- Email: [email protected]
Effect of postmortem ischemia on the function of adult rat islets following pancreatic transplantation
Metabolism Clinical and Experimental JULY 1983
VOL. XxX11, NO. 7
PRELIMINARY
REPORT
Effect of Postmortem Ischemia on the Function of Adult Rat Islets Following Pancreatic Transplantation James A. Schulak Pancreata from adult rats were subjected to in situ postmortem ischemia at ambient temperature for between SO and 360 minutes before syngeneic transplantation into diabetic recipients. Islet function, determined by recipient blood glucose and intravenous glucose tolerance tests, was not significantly different from that observed in nonischemic controls after up to 300 minutes of such an insult. Transplantation led to death of pancreata subjected to 360 minutes of ischemia in four of five recipients. Clinical and experimental implications are discussed.
I
T HAS BEEN DEMONSTRATED PREVIOUSLY that pancreatic islets of adult rats subjected to ex vivo ischemia at 37 “C for up to 90 minutes will function normally, as determined by posttransplantation glucose metabolism in the diabetic recipients of primarily vascularized syngeneic segmental pancreatic grafts.‘,’ The study reported here includes an evaluation of in situ ischemia induced by allowing sacrificed animals to remain at ambient temperature and by delaying organ harvest and transplantation for the indicated period of ischemic insult. A similar study has been reported for neonatal rats in which the islets were isolated and subsequently transplanted as intraportal emboli.’ Here the function of intact adult pancreata following ischemic challenge and subsequent transplantation is described.
to the infrarenal aorta and vena cava of the recipient. The usual anastomosis time was approximately 30 minutes. The ischemic insult was achieved by killing the donor animals through exsanguination immediately following partial organ dissection. The pancreata were then allowed to remain in situ for the designated period of ischemia before removal and subsequent transplantation. During this period the abdominal incision was loosely closed. In one group of donors, intraperitoneal temperatures were measured hourly. Following transplantation, blood glucose was measured periodically for three weeks, after which an intravenous glucose tolerance test (IVGTT) was performed using 500 mg/kg of dextrose. Subsequently, recipients underwent graft pancreatectomy in order to verify that control of their glucose metabolism was indeed due to the presence of the pancreatic transplant.
RESULTS
MATERIALS AND METHODS
The results of this experiment are summarized in Table I. Up to five hours of ischemia at ambient temperature was tolerated by the islets in groups 2-6 as manifested by mean glucose values that were similar
Syngeneic adult Lewis x Brown Norway F, hybrid male rats were used as both donors and recipients in all experiments; this eliminated the problem of allograft rejection. Recipient animals were made diabetic with an intravenous injection (65 mg/kg) of streptozotocin (Upjohn). Blood glucose was measured with a reflectance colorimeter (Dextrometer, Ames) using a drop of tail capillary blood. Only rats with stable hyperglycemia of greater than 400 mg/dL were used as recipients in these experiments. Blood glucose values of greater than 400 mg/dL cannot be measured accurately with this technique, thus all such values are listed as greater than 400. The technique of donor organ procurement and transplantation into the recipient has been described previously.* Briefly. ductligated segmental pancreas grafts were placed in a heterotopic position by anastomosing the celiac axis and portal vein of the graft
Fromthe Department ofsurgery, The Universirv offowa College of Medicine, Iowa City, IA Received for publication March 8, 1983. This investigation was supported in part by Biomedical Research Support Grant RR 0.5372 from the Biomedical Research Support Branch. Division of Research Facilities and Resources. National Institutes of Health. Address reprint requests to James A. Schulak, MD, Department of Surgery, The University of Iowa College of Medicine, Iowa City. IA 52242. 0 1983 by Grune & Stratton, Inc. 00X-0495/83/3207-OOOl$l.OO/O
Merebo/ism, Vol. 32, No. 7 (Julv), 1983
643
644
JAMES
Table 1. Effect of Preharvest Warm lschemia on Blood Glucose Concentrations Following Pancreatic
Number
1
Total Number
Lengthof lschemia (minutes1
Group
Control-no
Number Successful
kchemia
Transplantation
A. SCHULAK
in Diabetic Rats
Mean Blood Glucose Levels Img/dL; Mean + SD) After Transplantarion 1 Day
7 Days
3
Pretransplant ,400
88 r 24
90 f 25
14 Days 108
f 27
21 Days
Postpancreatectomy
90 * 7
2400
2
90
3
2400
116 f 25
3
120
5
>400
97 k 7
4
180
3
>400
5
240
3
>400
152
2 120
155
+ 82
115
+ 2
95k
6
300
3
>400
189
_+ 139
138
f 30
108
_+ 7
93 _t 11
7
360
1’
>400
112+33 121
97 k 28
i
92 f 12 19
93 k 8
48
108
116
i 62
115215
115
101 t 14 115
k 20
lOOk5
114
333 >400 ,400
13
>400 377 -t
*Four rats in this group died within 48 hours of transplantation (see text for explanation). tRat died during graft pancreatectomy.
