Approach to the Patient With Hypothalamic Obesity

Abstract

Hypothalamic obesity (HO) is defined as abnormal weight gain due to physical destruction of the hypothalamus. Suprasellar tumors, most commonly craniopharyngiomas, are a classic cause of HO. HO often goes unnoticed initially as patients, families, and medical teams are focused on oncologic treatments and management of panhypopituitarism. HO is characterized by rapid weight gain in the first year after hypothalamic destruction followed by refractory obesity due to an energy imbalance of decreased energy expenditure without decreased food intake. Currently available pharmacotherapies are less effective in HO than in common obesity. While not a cure, dietary interventions, pharmacotherapy, and bariatric surgery can mitigate the effects of HO. Early recognition of HO is necessary to give an opportunity to intervene before substantial weight gain occurs. Our goal for this article is to review the pathophysiology of HO and to discuss available treatment options and future directions for prevention and treatment.

An 8-year-old boy presented to the emergency room with a 3-month history of daily generalized headaches, possible photophobia, and 1 day of altered mental status. In the month before presentation, he had been seen by his primary care physician and an urgent care clinic and treated for seasonal allergies and a sinus infection without relief. The headaches were not relieved by acetaminophen or ibuprofen. In the preceding 24 hours, his parents noted decreased attention and increasing somnolence. On examination he was arousable, but speech was slow. Vital signs were normal, and examination was unremarkable. A head computed tomography scan without contrast revealed a suprasellar mass with internal calcifications, measuring 4.1 × 6 cm and causing obstructive hydrocephalus, consistent with a craniopharyngioma. He was admitted to the pediatric intensive care unit for monitoring and external ventricular drain placement. Ophthalmology was consulted and no gross visual deficits were found. The mass was completely resected surgically, and pathology confirmed craniopharyngioma. The pituitary stalk was severed during the resection.

Postoperatively, the patient was treated with 5 days of dexamethasone for swelling, followed by physiologic hydrocortisone dosing at 9 mg/m²/day for presumed adrenal insufficiency. He was discharged on hydrocortisone, along with desmopressin and levothyroxine for panhypopituitarism. He returned to the endocrine clinic for follow-up 2 weeks post hospital discharge. His weight had increased from 37 kg to 40.1 kg and his body mass index (BMI) was 24.9 (98th percentile). On review of his medical records, it was noted that his weight had been tracking along the 75th percentile before age 6 years but had increased to the 90th percentile when he presented with craniopharyngioma (Fig. 1). The family denied food-seeking behavior but noted that he ate a larger portion at meals. His activity level had decreased, attributed to fatigue and postoperative vision loss. He was started on growth hormone without improvement in weight gain or BMI. Over the next 6 months, his weight increased to 52 kg and BMI was 30.7 (> 99th percentile on the Centers for Disease Control BMI growth chart).

Figure 1.

The case patient's data plotted on a Centers for Disease Control and Prevention male growth chart, weight for age.

Background

Hypothalamic obesity (HO) is defined as abnormal weight gain due to physical destruction of the hypothalamus. Suprasellar tumors, most commonly craniopharyngiomas, are a classic cause of HO. Approximately 60% of patients who present with craniopharyngioma go on to have HO (1). Other causes include cranial radiation, vasculitis, and head trauma. The hallmark of HO is an increase in the rate of weight gain. The weight gain may be due to hypothalamic damage from the underlying disease process (tumor infiltration, etc) or treatment modalities (surgery, radiation) (2). At presentation, individuals may already have weight gain, stable weight, or even weight loss, but any of these individuals may go on to have HO following treatment. Evaluation of hypothalamic damage on magnetic resonance imaging (MRI) following surgery can predict increased risk of HO based on specific nuclei that are damaged. There are ongoing research efforts to use MRI scoring systems to predict risk of HO but this is not yet common in clinical practice (3). This change in the rate of weight gain is rapid and can be detected in a period of weeks to months. HO is associated with increased morbidity and mortality, primarily driven by cardiovascular complications (4). Hyperinsulinemia is common due to obesity and increased vagal tone (5).

HO is typically associated with panhypopituitarism due to the proximity of the structures. Along with HO, hypothalamic damage can lead to hypothalamic syndrome, which includes autonomic dysregulation such as temperature instability and heart rate/blood pressure variability (6). Disturbed sleep is common, particularly in patients with vision loss who have considerable difficulties with diurnal rhythms. The altered sleep patterns and nonspecific fatigue contribute to a decrease in energy expenditure. Decreased physical activity is a hallmark of HO (7). In one study using actigraphy, 97% of patients with HO did not meet physical activity goals of 60 minutes of moderate-vigorous exercise per day (8). Resting energy expenditure may also be reduced (9, 10).

