Decreased serum insulin-like growth factor-I in burn patients: Relationship with serum insulin-like growth factor binding protein-3 proteolysis and the influence of lipid composition in nutritional support

Objectives: To test the effects of the amount and type of fat in the nutrit ional support on serum insulin-like growth factor (IGF)-I concentrations in burn patients and to test the hypothesis that the serum proteolytic activi ty for insulin-like growth factor binding protein (IGFBP)-3 is a major mech anism for the decreased serum IGF-I observed in these patients. Design: Randomized, double-blind trial of three different nutritional suppo rts and analysis of serum IGF-I, IGFBP-3, and serum IGFBP-3 proteolysis. Setting: Burn center in a university hospital. Patients: A total of 23 severely burned (>25% total body surface area burne d) adult patients. Interventions: Patients were randomly assigned to three types of nutritiona l support differing in the amount of energy derived from fat and the presen ce or absence of fish oil: Group I (control), 35% fat; Group II, 15% fat; G roup III, 15% fat with 50% as fish oil. Nutritional support was both parent eral and enteral and was started within 24 hrs of admission. Measurements and Main Results: Serum IGF-I and IGFBP-3 were measured by rad ioimmunoassay every 3 days for 28 days in 23 severely burned adults. In six patients, IGFBP-3 was measured by ligand binding assay and the serum prote olytic activity for rhlGFBP-3 was measured as well. Serum IGF-I concentrati on was low in all subjects throughout the study period, but did increase wi th time (p < .01); significantly higher values were found in Group III (p < .05), Multivariate analysis showed that fish ail and low fat solutions wer e significantly correlated to serum IGF-I concentrations. Serum IGFBP-3 (ra dioimmunoassay) was higher than normal throughout the study with no differe nce between the groups. Between days 4 and 16, IGFBP-3 was cleaved into two fragments in all patients studied, and the molecular weights of the fragme nts were equal to those observed in the serum of a woman late in pregnancy. During this period of time, serum proteolytic activity for rhlGFBP-3 was > 30% in 24 of the 30 samples measured, whereas 20 of the 28 samples measured thereafter were normal (<25%), Serum IGFBP-3 concentration from ligand bin ding assay was correlated with serum proteolytic capacity in all subjects ( mean r(2) = 0.77; p < .01) and with serum IGF-I concentrations in five of s ix subjects (mean P = 0.81; p < .01). Conclusions: In burn injury, serum IGF-I concentrations are sensitive to th e amount and type of fat in their nutritional support. The presence of fish oil allowed for a more rapid recovery of serum IGF-I levels. The proteolys is of IGFBP-3 may be an important cause of the decreased serum IGF-I values and the protease(s) responsible for this seem to be similar to those obser ved in late pregnancy.