- Email: [email protected]
Cyclosporin A and islet function
Cyclosporin A and Islet Function Giacomo Basadonna, MD, Francesco Montorsi, MD, Kenji Kakizaki, MD, Ronald C. Merrell, MD, Houston, Texas Long-term cyclosporin A (CsA) administration in dogs was studied with respect to function of the islets of Langerhans. After 3 weeks of immunosuppression with therapeutic doses, the islets were isolated and assessed in vitro for insulin release in response to glucose challenge. Islet tissue retrieved from the CsA-treated animals showed a total insulin output significantly lower than that of the control animals (p <0.01 ). The first and second phases of insulin release were both impaired in animals treated with CsA compared with controls (p <0.001 and p <0.05, respectively). The negative impact of CsA on the beta cells was easily demonstrated in this in vitro study. Similar results are more difficult to achieve with purely in vivo models, probably due to the great redundancy of the islet mass in intact animals. The mechanism of this CsA toxicity remains to be defined.
he advantages of cyclosporin A (CsA) immunosuppression are considerable and have made this drug T the agent of choice for clinical transplantation of the heart, kidneys, bone marrow, and liver [I-4]. CsA has also been used in pancreatic transplantation, either alone or in combination with azathioprine and prednisone [5]. The addition of CsA to the immunosuppressive regimen improves glucose tolerance in patients with functioning pancreatic allografts [6-8]. It also offers therapeutic promise in incipient type I diabetes mellitus for suppressing the autoimmune reaction destroying beta cells [9,10]. However, CsA has also shown a poorly understood toxicity towards pancreatic islets both in vitro and in vivo. It has been associated with poor results when employed as an immunosuppressant in islet anografts [11], and intrasplenic canine islet autografts can be compromised by it [12]. In a recent study, Gores et al [13] did not confirm CsA toxicity to canine autografts, but in that study, the incidence of failure of untreated autografts may have obscured the effect. Certain patients with functioning pancreatic and renal allografts who are taking azathioprine and steroids have shown significant deterioration of glucose clearance after conversion to CsA [14]. This deterioration reversed in one case after resuming the origiFrom the Department of Surgery, The University of Texas Medical School at Houston, Houston, Texas. Request s for reprints should be addressed to Ronald C. Merrell, MD, Department of Surgery, The University of Texas Medical School at Houston, 6431 Fannin, Suite 4292, Houston, Texas 77030.
hal immunosuppression regimen. In vitro prolonged exposure to very high CsA concentrations impaired mouse islet beta cell function, and the detrimental effect was attributed to impaired protein biosynthesis [15]. High doses of CsA in rats caused severe morphologic and functional alterations of the pancreatic B cells with subsequent hyperglycemia and hypoinsulinemia [16]. CsA has a similar negative impact on some specific aspects of the subcellular events of wound healing, although this issue remains controversial [17,18]. One source of difficulty in determining subtle toxicity to islets in vivo is the great redundance of beta cells with respect to metabolic demand, which can mask even extraordinary toxicity. Measurement of acute toxicity on isolated islets does not take into account the long-term aspect of potential in vivo toxicity, This report addresses the effect of long-term cyclosporine administration on canine islets. The model entails a 3 week course of therapeutic doses of CsA and subsequent isolation of the islet to assess in vitro function on a per cell basis. MATERIAL AND METHODS Ten adult mongrel dogs received either a regimen of long-term immunosuppression for 3 weeks with CsA (Sandimmune | Sandoz, Geneva, Switzerland) or served as control subjects. The CsA regimen was 20 mg/kg/day during the first week, 10 mg/kg/day during the second week, and 5 