Correspondence *Division of Pediatric General and Thoracic Surgery, ML 2023, Cincinnati Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229-3039, USA

Email thomas.inge@cchmc.org

The case

A previously healthy 14-year-old boy presented to the primary care physician with fatigue, daytime somnolence, intermittent emesis, and hypothyroidism. The patient did not suffer from headache or visual impairment. A CT of the head revealed a large (3 4.5 4 cm) midline suprasellar mass arising from the sella turcica with amorphous calcification compressing the third ventricle (Figure 1). At diagnosis, the patient's height was 170 cm (50^th percentile for his age), his weight was 70.9 kg (85^th percentile), and his BMI was 25 kg/m²(92^nd percentile; Figure 2). The patient underwent subtotal resection of the lesion through a right-sided orbitozygomatic craniotomy. Histological findings were typical for an adamantinomatous craniopharyngioma (i.e. a slow growing, extra-axial, epithelial–squamous, calcified cystic tumor arising from remnants of the craniopharyngeal duct and/or Rathke cleft).

Figure 1 MRI scan of the patient's brain before craniotomy; posterior view (A) and lateral view (B)

An hourglass shaped, predominantly cystic mass with peripheral enhancement is seen in the sella turcica and suprasellar cistern in this gadolinium-enhanced, T1-weighted MRI scan of the brain. The lesion was causing a mass effect on the third ventricle.

Full figure and legend (25K)Figures & Tables indexDownload Power Point slide (204K)

Figure 2 BMI changes of the patient and clinical events over time

At the time of gastric bypass surgery, the patient's BMI was 67 kg/m², his weight was 223.5 kg, and his excess weight was 150 kg, representing 204% over ideal weight for his gender and age. At the longest available follow-up (2.5 years following operation) the patient had a weight loss of 49 kg; that represents 22% of his initial weight or 33% of his initial excess weight.

Full figure and legend (18K)Figures & Tables indexDownload Power Point slide (196K)

The patient was treated with radiation therapy of 54 Gy over the following 3 months. He subsequently developed panhypopituitarism and diabetes insipidus. Hormonal replacement therapy was initiated at the following doses (titrated to the targets shown): levothyroxine 100 g/m² per day, orally (target: free T₄ level in the highest third of the normal range [23–28 pmol/l]), desmopressin 0.2 mg twice daily, orally (target: pre-dose serum sodium level in the normal range [135–145 mmol/l], no nocturia, daily diuresis), growth hormone (started at age 16 years) 0.1 mg subcutaneous injection daily (target: level of insulin-like growth factor 1 in the mid-normal range [400–500 ng/ml]), prednisone 5 mg/m² (triple usual dose), and depot intramuscular testosterone, 200 mg every 2 weeks (started at age 17 years).

During the 24 months following tumor resection, the patient's appetite markedly increased and he gained weight at an extreme rate of 70 kg per year (Figure 2), despite outpatient and inpatient dietary and physical activity interventions. His height increased from 173 cm at 15 years of age to a plateau of about 180 cm at 17 years. He developed hyperinsulinemia associated with the extreme rate of weight gain, and octreotide therapy was initiated. The initial octreotide dose was 5 g/kg/day (350 g/day), it was increased to 1,000 g/day over 5 months¹ and continued at this daily dose for another 14 months. Octreotide markedly attenuated serum insulin excursions in response to oral glucose loading, and no evidence of hyperglycemia was observed (data not shown).

Over the entire 19-month octreotide therapy, the patient's weight gain markedly slowed (from 70.0 kg/year to 9.5 kg/year) but no weight loss was observed. After gaining approximately 150 kg, he developed obstructive sleep apnea (which required treatment with nocturnal continuous positive airway pressure); concentric left ventricular hypertrophy (left ventricular mass index of 68.5 g/m^2.7 [normal value: <50 g/m^2.7]), and hypertriglyceridemia (3.12 mmol/l, normal range 0.34–2.26 mmol/l).

The patient was referred for an evaluation at the Comprehensive Weight Management Center that included consultation with a bariatric surgeon, a pediatric weight management specialist, a psychologist, and an endocrinologist. In view of his extreme obesity, comorbidities, and lack of weight loss with prior interventions, laparoscopic Roux-en-Y gastric bypass surgery was offered. Octreotide therapy was discontinued. To reduce vagal contribution to hyperinsulinemia,^{2, 3} a truncal anterior vagotomy was also performed. The procedure was performed at the age of 18 years (approximately 3 years following craniopharyngioma resection and irradiation), with no postoperative complications.