Hyperphagia is not a universal finding in HO. Some patients do report Prader-Willi syndrome–like hyperphagia or food-seeking behaviors, but this is not typical. Caregiver-reported hunger symptoms are variable and but typically higher than control patients with obesity (Fig. 2). Diet recall does not clearly show increased food intake in patients with HO, but patients struggle with accurate diet recall, possibly due to executive function deficits (7, 8). Hyperphagia can be difficult to measure in routine clinical practice as patients and families may struggle to recognize or verbalize subtle changes in eating behaviors. While patients with HO may not clearly increase their caloric intake, they do not decrease their intake to match the decrease in energy expenditure, leading to a positive energy balance. Multiple orexigenic and anorectic neuropeptide pathways are critical to the hypothalamus but are beyond the scope of this clinically focused approach to the patient (11).

Figure 2.

Parent-reported hyperphagia questionnaire (12) scores (minimum score 11 and maximum score 55) from hypothalamic obesity (HO) patients (8) and patients with common obesity enrolled in the Vanderbilt Childhood Obesity Registry (27.5 ± 9.0 vs 21.5 ± 7.0; P = .005 by t test).

Approach to Management

Hypothalamic Obesity Prevention Strategies

HO is now recognized as an important sequela of suprasellar tumors due to mass effect or treatment effect. Craniopharyngiomas are benign tumors though continued growth causes significant morbidity due to their intracranial location. Older studies emphasized oncologic control as the primary outcome, despite excellent 10-year survival rates (13–16). Newer studies focus on functional outcomes, hypothalamic obesity, and neuroendocrine function. Subtotal resection is now a common surgical strategy, particularly in cases of subpial tumor extension or dense adherence (17, 18). Radiation therapy is often used to prevent local recurrence in cases of incomplete resection, but radiation therapy can further damage pituitary and hypothalamic structures. Postoperative surveillance is an option to delay radiation therapy by several years without evidence of worse outcomes (19). As part of the KRANIOPHARYNGEOM 2007 project, there is an ongoing randomized trial to compare quality-of-life outcomes of immediate postoperative radiotherapy vs radiotherapy after progression in pediatric patients with craniopharyngiomas (2). An interim intention-to-treat analysis did not support one strategy over another, and the study is ongoing. The analysis did find that hypothalamus-sparing treatment is important for optimizing quality of life (2).

Lifestyle Management

Patient and families are often overwhelmed at presentation. The onset of HO can go unrecognized by patients, families, and providers among other surgical and radiation side effects, particularly the multiple medications required to treat hypopituitarism. Recognizing HO and providing early education may allow for effective lifestyle interventions. Decreasing caloric intake is a critical component in management of HO. The subset of patients with HO and hyperphagia benefit from strategies to decrease caloric intake used to care for patients with Prader-Willi syndrome, particularly establishing food security. Food security is achieved when food is available on a strict, reliable schedule and there are no opportunities to access food outside this plan (20, 21). Patients without hyperphagia still need physicians and caregivers to recognize the abnormal weight gain pattern to allow for timely caloric restriction (22). The velocity of weight gain is greatest in the first 6 months post diagnosis (23, 24). One proposed algorithm includes assessment by a registered dietitian at the first outpatient visit with follow-up visits for any patients with abnormal weight gain (22).

There is a lack of data on nutritional interventions in HO. HO is a refractory form of obesity and patients randomly assigned to the lifestyle arm of clinical trials have not demonstrated even short-term weight loss (25–28). In our clinical experience, patients have reported improved hunger and weight maintenance with a low-carbohydrate diet. This is supported by animal models; 3 different rat models of HO had weight gain driven by increased carbohydrate intake (29). Low-carbohydrate diets (< 130 g/day or < 26% of daily calories) can decrease glycemia excursions, promote weight loss, and improve triglyceride concentrations when compared with low-fat diets (30). Physical activity should be scheduled with the goal of meeting the Centers for Disease Control and Prevention Physical Activity Guidelines for America (2nd edition) pediatric goal of 60 minutes/day of moderate-vigorous activity. Physical therapy can help patients who are struggling with a change in balance or vision. While physical activity is clearly critical, to our knowledge, there are no targeted exercise trials in humans with HO to provide clear guidance on specific exercise interventions.

