Kidney and Pancreas Transplantation

David K. Klassen

Vinaya Rao^*

John D. Pirsch^*

Division of Nephrology, Department of Medicine, University of Maryland School of Medicine, Baltimore, Maryland 21201; and ^*Division of Transplantation, Department of Surgery, University of Wisconsin Medical School, Madison, Wisconsin, 53706

HISTORICAL BACKGROUND

Allotransplantation of the pancreas was performed along with the kidney in humans for the first time on December 17, 1966 by Drs. Kelly and Lillehei in Minneapolis (1); since then more than 17,000 have been reported by the International Pancreas Transplant Registry as of October 10, 2001. The first case was a partial pancreatic graft with ligation of the pancreatic duct. The pancreatic allograft worked for 6 days, and then developed complications of pancreatitis and pancreas fistula necessitating removal of the graft. This was the first of a series of 14 patients done at the University of Minnesota that ended in 1973. In a report of ten of these cases, nine received pancreaticoduodenal allotransplants with a renal allograft from the same cadaver donor simultaneously (2). Only one graft functioned for more than 1 year. In the first era between 1966 and 1977, patient survival at 1 year was 42% and graft survival was 7%.

The next era of pancreas transplantation began in 1978 (1978-1989). This period saw the evolution of pancreatic graft duct management techniques including free intraperitoneal drainage (3), enteric drainage (4), duct occlusion with synthetic polymers (5) and bladder drainage (6,7). The concept of bladder-drainage was introduced by Gliedman with a ureter-pancreatic duct anastomosis (6) and popularized by Sollinger et al (7) where a pancreaticocystostomy was performed by anastomosing the pancreatic duct to the bladder mucosa. A modification to this technique was the inclusion of a duodenal button with pancreatic graft (8). Corry et al further modified the bladder technique with the duodenal segment technique (9). This technique became popular as the difficult separation of the duodenum from the pancreatic head was avoided, thus minimizing the potential for bleeding and injury to accessory ducts.

A retrospective study comparing enteric and bladder drainage (10) showed superior pancreatic graft survival rates with bladder drainage. The main advantage of this technique was the ability to monitor pancreatic graft rejection by urinary amylase levels and thus facilitating early diagnosis, effective treatment and hence improving pancreatic graft survival rates.

This era also saw the introduction of triple immunosuppression with cyclosporine, azathioprine and steroids. The series from the University of Minnesota showed an improvement in patient survival rates from the initial series (1966-1973) at 47%. In the era 1978-1989, it was 88%, 1984-1987 91%, and 1987-89 92% with overall graft survival of 70% (SPK, PAK, PTA). During the same period, a series from the University of Wisconsin using quadruple immunosuppression reported overall graft survival of 73.1% and overall patient survival of 95.6%.

During the period 1988-1995, the systemic venous delivery of insulin with bladder drainage of exocrine secretions was popularized with more than 90% of pancreas transplants being performed by this technique. This technique was well tolerated in a majority of patients. Metabolic and urologic complications, however, ensued from obligatory fluid and bicarbonate losses and the exposure of the usually acidic and enzyme-free lower genitourinary tract to pancreatic enzymes. These included UTI, hematuria, acute or chronic urinary retention due to cystitis, leaks, episodes of urethritis, delayed urinary fistulas and metabolic disorders. These complications resulted in enteric conversion rates from 10 to 25% in large series (11, 12, 13). As a result, there was a resurgence of the enteric drainage technique to correct intractable complications of the bladder drainage technique. Further innovations to the surgical technique included portal venous delivery of insulin (14). The rationale was to lower the hyperinsulinemia associated with systemic drainage, thus improving metabolism of lipids and glucose uptake by the liver.

PATIENT SELECTION

Diabetes mellitus can be treated with insulin but is cured only by a pancreas transplant. It restores a euglycemic state, improves quality of life by making the patient insulin-free and, to a variable extent, prevents progression of neuropathy and permits reversal of micro- and macrovascular complications. It is now considered to be not only life enhancing, but also life saving.

Pancreas transplantation can be performed simultaneously (SPK) with or subsequent to a kidney transplant (PAK). This is done in diabetic patients with imminent or established end-stage renal disease who plan to have or have had a kidney transplant. The patients must meet criteria for kidney transplantation, and not have excessive surgical risks for the dual transplant procedure. Pancreas graft survival may be better when transplantation is done simultaneously with a kidney transplant.

