Transplantation of the Pancreas

41. Transplantation of the Pancreas 719

Figure 41.1. Number of pancreas transplants reported in the International Pancreas Transplant Registry (IPTR) by continent, 1987–1999. (Reprinted from Cecka JM, Terasaki PI, eds. Clinical Transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000, with permission.)

Historical Perspective

Pancreas transplant (P Tx) remains the least recognized transplant to the lay public with the exception of small-bowel transplant. Although whole-organ P Tx was first performed in 1966, it remains much of a mystery to healthcare providers at large as well as to the general public because of the paucity of news surrounding it. More kidney transplants (K Tx) have been performed in 1 year in the United States than pancreas transplants worldwide from 1987 to 1999. According to the

International Pancreas Transplant Registry (IPTR), 13,372 kidney transplants were performed in the United States alone in 2000, whereas the total number of pancreas transplants performed worldwide from 1966 to 1999 was 12,939.1 Two thirds of the pancreas transplants performed are done in the United States, with one fourth in Europe (Fig. 41.1). With the advent of Medicare payment for pancreas transplants effective July 1, 1999, it is hoped that financial reimbursement will stimulate an increase in the total number of pancreas transplants performed. In fact, more information about islet cell transplant has emerged with the news of successful islet cell transplant in Edmonton, despite the small numbers and limited follow-up span.

Diabetes Mellitus

Overview

It is estimated that 6.3% or 18.2 million people in the United States have diabetes mellitus (DM). This includes 13 million people who are diagnosed and the estimated 5.2 million people who are undiagnosed.2

1 OPTN/SRTR 2001 Annual Report.

2 Centers for Disease Control and Prevention. National diabetes fact sheet: general information and national estimates on diabetes in the U.S., 2002. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2003.

720 J.W. Lim

The incidence of new cases diagnosed per year is 1.3 million people aged 20 years or older. Approximately one in every 400 to 500 children and adolescents has type 1 DM. Clinical reports and regional studies indicate that type 2 DM is becoming more common among Native American/American Indian, African American, and Hispanic/Latino children and adolescents. Diabetes was the sixth leading cause of death on U.S. death certificates in 2000. Overall, the risk for death among people with DM is about two times that of people without DM. In fact, those patients on dialysis who have DM have the highest rate of mortality on dialysis, approximating 28% per year. From a strictly economic point of view, it is estimated that direct and indirect expenditures attributable to DM in 2002 were $132 billion.3 Per capita expenditures totaled $13,243 for people with DM and $2,560 for people without DM. Healthcare spending for patients with DM is more than double the spending for patients without DM.

Classic Complications

Diabetes mellitus is a chronic, disabling disease that affects all the major organ systems and shortens the life span by 10 years for those people in the United States who are affected by it. The classic complications that occur consist of retinopathy (DM is the most common cause of blindness among adults aged 20 to 74), gastropathy/gastroparesis, nephropathy (DM is the most common cause of renal failure), neuropathy, vasculopathy (DM is the most common cause of amputations), labile control, and diabetic ketoacidosis (DKA).4

Strict control clearly has been shown to delay significantly the development and slow the progression of the microvascular complications of DM. There appears to be a trend for fewer macrovascular events (strokes, myocardial infarctions, etc.) in patients who are controlled aggressively with insulin. However, this is at the cost of a 10-pound greater weight gain and a threefold higher risk of severe hypoglycemia. These results were noted with the publication of the Diabetic Control and Complications Trial (DCCT) in the New England Journal of Medicine in 1993.5 For many patients, however, such aggressive control with its inherent problems proves to be as troublesome as the primary disease itself.

3 Reviews/commentaries/position statements report from the American Diabetes Association. Diabetes Care 2003;26:917–932.

4 Centers for Disease Control and Prevention. National diabetes fact sheet: general information and national estimates on diabetes in the U.S., 2002. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2003.

