Objectives: In the glucagon stimulation test (GST) growth hormone (GH) secretion is considered sufficient when at least 1 value is >10 ng/mL. Because GH typically peaks at 90 or 120 minutes, we evaluated whether peak occurrence at other times ("atypical") signifies abnormal GH secretion.
Study design: A retrospective review of 222 GSTs was conducted to determine these outcomes: 1) frequency of GH deficiency per confirmatory clonidine or arginine stimulation test in typical versus atypical GSTs, and 2) growth velocity standard deviation score (GVSDS) in patients with typical versus atypical GSTs.
Results: Of 222 tests, 57 GST results (25.7%) were atypical, and 54 GST results (24.3%) were deficient. Atypical deficient tests had a higher chance than typical deficient tests of predicting GH deficiency per confirmatory test (15/21, 71.4% versus 14/33, 42.4%; P < .05). Patients with deficient atypical GST results and sufficient confirmatory tests (n = 6) had a lower GVSDS than patients with deficient typical GST results (n = 18; -1.58 [-3.2-1.76], versus 0.23 [-1.54-3.95], P = .03). Overall, 75% of atypical deficient GST results were followed by atypical timing in a clonidine test. Three of the 222 tests (1.3%) peaked at 180 minutes.
Conclusions: The GST can be terminated at 150 minutes without sacrificing sensitivity. A GH peak at a time other than 90 or 120 minutes may be a new important indicator of GH deficiency.