ESPE Abstracts (2019) 92 P1-397

Questioning the Value of Brain MRIs in the Evaluation of Children with Isolated Growth Hormone Deficiency

Asaf Oren1,2, Dana Singer1, Mariana Rachmiel3,2, Uri Hamiel3, Shelly Shiran1,2, Liat Ben-Sira1,2, Anita Schachter-Davidov1,2, Ori Eyal1,2


1Dana-Dwek Children's Hospital, Tel-Aviv Medical Center, Tel Aviv, Israel. 2Sackler School of Medicine, Tel-Aviv University, Tel Aviv, Israel. 3Assaf Harofeh Medical Center, Zerifin, Israel


Background: Isolated growth hormone deficiency (IGHD) is a relatively common disorder. Current diagnostic protocols require a brain MRI of the hypothalamus and the hypophysis after establishment of the diagnosis, with the aim of identifying structural defects and specifically rule out an underlying space-occupying lesion. An MRI scan is costly and requires general anesthesia in young children. Data on the contribution of brain MRI in diagnosing children with IGHD are sparse.

Objectives: To examine the yield of brain MRI in the evaluation of children with IGHD and to define clinical and laboratory parameters that justify its performance.

Methods: A retrospective chart review of all children (<18 years) diagnosed with IGHD at two pediatric endocrinology units between 2008 and 2018 for auxologic, laboratory, and brain MRI findings.

Results: The study included 129 children (72 boys) with confirmed IGHD. The mean age at diagnosis was 7.5 ± 3.8 years (median 7.7 years, range 0.8-15.9). Boys were diagnosed at a younger age than girls (6.8 ± 3.7 vs. 8.5 ± 3.8 years, respectively, p=0.02). The mean height SDS at diagnosis was -2.2 ± 0.8. The mean height deficit SDS (defined as the difference between height SDS at diagnosis and mid-parental height SDS) was -1.7 ± 0.9. Five children (3.9%) had pathologic findings on their MRI: two had ectopic posterior hypophysis, two had hypoplastic hypophysis and one had Rathke cleft cyst. Six children (4.6%) had incidental findings of Chiari type 1 malformation. No space-occupying lesion was detected. The mean height deficit SDS among the children with pathological MRIs was -3.2 ± 1.4 vs. -1.6 ± 0.8 amog the children with normal MRIs (p=0.007). Both a height deficit threshold of ≥2 SDS and a peak GH level threshold of ≤ 6.5 µg/liter identified all the pathological cases (sensitivity of 100% and specificity of 83%).

Conclusion: Our preliminary data indicate that most brain MRIs performed for routine evaluation of children with IGHD are not essential for establishing diagnosis. Only the children with extreme height deficit (≥2 SDS) and peak GH ≤6.5 µg/liter had pathological brain MRIs. Further studies with larger cohorts are needed in order to validate this revision of current protocols.

Article tools

My recent searches

No recent searches.