mg/kg/day during the third week. The drug was given by mouth and administration was not difficult. This decreasing dose is similar to clinical regimens. Fasting blood glucose levels were determined using a YSI model 23A glucose analyzer (Yellow Springs Instruments, Yellow Springs, OH). Animals were screened for norm0glycemia before the study and were followed until death. All animals underwent total pancreatectomy after 3 weeks, and islets were isolated as previously described [19]. Briefly, under general inhalant anesthesia, a total pancreatectomy was performed [20]. The pancreas was digested by perfusing the ductal system with 0.2 percent type I collagenase (Sigma, St. Louis, MO) in Hank's balanced salt solution (Gibco, Grand Island, IL). After 40 minutes of perfusion at 37~ the pancreatic tissue was minced, filtered, and washed in Hank's balanced salt solution. The pancreatic fragments were suspended in RPMI 1640 (Gibco) containing 25 mM of N-2-hydroxyethyl-piperazine-N-2-ethanesul-fonic acid (HEPES) and 10 percent calf serum (Gibco). The islet-cell function was assessed according to the perifusion model described by Lacy et al [21]. Approximately 1 X 10 6 cells were loaded directly onto a 2 ml, 25 mm diameter filter holder containing a 1.2 ~z millipore filter and a prefilter. Flow was maintained with a peristaltic pump at 1 ml/min. After 30 minutes of perifusion with Hank's solution containing 25 mM HEPES and 2.75 mM glucose at pH 7.4, the ceils were challenged with Medium 199 (Gibco), containing
THE AMERICAN JOURNAL OF SURGERY
VOLUME 156 SEPTEMBER 1988
191
BASADONNA ET AL
200
20 -
9 CONTROL
18"
9
180
9 CsA
7< 160 Q
16"
12' ,~ 10' "n 8 ' m 6' Z
g
140
2
120 100
I
~ so _~ 60 ~ 40
~r
20
0
,
0
5
10
'
15
20
25
0
30
CsA
CONTROL
TIME (MINUTES)
Figure 1. Total Insulin output per microgram of DNA during 30 minutes of stimulation with 22.2 mM of glucose. Islets retrieved from animals treated with cyclosporin A (CsA) showed a significant decrease in insulin release compared with islets in the control animals. Asterick indicates, p <0.01. TABLE I
Insulin Output (/zU/min/#g DNA) After 22.2 mM Glucose Challenge in Control Animals and Cyclosporin A(CsA)Treated Animals* Time (rain) 0 1 3 5 8 11 15 30
Groups Control 3.07 17.05 13.11 10.67 6.57 4.49 3.03 2.95
4- 0.77 4- 2,90 4- 2,95 4- 2.97 4- 2.06 =1= 1.11 -I- 0.45 4- 0,22
CsA 1.08 1.62 2.11 2.07 1.84 1.75 1.57 1.54
+ 0.24 4- 0.30 :E 0.63 ::E 0.58 4- 0.39 4- 0.33 4- 0.32 4- 0.29
p Value <0.05 <0.001 <0.01 <0.05 NS <0.05 <0.05 <0.01
* COntrol and CsA values expressed as the mean 4- standard deviation. NS = not significant.
22.2 mM of glucose. Samples were collected at 0, 1, 3, 5, 8, 11, 15, and 30 minutes and stored at -40~ The number of cells employed in each perfusion was approximated by measuring DNA by Hill and Whatley's method [22]. Precise cell counting is not possible in these preparations, and islet number is not a particularly accurate measure either due to differences in islet size. However, DNA content per diploid cell is constant in a given species and can be used to reflect cell number. Insulin output was then expressed in micrograms of DNA. Insulin concentrations were determined using a modification of the double antibody insulin radioimmunoassay decribed by Morgan and Lazarow [23] and Marincola et al [24]. Data are expressed as the mean 4- standard error of the mean. Student's t test for impaired data was performed. RESULTS All animals, treated and untreated, remained normoglycemic throughout the study, and the CsA regimen was well tolerated. Islet tissue retrieved from control animals showed a total insulin output significantly higher than that from animals treated with CsA for 3 weeks (162.94 4- 25.05 gU insulin/30 min/#g DNA versus 192
THE A M E R I C A N J O U R N A L O F S U R G E R Y
Figure 2. Insulin release per minute per microgram of DNA after challenge with 22.2 mM of glucose. At each time p o i n t , except time 8, islets from animals treated with cyclnsporin A (CsA) showed a significant decrease in insulin release compared with islets in control animals.
