Serum Vitamins and Minerals at Diagnosis and Follow-up in Children With Celiac Disease

Abstract

Objectives: 

Children with celiac disease (CD) may experience deficiencies of several micronutrients. The objectives of the present study were to determine the prevalence of micronutrient deficiencies in children with CD at diagnosis, 6 months, and 18 months after the start of a gluten-free diet (GFD), and examine any correlation between micronutrient deficiencies, serum tissue transglutaminase (TtG) immunoglobulin A (IgA) antibody titers, and the degree of mucosal damage at diagnosis.

Methods: 

Children (<17 years) with CD had their serum vitamins, minerals, and anti-TtG IgA antibodies measured at diagnosis, 6 and 18 months after starting a GFD. Histopathological changes of duodenal biopsies at diagnosis were documented using modified MARSH classification.

Results: 

The medical records of 140 children (mean age at diagnosis 7.8 ± 4.01 years, 87 girls [621%]) with CD were examined. At diagnosis, serum vitamin D was the most commonly deficient vitamin in 70% of children. Serum ferritin was subnormal in 34.5% with zinc in 18.6% children but only 12 (10.9%) children had iron deficiency anemia. There was no correlation between micronutrient deficiencies at diagnosis and serum TtG IgA antibody titers or the degree of villous atrophy. The majority of serum levels of measured micronutrients had normalized after 6 months of starting GFD except for vitamin D, which improved but remained subnormal.

Conclusions: 

At diagnosis, most children with CD have vitamin D deficiency. The degree of micronutrient deficiencies does not correlate with the degree of villous atrophy or serum titers of anti-TtG IgA antibodies.

What Is Known

• Untreated celiac disease may lead to vitamin and mineral deficiencies.
• Screening for vitamin and mineral deficiencies at diagnosis has been recommended.

What Is New

• In children with celiac disease, vitamin D is the most common deficient vitamin at diagnosis and should be checked as a part of the annual assessment for these children.
• Vitamin and mineral deficiencies at diagnosis do not correlate with the degree of villous atrophy or serum levels of antitissue transglutaminase immunoglobulin A antibody titers.
• Once normalized, and with adherence to a gluten-free diet, the majority of vitamins and minerals remain within normal ranges at 18 months after diagnosis.

Celiac disease (CD) is an immune-mediated disorder triggered by ingestion of gluten in genetically susceptible individuals and results in inflammatory changes of the small intestine ^(1–6). Genetic and environmental backgrounds are one of the important factors affecting the etiology of the disease ^(1,5).

Many studies from around the world indicated that the prevalence of CD in children younger than 18 years is 3 to 13 per 1000 children, although the number may be higher as many asymptomatic children remain undiagnosed ⁽¹⁾. Children with CD may manifest with several gastrointestinal symptoms such as abdominal pain, vomiting, abdominal distension, and diarrhea. Many nongastrointestinal symptoms such as dermatitis herpetiformis, dental enamel hypoplasia of permanent teeth, osteoporosis, abnormal growth, delayed puberty, and anemia may also warrant testing for CD ⁽²⁾. Unusual presentations such as pyrexia of unknown origin were also reported ^(7,8).

Antitissue transglutaminase (TtG) antibodies have been reliably used as a serological marker in the diagnostic work-up for CD but endoscopy and duodenal biopsies ⁽⁹⁾.

A gluten-free diet (GFD) for life is the only treatment for CD in clinical practice ^(2,6,9). Treatment should be started after confirmation of diagnosis via intestinal biopsies. Children should be periodically evaluated for presence of any gastrointestinal/celiac-related extraintestinal symptoms, growth and development, and adherence to GFD ^(2,6,9). Measurements of serum TtG immunoglobulin A (IgA) antibodies at an interval of 1 year or longer in asymptomatic children can be performed to check for adherence to GFD ^(2,6).

Micronutrient deficiencies are common complications of CD. Damage to the small intestine villi due to chronic inflammation often leads to malabsorption of various micronutrients. Individuals with untreated CD have micronutrient deficiencies more commonly than those with no disease, including deficiencies of iron; folic acid; vitamins B₆, B₁₂, and D; copper; zinc; and carnitine ⁽¹⁾.

Wild et al ⁽¹⁰⁾ concluded that iron, zinc, selenium, folate, magnesium, and manganese were commonly deficient in 139 consecutive celiac patients. Mariana et al ⁽¹¹⁾ revealed that iron, fiber, and calcium were deficient in 47 adolescent children with CD. In another study by Ohlund et al ⁽¹²⁾, magnesium, vitamin D, and selenium were deficient in 30 children with CD. Shephard and Gibson ⁽¹³⁾ demonstrated that patients with CD had significantly lower levels of folate, vitamin A, magnesium, calcium, iron, and zinc.