to those in the control group (there were no significant differences between all group means and the control mean with t test analysis). Early metabolic dysfunction was observed in several of the recipients of the fourand five- hour ischemic organs (groups 5 and 6) in that they had somewhat higher glucose values at both one and seven days after transplantation than did the animals in groups 1-4 as manifested by their greater group mean values. On the other hand, at three weeks after transplantation, all recipients had blood glucose values of less than 130 mg/dL, and there continued to be no significant difference in the mean values (t test, P > 0.1 or greater). In contrast, extension of the ischemic period to six hours (group 7) resulted in graft failure and death of the animal within 48 hours in four of five recipients. One rat did survive in this group and eventually manifested normal glucose values after being moderately hypoglycemic during the first day after transplantation. Intraperitoneal temperatures were measured in group 7 in order to ascertain the amount of graft cooling that occurs after donor death. The temperature dropped to 30.1 + 0.9 OCduring graft dissection. Over the next six hours the mean intra-abdominal tempera400CROUPS . A 0 A
min.) min.) 4 (160 min.) 5 (240 min.) l 6 (300 min.)
o
5
101520
30 TIME.
Fig. 1. Data from intravenous trose. 600 mg/kgJ.
2 (90
3 (120
60
min. glucose tolerance tests Idex-
ture fell further with hourly values of 26.5 f 1.3, 23.8 + 1.8, 22.2 r 2.1, 21.7 + 2.3, 21.5 + 2.3, and 21.3 + 2.1 “C respectively. The IVGTT data are illustrated in Figure I, Glucose decay in all five groups was rapid and similar in pattern. While a slightly blunter response was observed in groups, 2, 3, and 6 when compared with groups 4 and 5, within 60 minutes the mean glucose values for all groups had returned to baseline, suggesting that overall islet function was comparable. Last, graft pancreatectomy performed on all recipients following the IVGTT resulted in recurrent hyperglycemia in all cases, indicating that the level of glucose homeostasis observed during the test period was indeed due to the presence of the transplanted pancreas rather than secondary to the function of the native islets.
