- Email: [email protected]
Pancreatic transplantation: The cambridge experience
Pancreatic Transplantation: The Cambridge Experience
R. Y. Calne, MD I. G. M. Br0ns, PhD
Oepartment of Surgery, University of Cambridge, Addenbrooke's Hospital, Cambrfdge, United Kingdom
ABSTRACT In contrast to heterotopic pancreas transplantation and conventional insulin therapy the paratopic positioning of a segmental pancreas graft provides physiological endocrine hormone delivery into the portal venous circulation. Metabolic studies performed on previously insulin dependent diabetic patients with renal failure with functioning paratopic pancreas transplants and heterotopic kidney grafts from the same donor showed near normal day to day glucose control with normal fasting glucose levels and normal HbA1 values. Hyperinsulinaemia was not seen in these patients. The effect of denervation of the pancreatic graft on the entero-insular axis and its influenceon the islets were investigated. The incretin effect was preserved indicating a significant hormonal stimulation of the enteroinsular axis. (The Journal of Diabetic Complications 2;3:121-124, 1988.)
INTRODUCTION The introduction of insulin for the treatment of diabetes was one of the greatest medical advances of this century and Banting, who was instrumental in this development, wrote in his notebooks that after obtaining an extract that could be used for the treatment of diabetes, the next step would be to transplant the pancreas so the diabetes could be cured. It is only recently that attempts at transplantation of the pancreas seem to have achieved this end, with the cure of diabetes in some patients. To obtain a more physiologic drainage of insulin into the portal system we recently concentrated on a technique in which a pancreas is transplanted very close to the patient's own pancreas, the so-called paratopic position, between the spleen and the stomach, anastomosing the graft splenic vessels end-to-side to the splenic vessels of the recipient and draining pancreatic juice into the stomach (Figure 1). 1 The operation can be technically difficult, since in many diabetic patients there is severe sclerotic disease of the splenic artery and access to the recipient splenic vessels may not be easy.
A SUMMARY OF THE CAMBRIDGEEXPERIENCE This paper was presented, in part, at the Diabetic Renal-Retinal Syndrome IV, New York, November 1987. Reprint requests: to be sent to Sir Roy Calne, Department of Surgery, University of Cambridge, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, United Kingdom.
Between August 1979 and October 1987, thirty-seven segmental pancreas grafts were performed, the details of which are shown in Table 1. Fourteen patients have died; a high mortality due in part to their severe long-'standing diabetes. The causes of death are shown in Figure 2. In 32 patients, transplantation of the pancreas was accompanied by simultaneous grafting of the kidney from the same donor and, in one, of the liver (Figure 3). The patients with continuing renal function are listed in Figure 4.
121
122
CALNE AND BRONS
Fluid overload Aspergillosis Infection: CMV pneumocystis pneumonia candida peritonitis liver abscess (death with functioning kidney grafts) n=7
n=l n=2 n=11
FIG. 2 Deaths related or unrelated to pancreas graft (n = 14).
\
Pancreas with simultaneous kidney Pancreas with simultaneous liver Pancreas only (2 retransplants)
FIG. 1 Diagram of the paratopic technique for pancreatic transplantation. The body and tail of the pancreas lie close to the stomach and the spleen. The splenic vessels of the segment are anastomosed end-to-side to those of the recipient. A small arteriovenous fistula has been constructed distal to the tip Of the tail of the pancreas to discourage thrombosis. A soft feeding tube ties in the pancreatic duct and is brought out through the lumen of the stomach via the anterior gastric wall The omenturn is wrapped around the graft.
Metabolic studies were conducted by Brons and Calne 2 and Clark et aL 3 in the patients with renal and paratopic pancreas grafts. Because the aim of pancreas transplantation is normalization of carbohydrate metabolism, which one hopes will stabilize or even reverse some of the severe and progressive complications of diabetes, it was decided to study glucose metabolism after paratopic grafting with portal venous hormone delivery. Pretransplant levels of serum C-peptide, HbA1, serum creatinine, and exogenous insulin requirements were recorded. Postransplantation fasting blood glucose levels, 24 hour profiles, and glucose tolerance tests with measurements of glucose, C-peptide, total insulin, free insulin, and proinsulin were performed. Excluding technical failures (six patients) all patients were insulin independent and had normal fasting blood glucose levels after grafting. HbA1 levels returned to normal within
TABLE 1
n=32 n= 1 n= 6
FIG. 3 Number of transplants (n = 39).
n=2O
n=12
Non-function occurring between 1 month and 4.5 years. (death with functioning kidney grafts n = 7) fully functioning kidney grafts from > 1 month to > 5 years.