After the operation, the patient's weight gain ceased (Figure 2), followed by 49 kg weight loss over the ensuing 2.5 years, while his height remained 182 cm (79^th percentile). Over the first postoperative year, standard hormone replacement for hypopituitarism was continued. There was no macronutrient or vitamin deficiency (albumin, vitamin B₁ and vitamin B₁₂ levels were normal). Mild iron-deficiency anemia developed (hemoglobin level was 123 g/l [normal range 133–177 g/l], serum iron concentration was 7.34 mol/l [normal range 8.77–32.40 mol/l], and ferritin concentration 10 g/l [normal range 5–244 g/l]). The patient was therefore given iron supplementation. After maximal weight loss, reassessment showed that the patient's serum triglyceride level (1.85 mmol/l), left ventricular mass index (46 g/m^2.7), and snoring had markedly improved. His food cravings also diminished considerably (Table 1). He has maintained his weight loss for 2.5 years after Roux-en-Y gastric bypass and returns annually for clinical visits.

Table 1 The patient's food cravings before and after bariatric surgery^a

Full tableFigures & Tables indexDownload Power Point slide (216K)

Discussion of diagnosis

Craniopharyngiomas are relatively uncommon neoplasms, accounting for approximately 10% of all pediatric brain tumors, with peak age of onset between 5 and 14 years.⁴ Although craniopharyngiomas are usually benign, their suprasellar location often makes these lesions quite debilitating.⁵ Tumor resection with or without radiotherapy represents the therapeutic standard of care. Unfortunately, critical areas of the pituitary gland and hypothalamus can be adversely affected by surgical or radiation treatments. Hypothalamic and pituitary damage is associated with significant morbidity, mostly by inducing partial hypopituitarism or panhypopituitarism.

Many patients who are treated for craniopharyngioma are obese at follow-up.^{6, 7} Despite optimal endocrine management, hyperphagia can result in extreme obesity⁸ and major metabolic complications of obesity.⁹ 'Hypothalamic obesity'¹⁰ after craniopharyngioma treatment could be a direct result of neurosurgical or radiation damage to ventromedial hypothalamic (VMH) neurons that are critical for appetite regulation and weight homeostasis. Alternatively, VMH damage could result in obesity due to disinhibition of vagal tone, and excess vagal stimulation of pancreatic -cell insulin secretion. Because of the intracranial insult, the hypothalamus would be unable to respond normally to circulating gut hormones, including insulin, ghrelin, and leptin. Preoperatively, we documented fasting hyperinsulinemia, an exaggerated insulin response to meal challenge, and markedly increased insulin excursions between meals. (The patient's insulin, ghrelin and leptin responses to meal challenges are shown on Supplementary Figure 1 online.) Hyperleptinemia, insulin resistance, and low ghrelin levels observed in this patient probably represent a compensatory response to hypothalamically induced obesity.

Treatment and management

Hypothalamic obesity is usually unresponsive to diet, physical activity, and most pharmacologic interventions.⁶ Little information is available about effective weight management treatment aside from octreotide therapy. Octreotide is thought to target the insulin excess that is associated with pathologic weight gain and has demonstrated modest effectiveness in weight reduction in some cases of hypothalamic obesity.¹ Indeed, in this case, octreotide effectively reduced insulin secretion in response to oral challenge, and arrested the dramatic rate of weight gain, but did not cause weight loss. There is no published information available regarding bariatric surgery for hypothalamic obesity; however, this approach seemed to be an appropriate treatment option, given that the best available behavioral and pharmacologic interventions had failed, and the patient was experiencing significant obesity-related comorbid conditions.

For patients with no neurological abnormalities, the currently prevailing hypothesis is that gastric bypass results in weight reduction by altering a complex interactive balance between gut hormones, the autonomic nervous system, and the brain circuitry regulating hunger and energy balance,¹¹ and does correct hyperinsulinemia effectively. When considering Roux-en-Y gastric bypass for this patient, the choice was made to surgically divide the anterior vagal trunk to limit vagally mediated insulin secretion. Although it is unlikely that anterior hemivagotomy is as effective as bilateral vagotomy for disruption of insulin stimulation, it is also possible that bilateral vagotomy could have unintended deleterious side effects after gastric bypass, such as gastric atony and stasis in the remnant stomach.

From a medical standpoint, it was important to assess preoperatively whether the patient's desmopressin and growth hormone doses were adequate, because gastric bypass results in restriction of fluid and caloric intake. A water-deprivation test confirmed diabetes insipidus. On desmopressin therapy, the patient was able to retain fluids and prevent dehydration satisfactorily. Glucose measurements were performed during an overnight fast to determine whether his growth hormone deficiency might lead to hypoglycemia in the event of impaired carbohydrate intake after bariatric surgery. During the fast, while he was receiving low dose growth hormone therapy, serum glucose and insulin-like growth factor 1 (IGF-I) levels remained normal; therefore, the risk of symptomatic hypoglycemia after gastric bypass surgery was considered low.

By 7 months postoperatively, fasting insulin levels had completely normalized, and by 14 months postoperatively there was no recurrence of hyperinsulinemia. Leptin levels were elevated before the operation and fell dramatically by 10 days after gastric bypass. With major weight loss, leptin levels decreased even further, ultimately falling below the level expected for the patient's degree of obesity (see Supplementary Table 1).

