Failure to thrive (FTT) is a term traditionally used to describe children with growth delay and developmental delay. It is not a specific disease process, but rather a cluster of symptoms involved in an interplay of metabolism, hormones, genetics, and psychosocial factors. The most frequent cause of FTT is not organic in nature, but mainly because of poor caloric intake. This may be secondary to poor appetite, lack of available food, or psychosocial problems. Organic causes for FTT such as endocrine or genetic disorders are rarely identified. In young children, prolonged FTT may impair normal development and future cognitive abilities (1).
The prevalence of obesity in children is increasing worldwide. The American Obesity Association estimates that 30% of children between the ages of 6 and 19 are overweight, and 15% meet the criteria for obesity (2). One of the main factors in obesity is an impaired balance between energy intake and expenditure. Recent investigations have focused on 2 gastric hormones, ghrelin and obestatin, and their role in the regulation of food intake in the children with obesity. The possible role of ghrelin in the appetite regulation of young children with FTT has been investigated (3).
Ghrelin is a 28-amino-acid peptide with an n-octanoylated serine residue, which is the posttranslation product of the peptide preproghrelin (4). The octanoylated structure is essential for its function and transport across the blood–brain barrier. Ghrelin is mainly synthesized in the gastric body and fundus; however, it is also found in multiple other locations, including the hypothalamus, hippocampus, cerebral cortex, pituitary and adrenal glands, small intestine, and pancreas. The receptor identified for ghrelin, growth hormone secretory receptor type 1a, is widely expressed throughout the body and plays an important role in a range of biological functions (5). As an orexigenic hormone, ghrelin has both short-term and long-term effects on energy balance, appetite, and weight gain (6). Endogenous ghrelin levels increase in the fasting state and decrease postprandially, suggesting a role in meal initiation (7). In the long-term, it decreases fat utilization and increases fat deposition and food intake (8). Other physiologic effects of ghrelin include increasing gastric motility (9), stimulation of gastric secretions (10), inhibition of pancreatic secretions (11), influence on glucose metabolism (12), induction of cellular differentiation, and inhibition of the apoptosis in adipose tissue (13).
Obestatin is also a 28-amino-acid peptide derived from the same precursor as ghrelin and is also mainly found in the stomach (14); however, obestatin is an anorexigenic hormone, which decreases food intake and inhibits gastrointestinal motility (5). Obestatin levels are not affected by a fixed-energy meal; therefore, it may have a role mainly in long-term regulation of food intake (15).
Based on the studies described above and the influence of both hormones on appetite, we hypothesized that children with FTT would have decreased ghrelin levels and/or increased obestatin levels and that the children with obesity would have higher ghrelin and lower obestatin as compared with that in age-matched controls. Thus far, the majority of research in this area has focused on the role of these hormones in obesity in adults. Our goals were to describe the serum levels of these 2 hormones in children with FTT and obesity and to compare them to those of age-matched controls.
Children with FTT, obesity, and age-matched normal weight controls who were scheduled for medically indicated procedures were enrolled in the study during a 2-year period.
Group-Specific Indications for Procedures
Esophagogastroduodenoscopy (EGD) is part of the evaluation of children with FTT to detect diseases such as gastroesophageal reflux disease, allergic gastroenteropathy, celiac disease, and pancreatic insufficiency. Children with obesity have a higher prevalence of digestive symptoms (abdominal pain, gastroesophageal reflux) and often require an EGD. The age-matched controls had normal weight and body mass index (BMI) and underwent various surgical procedures, including elective otolaryngological surgeries (mainly myringotomy tube placement) or EGD for chronic abdominal pain and reflux symptoms.
The FTT group included children <3.0 years old who had a clinical diagnosis of FTT without evident organic disease. FTT was defined as weight <5th percentile with normal esophageal, gastric, and duodenal histology. The obesity group consisted of children who had a BMI >95% on the Centers for Disease Control and Prevention 2000 growth chart and had normal esophageal, gastric, and duodenal histology. The age-matched control group was recruited from patients with a normal weight and BMI for age and with no confirmed gastrointestinal pathology.
