- Email: [email protected]
A morphological study of the gross and light microscopic lesions of infectious anaemia in Atlantic salmon (Salmo salar)
Research in Veterinary Science 1991, 51, 215-222
A morphological study of the gross and light microscopic lesions of infectious anaemia in Atlantic salmon (Salmo salar) tO. EVENSEN, K. E. THORUD, National Veterinary Institute, PO Box 8156, Dep, N-0033 Oslo 1, Norway, Y. A. OLSEN, Norwegian College of Veterinary Medicine, PO Box 8146, Dep, N-0033 Oslo 1, Norway
Infectious anaemia in Atlantic salmon (Salmo salar) was studied by recording gross and light microscopic changes, and the development of lesions was studied in relation to haematocrit values. Gross lesions were characterised by ascites formation, congestion and enlargement of liver and spleen, congestion of the foregut and petechiae in the peritoneum. Histologically, lesions were demonstrated in the liver, being characterised by congestion in early stages (that is, haematocrit values around 25), dilatation of the sinusoids, and in later stages (haematocrit values 25 to 15) formation of blood-filled spaces bearing morphological resemblance to peliosis hepatis. At low haematocrit values (around 10), these changes comprised large areas of the liver parenchyma, that is, blood-filled areas coalesced, presenting islets of degenerate and necrotic hepatocytes. At this stage, haemorrhagic necroses were found. Spleen and kidney lesions were characterised by congestion. In the foregut, congestion and bleeding in lamina propria were observed. Liver lesions became more disseminated and severe with decreasing haematocrit values. Hypoxia due to anaemia alone cannot fully explain the development of the liver lesions. INFECTIOUS anaemia in Atlantic salmon (Salmo salar), also termed infectious salmon anaemia (ISA), is a transmissible disease of unknown aetiology (Thorud and Djupvik 1988). The gross pathological changes in terminal stages of the disease are characterised by extreme pallor of the gills, ascites, petechiae in visceral fat, and congestion of the liver and the spleen (Thorud 1989). Histologically, the most prominent lesions are observed in the liver, the changes being characterised as focal/zonal liver necroses (Thorud and Djupvik 1988, Thorud 1989). Haematological examination shows a marked reduction in the number of circulating erythrocytes and leucocytes, and increased numbers of immature erythrocytes (Thorud and Djupvik 1988). Recent investigations have shown that the disease can be transmitted by injection of liver homogenates from diseased fish, and also by cohabitation with diseased
fish under experimental conditions (Thorud and Djupvik 1988), indicating an infectious aetiology. The diagnosis of the disease is based on clinical signs, gross pathology and histological lesions. However, there are no previous descriptions of the major morphological changes at different stages of the disease. Therefore, this study was undertaken to study the gross and light microscopic lesions during the course of the disease. Materials and methods A total of 62 Atlantic salmon (postsmolts) of the same age group and of the same descent were included in the investigation, which was carried out during an outbreak of ISA in a commercial fish farm on the west coast of Norway. The fish were kept in two different tanks (5 m in diameter) in sea water. Organ samples were obtained from fish at different stages of the disease, as evaluated by haematocrit values ranging from 44 to under 5. Dead fish (four) not showing any significant post mortem changes were also included in the study. Ten fish of the same descent, kept in a separate tank, and without any clinical signs of disease were included as controls. Each fish was killed by a blow on the head, blood samples were collected immediately from the caudal vein using evacuated blood collecting tubes (Venoject) containing sodium heparinate, and kept in ice water (approximately 0°C) until examination. Haematocrit values were determined by using heparinised microcapillary tubes, and centrifuged (12,000 g) for five minutes in a Compure M 1100 centrifuge. Gross lesions were evaluated by one person only, immediately after euthanasia, and findings were recorded for individual fish. Samples of gill, oesophagus, heart, pyloric caeca with pancreatic tissues, intestine, liver, spleen, kidney and musculature, were fixed by immersion in 10 per cent phosphate buffered formalin, processed routinely, embedded in paraffin wax, sectioned at 5 to 6/zm, and stained with haematoxylin and eosin (H&E).
215
O. Evensen, K. E. Thorud, Y. A. Olsen
216 Results
Gross pathology The gross pathological changes were classified into five different groups, according to the major findings, that is, ascites, congestion and enlargement of the liver, congestion and enlargement of the spleen, peritoneal petechiae and congested mucosa of the foregut (Diagram 1). The changes were related to heamatocrit values. No gross lesions were observed in fish with haematocrit values above 30. Ascites was observed in approximately 63 per cent of the fish examined with a haernatocrit below 30, though most frequently in those with a haematocrit below 20, and there seemed to be an inverse relationship between ascites formation and haematocrit values (Diagram 1). Diffuse congestion and enlargement of the liver were more frequently observed than spleen enlargement: 80 per cent and 50 per cent, respectively, of all fish examined, showing such changes (Diagram 1). Spleen changes were most frequently associated with low haematocrit values (under 15).
29
Gut congestion
I 92
Peritoneal petechiae
Spleen lesions
I
I
I
,
I
I
!
I
!
I
I
86
I
l
I
too
Liver lesions
lOO 67 40
Ascites
| 1-5
I 6-10
I 11-15
[ 16-20
I 21-25
l 26-30
Haematocrit D I A G R A M 1: Proportions o f t h e main m a c r o s c o p i c lesions observed, divided into six d i f f e r e n t h a e m a t o c r i t ranges. The n u m b e r above each c o l u m n indicates t h e percentage of individuals w i t h m a c r o s c o p i c lesions (within each h a e m a t o c r i t range)
Peritoneal petechiae (both visceral and parietal) were observed in approximately 50 per cent of the fish examined, being most frequently found in low haematocrit groups (Diagram 1). At a haematocrit below 15, approximately 83 per cent of the fish examined exhibited peritoneal petechiae, the severity increasing with decreasing haematocrit. Congestion in the intestinal mucosa was observed in approximately 11 per cent (seven out of 62) of the fish examined. Such lesions were seen exclusively in groups with haematocrit values between 11 and 30 with more than two-thirds (five of seven) of the total being observed in fish with values between I 1 and 20. Fish with values below 10, did not have these changes.
