Stability of IGF-I concentration despite divergent results of repeated GH stimulating tests indicates poor reproducibility of test results.

OBJECTIVE

Growth hormone (GH) secretion is routinely assessed in provocative tests. However, some limitations of these tests have been reported, including a weak correlation between spontaneous GH secretion and GH peak in stimulating tests, poor reproducibility of test results and normalisation of previously decreased GH response to stimulation in the repeated provocative tests. The observed discrepancies between the results of consecutive GH stimulating tests in the same patient may be explained either by real changes in GH secretion or by the poor reproducibility of GH response to stimulation. Normalisation of previously decreased GH secretion should entail the increase of insulin-like growth factor-I (IGF-I) concentration, while stability of IGF-I secretion despite divergent GH response to stimulation in repeated tests might indicate the poor reproducibility of GH stimulating tests. continue with The aim of the study was a comparison of GH peak in repeated stimulating tests and of corresponding, simultaneously measured IGF-I concentration.

METHODS

The investigation comprised 84 children with short stature who underwent repeated GH tests and IGF-I evaluation. In 60 patients each of two different tests (with clonidine and with glucagon, in standard doses) was performed twice, together with IGF-I measurement during both evaluations. In 20 patients (remaining in the same pubertal stage during the time period between the procedures in question) at least one test, together with IGF-I measurement, was repeated within one year.

RESULTS

The correlation between the results of GH tests, repeated during the two evaluations, was weak (r=0.22, p<0.05 for all patients and r=0.25, p<0.05 for the patients examined 2 times within 1 year), with the high within-subject variability (43.4 % and 59.5 %, respectively). Conversely, the correlation between two values of IGF-I SDS was good both for all patients (r=0.65, p<0.05) and especially for the patients examined twice within 1 year (r=0.96, p<0.05), with low within-subject variability for the latter subgroup of patients (11.2 %). The same GH stimulating tests, performed twice in the same patient, led to different conclusions (either the confirmation of GHD diagnosis or its exclusion) in most of examined patients. Poor reproducibility of GH stimulating tests, rather than the possibility of short-term changes in GH secretion, was confirmed.