Background: Growth hormone (GH) stimulation testing is necessary for the diagnosis of growth hormone deficiency (GHD). Insulin tolerance test (ITT) has been considered the gold standard for evaluating GHD in adults. However, it carries a risk of rare but severe adverse effects secondary to hypoglycaemia and is therefore avoided in many centres. There is no consensus for the first test in children.
Aim: Audit to compare ITT to Arginine test as a first line test for GH deficiency evaluation in children presenting with short stature.
Methods: All patients with possible GHD seen in the Endocrine Department of Birmingham Childrens Hospital between February 2015 and February 2017 were tested for GH secretion assessment. During 2015 all patients had ITT as a first test with 0.1 Units/kg of insulin, unless otherwise indicated, and if that was positive for GHD, a second test followed (usually Arginine/Glucagon). During 2016 Arginine provocation was used as a first line test, along with short synacthen test, and was followed by an ITT or Glucagon test. Patients with brain tumours or other conditions, not requiring a second test, and those undergoing end of growth test for adult GHD were excluded. Patients with a peak GH result <6.7 μg/l, underwent a second test. Children >10 years with no signs of puberty were primed with stilboestrol (1 mg BD for 2 days).
Results: 14/30 children with ITT first needed second test (47%), while 12/40 children with Arginine test first needed second test (30%) (P=0.15). 64.3% of the children who had an ITT first had a false positive result for GHD, while only 8.3% of those who had Arginine test first had a false positive result (P=0.0053). There was no correlation between lack of priming and false positive results.
Discussion: Even though ITT is considered the gold standard for the diagnosis of GHD, there is a high incidence of false positive results. Therefore, Arginine should be considered as a first line test, as it is a relatively safe test, easily applicable in most centres and could reduce the number of repeat stimulation tests. Larger studies are needed to confirm these results.