Objective and rapid assessment of pancreas graft viability using 31P-nuclear magnetic resonance spectroscopy combined...

With the current shortage of donors, there is a critical need to optimally use "less-than-ideal" donors for pancreas transplantation. Objective and rapid means for assessing graft viability and suitability for transplantation are mandatory. This study examined the possibility of graft viability assessment and posttransplant outcome prediction using (31)P-nuclear magnetic resonance (NMR) spectroscopy combined with the two-layer cold-storage method (TLM).

Canine pancreas grafts were preserved with TLM for 24 hours after 0, 60, or 120 minutes of warm ischemia (groups 1, 2, or 3, respectively). After preservation, we determined intragraft phosphate metabolites noninvasively using (31)P-NMR spectroscopy. Time required for this assessment was 5 minutes. Because our previous studies demonstrated that all grafts in groups 1 and 2 were successfully transplanted (the viable group), whereas all in group 3 failed to survive (the nonviable group), the possibility of posttransplant outcome prediction was examined on the basis of the comparison between these two groups.

The ratios of inorganic phosphate/gamma-adenosine triphosphate (Pi/gamma-ATP) and Pi/beta-ATP reflected the extent of graft damage, and the differences were statistically significant among groups 1, 2, and 3. On the basis of analyses of receiver operator characteristic curves, the optimum cutoff levels between the viable and nonviable groups were 1.6 and 2.2 for Pi/gammaATP and Pi/betaATP, respectively. The accuracy rates of these ratios were both 83%.

(31)P-NMR spectroscopy combined with TLM preservation could provide an objective, rapid, and possibly noninvasive means to assess pancreas graft viability and to determine suitability of damaged pancreata for organ transplantation.