Completion thyroidectomy in patients with thyroid carcinoma initially submitted to lobectomy

In their article entitled ‘Circulating ghrelin levels in the newborn are positively associated with gestational age’, Bellone et al . (2004) found no correlation between ghrelin levels in cord blood and anthropometric or biochemical parameters (GH, IGF-I, insulin, glucose and leptin levels) in adequate for gestational age newborns. I would like to point out some observations concerning this study. Endocrine activities of ghrelin are mediated by GH secretagogue receptor (GHSR)-1a, a G-protein-coupled receptor mainly expressed in the pituitary and hypothalamus, previously identified as the receptor for GH secretagogues, a group of molecules endowed with strong GH release activity (Howard et al ., 1996). Ghrelin may be esterified with octanoic acid on serine 3. Acylation of ghrelin is required for GHSR-1a activation, whereas des-acyl ghrelin, which is far more abundant than octanoylated ghrelin (Hosoda et al ., 2000), does not bind GHSR-1a and it is devoid of any pituitaric or pancreatic endocrine activity (Kojima et al ., 1999; Muccioli et al ., 2002; Torsello et al ., 2002; Broglio et al ., 2003). The polyclonal antibody used by the authors (Phoenix Pharmaceutical, Belmont, CA, USA) does not distinguish octanoylated and nonoctanoylated ghrelin. Therefore, the authors measured the total ghrelin levels in cord blood and they did not measure the levels of active ghrelin. I believe that the correlation of total ghrelin levels with anthropometric and biochemical parameters may be misleading because only octanoylated ghrelin and not total ghrelin has biological effects and in particular stimulates GH secretion. Interestingly, Linco Research (St Charles, MO, USA) produces an antibody that is specific for the biologically active form of ghrelin with the octanoyl group on serine 3 (cat. no. GHRA-88HK). The active form of ghrelin is very unstable and labile in plasma; as a consequence, special precautions must be taken. In particular, blood samples should be kept in ice and processed as quickly as possible after blood is withdrawn; in addition the plasma should be acidified. In the light of these considerations, the authors’ conclusion that ‘ghrelin levels are not associated to GH secretion or to IGF-I levels, suggesting that ghrelin is not involved in perinatal GH hypersecretion’ cannot be justified on the basis of the presented data. Further investigations are required to correlate the levels of octanoylated ghrelin levels in cord blood with those of GH, IGF-I, insulin, glucose and leptin.