Trying to Prevent the Clogged Drain: Optimizing the Yield and Function of Portal Vein-Infused Islets

Abstract

Total pancreatectomy with islet auto-transplantation (IAT) is a major surgical procedure during which the pancreas gland is resected, and pancreatic islets are isolated then infused into the patient’s portal vein. To avoid complications it is desirable to minimize the total volume of tissue transplanted intraportally [1]. Embolization via intra portal (IP) infusion of islets in auto and allo-transplantation can be associated with portal hypertension (PH), hepatic infarction, elevation of liver enzyme levels (transaminitis), disseminated intravascular coagulation, and portal venous thrombosis (PVT) [2–5]. Two recent articles in Digestive Diseases and Sciences, one in this issue, seek to address issues associated with isolation and intraportal infusion of islets. Khan et al. [6] report the use of preoperative MRI with contrast to predict of extent of pancreatic fibrosis and subsequent islet yield. This is a very useful report, but would benefit from discussion of whether any modification of the islet digestion technique would improve yield. We need a reproducible method for estimation of total potential islet mass/number within the pancreas that correlates with the difficulty of islet isolation. The article by Khan is a step toward a pre-isolation estimate of islet availability by use of MRI, but stating that the more diseased gland has fewer recoverable islets still leaves many questions relating to process optimization. However, as noted below, optimization of the islet isolation technique has yet to be defined for the fibrotic pancreas. Isolating islets from the pancreas of a patient with chronic pancreatitis (CP) poses multiple challenges. During islet isolation, collagenase digestion separates islets from exocrine cells; total tissue volume collected varies depending on donor characteristics, pancreatic fibrosis, the pathological status of the exocrine cells, and the efficiency of islet isolation [7]. The typical autologous islet transplant product often contains as much exocrine as endocrine tissue. The technical difficulty of islet isolation is daunting. Enzyme source and manufacturer, time, pressure, and temperature, in addition to the host factors listed above, affect islet yield from the diseased pancreas. Aggressive purification steps can significantly reduce islet yields. Transplantation of unpurified islet preparations is, therefore, still preferred during IAT. Portal vein flow complications/thrombosis or significant liver dysfunction are virtually unheard of for small-volume islet preparations (\10 cc volume). When pellet volumes exceed 20 cc, however, our practice is to perform further islet purification to reduce the risk of complications. Even with improvements in large-scale purification techniques using the COBE 2991 cell processor, purification unavoidably reduces yield [8]. The article by Desai et al. [9] in this issue reports the relationship between liver histology and the development of transaminitis after autoislet transplantation. Among their 26 autologous islet transplant cases, three had PVT. It was concluded that preexisting liver pathology was a factor contributing to the increase in liver enzymes but did not correlate with development of PVT. Generally, hepatocyte injury is multifactorial; likely causes include hypoxia, thrombosis, inflammation from infused cellular debris, a possible direct toxic effect of the isolation process, and host factors. Desai study [9] supports the notion that liver susceptibility to islet infusion and alteration of portal blood flow is variable in patients undergoing pancreatectomy and autoislet transplantation. The increase in liver enzymes in this series was relatively modest, returning to normal in all A. N. Balamurugan T. L. Pruett (&) Department of Surgery, Schulze Diabetes Institute, University of Minnesota, Minneapolis, MN, USA e-mail: tlpruett@umn.edu