50.81 4- 9.56 #U insulin/30 min/#g DNA, respectively; p <0.01) (Figure 1). Figure 2 shows the insulin release per microgram of DNA at the single time points after glucose challenge. The islet tissue from animals treated with CsA was impaired for both the first and the second peak of insulin release when compared with islet fragments isolated from the control animals (1.62 4- 0.30 #U insulin/min/#g DNA versus 17.05 4- 2.90 #U insulin; p <0.001 and 1.75 4- 0.33 #U insulin/min/#g DNA versus 4.49 4- 1.11 #U insulin/min//zg DNA, respectively; p <0.05). Every time point in the curve of insulin release, with the exception of 8 minutes, showed a statistically significant decrease in insulin output from the islet tissue of immunosuppressed animals compared with the control values (Table I). COMMENTS These data clearly demonstrate an untoward effect of CsA on the islets of Langerhans. The mechanism could be due to functional abnormalities of intact beta cells or to a decrease in the number of beta cells. Clinical data suggest that CsA does not affect pancreatic beta cells directly but induces insulin resistance, leading to a marked deterioration of glucose tolerance [14]. However, Rynasiewicz et al [25] found no significant difference in glucose, C-peptide, and insulin levels when they compared pancreas allograft recipients treated with conventional immunosuppression to those treated with CsA. Our studies with islet autografts in dogs strongly implicated a negative impact of CsA on islet function [12]. CsA should not be considered the immunosuppressive drug of choice in islet allotransplantation [11]. CsA also has a more subtle but still deleterious effect on islet autotransplantation. Intrasplenic islet autografts restore normal glucose metabolism in all cases, but a 50 percent failure rate occurs when CsA is introduced as the only variable [12]. It was impossible to interpret the detrimental effect of CsA on grafted islet cells when working with intact animals. When Helmchen et al [16] used very high doses of CsA in rats, there was a pronounced effect on pancreatic
V O L U M E 156
SEPTEMBER 1988
CYCLOSPORIN A AND ISLET FUNCTION
beta cell morphologic characteristics, with vacuolization of the cytoplasm and loss of secretory granules. From these data, it is perhaps surprising that intact animals showed no more abnormalities after CsA administration. The explanation for this discrepancy is probably the great redundancy of the islet mass, which can be decreased by at least 80 percent before carbohydrate metabolism is significantly impaired. Our study assessed the effect of CsA on the endocrine pancreas in vivo by removing the potentially intoxicated islets for evaluation on a more or less per islet basis in vitro. The negative effect of CsA on insulin output was well demonstrated by the marked decrease of insulin release from the islet tissue isolated from the CsA-treated animals. Our results underscore the importance of further efforts to determine the mechanism of CsA damage to the beta cells of the islets of Langerhans.
Acknowledgment:We thank Bobbie Lovett for manuscript preparation, and we also appreciate the technical support of Tam Phan and Mike Linzel. REFERENCES 1. Morris PJ. Cyclosporine A. Transplantation 1981; 32: 349-54. 2. Morris PJ. The impact of cyclosporine A on transplantation. Chicago: Yearbook Medical, 1984: 99. 3. Merion RM, White D J, Thiru S, Evans DB, Calne RY. Cyclosporine: five years experience in cadaveric renal transplantation. N Engl J Med 1984; 310: 148-54. 4. Starzl TE, Klintmalm GB, Porter KA, Iwatsuki S, Schroter GP. Liver transplantation with use of cyclosporin A and prednisone. N Engl J Med 1981; 305: 266-9. 5. Sutherland DER, Goetz FC, Najarian JS. Recent experience with 89 pancreas transplants at a single institution. Diabetologia 1984; 27:14953. 6. Traeger J, Dubernard JM, Pozza G, et al. Influence of immunosuppressive therapy on the endocrine function of segmental pancreatic allografts. Transplant Proc 1983; 15: 1326-9. 7. Pozza G, Traeger J, Dubernard JM, Secchi A, Bosi A, Pontiroli AE. Cyclosporin and glucose tolerance in pancreas allotransplantation. Lancet 1983; 2: 1080-1. 8. Pozza G, Traeger J, Dubernard JM, et al. Endocrine responses of type 1 (insulin-dependent) diabetic patients following successful pancre-
