Craniopharyngioma patients were admitted overnight to the Children’s Hospital of Philadelphia. After a standardized dinner and 12-h overnight fast height, weight, and skinfold measurements were obtained. REE was then determined by open-circuit indirect calorimetry (SensorMedics 2900; SensorMedics Corp, Yorba Linda, CA). To ensure that they were truly at rest, subjects were wheeled to the calorimeter and remained supine for the entire hour while oxygen and carbon dioxide measurements were taken. Because it was an assessment of resting energy expenditure, measurements collected during nonresting conditions (i.e., when the subject moved) were omitted according to a standard protocol. Body composition was determined by dual energy X-ray absorptiometry (DEXA) using a Hologic QDR-2000 whole-body scanner (Hologic, Wortham, MA) in pencil beam mode. Brain MRI results were reviewed to determine the location of the craniopharyngioma and the presence of hypothalamic involvement. Control subjects were obese children, aged 7 or older, who were previously studied [5] at our center using the same protocols for DEXA to assess body composition and indirect calorimetry for REE, except that control subjects were not admitted overnight prior to their studies.

Obese children with a history of craniopharyngioma had greater percentage fat-free mass (FFM) compared to obese controls (mean ± standard error, SE 57.2%±0.088% vs. 50.0%±0.87%) as shown in the Table 1. There was no significant difference observed between obese craniopharyngioma patients and obese controls with respect to age, height, weight, body mass index, fat mass, or fat mass percentage.

In obese craniopharyngioma patients, measured REE was significantly lower than WHO-predicted REE (mean ± SE 1,541±112.6 vs 1,809±151.8 kcal; p=0.01; Fig. 1). This represented a ΔREE of 268 kcal or a daily kcal deficit of 17% (Fig. 2). In obese control subjects, there was no significant difference between measured and predicted REE (1,647±33.2 vs 1,652±40.2 kcal, p=0.8; Fig. 1).

Regulation of resting energy expenditure depends on many factors, such as fat-secreted and pituitary hormones. For example, leptin and adiponectin can act centrally to increase energy expenditure [7]; this action can be dampened by adipose inflammation that can result from obesity itself [8], high-fat diet [9], or hormone deficiencies [10–12]. Sympathetic tone may be decreased in craniopharyngioma patients which could be a factor leading to decreased REE [13]. Multiple signaling pathways involving the hypothalamus, including leptin [14] and adiponectin [15] from adipocytes, ghrelin from the gastrointestinal tract [16], and the endocannabinoid pathway [17] are implicated in appetite control and metabolic rate. Notably, all obese craniopharyngioma subjects (versus none of the non-obese) had suprasellar tumor involvement, suggesting hypothalamic dysfunction as a factor in the observed differences in REE.

Several medications used by our subjects can affect energy balance. Three subjects were taking sibutramine which promotes weight loss by increasing satiety and energy expenditure [18]. One of these three was also taking methylphenidate for attention-deficit disorder. Methylphenidate is reported to increase REE and suppress food intake [19, 20]. In spite of this, our patients still had a lower than expected REE indicating that these medicines were unlikely to be confounders in our analysis.

Under-treatment of endocrine deficiencies could also contribute to weight gain. Growth hormone can increase resting energy expenditure [21] and modulates thyroid hormone metabolism by increasing activation of thyroxine to tri-iodothyronine [22]. It increases lean mass and reduces body fat [23]. Hypothyroidism is associated with increased fat mass and low resting energy expenditure which improve with thyroid hormone replacement [24]. Conversely, thyroid hormone in excess is catabolic to muscle. Thus, an untreated deficiency of growth hormone or thyroid hormone could contribute to weight gain and body fat increase. Cortisol increases resting energy expenditure; however, it also increases appetite [25] and, therefore, tends to favor a positive energy balance. Overall, untreated deficiencies of these hormones are unlikely to entirely explain our main observation given the close monitoring of hormone replacement these subjects received.

Surprisingly, though they had a deficit in REE, the craniopharyngioma patients were in fact not lower in FFM. This combination suggests that the fat-free mass in patients with craniopharyngioma had a lower intrinsic metabolic activity. Other mechanisms could play a role in the observed alteration in REE. Derangements in the hypoathalamic endocannabinoid pathway, which controls appetite and food intake by central and peripheral effects, may affect skeletal muscle energy balance. Of particular therapeutic interest, overactivation of the endocannabinoid system has been associated with obesity, making antagonists of the hypothalamic receptor (CB1) novel candidates for treatment [26]. Uncoupling proteins, important in thermogenesis and resting metabolism [27], are found in human adipocytes and skeletal muscle and found to be expressed in the hypothalamus and pituitary of primates [28]. Theoretically, damage to these structures may lead to decreased levels of the proteins and thus reduced metabolic activity.

Our control subjects were not hospitalized the night before their indirect calorimetry study. However, their resting energy expenditure as determined by indirect calorimetry was, on average, very close to that predicted by the WHO equation (Figs. 1 and ​and2).2). Therefore, they appear to be valid comparison subjects for the purposes of testing the applicability of the WHO prediction model to measured energy expenditure in craniopharyngioma patients.

Given our small sample size, these results must be validated by larger-scale studies. It is possible that differences in body composition that did not reach statistical significance in our subjects may do so when larger numbers are evaluated. Similarly, a larger study might also reveal unexpected associations between overweight and endocrine deficiency and medication usage. However, our preliminary results outline an interesting area for future study and clinical intervention.