Nutritional status, nutrient imbalances, food-related behaviors and dietary supplements use among patients with celiac disease on a gluten free diet in Lebanon: a national cross-sectional study

Background: Celiac disease is an autoimmune disorder triggered by gluten, that occurs in susceptible individuals and is associated with dietary restriction and subsequent nutritional deficiencies. This study investigated the diet quality, nutrition imbalances and nutrition status among young children,adolescents and adults with CD who were referred to several hospitals in Lebanon. Methods: A cross-sectional study in 50 individuals (31.74 ± 15.64 years) with CD who follow a gluten free diet was conducted, using biochemical parameters, anthropometric measurements, dietary and physical activity assessments. Results: Of the 50 participants, 38% and 16% were presenting low serum levels of iron and vitamin B12, respectively. The majority of participants were physically inactive and around 40% of them had low muscle mass. A weight loss of 10% to 30% indicating mild to moderate malnutrition was shown in 14% of individuals. The assessment of food-related behaviors shows that 80% of participants were reading nutrition labels and 96% of them were following gluten-free diets (GFD). Some barriers including family ignorance (6%), language of the nutrition labels (20%) and expensive GF products (78%) were limiting the adherence to GFD. The inadequacy of the daily energy intake along with insufficient intakes of calcium and vitamin D were remarked among individuals with CD. However, protein and iron intake were exceeding the recommendations among all age groups, except in males aged 4-8 years and 19-30 years. Half the study participants were using dietary supplements where 38%, 10%, 46%, 18%, 16% and 4% used vitamin D, vitamin B12, iron, calcium, folate and probiotics, respectively. Conclusion: GFD is the key treatment for CD. However, it is not without inadequacies and may cause certain deficiencies such as calcium and vitamin D leading to reduced bone density. This underlines the critical role of dietitians in education and maintenance of healthy GFD among individuals with CD.