The children with known medical conditions that influence appetite or growth, children with insufficient samples, and those whose age did not match the study guidelines were excluded. In the FTT group, 7 children were excluded: 3 were >3 years of age, 2 had insufficient sample size, 1 had pancreatic insufficiency, and 1 had a genetic disorder. In the obese group, 4 children were excluded: 2 had insufficient samples, 1 had acute hepatitis, and 1 was <2 years of age (outlier). In the control groups, 5 children were removed to match the ages of the FTT and obese groups, and 1 had an insufficient sample.
The blood samples were collected after the children fasted as per the following anesthesia guidelines. Children >1 year of age were nil per os after midnight. Children <1 year of age were allowed to have formula 6 hours before the procedure and pedialyte up to 2 hours before the procedure. All of the samples were collected between 8:00 AM and 11:00 AM to avoid variations owing to the diurnal changes in the hormone levels.
Three to 5 milliliters of blood were collected into BD Vacutainer (BD Biosciences, Franklin Lakes, NJ) ethylenediaminetetraacetic acid-plasma tubes and processed in the gastrointestinal laboratory. The tubes were centrifuged at 1600g for 15 minutes at 4°C. The plasma was separated and divided into multiple aliquots that were stored at −80°C until the assays for total ghrelin and obestatin were performed.
Plasma total ghrelin and obestatin were measured using commercially available kits and following the manufacturer's instructions. All of the specimens were measured in duplicates.
Total ghrelin was measured by a sandwich ELISA (Millipore, Billerica, MA). The ghrelin in the sample was immobilized to the wells of a microtiter plate coated with a pretitered amount of anchor antibodies. Simultaneously, the captured ghrelin was bound to a second biotinylated ghrelin antibody, which was recognized by the biotin antibodies conjugated to horseradish peroxidase (HRP). Quantification of total human ghrelin was determined by measuring HRP activities in the presence of the TMB substrate (3,3′,5,5′-tetramethylbenzidine). Total ghrelin levels (picograms per mole) were calculated from the ghrelin standards provided with the kit, which were run at the same time as the samples. Based on the information provided by the manufacturer, the detection range of this assay was 100 to 5000 pg/mL, the intraassay variation was 0.9% to 1.9%, and the intraassay variation was 5.2% to 7.8%.
Obestatin was measured by competitive enzyme immunoassay using the combination of a highly specific antibody to human obestatin and a biotin-avidin affinity system (ALPCO Diagnostics, Salem, NH). The 96-well plate was coated with goat antirabbit immunoglobulin G, to which biotinylated human obestatin, human obestatin standards, samples, and rabbit antihuman obestatin antibody were added for a competitive immunoreaction. After incubation and plate washing, HRP labeled with streptavidin was added so that HRP-labeled, streptavidin-biotinylated human obestatin-antibody complexes formed on the surface of the wells. Finally, HRP enzyme activity was determined by 3,3′,5,5′-tetramethylbenzidine as substrate, and the concentration of human obestatin was calculated based on the obestatin standards. Based on the information provided by the manufacturer, the sensitivity of the assay ranged from 0.412 to 100 pg/mL, the intraassay variation was 3.5% to 9.9%, and the interassay variation was 5.6% to 9.0%.
The Nemours/Alfred I. duPont Hospital for Children institutional review board approved the study protocol. Written informed consent was obtained from all of the parents or guardians, and an assent form was signed by the children 7 to 17 years of age, before the blood draw. The blood specimens of the enrolled patients were labeled with a study number to protect their identity and blind the research personnel assaying the samples.
Descriptive statistics were used to compare variables between the groups. A χ2 test was used to perform statistical comparisons of all of the categorical variables. A t test was used to compare groups for quantitative variables. Pearson correlation was used to analyze association between 2 variables. A P value <0.05 was considered a statistically significant difference. Logistic regression analysis was performed with SPSS statistical software system (version 7.0, SPSS Inc, Chicago, IL).
Study Patients’ Characteristics
Initially, 85 children were screened for the study based on the inclusion criteria; of those, 80 (20 with FTT, 21 with obesity, and 39 controls) were successfully enrolled and 5 declined to participate (Table 1). After further review of the medical records and later sample processing, we removed 17 children from the study based on the specific exclusion criteria (Table 1). The exclusion details were described in Methods section.
