Background: Since the publication of international growth charts by the World Health Organization (WHO) in 2006, the use of national growth charts for growth monitoring (GM) has been questioned.
Objective and hypotheses: To evaluate the potential impact of using WHO vs. national growth charts on the performance of a clinical decision rule for detecting children with one of the target conditions of GM: GH deficiency (GHD).
Method: In a case-referent study, we applied the Grote clinical decision rule on growth data of 33 children with GHD related to pituitary-stalk interruption syndrome (cases), and 2 200 apparently healthy children followed longitudinally from birth (referents). The Grote clinical decision rule is mainly based on the following auxological criteria combined in various ways: standardised height, distance to standardised target height, absolute height deflection, and small for gestational age with no catch up after 3 years (Grote 2008). The sensitivities, specificities and theoretical improvement in time to diagnosis of the rule using French or WHO growth charts were calculated and compared using McNemar or Wilcoxon tests for matched pairs/series.
Results: The application of the Grote clinical decision rule would have led to a higher sensitivity (78.8% vs 66.7%, P=0.04), and a lower specificity (98.3% vs 99.2%, P<0.01) with the WHO vs. French growth charts, respectively, with no statically significant theoretical improvement in time to diagnosis (9 months vs 4, P=0.12).
Conclusion: The use of WHO growth charts to apply the Grote clinical decision rule for the early detection of GHD would have notable impacts on false-positive rate. Indeed, among the 2012 French birth cohort (n=822 000), 14 220 vs 6 500 would have been inadequately referred using WHO vs. French growth charts respectively.
01 - 03 Oct 2015
European Society for Paediatric Endocrinology