SERUM INSULIN-LIKE GROWTH-FACTOR BINDING PROTEIN-3 RESPONDS DIFFERENTLY TO TRAUMA IN MEN AND WOMEN

Objectives: Insulin-like growth factor-1 (IGF-1) has been studied as a marker of nutrition in critical illness, but there is little research on IGF-binding protein-3, which regulates the bioactivity of IGF-1. T he objectives of the present study were to measure serum IGF-binding p rotein-3 concentrations in trauma patients and to determine whether fa ctors such as age, gender, and severity of injury should be considered when evaluating serum IGF-binding protein-3 concentrations as a marke r of nutritional or clinical status. Design: Prospective, randomized, descriptive study. Setting: Emergency room of a university hospital. P atients: One hundred eight trauma patients. Interventions: None. Measu rements and Main Results: In this diverse group of patients, Injury Se verity Score ranged from 1 to 45 (11.5 +/- 10.3 [SD]), age ranged from 18 to 77 yrs (35 +/- 15.3), and 68% were male. A venous blood sample was collected at the time of admission into the study and was analyzed for serum IGF-binding protein-3 concentration (by radioimmunoassay), serum osmolality, IGF-1 concentration, and C-reactive protein concentr ation. Relationships between variables were tested using Pearson's cor relation coefficients and multiple regression analysis. Age, Injury Se verity Score, serum osmolality, time since injury, and gender were not significant predictors of serum IGF binding protein-3 concentrations when all patients were analyzed together. However, when men and women were analyzed separately, notable gender differences were observed. In women, serum IGF-binding protein-3 concentrations were increased with increasing severity of injury (beta = 0.52, R(2) = .33, p < .01). In men, the opposite relationship was observed (beta = -0.29, R(2) = .17, P < .01). Other predictors in the model (age, serum osmolality, and t ime since injury) were not significant. Variability in IGF binding pro tein-3 concentration could not be explained by differences in body mas s index or acute-phase response (serum C-reactive protein). Serum IGF- 1 concentrations changed coordinately with IGF-binding protein-3 conce ntrations in females and males (r = .62, p < .001 and r = .54, p < .00 1, respectively). IGF-binding protein 3 concentration at the time of a dmission into the study could not predict mortality, but this value wa s correlated with length of hospitalization in women (r = .37, p < .05 ). Conclusions: Determination of the specificity and sensitivity of IG F-binding protein-3 as an index of nutrition or anabolism requires kno wledge of its relationship to nonnutritional factors. These factors ar e most discernible before the confounding effects of treatments, absen ce of feeding, and complications. The present study demonstrated that gender and severity of injury must be considered when interpreting ser um IGF-binding protein-3 concentrations in trauma patients. In a much wider context, the present findings suggest that the study of the meta bolic response to stress requires separate analyses, based on gender.