A total of 63 children were included in the study (Table 1): 13 with FTT, 17 with obesity, and 33 controls, of whom 14 were age matched with the FTT group (control group A) and 19 with the obesity group (control group B). The characteristics of the participants are described in Table 2. There were no statistical differences in age or sex between the study and control groups. The average age of the FTT group was 1.7 ± 0.5 years (range, 9–31.2 months). Control group A had an average age of 1.9 ± 0.7 years (range, 6–33.6 months). The average age of the obese group was 12.8 ± 2.8 years (range, 7.3–18.1 years), and the age-matched control group B had an average age of 12.4 ± 3.4 years (range, 7–17.9 years). As expected, the weight-for-age and BMI-for-age (for >2.0 years) or weight/height-for-age (for <2.0 years) percentiles were significantly different between the 2 study groups and controls (Table 2).
Because of the heterogeneous causes of FTT, we expanded growth parameters of the 13 children with FTT from birth to the postenrollment period when data were available to assess their growth status. We focused on tracking changes in weight, height, and head circumference percentiles. At birth, 2 children were <5th percentile for weight, 3 were <5th percentile for height, and 2 were <5th percentile for head circumference. Before the enrollment date, ranging from ages 1 to 8 months, 9 children were <5th percentile for weight, 2 were <5th percentile for height, and 1 was <5th percentile for head circumference (Table 3). At the time of the enrollment, all of the 13 children were <5th percentile for weight. The 2 children <5th percentile for height, of whom 1 also was <5th percentile for head circumference before the study, still had these parameters postenrollment. After the study period (1 month–present), most of the children (62%) no longer had FTT, whereas 5 (38%) were still <5th percentile for weight (Table 3). The average FTT duration was 9 ± 6.2 months, excluding 1 patient who had FTT from birth to at present (Table 3). The results hinted that a majority of the children with FTT had normal growth parameters at birth and were able to gain weight later.
Differences Between FTT and Control Groups
The total ghrelin levels were slightly lower in the FTT group than in the age-matched control group, but this did not reach statistical significance (Table 4). There was no significant difference in the total obestatin levels or ghrelin-to-obestatin ratio between the FTT and control groups (Table 4).
Differences Between Obese and Control Groups
The total fasting ghrelin levels were significantly lower (P = 0.0003), whereas the fasting obestatin levels were significantly higher (P = 0.029) in the children with obesity than those in the age-matched controls (Table 4). As the result of these differences, the ghrelin-to-obestatin ratio was significantly lower in the patients with obesity than in the controls (Table 4).
Age-Related Changes in Ghrelin and Obestatin Levels
Initially, all controls (n = 33) were analyzed as 1 group. The total fasting ghrelin levels were significantly higher in children <3 years of age compared with those >3 years (Fig. 1A; P = 0.0004). In contrast, obestatin levels were not significantly different between the 2 age groups (Fig. 1A). Further analysis showed that the relation between age and total ghrelin in the control children exhibited a negative correlation before age 3 years (Fig. 1B; P = 0.048; r = −0.537). This correlation was not significant after age 3 years in the controls (Fig. 1B; P = 0.149).
In the FTT group, no significant correlation was found between age and ghrelin (Fig. 1B; P = 0.176). In the obese group, the total ghrelin levels were negatively and significantly correlated with age (Fig. 1B; P = 0.0001; r = −0.8435). In comparison, the correlation with age was not significant in the age-matched control group (Fig. 1B). Obestatin levels had no significant correlation with age in both FTT and obese groups (data not shown).
It is well known that ghrelin levels increase in the fasting state (7). We analyzed and compared the fasting time between the 2 age control groups to determine whether the age-associated difference in ghrelin was influenced by the fasting time. As shown in Table 5, on average, the solid food fasting time in the younger group was 2 hours shorter than that in the older group (P = 0.057). The influence of fasting time on ghrelin levels was further analyzed by using logistic regression analysis in which the effect of fasting time was controlled between the 2 groups. The results revealed that the total ghrelin in the younger group was still significantly different from that of the older group (Table 5; P = 0.02). Regression analysis did not reveal statistical significance for fasting time between the 2 groups (P = 0.236). Fasting time for liquid was also analyzed; no difference was exhibited between the 2 age groups (data not shown). Furthermore, fasting time was indistinguishable between FTT and younger controls or between obese and the older controls (data not shown).