5 DCCT Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–986.

41. Transplantation of the Pancreas 723

A

B

Figure 41.3. (A) Donor iliac conduit is anastomosed to the splenic artery and the superior mesenteric artery of the pancreas on the back table. (B) Alternatively, the blood supply to the donor pancreas can be reconstituted by direct anastomosis of the splenic artery to the superior mesenteric artery. (Reprinted from Harland RC. Pancreas transplantation. In: Norton JA, Bollinger RR, Chang AE, et al, eds. Surgery: Basic Science and Clinical Evidence. New York: Springer-Verlag, 2001, with permission.)

extraneous tissue and stapling off the mesenteric vessels and duodenal stumps. The donor common iliac artery with its attached external and internal iliac artery is used as a Y graft to attach to the donor pancreas superior mesenteric artery (SMA) and splenic artery (SA) so as to use the common iliac artery as a conduit for the inflow of the pancreas (Fig. 41.3).

Three Options for Pancreas Transplant

Renal function determines which of the three options potential P Tx candidates will undergo.

Patients with renal failure or pending renal failure receive both a pancreas and kidney transplant, otherwise known as simultaneous pancreas and kidney (SPK) transplant (Table 41.1). Approximately 83% of all pancreas transplants performed in the U.S. are performed in this manner. This patient is then able to come off dialysis, much as a patient who receives a K Tx. This patient, however, would have the added benefit of no longer being a diabetic. The ramifications go

724 J.W. Lim

Table 41.1. Demographics: U.S. cadaver pancreas transplants, January 1996 to September 1999.

HLA, human leukocyte antigen; PAK, pancreas after kidney (transplant); PTA, pancreas transplant alone; SPK, simultaneous pancreas and kidney (transplant).

Source: Reprinted from Cecka JM, Terasaki PI, eds. Clinical Transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000, with permission.

further, since the natural history of patients with DM who receive only a K Tx indicates they have nearly a 100% recurrence of their DM in the transplant kidney histologically. Fortunately, this is clinically significant only 5% of the time. As with other transplants, results have improved with time (Table 41.2).

A second subset of patients receives a pancreas after kidney (PAK) transplant. These patients had renal failure because of DM and already have received a K Tx. Because some of the medications that are required posttransplant can exacerbate and worsen glucose control, diabetic complications worsen to the point that these patients now become candidates for PAK. The creatinine clearance (Cr Cl) that defines these patients is >50 cc/min. Approximately 10% of all pancreas transplants are performed in this manner.

Table 41.2. Outcome of U.S. cadaveric pancreas transplants by era of transplant (level II evidence).

Source: International Pancreas Transplant Registry (IPTR), Department of Surgery, University of Minnesota, Minneapolis, MN. With permission. Reprinted from Harland RC. Pancreas transplantation. In: Norton JA, Bollinger RR, Chang AE, et al, eds. Surgery: Basic Science and Clinical Evidence. New York: Springer-Verlag, 2001, with permission.

41. Transplantation of the Pancreas 725

The third and last subset of patients to receive a P Tx has the most to benefit from this procedure and the most to lose from diabetes. This subset makes up 7% of all the pancreas transplants performed in the U.S. today. These patients have DM and some of the classic complications, most often labile sugar control or hypoglycemic unawareness. Hypoglycemic unawareness is very uncommon, but it is defined as the inability to sense when the glucose level is nearing dangerously low levels, to the point of losing consciousness. Most diabetics can sense when their glucose levels are getting low and are able to take some sort of “high sugar” tablet or juice to keep them from losing consciousness or going into seizures. Those with hypoglycemic unawareness are not able to detect this and always are fearful that they may do themselves or others harm. However, they do not have renal failure yet or, at worst, they have some early renal insufficiency. These patients receive a pancreas transplant alone (PTA). The Cr Cl that delineates these patients is 60 to 70 cc/min. Since nearly 35% of all patients with DM go on to suffer renal failure after 15 to 20 years, and since it has been shown that a PTA can reverse all the diabetic changes in the native kidney after 5 to 10 years, this is a very viable option.7

