Mean ± SEM values for serum tumor necrosis factor α (TNF-α) after drug treatment. Asterisk indicates a significant difference between levels before treatment (day 0) and those 21 and 42 days after the burn (P<.05); dagger, a significant difference between the group receiving recombinant human growth hormone (rhGH) and the placebo group (P<.05).
A schematic showing the individual changes in serum tumor necrosis factor α (TNF-α) levels during the study in burned children receiving (A) recombinant human growth hormone (rhGH) (n=10) and (B) placebo (n=10). The mean ± SD reference serum TNF-α level is 1.4 ± 0.2 pg/mL (from Petersen et al1).
Chrysopoulo MT, Jeschke MG, Ramirez RJ, Barrow RE, Herndon DN. Growth Hormone Attenuates Tumor Necrosis Factor α in Burned Children. Arch Surg. 1999;134(3):283-286. doi:10.1001/archsurg.134.3.283
Recombinant human growth hormone (rhGH) has been shown to favorably modulate the acute-phase response and may improve the clinical outcome.
To examine whether rhGH attenuates the elevated tumor necrosis factor α (TNF-α) levels that correlate with increased multiorgan failure and mortality in burned adults and children.
Twenty children with burns of greater than 40% of the total body surface area were randomly divided into 2 groups to receive placebo (n=10) or rhGH, 0.2 mg/kg per day intramuscularly (n=10).
Pediatric burn hospital.
Main Outcome Measure
Serum TNF-α levels by enzyme-linked immunoassay at baseline (day 0) and at 21 and 42 days after injury. For statistical analysis, we used the Kruskal-Wallis and Friedman tests.
No significant differences in age (mean ± SD, 6.2 ± 1.6 vs 5.0 ± 1.2 years) or percentage of total body surface area burn (mean ± SD, 65.1% ± 8.2% vs 57.1% ± 5.2%) could be shown between the groups given rhGH and placebo. Baseline TNF-α levels were elevated from reference values in both groups. Twenty-one and 42 days after rhGH administration, serum TNF-α levels were significantly decreased from those at baseline (P<.05). No significant decrease in TNF-α levels was observed in the placebo group (P=.5).
Recombinant human growth hormone significantly lowers serum TNF-α levels after burn injury. This is consistent with the beneficial effect that rhGH has on the acute-phase response.
AN ACUTE hypercatabolic state is often observed following major illness or trauma, such as a severe burn. Increases in catecholamine, glucagon, and cortisol levels raise the energy requirements, and protein-degradation products (amino acids) are used for fuel. Most of these amino acids are supplied by muscle. The resultant muscle wasting and nitrogen imbalance are not improved by early aggressive nutritional supplementation. Extracellular to intracellular glucose-transport mechanisms are changed, as evidenced by the development of postinjury insulin resistance. Futile substrate cycling of glucose and triglycerides yields no net conversion of substrate to product for fuel. These changes are typical of the hypermetabolic response to a thermal injury.
One of the main components of this response is the alteration in serum concentrations of a large number of plasma proteins known as the acute-phase proteins.1 One of the principal mediators of these acute-phase proteins is tumor necrosis factor α (TNF-α). Indeed, altered body composition has been associated with elevated serum levels of TNF-α, and TNF-α has been shown to induce anorexia, weight loss, and an acute-phase response.2,3 Studies of animals have shown that TNF-α may be responsible for the tissue wasting associated with malignancy,4 and increased serum levels have been found in many patients with chronic diseases such as rheumatoid arthritis.5 Elevated levels also have been found in the serum of burned patients with sepsis.6 Yamada et al7 showed that serum levels of TNF-α were elevated following a thermal injury and correlated this with an increase in the incidence of sepsis and mortality and sepsis-induced muscle proteolysis.8
In addition to elevated levels of stress hormones, a reduction in anabolic hormones, including growth hormone (GH) and insulinlike growth factor type I (IGF-I) is observed after a thermal injury.
The purpose of this study was to determine whether treatment with recombinant human growth hormone (rhGH) affects TNF-α serum levels in burned children.
Twenty children with acute burn injury were entered into a randomized, double-blind study to receive either placebo (n=10) or rhGH, 0.2 mg/kg per day (n=10). Randomization was performed by computer random number selection. Inclusion criteria for these studies were age 1 to 18 years, admission to our burn hospital within 3 days of injury, burn size of more than 40% of total body surface area, with a third-degree burn of more than 10% requiring a minimum of 1 donor-site harvest. The rhGH was dissolved in isotonic sodium chloride solution and given intramuscularly each day for 6 weeks. Serum specimens were taken on admission before treatment (day 0) and on days 21 and 42 after the burn.
Serum TNF-α concentrations were determined using a TNF-α human enzyme-linked immunosorbent assay (Endogen, Inc, Woburn, Mass). A 3-step technique was used. Standard curves for quantifying human TNF-α were obtained. The standard curve observed was linear from 0 to 1000 pg/mL on a logarithmic scale. Absorbance was measured at 450 nm.
Patient records were reviewed for the development of organ failure and sepsis and for mortality. Acute renal failure was defined as oliguria for at least 36 hours (urine output <0.5 mL·kg−1·h−1), a serum urea nitrogen-to-creatinine ratio of less than 20, a serum creatinine level of greater than 177 µmol/L (>2 mg/dL), or any combination of these criteria. We determined whether burned children required ventilatory support and the duration of support. Cardiac function was examined, with cardiac failure defined as a cardiac arrest requiring pediatric advanced life support. Sepsis was defined as having a blood culture positive for pathogenic bacteria or fungus either during the clinical course or at autopsy.
