The gluten-free diet represents the therapy for patients affected by celiac disease (CD) and must be adhered to throughout the patient's lifetime. However, several studies have shown that dietetic compliance in these patients frequently is not strict and that in adolescence correct diet is frequently neglected (1-5). Furthermore, it is well known that the alimentary patterns of healthy adolescents in industrialized countries are nutritionally unbalanced. In particular, they are often characterized by excessive consumption of energy, proteins, and fats, and a reduced intake of complex carbohydrates and fiber (6-10). Adherence to a strict gluten-free diet theoretically would worsen some of these alimentary excesses, because food prohibited in the diet of a patient with CD is mainly composed of complex carbohydrates.
The purpose of the present study was to evaluate compliance with diet in a group of adolescents with CD and to estimate their nutritional choices in intake of energy, macronutrients, iron, calcium and fiber and frequency of consumption of food.
MATERIALS AND METHODS
Forty-seven patients (group 1), 10 males and 37 females, aged from 10 to 20 years (mean age ± standard deviation, 15.2 ± 2.3 years; median age, 14.9 years), affected by CD diagnosed according to ESPGHAN criteria (11), were enrolled in the study. They were seen as outpatients at regular intervals at the Pediatric Clinic of Rome University "La Sapienza." Compliance with a gluten-free diet was evaluated by dietary record and verified by an assessment of antigliadin antibodies immunoglobulin (Ig) A and antiendomysium antibodies (12,13). Antigliadin antibody IgA was measured by enzyme-linked immunosorbent assay (Eurospital, Trieste, Italy) (14), and antiendomysium IgA was evaluated by an indirect immunofluorescence method (Biosystem, Milan, Italy). Sections from the distal portion of monkey esophagus were used as a substrate, and fluorescein-labeled goat antihuman IgA antibody was used as the second antibody. The patients' serum was diluted 2:5 in phosphate buffer (pH 7.2). The presence of a brilliant green network pattern under fluorescence microscope was taken as positive (12). No patient with IgA deficiency participated in the study.
All subjects compiled a 3-day alimentary record (2 weekdays plus Sunday) to evaluate energy consumption; percentage distribution of macronutrients; and calcium, iron and fiber intakes. An informative talk about how to compile the record and a dietetic interview to verify food weights in grams preceded the recording.
As a control group (group 2), we evaluated 47 healthy adolescents, 13 males and 34 females (mean age ± standard deviation, 15.7 ± 2.3 years; median age: 15.2 years) who compiled an alimentary record in a similar modality. As in the case of group 1, all subjects in the control group were middle and high school students in the Rome area.
Thirty of the 47 adolescents with CD (64%) reported strict compliance with a gluten-free diet, 14 (30%) admitted consumption of prohibited food once or twice a week, 3 (6%) acknowledged consuming a gluten-containing diet. However, when serologic markers were evaluated, 5 (16%) of the 30 patients who reported a strict gluten-free diet were antigliadin antibody IgA- and/or antiendomysium antibody-positive.
On the basis of these data, the actual compliance with the diet of our patients was 53% (25/47). It was therefore possible to compare the alimentary consumption of the 25 adolescents who complied with dietary prescriptions (group 1A) and that of the remaining 22 subjects who are gluten-containing foods (group 1B).
The patients of groups 1A and 1B did not show significant differences in distribution by sex, median age, weight (mean ± standard deviation, 52.5 ± 19 vs. 48.8 ± 17.6 kg, respectively) and height (mean ± standard deviation, 156.6 ± 25 vs. 152 ± 24 cm, respectively). Nutrient intake was compared with the Italian recommended consumption of nutrients (LARN 1996) (15) and with the United States recommended daily allowances (RDAs) (16). Statistical analysis was performed using Student's t-test.
Total energy consumption; the macronutrient division; and the iron, calcium, and fiber intakes in adolescents with CD and in control subjects, together with the reference values are shown in Table 1. Average percentages of nutrient intake (in comparison with RDAs) in the group with CD and in the control group, respectively, were: energy, 110.4 ± 26.3% and 117 ± 28%; carbohydrates, 71.5 ± 16.4% and 73 ± 16.8%; lipids, 106 ± 25% and 104.6 ± 24.7%; proteins, 129 ± 29.3% and 140 ± 31.7%. The percentages of subjects with adequate intakes of iron, calcium and fiber were similar in the group with CD (27%, 21% and 13%, respectively) and in the control group (21%, 17%, and 22%).