as transplantation. Diabetologia 1983; 24: 244-8. 9. Nerup J, Lernmark A. Autoimmunity in insulin dependent diabetes. Am J Med 1981; 70: 135-41. 10. Stiller CR, Dupre J, Gent M, et al. Effects of cyclosporine immunosuppresion in insulin-dependent diabetes mellitus of recent onset. Science 1984; 23: 1362-7. 11. DuToit DF, Reece-Smith H, McShane P, Denton T, Morris PJ. Effect of cyclosporin A on allotransplanted pancreatic fragments to the spleen of totally pancreatectomized dogs. Transplantation 1982; 33: 302-7. 12. Merrel RC, Mahoney ME, Basadonna G, Cobb LF, Maeda M. Failure of canine islet aUografts and autografts with cyclosporine A. Surgery 1985; 98: 324-9. 13. Gores PF, Boudreaux JP, Hesse U J, Najarian JS, Sutherland DER. Canine islet autografts with and without administration of cyclosporine. Surgery 1987; 101: 557-61. 14. Gunnarsson R, Klintmalm G, Lundgreu G, Wilczek H, Ostman J, Groth CG. Deterioration in glucose metabolism in pancreatic transplant recipients given cyclosporin. Lancet 1983; 2: 571-2. 15. Andersson A, Borg H, Hallberg A, Hellerstrom C, Sandier S, Schnell A. Long-term effects of cyclosporine A on cultured mouse pancreatic islets. Diabetologia 1984; 27: 66-9. 16. Helmchen V, Schmidt WE, Siegel EG, Creutzfeldt W. Morphological and functional changes of paneratic B cells in cyclosporin A-treated rats. Diabetologia 1984; 27: 416-8. 17. Eisinger DR, Sheil AG. A comparison of the effects of cyclosperine A and standard agents on primary wound healing in the rat. Surg Gynecol Obstet 1985; 160: 135-8. 18. Towpick E, Kupiec-Weglinski JW, Schneider TM, et al. Cyclosporine and experimental skin aUografts. Transplantation 1985; 40: 714-8. 19. Horaguchi A, Merrell RC. Preparation of viable islet cells from dogs by a new method. Diabetes 1981; 30: 455-8. 20. Cobb LF, Merrell RC. Total pancreatectomy in dogs. J Surg Res 1984; 37: 235-40. 21. Lacy PE, Walker MM, Fink CJ. Perifusion of isolated rat islets in vitro. Diabetes 1972; 21: 987-98. 22. Hill BT, Whatley S. A simple, rapid miroassay for DNA. Fed Exp Biol Sci 1975; 56: 20-3. 23. Morgan CR, Lazarow A. Immunoassay of insulin: two antibody system. Diabetes 1963; 12:115-26. 24. Marincola F, Frank W, Clark W, Douglas M, MerreU R. The independence of insulin release an ambient insulin in vitro. Diabetes 1983; 32: 1162-7. 25. Rynasiewicz J J, Sutherland DE, Ferguson RM, et al. Cyclosporine A for immunosuppression: observations in rat heart, pancreas and islet allograft models and in human renal and pancreas transplantation. Diabetes 1982; 31(suppl 4): 92-108.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 156 SEPTEMBER 1988
193
he advantages of cyclosporin A (CsA) immunosuppression are considerable and have made this drug T the agent of choice for clinical transplantation of the heart, kidneys, bone marrow, and liver [I-4]. CsA has also been used in pancreatic transplantation, either alone or in combination with azathioprine and prednisone [5]. The addition of CsA to the immunosuppressive regimen improves glucose tolerance in patients with functioning pancreatic allografts [6-8]. It also offers therapeutic promise in incipient type I diabetes mellitus for suppressing the autoimmune reaction destroying beta cells [9,10]. However, CsA has also shown a poorly understood toxicity towards pancreatic islets both in vitro and in vivo. It has been associated with poor results when employed as an immunosuppressant in islet anografts [11], and intrasplenic canine islet autografts can be compromised by it [12]. In a recent study, Gores et al [13] did not confirm CsA toxicity to canine autografts, but in that study, the incidence of failure of untreated autografts may have obscured the effect. Certain patients with functioning pancreatic and renal allografts who are taking azathioprine and steroids have shown significant deterioration of glucose clearance after conversion to CsA [14]. This deterioration reversed in one case after resuming the origiFrom the Department of Surgery, The University of Texas Medical School at Houston, Houston, Texas. Request s for reprints should be addressed to Ronald C. Merrell, MD, Department of Surgery, The University of Texas Medical School at Houston, 6431 Fannin, Suite 4292, Houston, Texas 77030.