Function of the Pancreas

Before the surgical technique can be explained, the function of the pancreas must be understood. The pancreas has two main functions, one endocrine and the other exocrine. The main endocrine function is insulin production. Insulin is produced mainly in the tail of the pancreas by the beta islet cells and released into the portal vein, from which it then goes to the liver to be metabolized in what is called the

first-pass phenomenon. The exocrine function is the production and release of the protease enzymes and bicarbonate. These are released via the pancreatic duct back into the gastrointestinal tract via the ampulla of Vater in the duodenum. The donor pancreas retains these two functions, and the surgical connections are related to these two functions. Hence, the donor portal vein is the conduit by which the insulin produced in the donor pancreas is released into the vascular system in the recipient. Likewise, the attached donor duodenum is the conduit by which the exocrine function is facilitated.

Surgical Techniques

The endocrine function can be drained into the iliac venous system of the recipient (systemic drainage) or into the mesenteric venous system of the recipient (portal drainage). The exocrine function can be drained either into the bladder (bladder drainage) or into the gastrointestinal tract (enteral drainage).

Most centers currently use systemic drainage of the endocrine function, and more and more centers are using enteral drainage of

7 Fioretto P, Steffes, MW, Sutherland DE, et al. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Engl J Med 1998;339:69–75.

728 J.W. Lim

this method is the theoretical improvement of the lipid profile of the recipient, since there is no longer any hyperinsulinemia. Other advantages consist of less rejection episodes and faster operative times.

To date, graft and patient survivals based on systemic versus portal drainage are similar. When portal drainage is used, the exocrine function drains back into the enteric system by way of the recipient loop of small bowel, be it a Roux-en-Y limb or adjacent recipient jejunum. Hence, this method preserves the physiologic nature of the pancreas by draining the endocrine function back into the portal system and the exocrine function back into the enteric system, hence, the terminology portal/enteral drainage. This is in distinction to systemic/bladder drainage, whereby the endocrine function drains into the iliac venous system and the exocrine functions drains via the bladder.

The advantages of the enteric drainage technique over the bladder drainage technique are many. First, the classic complications associated with draining active enzymes and bicarbonate with the resultant volume loss are alleviated. Therefore, the problems with hematuria, recurrent urinary tract infections (UTIs), reflux pancreatitis (reflux of the urine back into the donor pancreas), acidosis, and hypovolemia are avoided entirely by performing the enteric drainage technique. The ramifications of this cannot be understated. Approximately 12% of all patients drained via the bladder route must be corrected surgically within 2 years of the original P Tx.8 The surgical correction consists of taking down the bladder anastomosis and reconnecting the exocrine drainage back into the enteric system (Fig. 41.8).

The main advantage for performing the bladder drainage technique is to measure the urinary amylase produced by the donor pancreas as a means to diagnose rejection. A decrease of 25% to 60% of the baseline urinary amylase is used by these centers to monitor for rejection. Although the ability to collect the urinary amylase is lost with the enteric drainage technique, direct biopsy of the donor pancreas is used to diagnose rejection definitively. Monitoring of the donor pancreas is performed by checking serum amylase and lipase. Any persistent increase from the baseline serum amylase and lipase is taken as being abnormal, and, if appropriate measures do not indicate infection or technical problems, then a percutaneous U.S.-guided biopsy of the donor pancreas is performed.

The other main advantage of bladder drainage over enteric drainage is that leaks at the donor duodenal anastomosis are tolerated better if urine is leaking rather than enteric contents. Most often, a placement of a Foley catheter to decompress the bladder is enough to control the urinary leak. The main advantage of performing a Roux-en-Y limb anastomosis to the donor duodenum is that, should a leak occur, the enteric contents from a defunctionalized loop of bowel usually can be better controlled than if an adjacent loop of recipient jejunum is used.

