K.G. postgraduate studies were funded by the Greek State Scholarship Foundation, the Hellenic Society of Gastroenterology and Nutrition and the University of Glasgow. No conflicts of interest to declare.
Impact of exclusive enteral nutrition on body composition and circulating micronutrients in plasma and erythrocytes of children with active Crohn's disease†
Article first published online: 8 NOV 2011
Copyright © 2011 Crohn's & Colitis Foundation of America, Inc.
Inflammatory Bowel Diseases
Volume 18, Issue 9, pages 1672–1681, September 2012
How to Cite
Gerasimidis, K., Talwar, D., Duncan, A., Moyes, P., Buchanan, E., Hassan, K., O'Reilly, D., McGrogan, P. and Ann Edwards, C. (2012), Impact of exclusive enteral nutrition on body composition and circulating micronutrients in plasma and erythrocytes of children with active Crohn's disease. Inflamm Bowel Dis, 18: 1672–1681. doi: 10.1002/ibd.21916
- Issue published online: 9 AUG 2012
- Article first published online: 8 NOV 2011
- Manuscript Accepted: 14 SEP 2011
- Manuscript Received: 12 SEP 2011
- enteral nutrition;
- Crohn's disease;
- body composition
- Top of page
- MATERIALS AND METHODS
- Supporting Information
Nutritional therapy is the primary treatment for active pediatric Crohn's disease (CD) in the UK/Europe, improving disease activity and anthropometry. This study assessed changes in micronutrient status during exclusive enteral nutrition (EEN).
Seventeen children (male/female: 8/9; median age: 12.7 years) with active CD were treated exclusively for 6–8 weeks on a polymeric feed (Modulen IBD; Nestle, UK). Body impedance was measured at baseline, during EEN, and posttreatment on normal diet and converted to z-scores of fat and lean mass. Blood samples for nutrient analysis were collected from 13 children at baseline, end of EEN, and posttreatment.
Lean but not fat mass improved at the end of EEN (initiation vs. end of EEN; fat mass [z-score]: −0.5 vs. −0.3; P = 0.141; lean mass [z-score]: −2.1 vs. −0.8; P < 0.0001). At baseline several children presented with suboptimal concentrations of carotenoids, trace elements, vitamin C, B6, and folate in plasma but not in erythrocytes. EEN improved concentrations for several nutrients, but more than 90% of patients had depleted concentrations of all carotenoids. The latter improved on normal diet but other micronutrients, which improved during EEN, returned toward pretreatment concentrations.
Lean but not fat mass improved at the end of EEN. Median concentrations for several plasma micronutrients improved on EEN but carotenoids were depleted. These findings may have implications for clinical practice and producers of enteral feeds. As plasma concentrations for many micronutrients can be affected by the acute phase response, measurements in erythrocytes may be a better marker of actual body stores. (Inflamm Bowel Dis 2012;)
Crohn's disease (CD) is a chronic inflammatory condition of the gastrointestinal (GI) tract with a continuous remitting and relapsing course.1 About 20% of patients are diagnosed during childhood, with an incidence of 3.1 per 100,000 in the UK.2 The exact pathogenesis remains unknown and current treatment options provide mainly symptomatic relief. Exclusive enteral nutrition (EEN) is the mainstream treatment of active pediatric CD in the UK and Europe, although it is rarely used in North America.3 It induces clinical remission in the majority of cases and provides nutritional support to the patient.4
Previous research and clinical experience have shown that EEN improves anthropometric indices in the CD child,4 but there are no data to confirm that the prescribed dietary regimen is nutritionally complete, adequately provides the micronutrient requirements of CD children, or corrects any preexisting nutrient deficits. As most commercially available feeds are designed to provide the national dietary recommendations for micronutrients based on healthy children, their adequacy in CD patients is uncertain. Altered metabolism,5 malabsorption,5, 6 and excessive utilization7 have been reported in CD patients,8 perhaps increasing the nutritional requirements compared to those established for healthy children. Abad-Lacruz et al,9 20 years ago, failed to show improvement in the vitamin status of inflammatory bowel disease (IBD) adults on EEN despite provision of micronutrients in doses significantly higher than the recommended dietary allowance. Moreover, as nutritional feeds are artificial composite foods made up of micronutrients, most of which have established recommended allowances, they may be insufficient in other important nutrients, particularly those that occur naturally in food but for which no recommended intakes have been established.
