Transition From Diencephalic Syndrome to Hypothalamic Obesity in Children With Suprasellar Low Grade Glioma: A Case Series

Abstract

Background: Children with suprasellar low grade glioma (LGG) frequently develop problems to maintain their body weight within the normal range, due to hypothalamic dysfunction. Hypothalamic damage may result in the diencephalic syndrome (DS), characterized by underweight or failure to thrive, but also in hypothalamic obesity (HO). Children with LGG presenting with DS at young age often develop HO later in life. The underlying pathophysiology for this change in body mass index (BMI) is not understood. Previous hypotheses have focused on the tumor or its treatment as the underlying cause. To better understand its etiology, we aimed to relate changes in BMI over time in children with suprasellar LGG presenting with DS to age, tumor progression, treatment, and endocrine function. We hypothesize that the development of HO in children with LGG presenting with DS is related to maturation status of the hypothalamus at time of injury and thus age.

Methods: In this retrospective case series, all cases diagnosed in the Netherlands with suprasellar located LGG, currently treated or followed, with a history of DS developing into HO were included.

Results: In total, 10 children were included. Median age at LGG diagnosis was 1.5 years (range 0.4-5.5), median BMI SDS was -2.64. The children developed overweight at a median age of 4.5 years (2.2-9.8). The median total difference in BMI SDS between underweight and obesity was +5.75 SDS (4.5-8.7). No association could be found between transition of DS to HO and onset of a pituitary disorder (present in 70.0%), surgery, chemotherapy, or tumor behavior. Two had developed central precocious puberty (CPP), both while having underweight or normal weight.

Conclusion: The shift from DS to HO in children with hypothalamic LGG may be associated with age and not to tumor behavior, treatment characteristics or pituitary function. The development of CPP in these children seems not to be related to obesity. Our findings may indicate that the clinical picture of hypothalamic dysfunction reflects the maturation state of the hypothalamus at time of lesioning. Future prospective studies are needed to better understand underlying causative mechanisms of the morbid changes in body weight.

Keywords: Low Grade Glioma (LGG); diencephalic syndrome; hypothalamic obesity; hypothalamic pituitary; suprasellar tumor.

Figure 1

Case of DS evolving into HO to support Table 2. On the X axis the age is shown and on the Y axis the BMI. Various time points as displayed in Table 2 are shown here.

Figure 2

Longitudinal BMI SDS over time. (A) Logistic regression (LOESS) plot of all patients. x-axis: age in years at moment of BMI follow-up, y-axis: BMI SDS corrected for age and gender. Orange beam shows 95% confidence interval. (B) Individualized BMI SDS data for each patient. This figure shows that the regression curve represents most of the patients, with only two outliers observed. Diencephalic syndrome defined as BMI SDS <−1.6 SD. Overweight defined as BMI SDS >+1.6 SD. Legend shows patients in correspondence with Table 2.

Figure 3

Prevalence of pituitary disorders (anterior pituitary deficiency, CPP, DI). ACTHD, adrenocorticotropic hormone deficiency; CPP, central precocious puberty; DI, diabetes insipidus; GHD, growth hormone deficiency; TSHD, thyroid-stimulating hormone deficiency. The circles of ACTHD and DI are completely overlapping: every child with ACTHD also had DI. None of the children had LH/FSHD. N = 3 had no pituitary disorder. N = 1 had CPP. N = 1 had CPP and TSHD. N = 1 had GHD and TSHD. N = 1 had TSHD, ACTHD and DI. N = 3 children had GHD, TSHD, ACTHD, and DI.

Figure 4

Longitudinal IGF-1 SDS over time in children with suprasellar LGG presenting with DS. Legend shows patients in correspondence with Table 2 and Figure 2. (A) IGF-1 SDS values over time in all patients. x-axis: age in years at time of measurement, y-axis: IGF-1 SDS corrected for age and gender. (B) IGF-1 SDS values over time for patients with GH hypersecretion only (for 1 patient with GH hypersecretion, data on IGF-1 values over time were missing). (C) BMI SDS (x-axis) according to IGF-1 SDS (y-axis) in all patients. Correlation coefficient for BMI-SDS and IGF-1 SDS: −0.37 (p <0.001). Only IGF-1 values without growth hormone treatment or use of somatostatin analogues are shown.