Introduction
Celiac disease (CD) is a chronic inflammatory autoimmune disorder characterized by partial to total villous atrophy of the small bowel. 1 The exposure to dietary wheat gluten and proteins of rye and barley triggers CD among genetically predisposed persons. 2 In certain circumstances, individuals show no symptoms and the diagnosis is only possible through serologic screening. 3 According to the European Society for the Study of Celiac Disease (ESsCD) guideline, the diagnostic criteria depend upon the exposure to dietary gluten or related proteins of rye and barley-dependent symptoms, through some histopathologic findings (villous atrophy and crypt hyperplasia) from a biopsy of the duodenum, specific antibody levels, HLA-DQ2 and/or HLA-DQ8. 4 The prevalence of CD is estimated to be one to two individuals per 100 worldwide 5 and roughly one in 99 among children and adolescents. 6 Moreover, according to a recent systematic review and meta-analysis, the pooled seroprevalence of CD in the general population is significantly higher in children (2%) compared with adults (1%, p-value=0.01) and significantly greater among females (1.65%) compared to males (0.8%, p-value=0.04). 7 The pooled sero-prevalence of CD in the Middle Eastern area (1.47%) is higher than that in South Asia (1.25%) and East-Asia (0.06%). 7 The highest prevalence among the Arab regions was reported in Saudi Arabia (3.2%), and the lowest in Tunisia (0.1%). 8 In Lebanon, the prevalence of CD was around 0.5% in 2011. 9 Furthermore, women have shown elevated prevalence compared to men in the Arab countries 8 and according to solid evidence, the risk becomes common and high if a first-or second-degree relative has been previously diagnosed. 6 Questionnaires and dietitians' assessment The survey contained 78 questions concerning demographics, disease's characteristics, history and presentation, foodrelated behaviors, lifestyle and diet history, weight before and after being diagnosed with CD, adherence to GFD and social and emotional barriers for sustaining the GFD. The questionnaire was derived from the one available from the Canadian Celiac Association website and used previously in the "Canadian Celiac Health Survey". 16 Medical history was evaluated based on the age at diagnosis of CD, family history of CD, symptoms (diarrhea, headache, flatulence and bloating), comorbid diseases (food allergies, food intolerance, other non-communicable disease). Alongside the first questionnaire, additional FFQ for caloric and protein intake, calcium, vitamin D and iron intakes were addressed to individuals with CD along with questions on dietary supplement use. This FFQ was derived from two Lebanese-validated questionnaires to address the questions for children, adolescents and adults 17,18 in which a substitution of 24 foods with GF foods was executed (see Underlying data 69 ). Dietary supplement use was assessed through patient's response of "Yes" or "No" to the statement: "I use dietary supplements in addition to GFD". To investigate the type of supplements used, patients were asked to answer the following question: "Since you were diagnosed, have you used any of the following dietary supplements?" The specific options included vitamin D, probiotics, iron, folate, calcium, and vitamin B12, which are the commonly used DS among individuals with CD. 19 The questionnaire used in the present study was reviewed by two national and one international expert in CD and was pre-tested in 14 non-expert individuals as a focus group. The aim of the focus group was to explore a specific set of issues: the questionnaire's readability, reliability and inter consistency (Cronbach-alpha=0. 69). Moderators often commence the focus group by asking broad questions about the topic of interest, before asking the focal questions. Although participants individually answer the facilitator's questions, they are encouraged to talk and interact with each other. In addition, three-day food records were collected and analyzed using Nutritionist Pro software (version 3.2, AXXYA) to assess caloric, micro-and macronutrients intake. The mean value and the standard deviation (SD) were used when interpreting the nutrient's intake.
Weight and height were measured using automated scale and stadiometer. Body composition including body fat, lean body mass and visceral fat were assessed using Bio-impedancemetry body composition analyzer (BF-511, OMRON). Percentage body fat (%BF) was measured by whole-body BIA to the nearest 0.1% using a digital scale/body composition monitor (Omron BF 511, 50 kHz, 500 µA, Kyoto, Japan), which includes an 8-sensor technology using both hands and feet. The participants stood with bare feet on electrodes on the scale with their knees and back straight while grasping a handle that also includes electrodes with both their hands horizontally raised, elbows extended straight, and maintaining a 90°-angle to the body. A previous study that compared body composition estimates using BIA devices with DXA and whole body magnetic resonance imaging indicated that the use of devices with additional hand electrodes provides a more accurate prediction of body composition and are suitable for public use. In order to determine percentage body fat, the device uses electrical impedance, along with the participant's height, weight, age, and gender to generate results. The readings were obtained in duplicates and the average recorded. According to the manufacturer's instructions, percentage body fat was measured two hours or more after breakfast. Omron Healthcare references for age and gender were used to classify the study participants into low, normal, high, and very high %BF.
The standard formula of body mass index (BMI) was calculated by dividing weight in kilograms (kg) by height in meters squared (m 2 ).

Laboratory values
According to a study by Bledsoe et al. (2019), 19 and Marteau et al. (2001) 20 albumin and micronutrient deficiencies, including vitamins B12 and D, as well as folate, iron, zinc and copper, are common in adults at the time of diagnosis with CD. Thus, the study participants were tested for three out of six micronutrients (serum iron, folate and vitamin B12) along with albumin levels in the present study. Venous blood was collected after a 12-hour period of fasting, centrifuged and stored at À80°C till analyzed. The iron level was measured in serum by the in vitro assay method with Roche Diagnostic (Mannheim, Germany) and Hitachi 704 devices (Roche Diagnostics, Switzerland).
All patients underwent a complete hemogram. However, only albumin, serum iron, folate and vitamin B12 were reported in this study to comply with the outcomes discussed in Bledsoe et al. 19 Serum iron was diagnosed according to the criteria followed by the laboratory in which the analysis occurred. 21 The reference values for children were: 8.9-21.4 μmol/L (SI units), males (adults): 14-32 μmol/L (SI units) and females (adults): 11-29 μmol/L (SI units) were compared to the values obtained in the current study. Serum folate and vitamin B12 levels were measured using a competitive immunoassay. As for vitamin B12, the reference values for 1-<11 years: 260-1200 pmol/L, 12-<18 years: 200-800 pmol/L, 18-<65 years: 200-600 pmol/L 22 were also used to compare the results obtained in this study. Concerning the folate reference values, the normal serum levels were, for children: 5-21 ng/ml and for adults: 2-20 ng/mL. 23 As for albumin, it was considered low if the value was <3.5 and high if >5.5 g/dl. 24