Panhypopituitarism

The presence of panhypopituitarism raises the suspicion for hypothalamic damage and HO due to close proximity of structures. It likely serves as a marker of hypothalamic damage as opposed to a direct cause of obesity. For example, patients with diabetes insipidus are more likely to have postsurgical weight gain and greater hypothalamic damage on MRI, particularly in the posterior hypothalamus. Diabetes insipidus does not directly cause HO but serves as an endocrine marker of hypothalamic damage. Undertreatment of central hypothyroidism can contribute to weight gain but 3,5,3′-triiodothyronine (T3) supplementation is controversial. There is a case report of weight loss with the addition of levo-T3 but also a study that failed to show change in brown fat thermogenesis, energy expenditure, or BMI with levo-T3 (31, 32). Per Endocrine Society guidelines, we do not recommend treating central hypothyroidism with levo-T3 (33). The use of high-dose steroids at diagnosis or postoperatively to reduce swelling can cause increased appetite and weight gain, but appropriate physiologic replacement doses should not cause increased weight gain or obesity. Hydrocortisone is typically the preferred steroid for physiologic dosing, because of the lower risk of overtreatment compared to prednisone and dexamethasone (34). The shorter half-life of hydrocortisone allows for dosing that better mimics normal cortisol release (33, 34). While hydrocortisone must be used in children before epiphyseal closure, and is recommended in adults, longer-acting glucocorticoids may be use in some cases (33).

Growth hormone (GH) deficiency is a common sequela of suprasellar tumors and the most common endocrine late effect after cranial radiation. In addition to improvement in adult height, GH treatment of cancer survivors can improve BMI z score and metabolic parameters such as low-density lipoprotein, high-sensitivity C-reactive protein, and alanine transaminase (35, 36). Untreated GH deficiency in adults is associated with increased cardiovascular mortality (37). In the St. Jude Lifetime Cohort study, untreated GH deficiency was associated with decreased lean muscle mass and increased weight-to-height ratio in children (38). Despite the well-accepted benefits of GH, there have been concerns about safety in cancer survivors. Insulin-like growth factor-1 has mitogenic, antiapoptotic, and proangiogenic effects in vitro, indicating a theoretical risk of disease progression or secondary malignancy in GH-treated patients (39). Cranial radiation already increases the risk of secondary malignancies such as glioma or meningioma. Reassuringly, a recent metanalysis by the Endocrine Society found no difference in the occurrence of secondary tumors in GH treated vs untreated cancer survivors and the Growth Hormone Research Society has concluded that the theoretical risk of secondary tumors is not sufficient to recommend against GH therapy (36, 40).

The 2018 Endocrine Society Clinical Practice Guidelines recommend waiting to begin GH treatment until patients reach 1 year disease free following treatment for malignant disease, or after discussion with the oncologist if there is chronic, stable disease. In 2022, the European Society of Endocrinology and Growth Hormone Research Society released a consensus statement on GH replacement in cancer survivors that recommended individualized decisions on GH therapy initiation; as early as 3 months post treatment in children with craniopharyngioma and stable disease (41). We recommend early discussions with the oncology team for children and adults with HO due to craniopharyngioma as the oncologic treatment goal is often stable disease and minimizing further tissue damage as opposed to achieving a disease-free state. While treatment with GH does not cure HO, GH deficiency compounds the harmful body composition and metabolic derangements of HO. In such cases, delaying GH therapy until the patient is disease free may be unreasonable.

Pharmacotherapies

There are no pharmacotherapies approved for the treatment of HO, and response to treatment varies between individuals. Hyperinsulinemia is common in HO and may contribute to the pathogenesis of hyperphagia and obesity, therefore, early studies in HO focused on addressing hyperinsulinemia. In one small study of the somatostatin receptor agonist octreotide, patients with HO demonstrated weight loss that correlated with the change in insulin response (42). In a follow-up, randomized clinical trial of 18 children, weight gain stabilized during 6 months of octreotide therapy (1.6 ± 0.6 kg vs 9.1 ± 1.7 kg, P < .001) (28). The main side effects of octreotide are gastrointestinal. Octreotide's use in clinical practice has been limited by the high cost and limited benefit (43). Diazoxide has similarly been used to suppress insulin secretion. An open-label study of combination diazoxide-metformin therapy in 9 children showed a small improvement in BMI (−0.3 ± 2.3 vs 2.2 ± 1.5; P = .02) that correlated with the pretreatment degree of hyperinsulinemia (44). The treatment effects were modest and 2 of 9 patients discontinued treatment because of side effects (peripheral edema and elevated hepatic enzyme levels). Weight loss was not seen in a randomized clinical trial of diazoxide in children with HO (45). Perhaps due to the lack of metformin use, hyperglycemia was a significant adverse effect, with 3 of 18 participants in the treatment group developing type 2 diabetes during the 2-month intervention. Metformin plus lifestyle counseling may have a similar ability to stabilize weight gain with less-deleterious side effects (46).

