Blood-Biomarkers for Glucose Metabolism in Preterm Infants

Abstract

This was an exploratory, prospective, longitudinal, cohort study that aimed to establish "healthy" reference levels related to growth parameters and glucose metabolites in preterm infants. This was conducted to further investigate growth and metabolic disturbances potentially related to neonatal illness. The study sample consisted of 108 preterm infants born before 32 weeks in 2018-2019 in the Capital Region of Denmark. Repetitive blood samples were acquired at the neonatal wards, while clinical data were obtained from the regional hospital medical record system. Thirty-four "healthy" preterm infants (31%) were identified. The "ill" infants were divided into four subgroups dependent on gestational age and small for gestational age. Reference levels for the growth parameters and metabolic biomarkers glucose, albumin, and adiponectin, and two glucose control indicators, glycated albumin and fructosamine, were determined for the "healthy" and "ill" subgroups. The "ill" extremely preterm infants had increased glucose levels (mean difference 0.71 mmol/L, 95% CI 0.23; 1.18 mmol/L) and glycated albumin (corrected; %) (mean difference 0.92 mmol/L, 95% CI 0.38 mmol/L;1.47 mmol/L) compared to the "healthy" infants. In "ill" extremely preterm infants and "ill" very preterm infants born small for gestational age, levels of biomarkers containing proteins were decreased. In the "Ill" extremely preterm infants and infants born small for gestational age, postnatal growth was continuously decreased throughout the postconceptional period. The short-term glucose-control indicator, glycated albumin (corrected; %), reflected well the high glucose levels due to its correction for the depleted plasma-protein pool.

Keywords: blood biomarkers; glucose; healthy reference levels; metabolism; plasma-protein depletion; preterm infants.

Figure 1

Blood glucose biomarkers from “healthy” children born GA < 32 weeks, AGA with no clinical signs of neonatal morbidities across the postconceptional period. Estimated mean and normal ranges based on the generalized additive model are shown as solid lines for each infant and gray shade with minimum and maximum values indicated by dashed lines. The red line represents the estimated mean of the “healthy” reference values. ICC = intraclass correlation.

Figure 2

Blood biomarkers and related ratios in “ill” subgroups divided according to gestational age, born SGA vs. A-/LGA across the postconceptional period. For comparison, normal ranges from the “healthy” subgroup are indicated in grey shading. Diff is the estimated mean difference compared to the “healthy” subgroup with 95% CI and p-values (unadjusted/adjusted) based on the generalized additive model.

Figure 3

The estimated associations between levels of glycated albumin (%), glycated albumin, fructosamine (corr), and fructosamine with backwards-running averages of glucose measurements; the strongest association for corrected glycated albumin (%) was found on the 22nd postnatal day. Valid interpretation of associations for the other potential glucose control indicators were difficult due to a depleted total plasma-protein pool and insufficient correction parameters.

Figure 4

Growth curves for body weight, body length, and head circumference in subgroups of preterm infants across the postconceptional period. Infants born extremely premature, small for gestational age, and experiencing neonatal morbidities remain more growth-restricted throughout the neonatal period.