Food-related behaviors and GFD dietary adherence
Some food patterns including shopping, cooking and eating out habits were assessed among the study participants. The adherence to GFD was assessed by trained dietitians who investigated the infringement regarding dietary recommendations. 12 Across the three-day food records, patients presenting a consumption between 3 to 10 g of gluten per day were considered as non-adherent. 12,25 Facilitators and barriers To search for possible barriers, patients were asked, "what are the major challenges that limit the adherence to a strict gluten-free diet?". Trained dietitians asked participants to report whatever came to mind concerning this question. The answers were analyzed according to the methodology used by Braun and Clarke. 26 Indicators of malnutrition and criteria for the determination of nutritional status The determination of the nutritional status was based on anthropometric data (weight, height, BMI), hypoalbuminemia and the percentage of weight loss taking into consideration age and sex factors in the previous 6 months. Taking into consideration the last 6 months, prior to the interview, weight loss of 10% to 20% indicated mild malnutrition, a weight loss of 20% to 30% characterized moderate malnutrition, and a weight loss of more than 30% indicates severe malnutrition. 25

Physical activity
Several types of physical activity such as aerobic (fitness), swimming, running, strength training gym, walking, cycling and yoga were investigated. The frequency of physical activity was based on frequency per week: one time per week, twothree times per week, four or more per week, daily or no physical activity.

Statistical analysis
Raw data, except the anthropometric data, were cleaned and exported to the Statistical Package of Social Sciences Software (SPSS) (Version 25.0. IBM Corp: Armonk, NY, USA) for analysis.The Shapiro Wilk test was used to evaluate the normality of data. The descriptive data were presented as percentages for categorical variables and mean AE SD for quantitative variables. Categorical variables were compared with the Chi-square test and continuous variables were compared with the student-T test. P<0.05 was considered a statistically significant difference with a 95% confidence interval.

Ethical considerations
At the beginning of the study, a consent form was signed by participants. Moreover, this study received the approval of the ethics committees at the Lebanese university (CU#18-46). In addition, an assent form was signed by the parents of children and adolescents under 18 years old.