Several glucagon-like peptide-1 receptor agonists (GLP1RAs) are Food and Drug Administration approved for the treatment of general obesity in adolescents and adults. GLP1RA have central and peripheral effects on satiety through activation of hindbrain and hypothalamic pathways and slowing of gastric emptying (47). The extrahypothalamic satiety effects of GLP1RA and their potential role in activating brown fat make this medication class a possible treatment for HO. We conducted a 52-week, randomized clinical trial of the GLP1RA once-weekly exenatide in children and young adults with HO (25). There was no significant change in BMI though fat mass did not increase as rapidly in the treatment group (estimated treatment difference −3.1 kg; 95% CI, −5.5 to −1.6 kg; P = .004). Treatment with GLP1RA significantly decreased ad libitum food intake, as expected (8). Interestingly, the GLP1RA-treated group had a decrease in total energy expenditure, despite an overall increase in lean mass and fat mass during the study period (8). These finding highlights how difficult HO is to treat. Higher, weight-loss dosing of liraglutide, newer GLP1RAs such as semaglutide, and the dual GLP1RA and glucose-dependent insulinotropic polypeptide (GIP) agonist tirzepatide all have better weight loss profiles than once-weekly exenatide (48–50). It is possible that these medications could provide additional weight loss benefits in HO, and further studies are needed.

Many patients with HO also have executive function deficits, inattention, and hypersomnolence (51, 52). Stimulant medications are indicated for these symptoms and can have the beneficial side effect of appetite suppression. In 2 case series, patients had stable or decreasing weight with dextroamphetamine (53, 54). There is supporting evidence in the animal model; treatment with methylphenidate decreased body weight in a rat model of HO (55). One retrospective case series of 12 patients showed persistent BMI improvements with standard dosing of methylphenidate (56). We commonly recommend stimulant medications in our practice for any patients who meet attention deficit disorder diagnostic criteria.

A combination of phentermine and topiramate is approved for treatment of general obesity in patients aged 12 years and older (57). It has not been trialed in patients with HO. Phentermine is a sympathomimetic agent that can suppress appetite and increase metabolism but the mechanism of action of topiramate on body weight is not well understood (57). It is not clear if an intact hypothalamus is required for efficacy. Other sympathomimetic agents, caffeine and ephedrine, have shown efficacy in case reports (58). Tesofensine is a monoamine reuptake inhibitor with potential for drug repurposing as a weight loss medication due to a side effect of appetite suppression and weight loss, complicated by sympathetic side effects such as increased blood pressure and increased heart rate. An investigational drug that combines tesofensine and metoprolol was tested in a small clinical trial of hypothalamic obesity and showed −6.3% mean weight change in the drug group vs placebo without unwanted cardiovascular effects (26).

Setmelanotide is a melanocortin receptor agonist (59). It is currently Food and Drug Administration approved for treatment of obesity due to POMC/PCSK1/LEPR deficiency and Bardet-Biedl syndrome. The hypothalamus controls body weight via the melanocortin-4 receptor (MC4R) pathway, which regulates both energy intake and energy expenditure (60). Setmelanotide is a proposed treatment for HO as it has potential to stimulate any remaining hypothalamic MC4R neurons as well as MC4R neurons in other areas of the brain and spinal cord (61). A small, phase-2, open-label clinical trial of setmelanotide in adults and children with HO was completed in 2022 (62). Eleven patients participated and at 16 weeks all patients had 5% or greater BMI reduction and 82% had 10% or greater BMI reduction. The mean BMI change was −17.2%. Further studies are needed to determine if setmelanotide is a durable and effective treatment for HO.