Pancreas transplantation can also be done in patients without indications for kidney transplantation (PTA). These patients should have a history of brittle diabetes, frequent acute and severe episodes of hypoglycemia or hyperglycemia, ketoacidosis requiring medical attention, and consistent failure of insulin-based management to prevent these acute complications. Thus, pancreas transplantation alone is restricted to patients whose diabetic complications are or will be worse then potential side effects of the required immunosuppression.

Absolute contraindications include significant noncorrectable or intractable coronary artery disease demonstrated by coronary angiography or recent myocardial infarction, active infection, history of malignancy treated in the past 3 years (excluding non-melanoma skin cancer) and positive HIV serology. Other contraindications include hepatitis Bsurface antigen positivity, active untreated peptic ulcer, substance abuse, major untreated psychiatric illness, inability to provide consent, and recent history of medical noncompliance. Systemic illness that limits life expectancy or compromise survival, irreversible hepatic or pulmonary dysfunction, or positive lymphocytotoxic crossmatch preclude consideration for transplantation (15,16).

Relative contraindications include age greater than 65 years, morbid obesity (BMI >35), active smoking, peripheral vascular disease and severe aortoiliac disease. Some centers do not consider patients with type 2 diabetes, but in the most recent International Pancreas Transplant Registry report in 2001, 6% of reported cases were classified as type 2 diabetics (17).

Older recipient age is a risk factor for all transplant procedures. Pancreas transplantation is no exception. Despite this concern, the overall trend in pancreas transplantation has been the transplantation of older recipients. The proportion of recipients >45years in 1987-1992 was 9%; in the 1999-2002 era, it was 27% (17).

For patients with a previous kidney transplant or recipients of a pancreas transplant alone, similar criteria apply. The major consideration for these recipients is the presence of renal insufficiency or proteinuria. Recipients with a creatinine clearance of less than 60 mls/minute may develop significant renal insufficiency when transplanted and treated with a calcineurin inhibitor.

OPTIONS FOR THERAPY

Simultaneous Pancreas-Kidney (SPK) and Pancreas After Kidney Transplant (PAK)

A patient with type 1 diabetes and kidney failure is faced with a number of options for kidney replacement therapy. There is consensus that dialysis is the poorest option, because of the higher mortality associated with this modality compared to kidney transplantation. For acceptable kidney transplant candidates, there are several options: kidney transplantation alone, living donor (LRD) or cadaver donor, SPK or PAK (sequential kidney transplantation followed by pancreas transplantation). LRD transplantation is associated with the best long-term patient and graft survival, but lacks the benefit of a pancreas transplant with improvement in quality of life from euglycemia. SPK transplantation has the next best long-term survival, but requires a period of time waiting for suitable organs. Kidney transplantation alone followed by a pancreas transplant is another option with acceptable patient and pancreas graft survival. Finally, patients who are not candidates for pancreas transplantation or do not have a suitable LRD may undergo cadaveric renal transplantation. Cadaveric renal transplantation alone has the worst long-term survival, but selection bias may be an important factor in outcome, because patients selected for pancreas transplantation have fewer co-morbidities. Several studies have compared the outcomes between type 1 diabetic patients receiving a cadaveric transplant alone vs. SPK (18, 19, 20, 21). Long-term
survival is better with SPK, but SPK recipients are younger and receive transplants from younger donors which may account for the difference in survival.

A recent analysis of the UNOS database comparing SPK transplantation to LRD or cadaver transplantation alone was published (20). In this analysis, LRD recipients and SPK recipients had similar 8-year crude survival rates of 72%. Cadaveric transplantation alone had an 8-year survival of 55%. SPK recipients had a higher initial mortality compared to LRD recipients through 18 months (hazards ratio 2.2; p <0.001). After 18 months, SPK recipients had a lower risk of death compared to LRD recipients. This study suggested that SPK transplantation is more advantageous than LRD transplantation alone for long-term survival despite the higher initial early mortality.