Results
A total of 50 participants (78% females; mean age=31.74 AE 15.64 years) was enrolled in the present study between February and July 2018. Out of 50, eight participants were aged below 18 years. Age categories were classified according to the dietary reference intake classes 27 of which 46% of the population's age ranged between 31 and 50 years. In the current study, females were younger than males (p-value=0.03). Moreover, around 66% of all the participants had university degrees and more than 70% were remunerated for more than 1,500 million Lebanese pounds (LBP) per months. Half the study's participants were married. Overall, in almost all the characteristics, there were no significant differences between females and males in this study (p>0.05). The sociodemographic characteristics of the study participants are displayed in Table 1. Table 2 includes the medical history, anthropometric measures and lifestyle characteristics of the study participants. In the current study, males were taller than females (p-value=0.042). It was observed that the weight of all participants increased in a range of 4 to 5.6 kg since being diagnosed with CD. This was significantly higher among males compared to females (p-value=0.023). Moreover, 45% of females were overweight compared to 37% males. The assessment of malnutrition was based on Kotze et al. 25 ; a weight loss of 10% to 30% indicating mild to moderate malnutrition were shown in 14% of the study population. The body composition analysis conducted for the study participants revealed that 40% of them had low muscle mass (5.0 to 32.8% for males and 5.0 to 25.8% for females). 28 Further, the mean muscle mass was higher in males compared to females (p-value=0.034). In total, 60% of participants were having high fat mass (more than 20% in males and more than 30% in females). In addition, the mean value of fat mass among women where higher than that among men (p-value=0.014). Furthermore, the majority of the study participants had low visceral fat and there were no significant differences regarding the mean and SD of visceral fat among both genders (p-value=0.25).
As for the medical history, around half the study participants had a family history of CD. The mean age at diagnosis was 22.28AE13.55 (range=1-57 years). Symptoms related to CD such as abdominal bloating, diarrhea, headache, and flatulence appeared in most participants (94%) and 6% were asymptomatic. Around 66% were facing moderate to severe symptoms. According to the medical history, other comorbidities along with CD were reported. Around 24% of participants had CD with other diseases of which dermatitis, arthritis, lactose intolerance and food allergy. As for the medications, 78% of the study participants reported were on medical treatments for CD symptoms.
As for the laboratory values, hypoalbuminemia was observed in 14% of the study population. Women were having significantly higher mean and SD levels of albumin compared to men (p-value=0.02). Out of 50 participants, 38% and 16% were presenting low serum levels of iron and vitamin B12, respectively. None of the study participants presented with folate deficiency. As for the lifestyle behaviors, the majority of participants were non-smokers, non-drinkers and physically inactive ( Table 2).
The food-related behaviors assessed among the study participants are presented in Table 3. Around 80% of the study participants reported reading nutrition labels when shopping groceries. Moreover, almost 96% of them were following a nutritional diet of which was the GFD. To assess their adherence to GFD, the examination of 96% of the participant's  Patients presenting an ingestion of approximately 3-10 g of gluten per day were considered as non-adherent. 12,25 T-test was used to compare means. Categorical variables were analyzed using Chi-Square. Page 10 of 30  In each age category, the age groups (13)(14) were removed due to the limited number of patients. b The missing values in the table are due to the absence of males above the age of 50 years. T-test was used to compare means. Categorical variables were analyzed using Chi-Square.
three-day records reported low intake of gluten (<3g per day). Only 4% weren't adhering to the GFD recommendations. As reported by the participants, some barriers including family ignorance (among 6% of the population), language on the nutrition labels (among 20% of the population) and expensive GF products (among 78% of the population) were limiting the adherence to GFD. It was observed that 72% of the study population had good cooking experience and 64% of the study participants were involved in preparing meals for themselves. On the other hand, almost all the subjects were unaware of a possible cross-contamination during cooking. Despite that, 82% of the subjects with CD preferred fast foods, although they consumed their meals frequently at home (92%). It was reported that the adherence to GFD by these participants induced a hunger feeling during the day. Most of our study population necessitated extra care by family members regarding cooking and eating. No significant differences were observed between genders.
Energy and protein intake compared to the recommended daily allowances (RDA) Table 4 shows the mean energy and nutrients intake calculated through the three-day food records reported by the study participants or their families. Among female children aged four to eight years, the mean daily energy (E) and protein (P) intake were estimated as 1396 AE424.15 Kcal/d and 56.41 AE18.31 g/d, respectively. In comparison with the reference values of the recommended daily allowances (RDA), the daily intake of E and P was 130% and 268%, respectively. On the other hand, it was observed that female children aged 9-13 years had insufficient daily E intake (75% vs RDA) but surpassed by 317% the daily protein requirements. Similarly, among the female participant's age groups 19-30, 31-50 and 50-70, there were inadequacies in the daily E intake along with overconsumption of protein per day. Otherwise, inadequacies in the daily E and P intake were observed among the male children (four to eight years) and those aged 19 to 30 years. Moreover, among those aged 9-13 and 31-50, the daily E intake was inadequate, but the P intake was high. No significant differences regarding the daily E and P intakes were observed between females and males, except in the 19-30 age group, where the daily energetic and protein ratios were high in females compared to males (p-value=0.024 and p=0.047, respectively).
The average daily nutrient intake of calcium, iron and vitamin D is reported in Table 5. Among females and males, none of the study participants showed adherence to the daily recommendations for calcium and vitamin D intake. On the other hand, almost all the study participants from both genders showed a high daily consumption of iron.
Dietary supplements were used by 54% of the study population of which 38%, 10%, 46%, 18%, 16% and 4% used vitamin D, vitamin B12, iron, calcium, folate and probiotics, respectively. No difference was observed between genders regarding DS use (Table 6). According to Figure 1, it appears that the usage of DS among individuals with CD was not excessive.