hal immunosuppression regimen. In vitro prolonged exposure to very high CsA concentrations impaired mouse islet beta cell function, and the detrimental effect was attributed to impaired protein biosynthesis [15]. High doses of CsA in rats caused severe morphologic and functional alterations of the pancreatic B cells with subsequent hyperglycemia and hypoinsulinemia [16]. CsA has a similar negative impact on some specific aspects of the subcellular events of wound healing, although this issue remains controversial [17,18]. One source of difficulty in determining subtle toxicity to islets in vivo is the great redundance of beta cells with respect to metabolic demand, which can mask even extraordinary toxicity. Measurement of acute toxicity on isolated islets does not take into account the long-term aspect of potential in vivo toxicity, This report addresses the effect of long-term cyclosporine administration on canine islets. The model entails a 3 week course of therapeutic doses of CsA and subsequent isolation of the islet to assess in vitro function on a per cell basis. MATERIAL AND METHODS Ten adult mongrel dogs received either a regimen of long-term immunosuppression for 3 weeks with CsA (Sandimmune | Sandoz, Geneva, Switzerland) or served as control subjects. The CsA regimen was 20 mg/kg/day during the first week, 10 mg/kg/day during the second week, and 5 mg/kg/day during the third week. The drug was given by mouth and administration was not difficult. This decreasing dose is similar to clinical regimens. Fasting blood glucose levels were determined using a YSI model 23A glucose analyzer (Yellow Springs Instruments, Yellow Springs, OH). Animals were screened for norm0glycemia before the study and were followed until death. All animals underwent total pancreatectomy after 3 weeks, and islets were isolated as previously described [19]. Briefly, under general inhalant anesthesia, a total pancreatectomy was performed [20]. The pancreas was digested by perfusing the ductal system with 0.2 percent type I collagenase (Sigma, St. Louis, MO) in Hank's balanced salt solution (Gibco, Grand Island, IL). After 40 minutes of perfusion at 37~ the pancreatic tissue was minced, filtered, and washed in Hank's balanced salt solution. The pancreatic fragments were suspended in RPMI 1640 (Gibco) containing 25 mM of N-2-hydroxyethyl-piperazine-N-2-ethanesul-fonic acid (HEPES) and 10 percent calf serum (Gibco). The islet-cell function was assessed according to the perifusion model described by Lacy et al [21]. Approximately 1 X 10 6 cells were loaded directly onto a 2 ml, 25 mm diameter filter holder containing a 1.2 ~z millipore filter and a prefilter. Flow was maintained with a peristaltic pump at 1 ml/min. After 30 minutes of perifusion with Hank's solution containing 25 mM HEPES and 2.75 mM glucose at pH 7.4, the ceils were challenged with Medium 199 (Gibco), containing
THE AMERICAN JOURNAL OF SURGERY
VOLUME 156 SEPTEMBER 1988
191
BASADONNA ET AL
200
20 -
9 CONTROL
18"
9
180
9 CsA
7< 160 Q
16"
12' ,~ 10' "n 8 ' m 6' Z
g
140
2
120 100
I
~ so _~ 60 ~ 40
~r
20
0
,
0
5
10
'
15
20
25
0
30
CsA
CONTROL
TIME (MINUTES)
Figure 1. Total Insulin output per microgram of DNA during 30 minutes of stimulation with 22.2 mM of glucose. Islets retrieved from animals treated with cyclosporin A (CsA) showed a significant decrease in insulin release compared with islets in the control animals. Asterick indicates, p <0.01. TABLE I
Insulin Output (/zU/min/#g DNA) After 22.2 mM Glucose Challenge in Control Animals and Cyclosporin A(CsA)Treated Animals* Time (rain) 0 1 3 5 8 11 15 30
Groups Control 3.07 17.05 13.11 10.67 6.57 4.49 3.03 2.95
4- 0.77 4- 2,90 4- 2,95 4- 2.97 4- 2.06 =1= 1.11 -I- 0.45 4- 0,22
CsA 1.08 1.62 2.11 2.07 1.84 1.75 1.57 1.54
+ 0.24 4- 0.30 :E 0.63 ::E 0.58 4- 0.39 4- 0.33 4- 0.32 4- 0.29
p Value <0.05 <0.001 <0.01 <0.05 NS <0.05 <0.05 <0.01
* COntrol and CsA values expressed as the mean 4- standard deviation. NS = not significant.