DISCUSSION
The purpose of this experiment was to better determine the “warm” ischemic tolerance of adult rat pancreatic islets. Previously, Matas et al3 have demonstrated that isolated neonatal islets transplanted as intraportal emboli can withstand up to three hours of in situ warm ischemia when donors were left at room temperature before organ removal, while ischemia at 37 OC was not as well tolerated. Similar data were obtained in our laboratory using a model of adult rat segmental pancreatic transplantation,2 where adult rat islets functioned normally after up to 90 minutes of warm ischernia at 37 OC. The discrepancy between the amount of postmortem ischemia tolerated by neonatal islets as described by Matas et al3 and that reported here for adult islets may have two explanations. First, cooling after death was probably more rapid in these experiments since the donors underwent partial pancreatic dissection before death as indicated by a fall in temperature to approximately 30 “C during this procedure, whereas in Matas’s experiments the peritoneal cavity was closed until organ removal. Second, the neonatal islets in Matas’s experiments were subjected to an added period of ischemia at 37 OCas required for
ISLET FUNCTIONING
FOLLOWING
ISCHEMIA
645
enzymatic islet separation and purification compared with the relatively short additional ischemic period required for anastomosis in these experiments. The experiments in this report differ from our previous investigation in that the effect of room temperature ischemia on adult rat pancreatic islet function was evaluated. The data clearly demonstrate that at ambient temperature, islet function is preserved up to five hours after donor sacrifice. The increased tolerance to long periods of ischemia in these experiments when compared with our previous experiments at 37 “C may be explained by the rapid decrease in temperature to approximately 20 OC. Early graft dysfunction was observed with the two longer periods of ischemia, groups 5 and 6, as manifested by slightly higher mean blood glucose values for up to one week. Presumably this was due to reversible ischemiainduced islet injury, since mean glucose values were indistinguishable from all other groups by the second week. In addition, the ability of recipients in all five successful groups to handle an intravenous glucose load was similar. Four of five recipients of six-hour ischemic pancreata died within 48 hours of transplantation. While the cause of death cannot be determined with certainty, current experience with attempting to achieve rat pancreas preservation using standard cold storage in Collins’ extracellular electrolyte solution suggests that early recipient death is secondary to either severe graft pancreatitis or lethal hypoglycemia, which results from a massive release of insulin from a severely damaged organ (unpublished data). In support of this theory is the finding that the one successful rat in that group was hypoglycemic for the first 24 hours. These data reveal an important aspect of islet physiology that may be applied both to clinical transplantation and to more efficient utilization of laboratory animals in research. Should human islets prove to be as
tolerant of warm ischemia as these in adult rats, then islets for clinical use may be harvested from pancreata of donors who have already suffered up to several hours of circulatory arrest, rather than requiring more expeditious removal that is now necessary for kidneys, hearts, and livers. This would both increase the number of potential islet donors as well as ease the logistics of such an endeavor. However, because of the very rapid decline to ambient temperature (20 “C) observed in this experiment, these rat islets may have enjoyed a greater amount of hypothermic protection than might be expected in a human cadaver, where cooling after death has been estimated to occur at a rate of between 0.3 “C and 1.8 OC per hour.4 While these data from rats may well pertain to human pancreata with regard to subsequent function of isolated islet allografts, it is not certain whether primarily vascularized human segmental pancreas grafts can also be used after prolonged periods of warm ischemia. This latter situation may be hampered by the risk of acinar pancreatitis, which may occur after a much shorter period of ischemia.’ Possibly of more immediate importance than clinical transplantation of ischemic islets is the implication that dead laboratory animals may still serve as a source of islets for in vitro experimentation. Since these data clearly demonstrate that adult rat islets will function in a relatively normal fashion after several hours of ischemia, it appears that using rats that were subjected to other investigations and subsequently killed as an islet source may not only be a feasible venture but also warranted in view of the need for more efficient use of resources in this age of research support that is increasingly difficult to obtain. ACKNOWLEDGMENT The author wishes to thank Kristin Engelstad for their expert technical assistance.
and James Kisthard
REFERENCES 1. Schulak
JA, Franklin W, Buckingham F, Stuart FP, et al: Warm ischemia in pancreas transplantation: a functional and histologic evaluation. Surg Forum 32:392-393, 1981. 2. Schulak JA, Franklin W, Stuart FP, et al: The effect of warm ischemia on segmental pancreas transplantation in the rat. Transplantation 35:7-l I, 1983. 3. Matas AJ, Sutherland DER, Payne WD, et al: islet transplan-
tation: the critical period of donor ischemia in neonatal rats. Transplantation 23:295-298, 1977. 4. Gee DJ: Lecture Notes on Forensic Pathology. Oxford. Blackwell Scientific Publications, 1979, p 101, 5. Broe PJ, Zuidema GD, Camearon JL: The role of ischemia in acute pancreatitis: studies with an isolated perfused canine pancreas. Surgery 91:3777382. 1982.