Discussion
This study, the first of its kind in Lebanon, highlights the nutrition imbalances, protein-energy malnutrition, adherence to GFD, food-related behaviors and dietary supplements use among participants with CD. Our findings show that 78% of individuals with CD were women. This finding was higher than the data reported by Jansson-Knodell et al., 30 and Ashtari et al. 7 which showed that the prevalence of women diagnosed with CD was 65% between 1990 and 2015 and the seroprevalence of CD reported among women was 165 women out of 2000, respectively. On the other hand, our finding aligned the data reported by Singh et al. which showed that the prevalence of CD was higher among women compared to men. 31 In addition, according to Bai et al., there are physiological differentials among genders in CD 32 : "men have indirect evidence of greater malabsorption than females and have female-predominant associated diseases when they present with celiac disease". 32 This study supported our finding concerning the remarked protein-energy deficiency among males compared to females in our study. Further, almost all the participants had symptoms related to CD such as abdominal bloating, diarrhea, headache, and flatulence; around 66% were facing moderate to severe symptoms and 78% of them reported being on a medical treatment for CD symptoms. These findings were two times higher than the prevalence reported by Stasi et al., who showed that 22% of individuals with CD faced persistent abdominal symptoms. 33 In the present study, 14% presented hypoalbuminemia and mild to moderate malnutrition (10%-30% weight loss in the previous six months). This data agreed with Wierdsma et al. 34 where 17% of individuals with CD were malnourished (>10% undesired weight loss). In general, patients with CD are misdiagnosed as malnourished and also their malnutrition is always underreported. Thus, physicians tend to ignore mild to moderate weight losses. This can aggravate the nutrition status of these patients and decrease their immunity. On the other hand, the percentage of underweight women (5%) in our study wasn't in accordance with Wierdsma et al. 34 in which 22% of the women were underweight and 29% of the patients were overweight (BMI > 25 kg/m 2 ). Our study showed that around 40% of individuals with CD had low muscle mass and 60% had high fat mass. This might be due to the low physical activity level and unhealthy food patterns. Also, some of the  medications ingested by the patients with CD can induce insulin resistance and promote overweightness. This result was in contradiction with the results reported in a national study which showed that subjects with CD had a lower fat mass (17.8AE2.0 kg) 35 but in concordance with two recent systematic reviews on children and adults that showed that after one year follow-up to GFD' adherence, fat mass of CD patients significantly increased compared to that at baseline. 36,37 According to Agarwal et al., patients with CD are more susceptible to develop metabolic syndromes and fatty liver diseases. 38 Thus, these patients should be regularly counseled by experienced dietitians for nutritional and metabolic factors about healthy diets and physical activity. However, the majority of participants in our study were physically inactive (72%). This prevalence of sedentary patterns among our study population was higher than that reported by Nestares et al. (37%) among children and adolescents 39 but lower than the prevalence of physical activity among women with CD in Poland (70%). 40 Out of 50 participants, 38% and 16% were presenting low serum levels of iron and vitamin B12, respectively. None of the study participants presented folate deficiency. These results showed consistency with a recent systematic review that showed high occurrence of anemia among individuals with CD. 41 Furthermore, micronutrient deficiencies are common features among individuals with CD. This was demonstrated in the study of Bledsoe et al. 19 where weight loss was seen in 25.2%, albumin was low in 19.7%, vitamin B 12 was low in 5.3% and folate was low in 3.6% of cases. Moreover, it was shown that 87% of patients with CD in the study of Wierdsma, et al. 34 had low vitamin B 6 (14.5%), folic acid (20%), and vitamin B 12 (19%). Additionally, 46% of patients with CD had low serum iron, and 32% had anemia. 34 This might be due to the lack of knowledge on the importance of micronutrients intake among patients with CD and also due to abdominal distress symptoms that are common in these patients.
