J.B. Doppenberg, W.H. Kopp, H. Putter, A.E. Braat, M.A. Engelse, E.J.P de Koning
Chair(s): drs. Christina Krikke, chirurg, UMC Groningen
Thursday 10 march 2016
13:00 - 13:30h at Foyer
Parallel session: Postersessies XI - Opgesplitst in 3 tijdblokken en 3 categoriëen (klinisch, basaal, donatie)
In order to better identify suitable donors for islet transplantation programs, it is imperative to find a model of islet isolation yield prediction based on donor characteristics. Current prediction models are either not evidence based or pertain to whole organ transplantation. Our aim is to generate an islet donor risk score (iDRI) that predicts islet isolation yield.
Clinical islet isolation outcome in IEQ was analyzed in relation to donor characteristics at the Leiden University Medical Center from 2008-2014 (n=289). Factors with a p-value <0.20 and <25% missingness in univariate analysis were entered into a multivariate backward elimination model. The least significant factor predicting IEQ was eliminated each round until only values with p<0.05 remained.
37 Factors were entered into the multivariate analysis and five remained: BMI, DBD/DCD, sex, amylase and CIT, making up the iDRI. The iDRI accounts for 28% of the variability in IEQ (correlation iDRI with IEQ p<0.0001, r2=0.279). The mean iDRI of isolations used for single organ transplantation was significantly higher than the mean of all other isolations (632,022±44,821 vs 415,407±8,243, p<0.0001).
The iDRi could be a useful tool to predict islet isolation outcome. Validation of the iDRI will be performed in islet isolation datasets from other centers.