Use of the oral glucose tolerance test to define remission in acromegaly.

An oral glucose tolerance test (OGTT) was used to assess growth hormone (GH) secretion in patients with acromegaly prior to (n = 26) and after (n = 71) transsphenoidal adenomectomy as well as in 196 controls. In controls, suppressed concentrations of GH showed a negative relationship both with body mass index (BMI) and with age. Having calculated the reference intervals for suppressed GH concentrations to be expected for any given age and BMI, we compared these individually predicted ranges to GH concentrations actually observed in patients with acromegaly during OGTT. Preoperatively, concentrations exceeded the normal range in all patients. Postoperatively, glucose-suppressed concentrations of GH were less than 2.0 ng/mL in 56 (79%) patients and less than 1.0 ng/mL in 44 (62%). However, only 37 of 71 (52%) patients had glucose-suppressed GH concentrations within the calculated reference intervals (defined by the 95th percentile of normal). Comparing these data with the patient's concentrations of insulin-like growth factor-1 (IGF-1; normal range first established and corrected for age and sex in 494 healthy individuals), congruency of both parameters was found in 59 (77%) patients with an unexplained discrepancy between GH and IGF-1 in the remaining in 16 (23%) patients. Our results confirm that concentrations of IGF-1 must be corrected for sex and age, whereas glucose-suppressed concentrations of GH depend on age and BMI. "Across-the-board" cut-off-values are clearly inadequate and should not be used. Rather, serum GH measurements obtained during an OGTT must be interpreted individually by comparison to control values taking into account both age and BMI. Copyright 2003, Elsevier Science (USA). All rights reserved.