Because extrapolation of results from adult studies to pediatric patients is inappropriate and there is no relevant evidence in children, this study had two aims. The first was to assess the effect of EEN on body composition and micronutrient status by measuring their concentrations in the blood of CD children. As plasma micronutrient concentrations can be influenced by the acute phase response in inflammatory conditions independently of tissue stores,10, 11 the second aim of this study was to assess the changes in red blood cells (RBC) of selected vitamins and trace elements as a potentially more reliable index of actual body stores in the presence of systemic inflammatory response and to relate changes in plasma and RBC micronutrients to inflammatory status.
MATERIALS AND METHODS
- Top of page
- MATERIALS AND METHODS
- Supporting Information
Subjects and Study Design
Eligible participants for the study were newly diagnosed children (<16 years) with CD or patients with longstanding disease in clinical relapse who started treatment on EEN as part of their standard clinical management.
Exclusive Enteral Nutrition
In all cases, patients followed a 6–8-week course of EEN using a polymeric casein-based liquid nutritionally complete feed enriched with transforming growth factor-β (Modulen, Nestle, UK). The feeds provided the UK recommended daily energy requirements for age and gender12 favoring a high protein intake (Online Supporting Material 1). Children who appeared malnourished, based on clinical dietetic assessment, were given a higher energy intake (≈10%–20% estimated average requirements) (Online Supporting Material 1). The feeds (≈2000 mL) were administered either orally, via a fine bore nasogastric tube, or by percutaneous endoscopic gastrostomy. During the EEN course all participants received micronutrients that were ≈1.8–6 times more than their UK Reference Nutritional Intakes (Online Supporting Material 1). No other foods were permitted during EEN treatment with the exception of water, tea, coffee, lemonade (7up), and clear mints. Compliance with EEN was checked by regular dietetic review. At the end of EEN, a week-long step-by-step food reintroduction program was followed before the patient returned to their normal diet. On completion of EEN, patients were asked to continue supplementary feeds (≈500 kcal/d) for a period of 2 weeks in addition to their normal diet.
Markers of Disease Activity
Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were measured by standard laboratory procedures and data were available from the patients' routine clinical measurements. CRP was measured using a turbidometric assay after binding to a specific antibody on the Architect analyzer. For CRP the limit of detection was 7 mg/L. The interassay coefficient of variation (CV) was less than 5% over the sample concentration range for CRP.
The day after the routine clinical appointment of each patient, a multidisciplinary health profession team (comprised of pediatric gastroenterologists, gastroenterology specialist dietitians, and IBD specialist nurses) met, reviewed, and discussed the disease progress of each patient who attended the outpatient IBD clinic (including the patients who participated in the study). The disease status of each patient was classified by consensus of the team as 1) in clinical remission, 2) some symptoms of active disease, 3) active disease/relapse. No member of the multidisciplinary health profession team was aware of the participation of the individual patients in the study.
Anthropometry and Body Composition
Measurements of body weight, height, and foot to foot bioelectrical impedance (TANITA 300M, Japan) were performed at initiation of EEN, after 1 month, at the end of EEN (after 6–8 weeks), and when patients were established on their normal diet (within 2–4 months of EEN cessation).
Collection and Preparation of Blood Samples
Venous blood samples were obtained on treatment initiation, end of EEN, and when patients were established on their normal diet (within 2–4 months of EEN cessation). As it was deemed unethical to venipuncture children for entirely research purposes, blood samples were drawn only as extra volume to the clinical monitoring samples of the patient. As patients were on strictly prescribed regimens, and it was inappropriate to interfere with their standard clinical management, fasting measurements were not always feasible. However, in all cases blood samples were collected ≈3 hours after the last meal.
Venous blood was collected in trace element tubes (sodium heparin Vacuette, 456080, Austria). Samples were centrifuged (3500g for 15 minutes at 4°C) and plasma was removed and packed red cells prepared by carefully removing all remaining plasma and buffy coat and stored at −70°C. Samples were analyzed within 6 months. Samples from individual patients were assayed in the same batch to minimize analytical variation.
In total, 17 micronutrients (vitamins and trace elements) were measured in plasma, whole blood, or erythrocytes. The micronutrient status of circulating micronutrients was evaluated by reference to the laboratory reference intervals (Online Supporting Material 2). Pediatric reference intervals were available for copper, zinc, selenium, and vitamin A in plasma (Online Supporting Material 2). In the absence of appropriate pediatric reference intervals for the remaining micronutrients, our adult reference intervals were used (Online Supporting Material 2).
Plasma vitamin A (retinol), E (α-tocopherol), and carotenoids (lutein, lycopene, α- and β-carotene) were determined by high-performance liquid chromatography (HPLC) as previously described.13 The intraassay CV was less than 9% for all analytes over the sample concentration range. Plasma vitamin E was corrected for serum total cholesterol.
Vitamin B1 (thiamin diphosphate; TDP) is present almost exclusively in erythrocytes and so vitamin B1 status was assessed by measuring TDP in whole blood. An HPLC system with postcolumn ferricyanide derivatization and fluorometric detection was used as described previously.14 The TDP concentration in whole blood was related to hemoglobin (Hb) in the sample (ng TDP/g Hb). The within-batch imprecision was 5.1% at 380 ng/gHb.
Vitamin B2 (flavin adenine dinucleotide; FAD) measurement in whole blood and erythrocytes was based on the method of Speek et al.15 Briefly, whole blood or diluted red cell hemolysate were precipitated with methanol and centrifuged and the supernatant was injected for HPLC analysis. FAD was separated on an isocratic HPLC system with a reversed-phase C18 column and fluorescence detection. The within-batch imprecision for whole blood FAD was 4.8% at 384 nmol/L and 4.8% at 2.8 nmol/g Hb red cell FAD.
Vitamin B6 (pyridoxyl phosphate; PLP) concentrations in plasma and red cells were measured by HPLC using precolumn semicarbazide derivatization and fluorescent detection as described previously.16 PLP concentrations in red cells were adjusted to hemoglobin (Hb) rather than to the volume of packed red cells because accurate pipetting of packed red cells is difficult, due to high viscosity. The within-batch imprecision for plasma PLP was 4.9% at 59 nmol/L and 6.3% at 16 nmol/L, and was 5.2% at 367 pmol/g Hb for red cell PLP.
Vitamin C status was assessed by measuring ascorbic acid in plasma. The method was based on that of Margolis and Davis.17 Briefly, plasma stabilized and deproteinized with 60 g/L metaphosphoric acid was centrifuged and an aliquot of supernatant injected on a C18 reversed-phase analytical column. After separation, ascorbic acid was determined by electrochemical detection. The within-batch imprecision for the assay was 3.7% at a concentration of 38 μmol/L.
Vitamin D status was assessed by measuring 25-hydroxy vitamin D with an enzyme immunoassay kit (Immunodiagnostic Systems, Boldon, UK).
Inductively coupled plasma mass spectrometry (Agilent Technologies, Cheadle, UK) was used to measure plasma zinc, copper, and selenium and red cell copper, zinc, selenium, magnesium, and iron using germanium as an internal standard. Plasma was diluted 10-fold in 2% butanol, 1% ammonia, 0.05% ethylene diaminetetraacetic acid, 0.05% Triton-X-100. Fifty μL of red cells were dried and then digested in 500 μL concentrated nitric acid for 2 hours at 70°C. The CV for all methods was less than 5%.
Albumin, magnesium, ferritin, hemoglobin, plasma folate, vitamin B12, and cholesterol were measured by routine laboratory procedures.
Differences in median micronutrient concentration and the percentage of patients with suboptimal concentrations between initiation and end of EEN, and end of EEN and normal diet, were assessed with Wilcoxon's signed-rank test and Fisher's exact test as appropriate. The effect of EEN on serial measurements of anthropometry and body composition was evaluated using analysis of variance of repeated measures with Bonferroni post-hoc correction. The associations between CRP concentrations and micronutrient concentrations were assessed using Spearman rank correlation. Body weight, height, and body mass index (BMI) were converted to z-scores according to UK national data.18, 19 Measurements of body impedance were converted to ranks of lean and fat mass according to Wright et al20 using recent national reference data. Statistical significance was set at P ≤ 0.05.
Permission to conduct the study was granted by the local research ethics committee and the research and development office. Each subject and guardian received written information about the study and signed a consent form.
- Top of page
- MATERIALS AND METHODS
- Supporting Information
Seventeen patients (male/female: 9/8; median age: 12.7 years, range: 7–14.8 years) consented to participate over a 2-year period of recruitment. Measurements of body composition were obtained for 15 patients. Data on body composition for one patient who started concomitant treatment with oral steroids was excluded to avoid bias but this patient was included in the micronutrient part of this study. Another patient declined to complete measurement of body composition.
Fifteen patients had extensive disease of the upper GI tract and colon; one had isolated colitis and one had isolated ileitis. Eleven children were on no other concomitant medical treatment during EEN and the remaining six were started on or continued their stable dose of concomitant treatment in the form of immunosuppressants (n = 5), 5-aminosalicylates (n = 4), steroids (n = 1), or antibiotics (n = 2). All but three children were newly diagnosed and this was their first course of EEN. Thirteen patients (7 girls; median age: 12.8 years, range: 7–14.8 years) had micronutrients measured at EEN initiation, end of EEN, and on normal diet but the actual number of micronutrients measured for each patient varied depending on the amount of the blood sample collected from each patient (mainly due to technical difficulties in taking blood from small children). Four children had no blood samples taken for clinical investigations and therefore did not contribute to the micronutrient data. On treatment completion, eight patients achieved clinical remission, in four patients clinical disease activity improved at the end of EEN, and for five children treatment either failed (n = 4) or was discontinued (n = 1). After the end of EEN, and while patients were established on a normal diet, seven were judged to be in clinical remission, three improved but still had some symptoms of active disease, and the remaining seven had active disease.
Changes in Anthropometry and Body Composition
Body weight z-score, BMI z-score, and lean index increased significantly (P < 0.001) 30 days after EEN initiation (median: 31 days; interquartile range [IQR]: 6 days) and plateaued during the second half of the EEN course (Fig. 1, Table 1). In contrast, no significant changes were observed in the fat index during the same period (Fig. 1). When patients were divided into those who achieved clinical remission and those did not, the observed changes in BMI z-score and lean index persisted only for those who achieved remission (data not presented). Neither anthropometry nor body composition differed significantly between the end of EEN and when a normal diet had been established (Table 1).
|Measurement||EEN Initiation||After 30 Days (31: 6 d After EEN Initiation)a||End of EEN (53: 9 d After EEN Initiation)a||Normal Diet (56: 60 d After EEN Completion)a|
|Weight z-score (SD)||-1.3||2.0||-0.5b||1.9||-1.0b||1.9||-0.9b||1.9|
|Height z-score (SD)||-0.4||1.7||-0.2||1.8||-0.6||1.6||-0.4||1.7|
|BMI z-score (SD)||-2.2||2.3||-1.0b||1.9||-1.1b||2.0||-0.9b||2.0|
Changes in Circulating Micronutrient Concentrations
At the end of EEN there were statistically significant decreases in the median concentrations of the plasma carotenoids: lutein (57%), lycopene (61%), and β-carotene (33%) (Table 2). Plasma α-carotene concentrations also fell but this could not be assessed statistically because the concentrations at the end of EEN were all below the detection limit of the assay (Table 2, Fig. 2). On normal diet, the carotenoids showed significant increases compared to their concentrations at the end of EEN treatment: lutein (197%), lycopene (464%), α-carotene (30%), β-carotene (100%) (Table 3, Fig. 2).
|Micronutrient Plasma Unless Otherwise Stated||Initiation of EEN||End of EEN (49: 10 Days After EEN Initiation)a||P-valueb for Concentration||P-valuec for % Patients Suboptimal|
|Vitamin A/retinol (μmol/L)||1.2||0.7||1/12||(8)||1.7||0.9||0/12||(0)||0.083||1.000|
|Vitamin E/α-tocopherol: cholesterol (μmol/mmol)||7.9||2.6||0/12||(0)||6.5||3.7||0/12||(0)||0.492||n/a|
|Vitamin D (nmol/L)||56||81||0/10||(0)||93||32||0/10||(0)||0.131||n/a|
|Vitamin C (μmol/L)||29||25.5||3/12||(25)||49||15||0/12||(0)||0.016||0.217|
|Vitamin B1 (blood; ng/g Hb)||559||218||0/13||(0)||669||191||0/13||(0)||0.005||n/a|
|Vitamin B2 (erythrocytes; nmol/g Hb)||2.5||0.9||0/13||(0)||2.5||0.6||0/13||(0)||0.769||n/a|
|Vitamin B2 (blood; nmol/L)||461||67||1/13||(8)||477||83||0/13||(0)||0.191||1.000|
|Vitamin B6 (nmol/L)||36||23.2||3/11||(27)||63||38||0/11||(0)||0.024||0.214|
|Vitamin B6 (erythrocytes; pmol/g Hb)||662||411||0/13||(0)||721||332||0/13||(0)||0.057||n/a|
|Vitamin B12 (pg/ml)||644||445||0/13||(0)||843||534||0/13||(0)||0.278||n/a|
|Zn (erythrocytes; nmol/g Hb)||597||107||0/13||(0)||531||35.2||0/13||(0)||<0.001||n/a|
|Cu (erythrocytes; nmol/g Hb)||47.3||5.4||0/13||(0)||55.3||11.7||0/13||(0)||0.013||n/a|
|Se (erythrocytes; nmol/g Hb)||4.2||1.1||2/13||(15)||4.5||0.8||1/13||(8)||0.946||1.000|
|Mg (erythrocytes; nmol/g Hb)||8.8||2||0/13||(0)||8.4||1.3||0/13||(0)||0.013||n/a|
|Serum albumin (g/dL)||29||9||8/11||(73)||36||3.7||5/11||(45)||0.032||0.387|
|Micronutrient Plasma Unless Otherwise Stated||End of EEN||Normal Diet (59: 48 d After EEN Completion)a||P-valueb for Concentration||P-valuec for % Patients Suboptimal|
|Vitamin A/retinol (μmol/L)||1.7||0.7||0/10||(0)||1.2||0.6||0/10||(0)||0.004||n/a|
|Vitamin E/α-tocopherol: cholesterol (μmol/mmol)||6.6||3.2||0/10||(0)||6.9||3.5||0/10||(0)||0.301||n/a|
|Vitamin D (nmol/L)||102||31||0/9||(0)||59||39||0/9||(0)||0.012||n/a|
|Vitamin C (μmol/L)||51||13.2||0/11||(0)||46||17||1/11||(9)||0.322||1.000|
|Vitamin B1 (blood; ng/g Hb)||669||207||0/11||(0)||573||179||0/11||(0)||0.083||n/a|
|Vitamin B2 (erythrocytes; nmol/g Hb)||2.5||0.5||0/11||(0)||2.1||0.6||0/11||(0)||0.375||n/a|
|Vitamin B2 (blood; nmol/L)||478||107||0/11||(0)||435||93||0/11||(0)||0.240||n/a|
|Vitamin B6 (nmol/L)||63||37||0/9||(0)||21||17||4/9||(44)||0.055||0.082|
|Vitamin B6 (erythrocytes; pmol/g Hb)||705||186||0/11||(0)||438||367||0/11||(0)||0.147||n/a|
|Vitamin B12 (pg/ml)||843||696||0/9||(0)||505||300||0/9||(0)||0.019||n/a|
|Zn (erythrocytes; nmol/ g Hb)||533||35||0/11||(0)||547||235||0/11||(0)||0.365||n/a|
|Cu (erythrocytes; nmol/ g Hb)||52.3||11.4||0/11||(0)||46.4||10||0/11||(0)||0.032||n/a|
|Se (erythrocytes; nmol/ g Hb)||4.5||0.8||1/11||(9)||4.5||1.6||2/11||(18)||1.000||1.000|
|Mg (erythrocytes; nmol/ g Hb)||8.0||1.1||0/11||(0)||7.9||0.8||0/11||(0)||1.000||n/a|
|Serum albumin (g/dL)||36||4.5||3/9||(33)||32||10.2||6/9||(67)||0.148||0.347|
At the end of EEN there were statistically significant increases from baseline in the median concentrations for plasma vitamin C (90%), plasma folate (320%), blood vitamin B1 (10%), plasma B6 (66%), erythrocyte copper (13%), plasma selenium (71%), whereas erythrocyte zinc (10%), erythrocyte magnesium (7%), and plasma ferritin (46%) significantly decreased (Table 2). From the end of EEN to normal diet, several micronutrients showed a statistically significant decrease: vitamin A (25%), erythrocyte copper (9%), plasma selenium (26%), vitamin D (30%), and vitamin B12 (23%) (Table 3).
Changes in the Percentage of Patients with Suboptimal Concentrations
Before EEN, more than two-thirds of patients had suboptimal carotenoid concentrations (i.e., below reference interval; Online Supporting Material 2). At the end of EEN this proportion had increased to over 90%, although these changes were not statistically different compared with pre-EEN concentrations (Table 2). Indeed, for the majority of children all plasma carotenoid concentrations were below the detection limit of the in-house assay (10 μg/L) at the end of EEN (Table 2). After returning to a normal diet, significantly fewer patients had plasma lutein, lycopene, and α-carotene concentrations below the reference interval (Table 3).
No patient had suboptimal plasma concentrations for folate or vitamins C and B6 at the end of EEN compared with 40%, 25%, and 27%, respectively, at baseline (Table 2), but these changes did not achieve statistical significance. After returning to normal diet, the plasma concentration of vitamin B6 deteriorated and was below the reference interval for four children, although this change did not achieve statistical significance (Table 3).
Plasma zinc and selenium concentrations were below the normal reference intervals for more than 80% of the patients at treatment initiation and copper concentrations were deficient in 10% (Table 2). Only plasma selenium improved statistically during EEN, with only one child having suboptimal concentrations at the end of EEN (Table 2). Erythrocyte selenium concentration was low in two patients before EEN and one at the end of EEN (Table 2). Erythrocyte zinc and copper were within the normal reference intervals for all patients at all points of the study.
At the start of EEN, concentrations of the remaining micronutrients in all patients (vitamin E/cholesterol, vitamin D, vitamin B12, and magnesium in plasma, blood vitamin B1, and vitamin B2, vitamin B6, zinc, copper, and magnesium in erythrocytes) were within the reference interval except for one patient who had low vitamin A and blood vitamin B2 concentrations. No significant changes were seen in the proportions of those with concentrations below the reference intervals at the end of EEN or when children were established on normal diet. The proportion of children with anemia or suboptimal concentrations of ferritin did not change significantly at any point of the observational period (Tables 2, 3).
No significant differences were found in the proportion of children with suboptimal concentrations of plasma and erythrocyte trace elements between the end of EEN and when patients were established on normal diet, although proportionally more patients had suboptimal plasma micronutrient concentrations for plasma zinc and selenium when patients were on their normal diet (Table 3).
Changes in Inflammatory Markers
For those patients in whom circulating micronutrients and systemic inflammatory markers could be measured (n = 13), plasma albumin concentrations significantly improved at the end of EEN but not the median CRP and ESR concentrations (Table 2). However, in those patients who achieved clinical remission at the end of EEN, median CRP, serum albumin, but not ESR decreased significantly compared with at treatment initiation (both P < 0.036). The median value for all systemic markers of inflammation were outside the reference interval on normal diet but there was no significant difference compared with the end of EEN (Table 3).
Relationship Between Inflammatory Status and Micronutrient Levels in Plasma and RBCs
Plasma vitamin E concentration (corrected for cholesterol concentrations) tended to be positively associated with CRP at EEN completion (r = 0.63; P = 0.052). On normal diet, plasma vitamin D concentration was inversely associated (r = −0.77; P = 0.016), whereas erythrocyte concentrations of B2 tended to be positively associated (r = 0.66; P = 0.054) with CRP concentration. With regard to trace elements and minerals, only plasma zinc concentration was inversely associated with CRP (r = −0.70; P = 0.036) on normal diet. Ferritin concentration was strongly associated with CRP at all points of follow-up (initiation of EEN: r = 0.75; P = 0.005; end of EEN: (r = 0.84; P = 0.002; normal diet: r = 0.83; P = 0.005).
- Top of page
- MATERIALS AND METHODS
- Supporting Information
The clinical efficacy of EEN in children with CD has been established by means of randomized controlled trials21 and clinical experience4; however, its effect on indices of nutritional status has been mainly restricted to anthropometry. There are only limited data on body composition changes22, 23 and scarce reference to indices of body micronutrient status.24, 25
In accordance with previous research,23 this study, using a practical, bedside method of body composition assessment, found that CD children presented features of nutritional cachexia with lean mass depletion at EEN initiation and preservation of fat stores. Within 1 month on EEN, lean mass stores had returned to within the normal range and no further changes were seen during the second half of EEN. This increase in lean mass stores was sustained when patients returned to their normal diet. In contrast, no significant changes were observed in fat mass stores which remained at pre-EEN values. These findings are in accordance with previous research by Azcue et al,26 who observed accretion of lean mass assessed using stable isotope techniques of body composition, and can be attributed to an anabolic effect of nutritional therapy in children with CD with increased protein synthesis and suppression of catabolism.27 In the present study, lean mass improved only in patients who achieved clinical remission. Nevertheless, it is difficult to elucidate a cause or effect association between the replenishment of lean mass and improvement in disease activity.
Micronutrient measurements are not commonly reported in IBD28, 29 and are largely limited to case reports.30, 31 Suboptimal circulating concentrations for most vitamins and trace elements have been reported previously in adults32, 33 and sporadically in pediatric CD patients.34–38 In the current study, at EEN initiation a substantial number of CD children had circulating concentrations below the laboratory reference interval for a large number of micronutrients. More than 80% of the children had plasma zinc, selenium, and carotenoid concentrations below the reference intervals at EEN initiation. Indeed, the plasma concentration of all carotenoids was below the detection limit of the assay for the majority of participants. Low concentrations of antioxidants may indicate increased usage due to increased oxidative stress, as suggested previously,39–41 or suboptimal dietary intake.42, 43 With the exception of plasma copper and ferritin, the acute phase response is known to result in a decrease of plasma micronutrient concentrations,10, 44, 45 independently of their actual body stores. This may be the reason for the lower concentrations of plasma selenium, zinc, and vitamins A, B6, and C observed, although we failed to show clear strong relationships between CRP and the plasma concentrations of several of these micronutrients.
This lack of an association may be due to the simultaneous interaction of several factors that influence plasma micronutrients concentrations in counterbalancing ways (e.g., acute phase response, nutritional rehabilitation, mucosal healing, and increased intestinal absorption) or lack of statistical power due to the small number of participants in this study. Nevertheless, plasma ferritin, a positive acute phase reactant, was strongly and positively associated with serum CRP at all timepoints of the follow-up, which explains our observation that the majority of the children were anemic on EEN initiation despite normal concentrations of plasma ferritin in the majority of our participants. Thus, the measurement of ferritin concentrations in CD patients with active disease has poor diagnostic potential for iron deficient anemia and should be interpreted only in conjunction with other hematological parameters46 and disease activity markers. Nevertheless, in 40% of patients ferritin concentrations were below the reference interval at EEN initiation. This may be explained by the absence of or low-grade systemic inflammation in some of our patients, and may well indicate patients with depleted iron stores. Indeed, all patients with suboptimal ferritin levels had CRP concentrations below 8 mg/L and a median concentration significantly lower than that of patients with normal ferritin concentrations (normal vs. Low ferritin, median; IQR: 7; 1 vs. 25; 35 mg/L, P = 0.010).
In this study we also measured the erythrocyte concentrations of several B complex vitamins and trace elements, since such measures are unaffected by the acute inflammatory response. Approximately 30% of our participants had suboptimal plasma concentrations of vitamin B6 on EEN initiation but none had suboptimal erythrocyte concentrations. Likewise, for most trace elements plasma concentrations were below the reference intervals but erythrocyte concentrations were not. These contradictory results may indicate an artifactual effect of the inflammatory response on the plasma concentrations, although, as mentioned above, we did not find an overt association between CRP and plasma micronutrient concentrations.
Significant changes in micronutrient concentrations were observed during EEN. Plasma concentrations for many micronutrients improved at the end of EEN; however, carotenoid concentrations decreased significantly, with more than 90% of participants having depleted stores for all carotenoids at EEN completion. Carotenoids, being negative acute phase reactants, would have been expected to improve at the end of EEN as a combined effect of nutritional hyperalimentation and amelioration of disease activity. As a result, these findings provide strong evidence that carotenoid body stores are severely depleted at the end of EEN, probably due to increased utilization during the active course of the disease and inadequate replenishment through dietary intake.43 This prompted us to investigate the carotenoid content of the EEN feed. The nutritional feed used is better characterized as an “artificial composite food” and is nutritionally complete and balanced for only the basic nutrients for which recommended amounts have been established (Online Supporting Material 1). Indeed, Modulen Nestle composition tables do not list carotenoids as constituents (pers. commun., Nestle) and none of its ingredients is a natural carrier of them. Depletion of antioxidant carotenoids may compromise the body's defensive antioxidant mechanisms and increase oxidative stress and damage of biological macromolecules. Unfortunately, we did not measure plasma oxidative stress markers. A previous study,25 however, showed reduced activity of enzymatic antioxidant systems in erythrocytes of CD children that did not improve with 8 weeks on EEN despite improvement of the clinical activity and systemic inflammatory markers (CRP and tumor necrosis factor alpha [TNF-α]). Likewise, Akobeng et al24 found no significant changes in antioxidant mechanisms or markers of oxidative stress in children with CD after a 4-week course with EEN. Both of these studies used the same commercial feed as the current study, and so despite clinical improvement, reduced production of reactive oxidative species, and decreased antioxidant consumption, inadequate provision of dietary antioxidants fails to replenish the depleted body stores. If this is the case, supplementation of the feed with antioxidants may increase the clinical efficacy of EEN.
The strong effect of EEN on micronutrient status is also shown by the observation that within a short period of EEN cessation the majority of micronutrients reversed or tended to reverse to pre-EEN concentrations. Interestingly, although none of our participants had suboptimal concentrations of vitamins D and B12 at any timepoint of follow-up, their median concentrations on normal diet significantly decreased compared with the end of EEN, reflecting a gradual deterioration of body stores, particularly as the majority of our patients were newly diagnosed. Sentongo et al47 reported that 16% of CD children and young adults had low circulating concentrations of vitamin D, in line with the definition of hypovitaminosis, but none of our participants did (using a vitamin D threshold of 25 nmol/L). Different definitions of hypovitaminosis between the two studies may explain this discrepancy. For example, if like Sentongo et al a threshold of 38 nmol/L or 75 nmol/L according to recent recommendations48 had been used to define vitamin D hypovitaminosis, 25% or 58%, respectively, of participants in our study would have been classified as deficient at treatment initiation.
There were some limitations of this study: the sample size was small, although a post-hoc power calculation (90% power, 5% error) showed that we had adequate sample size to detect differences for several micronutrients (e.g., lutein, lycopene, plasma selenium, folate); missing data; the use of adult reference intervals for some micronutrients in the absence of a pediatric reference; the concomitant use of other medications; and the lack of a healthy control group on parallel treatment with EEN. However, pediatric CD is an uncommon disease, with ≈20 new cases per year in the largest pediatric referral center in Scotland. This, in conjunction with alternative treatment options (steroids, anti-TNF-α agents), and the unpredictable course of the disease, substantially reduced our recruitment rates. Furthermore, venipuncture entirely for research purposes was deemed unethical and this further reduced the number of patients with follow-up measurements. Recruitment of a control sample of healthy children on parallel treatment with EEN was not possible due to obvious ethical constraints but it would be impossible to observe such radical changes in micronutrient status in healthy children within such a short period of time. Although we intended to study differences in the patterns of micronutrient changes between patients whose disease improved and those in whom it did not, the lack of an association between nutrient concentrations and systemic inflammatory markers and the small sample size precluded such analysis. Some patients were on a stable dose of medication or started on concomitant therapy with other medications, as per standard clinical practice, which may have biased our findings. The small size of this pilot and the heterogeneity of medication type and doses precluded a subanalysis. When we plotted the micronutrient concentration changes during EEN for each of our participants we observed either changes in the same direction in the change of the majority of micronutrients, or high variability in the direction of the changes that could not be explained by the use of concomitant treatment or clinical response to EEN (i.e., difference between responders and nonresponders to treatment) (Online Supporting Materials 3, 4).
Moreover, it would be unlikely that an effect of the parallel use of medication would prevail over the hyperalimentation effect of the nutritional complete feed we used in the study, which provided the majority of micronutrients in excess of the dietary reference values (Online Supporting Material 1).
Nutritional status is suboptimal in pediatric CD patients with active disease and significant changes occur following treatment with EEN. Changes in the concentrations of the circulating micronutrients are difficult to explain and may be influenced by the effect of nutritional replenishment, resolution of the acute phase response, and decreased utilization with disease improvement. This may also explain the lack of an expected relationship between some plasma micronutrients and acute phase inflammatory markers. However, this study suggests that changes in red blood cell micronutrient levels are likely to be better markers of true body stores than concentrations in plasma.
This study thus advocates the use of EEN. It supports the protein/energy stores of CD children with active disease but the composition of the EEN appears to be insufficient to replenish or sustain carotenoid body status.
- Top of page
- MATERIALS AND METHODS
- Supporting Information
The authors thank the participants and their caregivers.
- Top of page
- MATERIALS AND METHODS
- Supporting Information
- 4The use of exclusive enteral nutrition for induction of remission in children with Crohn's disease demonstrates that disease phenotype does not influence clinical remission. Aliment Pharmacol Ther. 2009; 30: 501–507., , , et al.
- 12Department of Health. Dietary Reference Values for Food Energy and Nutrient for the United Kingdom. Report on Health and Social Subjects No. 41. 1991.
- 16Optimisation and validation of a sensitive high-performance liquid chromatography assay for routine measurement of pyridoxal 5-phosphate in human plasma and red cells using pre-column semicarbazide derivatisation. J Chromatogr B Analyt Technol Biomed Life Sci. 2003; 792: 333–343., , , et al.
- 33Low vitamin B(6) plasma levels, a risk factor for thrombosis, in inflammatory bowel disease: role of inflammation and correlation with acute phase reactants. Am J Gastroenterol. 2003; 98: 112–117., , , et al.Direct Link:
- 42A comprehensive approach to evaluate nutritional status in Crohn's patients in the era of biologic therapy: a case-control study. Am J Gastroenterol. 2007; 102: 2551–2556., , , et al.Direct Link:
- Top of page
- MATERIALS AND METHODS
- Supporting Information
Additional Supporting Information may be found in the online version of this article.
|IBD_21916_sm_suppinfo1.doc||45K||Supporting Information 1.|
|IBD_21916_sm_suppinfo2.doc||47K||Supporting Information 2.|
|IBD_21916_sm_suppinfo3.doc||483K||Supporting Information 3.|
|IBD_21916_sm_suppinfo4.doc||472K||Supporting Information 4.|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.