3.1. Overweight and Obesity in Newly Diagnosed CD Patients: The “Compensatory” Hypothesis
Semeraro first hypothesized that the atrophy of the duodenum—jejunum in CD patients could be compensated by enhanced absorption in the distal intestinal segments [13
]. The fat absorption coefficient could in fact be preserved in a patient with a partially atrophic bowel [13
]. This process could be similar to that occurring in the residual bowel after surgical resection, which involves structural changes that lead to an increased absorptive attitude of the intestine. The intestinal adaptation consists of morphological changes of the mucosa, including increased villus height, crypt depth, and epithelial cell number. In CD patients, atrophy determines the loss of normal intestinal function. This can hypothetically induce increased absorption of the functionally preserved intestinal tract. If this process overcompensates, it could lead to the extraction of energy exceeding the child’s needs, thus increasing the risk of overweight/obesity [13
This compensatory hypothesis appears to be supported by some of the first published cases of adolescents affected by CD who continued to present with overweight or obesity despite persistent villous atrophy on jejunal biopsies [14
]. The compensatory surface area of the small intestine appears to increase with patient age. Therefore, the intestine may develop the ability to absorb an adequate amount of compensatory energy [13
]. This notion is corroborated by the particular distribution of symptoms upon CD diagnosis, which appears to be related to age [5
]. Children aged less than 2 years often exhibit the classic CD presentation, which includes malabsorption. In contrast, older children, adolescents, and adults often present with atypical symptoms. This appears to be concordant with the compensatory hypothesis. In fact, the classic symptoms may be due to a lack of intestinal adaptation, which is less developed in young children as mentioned above. The absence of intestinal adaptation induces the occurrence of severe and classic symptoms including malabsorption and celiac crisis, which can be found in very young children newly diagnosed with CD. As intestinal adaptation is a time-dependent phenomenon, the probability that an individual’s mucosa is modified increases with age. Therefore, CD symptoms could be attenuated in older children and adolescents.
Concordant with this hypothesis, there is no correlation between the presentation of CD and the degree of villous atrophy [37
] or the extent of the intestine involved as visualized through video-capsule endoscopic procedures [38
]. The morphological appearance of the mucosa may be unrelated to its functional expression responsible for the severity of the presenting symptoms.
In addition, the nutritional status of the underlying population is clearly very important for the correct interpretation of BMI in children with CD at diagnosis. CD may indeed develop in patients with overweight/obesity, reflecting an individual’s predisposition (i.e
., genetic, nutritional, and environmental factors). The worldwide prevalence of overweight/obesity in children has increased over the last 2 decades; an estimated 60 million children will be overweight or obese by 2020 [39
]. In this scenario, the symptoms of malabsorption that could manifest in overweight patients at CD onset may reduce the prevalence of overweight/obesity in CD patients compared to the reference population but increase it in comparison to what is usually expected in CD patients.
3.2. The Effect of Gluten Withdrawal on Overweight/Obese CD Patients
Overweight or obesity may develop in CD patients after gluten withdrawal. The main surveys on children discussed above report the normalization of BMI in underweight and overweight patients on a GFD, although they also report the development of overweight and obesity independent of baseline nutritional status [21
]. In consideration of the abovementioned “compensatory” hypothesis, it can be supposed the mucosal healing following gluten withdrawal is responsible for the normalization of BMI in both underweight and overweight patients as a result of the recovery of energy balance. Therefore, the restoration of the absorptive functions of the whole bowel could constitute a physiological redistribution of the absorptive attitude in whole bowel mucosa. This could result in an increased energetic yield in patients with symptoms of malabsorption. However, in patients with a mucosa adapted to supply a higher energetic yield, the improved absorptive function of the whole bowel could induce the normalization of caloric balance. Nevertheless, it remains to be determined if a GFD itself is a cause of the development of overweight/obesity in CD patients.
