LAUSR

Abstract Disclosure: R.D. Batson: None. A.M. Bruton: None. E.S. Pistone: None. A.J. Forrester: None. T.J. Wright: None. R.B. Pyles: None. K.M. Randolph: None. K.A. McGovern: None. C.P. Danesi: None. C.R. Gilkison: None. T.L. Wexler: None. B.E. Masel: None. K.C. Yuen: None. R.J. Urban: None. Introduction: Prior studies have reported that body mass index (BMI) negatively correlated with peak growth hormone (GH) response on the glucagon stimulation test (GST). Very little is known regarding the influence specifically of body composition and weight on peak GH response to glucagon stimulation. We sought to clarify the influence of BMI, weight and body composition on peak GH response to GST in a predominantly mild TBI adult population. Methods: A retrospective analysis of fixed-dose glucagon stimulation test (GST) results from 288 subjects at two centers was conducted. Fixed-dose GST protocols utilized 1 mg glucagon for individuals ≤90 kg and 1.5 mg for those >90 kg. Center A and Center B performed 240-minute and 180-minute GSTs, respectively. Multiple linear regression was used to evaluate the association between the dependent variable of peak GH and the following: BMI, weight, lean mass, fat mass, and fat percent in 5 separate models. Age and sex were added to all models and glucagon dose (1/1.5 mg) was assessed for inclusion. BMI was assessed as continuous, categorical, and binary, plus the interaction terms between them. F-tests and the coefficient of determination (adjusted R2) were used for model selection. Results: The best fit to the data was a model with age, sex, and the interaction between continuous and binary BMI (< or ≥30 kg/m2; N=288; adjusted R2=0.14; p<0.001). The range of BMI for non-obese (<30 kg/m2) participants was 15.8-29.9, and for obese (≥30 kg/m2) participants was 30.0-55.6. Every one-unit increase in BMI in the non-obese group was associated with a 0.54 ng/mL decrease in peak GH (95% CI: -0.82, -0.26; p<0.001). The association between BMI and peak GH in the obese group was non-significant. A restricted cubic spline regression model suggested the association between BMI and peak GH was nonlinear, especially for participants with BMIs between 24-30 kg/m2. Logistic regression revealed that overweight and obese individuals had increased odds of failing the GST compared to normal weight individuals. In a model adjusted for age and sex, every one-kilogram increase in weight was associated with 0.09 ng/mL decrease in peak GH (95% CI: -0.13, -0.05; p<0.001) but the weight model (N=288; adjusted R2=0.11; p<0.001) predicted peak GH less accurately than BMI. In participants with lean mass, fat mass, and fat percent data (N=50) none of the three were significantly associated with peak GH. Glucagon dose was not included in any of the models. Conclusions: In subjects with TBI who underwent GSTs, body weight was significantly associated with peak GH, but fat mass, lean mass, and fat percent were not. Additionally, BMI was negatively associated with peak GH, but only in subjects with BMIs <30 kg/m2. These results, although limited to a TBI population, differ from prior studies demonstrating an inverse relationship between BMI and peak GH in those with BMIs ≥ 30 kg/m2. Presentation: Saturday, July 12, 2025