Light microscopic findings The liver changes developed gradually. At haematocrit values above 25, changes were characterised by moderate congestion and a tendency towards dilatation of the sinusoids (Fig 1). At lower haematocrit values (25 to 15), the main lesions comprised marked congestion and dilated sinusoids (Fig 2). The hepatocytic tubules became disorganised, and small cavities filled with erythrocytes were present (Figs 2 and 3). In these areas, hepatocytes appeared to be atrophic with consequent thinning of the hepatocytic tubules (Figs 2 and 3), and islets of hepatocytes were found in the congested areas (Fig 2). The blood-filled cavities did not seem to have any endothelial lining (Fig 3). Hepatocellular degeneration and necrosis was also observed in these areas (Fig 4). At low haematocrit values (10), these lesions involved large areas of the liver parenchyma. Large blood-filled areas were formed by coalescence of the above-mentioned blood-filled cavities. Concomitant hepatocyte degeneration and necrosis were found, such changes being characterised as haemorrhagic necroses (Figs 5 and 6). In dead fish, confluent necrosis appeared that gave these changes a bridging appearance (Fig 7). Thrombi were observed in venules and small veins (Fig 8). Swollen and degenerate endothelial cells were also present in small and medium-sized veins (Fig 8). A moderate oedema was found around blood vessels and biliary ducts (Fig 9). Inflammatory changes were not observed at any stage of the disease. The splenic lesions were characterised by congestion and increased erythrophagocytosis (Fig 10). The foregut showed pronounced congestion, sometimes with diffuse bleeding in the lamina propria (Fig 11). Epithelial cells in crypts were found to be intact, while degenerate cells were present in the midpart of the villi, and there was complete shedding of epithelial cells on the top of the villi (Fig 1 I). The kidneys exhibited congestion of the sinusoids, and individual haematopoietic cells showed degeneration and necrosis (Fig 12).
Morphological lesions in infectious salmon anaemia
217
FIG 1 : Liver. Congestion in small venules and sinusoids. Small vacuoles in hepatocytes. Haematocrit 28. H&E x 225
FIG 2: Liver. Marked congestion with dilated sinusoids and formation of blood-filled cavities. Atrophic hepatocytes and thinner tubules than normal (large arrow). Islets of hepatocytes are observed in the congested areas (small arrow). Haematocrit 18. H&E X 90
FIG 3: Liver. Detail from Fig 2 shewing blood-filled spaces apparently without endothelial lining (arrows). Haematocrit 18. H&E x 3 6 0
FIG 4: Liver. Islets of hepatocytes in a congested area. Degenerate and necrotic liver cells (arrows). Haematocrit 16. H&E x 360
218
O. Evensen, K. E. Thorud, Y. A. Olsen
FIG 5: Liver. A large area w i t h hepatocellular degeneration and necrosis, and with bleeding in the same area (haemorrhagic necrosis). Haematocrit 5-6. H&E × 9 0
FIG 7: Liver. Confluent necrosis with bridging appearance. N Area with necrosis. Terminal stage, haematocrit value not determined. H&E x 36
FIG 6: Liver. Detail from Fig 5 showing degenerate and necrotic hepatocytes. H&E x 3 6 0
FIG 8: Liver. Intravascular thrombi are present in small venules (asterisk). Swollen and degenerate endothelial cells are also observed (arrows). Haematocrit 18. H&E × 360
Morphological lesions in infectious salmon anaemia
FIG 9: Liver. Oedema around vessels and biliary ducts. Note the swollen endothelial cells in the arteriole. An area in the vein is without endothelial lining (arrow). Haematocrit 28. H&E x 225
FIG 11: Foregut. Congestion and bleeding in the lamina propria (asterisk). Epithelial cells in the crypts are preserved, while degenerate and necrotic cells are in the midpart of the villi, The epithelial cells shed off from the top of villi. Haematocrit 21. H&E x 3 6 0
219
FIG 10: Spleen. Marked congestion. Increased erythrophagocytosis (arrows), Haematocrit 12. H&E x 360
FIG 12: Kidney. A moderate sinusoidal congestion. Single cell degeneration and necrosis is observed in the haematopoietic tissue (arrow). Haematocrit 12. H&E × 360
220
O. Evensen, K. E. Thorud, Y. A. Olsen
No lesions were observed in the other organs examined. Discussion The present study shows that infectious anaemia in Atlantic salmon is characterised by a progressive anaemia as demonstrated by lowered haematocrit, and concomitant development of ascites, congestion of the foregut in early stages, and enlargement and congestion of the liver and the spleen. Histologically, the liver changes are the most prominent findings, characterised by congestion, degeneration of hepatocytes and the presence of haemorrhagic necroses in the late stages of the disease. The dissemination and severity of liver lesions increased gradually with decreasing haematocrit values. Descriptions of liver histology in teleost fish species are limited, although the anatomical arrangement of hepatocytic tubules in the liver of salmonids is well known (Ellis et al 1989). However, the functional units in teleost livers are poorly understood and, traditionally, terms used to describe the morphology of liver lesions in salmonids are derived from the mammalian liver lobule (Hinton and Pool 1976, Hinton et al 1984). Recent studies of functional units have revealed that lobules are absent in the liver of rainbow trout (Oncorhynchus mykiss), and that hepatocytes are arranged in branching and anastomosing tubules with small lumina (Hampton et al 1985). Furthermore, it has been difficult, on a morphological basis, to differentiate between branches of portal veins and central veins (terminal hepatic veins), and terms such as centrilobular, periportal and zonal are therefore not recommended when describing liver diseases in teleosts (Hampton et al 1988). The present investigation on infectious anaemia in Atlantic