Correlation of Ghrelin and Obestatin With Anthropometric Parameters
There was no significant association between the ghrelin levels and weight-for-age percentiles or the BMI (weight-for-length for <2 years) in the control or study groups (Table 6). Obestatin levels were negatively correlated with weight-for-age percentile only in the young control group (<3 years; P = 0.019; Table 6), but not in the older control group (>3 years; Table 6). No significant correlation was found between growth parameters and obestatin levels in the children with FTT (Table 6). In the children with obesity, the obestatin levels were positively correlated with weight-for-age percentiles, but the correlation with BMI values was less significant (Table 6; P = 0.171).
In our study, we did not observe a significant difference in ghrelin levels in younger children with FTT compared with age-matched controls. Contrary to our findings, Tannenbaum et al (3) found that the acetylated and total ghrelin levels in 9 infants (age range, 9–18 months) with FTT were higher than those in 5 controls. The authors hypothesized that the hypothalamic ghrelin receptors in infants with FTT became desensitized to ghrelin and that the elevated plasma levels may represent an adaptive mechanism; however, their study had mainly young infants, a small sample size, and shorter duration of fasting before the blood draw. These differences may contribute to the contrary findings. Alternatively, the lack of differences between FTT and age-matched controls may be secondary to the heterogeneous nature of the FTT group. We excluded children whose FTT had known factors based on the medical records. Our data also show that the majority of children with FTT did not have severe growth abnormalities except for weight-for-age <5th percentile; however, there may have been unidentified medical conditions that affected their FTT and our results. For instance, the 2 patients who had both weight and height continuously <5th percentile may indicate some organic, yet unidentified diagnosis. A study with a larger population may correct for some of these confounding factors to reveal the difference in ghrelin levels between the FTT and controls.
In the obese group (average age 12.8 years), the fasting ghrelin level was significantly lower than that of age-matched controls. This is consistent with previous studies reporting decreased fasting ghrelin levels that returned to normal levels after weight loss in children with obesity and overweight children (16,17). In healthy volunteers without obesity who were overfed with high-fat dietary supplements, the increased caloric intake suppressed the plasma ghrelin level by 18% despite only a 3% increase in body weight (18). Bacha and Arslanian (19) suggested that decreased suppression of ghrelin after glucose challenge in overweight versus normal-weight children may be another manifestation of insulin resistance in obesity. These results suggest that ghrelin levels in obesity may reflect a possible physiological adaptation to positive energy balance. In older children with anorexia nervosa, ghrelin levels were significantly higher than those of the age-matched controls (15,20,21), and the levels decreased when weight gain occurred (22). Ghrelin levels are also high in patients who have cancer-induced cachexia (23).
Our study showed that fasting total ghrelin level is age dependent in healthy controls and that it negatively correlates with age in children <3 years of age. Our data indicated that although ghrelin was expected to fluctuate with fasting status, the difference was not significant enough to affect the ghrelin levels in the 2 age groups. Total ghrelin was still significantly higher in the younger controls compared with the older group when the variance from fasting time was controlled. This is supported by other studies that also found elevated levels of plasma ghrelin in young children (24,25). It is possible that total ghrelin levels are upregulated during periods of rapid growth in childhood when the caloric intake per kilogram body weight is higher. In addition, we found that ghrelin levels had a negative correlation with age in children with obesity, whereas such correlation was weaker in the age-matched healthy controls. Previous studies reported a negative correlation between age and ghrelin in children with Prader-Willi, an obesity syndrome associated with genetic disorder and abnormal metabolism (25,26). The observation of declining ghrelin levels with age gives further support to a possible feedback regulation of ghrelin in the children with obesity.