The unpalatability of some gluten-free foods may induce a preference toward hyperproteic and hyperlipidemic foods [20
]. This may consequently lead to increased energy intake followed by excessive weight gain [41
]. Mariani et al
] examined the eating habits and diet composition of 47 adolescents with CD and compared them to those of 47 healthy age-matched control subjects. They divided the CD patients into 2 subgroups according to compliance with a GFD: group 1A patients rigorously adhered to a GFD, while group 1B patients did not comply with a GFD. Compared to Recommended Dietary Allowances, total energy, lipid, and protein intake were higher and carbohydrate intake was lower in CD patients and controls. Total caloric intake and lipid and protein consumption were higher in group 1A than in group 1B. As a consequence, overweight/obesity was more frequent in group 1A (72%) than in group 1B (51%) and the controls (47%).
Several studies confirm long-term GFDs may not be nutritionally balanced. Indeed, there is clinical evidence indicating high simple sugar, protein, and saturated fat intake as well as low complex carbohydrate and fiber intake in such diets [20
]. Concordant with this pediatric evidence, higher total caloric [44
], carbohydrate, and fat [45
] intake is reported among adults with CD than among healthy control subjects. In contrast, a few studies in adults [46
] and children [48
] report reduced caloric intake in CD patients on a GFD.
Besides increased total caloric intake, the macronutrient composition of the diet may be involved in the pathogenesis of overweight and obesity in patients with CD. Carbohydrates are the major energy source in the diet of children in developed countries and are the dietary components that most strongly affect blood glycemia. Both the quantity and type of carbohydrates are the determinants of postprandial glycemia [49
]. The glycemic index (GI) is a parameter used to classify foods according to their postprandial glycemic response [50
Many gluten-free foods are characterized by a GI higher than that of equivalent gluten-containing foods [50
], although this is refuted by some authors [52
]. Gluten-free foods have a higher GI, because gluten protein does not allow the easy access of amylase to hydrolyze starch granules in the lumen of the small intestine [50
]. However, many foods with a high GI have been shown to only slightly increase blood glucose and vice versa [51
]. Thus, the GI provides a measurement of the quality but not the quantity of the carbohydrates consumed. Meanwhile, blood glycemia is influenced by the synergistic interaction between the quantity and quality of carbohydrates. Therefore, epidemiological studies are utilizing a new concept to assess outcomes as a result of glucose metabolism: the glycemic load (GL). The GL may be calculated with the product of GI (as a percentage) of available carbohydrates, representing both the quality and quantity of carbohydrates consumed. The GL may be interpreted as a measure of insulin requested in free-living conditions, because the amount of carbohydrates consumed at each meal usually varies in such conditions [51
]. Nevertheless, if the blood glucose response to food is a determinant of body weight remains controversial [56
Several studies conducted in overweight or obese children show discordant results regarding the associations of GI and GL with obesity. One cross-sectional study reports no association of body fat with GI or GL [57
]. Others studies show positive associations of GI and GL with waist circumference, BMI, and the sum of 4 skinfolds [58
]. However, other cohort studies report inconclusive results [60
]. A meta-analysis [63
] that identified six eligible randomized clinic trials including a total of 202 participants concludes that low-GI or low-GL diets confer marked benefits on weight, BMI, total fat mass, and lipid profile. Regardless, further research on long-term improvements is required. A more recent systematic review provides evidence that long-term interventions with a low-GI/GL diet confer beneficial effects on fasting insulin and pro-inflammatory markers such as C-reactive protein; such interventions might prove to be helpful in the primary prevention of obesity-associated diseases [64
]. These aspects could help explain the occurrence of overweight/obesity in celiac patients on a GFD.