22.2 mM of glucose. Samples were collected at 0, 1, 3, 5, 8, 11, 15, and 30 minutes and stored at -40~ The number of cells employed in each perfusion was approximated by measuring DNA by Hill and Whatley's method [22]. Precise cell counting is not possible in these preparations, and islet number is not a particularly accurate measure either due to differences in islet size. However, DNA content per diploid cell is constant in a given species and can be used to reflect cell number. Insulin output was then expressed in micrograms of DNA. Insulin concentrations were determined using a modification of the double antibody insulin radioimmunoassay decribed by Morgan and Lazarow [23] and Marincola et al [24]. Data are expressed as the mean 4- standard error of the mean. Student's t test for impaired data was performed. RESULTS All animals, treated and untreated, remained normoglycemic throughout the study, and the CsA regimen was well tolerated. Islet tissue retrieved from control animals showed a total insulin output significantly higher than that from animals treated with CsA for 3 weeks (162.94 4- 25.05 gU insulin/30 min/#g DNA versus 192
THE A M E R I C A N J O U R N A L O F S U R G E R Y
Figure 2. Insulin release per minute per microgram of DNA after challenge with 22.2 mM of glucose. At each time p o i n t , except time 8, islets from animals treated with cyclnsporin A (CsA) showed a significant decrease in insulin release compared with islets in control animals.
50.81 4- 9.56 #U insulin/30 min/#g DNA, respectively; p <0.01) (Figure 1). Figure 2 shows the insulin release per microgram of DNA at the single time points after glucose challenge. The islet tissue from animals treated with CsA was impaired for both the first and the second peak of insulin release when compared with islet fragments isolated from the control animals (1.62 4- 0.30 #U insulin/min/#g DNA versus 17.05 4- 2.90 #U insulin; p <0.001 and 1.75 4- 0.33 #U insulin/min/#g DNA versus 4.49 4- 1.11 #U insulin/min//zg DNA, respectively; p <0.05). Every time point in the curve of insulin release, with the exception of 8 minutes, showed a statistically significant decrease in insulin output from the islet tissue of immunosuppressed animals compared with the control values (Table I). COMMENTS These data clearly demonstrate an untoward effect of CsA on the islets of Langerhans. The mechanism could be due to functional abnormalities of intact beta cells or to a decrease in the number of beta cells. Clinical data suggest that CsA does not affect pancreatic beta cells directly but induces insulin resistance, leading to a marked deterioration of glucose tolerance [14]. However, Rynasiewicz et al [25] found no significant difference in glucose, C-peptide, and insulin levels when they compared pancreas allograft recipients treated with conventional immunosuppression to those treated with CsA. Our studies with islet autografts in dogs strongly implicated a negative impact of CsA on islet function [12]. CsA should not be considered the immunosuppressive drug of choice in islet allotransplantation [11]. CsA also has a more subtle but still deleterious effect on islet autotransplantation. Intrasplenic islet autografts restore normal glucose metabolism in all cases, but a 50 percent failure rate occurs when CsA is introduced as the only variable [12]. It was impossible to interpret the detrimental effect of CsA on grafted islet cells when working with intact animals. When Helmchen et al [16] used very high doses of CsA in rats, there was a pronounced effect on pancreatic
V O L U M E 156
SEPTEMBER 1988
CYCLOSPORIN A AND ISLET FUNCTION
beta cell morphologic characteristics, with vacuolization of the cytoplasm and loss of secretory granules. From these data, it is perhaps surprising that intact animals showed no more abnormalities after CsA administration. The explanation for this discrepancy is probably the great redundancy of the islet mass, which can be decreased by at least 80 percent before carbohydrate metabolism is significantly impaired. Our study assessed the effect of CsA on the endocrine pancreas in vivo by removing the potentially intoxicated islets for evaluation on a more or less per islet basis in vitro. The negative effect of CsA on insulin output was well demonstrated by the marked decrease of insulin release from the islet tissue isolated from the CsA-treated animals. Our results underscore the importance of further efforts to determine the mechanism of CsA damage to the beta cells of the islets of Langerhans.