Statistical analysis was performed using the Kruskal-Wallis test to determine differences in treatment and the Friedman test to determine differences in time. The paired t test with the Bonferroni correction was used post hoc to determine differences within groups. Data are expressed as the mean ± SD. Significance was accepted at P<.05.
No differences in age, sex, burn size, or depth of burn between the group receiving rhGH and the placebo group could be demonstrated (Table 1). Before treatment (day 0), TNF-α levels in both groups were significantly elevated compared with reference values (14.0 ± 13.4 pg/mL in the group receiving rhGH and 9.4 ± 5.9 pg/mL in the placebo group; P<.06). Twenty-one and 42 days after therapy, TNF-α levels were significantly decreased in patients receiving rhGH compared with baseline (4.2 ± 3.4 pg/mL and 2.3 ± 2.9 pg/mL, respectively; P<.05). Serum TNF-α levels did not change with time in the placebo group (Figure 1). Figure 2 shows the change in individual serum TNF-α levels with time in the group receiving rhGH and the placebo group. No significant differences in mortality or incidence of renal failure, cardiac arrest, or sepsis could be shown between patients receiving rhGH and those receiving placebo (P>.05). No significant difference could be shown in the number of patients requiring mechanical ventilatory support (P>.05). Patients receiving rhGH had a greater increase in body weight than those receiving placebo (4.6% ± 4.3% vs −1.1% ± 3.9%; P<.05), but no differences in energy intake or protein and fat intake between the groups could be shown during the 6 weeks.
The beneficial effects of rhGH on metabolism in patients with burn injury have been described in some detail.9- 13 Herndon et al14 showed that rhGH shortened donor site healing times and subsequently decreased the length of hospital stays. These shorter wound healing times were associated with elevated levels of serum IGF-I. Growth hormone has been suggested to act primarily in the liver, both directly and indirectly through IGF-I production.15- 17 Protein anabolism is 1 outcome of GH administration that has been studied in an extensive randomized, double-blind clinical trial11 in which the administration of GH was shown to increase protein turnover as a result of elevated protein synthesis and breakdown. Protein synthesis, however, was increased by a greater degree than protein breakdown in severely burned patients.18 The net result was a reduction in protein loss of as much as 50% and a reduction in weight loss. In the present study, burned patients receiving rhGH gained substantially more weight than those receiving placebo.
Tumor necrosis factor α has been previously shown to directly inhibit the basal level of protein synthesis and IGF-I– and serum-stimulated protein synthesis in myoblasts and myotubules.19 This action of TNF-α is similar to that observed in the skeletal muscle of animals with sepsis.20 Exposing myoblasts to TNF-α for 10 minutes inhibited the stimulation of protein synthesis by serum and IGF-I for as long as 48 hours after administration. This suggests that TNF-α acts rapidly and that even transient exposures to this cytokine may decrease protein synthesis long after its disappearance from the circulation.19 Increased serum levels of TNF-α due to inflammatory states can persist for up to several days.21 Our study indicates higher serum levels in children for as long as 6 weeks after a burn.
Fan et al21 analyzed changes in circulating and tissue concentrations of IGF-I and selected IGF-I–binding proteins in rats after the infusion of TNF-α at various rates. Their study showed that TNF-α is an important regulator of the GH-insulinlike growth factor axis in vivo. Substantially reduced plasma levels of GH and IGF-I were detected even in the lowest TNF-α infusion rates.21 In similar studies,22- 24 levels of IGF-I were decreased following burns, trauma, surgical therapy, and the administration of lipopolysaccharides. It has been postulated that this decrease in plasma IGF-I levels is partly caused by a reduction in the synthesis and release of IGF-I by the liver. This theory emerged because the liver is the major source of IGF-I entering the circulation,25 and TNF-α has been shown to decrease IGF-I concentrations in the liver, muscles, and pituitary gland.21 These TNF-α–induced reductions may be mediated by the concomitant decrease in plasma GH levels.25,26
Frost et al19 reported that the exposure of human myoblasts to TNF-α does not change IGF-I–receptor binding characteristics. This suggests that the cytokine acts on a component of the IGF-I signal transduction pathway that lies distal to the receptor and to the autophosphorylation processes. Tumor necrosis factor α has not been shown to impair IGF-I– or serum-stimulated thymidine uptake, which implies that TNF-α acts specifically on the part of the signaling transduction pathway involved in the stimulation of protein synthesis.19 By impairing this process, together with other components of the IGF-I signaling pathway, TNF-α may inhibit protein synthesis.
The decrease in serum TNF-α levels observed in burned children receiving rhGH may be one mechanism by which rhGH exerts its favorable modulation on protein metabolism and the acute-phase response.
This study was supported by grant 8040 from the Shriners Hospitals for Children, Galveston, Tex.
Presented as a poster at the 18th Annual Meeting of the Surgical Infection Society, New York, NY, May 1-2, 1998.
Corresponding author: Robert E. Barrow, PhD, Shriners Hospitals for Children, 815 Market St, Room 718, Galveston, TX 77550.