biliary cirrhosis and other diseases. Ann Intern Med 77:533. 1972 YAS Berry, et p 55 & al: Observer error and sampling variability tested 27. Kaplowitz N. Finlayson NDC, Walmsley P. et al: 44. Nielsen JO. C in evaluation of hepatitis and cirrhosis by liver Hepatobiliary diseases associated with serum an- dence and m biopsy. Am J Dig Dis 16:1082-1086, 1971 timitochondrial antibody (AMA). Am J Med antigen in p< 10. Cook CG, Mulligan R, Sherlock S: Controlled 54:725-730, 1973 development prospective trial of corticosteroid therapy in ac 28. Gocke DJ, Hsu K, Morgan C, et al: Association 285:1157-1160 tive chronic hepatitis. Q J Med 40:159-185, 1971 between polyarteritis and Australia antigen. Lan 45. Nielsen JO, .' 11. Murray-Lyon IA. Stem RB, Williams R: Con cet 2:1149-1153, 1970 distribution ot trolled trial of prednisone and azathioprine in 29. Prince AM, Trepo C: Role of immune complexes cles in patien active chronic hepatitis. Lancet 1:735-737, 1973 involving SH antigen in pathogenesis of chronic carriers. N Er PROGRESS IN HEPATOLOGY Vol. 66, No. 3 GASTKOENTEJIOtOCY t Copyright 1974 by -negative liver disease. Gut 14:360-364, 1973 ity in patients with active chronic hepatitis and 62. Alpert E, Isselbacher KJ. Schur PH: The patho primary biliary cirrhosis treated with azathio- genesis of arthritis associated with viral hepatitis. prine. Gut 13:907-914, 1972 Complement-component studies. N Engl J Med 73. Makinodan T, Santos GW, Quinn RP: 1m- 285:185-189, 1972 munosuppressive drugs. Pharmacol Rev 63. Finlayson NDC, Krohn K, Fauconnet MH, et al: Significance of serum complement levels in chronic liver disease. Gastroenterology 22:189-247, 1970 74. Boye NP, Nordoy A, Gjone E: Splenectomy in active chronic hepatitis. Scand J Gastroenterol CHRONIC IN PERSPECTIV 63:653-659, 1972 7:747-750, 1972 64. Grob PJ, Jemelka HJ, Muller JW: B,A and SH 75. Page AR. Good RA, Pollara B: Long term results During the p; antigen in chronic hepatitis. Gastroenterology 61:91-95, 1971 65. Mackay IR, Morris PJ: Association of autoim mune active chronic hepatitis with HL-A1.8. Lancet 2:793-795, 1972 66. Sherlock S: The immunology of liver disease. Am J Med 49:693-706, 1970 67. Westergaard H, Jamum S. Ramsoe K, et al: Albumin and immpnoglobulin turnover in pa of therapy in patients with chronic liver disease associated with hypergammaglobulinemia. Am J Med 47:765-774, 1969 76. Ammon HV, Baggenstoss AH. Summerskill WHJ: Characterization and incidence of remis sion and relapse in chronic active liver disease (CALD) (abstr). Gastroenterology 62:173, 1972 77. Ferris AA, Kaldor J, Gust ID, et al: Faecal antigen in viral hepatitis. Lance 2:243-244, 1970 exceedingly fat implicitly1'* or "what is chroni by Summerskil. j previously,1 he I "what is chron Having neatly c labels in an attt tients with chronic active liver disease treated 78. Mazzur S: Menstrual blood as a vehicle ofAustra sion surroundin with prednisone. Scand J Gastroenterol lia antigen transmission. Lancet 1:749, 1973 ' leaves us with an 7:623-630, 1972 79. Fulford KWM, Dane DS, Catterall RD, et al: is, at least in my 68. Thomas HC, McSween RXM, White RG: Role of Australia antigen and antibody among patients somewhat dispu the liver in controlling the immunogenicity of attending a clinic for sexually transmitted dis clearly written, o r commensal bacteria in the gut. Lancet 1:1288-1291, 1973 69. Ammon HV, Summerskill WHJ, Baggenstoss AH: Initial morphologic appearances determine eases. Lancet 1:1470-1473, 1973 80. Heathcote J, Sherlock S: Spread of acute type-B hepatitis in London. Lancet 1:1468-1470, 1973 81. Ward R, Borchert P, Wright A, et al: Hepatitis B terpreted as CC problem. The te disease" is used l responses to treatment of chronic active liver antigen in saliva and mouth washings. Lancet pass a variety of l disease (CALD); a prospective study (abstr). 