FIG. 4 Kidney function in patients with pancreas graft (n = 32).
a short time, while 24 hour metabolic profiles showed near normal glucose control throughout the day with characteristic serum C-peptide peaks after meals. It was, however, shown that fasting C-peptide levels were raised when compared with healthy volunteers, but the same elevated levels were seen in nondiabetic patients with kidney transplants. Because the kidney is the major site of C-peptide metabolism and clearance, slightly impaired renal function with raised serum creatinine levels in and of itself will result in increased blood levels of C-peptide. Although this connecting peptide is released in equimo-
Cambridge Experience with Segmental Pancreas Transplantation: August 1979 to October 1987 Pancreatic function
Technique Heterotopic duct injection Roux-en-y loop bladder Paratopic stomach Roux-en-y loop Total
Transplants
Patients
11 7 2
10 6 2
18 1 39
18 1 37
Part Function 1 (>4.5 yrs) "'- ~ 2 (>3 yrs) 3
Full Function
Failed Transplanls
1 (>8 yrs) 1 (>4 yrs) 1 (-t-1 mos)
10 5 1
4 (>2,2.5,3,3.5 yrs) 1 (>6 mos) 8
12 28
PANCREATICTRANSPLANTATIONIN CAMBRIDGE lar amounts with insulin and is generally regarded as the only measure of total insulin release, it does not reflect true insulin output when renal function is impaired. Oral glucose tolerance tests with a glucose load of 75 g showed a slight impairment of glucose disappearance in some patients. However, blood glucose returned to normal levels (below 8 mmol/L) within the third or fourth hour after the test. Total as well as free insulin levels ranged within the values of healthy volunteers, as did proinsulin levels measured by a new sensitive and specific immunoradiometric assay for the intact proinsulin molecule (estimations kindly performed by Professor Hales, Department of Clinical Biochemistry, University of Cambridge). In contrast to conventional insulin therapy and heterotopic pancreas transplantation, peripheral hyperinsulinemia was not seen in our patients. 4 Two nondiabetic patients with partial pancreaticoduodenectomy for pancreatic adenocarcinoma (Whippie's operation) with a similar amount of pancreatic tissue left in situ and a similar exocrine drainage technique as the paratopically transplanted patients, but with normal k dney I function, displayed similar glucose, insulin, free insulin; and proinsulin results, but had normal fasting C-peptide levels. A study was also made of the hormone responses of the denervated pancreas and the influence of the enteroinsular iaxis on the islets of Langerhans. 3 The difference between the insulin response to an oral glucose load and that after a similar glucose load given intravenously is called the "incretin effect", s produced by the enteroinsular axis and thought to be under neuronal and hormonal control. However, it is currently unknown whether insulin secretion is significantly affected by peptiderigic, cholinergic, or adrenergic innervation of the islets of Langerhans. The most likely hormonal candidate to produce such an incretin effect is the glucose-dependent insulinotropic peptide (GIP), which is secreted by epithelial cells from the upper small intestine. The aim of the study was to investigate whether denervation of the pancreas impaired the hormone responses and reduced the incretin effect. Five previously insulin-dependent diabetic patients with renal failure simultaneously transplanted with paratopic segmental pancreata and kidney grafts from the same donors were investigated 6 months to 2 years after transplantation. Oral glucose tolerance tests (OGTT) with a 75 g glucose load were followed 1 week later with intravenous glucose infusion (IVGTT) to mimic the oral glucose curve. The incretin effect was preserved in the patient group and did not differ significantly from the control group. Augmented insulin response to an oral glucose load despite denervation indicated a significant hormonal contribution to the enteroinsular axis. The size of the incretin effect was similar in both groups. GIP levels were shown to be within the same range in both control and patient samples. (Estimations kindly performed by Professor Bloom, Hammersmith Hospital). These results indicate the normal relationship of hor . . . . mones involved in the enteroinsular axis in patients transplanted with a denervated segmental pancreas graft in the physiologic paratopic position.
123
FIG. 5 Successful pregnancy 3 years after paratopic segmental pancreas with simultaneous kidney transplantation resulted in normal baby.
CONCLUSIONS.