salmon, shows that the key lesions are confined to the parenchyma and vascular system of the liver. The authors have not been able to find adequate terms to describe the lesions observed, and the 'bridging' appearance might indicate the presence of bridging necroses linking central vein to central vein as observed in cardiac disease and shock (Lefkowitch and Mendez 1986), infectious hepatitis (Scheuer 1987) and also in chronic active hepatitis in humans (Bianchi et al 1987). The observed relationship between increasing development of the liver lesions with decreasing hematocrit values, might indicate that the observed necroses in the liver parenchyma should be termed 'anaemic necroses'. Studies in progress (K. "lZhorud, unpublished observation) have shown that plasma alanine-aminotransferase (ALAT) activity is inversely related to haematocrit values (that is, low haematocrit-high ALATactivities), a possible indication of hypoxic liver
cell damage. However, in erythrocytic inclusion body syndrome virus infection, a disease causing severe anaemia in salmonid parr and postsmolts, haematocrit values below 10 have been frequently observed, but without liver lesions corresponding to those seen in ISA being found (Holt and Piacentini 1989, Lunder et al 1990). Furthermore, the presence of intravascular tbrombi might contribute to the development of a severe hypoxia in the liver parenchyma, and 'ischaemic necroses' might be a more appropriate term. Based on these observations it seems clear that the pathogenesis of the liver lesions in ISA is not obvious, and further studies are needed to elucidate the possible relationship between anaemia and liver lesions in this disease. Sinusoidal dilatation seems to precede the development of small blood-filled spaces in the liver parenchyma in ISA. The latter changes have morphological characteristics in common with hepatic sinusoidal abnormalities described as 'peliosis hepatis' or peliotic-like lesions (Bras and Brandt 1987). These changes are considered to be nonspecific, and have been described in several conditions such as tuberculosis, neoplasms (different carcinomas), and in association with steroid therapy (androgenic/anabolic steroids) and infections with human immunodeficiency virus in humans (Zafrani et al 1984, Gordon et al 1986). The pathogenesis of peliosis or peliotic-like lesions is a matter of debate (Scoazec et al 1988) and several hypotheses have been proposed. First, hepatocellular necroses with disruption of the reticulin framework have been suggested (Barbadin and Scheuer 1984); secondly, congestion of the sinusoids resulting from blocked outflow of blood (Degott et al 1978); and, thirdly, a direct lesion of the sinusoidal barrier (Zafrani et al 1984). At present, opinion is in favour of the lastmentioned hypothesis (Zafrani et al 1984, Scoazec et al 1988). Lesions of the sinusoidal barrier, and especially of the endothelial cells in sinusoids are difficult to interpret morphologically by light microscopy. However, it is well recognised that endothelial cell injury constitutes the major aetiological factor in thrombus formation (Cotran et al 1989). The observed endothelial changes in small venules and arteries together with thrombi formation in small venules, might therefore indicate that endothelial cell damage was indeed ~present in the sinusoidal lining although not recognisable by light microscopy. Regarding the second theory (Degott et al 1978), thrombus formation in small and medium-sized hepatic venules and veins will contribute to impaired flow of blood from the sinusoids, ischaemia and thereby to the sinusoidal abnormalities in ISA. However, thrombus formation is considered of minor importance in the pathogenesis of the liver lesions since those changes are rarely observed.
Morphological lesions in infectious salmon anaemia Ascites is a prominent sign in ISA. It was found in over 60 per cent of the fish examined in the present study, being more frequently present at low haematocrit values. Ascites frequently develops in association With liver disease, its formation being, in general, dependent upon the colloid osmotic pressure and portal venous pressure (Sherlock 1987). However, in many fishes the effective colloid osmotic pressure across the capillary membrane is relatively slight because of high extravascular protein concentration due to high vascular permeability of blood proteins (Hargens et al t974). Therefore, the ascites seen in Atlantic salmon with ISA cannot, it seems, be explained by the observed reduction in plasma protein concentration. Increased portal venous pressure (portal hypertension) plays a part in the development of ascites (Sherlock 1987). Portal hypertension is classified as pre-, intra- and post hepatic; intrahepatic hypertension again being subdivided into presinusoidal, sinusoidal and postsinusoidal (Groszmann and Atterbury 1982). In this context, the observed peliotic-like abnormalities and the thrombi observed in venules and veins in ISA, might be of some relevance. Peliosis hepatis is considered to be a possible cause of intrahepatic, sinusoidal portal hypertension, although its importance is debated, being probably alone insufficient to cause portal hypertension (Groszmann and Atterbury 1982). Intravascular thrombi give rise to intrahepatic, postsinusoidal portal hypertension (McLean 1970). In ISA, thrombi formation is rarely observed and therefore seems to be of limited importance in ascites formation. However, the possible causes outlined here are related only to changes in vascular resistance, and changes in blood flow have not been considered, the main reason being the fact that these parameters are difficult to evaluate (Groszmann and Atterbury 1982). In this context, the possible involvement of insufficient cardiac function is also difficult to elucidate, especially since no lesions were observed in the myocardium in fish with ISA. In conclusion, the present study shows that Atlantic salmon with ISA develop a progressive anaemia, the major lesions being confined to the liver parenchyma, in which haemorrhagic necroses occur in the later stages of the disease. Ascites is a prominent finding, as is congestion of the spleen with concomitant erythrophagocytosis. However, further studies including electron microscopy should be pursued to elucidate the pathogenetic mechanisms involved in the development of the liver lesions. Acknowledgements Financial support was provided by the Agricultural Research Council of Norway, grant number 221.701. The authors thank Dr B. Bratberg for valuable dis-
