L-ARGININE AND INSULIN-TOLERANCE TESTS IN THE DIAGNOSIS OF ADULT GROWTH-HORMONE DEFICIENCY - INFLUENCE OF CONFOUNDING FACTORS

OBJECTIVE In the attempt to define a On stimulation test with high spe cificity and reproductibility, few studies have addressed the influenc e of potential interfering external factors on the test result. We the refore tested the influence of physical activity (admission to hospita l on test morning) and mild heat exposure on the GH response to L-argi nine stimulation test (Arg) and insulin-tolerance test (ITT). DESIGN O ne Arg stimulation test and one ITT were performed in all subjects dur ing standard conditions (overnight hospital stay, 10 hours fasting). I n addition, each subject was randomized to undergo either two addition al Arg tests, or two ITTs, performed under two different conditions: a dmission to hospital on the morning of the test and during standard co nditions except for heat exposure before testing. The four tests were performed in random order. PATIENTS Twenty-two patients (six women, 16 men) (mean age +/- SEM, 38.3 +/- 5.3 years and 36.1 +/- 2.7 years, re spectively) presenting with pituitary disease and a group of healthy a ge and gender-matched normal subjects (six women, 13 men) (age 38.3 +/ - 4.8 years and 35.7 +/- 2.4 years, respectively) participated. MEASUR EMENTS During the on-stimulation tests serum on, cortisol, blood gluco se, and plasma glucagon were measured and compared in the three differ ent test conditions. RESULTS During standard conditions, peak GH respo nse was higher in the ITT compared to the Arg test in the control grou p (23.4 +/- 3.6 mU/l vs 11.6 +/- 2.0 mU/l, P = 0.004), and the specifi city of the ITT was higher (18/19 versus 13/19, P = 0.047). Minor heat exposure before the ITT (temperature rise 0.24 +/- 0.05 degrees C, ra nge 0.0-0.5 degrees C) did not change the on response in the healthy a dults whereas admission to hospital on the morning of the test reduced the on response significantly (P < 0.05). The lowest blood glucose di d not change in the three situations and did not correlate with peak o n during the ITT. In the patients there were no significant difference s between the on response during different conditions. Plasma glucagon did not significantly differ between the different test conditions in the control group (P = 0.88), but there was a significant decrease in the glucagon response to the test performed after hospital admission on the test morning in the patients (P < 0.025). Serum cortisol respon se in the control group did not differ in the three situations.CONCLUS IONS Since provocative on responses are influenced by external factors , conditions should be standardized to optimize the reproductibility a nd specificity of the tests. Furthermore the higher specificity of the insulin-tolerance test as compared to the arginine stimulation test w as confirmed.