Background: The false positive rate for the newborn screening of disorders of amino acid metabolism for premature infants is higher than full term infants. This may be due to very low birth weight infants receiving high concentrations of amino acids from total parenteral nutrition (TPN) administration and/or immature metabolism. An investigation of the possible influence of TPN on screening of premature infants resulted in the detection of three unusual peaks in the tandem mass spectrometry (MS/MS) acylcarnitine profile. These markers were closely correlated with the detection of very high multiple amino acid increases in the profiles of newborns administered with TPN and who were ultimately found to be normal and free of inherited metabolic disorders.
Methods: TPN solutions contain a concentrated mixture of amino acids and dextrose and other nutrients in saline. Due to its high concentration and suggestion of a carbohydrate, it was hypothesized that dextrose (D-glucose) was the contaminant and source of the markers detected. Dextrose, stable isotope-labeled 13C6-dextrose and various TPN solutions were analyzed directly or after enrichment in whole blood by multiple MS/MS acquisition modes including MS-only, product and precursor ion and neutral loss scans.
Results: Analysis of dried-blood spots (DBS) prepared from whole blood spiked with TPN solutions containing 12.5% dextrose and amino acid formulations designed to deliver 2.5 gm/kg/day of an amino acid mixture had moderate increases of all 3 dextrose markers detected at m/z 325, 399 and 473 as compared to controls. MS-only scans, product and precursor ion scans of dextrose and 13C6-dextrose in positive ion mode confirmed that these 3 peaks are derived from dextrose. Mass spectral analysis of labeled and unlabeled dextrose suggested that these peaks were dimers derived from dextrose.
Conclusion: The identification of dextrose markers in DBS indicates that high concentrations of dextrose were present in blood and the likely source was contamination by TPN solutions most likely occurring during a sample collection process.
2010 Elsevier B.V. All rights reserved.