221
cussions, and the technical assistance of Ms Birgit ROe is acknowledged. References BARBADIN, K. A. & SCHEUER, P. J. (1984) Endothelial cell changes in acute hepatitis: A light and electron microscope study. Journal of Pathology 144, 213-220 BIANCHI, L., SPICHTIN, H. P. & GUDAT, F. (1987) Chronic hepatitis. In Pathology of the Liver. 2nd edn. Eds R. N. M. MacSween, P. P. Anthony and P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 310-341 BRAS, G. & BRANDT, K. H. (1987) Vascular disorders. In Pathology of the Liver. 2nd edn. Eds R. N. M. MacSween, P. P. Anthony & P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 478-502 COTRAN, R. S., KUMAR, V. & ROBBINS, S. L. (1989) Thrombogenesis. In Pathologic Basis of Disease. 4th edn. Eds R.S. Cotran, V. Kumar and S. L. Robbins. Philadelphia, Saunders. pp 99-105 DEGOTT, C., RUEFF, B., KREIS, H., DUBOUST, A., POTET, F. & BENHAMOU, J. P. (1978) Peliosis hepatis in recipients of renal transplants. Gut 19, 748-753 ELLIS, A. E., ROBERTS, R. J. & TYTLER, P. (1989) The anatomy and physiology of teleosts: The liver. Fish Pathology. 2nd edn. London, BaiUi~reTindall. pp 35-37 GORDON, S. C., REDDY, K. R., GOULD, E. E., McFADDEN, R., O'BRIEN, C., DE MEDINA, M., JEFFERS, L. J. & SCHIFF, E. R. (1986) The spectrum of liver disease in acquired immunodeficiency syndrome. Journal of Hepatology 2, 475-484 GROSZMANN, R. J. & ATTERBURY, C. E. (1982) The pathophysiology of portal hypertension: A basis for classification. Seminars in Liver Diseases 2, 177-186 HAMPTON, J. A., McCUSKEY, P. A., McCUSKEY, R. S. & HINTON, D. E. (1985) Functional units in rainbow trout (Salmo gairdnert) liver: I. Arrangement and histochemical properties of hepatocytes. Anatomical Record 213, 166-175 HAMPTON, J. A., LANTZ, R. C., GOLDBLATT, P. J., LAUREN, D. J. & HINTON, D. E. (1988) Functional units in rainbow trout (Salmo gairdneri, Richardson) liver: II. The biliary system. ,4 natomical Record 221, 619-634 HARGENS, A. R., MILLARD, R. W. & JOHANSEN, K. (1974) High capillary permeability in fishes. Comparative Biochemical Physiology 48A, 675-680 HINTON, D. E. & POOL, C. R. (1976) Ultrastructure of the liver in channel catfish (Ichthalurus punctatus Rafinesque). Journal of Fish Biology 8, 209-219 HINTON, D. E., WALKER, E. R., PINKSTAFF, C. A. & ZUCHELKOWSKI, E. M. (1984) Morphological survey of teleost organs important in carcinogenesis with attention to fixation. National Cancer Institute Monograph 54, US Department of Health and Human Services, Bethesda, USA. pp 291 320 HOLT, R. A. & PIACENTINI, S. (1989) Informational report on erythrocytic inclusion body syndrome. Oregon Department of Fish and Wildlife/Oregon State University, Corvallis, Oregon. pp 1-16 LEFKOWITCH, J. H. & MENDEZ, L. (1986) Morphologic features of hepatic injury in cardiac disease and shock. Journal of Hepatology 2, 313-327 LUNDER, T., THORUD, K., POPPE, T. T. & ROHOVEC, J. S. (1990) Particles similar to the virus of erythrocytic inclusion body syndrome, EIBS, detected in Atlantic salmon (Salmo salar) in Norway. Bulletin of the European Association of Fish Pathologists 10, 21-23 McLEAN, E. K. (1970) The toxic actions of pyrrolizidine (Senecio) alkaloids. Pharmacological Review 22, 429-483 SHERLOCK, S. (1987) Pathophysiology of the liver: Ascites. In Pathology of the Liver. 2nd edn. Eds R. N. M. MacSween, P. P. Anthony and P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 49-50
222
O. Evensen, K. E. Thorud, Y. A. Olsen
SCHEUER, P. J. (1987) Viral hepatitis. In Pathology of the Liver. 2nd edn. EdsR. N. M. MacSween, P. P. Anthony & P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 202-210 SCOAZEC, J.-Y., MARCH, C., GIRARD, P.-M., HOUTMANN, J., DURAND-SCHNEIDER, A.-M., SAIMOT, A.-G., BENHAMOU, J.-P. & FELDMANN, G. (1988) Peliosis hepatis and sinusoidal dilatation during infection by the human immunodeficiency virus (HIV). An ultrastructural study. American Journal of Pathology 131, 38-47 THORUD, K. (1989) Infectious salmon anaemia. An overview. Norsk Veterinaer Tidsskrift 101,517-521
THORUD, K. & DJUPVIK, H. O. (1988) Infectious anaemia in Atlantic salmon (Salmo salar L). Bulletin of the European Association ofFish Pathologists 8, 109-111 ZAFRANI, E. S., CAZIER, A., BAUDELOT, A. M. & FELDMANN, G. (1984) Ultrastruc~ural lesions of the liver in human peliosis: A report of 12 cases. American Journal of Pathology 114, 349-359
Received November 23, 1990 Accepted May 7, 1991
A morphological study of the gross and light microscopic lesions of infectious anaemia in Atlantic salmon (Salmo salar) tO. EVENSEN, K. E. THORUD, National Veterinary Institute, PO Box 8156, Dep, N-0033 Oslo 1, Norway, Y. A. OLSEN, Norwegian College of Veterinary Medicine, PO Box 8146, Dep, N-0033 Oslo 1, Norway