Total energy intake; the macronutrient division; and iron, calcium and fiber intakes in adolescents with CD complying with a gluten-free diet (group 1A) and in those consuming a gluten-containing diet (group 1B) are shown in Table 2. Average percentages of nutrient intake (in comparison with RDAs) in group 1A and in group 1B patients, respectively, were: calories, 115.2 ± 27.4% and 106 ± 25.2%; carbohydrates, 68 ± 15.6% and 75 ± 17%; lipids, 107 ± 25.4% and 105.3 ± 24.9%; proteins, 154 ± 35% and 109 ± 24%; calcium, 70 ± 36% and 50 ± 25%; and iron in males, 90 ± 31% and 72.8 ± 25% and in females, 88 ± 30.8% and 65 ± 22%. The frequency of consumption of different kinds of food inferred from alimentary records of the two groups of patients with CD is reported in Table 3. The distribution of real body weight (RBW) in patients with CD (groups 1A and 1B) and in control subjects is shown in Table 4.
The percentage of adolescents with CD who strictly follow a gluten-free diet varies from 43% to 81%, according to several studies (2,17-20); in the present study this percentage was 53%, confirming the need to make every effort to undertake a proper follow-up of patients with CD. On the whole, the alimentary habits of our patients show nutritional imbalances. In fact, compared with the Italian LARN (15) and to the United States RDA (16), our patients' diets are hyperproteinic and hyperlipidic and contain low amounts of carbohydrates, iron, calcium, and fiber.
As far as we know, only a few prior reports have been published on the alimentary habits of patients with CD, and these always concerned children on a gluten-free diet. Polito et al. (21) and Rea et al. (22) observed an excess in energy, animal protein, and lipid intake, which is partially responsible for the high percentage of overweight patients. However, Ansaldi et al. (23) observed a hypocaloric diet with the same macronutrient imbalances.
In our study, the presence of a control group of healthy adolescents and the division of those with CD in two subgroups according to their compliance with the diet allowed us to define better the nutritional choices that are induced by the elimination of gluten from the diet.
Comparing the alimentary habits of these groups, we noted that the nutritional imbalance of the diet of patients with CD was not the consequences of the elimination of gluten from the diet, because even the healthy adolescents' diet was hyperproteinic and hyperlipidic and therefore, low in carbohydrates. Moreover, an adequate amount of iron, calcium, and fiber was part of the diet in only 20% of all adolescents, with or without CD. These results are not surprising. They confirm results of nutritional investigations of adolescents in industrialized countries during recent years (6-10).
The relevant data are those obtained through the comparison of nutritional choices of patients with CD who consume a strict gluten-free diet with those of patients with CD who consume a gluten-containing diet. Energy intake was higher in group 1A than in group 1B, particularly in the girls; but this difference did not reach statistical significance. As far as the percentage distribution of macronutrients is concerned, protein intake was significantly higher and carbohydrate intake was significantly lower in group 1A patients than in group 1B patients. Furthermore, patients with CD complying with a strict gluten-free diet consumed a higher amount of lipids than the gluten-consuming group, although these data are not statistically significant. Even fiber consumption was significantly reduced in subjects consuming a gluten-free diet, whereas intakes of iron and calcium were increased. Therefore, in patients with CD, adherence to a strict gluten-free diet worsens the already nutritionally unbalanced diet of all adolescents, increasing the already elevated protein and lipid consumption.
To understand the origin of these incorrect dietary habits, we evaluated the frequency of consumption of different kinds of foods. It was obvious that among patients with CD consuming a gluten-free diet the reduced consumption of some foods containing carbohydrates such as bread, pizza, or pasta was not counterbalanced by an increase of alimentary items of the same group (rice, corn, and potatoes). Group 1A patients preferred food with a high amount of protein (meat, eggs, legumes) and snacks with a high content of lipids much more often than group 1B patients. The greater consumption of meat and legumes by group 1A patients was responsible for the higher intake of iron and fiber.
There was no connection between the calcium intake recorded, which was greater among patients in group 1A, and the frequency of milk consumption and its products, which was similar in the two subgroups. Basically, a diligent analysis of alimentary records showed a greater consumption of parmesan cheese in patients in group 1A, possibly to make pasta more appetizing.
The distribution of real body weight in patients with CD revealed a greater percentage of overweight and obese adolescents among those who complied with dietetic prescriptions (72%) than among subjects consuming an unrestricted diet (51%). We speculate that the dietetic habits (high energy, protein and lipid intake, combined with low carbohydrate consumption) in patients with CD consuming a gluten-free diet with a normal small bowel mucosa could cause a disturbance in the growth pattern.