Less clear is whether there is a survival advantage comparing PAK transplantation vs. SPK vs. LRD transplantation alone. Many recipients with a suitable LRD may prefer to proceed with kidney transplantation than wait on dialysis for an SPK transplant. If the transplant is successful, then pancreas transplantation can be performed. Until recently, the pancreas survival after kidney transplantation was significantly poorer than with an SPK transplant from the same donor. Improvements in surgical techniques and immunosuppression have narrowed the pancreas survival advantage of SPK vs. PAK. A recent study examined the optimal strategy for type 1 diabetic recipients with ESRD using a decision model (22). Outcome measures were life expectancy and quality adjusted life expectancy (QALY). In this analysis, LRD recipients alone had the best long-term life expectancy and QALY. PAK and SPK had the next best outcome. Worst was cadaveric transplantation and dialysis. When LRD transplants were removed from the analysis, SPK had the greatest life expectancy compared to PAK, cadaver transplant and dialysis.

For a type 1 diabetic with ESRD, pre-emptive transplantation with an LRD, CAD or SPK is optimal. Avoidance of dialysis and the attendant complications should be the goal. For patients with a potential living kidney donor, consideration must be given to proceeding with the transplant rather than wait for an SPK transplant. A PAK can then follow after a period of stability. Recipients without an LRD and who are acceptable candidates should be listed for SPK, cadaver transplantation or both. Cadaveric kidney transplant recipients alone can then be considered for PAK.

The relative benefits of PAK or PTA transplantation must be weighed against the potential for increased morbidity and mortality of the procedure. A recent analysis of UNOS data indicated an increased mortality of patients undergoing these procedures compared to patients on the waiting list treated with conventional therapy alone (23). The relative risk for death over a 4-year follow-up for PTA and PAK was 1.57 and 1.42, (p = .06 and p = .03 respectively). Patients undergoing an SPK transplant had a 60% reduction in mortality compared to patients waiting on the list.

Pancreas Transplant Alone (PTA)

About 250 pancreas transplant alone (PTA) procedures per year were performed in the United States since 1999. This procedure is considered in type 1 diabetics with severe secondary complications of diabetes or life-threatening hypoglycemic unawareness. In general, significant nephropathy (creatinine clearance <55 ml/min or proteinuria) precludes consideration for a solitary pancreas transplant. The major reason for exclusion is the requirement for calcineurin inhibitor therapy that will significantly lower the creatinine clearance post-procedure. Patients with significant nephropathy are best advised to wait for an SPK procedure or LRD. A recent study, noted above (23), found a trend toward an increased risk of death extending out to four years after PTA compared to patients who remain on the waiting list with conventional therapy. Analysis of longer-term data may provide further insights into the risks and benefits of the PTA procedure.

SURGICAL TECHNIQUE

A major surgical challenge for pancreas transplantation has been the management of the exocrine drainage of the pancreas. Early attempts at enteric drainage of the pancreas was a technical challenge and not successful. Pancreatic duct embolization with polymers was also attempted, but was largely abandoned with the introduction of the bladder drainage technique. Bladder drainage of the pancreas transplant offered several advantages. The first was the ability to monitor amylase in the urine. A falling serum amylase was often associated with acute rejection of the pancreas. From a surgical standpoint, the bladder is a sterile environment and early leaks from the duodenal-vesical anastomosis could be easily managed surgically or with a Foley catheter. In contrast, leaks from enterically-drained pancreas transplants were associated with polymicrobial infection from the bowel. In the early era, pancreas transplant recipients had multiple rejections and thus required high doses of corticosteroids and antilymphocyte preparations. Leaks were not uncommon in the setting of higher doses of immunosuppression because of poor wound healing and rejection of both the pancreas transplant and duodenal segment.

The transition to enteric drainage of the pancreas transplant began in the early 1990’s with improvement in immunosuppression. The introduction of mycophenolate mofetil and tacrolimus reduced the overall incidence of rejection. In addition, long-term bladder drainage of the pancreas was associated with significant urological complications including hematuria, recurrent urinary tract infections, metabolic acidosis from bicarbonate losses and late leaks from the anastomosis. In addition, urethral disruption occurred in men with bladder-drained pancreas transplants. Most of the significant complications of bladder drainage required enteric conversion of the pancreas transplant from bladder drainage. This is accomplished by removal of the
pancreas from the bladder and completing an anastomosis with the small bowel (12,24). With increasing experience with the enteric conversion technique, transplant centers began to move away from bladder drainage to primary enteric drainage of the pancreas. Currently, in the United States, nearly 80% of SPK transplants are performed with primary enteric drainage (17). Nearly 60% of PAK and PTA transplants are drained into the bowel. Many centers still prefer the bladder technique for these transplants because of the ability to continue to monitor urine amylase.