Infectious anaemia in Atlantic salmon (Salmo salar) was studied by recording gross and light microscopic changes, and the development of lesions was studied in relation to haematocrit values. Gross lesions were characterised by ascites formation, congestion and enlargement of liver and spleen, congestion of the foregut and petechiae in the peritoneum. Histologically, lesions were demonstrated in the liver, being characterised by congestion in early stages (that is, haematocrit values around 25), dilatation of the sinusoids, and in later stages (haematocrit values 25 to 15) formation of blood-filled spaces bearing morphological resemblance to peliosis hepatis. At low haematocrit values (around 10), these changes comprised large areas of the liver parenchyma, that is, blood-filled areas coalesced, presenting islets of degenerate and necrotic hepatocytes. At this stage, haemorrhagic necroses were found. Spleen and kidney lesions were characterised by congestion. In the foregut, congestion and bleeding in lamina propria were observed. Liver lesions became more disseminated and severe with decreasing haematocrit values. Hypoxia due to anaemia alone cannot fully explain the development of the liver lesions. INFECTIOUS anaemia in Atlantic salmon (Salmo salar), also termed infectious salmon anaemia (ISA), is a transmissible disease of unknown aetiology (Thorud and Djupvik 1988). The gross pathological changes in terminal stages of the disease are characterised by extreme pallor of the gills, ascites, petechiae in visceral fat, and congestion of the liver and the spleen (Thorud 1989). Histologically, the most prominent lesions are observed in the liver, the changes being characterised as focal/zonal liver necroses (Thorud and Djupvik 1988, Thorud 1989). Haematological examination shows a marked reduction in the number of circulating erythrocytes and leucocytes, and increased numbers of immature erythrocytes (Thorud and Djupvik 1988). Recent investigations have shown that the disease can be transmitted by injection of liver homogenates from diseased fish, and also by cohabitation with diseased
fish under experimental conditions (Thorud and Djupvik 1988), indicating an infectious aetiology. The diagnosis of the disease is based on clinical signs, gross pathology and histological lesions. However, there are no previous descriptions of the major morphological changes at different stages of the disease. Therefore, this study was undertaken to study the gross and light microscopic lesions during the course of the disease. Materials and methods A total of 62 Atlantic salmon (postsmolts) of the same age group and of the same descent were included in the investigation, which was carried out during an outbreak of ISA in a commercial fish farm on the west coast of Norway. The fish were kept in two different tanks (5 m in diameter) in sea water. Organ samples were obtained from fish at different stages of the disease, as evaluated by haematocrit values ranging from 44 to under 5. Dead fish (four) not showing any significant post mortem changes were also included in the study. Ten fish of the same descent, kept in a separate tank, and without any clinical signs of disease were included as controls. Each fish was killed by a blow on the head, blood samples were collected immediately from the caudal vein using evacuated blood collecting tubes (Venoject) containing sodium heparinate, and kept in ice water (approximately 0°C) until examination. Haematocrit values were determined by using heparinised microcapillary tubes, and centrifuged (12,000 g) for five minutes in a Compure M 1100 centrifuge. Gross lesions were evaluated by one person only, immediately after euthanasia, and findings were recorded for individual fish. Samples of gill, oesophagus, heart, pyloric caeca with pancreatic tissues, intestine, liver, spleen, kidney and musculature, were fixed by immersion in 10 per cent phosphate buffered formalin, processed routinely, embedded in paraffin wax, sectioned at 5 to 6/zm, and stained with haematoxylin and eosin (H&E).
215
O. Evensen, K. E. Thorud, Y. A. Olsen
216 Results
Gross pathology The gross pathological changes were classified into five different groups, according to the major findings, that is, ascites, congestion and enlargement of the liver, congestion and enlargement of the spleen, peritoneal petechiae and congested mucosa of the foregut (Diagram 1). The changes were related to heamatocrit values. No gross lesions were observed in fish with haematocrit values above 30. Ascites was observed in approximately 63 per cent of the fish examined with a haernatocrit below 30, though most frequently in those with a haematocrit below 20, and there seemed to be an inverse relationship between ascites formation and haematocrit values (Diagram 1). Diffuse congestion and enlargement of the liver were more frequently observed than spleen enlargement: 80 per cent and 50 per cent, respectively, of all fish examined, showing such changes (Diagram 1). Spleen changes were most frequently associated with low haematocrit values (under 15).
29
Gut congestion
I 92
Peritoneal petechiae
Spleen lesions
I
I
I
,
I
I
!
I
!
I
I
86
I
l
I
too
Liver lesions
lOO 67 40
Ascites
| 1-5
I 6-10
I 11-15
[ 16-20
I 21-25
l 26-30
Haematocrit D I A G R A M 1: Proportions o f t h e main m a c r o s c o p i c lesions observed, divided into six d i f f e r e n t h a e m a t o c r i t ranges. The n u m b e r above each c o l u m n indicates t h e percentage of individuals w i t h m a c r o s c o p i c lesions (within each h a e m a t o c r i t range)
Peritoneal petechiae (both visceral and parietal) were observed in approximately 50 per cent of the fish examined, being most frequently found in low haematocrit groups (Diagram 1). At a haematocrit below 15, approximately 83 per cent of the fish examined exhibited peritoneal petechiae, the severity increasing with decreasing haematocrit. Congestion in the intestinal mucosa was observed in approximately 11 per cent (seven out of 62) of the fish examined. Such lesions were seen exclusively in groups with haematocrit values between 11 and 30 with more than two-thirds (five of seven) of the total being observed in fish with values between I 1 and 20. Fish with values below 10, did not have these changes.
