Pancreas transplantation: the histologic morphology of graft loss and clinical correlations

BACKGROUND

Graft losses due to leaks, bleeding, thrombosis, infections, and early pancreatitis are grouped together under the category of technical failure. Among these complications, massive vascular thrombosis continues to be the most important cause of early graft loss due to technical failure. Pathological evaluation of most allografts lost early in the posttransplantation period shows vascular thrombosis with associated proportional parenchymal necrosis. The morphological findings in allografts that are considered to be lost due to technical failure has not been systematically addressed. In particular, the role of acute rejection in early graft loss has not been well studied.

METHODS

Seventy-four consecutive pancreas graft pancreatectomies were studied histologically to evaluate for thrombosis (recent versus organized), type of vessel involved by thrombosis (arteries, veins, or both), acute rejection grade, chronic rejection grade, endotheliitis, transplant arteritis, coagulation necrosis, acute pancreatitis, presence of infectious organisms, transplant (obliterative) arteriopathy, neoplasia, relative proportions of alpha and beta islet cells, and immunoglobulin and complement deposition. The histological findings were correlated with donor and recipient data as well as clinical presentation.

RESULTS

In 23 out of 39 grafts lost in the first 4 weeks posttransplantation, the only pathological changes found were vascular thrombosis and bland ischemic parenchymal necrosis. In these cases, no underlying vascular pathology or any other specific histological change was identified. Most of these grafts (78%) were lost in less than 48 hr and all in the first 2 weeks posttransplantation. Massive vascular thrombosis occurring in an otherwise histologically normal pancreas was the most common cause of graft loss in the first 4 weeks posttransplantation (59%). In most of the remaining cases (33%), although the clinical presentation suggested technical failure, there was clear histological evidence that the massive thrombosis resulted from vascular injury due to immune damage (acute and hyperacute rejection). Increased incidence of early graft thrombosis was seen in grafts from older donors and longer cold ischemia times. After the first month posttransplantation, graft pancreatectomies revealed a wider variety of pathological processes that included severe acute rejection, combined acute and chronic rejection, chronic rejection, and infections. Acute and chronic vascular thrombosis in large and small vessels was commonly seen at all times posttransplantation; chronic, organized thrombosis was strongly associated with chronic rejection.

CONCLUSIONS

(a) Early acute thrombosis occurring in a histologically normal pancreas defines a true technical failure. This study showed that acute rejection leading to massive thrombosis, which clinically simulates technical failure, results in a significant proportion of early graft losses. (b) Systematic histological evaluation of failed grafts is absolutely necessary for the accurate classification of the cause of graft loss. (c) There is morphological evidence that chronically ongoing thrombosis is an important, common, contributing factor for late graft loss.

Review of immunosuppressive usage in pancreas transplantation

Throughout 1997, nearly 10,000 pancreas transplants have been performed worldwide, with 88% being simultaneous kidney transplants (SKPT). The current 1 yr patient survival rate exceeds 90% and pancreas graft survival (complete insulin independence) rate exceeds 80% for SKPT, 70% for sequential pancreas after kidney transplant (PAKT), and 65% for pancreas transplant alone (PTA). According to registry data, rejection accounts for 32% of graft failures in the first year after pancreas transplantation. However, improvements are expected to continue with the evolution of treatment protocols. Most pancreas transplant centers employ quadruple drug immunosuppression with anti-lymphocyte induction with either a monoclonal or polyclonal antibody agent. In recent years, there has been an overall decline in the use of antibody induction therapy from 90% during the period 1987-1993 to 83% of pancreas transplants performed during 1994-1997. Maintenance immunosuppression is triple therapy consisting of a calcineurin inhibitor (cyclosporine or tacrolimus), corticosteroids, and an anti-metabolite (AZA or MMF). Prior to 1995, nearly all pancreas transplant recipients were managed with Sandimmune. In the last 2 yr, tacrolimus-based therapy has been used in approximately 20% of cases and a new microemulsion formulation of cyclosporine (Neoral) has replaced Sandimmune in contemporary post-transplant immunosuppression. In addition, MMF is replacing AZA as part of the standard immunosuppressive regimen after pancreas transplantation. At present, a number of centers are conducting various trials with new drug combinations including either Neoral or tacrolimus in combination with steroids and MMF with or without antibody induction therapy. From 1994 to 1997, the 1 yr rates of immunologic graft loss have decreased to 2% after SKPT, 9% after PAKT, and 16% after PTA. The current array of new immunosuppressive agents are providing more effective control of rejection and permitting solitary pancreas transplantation to become an accepted treatment option in diabetic patients without advanced complications. The apparent potency of new drug combinations has also resulted in a resurgence of interest in steroid withdrawal. Immunosuppressive strategies will continue to evolve in order to achieve effective control of rejection while minimizing injury to the allograft and risk to the patient. In addition, new regimens must not only address the issue of specific drug toxicities but also long-term economic, metabolic, and quality of life outcomes. Pancreas transplantation will remain an important alternative in the treatment of diabetic patients until other strategies are developed that can provide equal glycemic control with less immunosuppression and overall morbidity.

Donor factors affecting outcome after pancreas transplantation

In this study, we demonstrated that Px grafts from donors older than 45 years are associated with an increased risk of developing poor glycemic control and premature loss of Px function. Previous studies corroborate our finding that age of the donor is the principal donor characteristic impacting postoperative Px survival. Whereas prior studies also implicated hyperamylasemia as a factor which contributes adversely to outcome, we were unable to demonstrate a significant influence of donor hyperamylasemia on long-term graft survival, although it did correlate with the degree of immediate postoperative pancreatitis and with the need for oral hypoglycemic agents. Similarly, elevated blood glucoses in the donor, which can be a result of many other factors unrelated to the quality of the graft, did not predict a poor outcome in the recipient. NHB donor pancreata did as well as HB pancreata with regards to all postoperative functional parameters. A marginally increased risk of developing major complications was associated with older donors. Despite the frequent use of non-ideal donors, including older and NHB donors, excellent overall Px graft survival can be achieved. Although the quality of the pancreas graft was not directly addressed in this study, we believe irrespective of hyperglycemia or hyperamylasemia, subjective assessment of organ quality by an experienced transplant surgeon is the most important determinant of suitability.

Pancreas and islet transplantation in man

The results of pancreas transplantation have greatly improved, the overall patient and graft 1-year survival rates now being 89 and 62%, respectively. A technically successful graft ensures a near-normal glucose metabolism in most cases, and improves the patient's quality of life. However, pancreas transplantation is not a life-saving procedure and because of the necessary permanent immunosuppression it is usually performed in patients in whom a kidney transplant is needed or has been previously established. In such patients the other diabetic chronic complications are often advanced and limit the potential benefit of pancreas transplantation, but it seems premature to extend the indications to early stage diabetes. Islet transplantation has many potential advantages, mainly the possibilities of immune alteration and immune protection of the transplant that could allow transplantation to be performed without immunosuppression and the use of xenogenic tissue. Major limiting factors are the high numbers of islets necessary to ensure insulin independence and the low yield of islet isolation from adult pancreas. Encouraging, albeit preliminary results have been recently reported in man.