Although gastric bypass did not affect the overall ghrelin profile, there was a steady decrease in peak and basal active ghrelin concentrations postoperatively, consistent with previous reports.¹² Reduction in the preprandial rise of ghrelin level may have contributed to the markedly reduced food cravings measured in this patient. Despite the probable disruption of ghrelin-responsive circuitry in the VMH after hypothalamic damage, ghrelin's primary targets are believed to be situated in the arcuate nucleus.¹³ Other ghrelin-responsive neuronal circuits are proposed to exist in the brainstem¹⁴ and the midbrain.¹⁵ These would be unharmed by a craniopharyngioma or related invasive treatments. These ghrelin-responsive targets could have a role in the complex neuroendocrine interactions responsible for weight loss following gastric bypass in this patient and potentially in neurologically normal patients as well.

Conclusions

After treatment for craniopharyngioma, children are at high risk for obesity; more rarely, extreme obesity develops with associated comorbidities. There is no consensus regarding optimal weight management strategies. In spite of iatrogenic injury to central appetite regulating centers, gastric bypass surgery can be a safe and sufficiently effective treatment to arrest inexorable weight gain, counteract pathophysiology of extreme hypothalamic obesity, and improve obesity-related comorbidities without serious nutritional compromise.

Acknowledgments

Written consent for publication was obtained from the patient's responsible relative.The authors would like to express gratitude for the helpful scientific input of R Lustig, R Seeley, J Kral, M Vierra, and D D'Alessio. In addition, we appreciate the expert clinical care provided by V Garcia, S Kirk, S Xanthakos, H Roehrig, and J Sweeney.This work was supported in part by an NIH General Clinical Research Center grant #M01 RR 08084 including Clinical Research Feasibility Funding (CReFF) funding to MZ.

References

1. Lustig RH et al. (2003) Octreotide therapy of pediatric hypothalamic obesity: a double-blind, placebo-controlled trial. J Clin Endocrinol Metab 88: 2586–2592 | Article | PubMed | ISI | ChemPort |
2. Gold RM et al. (1980) Vagal mediation of hypothalamic obesity but not of supermarket dietary obesity. Am J Physiol 238: R447–R453 | PubMed | ISI | ChemPort |
3. Lustig RH (2006) Childhood obesity: behavioral aberration or biochemical drive? Reinterpreting the First Law of Thermodynamics. Nat Clin Pract Endocrinol Metab 2: 447–458 | Article | PubMed | ISI | ChemPort |
4. Haupt R et al. (2006) Epidemiological aspects of craniopharyngioma. J Pediatr Endocrinol Metab 19 (Suppl 1): S289–S293
5. Hopper N et al. (2006) The pre-operative endocrine assessment of craniopharyngiomas. J Pediatr Endocrinol Metab 19 (Suppl 1): S325–S327
6. Eyal O et al. (2006) Obesity in patients with craniopharyngioma. Endocrinologist 16: 286–293 | ISI |
7. Hoffman HJ et al. (1992) Aggressive surgical management of craniopharyngiomas in children. J Neurosurg 76: 47–52 | PubMed | ISI | ChemPort |
8. Pinto G et al. (2000) Hormonal factors influencing weight and growth pattern in craniopharyngioma. Horm Res 53: 163–169 | Article | PubMed | ISI | ChemPort |
9. Srinivasan S et al. (2004) Features of the metabolic syndrome after childhood craniopharyngioma. J Clin Endocrinol Metab 89: 81–86 | Article | PubMed | ISI | ChemPort |
10. Lustig RH (2001) The neuroendocrinology of childhood obesity. Pediatr Clin North Am 48: 909–930 | Article | PubMed | ISI | ChemPort |
11. le Roux CW et al. (2006) Gut hormone profiles following bariatric surgery favor an anorectic state, facilitate weight loss, and improve metabolic parameters. Ann Surg 243: 108–114 | Article | PubMed | ISI |
12. Cummings DE et al. (2002) Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 346: 1623–1630 | Article | PubMed | ISI |
13. Hewson AK and Dickson SL (2000) Systemic administration of ghrelin induces Fos and Egr-1 proteins in the hypothalamic arcuate nucleus of fasted and fed rats. J Neuroendocrinol 12: 1047–1049 | Article | PubMed | ISI | ChemPort |
14. Hou Z et al. (2006) Ghrelin-containing neuron in cerebral cortex and hypothalamus linked with the DVC of brainstem in rat. Regul Pept 134: 126–131 | Article | PubMed | ISI | ChemPort |
15. Abizaid A et al. (2006) Ghrelin modulates the activity and synaptic input organization of midbrain dopamine neurons while promoting appetite. J Clin Invest 116: 3229–3239 | Article | PubMed | ISI | ChemPort |
16. White MA et al. (2002) Development and validation of the food-craving inventory. Obes Res 10: 107–114 | PubMed | ISI |

Supplementary information

Supplementary Information (doc 319 KB)

Contact the journal about this article

Subject areas under which this article appears: Obesity | Neuroendocrinology (including the hypothalamus)