Growth hormone deficiency (GHD) in adult thalassaemic patients

Abstract

Background and objective: Short stature and growth hormone deficiency (GHD) are frequent occurrences in thalassaemic children, while data on the prevalence of GHD in adult patients are lacking. Therefore, we elected to study the growth hormone and insulin-like growth factor-I (GH-IGF-I) axis in a large group of adult thalassaemic subjects.

Design: Cross-sectional study on the prevalence of GHD in 94 adult thalassaemic patients (69 with thalassaemia major and 25 with thalassaemia intermedia, 39 men and 55 women, aged 31.5 +/- 6.8 years, on sex steroid replacement when necessary).

Methods: All patients underwent GHRH (1 microg/kg as an i.v. bolus) plus arginine (0.5 g/kg as a 30 min i.v. infusion) testing. Severe GHD was defined by GH peaks lower than 9 microg/l, whereas partial GHD was defined by GH peaks ranging from 9-16.5 microg/l. Blood samples for IGF-I, ferritin and pseudocholinesterase measurements were collected. Urinary free cortisol (UFC) levels were also assayed.

Results: Severe GHD was demonstrated in 21 of the 94 patients (22.3%), while 18 additional patients (19.1%) displayed partial GHD. GH peaks were positively correlated with IGF-I standard deviation score (SDS) (r = 0.22, P < 0.05), although 1 of the 21 patients with severe GHD showed normal IGF-I SDS values, and 44 of the 55 patients with normal GH reserve displayed low IGF-I SDS. A strong positive correlation (r = 0.48, P < 0.0001) between IGF-I SDS and pseudocholinesterase was identified. No correlations were found between ferritin and UFC levels on the one hand and GH peaks and IGF-I SDS on the other.

Conclusion: Findings from this study demonstrate that GHD, either partial or severe, is not a rare occurrence in adult thalassaemic patients. GHD is associated with a higher prevalence of low serum IGF-I levels, recorded also in patients with normal GH secretion. The lack of correlation between ferritin and both GH peaks and IGF-I SDS suggests that mechanisms additional to iron overload, whose relevance cannot however be definitely ruled out, play a role in the pathophysiology of somatotrophin-somatomedin deficiency in this clinical condition. The positive correlation between IGF-I SDS on the one hand and GH peaks and pseudocholinesterase values on the other hand indicates that reduced liver protidosynthetic activity, in addition to somatotrophin secretory status, is a major determinant of the impaired IGF-I production in thalassaemia. Therefore biosynthetic GH replacement therapy in GH-deficient thalassaemic adults is worth considering.