The results of the present study provide a discouraging description of the alimentary habits of adolescents with CD, in the poor compliance with diet in a high percentage of subjects and in the presence of nutritional imbalances, particularly in those consuming a strict gluten-free diet. This last observation could be partially explained by the adolescents' infrequent consumption of gluten-free commercial foods, because of poor palatability, and the cost and difficulty of obtaining more palatable food. To solve these problems, it would be useful if a nutritionist advised patients on dietary control during periodic medical checks, particularly during adolescence.
In conclusion, our results show that, paradoxically, in adolescents with CD, a strict gluten-free diet may be a nutritional risk factor, altering the already unbalanced diet commonly consumed by healthy adolescents, because it leads these subjects to incorrect alimentary choices.
Although in the past years there has been great improvement in diagnostic tools for CD, we believe the moment has come to devote more attention to problems regarding a correct follow-up of CD, not only the compliance with diet, but also the nutritional balance of the diet itself.
1. Montgomery AMP, Goka AKJ, Kumar PJ, Earthing MJG, Clark ML. Low gluten diet in the treatment of adult celiac disease: Effect on jejunal morphology and serum antigluten antibodies. Gut
2. Colaco J, Egan-Mitchell B, Stevens FM, Eottrell PF, McCarthy CF, McNichol F. Compliance with gluten-free diet in coeliac disease. Arch Dis Child
3. Kumar PJ, Walker-Smith J, Milla P, Harris G, Colyer J, Halliday R. The teenage coeliac: Follow-up study of 102 patients. Arch Dis Child
4. Kokkonen J, Viitanen A, Simila S. Coping with a coeliac diet after adolescence. Helv Paediatr Acta
5. Mayer M, Greco L, Troncone R, Auricchio S, Marsh MN. Compliance of adolescents with coeliac disease with a gluten-free diet. Gut
6. Giovannini M, Galluzzo C, Scaglioni S, et al. Indagine nutrizionale nel comune di Milano: Dati antropometrici, intake calorici e abitudini alimentari in età scolare. Riv Ital Pediatr
7. Pinelli L, Cirillo D, Golinelli M, et al. Indagini su apporti alimentari e dati antropometrici nell' età scolare. Rilievi su un campione di 1177 bambini del comune di Verona. Riv Ital Pediatr
8. Spyckerelle Y, Herbert B, Didelot-Barthélemy L, Bairati I, Deschamps JP. Alimentation des adelescentes en Lorraine. Arch Fr Pediatr
9. Skinner JD, Salvetti NN, Enzel EM. Appalachian adolescent's eating patterns and nutrient intakes. J Am Diet Assoc
10. Doyle W, Jenkins S, Crawford MA, Puvandendran K. Nutritional status of schoolchildren in an inner city area. Arch Dis Child
11. Walker-Smith JA, Guandalini S, Schmitz J, Shmerling DH, Visakorpi JK. Revised criteria for diagnosis of coeliac disease. Arch Dis Child
12. Burgin-Wolff A, Gaze H, Hadzielimovic F, et al. Antigliadin and antiendomysium antibody determination for coeliac disease. Arch Dis Child
13. Gemme G, Delogu A, Nobili F, et al. Anticorpi antiendomisio: Ruolo attuale nella diagnosi di malattia celiaca a confronto con gli anticorpi antigliadina. Pediatr Med Chir
14. Savilahti E, Perkkio M, Kalimo K. IgA antigliadin antibodies: A marker of mucosal damage in childhood coeliac disease. Lancet
15. Società Italiana di Nutrizione Umana. Livelli di assunzione raccomandati di energia e nutrienti per la popolazione italiana. LARN Revisione
16. Recommended daily allowances. Nutr Rev
17. Savilahti E, Simmel O, Koskimies S, Riva A, Akerblorn HK. Celiac disease in insulin-dependent diabetes mellitus. J Pediatr
18. Maki M, Lahdehaho ML, Halstromm O, Viander M, Visakorpi JK. Postpubertal challenge in coeliac disease. Arch Dis Child
19. Ljungman G, Myrdal U. Compliance in teenagers with coeliac disease: A Swedish follow-up study. Acta Paediatr
20. Bardella MT, Molteni N, Prampolini L, et al. Need for follow-up in coeliac disease. Arch Dis Child
21. Polito C, Olivieri AC, Marchese L, et al. Weight overgrowth of coeliac children on gluten-free diet. Nutr Res
22. Rea F, Polito C, Marotta A, et al. Restoration of body composition in celiac children after one year of gluten-free diet. J Pediatr Gastroenterol Nutr
23. Ansaldi N, Palma L, Dell' Olio D, Malorgio E, Che cosa mangiano i bambini celiaci? Analisi dietologica su un gruppo di celiaci a dieta. Riv Ital Pediatr