. Liver, 1982: 2: 61-67 Key words: viral hepatitis; acute hepatitis non-A, non-B: light microscopy.

Acute hepatitis non-A, non-B; are there any specific light microscopic features?

MARTIN SCHMID', MAURO PIROVINO', JOSEF

ANCHI*

'Department of Internal Medicine, Stadtspital Waid, Zurich and *Department of Pathology, University of Basel, Basel, Switzerland

ALTORFER' , FRED GUDAT* AND LEONARDO BI-

ABSTRACT - Coded examination of liver biopsies from a total of 24 patients with acute hepatitis non-A, non-B revealed two main histological trends: (a) acute viral hepatitis with confluent necrosis (sublobular and bridging) carrying a relatively good prognosis and taking a chronic course in only four out of 14 patients (29%); and (b) acute viral hepatitis with severe portal infiltration rich in lymphocytes and plasma cells, lymph follicles with germinal centers and bile duct lesions, as described by Poulsen & Christoffersen. The latter group showed a very high tendency to transition to chronic hepatitis (sixout of seven patients, 86%) or a course characterized by one or multiple acute relapses (one out of seven patients, 14%). Bile duct lesions, if present in biopsies of patients with acute hepatitis, are of diagnostic and prognostic value. They point to the etiological possibility of a hepatitis non-A, non-B and, at the same time, they indicate a high likelihood of evolution to chronic liver disease.

Accepted for publication 10 October 1981

Little is known about the histopathologic fea- tures of acute hepatitis non-A, non-B in humans (1, 2). In a search for possible specific light microscopic changes of acute hepatitis non-A, non-B in man, the biopsies from a total of 54 patients with acute viral hepatitis, type A, B,

and without knowledge of any clinical and laboratory data by three different examiners.

Material and methods (a) Patients: Definition of etiol- ogical type of hepatitis, epidemi- 01 OgY Among the 54 patients investigated, 15 had acute hepatitis A, 15 acute hepatitis B, and 24 acute

in the three groups was comparable (40, 45 and 41 years). The diagnosis of acute viral hepatitis A was

and non-A9 were reviewed hepatitis non-A, n0n-B. The mean age ofthe patients

0106-9543/82/010061-07 $02.50/0 0 1982 Munksgaard, Copenhagen

62 SCHMID ET AL.

based on the presence of anti-HAV antibody of the IgM class in the serum by radioimmunoassay (3). Acute viral hepatitis, type B, was defined by the finding of HBsAg in serum by radioimmunoassay (Ausria, Abbott Laboratories) followed by serocon- version during convalescence, and by the presence of anti-HBc antibody of the IgM class by ELISA (4). The serological markers of hepatitis A and B were deter- mined at admission to the hospital, and 3 weeks and 3 months later and, in cases where chronicity was suspected, every 6-12 months during the follow-up period. The length of follow-up was 13-46 months (mean 28.5 months).

The diagnosis of acute Viral hepatitis non-A, non-B was made by exclusion, i.e. the lack of serological markers of hepatitis A and B in the presence of typical clinical and biochemical signs of acute hepatitis. In all patients, immune histological examination of the liver biopsy was negative for HBsAg and HBcAg (5 ) . Negative serology for cytomegalovirus, Epstein-Barr virus, herpes virus and adenovirus was prerequisite.

All patients with a history of excessive alcohol intake or with histological features indicating alco- holic or drug-induced liver disease (such as marked fatty change, abundance of eosinophils, granulomas) were excluded.

A potential parenteral source of infection was identified in 17 out of 24 patients with acute non-A, non-B hepatitis (71 %;Table 1). A history of previous blood transfusion was obtained in two patients, one patient had a surgical intervention, eight patients were intravenous drug users, and six patients reported other needle puncture exposure. The incubation period in these cases varied conspicuously and ranged from 20 to 120 days. No definitive trend toward two separate groups with a short or long incubation period was recognizable. Seven patients (29%) did not reveal any obvious source of exposure.

(b) Liver biopsy Liver biopsy is a routine procedure in all patients admitted to our hospital with acute or chronic

hepatitis. To avoid any kind of selection, all biopsies performed during a certain period of time were included in this study.

Initial liver biopsy was obtained in all patients by laparoscopy. Twelve patients were subjected to repeat biopsies. The mean interval between peak serum ALT levels and initial liver biopsy was 13 days (range 0-90 days). In only five of the 54 patients was the initial biopsy taken later than 20 days after peak ALT. These five patients were equally distributed among the four histological trend-groups which are defined below. One or more follow-up biopsies were performed in 12 patients because chronicity was suspected from clin- ical findings 6-36 months later.