An ongoing debate is the venous drainage of the pancreas. Because of the proximity to the bladder, most pancreas transplants were drained into the systemic venous circulation via the iliac vein. The venous drainage of the native pancreas is into the portal vein. Drainage into the systemic circulation results in hyperinsulinemia and this could have significant long-term metabolic consequences such as accelerated atherosclerosis, hypertension and hyperlipidemia. Since 1996, the number of SPK and PAK procedures per year utilizing portal drainage has been about 25%. More PTA procedures had portal drainage with about 45% in the latest analysis. Despite some transplant centers’ enthusiasm for portal drainage (25), there has been no significant improvement in pancreas graft survival (26).

The procurement and back table preparation of the pancreas is beyond the scope of this review. See reference (27) for a detailed description of the donor operation. Pancreas transplantation can be performed through a midline incision or bilateral lower quadrant extraperitoneal incision (28). In general, the pancreas transplant is performed on the right side of the abdomen due to a more favorable venous anatomy of the right iliac system. Because the liver is often procured with the pancreas, the celiac axis remains with the liver. This requires surgical reconstruction of the arterial anatomy of the pancreas. This is accomplished with a donor iliac “Y” graft that is used to reconstruct the superior mesenteric artery and the splenic artery.

The arterial revascularization is usually performed by anastomosis of the “Y” graft to the right common iliac or external iliac artery. Venous drainage of the pancreas is then accomplished using the iliac vein for systemic drainage and the superior mesenteric vein for portal drainage.

During back table preparation, the duodenal segment is prepared by suturing the proximal and distal ends of the duodenum. The duodenum is then opened along its antimesenteric border and sutured to the bladder or small bowel. In some cases, a Roux-en-y limb is utilized for the enteric pancreas drainage to isolate the anastomosis from the bowel contents and minimize the risks of an enteric leak.

SURGICAL COMPLICATIONS OF PANCREAS TRANSPLANTATION

Pancreas graft thrombosis or leak from the duodenal anastomosis are the most significant complications immediately following pancreas transplantation (29). Although the incidence of early vascular thrombosis is decreasing, registry data shows that the current rate of graft thrombosis ranges from 6 to 9% (17). Graft thrombosis is more common in solitary pancreas transplantation. A number of factors contribute to this problem, including the low microvascular blood flow in the pancreas allograft, and the complex arterial reconstruction required for pancreas transplantation. Pancreatitis related to reperfusion injury may increase resistance to blood flow in the pancreas allograft resulting in secondary thrombosis. Pancreas graft thrombosis is usually an early event, with a peak incidence in the first 2 weeks posttransplant. This typically presents as a sudden onset of hyperglycemia. There may be associated abdominal pain and allograft tenderness. Duplex ultrasonography will rapidly confirm this diagnosis. It is unusual that thrombectomy of the allograft is successful; most commonly, transplant pancreatectomy is required. Pancreas graft thrombosis is somewhat more common in enterically-drained transplants, as well as in transplants utilizing portal venous drainage. The rate of thrombosis is, however, decreasing as more centers gain increased experience with these surgical approaches.

Intra-abdominal hemorrhage is a relatively common complication following transplantation. This has been reported to occur in up to 15% of cases (30). The use of anticoagulation as a strategy to avoid early pancreas allograft thrombosis is associated with an increased incidence of intra-abdominal bleeding. Minor bleeding is treated conservatively with correction of anti-coagulation and transfusion. However, significant intra-abdominal hemorrhage requires a re-laparotomy. Occasionally, the distal portion of the ligated splenic vein will thrombose; however, this is generally not of great clinical significance.

Intra-abdominal Infections

Intra-abdominal infections after pancreas transplantation are one of the most serious complications. This complication is associated with high morbidity and also results in significant decreases in patient and allograft survival. The incidence of intra-abdominal infections following pancreas transplantation has been decreasing over the last 10 years. Registry data prior to the early 1990s showed an approximate 20% rate of intra-abdominal infection (31). More recent data has shown that intra-abdominal infections have decreased substantially (32). This has been attributed to improved surgical technique, reduced rates of rejection with newer immunosuppressive medications, improvements in antibiotic prophylaxis, and improved donor and recipient selection.

Only gold members can continue reading. Log In or Register to continue