In the vast majority of cases, basal serum GH and IGF-1 levels are markedly increased in patients with obvious clinical signs and symptoms of acromegaly. The oral glucose tolerance test (OGTT) is useful for diagnosis in the minority of patients who have weak GH hypersecretion. The cutoff for a "normal" GH nadir in the OGTT remains to be agreed. The type of GH assay, its sensitivity, the type of standard used by the manufacturer, the patient's age and especially gender, must all be taken into account. Recent studies using new highly sensitive assays suggest an upper normal GH nadir of 0.71 microg/l for female healthy patients, but no "universal" cut-off has yet been defined for healthy males (from 0.057 to 0.25 microg/l). The 1 microg/l cutoff proposed for the diagnosis of acromegaly in a 2000 consensus should be abandoned in favor of a 0.30 microg/l cutoff. Clinicians should know which assay is used, together with its sensitivity and the standard, before making therapeutic decisions. A more pragmatic view should probably be adopted when assessing the treatment response. Indeed, if "cure" is defined not with the <1 microg/l GH nadir but on the basis of healthy control values, many patients will not be considered controlled. However, the clinical relevance of such goal (e.g. achieving GH nadir <0.4 microg/l rather than <1 microg/l) in terms of prognosis and prediction of outcome on long term is not firmly established. Thus, from a pragmatic point of view, achieving a normal age-adjusted IGF-1 level and a GH nadir below 1 microg/l during OGTT will probably remain relevant for defining remission and good disease control in terms of morbidity and mortality in acromegaly.