Current guidelines suggest screening for micronutrient deficiencies at diagnosis. Folate, vitamin B₁₂ and D, serum iron, zinc, selenium, calcium, and magnesium were commonly deficient in celiac patients ^(10–13). The prevalence of micronutrient deficiencies at different time intervals in children with CD is, however, currently underinvestigated. The need for annual measurements for those micronutrients, which are costly, is debatable.

The objectives of the present study were to examine the prevalence of micronutrient deficiencies at diagnosis, 6 and 18 months following the start of GFD, and to determine any possible correlation between micronutrient deficiencies, serum TtG IgA antibody titers, and the degree of mucosal damage at diagnosis.

METHODS

Study Population

In a single-center cohort study, the medical records of all consecutive children with CD seen in the Celiac Outpatient Clinic at the Children's Hospital, Health Sciences Center, Winnipeg, Manitoba, Canada between June 2012 and March 2016 were examined. Routine celiac blood panel included complete blood count, serum albumin, liver function tests, thyroid function tests, serum levels of C-reactive protein (CRP), IgA level, anti-TtG IgA antibody titers, iron, ferritin, zinc, selenium, copper, red blood cell folate, and vitamins A, D, E, and B₁₂. Blood tests were performed at diagnosis, 6 months after starting GFD, and then on an annual basis. All participants had an upper gastrointestinal endoscopy and mucosal biopsy specimens taken (4 biopsy specimens from the second part of the duodenum and 2 from the duodenal bulb) as per the North American and European Societies of Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN and ESPGHAN) guidelines ^(2,9). Histopathological classification of duodenal changes was performed after the modified MARSH criteria ⁽¹⁴⁾. Dietetic assessment with proper teaching of GFD took place at diagnosis, 6 months, and 18 months after diagnosis. Nutritional status, the need for micronutrient supplementations, and adherence to GFD were assessed by an experienced dietitian in each clinic visit.

Data Collection

Data were collected from participants’ medical records including participants’ demographics, postal code as an indication for socioeconomic status, presenting symptoms and signs, particularly those of vitamin and mineral deficiencies, laboratory values at diagnosis, 6 months, and 18 months after the diagnosis, season of blood sample collection for serum vitamin D levels, histopathological findings of the duodenal biopsy specimens, the degree of adherence to GFD, and any intake of vitamin and mineral supplements.

The details regarding methods of measuring serum vitamins and minerals were provided in Supplemental Digital Content, https://links.lww.com/MPG/A852.

Statistical Analysis

Calculations and data analysis were performed using Statistical Package for the Social Sciences (IBM SPSS Statistics for Windows, Version 22.0, Armonk, NY). Univariate summaries (means, ranges, and standard deviation) were calculated for the study group for continuous variables, whereas frequencies were calculated for categorical variables, along with 95% confidence intervals for the means and proportions. Variables were examined for normal distribution. Nominal variable comparisons were performed using Fisher exact test. Analysis of variance was used where appropriate. Univariate regression analysis and Pearson correlation were performed to examine any possible association between micronutrient deficiencies, anti-TtG titers and duodenal histopathology after adjusting for adherence to GFD, CRP, and socioeconomic status. A P value <0.05 was considered significant.

Ethical Considerations

The study protocol was approved by the University of Manitoba Health Research Ethics Board (HS17692).

RESULTS

The medical records of 140 children (mean age of 7.8 ± 4.01 years, 87 girls [62%]) consecutively diagnosed with CD were reviewed. The mean duration of follow-up was 3.01 ± 0.4 years. All major presenting symptoms are summarized in Table 1. Fifty-seven participants (40.7%) had positive family history for a first-degree relative with CD. All participants had their diagnosis confirmed via endoscopy and duodenal biopsies. Histopathological findings of duodenal biopsies included complete villous atrophy (MARSH IIIC) in 60 participants (42.8%) and 63 participants (45%) had partial villous atrophy (MARSH IIIa and b). The rest of participants (17 [12.1%] children) were MARSH I-II and seropositive for CD. The demographic, serological, and histopathological features are summarized in Table 2

TABLE 1: Main presenting symptoms in 140 children with celiac disease

TABLE 2: Demographic, serological, and histopathological features of 140 children with celiac disease

Six months after the diagnosis, 20 children (14.3%) were classified as noncompliant with a GFD but only 3 (2.1%) were noncompliant after 18 months from diagnosis. The number of children that required supplementation for deficiency versus improvement with GFD alone are summarized in Table 3.