The biopsy material was fixed in 4% neutral formalin solution, embedded in Paraplast@ plus tissue medium, cut into 5 pm sectibns and stained with hematoxylin and eosin, chromotrope aniline blue (connective tissue stain) and Prussian blue.

Results In addition to the classical picture of acute viral hepatitis, characterized by spotty liver cell necrosis, liver cell swelling, disorganization of liver cell plates, portal and intralobular inflam- mation, the following four histological trends were observed (Figs. 1-4):

Trend I: Portal inflammatory reaction rich in plasma cells, combined with predominantly periportal liver cell necrosis and relatively mild changes in centrilobular areas.

Trend ZI. Confluent liver cell necrosis, notably bridging necrosis.

Trend IZZ. Bile duct lesion with abnormal duct epithelium as described by Poulsen & Christof-

Table 1 Potential source of exoosure in 24 oatients with acute heoatitis non-A. non-B

No. Incubation period Potential source of exposure (n=24) (days)

Blood transfusion 2 45-58 Surgical intervention 1 20 Illicit self-injection 8 - Other needle puncture exposure 6 28-120 Non-obvious source 7

ACUTE HEPATITIS NON-A, NON-B 63

Fig. 1. Trend I: Liver cell necrosis predominantly at the acinar periphery (a) (H&E, X320), in contrast to only mild changes in the centrolobular area (b). Occasional liver cells containing fat droplets (H&E, X200).

Fig. 2. Trend 11: Acute hepatitis with confluent necrosis: sublobular (a) (H&E, X 80), and bridging (b) (H&E, x 200).

fersen (6), always accompanied by dense lym- phocytic portal infiltration (mimicking lymph follicles with germinal centers) and many intralobular and portal infiltration. plasma cells.

Trend IV. Predominantly centrilobular liver cell necrosis with many macrophages and minimal

In 41 out of the 54patients, the microscopical

Table 2 Trends in histological findings from 54 patients with acute hepatitis types A, B, and non-A, non-B

No. of Hepatitis Hepatitis Hepatitis patients A B non-A, non-8

Trend I 6 3 2 1 Trend II 24 4 6 14 Trend III 8 0 1 7 Trend IV 3 1 1 1 No trend 13 7 5 1

Total 54 15 15 24

64 SCHMID ET RL.

findings could be classified into one of the four histological trend patterns (Table 2). All but one of the 24 patients with acute non-A, non-B hepatitis could be ranged into one of the four patterns. In trends I and IV, all etiological forms of acute hepatitis were represented al-

Fig. 3. Trend 111: (a) Bile duct lesion. Abnormal bile duct lying in a dense lymphocytic infiltrate (H&E, X200). (b) Bile duct with multilayered hydropic epithelium with typical segmental in- volvement of the duct wall (H&E, X320). (c) Dense lymphocytic portal infiltration with lymph follicle and germinal center (H&E, X80).

most evenly, whereas in trends I1 and 111, there was a clear predominancy of acute non-A, non- B hepatitis; 14 cases (61%) showed findings typical for trend 11, and seven (30%) for trend 111.

In an occasional patient, the pattern of trend 111 was associated with the additional finding of a diffuse proliferation of reticulo-endothelial cells, marked acidophilic degeneration of he- patocytes and frequent fragmentation of Coun- cilman-bodies.

Outcome as related to the initial histological trend The outcome in the 24 patients with acute non- A, non-B hepatitis was as follows: 12 patients (50%) showed clinical healing, ten (42%) showed transition to chronic liver disease, and two (8%) a fluctuating course with multiple relapses, documented in all chronic and re-

Fig. 4. Trend I V Acute viral hepatitis with centri- lobular necrosis. Mild inflammatory reaction (H&E, x 200).

. ACUTE HEPATITIS NON-A, NON-B 65

Table 3 Outcome as related to initial histological trends I1 and I11 (21 patients with acute hepatitis NANB)

Initial No. of Multiple relapses Chronic histological trend patients Healing (6 months) liver disease

Trend I1 14 9 (64%) 1 ( 7%) 4 (29%)

Trend I11 7 - 1(14%) 6 (86%)

lapsing forms with one or more follow-up liver biopsies. Repeat liver biopsy of the ten patients with chronic liver disease revealed chronic active hepatitis in nine (evolving into cirrhosis in three), and chronic persistent hepatitis in one patient. When correlated with the initial histo- logical trend, only four out of 14 patients (29%) with confluent necrosis (trend 11) revealed tran- sition to chropic liver disease, as opposed to trend 111 (bile duct lesion) in which six out of seven (86%) showed transition to chronic liver disease (Table 3).

