Growth is one of the most important indicators of child's well-being. The growth pattern is the result of the complex interaction between genetic and environment factors. The gold standard for the diagnosis of GH-deficient (GHD) children is not yet defined. In this review we focused on sensitivity and specificity of the commonly used test for GH secretion and neuroimaging evaluation. Initially, accurate determinations of height, weight, head circumferences and growth velocity should be performed and plotted on an appropriate growth chart. Once chronic non-endocrine diseases such as celiac disease, chronic inflammatory bowel diseases, liver disorders and renal failure have been excluded, children whose height is lower than the 2nd percentile or 2 SD below the mean associated with a decreased height velocity (less than 25th percentile), or height less than -2.4 SD alone, or height velocity less than 10th percentile alone, should be investigated for possible abnormalities in the GH-IGF-I axis. A single IGF-I determination will be the great diagnostic value in all patients with severe GHD or GH insensitivity. Stimulation tests and imaging studies will confirm the etiology. One of the major problems of the provocative tests lies in their poor reproducibility and in the great number of falsely abnormal responses observed also in normal children. Genetic tests are also indicated in all cases of genetic forms. Patients with growth failure and abnormal MRI findings have a very high probability of having GHD and can be identified easily by serum IGF-I or by GH testing. All patients with subnormal GH responses to pharmacological stimulation either with normal or low IGF-I concentration but normal MRI findings should be followed up before a diagnosis of GHD is firmly established.