Acknowledgment:We thank Bobbie Lovett for manuscript preparation, and we also appreciate the technical support of Tam Phan and Mike Linzel. REFERENCES 1. Morris PJ. Cyclosporine A. Transplantation 1981; 32: 349-54. 2. Morris PJ. The impact of cyclosporine A on transplantation. Chicago: Yearbook Medical, 1984: 99. 3. Merion RM, White D J, Thiru S, Evans DB, Calne RY. Cyclosporine: five years experience in cadaveric renal transplantation. N Engl J Med 1984; 310: 148-54. 4. Starzl TE, Klintmalm GB, Porter KA, Iwatsuki S, Schroter GP. Liver transplantation with use of cyclosporin A and prednisone. N Engl J Med 1981; 305: 266-9. 5. Sutherland DER, Goetz FC, Najarian JS. Recent experience with 89 pancreas transplants at a single institution. Diabetologia 1984; 27:14953. 6. Traeger J, Dubernard JM, Pozza G, et al. Influence of immunosuppressive therapy on the endocrine function of segmental pancreatic allografts. Transplant Proc 1983; 15: 1326-9. 7. Pozza G, Traeger J, Dubernard JM, Secchi A, Bosi A, Pontiroli AE. Cyclosporin and glucose tolerance in pancreas allotransplantation. Lancet 1983; 2: 1080-1. 8. Pozza G, Traeger J, Dubernard JM, et al. Endocrine responses of type 1 (insulin-dependent) diabetic patients following successful pancre-
as transplantation. Diabetologia 1983; 24: 244-8. 9. Nerup J, Lernmark A. Autoimmunity in insulin dependent diabetes. Am J Med 1981; 70: 135-41. 10. Stiller CR, Dupre J, Gent M, et al. Effects of cyclosporine immunosuppresion in insulin-dependent diabetes mellitus of recent onset. Science 1984; 23: 1362-7. 11. DuToit DF, Reece-Smith H, McShane P, Denton T, Morris PJ. Effect of cyclosporin A on allotransplanted pancreatic fragments to the spleen of totally pancreatectomized dogs. Transplantation 1982; 33: 302-7. 12. Merrel RC, Mahoney ME, Basadonna G, Cobb LF, Maeda M. Failure of canine islet aUografts and autografts with cyclosporine A. Surgery 1985; 98: 324-9. 13. Gores PF, Boudreaux JP, Hesse U J, Najarian JS, Sutherland DER. Canine islet autografts with and without administration of cyclosporine. Surgery 1987; 101: 557-61. 14. Gunnarsson R, Klintmalm G, Lundgreu G, Wilczek H, Ostman J, Groth CG. Deterioration in glucose metabolism in pancreatic transplant recipients given cyclosporin. Lancet 1983; 2: 571-2. 15. Andersson A, Borg H, Hallberg A, Hellerstrom C, Sandier S, Schnell A. Long-term effects of cyclosporine A on cultured mouse pancreatic islets. Diabetologia 1984; 27: 66-9. 16. Helmchen V, Schmidt WE, Siegel EG, Creutzfeldt W. Morphological and functional changes of paneratic B cells in cyclosporin A-treated rats. Diabetologia 1984; 27: 416-8. 17. Eisinger DR, Sheil AG. A comparison of the effects of cyclosperine A and standard agents on primary wound healing in the rat. Surg Gynecol Obstet 1985; 160: 135-8. 18. Towpick E, Kupiec-Weglinski JW, Schneider TM, et al. Cyclosporine and experimental skin aUografts. Transplantation 1985; 40: 714-8. 19. Horaguchi A, Merrell RC. Preparation of viable islet cells from dogs by a new method. Diabetes 1981; 30: 455-8. 20. Cobb LF, Merrell RC. Total pancreatectomy in dogs. J Surg Res 1984; 37: 235-40. 21. Lacy PE, Walker MM, Fink CJ. Perifusion of isolated rat islets in vitro. Diabetes 1972; 21: 987-98. 22. Hill BT, Whatley S. A simple, rapid miroassay for DNA. Fed Exp Biol Sci 1975; 56: 20-3. 23. Morgan CR, Lazarow A. Immunoassay of insulin: two antibody system. Diabetes 1963; 12:115-26. 24. Marincola F, Frank W, Clark W, Douglas M, MerreU R. The independence of insulin release an ambient insulin in vitro. Diabetes 1983; 32: 1162-7. 25. Rynasiewicz J J, Sutherland DE, Ferguson RM, et al. Cyclosporine A for immunosuppression: observations in rat heart, pancreas and islet allograft models and in human renal and pancreas transplantation. Diabetes 1982; 31(suppl 4): 92-108.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 156 SEPTEMBER 1988
193