As the sensitivity of a single GH test is poor, current NICE guidelines (2010) state that to make a diagnosis of isolated GH deficiency (IGHD), two stimulation tests need to show subnormal peak GH levels. In our centre we use insulin tolerance (ITT) or glucagon stimulation (GST) as the 1st test, and arginine stimulation (AST) as the 2nd test.
The purpose of this study was to identify the proportion of children with discrepant test results; and to establish whether lowering the peak GH cut-off might reduce the discrepancy between tests. Our own assay-specific normal peak GH cut off is >5.7 μg/l.
80 children who received two GH stimulation tests in our unit between January 2002 and June 2011 were identified:
Of the 52 with an abnormal ITT, 22 (42%) subsequently had a normal AST.
Of the 28 with an abnormal GST, 7 (21%) subsequently had a normal AST.
The correlations for peak GH levels between the 1st and 2nd tests were moderate, at 0.49 and 0.51 for ITT and GST vs AST respectively.
Lowering the cut off value for the ITT test result to <2.7 μg/l to characterise severe GHD increased the proportion of abnormal second test results from 58 to 68%, but this cut off would miss 15 (50%) children with a peak GH level of 2.75.7 μg/l on their 1st test.
In summary, a significant proportion of children have an abnormal 1st test and a normal 2nd test. There is no absolute peak GH cut-off value on the first test that will predict an abnormal 2nd test result. Until a better biomarker becomes available, the NICE recommendation for two GH stimulation tests to diagnose IGHD stands.