On the other hand, several studies evaluating the effects of a GFD on metabolic control, growth, and nutritional status in celiac patients with type I diabetes provide a natural model of the interactions between diet, glycemic response, and nutritional status, demonstrating how this interrelationship can be much more complex. However, these studies have completely discordant results. Some studies [65
] indicate improvements in BMI and glycosylated hemoglobin (HbA1c) levels in patients with CD and type I diabetes on a GFD. Meanwhile, Novòa Medina et al
] report no effect on the metabolic control, height, or weight of such patients. Other studies evaluated the influence of GFDs on metabolic parameters including insulin dose, HbA1c, glucose excretion, and hypoglycemic episodes. Saadah et al
] report that a GFD resulted in a significant improvement of growth and influenced diabetic control, particularly higher insulin levels in patients with CD than the levels at baseline. Other authors [68
] found no significant difference in the insulin dose, HbA1c, 24-h urinary glucose excretion, or the number of hypoglycemic episodes. Abid et al
] found that a GFD reduced gastrointestinal symptoms in the short term and particularly episodes of severe hypoglycemia in children with type I diabetes with CD; however, there were no changes in the standard deviation scores for height, weight, BMI, or the average HbA1c before and after GFD consumption. Furthermore, epidemiological studies show energy intake is predictor of weight gain [71
]. Thus, the GI and GL of the previous meal can theoretically influence energy intake in the next meal. A recent meta-analysis on this topic suggests that the GI, but not the GL might influence the energy intake of the next meal [72
]. This may be because low-GI foods result in sustained blood glucose levels and hunger is delayed as compared with that after a high-GI meal [73
]. Furthermore, recent evidence suggests energy intake is associated with changes in the resting metabolic rate [74
]. The mechanism involved in this phenomenon may be the specific effect of blood glucose level on satiety (i.e
., the glucostatic theory) or of other stimuli (e.g., peptides) involved in the control of appetite. Insulin and glucose stimulate the release of the leptin hormone that produces satiety and suppress the release of the ghrelin hormone that stimulates the appetite. Regardless, this does not precisely characterize relationships among GI, satiogenic leptin, and appetitic ghrelin. Furthermore, several gastrointestinal hormones called incretins are involved in the physiological control of hunger and satiety; they are involved in glucose metabolism and can act on pancreatic beta cells to stimulate insulin secretion. Among these hormones there is glucagon-like peptide-1 (GLP-1), which acts directly on the central nervous system and indirectly by slowing gastric emptying, inhibiting appetite and food intake, and inducing body weight reduction. The stimulation of insulin secretion by incretins is typically glucose dependent and manifests when glycemic levels are high but not when normal or low. Therefore, incretins have the potential to reduce hyperglycemia without causing hypoglycemia. A recent study revealed children with CD have a secretion pattern of gut-brain axis hormones that differs from that of controls. Alterations in this axis were more pronounced in children with both CD and type I diabetes mellitus; nevertheless, the roles of these gut-brain axis hormones in food intake and glycemic control in patients with CD and type I diabetes mellitus must be clarified [75
Overall existing clinical evidence explains the variability of the anthropometric trends in CD patients after gluten withdrawal. However, it does not clarify why some CD patients develop overweight/obesity after beginning a GFD. Furthermore, as is the case in newly diagnosed CD patients, the global trend toward increased overweight/obesity could explain why CD patients on a GFD may become overweight. The changes in nutritional habits that induce the development of obesity are probably shared by CD patients and the general population. Table 2
summarizes the main pathogenetic links between CD and overweight/obesity.
Suggested pathogenetic links between CD and overweight/obesity.
Suggested pathogenetic links between CD and overweight/obesity.
|Time of Overweight/Obesity Diagnosis in Celiac Patients||Pathogenetic Link||Reference|
|Overweight/Obesity at CD Presentation||“Compensatory hypothesis”: high energetic yield due to the slow adaptation of the atrophic mucosa|||
|Global trend toward overweight/obesity in children|||
|Diagnosis not based on clinical symptoms but on screening test||[25,26]|
|Overweight/Obesity on a GFD||“Compensatory hypothesis”: normalization of caloric balance due to the restoration of mucosal functions|||
|Worldwide trend toward overweight/obesity in children|||
|Unpalatability of gluten-free foods, prompting the consumption of foods with high caloric content (i.e., fat and protein).||[20,29,40]|
|High GI/GL of GFD?||[49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74]|
|Altered secretion of gut–brain axis hormones?|||