EXPERIMENTAL AND CLINICAL EXPERIENCE WITH URINE AMYLASE MONITORING FOR EARLY DIAGNOSIS OF REJECTION IN PANCREAS TRANSPLANTATION

Abstract

Pancreas allograft rejection in dogs with pancreaticocystostomy can be predicted in advance of hyperglycemia by monitoring the urinary amylase (UA) concentration (U/L): In initial experiments, UA values declined to less than 1000 1.3 +/- 0.2 days before hyperglycemia in nonimmunosuppressed dogs, 3.3 +/- 1.0 days in dogs treated with cyclosporine (CsA), and 9.3 +/- 0.7 days in dogs treated with CsA, azathioprine (Aza), and prednisone (triple therapy). Autotransplanted control dogs maintained high urine amylase concentrations indefinitely (mean 125,544 +/- 36,931). In a subsequent experiment, in 19 dogs with bladder-drained pancreas allografts on CsA only for prophylactic immunosuppression, a five-day course of antirejection treatment with Aza (5.0 mg/kg) and antilymphocyte globulin ALG (1 mg/kg) was started in group A (n = 10) when a raise in serum glucose was detected, and in group B (n = 9) when a drop of UA below 1000 was observed. The functional allograft survival rate was 9.2 +/- 0.5 days in group A (treatment started after hyperglycemia) and 29.0 +/- 5.7 days in group B (treatment started after drop in UA) (P = .002). The UA dropped in all dogs before hyperglycemia, at a mean of 2.7 days in group A and 20.8 days in group B. Clinically, 8 patients received a whole cadaver pancreas transplant with urinary drainage of the exocrine secretions. All were followed with UA monitoring. Three recipients lost the grafts for technical reasons. Three recipients lost the grafts for technical reasons. One had a primary non-function and UA was below 1000 U/24 hr; two developed abscesses and the grafts were removed while functioning with high UA values. Five grafts are currently functioning; 3 recipients had no rejection episodes and their UA values ranged from 30,000 to 100,000 U/24 hr during their entire postoperative course. The other two had rejection episodes. In both cases UA decreased to baseline levels 1 and 4 days in advance of the hyperglycemia. After antirejection treatment UA rose again to high values and plasma glucose levels declined. Both patients are currently insulin-independent, with UA values ranging from 10,000 to 200,000 U/24 hr. Both experimentally and clinically UA is an early predictor of pancreas allograft rejection. The institution of early treatment of rejection episodes in dogs, based on UA, significantly improved allograft survival. Urine amylase monitoring in pancreas transplant recipients could lead to an early treatment of rejection and improve graft survival.