Light microscopic findings The liver changes developed gradually. At haematocrit values above 25, changes were characterised by moderate congestion and a tendency towards dilatation of the sinusoids (Fig 1). At lower haematocrit values (25 to 15), the main lesions comprised marked congestion and dilated sinusoids (Fig 2). The hepatocytic tubules became disorganised, and small cavities filled with erythrocytes were present (Figs 2 and 3). In these areas, hepatocytes appeared to be atrophic with consequent thinning of the hepatocytic tubules (Figs 2 and 3), and islets of hepatocytes were found in the congested areas (Fig 2). The blood-filled cavities did not seem to have any endothelial lining (Fig 3). Hepatocellular degeneration and necrosis was also observed in these areas (Fig 4). At low haematocrit values (10), these lesions involved large areas of the liver parenchyma. Large blood-filled areas were formed by coalescence of the above-mentioned blood-filled cavities. Concomitant hepatocyte degeneration and necrosis were found, such changes being characterised as haemorrhagic necroses (Figs 5 and 6). In dead fish, confluent necrosis appeared that gave these changes a bridging appearance (Fig 7). Thrombi were observed in venules and small veins (Fig 8). Swollen and degenerate endothelial cells were also present in small and medium-sized veins (Fig 8). A moderate oedema was found around blood vessels and biliary ducts (Fig 9). Inflammatory changes were not observed at any stage of the disease. The splenic lesions were characterised by congestion and increased erythrophagocytosis (Fig 10). The foregut showed pronounced congestion, sometimes with diffuse bleeding in the lamina propria (Fig 11). Epithelial cells in crypts were found to be intact, while degenerate cells were present in the midpart of the villi, and there was complete shedding of epithelial cells on the top of the villi (Fig 1 I). The kidneys exhibited congestion of the sinusoids, and individual haematopoietic cells showed degeneration and necrosis (Fig 12).
Morphological lesions in infectious salmon anaemia
217
FIG 1 : Liver. Congestion in small venules and sinusoids. Small vacuoles in hepatocytes. Haematocrit 28. H&E x 225
FIG 2: Liver. Marked congestion with dilated sinusoids and formation of blood-filled cavities. Atrophic hepatocytes and thinner tubules than normal (large arrow). Islets of hepatocytes are observed in the congested areas (small arrow). Haematocrit 18. H&E X 90
FIG 3: Liver. Detail from Fig 2 shewing blood-filled spaces apparently without endothelial lining (arrows). Haematocrit 18. H&E x 3 6 0
FIG 4: Liver. Islets of hepatocytes in a congested area. Degenerate and necrotic liver cells (arrows). Haematocrit 16. H&E x 360
218
O. Evensen, K. E. Thorud, Y. A. Olsen
FIG 5: Liver. A large area w i t h hepatocellular degeneration and necrosis, and with bleeding in the same area (haemorrhagic necrosis). Haematocrit 5-6. H&E × 9 0
FIG 7: Liver. Confluent necrosis with bridging appearance. N Area with necrosis. Terminal stage, haematocrit value not determined. H&E x 36
FIG 6: Liver. Detail from Fig 5 showing degenerate and necrotic hepatocytes. H&E x 3 6 0
FIG 8: Liver. Intravascular thrombi are present in small venules (asterisk). Swollen and degenerate endothelial cells are also observed (arrows). Haematocrit 18. H&E × 360
Morphological lesions in infectious salmon anaemia
FIG 9: Liver. Oedema around vessels and biliary ducts. Note the swollen endothelial cells in the arteriole. An area in the vein is without endothelial lining (arrow). Haematocrit 28. H&E x 225
FIG 11: Foregut. Congestion and bleeding in the lamina propria (asterisk). Epithelial cells in the crypts are preserved, while degenerate and necrotic cells are in the midpart of the villi, The epithelial cells shed off from the top of villi. Haematocrit 21. H&E x 3 6 0
219
FIG 10: Spleen. Marked congestion. Increased erythrophagocytosis (arrows), Haematocrit 12. H&E x 360
FIG 12: Kidney. A moderate sinusoidal congestion. Single cell degeneration and necrosis is observed in the haematopoietic tissue (arrow). Haematocrit 12. H&E × 360
220
O. Evensen, K. E. Thorud, Y. A. Olsen
No lesions were observed in the other organs examined. Discussion The present study shows that infectious anaemia in Atlantic salmon is characterised by a progressive anaemia as demonstrated by lowered haematocrit, and concomitant development of ascites, congestion of the foregut in early stages, and enlargement and congestion of the liver and the spleen. Histologically, the liver changes are the most prominent findings, characterised by congestion, degeneration of hepatocytes and the presence of haemorrhagic necroses in the late stages of the disease. The dissemination and severity of liver lesions increased gradually with decreasing haematocrit values. Descriptions of liver histology in teleost fish species are limited, although the anatomical arrangement of hepatocytic tubules in the liver of salmonids is well known (Ellis et al 1989). However, the functional units in teleost livers are poorly understood and, traditionally, terms used to describe the morphology of liver lesions in salmonids are derived from the mammalian liver lobule (Hinton and Pool 1976, Hinton et al 1984). Recent studies of functional units have revealed that lobules are absent in the liver of rainbow trout (Oncorhynchus mykiss), and that hepatocytes are arranged in branching and anastomosing tubules with small lumina (Hampton et al 1985). Furthermore, it has been difficult, on a morphological basis, to differentiate between branches of portal veins and central