9
A particularly encouraging aspect of pancreas transplantation is the complete restoration to full normal activity of some patients who have good function both in pancreas and kidney grafts. A patient whose life has become a misery from dialysis three times a week, restricted fluid intake, strict diet, multiple insulin injections and, despite all this, progressive worsening of the complications of diabetes can become a normal individual again by taking small doses of immunosuppressants. Rehabilitation has been full in all of our patients with a functioning pancreas allograft. Several have resumed participation in sports, and one had a normal baby 3 years after transplantation
(Figure 5). In diabetics with renal failure the vascularized pancreas would seem to be the best option currently available. In the future there may be advances in beta cell or islet grafting, as well as developments in genetic engineering, it should not be too difficult to transfect the insulin-producing gene into a dividing cell, but to provide that cell with a control feedback mechanism that will be sensitive to the blood level of glucose may be difficult to achieve. In addition, there might be a danger ~_at such cells would divide in an uncontrolled manner and form tumors. An approach that is being studied intensively i s t o enclose islets in porous membranes, the most encouraging at present being polymer membranes
124 through which the islets can receive nutrition and release insulin as needed according to the blood sugar levels. Because cells cannot penetrate the envelope, the grafts are not rejected, and islets from another species, for example, rodents, can be used in humans. It seems unlikely that such a technique would provide a permanent source of insulin and it is probable that islet packages would need to be renewed every few months. Because diabetic microangiopathy is such a devastating complication the challenge to produce better results in this field is enormous, and the limited success already obtained with vascularized pancreas grafts is a signpost to future progress.
CALNE AND BRONS
REFERENCES 1. Calne RY: Paratopic segmental pancreas grafting: A technique with portal venous drainage. Lancet 1:595-597, 1984. 2. Brons IGM, Calne RY: Pancreas transplantation: The cambridge experience. Transplant Proc XlX, 4:11-16, 1987. 3. Clark JDA, Wheatley T, Brons IGM, Bloom SR, Calne RY: Insulin and glucose-dependent insulino opic peptide (GIP) secretion following pancreas transplantation in man. Diabetic Medicine 4:352, 1987. 4. Brons IGM, Calne RY, Rolles K, Clark JDA, Wheatley T: Glucose stimulated hormone responses after paratopic pancreas transplantation. Transplant Proc XIX, 5:3880-3881, 1987. 5. Creutzfeld W, Ebert R: New developments in the incretin concept. Diabetologia 28:565-573, 1985.
R. Y. Calne, MD I. G. M. Br0ns, PhD
Oepartment of Surgery, University of Cambridge, Addenbrooke's Hospital, Cambrfdge, United Kingdom
ABSTRACT In contrast to heterotopic pancreas transplantation and conventional insulin therapy the paratopic positioning of a segmental pancreas graft provides physiological endocrine hormone delivery into the portal venous circulation. Metabolic studies performed on previously insulin dependent diabetic patients with renal failure with functioning paratopic pancreas transplants and heterotopic kidney grafts from the same donor showed near normal day to day glucose control with normal fasting glucose levels and normal HbA1 values. Hyperinsulinaemia was not seen in these patients. The effect of denervation of the pancreatic graft on the entero-insular axis and its influenceon the islets were investigated. The incretin effect was preserved indicating a significant hormonal stimulation of the enteroinsular axis. (The Journal of Diabetic Complications 2;3:121-124, 1988.)
INTRODUCTION The introduction of insulin for the treatment of diabetes was one of the greatest medical advances of this century and Banting, who was instrumental in this development, wrote in his notebooks that after obtaining an extract that could be used for the treatment of diabetes, the next step would be to transplant the pancreas so the diabetes could be cured. It is only recently that attempts at transplantation of the pancreas seem to have achieved this end, with the cure of diabetes in some patients. To obtain a more physiologic drainage of insulin into the portal system we recently concentrated on a technique in which a pancreas is transplanted very close to the patient's own pancreas, the so-called paratopic position, between the spleen and the stomach, anastomosing the graft splenic vessels end-to-side to the splenic vessels of the recipient and draining pancreatic juice into the stomach (Figure 1). 1 The operation can be technically difficult, since in many diabetic patients there is severe sclerotic disease of the splenic artery and access to the recipient splenic vessels may not be easy.
A SUMMARY OF THE CAMBRIDGEEXPERIENCE This paper was presented, in part, at the Diabetic Renal-Retinal Syndrome IV, New York, November 1987. Reprint requests: to be sent to Sir Roy Calne, Department of Surgery, University of Cambridge, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, United Kingdom.
