ESPE2022 Poster Category 1 Fetal, Neonatal Endocrinology and Metabolism (30 abstracts)
1Endocrinology Unit, Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico, Milan, Italy; 2Clinical Chemistry Unit, Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico, Milan, Italy; 3International Medical School, University of Milan, Milan, Italy; 4Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy; 5Neonatology and Neonatal Intensive Care Unit, Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico, Milan, Italy; 6University of Milan, Milan, Italy
Background: Severe congenital growth hormone deficiency (cGHD) is a rare but potentially life-threatening condition. Even though random growth hormone (GH) can confirm cGHD during the first week of life, the diagnosis remains extremely challenging in the absence of reliable reference values in healthy neonates and thus of a best diagnostic cut-off.
Aims: First, to provide solid reference values for GH concentrations in term newborns, by means of a non-invasive procedure (GH from screening cards), using a current ultrasensitive GH assay. Secondly, to investigate eventual maternal and neonatal predictors of GH concentrations.
Methods: Using Immulite 2000 assay, GH was measured simultaneously from 200 dried blood spots (DBS) and serum samples of controls, thus validating this method for DBS. With the same assay, GH concentrations were measured in 444 filter papers of term newborns after 48 hours of life. Maternal and neonatal anamnestic data were collected from clinical records.
Results: In our cohort (444 neonates, 212 males and 232 females), the auxological parameters were spread according with the reference neonatal anthropometric charts of Italian population (median length -0.05 SDS, median head circumference -0.07 SDS, median weight -0.12 SDS). Median GH value was 16.9 µg/l (IQR 11.2 - 23.1 µg/L), with no significant gender difference. We defined a lower limit of 6.5 µg/l as 5° centile through Harrell-Davis’ method with a confidence interval at 90 % between 5.9 and 7 µg/l by bootstrap BCA method. Considering our lower limit of GH<6.5 µg/l, at logistic regression analysis, jaundice presented the greater association with GH concentrations (P<0.001). Indeed, neonates with jaundice had 5-fold increased risk of presenting a GH<6.5 µg/L. No other significant correlation was found between GH and maternal or neonatal characteristics. Yet, the association between neonatal hypoglycaemia and lower GH concentrations was suggestive, though not significant (P=0.077).
Conclusions: We defined the lower limit of reference values for GH in term healthy new-borns independently from sex, gestational age and auxological parameters using for the first time a widely available assay. The relationship found between symptoms/signs suggestive for cGHD and lower GH concentrations, even in healthy subjects, confirmed the crucial role of clinical presentation in the diagnosis of cGHD. Providing solid reference values in term newborns, we put the basis for further studies aiming at defining a reliable diagnostic cut-off of cGHD.