Background
Islet transplantation is a viable treatment alternative for a select group of patients with type 1 diabetes. However, variables unique to the donor pancreas, such as age, fibrosis and edema, can influence the number and purity of the isolated islets. Thus isolation of a sufficient number of islets for transplantation from the pancreas remains challenging because of the lack of methods enabling reproducible isolation.
Methods
Islets were isolated from 38 consecutive deceased donors using the semi-automated Ricordi method of islet isolation, and purified on a COBE 2991 cell processor using Ficoll-based continuous density gradients. Three different gradient protocols were used. These included a pre-defined gradient using different densities of Ficoll (1.100 g/mL and 1.077 g/mL) mixed with HBSS (group 1), a pre-defined gradient using single-density Ficoll (1.100 g/mL) mixed with University of Wisconsin solution (UW) (group 2) and a variable gradient using single-density Ficoll (1.100 g/mL) with UW and densities selected based on the results of test gradients (group 3).
Results
Group 3 yielded a better recovery of islets (74%) than groups 1 (43%) or 2 (37%) (P=0.0144). Viability was significantly higher in groups 2 and 3 (P=0.0115). Purity was not significantly different among the groups.
Discussion
This method, using a simple test gradient, is a significant process improvement that can improve islet recovery without loss of viability or purity and increase the number of islet products suitable for transplantation.