Between August 1979 and October 1987, thirty-seven segmental pancreas grafts were performed, the details of which are shown in Table 1. Fourteen patients have died; a high mortality due in part to their severe long-'standing diabetes. The causes of death are shown in Figure 2. In 32 patients, transplantation of the pancreas was accompanied by simultaneous grafting of the kidney from the same donor and, in one, of the liver (Figure 3). The patients with continuing renal function are listed in Figure 4.
121
122
CALNE AND BRONS
Fluid overload Aspergillosis Infection: CMV pneumocystis pneumonia candida peritonitis liver abscess (death with functioning kidney grafts) n=7
n=l n=2 n=11
FIG. 2 Deaths related or unrelated to pancreas graft (n = 14).
\
Pancreas with simultaneous kidney Pancreas with simultaneous liver Pancreas only (2 retransplants)
FIG. 1 Diagram of the paratopic technique for pancreatic transplantation. The body and tail of the pancreas lie close to the stomach and the spleen. The splenic vessels of the segment are anastomosed end-to-side to those of the recipient. A small arteriovenous fistula has been constructed distal to the tip Of the tail of the pancreas to discourage thrombosis. A soft feeding tube ties in the pancreatic duct and is brought out through the lumen of the stomach via the anterior gastric wall The omenturn is wrapped around the graft.
Metabolic studies were conducted by Brons and Calne 2 and Clark et aL 3 in the patients with renal and paratopic pancreas grafts. Because the aim of pancreas transplantation is normalization of carbohydrate metabolism, which one hopes will stabilize or even reverse some of the severe and progressive complications of diabetes, it was decided to study glucose metabolism after paratopic grafting with portal venous hormone delivery. Pretransplant levels of serum C-peptide, HbA1, serum creatinine, and exogenous insulin requirements were recorded. Postransplantation fasting blood glucose levels, 24 hour profiles, and glucose tolerance tests with measurements of glucose, C-peptide, total insulin, free insulin, and proinsulin were performed. Excluding technical failures (six patients) all patients were insulin independent and had normal fasting blood glucose levels after grafting. HbA1 levels returned to normal within
TABLE 1
n=32 n= 1 n= 6
FIG. 3 Number of transplants (n = 39).
n=2O
n=12
Non-function occurring between 1 month and 4.5 years. (death with functioning kidney grafts n = 7) fully functioning kidney grafts from > 1 month to > 5 years.
FIG. 4 Kidney function in patients with pancreas graft (n = 32).
a short time, while 24 hour metabolic profiles showed near normal glucose control throughout the day with characteristic serum C-peptide peaks after meals. It was, however, shown that fasting C-peptide levels were raised when compared with healthy volunteers, but the same elevated levels were seen in nondiabetic patients with kidney transplants. Because the kidney is the major site of C-peptide metabolism and clearance, slightly impaired renal function with raised serum creatinine levels in and of itself will result in increased blood levels of C-peptide. Although this connecting peptide is released in equimo-
Cambridge Experience with Segmental Pancreas Transplantation: August 1979 to October 1987 Pancreatic function
Technique Heterotopic duct injection Roux-en-y loop bladder Paratopic stomach Roux-en-y loop Total
Transplants
Patients
11 7 2
10 6 2
18 1 39
18 1 37
Part Function 1 (>4.5 yrs) "'- ~ 2 (>3 yrs) 3
Full Function
Failed Transplanls
1 (>8 yrs) 1 (>4 yrs) 1 (-t-1 mos)
10 5 1
4 (>2,2.5,3,3.5 yrs) 1 (>6 mos) 8
12 28
PANCREATICTRANSPLANTATIONIN CAMBRIDGE lar amounts with insulin and is generally regarded as the only measure of total insulin release, it does not reflect true insulin output when renal function is impaired. Oral glucose tolerance tests with a glucose load of 75 g showed a slight impairment of glucose disappearance in some patients. However, blood glucose returned to normal levels (below 8 mmol/L) within the third or fourth hour after the test. Total as well as free insulin levels ranged within the values of healthy volunteers, as did proinsulin levels measured by a new sensitive and specific immunoradiometric assay for the intact proinsulin molecule (estimations kindly performed by Professor Hales, Department of Clinical Biochemistry, University of Cambridge). In contrast to conventional insulin therapy and heterotopic pancreas transplantation, peripheral hyperinsulinemia was not seen in our patients. 4 Two nondiabetic patients with partial pancreaticoduodenectomy for pancreatic adenocarcinoma (Whippie's operation) with a similar amount of pancreatic tissue left in situ and a similar exocrine drainage technique as the paratopically transplanted patients, but with normal k dney I function, displayed similar glucose, insulin, free insulin; and proinsulin results, but had normal fasting C-peptide levels. A study was also made of the hormone responses of the denervated pancreas and the influence of the enteroinsular iaxis on the islets of Langerhans. 