veins (terminal hepatic veins), and terms such as centrilobular, periportal and zonal are therefore not recommended when describing liver diseases in teleosts (Hampton et al 1988). The present investigation on infectious anaemia in Atlantic salmon, shows that the key lesions are confined to the parenchyma and vascular system of the liver. The authors have not been able to find adequate terms to describe the lesions observed, and the 'bridging' appearance might indicate the presence of bridging necroses linking central vein to central vein as observed in cardiac disease and shock (Lefkowitch and Mendez 1986), infectious hepatitis (Scheuer 1987) and also in chronic active hepatitis in humans (Bianchi et al 1987). The observed relationship between increasing development of the liver lesions with decreasing hematocrit values, might indicate that the observed necroses in the liver parenchyma should be termed 'anaemic necroses'. Studies in progress (K. "lZhorud, unpublished observation) have shown that plasma alanine-aminotransferase (ALAT) activity is inversely related to haematocrit values (that is, low haematocrit-high ALATactivities), a possible indication of hypoxic liver
cell damage. However, in erythrocytic inclusion body syndrome virus infection, a disease causing severe anaemia in salmonid parr and postsmolts, haematocrit values below 10 have been frequently observed, but without liver lesions corresponding to those seen in ISA being found (Holt and Piacentini 1989, Lunder et al 1990). Furthermore, the presence of intravascular tbrombi might contribute to the development of a severe hypoxia in the liver parenchyma, and 'ischaemic necroses' might be a more appropriate term. Based on these observations it seems clear that the pathogenesis of the liver lesions in ISA is not obvious, and further studies are needed to elucidate the possible relationship between anaemia and liver lesions in this disease. Sinusoidal dilatation seems to precede the development of small blood-filled spaces in the liver parenchyma in ISA. The latter changes have morphological characteristics in common with hepatic sinusoidal abnormalities described as 'peliosis hepatis' or peliotic-like lesions (Bras and Brandt 1987). These changes are considered to be nonspecific, and have been described in several conditions such as tuberculosis, neoplasms (different carcinomas), and in association with steroid therapy (androgenic/anabolic steroids) and infections with human immunodeficiency virus in humans (Zafrani et al 1984, Gordon et al 1986). The pathogenesis of peliosis or peliotic-like lesions is a matter of debate (Scoazec et al 1988) and several hypotheses have been proposed. First, hepatocellular necroses with disruption of the reticulin framework have been suggested (Barbadin and Scheuer 1984); secondly, congestion of the sinusoids resulting from blocked outflow of blood (Degott et al 1978); and, thirdly, a direct lesion of the sinusoidal barrier (Zafrani et al 1984). At present, opinion is in favour of the lastmentioned hypothesis (Zafrani et al 1984, Scoazec et al 1988). Lesions of the sinusoidal barrier, and especially of the endothelial cells in sinusoids are difficult to interpret morphologically by light microscopy. However, it is well recognised that endothelial cell injury constitutes the major aetiological factor in thrombus formation (Cotran et al 1989). The observed endothelial changes in small venules and arteries together with thrombi formation in small venules, might therefore indicate that endothelial cell damage was indeed ~present in the sinusoidal lining although not recognisable by light microscopy. Regarding the second theory (Degott et al 1978), thrombus formation in small and medium-sized hepatic venules and veins will contribute to impaired flow of blood from the sinusoids, ischaemia and thereby to the sinusoidal abnormalities in ISA. However, thrombus formation is considered of minor importance in the pathogenesis of the liver lesions since those changes are rarely observed.
Morphological lesions in infectious salmon anaemia Ascites is a prominent sign in ISA. It was found in over 60 per cent of the fish examined in the present study, being more frequently present at low haematocrit values. Ascites frequently develops in association With liver disease, its formation being, in general, dependent upon the colloid osmotic pressure and portal venous pressure (Sherlock 1987). However, in many fishes the effective colloid osmotic pressure across the capillary membrane is relatively slight because of high extravascular protein concentration due to high vascular permeability of blood proteins (Hargens et al t974). Therefore, the ascites seen in Atlantic salmon with ISA cannot, it seems, be explained by the observed reduction in plasma protein concentration. Increased portal venous pressure (portal hypertension) plays a part in the development of ascites (Sherlock 1987). Portal hypertension is classified as pre-, intra- and post hepatic; intrahepatic hypertension again being subdivided into presinusoidal, sinusoidal and postsinusoidal (Groszmann and Atterbury 1982). In this context, the observed peliotic-like abnormalities and the thrombi observed in venules and veins in ISA, might be of some relevance. Peliosis hepatis is considered to be a possible cause of intrahepatic, sinusoidal portal hypertension, although its importance is debated, being probably alone insufficient to cause portal hypertension (Groszmann and Atterbury 1982). Intravascular thrombi give rise to intrahepatic, postsinusoidal portal hypertension (McLean 1970). In ISA, thrombi