Objective: The objective of this study was to analyze the performance of 2 stimulation tests used in the diagnosis of growth hormone deficiency.
Method: A retrospective study was conducted on a non-random sample of 310 patients, between 2 and 20 years old, who were hospitalized in the Mureș County Hospital’s Endocrinology Department and in the National Institute of Endocrinology C.I. Parhon with short stature between 2009-2015. Inclusion criteria: all subjects who underwent growth hormone stimulation tests in accordance with the national protocol. Microsoft Office Excel was used for data collection and MedCalc v 12.5 was used for statistical analysis.
Results: From the total of 310 patients, 102 were diagnosed in Târgu Mureș and 208 in Bucharest. Sex ratio favored boys (boys:girls 1.64:1). In 173 subjects growth hormone deficiency was confirmed. For both tests the percentage of maximum response was the highest for the 60 minutes blood sample regardless if the test were positive or not. Both tests have 100% sensitivity and negative predictive value, with the highest specificity for the 60 minutes clonidine and 30 minutes insulin. The false positive rate was 60% for the insulin test and 27.2% for clonidine for Târgu Mureș sample and 86.9% for the insulin test and 62.5% for clonidine for Bucharest sample. The concordance of the 2 tests was 49.36%.
Conclusions: Stimulating growth hormone testing has a number of limitations but is still needed in some auxological circumstances. We recommend performing the clonidine test first to exclude idiopathic short stature and then the insulin tolerance test for the diagnosis of growth hormone deficiency.