3 The difference between the insulin response to an oral glucose load and that after a similar glucose load given intravenously is called the "incretin effect", s produced by the enteroinsular axis and thought to be under neuronal and hormonal control. However, it is currently unknown whether insulin secretion is significantly affected by peptiderigic, cholinergic, or adrenergic innervation of the islets of Langerhans. The most likely hormonal candidate to produce such an incretin effect is the glucose-dependent insulinotropic peptide (GIP), which is secreted by epithelial cells from the upper small intestine. The aim of the study was to investigate whether denervation of the pancreas impaired the hormone responses and reduced the incretin effect. Five previously insulin-dependent diabetic patients with renal failure simultaneously transplanted with paratopic segmental pancreata and kidney grafts from the same donors were investigated 6 months to 2 years after transplantation. Oral glucose tolerance tests (OGTT) with a 75 g glucose load were followed 1 week later with intravenous glucose infusion (IVGTT) to mimic the oral glucose curve. The incretin effect was preserved in the patient group and did not differ significantly from the control group. Augmented insulin response to an oral glucose load despite denervation indicated a significant hormonal contribution to the enteroinsular axis. The size of the incretin effect was similar in both groups. GIP levels were shown to be within the same range in both control and patient samples. (Estimations kindly performed by Professor Bloom, Hammersmith Hospital). These results indicate the normal relationship of hor . . . . mones involved in the enteroinsular axis in patients transplanted with a denervated segmental pancreas graft in the physiologic paratopic position.
123
FIG. 5 Successful pregnancy 3 years after paratopic segmental pancreas with simultaneous kidney transplantation resulted in normal baby.
CONCLUSIONS.
9
A particularly encouraging aspect of pancreas transplantation is the complete restoration to full normal activity of some patients who have good function both in pancreas and kidney grafts. A patient whose life has become a misery from dialysis three times a week, restricted fluid intake, strict diet, multiple insulin injections and, despite all this, progressive worsening of the complications of diabetes can become a normal individual again by taking small doses of immunosuppressants. Rehabilitation has been full in all of our patients with a functioning pancreas allograft. Several have resumed participation in sports, and one had a normal baby 3 years after transplantation
(Figure 5). In diabetics with renal failure the vascularized pancreas would seem to be the best option currently available. In the future there may be advances in beta cell or islet grafting, as well as developments in genetic engineering, it should not be too difficult to transfect the insulin-producing gene into a dividing cell, but to provide that cell with a control feedback mechanism that will be sensitive to the blood level of glucose may be difficult to achieve. In addition, there might be a danger ~_at such cells would divide in an uncontrolled manner and form tumors. An approach that is being studied intensively i s t o enclose islets in porous membranes, the most encouraging at present being polymer membranes
124 through which the islets can receive nutrition and release insulin as needed according to the blood sugar levels. Because cells cannot penetrate the envelope, the grafts are not rejected, and islets from another species, for example, rodents, can be used in humans. It seems unlikely that such a technique would provide a permanent source of insulin and it is probable that islet packages would need to be renewed every few months. Because diabetic microangiopathy is such a devastating complication the challenge to produce better results in this field is enormous, and the limited success already obtained with vascularized pancreas grafts is a signpost to future progress.
CALNE AND BRONS
REFERENCES 1. Calne RY: Paratopic segmental pancreas grafting: A technique with portal venous drainage. Lancet 1:595-597, 1984. 2. Brons IGM, Calne RY: Pancreas transplantation: The cambridge experience. Transplant Proc XlX, 4:11-16, 1987. 3. Clark JDA, Wheatley T, Brons IGM, Bloom SR, Calne RY: Insulin and glucose-dependent insulino opic peptide (GIP) secretion following pancreas transplantation in man. Diabetic Medicine 4:352, 1987. 4. Brons IGM, Calne RY, Rolles K, Clark JDA, Wheatley T: Glucose stimulated hormone responses after paratopic pancreas transplantation. Transplant Proc XIX, 5:3880-3881, 1987. 5. Creutzfeld W, Ebert R: New developments in the incretin concept. Diabetologia 28:565-573, 1985.