Surgical Therapies

Bariatric surgery is currently the most effective and durable therapy for obesity with a mean reduction in body weight of 25% after 3 years (63). Guidelines recommend considering bariatric surgery for adults with a BMI greater than or equal to 40, greater than or equal to 35 with obesity-related comorbidities, or greater than or equal to 30 with uncontrolled type 2 diabetes (64). In adolescents, the equivalent thresholds are a BMI greater than or equal to 140% of the 95th percentile or BMI greater than or equal to 120% of the 95th percentile with an obesity-related comorbidity (65). Many patients with HO will meet these criteria and bariatric surgery may be considered. Bariatric surgery has potential to target the hyperphagia and hyperglycemia common in HO. There are concerns that patients with HO may be at higher risk of complications from surgery due to adrenal insufficiency and diabetes insipidus. While not everyone with hypopituitarism will have excess adverse effects from surgery, vomiting, impaired medication absorption, and adrenal crisis have been reported postbariatric surgery in patients with HO (66). Minor changes in hormone replacement therapy doses should be expected (67). Patient selection and an experienced surgical and endocrinology team can mitigate the risks posed by panhypopituitarism, presuming that adequate benefit is expected.

There are 3 main types of bariatric surgery: Roux-en-Y gastric bypass (RYGB), vertical sleeve gastrectomy (VSG), and laparoscopic adjustable gastric banding (LAGB). LAGB relies primarily on changes in gastric volume and transit and is least effective for weight loss (68). The mechanisms of weight loss from RYGB and VSG are not fully understood but can include changes in nutrient absorption, gastric transit, gut microbiome, gut hormones, and central nervous system appetite and reward signaling pathways (68). With the exception of hypothalamic pathways, the other potential changes should be as effective in patients with HO as patients with common obesity. LAGB does not appear to cause effective weight loss in HO (69, 70). Durable weight loss has been seen with RYGB or VSG in HO in several case reports with up to 48 months of follow up (66). There has been one retrospective case-control study comparing weight loss in 8 patients with HO with 75 control patients following RYGB or VSG (67). Weight loss at 2 years’ follow-up was similar for cases and controls undergoing RYGB. In contrast, following VSG, patients with HO had less weight loss (10% vs 20%) than controls.

Back to the Case

The patient was identified as having HO as the etiology of his rapid post–tumor resection weight gain. The family met with a registered dietitian to develop a calorie restricted, modified-Atkins diet plan with a goal of weight stabilization. This plan was shared with the boy’s school, and his individualized education plan specified not to use food as a reward. His rate of weight gain slowed but he did not have any weight loss. When he turned 11, we added oral dextroamphetamine 7.5 mg 3 times daily for attention deficit disorder and appetite suppression. This led to a 4-kg weight loss followed by weight gain tracking along the 95th percentile (see Fig. 1). His glycated hemoglobin A_1c increased from 5.3% to 5.9% at around age 15 years. Metformin and a GLP1RA were added, and the patient's glycated hemoglobin A_1c normalized, and he lost 3 kg. We also began testosterone therapy for hypogonadotropic hypogonadism, and the family added 1 hour of walking daily. At age 18 years his weight had been stable for 2 years and his BMI had improved to 30, down from his maximum of 39 at age 14 years.

Conclusion

Therapeutic options for HO are still limited but early recognition of HO and a patient-centered approach may greatly affect outcomes. While untreated GH deficiency or overtreatment of adrenal insufficiency can contribute to increasing BMI, panhypopituitarism is not the underlying etiology of HO. Damage to the hypothalamus results in decreased energy expenditure without concomitant decrease in energy intake. Hyperphagia is not always present. Currently available pharmacotherapies for weight gain tend to be less efficacious in HO but can still be useful. Several novel drugs are under investigation. A substantial amount of weight gain can be seen in the first year, therefore early recognition and treatment are critical. We recommend a multidisciplinary approach to patients at high risk for HO to quickly identify abnormal weight gain and provide support for environmental, dietary, and medical interventions.

Financial Support

J.T. is supported by the National Institutes of Health (No. F32DK128970).

Disclosures

A.H.S. has served as a consultant for Rhythm Pharmaceuticals, Inc and Saniona, and has received research funding from Rhythm Pharmaceuticals, Inc. J.T. has nothing to disclose.

Abbreviations

BMI

body mass index

GH

growth hormone

GIP

glucose-dependent insulinotropic polypeptide

GLP1RA

glucagon-like peptide-1 receptor agonist

HO

hypothalamic obesity

LAGB

laparoscopic adjustable gastric banding

MC4R

melanocortin-4 receptor

MRI

magnetic resonance imaging

RYGB

Roux-en-Y gastric bypass

T3

3,5,3′-triiodothyronine

VSG

vertical sleeve gastrectomy

Data Availability

Some or all data sets generated during and/or analyzed during the present study are not publicly available but are available from the corresponding author on reasonable request.