In our study, around 80% of participants read nutrition labels when shopping groceries. Reading and understanding food label's topics have been discussed widely and have been associated with strict GFD adherence. 42 Our finding was in accordance with Butterworth et al., who showed that amongst 74% of individuals with CD, some factors were in compliance with adherence to GFD: being a member of a celiac society, understanding food label's language, and having access to gluten-free products with reasonable prices, regular dietetic and medical follow-up. 43 Causes of adherence/nonadherence to a GFD are multifactorial and differed according to their nature and magnitude. 44-47 These factors are very correlated with the education level, the economic situation and the readiness of patients to adhere to this diet. 46,47 Almost 96% of our population were following a GFD. To assess their adherence to GFD, the examination of the participant's three-day records reported low intake of gluten (<3g per day) among 96% of the study participants. Only 4% weren't adhering to the GFD recommendations. This finding was consistent with a study conducted in individuals with CD in Spain (92.5%). 48 According to a Spanish study, 72% of individuals with CD showed acceptable adherence which was associated with higher levels of self-efficacy. 49 Three studies from the UK showed that white patients were adhering to GFD in a range from 53 to 81%. 50 Non-adherence to GFD was also associated with diagnosis at younger ages and smoking. 50 This was not shown in our study. In Canada, GFD adherence was assessed and revealed that the percentage of strict adherence was only 56%. 51 The assessment of GFD adherence of 70 Swedish adolescents with CD showed that 86% of them were adherent to GFD five years after screening. 52 Moreover, of 5310 adult and adolescent Australians and New Zealanders with CD, 61% ashered to a GFD. 53 Three studies on adolescents from Brazil identified adherence rates from 36 to 86% taking into consideration that younger children (up to 12 years) were more likely to comply with the diet 54 while teenagers interrupted their diet purposefully. Parents of pediatric patients with CD from the Slovak Republic, reported a GFD adherence of 69% by children. 55 In addition, among 38 studies, the adherence was ranging from 42 and 91%. 49 According to a recent systematic review of 49 studies, the adherence rates were ranging from 23 to 98%. 56 This wide variability in adherence rates may be explained by the different populations examined (e.g., adults, adolescents, children, ethnic minorities), but also by the different tools used for evaluating adherence.
As reported by the participants, some barriers including family ignorance (among 6% of the population), language on the nutrition labels (among 20% of the population) and expensive GF products (among 78% of the population) were limiting the adherence to GFD. These findings were in concordance with Muhammad et al., 44 who showed that not understanding food labelling was associated with lower GF dietary adherence scores. A higher proportion of individuals in the United Kingdom (UK) reported difficulties in acquiring knowledge about GF products (ranging between 5% and 76%) and understanding of food labels (4%-53%). 57 It was observed that 72% of the study population had good cooking experience and 64% of the study participants prepared their meals. On the other hand, almost all the subjects were unaware of a possible cross-contamination during cooking food. Despite 82% of the subjects with CD preferring fast foods, they frequently consumed their meals at home (92%). Beyond cooking meals at home and cross-contamination, our findings were consistent with the findings reported in a recent systematic review indicating that the most significant barriers limiting the adherence to GFD included: "lower knowledge of CD" (35%); "restaurant/supermarket shopping" (30%); "poor patient education from practitioner" (17.5%); and "low intention/motivation to adhere to a GFD (17.5%)". 45 In our study, patients with CD had an inadequate total daily energy intake. This finding was concordant with several studies from Poland, Germany and Spain. 58-60 As for nutrients intake, among females and males, none of the study participants showed adequacy to the daily recommendations for calcium and vitamin D intake. These results, altogether, showed an unbalanced diet in terms of micronutrients, that were in accordance with those shown in several studies. 14,[61][62][63] On the other hand, almost all the study participants from both genders showed a high daily consumption of iron. This finding contradicts the results reported in other studies. 60,[64][65][66] In general, patients with CD are always alarmed by their physicians concerning anemia. Thus, these patients tend to consume more food sources of iron and ignore other nutrients.
As for dietary supplement use, it appears that the usage of DS among individuals with CD wasn't in excess of that reported in the general population in Lebanon. 67 In our study, the most commonly used DS was vitamin D and the least used one was probiotics. Despite that most patients with CD are at risk of micronutrient deficiencies, however, the lack of knowledge concerning dietary supplement use is one of the challenges related to malnutrition among these patients. 68 This can explain the occurrence and persistence of gastrointestinal manifestations among the majority of participants in the current study.