8 Cecka JM, Terasaki. PI, Clinical Transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000:68.

730 J.W. Lim

Duodenal anastomosis leak occurs in up to 1.4% of all cases, with the highest incidence in enteric drained PTA (Table 41.3). As noted above, a Roux-en-Y connection facilitates control. If undetected, the leaks may lead to deep wound infections that may cause worsening renal or pancreatic function and potential multiorgan dysfunction. Much of the cause of the leak is secondary to the attenuated donor bowel wall, which may facilitate pulling through of sutures and resultant leak. The immunocompromised state does not lend itself to faster or better healing of the duodenal anastomosis. The patient usually has a nasogastric tube to help with the perioperative ileus associated with any intraabdominal procedure.

The usual postoperative course consists of 7 to 10 days of hospitalization, with frequent glucose monitoring in the first 48 hours postoperation. Thereafter, the patient recovers as with any other intraabdominal procedure and proceeds to ambulate as quickly as possible. Some centers keep their patients at bed rest for at least 48 to 72 hours postoperatively so as to minimize any chance of the donor pancreas moving and potentially kinking at the vascular anastomoses. Knowledge about immunosuppression medications is imperative, and patients are not discharged until they have proven that they know their medications. (See Immunosuppressive Drugs and Their Toxicities in Chapter 40 for an overview of immunosuppression medications.) The same immunosuppression medications used for K Tx also are used in P Tx, i.e., Prograf (tacrolimus or FK506), Neoral (cyclosporine), Cellcept, and prednisone.

See Algorithm 41.1 for postoperative monitoring.

Immunosuppression Medications

Recent immunosuppressive regimens have stressed corticosteroid sparing or elimination, with rejection-free survival rates approaching 90% at some centers. The PIVOT Study Group compared the conventional 5-day dose of daclizumab (DAC) versus a 2-dose DAC versus no antibody induction in SPK recipients receiving Prograf, Cellcept, and prednisone. The conclusions were that both DAC regimens significantly reduced the incidence of acuter rejection, with the 2-dose

Table 41.3. Reasons for technical failures: U.S. cadaver pancreas transplants, January 1996 to September 1999.

BD, bladder drained; ED, enteric drained.

Source: Reprinted from Cecka JM, Terasaki PI, eds. Clinical Transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000, with permission.

41. Transplantation of the Pancreas 731

Perioperative period Elevated glucose (above 200)

STAT Doppler ultrasound (US)

Repeat US within 24 hours

OR

Heparinize

Viable Nonviable

Algorithm 41.1. Algorithm for postoperative monitoring in pancreas transplant.

DAC regimen showing the highest event-free survival rate.10 Kaufman and colleagues11 presented their single-center, retrospective experience with a number of different protocols designed to minimize and then eventually withdraw steroids. They concluded that rapid steroid elimination can be achieved safely in SPK and PTA patients. Enough data have been reviewed to make some clear-cut recommendations on P Tx. For SPK, there appears to be very little difference with respect to graft survival using Prograf or Neoral as the first-line medication, whether

10Stratta RJ, Alloway RR, Hoge LA, PIVOT Investigators. One year outcomes in simultaneous kidney-pancreas transplant recipients receiving an alternative regimen of daclizumab. Program and abstracts of American Transplant Congress 2003, the fourth joint American Transplant meeting, May 30–June 4, 2003, Washington, DC. Abstract 662.

11Kaufman DB, Leventhal JR, Gallon LG, et al. Pancreas transplantation in the pred- nisone-free era. Program and abstracts of American Transplant Congress 2003, the fourth joint American Transplant meeting, May 30–June 4, 2003, Washington, DC. Abstract 665.

41. Transplantation of the Pancreas 733

the total number of patients reported was exceedingly small, this preliminary report gives hope that potentially greater and bigger things will follow. As of mid-2003, a total of 49 transplants (36 patients at nine international sites) had been performed. The immunosuppression protocol included daclizumab, sirolimus, and Prograf. Median follow-up at that time was 9.4 months. The eligibility criteria are quite strict and consist of the following: presence of type 1 DM with stable requirements, weight limit of 70 kg or less, Cr Cl greater than 80 mL/min, and no progressive diabetic complications. Of those who completed the transplant, 82.3% are insulin-free, with 52% of those who received any transplant being insulin-free. Most patients required two transplants, and some required three transplants. No deaths have been reported; however, there is a wide variation of success, with results as high as 90% at Edmonton to as low as 23% at some of the other sites.

