Age changes the diagnostic accuracy of mean profile and nadir growth hormone levels after oral glucose in postoperative patients with acromegaly

Abstract

Background: This analytical, retrospective study was designed to select cut-off thresholds of mean GH levels during a diurnal profile and nadir GH levels after oral glucose tolerance test (OGTT) according to age to diagnose surgical remission of acromegaly.

Methods: One hundred forty-one patients (76 women, aged 44 +/- 15 years and 65 men, aged 43 +/- 13 years) were included in this study. For the purpose of this study, remission was based on insulin-like growth factor-I (IGF-I) levels in the normal range for age. Diagnostic accuracy was analysed by receiving-operator characteristics (ROC) curves in the entire series, and in young (20-40 years), middle-aged (41-60 years) and older patients (> 60 years), separately.

Results: Sixty patients (42.6%) had normal IGF-I levels after surgery. In the entire series, in young and in middle-aged patients, the ROC analysis showed that optimum cut-off for mean GH levels was 2.3 microg/l (diagnostic accuracy range, 94-97%) whereas that for nadir GH after OGTT were, respectively, 0.85, 0.9 and 0.8 microg/l (diagnostic accuracy range, 90-95%). In the older patients, the optimum cut-off selected for mean GH levels was 1.4 microg/l and that for nadir GH after OGTT was 0.5 microg/l (diagnostic accuracy, 100% for both). The comparative analysis of the ROC curves did not show any significant difference between mean GH and nadir GH after OGTT (P = 0.21).

Conclusions: The criteria currently accepted for diagnosing post-surgical remission of acromegaly have high diagnostic accuracy only in the patients aged below 60 years. In older patients, lower cut-offs (i.e. = 1.4 microg/l for fasting GH and = 0.5 microg/l for nadir GH after OGTT) predict normal IGF-I levels. Mean GH levels during a diurnal profile have similar diagnostic accuracy of nadir GH levels after OGTT. This suggests that OGTT is not necessary to establish surgical cure.