As an example of the wide spectrum of heterogeneous histological lesions, a short case report will be given of a 17-year-old boy with acute hepatitis non-A, non-B. A first biopsy was taken 4 months after the onset of an acute, icteric hepatitis. There was extensive post- necrotic scarring and partial nodular regenera- tion; complete cirrhosis, however, was not present. There was an extreme inflammatory reaction and formation of rosettes was con-

spicuous. The most impressive feature in this case was the appearance of large numbers of bizarre, multinucleated parenchymal giant cells. Two months later, an identical picture of giant cell hepatitis was found in association with a highly active chronic inflammation, accompanied by regeneration (Fig. 5) . Omlapa- roscopy, the liver surface was finely granular but there was no evidence of macronodular postnecrotic cirrhosis, as might have been ex- pected.

Discussion More than a third of all patients with acute hepatitis A, B, and non-A, non-B (24 out of 54) showed biopsy findings characterized by con- fluent necrosis, notably bridging necrosis. Such an unusually high percentage may be explained by the admission policy of our hospital. Only a few patients with acute hepatitis are hospital- ized, of whom most have a severe and pro-

Fig. 5. (a+ b) Acute hepatitis non-A, non-B with confluent necrosis. Picture of giant cell hepatitis (HBrE, X200).

66 SCHMID ET AL.

tracted clinical course. We are unable, however, to find a valid explanation for the high ratio of acute hepatitis non-A, non-B within the group showing confluent necrosis (trend 11). The clinical course and particularly the relatively good prognosis do not differ from acute hepa- titis with bridging hepatic necrosis in hepatitis B infection (7).

Bile duct lesions with abnormal duct epithe- lium (trend 111) were found almost exclusively in patients with hepatitis non-A, non-B. In hepatitis A, bile duct lesions were not seen at all; the lesion was present, however, in one patient with hepatitis B. Histologically, the bile duct lesions always appeared in association with a dense lymphocytic infiltration, often forming lymph follicles with germinal centers. Within the lymphocytic infiltrate, many plasma cells could be discerned. The inflammation may extend into the adjacent parenchyma and slight piecemeal necrosis may be seen. In our ex- perience, the finding of a very dense portal infiltration has served as a guide to look for typical bile duct lesions which may appear only after serial sectioning of the biopsy material.

In one of the seven patients with non-A, non- B hepatitis characterized by the pattern of trend 111, the initial biopsy was taken rather late in the course of the disease (90 days after peak ALT level). In the other six patients, however, the biopsies were taken between day 6 and 14 after the maximum transaminase level was reached. With one exception, all seven patients had a mild and anicteric course of acute hepatitis. One patient showed features of both confluent ne- crosis (trend 11) and of bile duct lesion (trend 111) in his initial biopsy. In two other patients, trend I1 in the initial biopsy was followed by features characteristic for trend I11 in the repeat biopsy. It thus seems unlikely that these two morphological features are caused by different etiological agents. All but one of our seven patients (86%) with bile duct lesion in non-A, non-B hepatitis eventually showed a chronic course. This is in agreement with the experience

of Christoffersen & Poulsen (8), who reported an increased incidence of chronic hepatitis among patients with hepatitis and bile duct lesions.

Few data are available in the literature on light microscopic findings in acute non-A, non- B hepatitis. To our knowledge, two groups of authors have stressed the occurrence of slight to moderate fatty change (2,9), in addition to the classical picture of viral hepatitis. Fatty change was also reported in two out of ten patients with chronic non-A, non-B hepatitis (10). In our series, slight fatty change was detected in six out of 24 cases of hepatitis non-A, non-B, a ratio not differing significantly from cases of acute hepatitis A or B, respectively. We were unable to find any correlation between the degree of fatty changes and other histological features.