Other centers have picked up the Edmonton protocol and, with their own variations, also have had some success. Long-term issues that need to be addressed involve the ability to achieve success with only one transplant as opposed to the two usually required. Also, the cost of starting up an islet cell transplant program is prohibitive for many centers, and therefore certain centers may become the regional center for all the prospective patients in that area. Finally, the inability to monitor for rejection delays any therapeutic measures until it is too late. The long-term question of potentially increasing patients’ antibody level to the extent that it makes them untransplantable should they need a K Tx in the future is another issue that can be addressed only with time.

Although solid organ P Tx still is considered to have the superior outcome for more people, the realization of successful islet cell transplant is closer than ever. The ability to avoid major surgery and yet undergo successful P Tx via percutaneous placement of these islet cells indeed would be a major breakthrough in the field of transplant. Although immunosuppression medications still would be required, any method that would decrease the known morbidity and mortality of an operative procedure could be considered only an advantage.

Case Discussion

The patient in the case presented exhibits the classic complications associated with someone who has DM with a severe infection. Most often, the infection is not noticed because of the lack of sensation in the lower extremities. These patients often become very ill very quickly and appear septic. Often, they are dehydrated as manifested by the elevated BUN and creatinine. In addition, as they lose their ability to control their sugars with routine subcutaneous doses of insulin, they can go into ketoacidosis, as manifested by the serum acetone. The treatment consists of an insulin drip and adequate intravenous hydration with empiric antibiotic coverage. The patient should be admitted and, when stable, have an imaging study to evaluate for possible osteomyelitis.

734 J.W. Lim

Summary

The realization of successful islet cell transplant forever has changed the face of P Tx. No longer are the prospects for islet cell transplant “just around the corner.” The success of islet cell transplant after so many years of futility has spawned new interest in hepatocyte transplant for liver transplant as well as in the potential of stem cells. Not lost in this great surge of cell research are the ever improving results for solid-organ P Tx. Clearly, results have improved from era to era. Technical failure rates have improved much over the years, but they still remain the number-one reason for graft loss. Improvements in immunosuppression have helped reduce the added complication rate in the care of the immunosuppressed patient. It is hoped that, as advances continue to be made in the field of islet cell and whole-organ P Tx, more potential patients are made aware that their options are no longer restricted to insulin alone. Percutaneous insulin is not the answer for many patients, and, as more news surrounding the improved results of P Tx become known, more healthcare providers will be better able to inform their prospective patients.

Selected Readings

Cecka JM, Terasaki PI, eds. Clinical Transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000.

DCCT Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulindependent diabetes mellitus. N Engl J Med 1993;329:986–997.

Harland RC. Pancreas transplantation. In: Norton JA, Bollinger RR, Chang AE, et al, eds. Surgery: Basic Science and Clinical Evidence. New York: SpringerVerlag, 2001.

Nymann T, Hathaway Dr, Shokouh-Amiri MH, et al. Kidney-pancreas transplantation: the effects of portal versus systematic venous drainage of the pancreas on the lipoprotein composition. Transplantation 1995;60(12):1406– 1412.

Porte D, Sherwin RS, Baron A, et al, eds. Ellenberg and Rifkin’s Diabetes Mellitus: Theory and Practice, 6th ed. New York: McGraw-Hill, 2002.

Troppmann C, Gruessner AC, Benedetti E, et al. Vascular graft thrombosis after pancreatic transplantation: univariate and multivariate operative and nonoperative risk factor analysis. J Am Coll Surg 1996;182(4):285–316.

Tyden G, Bolinder J, Solders G, Brattstrom C, Tibell A, Groth C-G. Improved survival in patients with insulin-dependent diabetes mellitus and end-stage diabetic nephropathy 10 years after combined pancreas and kidney transplantation. Transplantation 1999;67(5):645–648.