Obestatin was discovered in 2005, and most of the studies have been reported in patients with obesity. Contrary to the appetite-stimulating effects of ghrelin, obestatin suppresses food intake and decreases body-weight gain (14). Our results showed that obestatin levels were significantly higher and ghrelin levels were lower in the children with obesity than those in normal controls. A study in children with obesity with similarly aged children as in our study also reported increased obestatin and decreased ghrelin levels as compared with controls (27). Similar results were also reported in the women with obesity displaying increased obestatin and decreased ghrelin and ghrelin/obestatin ratio compared with the nonobese controls (28). Elevated levels of obestatin may indicate a feedback response to obesity in an attempt to curtail appetite and food intake. There are no past studies on the obestatin levels in children with FTT. Our data indicate that the plasma obestatin level in the FTT group was not significantly different from that of the controls.
The relation between obestatin and anthropometric parameters has been examined in only a few studies. We observed a positive correlation between obestatin and weight-for-age percentiles in the obese group. Reinehr et al (27) also demonstrated positive correlation between obestatin and BMI in children. The positive correlation can be explained by a suppression role of obestatin in regulating food intake in response to increasing body weight, which may indicate a feedback regulation. We, however, observed a negative correlation between obestatin and weight-for-age percentiles in the age-matched control. Obestatin levels have been reported to be negatively correlated with BMI in adults with normal weight and those with anorexia (29–31), which concurs with our results in the control group. Clearly, large-scale studies are needed to understand the relation of obestatin with nutrition status and energy balance in both normal, healthy children and those with energy imbalance.
There were several limitations of our study. The sample size was relatively small and did not allow the assessment of potential confounders. We also evaluated the total ghrelin level, although only acetylated ghrelin is bioactive; however, previous studies show that total ghrelin is a reasonable proxy for the acetylated form because the ratio of the 2 forms has been reported to remain constant in animals and humans (32–34). We measured morning hormone levels, but multiple measurements covering the diurnal variation in serum levels may generate additional information.
In summary, we did not observe significant differences in the ghrelin and obestatin levels, or in the ghrelin/obestatin ratio in children with FTT as compared with the healthy controls; however, these hormone levels were significantly different between the patients with obesity and healthy controls. Ghrelin level was strongly associated with age, whereas obestatin was more closely correlated with anthropometric parameters. Overall, the results from the obese group may suggest a possible adaptive process.
Additional studies with a larger sample size at different age groups with multiple daily serum level measurements and parallel assessment of hormone expression in gastric tissue are necessary to further elucidate the role of ghrelin and obestatin in the regulation of appetite and weight control in children.
The authors are thankful to Laura Rodowski, RN, for assistance in the recruitment of patients.
1. Black MM, Dubowitz H, Krishnakumar A, et al. Early intervention and recovery among children with failure to thrive: follow-up at age 8. Pediatrics
2. Devi S. Progress on childhood obesity patchy in the USA. Lancet
3. Tannenbaum GS, Ramsay M, Martel C, et al. Elevated circulating acylated and total ghrelin concentrations along with reduced appetite scores in infants with failure to thrive. Pediatr Res
4. Kojima M, Hosoda H, Date Y, et al. Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature
5. Arslan N, Erdur B, Aydin A. Hormones and cytokines in childhood obesity. Indian Pediatr
6. Wren AM, Seal LJ, Cohen MA, et al. Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab
7. Cummings DE, Purnell JQ, Frayo RS, et al. A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes
8. Kos K, Harte AL, O’Hare PJ, et al. Ghrelin and the differential regulation of des-acyl (DSG) and oct-anoyl ghrelin (OTG) in human adipose tissue (AT). Clin Endocrinol (Oxf)
9. Ejskjaer N, Vestergaard ET, Hellström PM, et al. Ghrelin receptor agonist (TZP-101) accelerates gastric emptying in adults with diabetes and symptomatic gastroparesis. Aliment Pharmacol Ther
10. Bilgin HM, Tumer C, Diken H, et al. Role of ghrelin in the regulation of gastric acid secretion involving nitrergic mechanisms in rats. Physiol Res
11. Kapica M, Laubitz D, Puzio I, et al. The ghrelin pentapeptide inhibits the secretion of pancreatic juice in rats. J Physiol Pharmacol
12. Tong J, Prigeon RL, Davis HW, et al. Ghrelin suppresses glucose-stimulated insulin secretion and deteriorates glucose tolerance in healthy humans. Diabetes
13. Kim MS, Yoon CY, Jang PG, et al. The mitogenic and antiapoptotic actions of ghrelin in 3T3-L1 adipocytes. Mol Endocrinol
14. Zhang JV, Ren PG, Avsian-Kretchmer O, et al. Obestatin, a peptide encoded by the ghrelin gene, opposes ghrelin's effects on food intake. Science
15. Huda MS, Durham BH, Wong SP, et al. Plasma obestatin levels are lower in obese and post-gastrectomy subjects, but do not change in response to a meal. Int J Obes (Lond)
16. Zou CC, Liang L, Wang CL, et al. The change in ghrelin and obestatin levels in obese children after weight reduction. Acta Paediatr
17. Zou CC, Liang L, Zhao ZY. Factors associated with fasting plasma ghrelin levels in children and adolescents. World J Gastroenterol
18. Robertson MD, Henderson RA, Vist GE, et al. Plasma ghrelin response following a period of acute overfeeding in normal weight men. Int J Obes Relat Metab Disord
19. Bacha F, Arslanian SA. Ghrelin suppression in overweight children: a manifestation of insulin resistance? J Clin Endocrinol Metab
20. Katergari SA, Milousis A, Pagonopoulou O, et al. Ghrelin in pathological conditions. Endocr J
21. Soriano-Guillén L, Barrios V, Campos-Barros A, et al. Ghrelin levels in obesity and anorexia nervosa: effect of weight reduction or recuperation. J Pediatr
22. Otto B, Cuntz U, Fruehauf E, et al. Weight gain decreases elevated plasma ghrelin concentrations of patients with anorexia nervosa. Eur J Endocrinol
23. Garcia JM, Garcia-Touza M, Hijazi RA, et al. Active ghrelin levels and active to total ghrelin ratio in cancer-induced cachexia. J Clin Endocrinol Metab
24. Soriano-Guillen L, Barrios V, Chowen JA, et al. Ghrelin levels from fetal life through early adulthood: relationship with endocrine and metabolic and anthropometric measures. J Pediatr
25. Haqq AM, Farooqi IS, O’Rahilly S, et al. Serum ghrelin levels are inversely correlated with body mass index, age, and insulin concentrations in normal children and are markedly increased in Prader-Willi syndrome. J Clin Endocrinol Metab
26. Butler MG, Bittel DC. Plasma obestatin and ghrelin levels in subjects with Prader-Willi syndrome. Am J Med Genet A
27. Reinehr T, de Sousa G, Roth CL. Obestatin and ghrelin levels in obese children and adolescents before and after reduction of overweight. Clin Endocrinol (Oxf)
28. Vicennati V, Genghini S, De Iasio R, et al. Circulating obestatin levels and the ghrelin/obestatin ratio in obese women. Eur J Endocrinol
29. Gao XY, Kuang HY, Liu XM, et al. Plasma obestatin levels in men with chronic atrophic gastritis. Peptides
30. Zamrazilová H, Hainer V, Sedlácková D, et al. Plasma obestatin levels in normal weight, obese and anorectic women. Physiol Res
2008; 57 (suppl 1):S49–S55.
31. Nakahara T, Harada T, Yasuhara D, et al. Plasma obestatin concentrations are negatively correlated with body mass index, insulin resistance index, and plasma leptin concentrations in obesity and anorexia nervosa. Biol Psychiatry
32. Goldstone AP, Patterson M, Kalingag N, et al. Fasting and postprandial hyperghrelinemia in Prader-Willi syndrome is partially explained by hypoinsulinemia, and is not due to peptide YY3-36 deficiency or seen in hypothalamic obesity due to craniopharyngioma. J Clin Endocrinol Metab
33. Ariyasu H, Takaya K, Hosoda H, et al. Delayed short-term secretory regulation of ghrelin in obese animals: evidenced by a specific RIA for the active form of ghrelin. Endocrinology
34. Marzullo P, Verti B, Savia G, et al. The relationship between active ghrelin levels and human obesity involves alterations in resting energy expenditure. J Clin Endocrinol Metab