Strength and limitations
The crucial strength of this study is that it fulfills a comprehensive assessment of dietary intake and food-related patterns, biochemical parameters, body composition, and physical activity. However, this study had some limitations: 1) the selected group of patients doesn't represent the general population with CD in Lebanon; 2) it was a cross-sectional study and, thereby, our findings must be interpreted with caution; 3), physical activity levels were self-reported; 4) there is an important limitation of data on caloric, macro-and micronutrient composition of the recalled three-day record. The low sample size must also be considered.

Conclusion
Monitoring nutritional status using dietary, anthropometric, blood tests and assessing physical activity are the key components in the management of CD. The ideal GFD should be balanced with macronutrients and micronutrients, with reasonable access and prices. Fortification/enrichment of GF foods that are frequently consumed should be encouraged to avoid deficiencies. This underlines the critical role of dietitians in education and maintenance of healthy GFD among individuals with CD.

Results:
In data with a statistically significant difference, please indicate under tables that: 1) Which test did you use and 2) The cut of point of significant P value (like p<0,05 or 0,001).

Discussion:
The discussion part contains multiple repeats of the results. On the other hand, The authors haven't interpreted the result (Like How, and why ..). For example: What are the factors that influenced adherence to GFD in Lebanon? Please after comparing the results with other studies, also give original comments related to your data.

○
Please first write the whole explanation of this term "P-E malnutrition" (Protein-energy malnutrition) ○ "In our study, the most commonly used DS was vitamin D and the least used one was prebiotics" The authors mentioned prebiotics, I think it was written by mistake. Please clarify this term whether probiotic or prebiotic? ○ Villous Atrophy takes place in the small bowel only, no need to specify its location small bowel. Alternatively, the authors may write, partial to total villous atrophy. (First line of the main introduction) ○ The first line of the Selection procedure contains results (February to July 2018). Please shift this in the results section.

Is the work clearly and accurately presented and does it cite the current literature? Partly
Is the study design appropriate and is the work technically sound? Partly their knees and back straight while grasping a handle that also includes electrodes with both their hands horizontally raised, elbows extended straight, and maintaining a 90°-angle to the body. A previous study that compared body composition estimates using BIA devices with DXA and whole body magnetic resonance imaging indicated that the use of devices with additional hand electrodes provides a more accurate prediction of body composition and are suitable for public use. In order to determine percentage body fat, the device uses electrical impedance, along with the participant's height, weight, age, and gender to generate results. The readings were obtained in duplicates and the average recorded. According to the manufacturer's instructions, percentage body fat was measured two hours or more after breakfast. Omron Healthcare references for age and gender were used to classify the study participants into low, normal, high, and very high %BF.
We followed these articles: A. The three centers were using same equipment and no issues were presented in the laboratory results.
○ ○ "Food-related behaviors and GFD dietary adherence: "Across the three-day food records, patients presenting a consumption between 3 to 10 g of gluten per day were considered as non-adherent" In this line, please explain how you calculated gluten intake from the diet record? Which method did you use?" For dietary compliance, patients were accurately interviewed by a trained physician and dietitian to assess the habitual intake of serving of glutencontaining food per week (e.g. pasta, pizza, bread, or cookies). Compliance to gluten-free diet was defined as follows: very low = more than one serving of gluten containing food/week; low = up to one serving of gluten-containing food/week; good = no gluten-containing food/week. Then based on the three days record the g/d data of gluten was investigated. We followed the same methodologies adopted in the below 2 studies: