Jun 27, 1975 - fluid and paraffin sections were stained with haematoxylin and eosin, aldehyde-fuichsin,. Glomori's reticulin stain, elastic-Van Gieson and ...
Br. J. exp. Path. (1975) 56, 530
PANCREATIC ISCHAEMIA; SENSITIVITY AND REVERSIBILITY OF THE CHANGES D. N. SLATER, D. BARDSLEY, Y. MANGNALL, A. SMIYTHE AND M. FOX From the Transplant Unit, Royal Hospital, Sheffield, Division of Pathology, University of Sheffield Received for publication June 27, 1975
Summary.-In the rat, pancreatic clamping producing warm ischaemic times ranging from 5 to 90 min gives rise to changes similar to those of an haemorrhagic pancreatitis. The severity of the changes is proportional to the time of ischaemia. After 1 to 12 h clamping 95% of the glandular changes are reversible, and at 2 months the pancreas appears normal apart from localized areas of fibrosis. The islets of Langerhans remain morphologically normal even after 90 min ischaemia. The implications for pancreatic and islet cell transplantation are discussed. THE SUCCESS of pancreatic transplantation is limited because of problems with vascular anastomoses, immunological rejection, autodigestion and pancreatitis. Pancreatitis has been attributed to operative trauma, ischaemia (either during transplantation or due to post-transplantation thrombosis), duct ligation or to rejection, and it is difficult to distinguish these histologically (Lee et al., 1972). Although the sensitivity of the pancreas to ischaemia has been blamed in some cases for transplant failure (Joison, 1967), the critical ischaemic time limit has not been clearly established. A period of warm ischaemia is inevitable in clinical transplantation and knowledge of its effects and limitations is therefore of paramount importance. Transplantation of isolated islets of Langerhans has been performed successfully in the rat (Kemp et al., 1973) and a clinical trial in man has been undertaken in an attempt to cure diabetes mellitus (Najarian, 1974, personal communication). This study was undertaken to quantitate the effects of reversible ischaemia on the 2 major components of the pancreas and to assess whether or not these effects are permanent. MATERIALS AND METHODS Operative technique.-Young adult male albino rats weighing approximately 200 g were
used for the study. Under general anaesthesia (nitrous oxide, oxygen, halothane) the upper abdomen was opened transversely and the spleen and tail of the pancreas withdrawn. "BBulldog" vascular clamps were applied across the body of the pancreas approximately 1 cm away from the spleen, thereby occluding the blood supply to the tail of the pancreas. When necessary, a second clamp was applied across the vessels running between the stomach and the upper pole of the spleen to interrupt any collateral circulation to the pancreatic tail. The exposed tissues were covered with a inoist swab to maintain them at body temperature for the duration of ischaemia. After removal of the clamps the viscera were returned to the abdomen, which was then closed. In a series of sham operated controls the pancreas and spleen were withdrawn from the abdominal cavity but were not clamped. To assess the contribution of duct obstruction to any changes observed, in another series of rats the main pancreatic dtuct was clamped near its termination but without rendering the pancreas ischaemic. Seruim amylase levels were determined in the early post-operative period, using an hydrolytic technique (Phadebas amylase test-Pharmacia AB). The normal range was also determined. Time course. Triplicate experiments were performed with pancreatic clamping times of 1, 5, 15, 30, 60 and 90 min and with duct clamping times of 60 min. The rats were killed at the end of the experiment (time 0), 6 and 12 h, 1, 2, 3, 5 and 6 days and 1, 2, 4, 8 and 12 weeks. Samples were taken from the tail and head of the pancreas. Histology. Specimens were fixed in Bouin's fluid and paraffin sections were stained with haematoxylin and eosin, aldehyde-fuichsin, Glomori's reticulin stain, elastic-Van Gieson and
PANCREATIC ISCHAEMIA
Perl's stain. Twenty "step" levels through each pancreas were examined and using a point counting technique with a Zeiss I integrating eyepiece, the percentage area of tissue showing changes was measured. RESULTS
There were no post-operative complications or deaths and all rats remained well.
531
ation, with the cells appearing basophilic and more pyramidal with a consequently smaller central acinar lumen (Fig. 5). The areas of collapse showed fibroblastic proliferation with an increase in local collagen. 1-3 month-The glands appeared normal (Fig. 7) with the exception of the areas of collapse which showed increased fibrous tissue and necrotic acinar cells (Fig. 8).
Sequence of changes in ischaemic area The following sequence of changes was Adipose tissue seen with ischaemic times between 5 and 3 day-Small areas of fat necrosis were 90 min: visible macroscopically and microscopically in the peripancreatic fat. Acinar tissue These were usually adjacent to areas Zero time-The pancreas appeared of glandular degeneration. normal macroscopically and micro6 day-Large areas of fat necrosis were scopically present and each area was surrounded 6 h-The pancreas appeared mildly by chronic inflammatory cells, oedematous both macroscopically including giant cells (Fig. 9). Areas and microscopically. of calcification were not seen. 12-24 h-Polymorphonuclear leucocyte 1-3 month-The areas of fat necrosis infiltration was observed in interappeared to have been replaced by acinar and occasionally intra-acinar fibrous tissue. sites (Fig. 1). 2 day-Early degeneration of acinar Blood vessels cells was observed. This was Numerous capillary haemorrhages characterized by the cells appearing were observed between Days 3 and 7. less basophilic and assuming a The larger vessels appeared normal. cuboidal or flattened shape, resulting in an increase in diameter of the Islets of Langerhans central acinar lumen (Fig. 2). Throughout the period of observation 6 day-Glandular degeneration and (even during the period of maximal cellular infiltration was maximal acinar tissue change) the islets (Fig. 3, 4). The cell infiltrate inappeared morphologically normal, cluded numerous lymphocytes, with unaffected alpha cells and plasma cells and macrophages with fully granulated beta cells. Reticuthe latter containing large amounts of lin stains confirmed normal islet stainable iron pigment. Small areas structure (Fig. 5, 10). contained necrotic acinar cells and reticulin stains demonstrated a variChanges in Relation to Duration of able degree of structural collapse Ischaemia (Fig. 5). 1 week-Mitotic figures were observed As shown in the Table, the area of the within the acinar cells and approxi- ischaemic pancreas with demonstrable mately one mitosis was observed histological changes was proportional to per acinus. the clamping time. Although 94-96% of 2 weeks-There was now beginning of the pancreas showed degeneration with reversal of the glandular degener- clamping times of 60 and 90 min, by
532
D. N. SLATER, D. BARDSLEY, Y. MANGNALL, A. SMYTHE AND M. FOX
FIG. 1.-24 h: Oedema and leucocyte infiltration. H. and E. x 255. FIG. 2.-2 day: Early degeneration of acinar cells. H. and E. x 64. FIG. 3.-6 day: Maximal acinar degeneration. H. and E. x 25.
PANCREATIC ISCHAEMIA
533
FIG. 4.-6 day: Acinar cells with increased size of central lumen. H. and E. x 255. FIG. 5.-6 day: Islets of Langerhans with normal architecture but a degree of glandular collapse is present. Reticulin stain x 64. FIG. 6.-2 week: Regeneration of acinar cells. H. and E. x 128.
534
D. N. SLATER, D. BARDSLEY, Y. MANGNALL, A. SMYTHE AND M. FOX
o
CO co
X
C)
4DX
x
C)
o03 o ¢ 0-
e
*W ;
E
4
-
P¢ O
i
535
PANCREATIC ISCHAEMIA
TABLE.-Severity of Exocrine Changes with Different Clamping Times at Body Temperature
Clamping time (min) 1
5 15 30 60 90
Time of initial change (h) Nil 24 24 12 12 12
% Area of changes on Day 2 (*) Nil 3 12 35 57 62
% Area of changes on Day 7 (*)
Time of recovery
Nil 7 17 79 94 96
Nil 3 4 8 8 8
(week)
% Area of "
collapse " (*)
Nil Nil Nil Nil 4 5
(*) Mean of 3 experiments.
8 weeks 94-95 % of the gland had completely recovered. The macroscopic degree of fat necrosis was also proportional to the clamping time and this correlated with histological appearances. The islets of Langerhans appeared morphologically normal at all clamping times.
Duct clamping and controls Neither the pancreases from the controls nor from the duct clamping experiments showed any significant abnormality within the glands, blood vessels or islets. It is concluded therefore that the duct obstruction which inevitably occurred in the ischaemic group was of insufficient duration to have any significant effect, and that the changes seen in this group were the results of ischaemia alone.
Serum amylase levels With ischaemic times of 90 min there were variable elevations of the serum amylase within the first 48 h, with a mean elevation to 8,500 i.u./l (s.d. ± 300). Normal mean amylase in rat 6,600 i.u./l (s.d. ± 250). DISCUSSION
Pancreatic ischaemia in the rat, lasting from 5 to 90 min at body temperature, produces histological changes which are proportional to the ischaemic time. After 60 to 90 min clamping the changes resembled those of a mild haemorrhagic pancreatitis. This is of interest in that it has been stated that ischaemia does
not initiate acute pancreatitis, but has
only a potentiating effect (White, 1966; Robbins, 1967). However, the present findings suggest that ischaemia may be at least one primary aetiological factor in pancreatitis occurring after transplantation. The proportional elevations of serum amylase levels were small when compared with those seen in human acute pancreatitis. There are 2 possible explanations for this. Firstly, approximately only one third of the pancreas was rendered ischaemic; secondly, Huttunen (1973) has reported similarly high levels in normal adult rats and the work of McGeachin and Potter (1965) suggests that this is possibly because of an additional amylase which is of hepatic origin. Although the exact contribution of hepatic amylase to normal levels is not known, if, in the rat, the pancreas contributes only a small proportion of the total serum amylase, the apparently small elevations achieve greater significance. The effects on the pancreas of irreversible ischaemia have been reported on numerous occasions (Smyth, 1940; Pfeffer et al., 1962), but there have been few studies of the effects of reversible ischaemia. Popper, Necheles and Russell (1948) reported that in the dog 30-45 min ischaemia had no macroscopic or microscopic effects, while Inou, Ota and Mori (1968), also in the dog, found only interacinar oedema after 44 min ischaemia. These contrast with the more marked changes found in the present experiment. This study has shown that, in the rat,
5..-53
D. N. SLATER. D. BARDSLEY, Y. MANGNALl, A. SMYTHE AND M. FOX
the effects of up to 90 min ischaetnia on and to Mrs M. Row for photographic assisthe acinar tissute are almost totally rever- tance. This study was supported by sible within 2 months. Attempts are grants from the Medical Research Council being made to cure diabetes mellitus by and British Diabetic Associationi. transplantation of islets of Langerhans (Naj arian, 1974, personal communication). REFERENCES This technique usually involves collagenase H-TTVNEN, R. (1973) Proteolytic Enzymes in Experimental Rat Pancreatitis. Ann. med. Exp. Biol. digestion of the freshly isolated pancreas. Fenniae, 51,133. It is therefore of interest that the islets Ioio, T., OTA, K. & MORI, S. (1968) Manifestations retain their normal histological appearance of Rejection of Pancreaticoduodenal Allografts. Tratisplatatioet, 6, 503. after 90 min warm ischaemia. This sugJ. (1967) Pancreatic Allotransplantation gests that up to at least 90 min they are .1OISON, without Cessation of Blood Flow. Surgery, Gynec. Obstet., 124, 1295. more resistant to ischaemia than the C. B., KNIGHT, M. J., SHARP, D. W., acinar tissue and that they are not affected KEMP, BALLINGER, W. F. & LACEY, P. E. (1973) Effect of by acinar degeneration and enzyme release. Transplantation Site on the Results of Pancreatic Islet Isograft in Diabetic Rats. Diabetologia, 9, Although presence of granules and normal 486. histological outlines are not conclusive LEE, S., TtTNG, K. S. K., KoOPMANS, H., CHANDLER, proof of islet cell function, other studies J. G. & ORLOFF, M. J. (1972) Pancreaticoduodenal Transplantation in the Rat. Transplantation, 13, suggest that histological changes do 421. occur when islet cells have been exposed McGEACHIN, R. L. & POTTER, B. A. (1965) Amylase Synthesis and Transport in the Isolated, perfused to longer periods of ischaemia (Slater, Liver. Adv. Enz. Reg., 3, 137. 1975, unpublished). Work is also in PFEFFER, R. B., LAZZARINI-ROBERTSON, A., SAFADI, progress to assess the physiological status D., MIXTER, G., SECOY, C. F. & HINTON, J. W. (1962) Gradations of Pancreatitis, Edematous, of the ischaemic islets by estimating through Hemorrhagic, Experimentally Produce(d insulin release in response to glucose and by Controlled Injection of Microspheres into these results will be communicated at a, Blood Vessels in Dogs. Surgery, St. Louis, 51, 764. POPPER, H. L., NECHELES, H. & RusSELL, K. C. laXlter date. (1948) Transition of Pancreatic Edema into
WVe are grateful to Miss S. Geary and Mrs L. Lonsdale for assistance with histology; to Dr D. M. Goldberg for the seruim amylase determinations, to Mrs J. Jacques for preparing the manuscript
Pancreatic Necrosis. Surgery, Gynec. Obstet., 87, 79. ROBBINS, S. L. (1967) Pathology. 3rd edn. Philadelphia andt LoIIdon: Saunders. p. 970. SMYTH, C. J. (1940) Etiology of Acute Hemorrhagic Pancreatitis with Special Reference to the Vascular Factors. Archs Path., 30, 651. WHITE, T. T. (1966) Pancreatitis. 1st edi. LoI1(lon1: Edward Arnold. p. 70.