VOL. XxX11, NO. 7
PRELIMINARY
REPORT
Effect of Postmortem Ischemia on the Function of Adult Rat Islets Following Pancreatic Transplantation James A. Schulak Pancreata from adult rats were subjected to in situ postmortem ischemia at ambient temperature for between SO and 360 minutes before syngeneic transplantation into diabetic recipients. Islet function, determined by recipient blood glucose and intravenous glucose tolerance tests, was not significantly different from that observed in nonischemic controls after up to 300 minutes of such an insult. Transplantation led to death of pancreata subjected to 360 minutes of ischemia in four of five recipients. Clinical and experimental implications are discussed.
I
T HAS BEEN DEMONSTRATED PREVIOUSLY that pancreatic islets of adult rats subjected to ex vivo ischemia at 37 “C for up to 90 minutes will function normally, as determined by posttransplantation glucose metabolism in the diabetic recipients of primarily vascularized syngeneic segmental pancreatic grafts.‘,’ The study reported here includes an evaluation of in situ ischemia induced by allowing sacrificed animals to remain at ambient temperature and by delaying organ harvest and transplantation for the indicated period of ischemic insult. A similar study has been reported for neonatal rats in which the islets were isolated and subsequently transplanted as intraportal emboli.’ Here the function of intact adult pancreata following ischemic challenge and subsequent transplantation is described.
to the infrarenal aorta and vena cava of the recipient. The usual anastomosis time was approximately 30 minutes. The ischemic insult was achieved by killing the donor animals through exsanguination immediately following partial organ dissection. The pancreata were then allowed to remain in situ for the designated period of ischemia before removal and subsequent transplantation. During this period the abdominal incision was loosely closed. In one group of donors, intraperitoneal temperatures were measured hourly. Following transplantation, blood glucose was measured periodically for three weeks, after which an intravenous glucose tolerance test (IVGTT) was performed using 500 mg/kg of dextrose. Subsequently, recipients underwent graft pancreatectomy in order to verify that control of their glucose metabolism was indeed due to the presence of the pancreatic transplant.
RESULTS
MATERIALS AND METHODS
The results of this experiment are summarized in Table I. Up to five hours of ischemia at ambient temperature was tolerated by the islets in groups 2-6 as manifested by mean glucose values that were similar
Syngeneic adult Lewis x Brown Norway F, hybrid male rats were used as both donors and recipients in all experiments; this eliminated the problem of allograft rejection. Recipient animals were made diabetic with an intravenous injection (65 mg/kg) of streptozotocin (Upjohn). Blood glucose was measured with a reflectance colorimeter (Dextrometer, Ames) using a drop of tail capillary blood. Only rats with stable hyperglycemia of greater than 400 mg/dL were used as recipients in these experiments. Blood glucose values of greater than 400 mg/dL cannot be measured accurately with this technique, thus all such values are listed as greater than 400. The technique of donor organ procurement and transplantation into the recipient has been described previously.* Briefly. ductligated segmental pancreas grafts were placed in a heterotopic position by anastomosing the celiac axis and portal vein of the graft
Fromthe Department ofsurgery, The Universirv offowa College of Medicine, Iowa City, IA Received for publication March 8, 1983. This investigation was supported in part by Biomedical Research Support Grant RR 0.5372 from the Biomedical Research Support Branch. Division of Research Facilities and Resources. National Institutes of Health. Address reprint requests to James A. Schulak, MD, Department of Surgery, The University of Iowa College of Medicine, Iowa City. IA 52242. 0 1983 by Grune & Stratton, Inc. 00X-0495/83/3207-OOOl$l.OO/O
Merebo/ism, Vol. 32, No. 7 (Julv), 1983
643
644
JAMES
Table 1. Effect of Preharvest Warm lschemia on Blood Glucose Concentrations Following Pancreatic
Number
1
Total Number
Lengthof lschemia (minutes1
Group
Control-no
Number Successful
kchemia
Transplantation
A. SCHULAK
in Diabetic Rats
Mean Blood Glucose Levels Img/dL; Mean + SD) After Transplantarion 1 Day
7 Days
3
Pretransplant ,400
88 r 24
90 f 25
14 Days 108
f 27
21 Days
Postpancreatectomy
90 * 7
2400
2
90
3
2400
116 f 25
3
120
5
>400
97 k 7
4
180
3
>400
5
240
3
>400
152
2 120
155
+ 82
115
+ 2
95k
6
300
3
>400
189
_+ 139
138
f 30
108
_+ 7
93 _t 11
7
360
1’
>400
112+33 121
97 k 28
i
92 f 12 19
93 k 8
48
108
116
i 62
115215
115
101 t 14 115
k 20
lOOk5
114
333 >400 ,400
13
>400 377 -t
*Four rats in this group died within 48 hours of transplantation (see text for explanation). tRat died during graft pancreatectomy.
to those in the control group (there were no significant differences between all group means and the control mean with t test analysis). Early metabolic dysfunction was observed in several of the recipients of the fourand five- hour ischemic organs (groups 5 and 6) in that they had somewhat higher glucose values at both one and seven days after transplantation than did the animals in groups 1-4 as manifested by their greater group mean values. On the other hand, at three weeks after transplantation, all recipients had blood glucose values of less than 130 mg/dL, and there continued to be no significant difference in the mean values (t test, P > 0.1 or greater). In contrast, extension of the ischemic period to six hours (group 7) resulted in graft failure and death of the animal within 48 hours in four of five recipients. One rat did survive in this group and eventually manifested normal glucose values after being moderately hypoglycemic during the first day after transplantation. Intraperitoneal temperatures were measured in group 7 in order to ascertain the amount of graft cooling that occurs after donor death. The temperature dropped to 30.1 + 0.9 OCduring graft dissection. Over the next six hours the mean intra-abdominal tempera400CROUPS . A 0 A
min.) min.) 4 (160 min.) 5 (240 min.) l 6 (300 min.)
o
5
101520
30 TIME.
Fig. 1. Data from intravenous trose. 600 mg/kgJ.
2 (90
3 (120
60
min. glucose tolerance tests Idex-
ture fell further with hourly values of 26.5 f 1.3, 23.8 + 1.8, 22.2 r 2.1, 21.7 + 2.3, 21.5 + 2.3, and 21.3 + 2.1 “C respectively. The IVGTT data are illustrated in Figure I, Glucose decay in all five groups was rapid and similar in pattern. While a slightly blunter response was observed in groups, 2, 3, and 6 when compared with groups 4 and 5, within 60 minutes the mean glucose values for all groups had returned to baseline, suggesting that overall islet function was comparable. Last, graft pancreatectomy performed on all recipients following the IVGTT resulted in recurrent hyperglycemia in all cases, indicating that the level of glucose homeostasis observed during the test period was indeed due to the presence of the transplanted pancreas rather than secondary to the function of the native islets.
DISCUSSION
The purpose of this experiment was to better determine the “warm” ischemic tolerance of adult rat pancreatic islets. Previously, Matas et al3 have demonstrated that isolated neonatal islets transplanted as intraportal emboli can withstand up to three hours of in situ warm ischemia when donors were left at room temperature before organ removal, while ischemia at 37 OC was not as well tolerated. Similar data were obtained in our laboratory using a model of adult rat segmental pancreatic transplantation,2 where adult rat islets functioned normally after up to 90 minutes of warm ischernia at 37 OC. The discrepancy between the amount of postmortem ischemia tolerated by neonatal islets as described by Matas et al3 and that reported here for adult islets may have two explanations. First, cooling after death was probably more rapid in these experiments since the donors underwent partial pancreatic dissection before death as indicated by a fall in temperature to approximately 30 “C during this procedure, whereas in Matas’s experiments the peritoneal cavity was closed until organ removal. Second, the neonatal islets in Matas’s experiments were subjected to an added period of ischemia at 37 OCas required for
ISLET FUNCTIONING
FOLLOWING
ISCHEMIA
645
enzymatic islet separation and purification compared with the relatively short additional ischemic period required for anastomosis in these experiments. The experiments in this report differ from our previous investigation in that the effect of room temperature ischemia on adult rat pancreatic islet function was evaluated. The data clearly demonstrate that at ambient temperature, islet function is preserved up to five hours after donor sacrifice. The increased tolerance to long periods of ischemia in these experiments when compared with our previous experiments at 37 “C may be explained by the rapid decrease in temperature to approximately 20 OC. Early graft dysfunction was observed with the two longer periods of ischemia, groups 5 and 6, as manifested by slightly higher mean blood glucose values for up to one week. Presumably this was due to reversible ischemiainduced islet injury, since mean glucose values were indistinguishable from all other groups by the second week. In addition, the ability of recipients in all five successful groups to handle an intravenous glucose load was similar. Four of five recipients of six-hour ischemic pancreata died within 48 hours of transplantation. While the cause of death cannot be determined with certainty, current experience with attempting to achieve rat pancreas preservation using standard cold storage in Collins’ extracellular electrolyte solution suggests that early recipient death is secondary to either severe graft pancreatitis or lethal hypoglycemia, which results from a massive release of insulin from a severely damaged organ (unpublished data). In support of this theory is the finding that the one successful rat in that group was hypoglycemic for the first 24 hours. These data reveal an important aspect of islet physiology that may be applied both to clinical transplantation and to more efficient utilization of laboratory animals in research. Should human islets prove to be as
tolerant of warm ischemia as these in adult rats, then islets for clinical use may be harvested from pancreata of donors who have already suffered up to several hours of circulatory arrest, rather than requiring more expeditious removal that is now necessary for kidneys, hearts, and livers. This would both increase the number of potential islet donors as well as ease the logistics of such an endeavor. However, because of the very rapid decline to ambient temperature (20 “C) observed in this experiment, these rat islets may have enjoyed a greater amount of hypothermic protection than might be expected in a human cadaver, where cooling after death has been estimated to occur at a rate of between 0.3 “C and 1.8 OC per hour.4 While these data from rats may well pertain to human pancreata with regard to subsequent function of isolated islet allografts, it is not certain whether primarily vascularized human segmental pancreas grafts can also be used after prolonged periods of warm ischemia. This latter situation may be hampered by the risk of acinar pancreatitis, which may occur after a much shorter period of ischemia.’ Possibly of more immediate importance than clinical transplantation of ischemic islets is the implication that dead laboratory animals may still serve as a source of islets for in vitro experimentation. Since these data clearly demonstrate that adult rat islets will function in a relatively normal fashion after several hours of ischemia, it appears that using rats that were subjected to other investigations and subsequently killed as an islet source may not only be a feasible venture but also warranted in view of the need for more efficient use of resources in this age of research support that is increasingly difficult to obtain. ACKNOWLEDGMENT The author wishes to thank Kristin Engelstad for their expert technical assistance.
and James Kisthard
REFERENCES 1. Schulak
JA, Franklin W, Buckingham F, Stuart FP, et al: Warm ischemia in pancreas transplantation: a functional and histologic evaluation. Surg Forum 32:392-393, 1981. 2. Schulak JA, Franklin W, Stuart FP, et al: The effect of warm ischemia on segmental pancreas transplantation in the rat. Transplantation 35:7-l I, 1983. 3. Matas AJ, Sutherland DER, Payne WD, et al: islet transplan-
tation: the critical period of donor ischemia in neonatal rats. Transplantation 23:295-298, 1977. 4. Gee DJ: Lecture Notes on Forensic Pathology. Oxford. Blackwell Scientific Publications, 1979, p 101, 5. Broe PJ, Zuidema GD, Camearon JL: The role of ischemia in acute pancreatitis: studies with an isolated perfused canine pancreas. Surgery 91:3777382. 1982.