TABLE 3: Number of children who required supplementations for vitamin and mineral deficiencies and those who improvement on gluten-free diet alone

The mean z scores for height, weight, and body mass index were calculated. There was no significant difference in the mean values of z scores of height (at diagnosis: –0.32 ± 0.98, after 6 months: –0.31 ± 0.95, after 18 months: –0.29 ± 0.87), weight (at diagnosis: –0.28 ± 1.02, after 6 months: –0.16 ± 0.98, after 18 months: –0.21 ± 0.94) or body mass index (at diagnosis: 0.33 ± 1.10, after 6 months: 0.25 ± 0.94, after 18 months: 0.14 ± 0.93) at different time intervals.

At diagnosis, only 9 (6.4%) participants had skin pallor and 2 (1.4%) children had clinical signs of rickets including frontal bossing and broad epiphyses.

Serum vitamin D was the most common deficient vitamin in 43 out of 62 (69.4%) children with a mean value of 55.5 ± 13.6 nmol/L (P < 0.05). Plasma ferritin (9.9 ± 5.2 μg/L, P < 0.05) was subnormal in 31 out of 90 (34.4%) children with subnormal plasma iron (4.9 ± 1.2, P < 0.05) in 11 out of 96 (11.5%) children who had iron deficiency anemia (hemoglobin below reference value for age and sex). Vitamins A and E and selenium were deficient in 11.1%, 6.8%, and 2.3% of participants, respectively. Five participants (7.7%) had vitamin B₁₂ deficiency with a mean value of 68.4 ± 77.0 (P < 0.05). Zinc and red blood cell folate were subnormal in 9 (18.6%) and 3 (7.5%) participants, respectively.

Micronutrients that were normal at diagnosis (selenium and copper) remained normal at 6 and 18 months postdiagnosis. With the exception of vitamin D, all serum/plasma values of measured micronutrients that had normalized after 6 months of starting GFD, remained normal at 18 months after diagnosis. The number of children with deficient micronutrients at diagnosis, 6 months, and 18 months after diagnosis are summarized in Table 4. The number of children with micronutrient deficiencies declined significantly from diagnosis to 18 months after diagnosis.

TABLE 4: Number of children with celiac disease and trace element and vitamin deficiencies at diagnosis, 6 months, and 18 months after diagnosis

At diagnosis, there was no correlation between micronutrient deficiencies and serum levels of TtG IgA antibody titers or the degree of villous atrophy.

DISCUSSION

CD is a relatively common chronic immune-mediated enteropathy that affects approximately 1% of the population ^(1,2). The immune dysregulation associated with CD as a result of gluten exposure causes chronic inflammation and destruction of the small bowel intestinal villi, resulting in nutrient malabsorption ^(1,15). Currently, a GFD is the only well-established treatment in clinical practice once the diagnosis is confirmed ^(2,6,15).

The American College of Gastroenterology clinical guidelines recommends that all individuals newly diagnosed with CD should be tested and treated for micronutrient deficiencies ⁽⁶⁾. Interestingly, in a Canadian survey that looked at different practices among Canadian pediatric and adult gastroenterologists for patients with CD, 41% of gastroenterologists never measured serum vitamin E levels, 28% never monitored vitamin D levels, and 33% never monitored serum vitamin A levels ⁽³⁾.

In the celiac clinic at Children's Hospital, Winnipeg, Manitoba, Canada, children are usually seen at diagnosis, 6 months after diagnosis, and then annually as part of our protocols. We monitor their symptoms, growth, adherence to GFD, and we also order several blood investigations including serum/plasma levels of several micronutrients. In our study, vitamin D deficiency in patients with CD was 70%, which is significantly higher than the reported figures in normal children of Canadian prairies. Previous studies from Canadian Prairies highlighted that the prevalence of vitamin D deficiency in normal children was 40% to 43% ^(16,17).

In the present study, approximately one third had low plasma ferritin at diagnosis. Approximately 20.5% children had abnormal plasma ferritin after 18 months of diagnosis. According to Statistics Canada, approximately 4% of healthy Canadians experience plasma ferritin deficiency ⁽¹⁸⁾. Vitamin B₁₂, vitamin A, and iron were deficient in 7.7%, 11.1%, and 11.5%, respectively in the present study. Healthy children rarely have vitamin B₁₂ deficiency (1.2%) ⁽¹⁹⁾.

Zinc is subnormal in 10% to 25% of Canadian adolescents ⁽²⁰⁾. Health Canada has indicated that 10% to 35% of Canadians of almost all age and sex groups have inadequate consumption of zinc ⁽²¹⁾. According to the present study, 9 (18.6%) patients had subnormal serum zinc level at diagnosis. It remained suboptimal in 18.2% children after 18 months after GFD.

There was no correlation between the degree of villous atrophy and the micronutrient deficiencies for all measured micronutrients. The majority of micronutrients improved 6 months after diagnosis and continued to improve 18 months after the start of GFD. Despite high prevalence of micronutrient deficiencies, only a minor percentage of participants had clinical signs secondary to their micronutrient deficiencies.

In a recent study from Turkey, Topal et al ⁽²²⁾ reported micronutrient deficiencies in 52 children newly diagnosed with CD. Similar to our results, vitamin D and iron were deficient in 52% and 34% of children, respectively. Zinc was, however, the most commonly deficient micronutrient in 67% of participants, whereas vitamins A and E were subnormal in 7% and 13% of included children, respectively ⁽²²⁾. These results concur with another adult study, which showed that 67% of adult patients newly diagnosed with CD had at least 1 vitamin or micronutrient deficiency at diagnosis ⁽²³⁾. Vitamin D deficiency was indeed common in patients newly diagnosed with CD ⁽²⁴⁾. In another study by Botero-Lopez et al ⁽⁵⁾, iron was deficient in 65% of participants, followed by zinc (20%) and copper (15%). There was no difference between those with typical CD compared to those with atypical disease in micronutrient deficiencies.

In our study, all participants with initial normal micronutrients and good adherence to GFD continued to have normal serum levels of micronutrients 6 and 18 months after diagnosis indicating that monitoring micronutrients in these children in follow-up is unnecessary. Moreover, children who started with suboptimal levels of 1 or more micronutrients at diagnosis but were prescribed appropriate supplementations with good adherence to GFD following the diagnosis ended up normalizing their serum/plasma levels of micronutrients at 6 to 18 months after diagnosis. It was observed that children who were not adherent to GFD had lower levels of micronutrients compared with those with good adherence to GFD but the difference was not statistically significant for the majority of micronutrients. The only 2 micronutrients that remained suboptimal 18 months after diagnosis were vitamin D and ferritin, indicating that long-term monitoring for vitamin D and ferritin is important in children with CD.

Interestingly GFD itself could be poor in some micronutrients especially in vitamin D, vitamin B₁₂, and folate ⁽²⁵⁾, which may explain the persistence of vitamin D deficiency in a fraction of our patients at 18 months after diagnosis. Those who were, however, vitamin B₁₂ and folate sufficient at diagnosis, remained sufficient on GFD, 6 and 18 months after diagnosis. Moreover, those who were vitamin B₁₂ deficient at diagnosis but normalized 6 months after starting on GFD and vitamin B₁₂ supplementation, remained vitamin B₁₂ sufficient at 18 months after diagnosis despite stopping vitamin B₁₂ supplementation. This can also be attributed to the sequestration of these micronutrients in the body (vitamin B₁₂ in the liver and, folate in red cells) and their long half-lives ^(26,27).

Our study does have several strong points. It is one of the few studies that looked not only at the initial values of micronutrients at diagnosis but also at 2 time points after diagnosis. It also looked at several micronutrients including selenium and copper with simultaneous measurement and adjustment for inflammatory markers such as CRP as any ongoing inflammation may affect the serum levels of some measured micronutrients.

In contrast, the study is limited by its small sample size, retrospective design, and the variable number of participants who had their micronutrients measured at the 3 time points of the study. Measuring vitamin D can also be affected by other factors including time of the year and ethnic background (skin color). Finally, the study results should be perceived in the setting of lack of a normal control group. The study, however, confirms and adds to what is already known about micronutrient deficiencies in a common disease especially with the lack of correlation between micronutrient deficiencies and degree of CD-induced mucosal damage. Larger properly designed controlled prospective long-term studies are needed to confirm our results. Correlation between severity of disease and micronutrient levels is also an important aspect to ponder. Because the answer to it is beyond the scope of the present study, further study is required.

CONCLUSIONS

At diagnosis, the majority of children with CD have vitamin D deficiency. The degree of micronutrient deficiency in children with CD does not correlate with the degree of villous atrophy or the serum titers of anti-TtG IgA antibodies. The majority of subnormal serum/plasma micronutrient levels normalized 6 months after starting children on GFD and remained normal 12 months later. Annual monitoring of serum micronutrient levels should be reserved for those micronutrients that remain deficient or in those patients who are not adherent to GFD. Long-term properly designed prospective controlled large-scale studies are needed to confirm our conclusions.