Altered Serum Amino Acid and Acylcarnitine Profiles in Hyperinsulinemic Hypoglycemia and Ketotic Hypoglycemia

Abstract

Background: In addition to inborn metabolic disorders, altered metabolic profiles were reported to be associated with the risk and prognosis of some non-metabolic diseases, while as a rare metabolic disease, the overall secondary metabolic spectrum in congenital hyperinsulinemic hypoglycemia (HH) is largely undetermined. Therefore, we investigated metabolic profiles in HH patients and used ketotic hypoglycemia (KH) patients as a control cohort to unveil their distinct metabolic features.

Methods: A total of 97 hypoglycemia children, including 74 with hyperinsulinemic hypoglycemia and 23 with ketotic hypoglycemia, and 170 euglycemia control subjects were studied retrospectively. Clinical and biochemical data were collected. The normoglycemic spectra of amino acids and acylcarnitines were determined by liquid chromatography tandem mass spectrometry. The serum insulin and fatty acid concentrations during standardized fasting tests in hypoglycemia patients were also collected. Receiver operating characteristic curve analysis was performed to screen potential biomarkers.

Results: Among the normoglycemic spectra of amino acids, blood valine (p < 0.001), arginine (p < 0.001), threonine (p = 0.001), glutamate (p = 0.002), methionine (p = 0.005), ornithine (p = 0.008), leucine (p = 0.014), alanine (p = 0.017), proline (p = 0.031), citrulline (p = 0.042), aspartate (p = 0.046), and glycine (p = 0.048) levels differed significantly among the three groups. Significantly decreased levels of long- (C14:1, p < 0.001; C18, p < 0.001), medium- (C8, p < 0.001; C10, p < 0.001; C10:1, p < 0.001), and short-chain (C4-OH, p < 0.001; C5OH, p < 0.001) acylcarnitines were found in the hyperinsulinemic hypoglycemia group. Hyperinsulinemic hypoglycemia children with focal lesions and diffuse lesions had similar amino acid and acylcarnitine spectra. C10:1 < 0.09 μmol/L, threonine > 35 μmol/L, and threonine/C10:1 > 440 showed sensitivities of 81.1, 66.2, and 81.1% and specificities of 72.7, 78.3, and 81.8%, respectively, in distinguishing HH from KH.

Conclusions: We found significantly different altered serum amino acid and acylcarnitine profiles at normoglycemia, especially decreased C10:1 and increased threonine levels, between HH and KH children, which may reflect the insulin ketogenesis inhibition effect in HH patients; however, the detailed mechanisms and physiological roles remain to be studied in the future.

Keywords: acylcarnitine; amino acid; hyperinsulinemic hypoglycemia; hypoglycemia; ketotic hypoglycemia.

Figure 1

Differences in amino acid and acylcarnitine concentrations at normoglycemia among the ketotic hypoglycemia, hyperinsulinemic hypoglycemia and control groups. The box plots show the median (horizontal line), interquartile range (box), 5th percentiles (whiskers), 95th percentiles (whiskers), and outliers outside the 5th and 95th percentiles (dots). *p < 0.05 when the two groups were compared. **p < 0.01 when the two groups were compared. ***p < 0.001 when the two groups were compared. NS, Not significant; HH, hyperinsulinemic hypoglycemia; KH, ketotic hypoglycemia. (A) Blood glucsoce, (B) Valine, (C) Arginine, (D) Methionine, (E) Threonine, (F) Hydroxybutyrylcarnitine, (G) 3-Hydroxyisovalerylcarnitine/2-methyl-3-hydroxybutyrylcarnitine, (H) Octanoylcarnitine, (I) Decanoylcarnitine, (J) Decenoylcarnitine, (K) Tetradecenoylcarnitine, (L) Octadecanoylcarnitine.

Figure 2

ROC curves of amino acids and acylcarnitines (A) and their combinations (B).