Estimation of urea production rate with [15N2]urea and [13C]urea to measure catabolic rates in diabetes mellitus.

For verifying catabolic states in insulin-dependent patients and dogs the method estimating urea production rates with 13C and with doubly 15N labeled urea, respectively, has been established. For a fast steady state of urea tracer dilution, a prime of 600 times the continuous infusion rate had to be injected. Urea was isolated from plasma samples by protein precipitation and cation exchange chromatography with a consecutive derivatization of the dried urea fraction (trimethylsilyl derivatives). The masses of the fragment ions m/z 189 (14N14N), 190 (14N15N) and 191 (15N15N) urea are monitored to estimate the [15N2] urea frequency in the overall body urea pool in mol percent excess (MPE). 1 to 15 ng of derivatized urea were measured efficiently. An excellent correlation between expected standard and measured MPE (r = 0.9977) was achieved from solutions containing 1 to 7% [15N2]urea. The interassay coefficient of variation amounted to < 10% for a [15N2]urea portion of > or = 3%. Normoglycemic diabetic patients who were treated with insulin overnight showed significantly higher urea production compared to healthy controls (9.22 +/- 2.07 vs. 5.4 +/- 0.32 mumol.kg-1.min-1; p < 0.05). Measurements in chronic diabetic dogs proved an increased rate of amino acid catabolism (+20% urea production) in systemic versus portal application of insulin in paired studies. This increased nitrogen load in diabetics may accelerate progression of diabetic nephropathy. Thus, the established stable isotope technique may serve as a sensitive and useful indicator of amino acid catabolism in clinical and experimental research.