In chimpanzees inoculated with non-A, non- B agents, a distinct increase in sinusoidal lining cells (1 1) and the presence of multinucleated giant hepatocytes has been described (12-15). A similar picture of marked proliferation of Kupffer cells associated with only minimal hepatocellular necrosis was observed in one of our non-A, non-B patients in whom the biopsy was obtained very early in the course of illness (day 5). Whether these features are typical for the very early phase of non-A, non-B infection in man, as it has been described in hemophiliacs following the administration of commercial factor VIII (9), remains a matter of specula- tion. In some other patients, the finding of diffuse reticulo-endothelial cell proliferation was associated with abundant acidophilic de- generation of hepatocytes and occasional frag- mented Councilman-bodies.

Conclusion There is no definitely specific light microscopic pattern that would permit an accurate etiologic diagnosis in the individual patient with acute viral hepatitis. It is noteworthy, however, that bile duct lesions as described by Poulsen &

ACUTE HEPATITIS NON-A, NON-8 67

Christoffersen have been found, with one single exception, in patients with hepatitis non-A, non-B only, suggesting that this lesion is of specific value for the histological recognition of hepatitis non-A, non-B.

Furthermore, our results indicate that the initial morphological features have prognostic significance. With one exception again, bile duct lesions were associated with a mild and anicteric course of acute hepatitis but with a high tendency to the development of chronic liver disease (86%).

Bile duct lesions are a rather rare finding in biopsies of patients with hepatitis of the differ- ent viral etiologies. If present, however, they have to be considered quite specific for hepatitis non-A, non-B. At the same time, they are of prognostic importance, indicating a high ten- dency to the development of chronic liver disease.

References FERENCI P, DRAGOSICS B, PAUSER G. Non- A/Non-B Hepatitisepidemie bei einer Plasma- pheresestelle. Verh Dtsch Ges Inn Med 1979: 85: 553. IWARSON S, LINDBERG J, LUNDIN P. Progression of hepatitis non-A, non-B to chronic active hepatitis. J Clin Paihol 1979: 32: 351. ROGGENDORF M, F R ~ S N E R G G, DEINHARDT F et al. Comparison of solid phase test systems for demonstrating antibodies against hepatitis A virus (anti-HAV). J Med Virol 1980: 5: 47. ROGGENDORF M, DEINHARDT F, FR~SNER G Get al. Immunoglobulin M antibodies to hepatitis B core antigen: Evaluation of enzyme immuno- assay for diagnosis of hepatitis B virus infection. J CIin Microbiol 1981: 13: 618.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

GUDAT F, BIANCHI L, SONNABEND W et al. Pattern of core and surface expression in liver tissue reflects state of specific immune response in hepatitis B. Lab Invest 1975: 32: 1. POULSEN H, CHRISTOFERSEN P. Abnormal bile duct epithelium in liver biopsies with histological signs of viral hepatitis. Acia Pathol Microbiol Scand 1969: 7 6 383. SCHMID M, PIROVINO M, ALTORFER J et al. Acute viral hepatitis B with bridging necrosis: a follow- up study. Liver 1981: 1: 222. CHRISTOFFERSEN P, POULSEN H, WINKLER K. Clinical findings in patients, with hepatitis and abnormal bile duct epithelium. Scand J Gasiro- enterol 1970: 5: 117. THOMAS H C, BAMBER M, TREPO C, MURRAY A, ARBORGH B A M, SCHEUER P J, KERNOFF P B A, SHERLOCK S. Manuscript in preparation. KNODELL R G, CONRAD M E, ISHAK K G. Development of chronic liver disease after acute non-A, non-B post-transfusion hepatitis. Gastro- enterology 1977: 7 2 902. POPPER H, DIENSTAG J L, FEINSTONE S M et al. The pathology of viral hepatitis in chimpanzees. Virchows Arch [Pathol Anai] 1980: 381 91. ALTER H J, HOLLAND P V, PURCELL R H et al. Transmissible agent in non-A, non-B hepatitis. Lancet 1978: i: 459. TABOR E, DRUCKER J A, HOOFNAGLE J H et al. Transmission of non-A, non-B hepatitis from man to chimpanzee. Lnncei 1978: i 463. TABOR E, APRIL M, SEEFF L B et al. Acute non-A, non-B hepatitis. Gastroenierology 1979: 76: 680. WYKE R J, THORNTON A, PORTMANN B et al. Transmission of non-A non-B hepatitis to chim- panzees by factor-IX concentrates after fatal complication in patients with chronic liver dis- ease. Lancer 1979: i: 520.

Address: Martin Schmid Depariment of Internal Medicine Siadispiial Waid CH~8037 Zurich Switzerland