Celiac disease (CD) is a systemic immune-mediated disorder caused in genetically susceptible persons by the ingestion of gluten-containing grains [1
]. The only available treatment is the gluten-free diet (GFD), which consists of the dietary exclusion of grains containing gluten (i.e., wheat, rye, barley) [2
The nutritional adequacy of the GFD remained controversial and a matter of debate for a long time [3
]. Indeed, apart from maintaining the safe limit of gluten intake (below 10–50 mg/day), a suitable GFD must also be nutritionally balanced and cover all energy and nutrient requirements to prevent deficiencies and ensure a healthy life. In children, the GFD must also allow appropriate growth and pubertal development [4
]. A body of evidence has so far suggested that the GFD may be nutritionally unbalanced either because of the need to exclude several cereals or because of the different nutritional composition of GF products as compared to their unrefined analogs [3
To the best of our knowledge, there are no large case-control studies performed on children regarding the nutritional adequacy of the GFD. Previous studies have mostly been performed on adolescents or adults, with the limit of (1) small sample sizes, (2) lack of a control group, (3) retrospective methods of dietary recording, and (4) inclusion of patients at diagnosis.
Therefore, we aimed to evaluate the nutritional status, the dietary intake and adherence to the national recommended dietary allowances as well as to the Mediterranean diet of Italian children with CD on the GFD by a large, prospective case-control study.
The present case-control study shows that the nutritional status of CD children does not differ from healthy children. However, the diet of CD children in this study was nutritionally less balanced than controls, with a higher intake of fat and a lower intake of fiber, highlighting the need for dietary counseling.
Data from the literature on the effects of GFD on anthropometric parameters of patients with CD are controversial. On the one hand, it has been reported that a good compliance to the GFD is associated with a positive effect on anthropometric parameters with a recovery of lean body mass, normalization of BMI in both underweight and overweight children, and acceleration of linear growth [23
]. On the other hand, there are also studies suggesting that the GFD may have a negative effect on body composition and anthropometric parameters in CD patients, with an increased prevalence of overweight and obesity [5
]. These conflicting data may in part be caused by differences in the duration of the GFD at the time of anthropometric assessment or by the lack of a control group. Our study is the first to evaluate the BMI in a large sample of CD children on a GFD for at least two years as compared to healthy children, showing that there is no difference in the percentage of underweight, normal weight and overweight/obesity between groups. We also evaluated the energy expenditure in the two study groups through lifestyle analysis, showing no differences between CD children and the control group. The similar BMI in the presence of a similar lifestyle suggested that energy intake was similar in the two study groups. Indeed, we did not observe any difference in total daily energy intake.
Nonetheless, concern about the nutritional quality of the GFD emerges from our results. Indeed, by the analysis of 3-day food diaries, we found a higher intake of fat and a lower intake of fiber and carbohydrates in CD children on a GFD as compared to healthy children, while there was no difference in the daily intake of protein. As regards carbohydrates, when comparing the daily intake of macronutrients of CD and control children with the Italian recommendations, the percentage of energy supplied by carbohydrates was, however, in line with the LARN recommendations in both groups. Noticeably, healthy children exceeded the daily intake of simple sugars as compared to LARN recommendations, while CD children did not. The main concern about GFD was the higher consumption of total and saturated fats observed in CD children, with the intake of saturated fat exceeding the nutritional goal recommended by LARN only in the CD group. The intake of fiber was also a concern, being lower in CD children as compared to controls and to LARN recommendations.
Our findings are in line with several previous studies that compared the intake of macronutrients in CD patients with the national recommendations, showing, overall, that CD patients consume less fiber and more fats than recommended [5
]. When comparing the nutritional quality of CD patients on a GFD to that of healthy controls, previous studies showed conflicting results. Consistent with our findings, several studies in adults reported a higher intake of fats in CD patients as compared to healthy subjects [6
], while others reported a lower intake of carbohydrates and protein [5
] or only a lower intake of fiber [7
] or no differences in CD adolescents as compared to a control group [12
]. Finally, Zuccotti et al. showed a higher intake of carbohydrates and lower consumption of fat in 18 CD children as compared to 18 healthy controls by a 24 h recall [11
]. Differences between studies may be explained by the different age of patients studied (children versus adolescents and adults), the small sample size of many previous studies, the different methods of dietary collection (prospective food diary versus retrospective recall), and finally by the inclusion of patients both at diagnosis and on GFD in some of the studies. Our study firstly evaluated prospectively the macronutrient intake in a large sample of CD patients of pediatric age with at least 2 years’ experience of GFD as compared to healthy subjects by 3-day food diary that is one of the best-practice methods to obtain dietary data [29
One of the main factors that could explain the unbalanced intakes of nutrients is the dietary pattern. For this reason, in our study, we compared the dietary habit of CD children and healthy subjects with respect to the IFP, showing that: (a) CD children have a higher consumption of processed meat and salty snacks as compared to healthy children; (b) both groups did not reach the portions recommended by the IFP for legumes, vegetables, eggs and fish, while exceeding the consumption of sugary drinks, meat and processed meat; (c) the consumption of minor and pseudo-cereals was very low in the CD group, and the major contributors to cereals were gluten-free products. These results may explain the higher intake of fat and lower intake of fiber observed in CD children, however, they also highlight that the dietary habits of Italian children, either celiacs or controls, are not fully adherent to the Mediterranean diet. Indeed, the KIDMED index was moderate in both groups.
Furthermore, several studies have shown that the nutritional profile of GF products specifically formulated for CD patients is different with respect to regular foods, with a higher content of fat and saturated fat, salt, sugar and a lower content of fiber [14
]. In our study, commercial GF products specifically formulated for CD patients provided 46% of the total daily energy, thus playing a major role in influencing the imbalance in the diet of CD children. Many GF foods are prepared from refined maize flour and white rice, which are lower in fiber (2.6 g and 0.7 g per 100 g, respectively) than wheat (3.5 g per 100 g) or whole wheat (9.6 g per 100 g) [19
]. The exclusion of gluten and the use of only GF raw materials as ingredients result in GF food which is less palatable than regular foods; consequently, the manufacturing of GF foods requires the addition not only of some additives, such as hydrocolloids, but also of some macronutrients, such as fats in the final products to mitigate the loss of gluten. Our study highlights the need of enhancing the nutritional quality of GF products.
The main weakness of the present study was the lack of data on micronutrient intake, an important piece in the puzzle of the nutritional quality of the GFD. This limitation was related to the lack of tables on GF products indicating the micronutrient content. Therefore, it was not possible to accurately evaluate the corresponding intake in the diet. Second, potential recording errors including inaccurate estimates of portions consumed and omission of foods (either deliberate or unintentional) could result in an underestimation of nutritional intake, as in all food diary recording. Finally, results on food group intake were collected by the 3-day food diary, that is not the best instrument to estimate the consumption frequency of some foods, such as those that are not eaten daily (e.g., legumes, fish, egg).