Protein turnover, lipolysis, and endogenous hormonal secretion in critically ill children

Objectives: The catabolic state is a major contributor to morbidity and mortality of critical illness and may be related to endocrine changes. We studied whether protein and lipid turnover correlate with insulin and growth and thyroid hormone plasma levels in critically ill infants. Design: Prospective clinical study. Setting: Pediatric intensive care unit. Patients: Twelve critically ill children and ten age-matched controls. Measurements: We measured lipolysis and protein turnover by infusing albumin-bound uniformly 13C palmitic acid and 2H3-leucine for 3 hrs and 2H5-glycerol for 5 hrs to critically ill infants. Simultaneously, we measured serum growth hormones, insulin, C-peptide, thyroid-stimulating hormone, T4, T3, albumin, retinol binding protein (RBP), and prealbumin. Hormone and serum protein levels were also measured in six children when recovered from critical illness. Ten healthy age-matched children served as controls for hormone serum levels comparison. Results: Palmitic acid and glycerol turnover were 5.6 ± 2.2 μmol/kg/min and 12.2 ± 7.3 μmol/kg/min, respectively, whereas α-ketoisocaproic turnover was 4.9 ± 2.8 μmol/kg/min. α-Ketoisocaproic turnover positively correlated (R = 0.7, p = .03) with duration of pediatric intensive care unit admission and with prealbumin and RBP serum levels (R = 0.9, p = .001). Insulin-like growth factor binding protein (IGFBP)-2 was significantly higher and IGFBP-3 was significantly lower in critically ill children (p = .03 and p = .04 vs. recovery phase, respectively). No other hormonal differences were found. Serum albumin was significantly lower in sick children. We found a significant correlation between prealbumin and RBP and IGFBP-3 (R = 0.6, p = 0.03 and R = 0.6, p = .04, respectively). α-Ketoisocaproic turnover positively correlated with IGFBP-1 (R = 0.79, p = .01) and did not correlate with insulin-like growth factor I (R = -0.5, p = .15 [not significant]) No other correlations were found. Lipid turnover measurements did not correlate with any endogenous hormone levels or with duration of critical illness. Conclusion: Protein turnover but not lipolysis correlated with a persisting critically ill condition, serum prealbumin, RBP, and plasma IGFBP-1.