2:726-727, 1972 to share certain c Gastroenterology 65:526,1973 82. Schweitzer IL, Spears RL: Hepatitis-associated phological, and 70. Powell LW, Axelsen E: Corticosteroids in liver antigen (Australia antigen) in mother and infant. Whether a seari disease: studies on the biological conversion of prednisone to prednisolone and plasma protein binding. Gut 13:690-696, 1972 71. Elion GB: Biochemistry and pharmacology of purine analogues. Fed Proc 26:898-903, 1967a 72. Whelan G, Sherlock S: Immunosuppressive activ- N Engl J Med 283:570-572, 1970 83. Chalmers TC, Alter HJ: Management of the asymptomatic carrier of the hepatitis-associated (Australia) antigen. N Engl J Med 285:613-617. 1971 among the variot tory liver diseases productive in ten ology. prevention, uncertain. My cor detract from the gantly controlled ducted by the Ma_\ question the wist disorders under a * time. A major goal of inflammatory liver tion of etiological t netic mechanisms ment of these disor the epidemiology oi been given adequ. tokk Received August 21, Address requests for Koff, Medical Service, 1 to tion Hospital. 150 South Q Massachusetts 02130. <$> ir Marstellsr e/ aL: Splenomegaiic Liver Disease 133 ! SchaUmann, 0. 193-4. Ueber die Herz.v.J ng einiger Inhalationsnarkotica. Medi- lin & Ch.emie. Abhandlungcn aus c ledizinisch-chemisehen Forschungsstat- ten der I.G. Farbenindustrie A.G^ )T. II. Bayer-Meister Lucius. Leverkusen, West Germany. SchaUmann, O. 1938. In Toxikoloa hd Hygiene der technischen L&sungsmittel. K. B. Lehmann & F. Flury, . 130 (Monochlorathyltn). Julius Springer. Berlin, Germany. Schmid, M. 1968. Zur Leberlj ogie der verschieder.cn Formen des portalen Hochdruckes. In The Theg 'of Portal Hypertension. N. G. MarkoS, Ed. Georg Thieme Verlag. Stu , West Germany. Schmidt. F. W. 1969. Gerig' K^ktivitatsanstieg der Serum-Transactinasen. Deut. Med. Wochschr. 94: 225 Sch.nack, H., L. Stockinj :F. Wewalxa. 1967. Adventitious connective tissue cells in the space ofj gie and their relation to fibre formation. Rev. Intern,. H6pa:ol. 17: 855. ^ Simonds, H. R., A. I^Hth & M. H. Bioelow. 1949. Handbook of Plastics. 2nd f/gor~ edit, D. van Nostj^^Co., Inc. Princeton, N-J. Slater, T. F. 197 e radical mechanisms in tissue injury. Pion Ltd. London, 1 1 England. Smoron, G. L. 'A. BaTtifora. 1972. Thorotrast-induced hepatoma. Cancer 30: 1252. Smyth, H. F.,, 1956. Improved communication. Hygienic standards for daily inhalation. Ind. Hyg. Assoc. J. 17: 129. Soloway, , A. H. Baccenstoss, L. J. Schoe.nfield et a!. 1971. Observer error an pling variability tested in evaluation of hepatitis ar.d cirrhosis by liver bi 1/137. Stein, G Am. J. Digest. Diseases 16: 1082. uhe, C.-E. Lance & G. Veltman. 1973. Bandformige Osteolysen in den aiangen des Handskeletts. Fortschr. Gebiete Roentgenstrahltn Nu- 118: 60. 138. StenosBPL J. 1965. Fibrogenesis along the hepatic sinusoids in carbon-tetra- -induced cirrhosis. An electron microscopic study. Expti. Mot. Pathol. v&Js, . H., D. C. Tormey &. J. Blom. 1973. Hemangiosarcoma and pathologic .ure of the spleen. Cancer 31: 1213. ,, I., I. Drejman & M. Valaskai. 1963. Contributii la studiul imbclnavirilor oduse de clorura de vinil. Med, Intema IS: 967. Jiu, I., I. Drejman & M. Valaskai. 1967. Etude des maladies dues au chlorure Poe vinyle. Med. Lavoro 58: 261; Abstr. Hyg. 43:430 (1968). UMMERSXlLL, W. H. J. 1974. Chronic active liver disease reexamined: prognosis hopeful. Gastroenterology 66: 450. 1441 fwUtSKt^^E^196LHandbuch der gesamten Arbeitsmedizin. E. W. Baader, Ed. Vol. IV/2: Arbeitshygiene. Urban & Schwarrenberg. Mflnchen, West Ger many. . Gut G. Klatskin & W. W. L, Glenn. 1959. Portal hypertension and geal varices. Their occurrence in the absence of both infra- ------ vein. New Engl. J. Med. 261: 6O2472 |j/l -- -- Thank you for choosing pt-intl* Advanced Information Consultants 1/ -- -- for your document delivery needs ---- MI Mail GEO02---- s.a.ss----- (WCha^T P ?/oy?6/>r BFG36898