formation is rarely observed and therefore seems to be of limited importance in ascites formation. However, the possible causes outlined here are related only to changes in vascular resistance, and changes in blood flow have not been considered, the main reason being the fact that these parameters are difficult to evaluate (Groszmann and Atterbury 1982). In this context, the possible involvement of insufficient cardiac function is also difficult to elucidate, especially since no lesions were observed in the myocardium in fish with ISA. In conclusion, the present study shows that Atlantic salmon with ISA develop a progressive anaemia, the major lesions being confined to the liver parenchyma, in which haemorrhagic necroses occur in the later stages of the disease. Ascites is a prominent finding, as is congestion of the spleen with concomitant erythrophagocytosis. However, further studies including electron microscopy should be pursued to elucidate the pathogenetic mechanisms involved in the development of the liver lesions. Acknowledgements Financial support was provided by the Agricultural Research Council of Norway, grant number 221.701. The authors thank Dr B. Bratberg for valuable dis-
221
cussions, and the technical assistance of Ms Birgit ROe is acknowledged. References BARBADIN, K. A. & SCHEUER, P. J. (1984) Endothelial cell changes in acute hepatitis: A light and electron microscope study. Journal of Pathology 144, 213-220 BIANCHI, L., SPICHTIN, H. P. & GUDAT, F. (1987) Chronic hepatitis. In Pathology of the Liver. 2nd edn. Eds R. N. M. MacSween, P. P. Anthony and P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 310-341 BRAS, G. & BRANDT, K. H. (1987) Vascular disorders. In Pathology of the Liver. 2nd edn. Eds R. N. M. MacSween, P. P. Anthony & P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 478-502 COTRAN, R. S., KUMAR, V. & ROBBINS, S. L. (1989) Thrombogenesis. In Pathologic Basis of Disease. 4th edn. Eds R.S. Cotran, V. Kumar and S. L. Robbins. Philadelphia, Saunders. pp 99-105 DEGOTT, C., RUEFF, B., KREIS, H., DUBOUST, A., POTET, F. & BENHAMOU, J. P. (1978) Peliosis hepatis in recipients of renal transplants. Gut 19, 748-753 ELLIS, A. E., ROBERTS, R. J. & TYTLER, P. (1989) The anatomy and physiology of teleosts: The liver. Fish Pathology. 2nd edn. London, BaiUi~reTindall. pp 35-37 GORDON, S. C., REDDY, K. R., GOULD, E. E., McFADDEN, R., O'BRIEN, C., DE MEDINA, M., JEFFERS, L. J. & SCHIFF, E. R. (1986) The spectrum of liver disease in acquired immunodeficiency syndrome. Journal of Hepatology 2, 475-484 GROSZMANN, R. J. & ATTERBURY, C. E. (1982) The pathophysiology of portal hypertension: A basis for classification. Seminars in Liver Diseases 2, 177-186 HAMPTON, J. A., McCUSKEY, P. A., McCUSKEY, R. S. & HINTON, D. E. (1985) Functional units in rainbow trout (Salmo gairdnert) liver: I. Arrangement and histochemical properties of hepatocytes. Anatomical Record 213, 166-175 HAMPTON, J. A., LANTZ, R. C., GOLDBLATT, P. J., LAUREN, D. J. & HINTON, D. E. (1988) Functional units in rainbow trout (Salmo gairdneri, Richardson) liver: II. The biliary system. ,4 natomical Record 221, 619-634 HARGENS, A. R., MILLARD, R. W. & JOHANSEN, K. (1974) High capillary permeability in fishes. Comparative Biochemical Physiology 48A, 675-680 HINTON, D. E. & POOL, C. R. (1976) Ultrastructure of the liver in channel catfish (Ichthalurus punctatus Rafinesque). Journal of Fish Biology 8, 209-219 HINTON, D. E., WALKER, E. R., PINKSTAFF, C. A. & ZUCHELKOWSKI, E. M. (1984) Morphological survey of teleost organs important in carcinogenesis with attention to fixation. National Cancer Institute Monograph 54, US Department of Health and Human Services, Bethesda, USA. pp 291 320 HOLT, R. A. & PIACENTINI, S. (1989) Informational report on erythrocytic inclusion body syndrome. Oregon Department of Fish and Wildlife/Oregon State University, Corvallis, Oregon. pp 1-16 LEFKOWITCH, J. H. & MENDEZ, L. (1986) Morphologic features of hepatic injury in cardiac disease and shock. Journal of Hepatology 2, 313-327 LUNDER, T., THORUD, K., POPPE, T. T. & ROHOVEC, J. S. (1990) Particles similar to the virus of erythrocytic inclusion body syndrome, EIBS, detected in Atlantic salmon (Salmo salar) in Norway. Bulletin of the European Association of Fish Pathologists 10, 21-23 McLEAN, E. K. (1970) The toxic actions of pyrrolizidine (Senecio) alkaloids. Pharmacological Review 22, 429-483 SHERLOCK, S. (1987) Pathophysiology of the liver: Ascites. In Pathology of the Liver. 2nd edn. Eds R. N. M. MacSween, P. P. Anthony and P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 49-50
222
O. Evensen, K. E. Thorud, Y. A. Olsen
SCHEUER, P. J. (1987) Viral hepatitis. In Pathology of the Liver. 2nd edn. EdsR. N. M. MacSween, P. P. Anthony & P. J. Scheuer. Edinburgh, Churchill Livingstone. pp 202-210 SCOAZEC, J.-Y., MARCH, C., GIRARD, P.-M., HOUTMANN, J., DURAND-SCHNEIDER, A.-M., SAIMOT, A.-G., BENHAMOU, J.-P. & FELDMANN, G. (1988) Peliosis hepatis and sinusoidal dilatation during infection by the human immunodeficiency virus (HIV). An ultrastructural study. American Journal of Pathology 131, 38-47 THORUD, K. (1989) Infectious salmon anaemia. An overview. Norsk Veterinaer Tidsskrift 101,517-521
THORUD, K. & DJUPVIK, H. O. (1988) Infectious anaemia in Atlantic salmon (Salmo salar L). Bulletin of the European Association ofFish Pathologists 8, 109-111 ZAFRANI, E. S., CAZIER, A., BAUDELOT, A. M. & FELDMANN, G. (1984) Ultrastruc~ural lesions of the liver in human peliosis: A report of 12 cases. American Journal of Pathology 114, 349-359
Received November 23, 1990 Accepted May 7, 1991