Abstract
Malnutrition is common in patients with Crohn’s disease and negatively influences immunity and quality of life. The optimal tools for nutrition assessment in patients with Crohn’s disease are not clearly defined and lead to variations in practice. With this review, we aimed to appraise the existing evidence for nutrition assessment of patients with Crohn’s disease compared with healthy controls and provide a comprehensive guide with relevant measures applicable to clinical practice. A literature search using Medline, Embase, and Scopus from inception to October 1, 2018, was conducted. Forty-one articles that assessed body composition, muscle strength, micronutrient status and/or dietary intake in adults with Crohn’s disease compared with an age- and sex-matched healthy individuals were included. There were heterogeneous findings on nutritional status in patients with Crohn’s disease compared with healthy controls. Only one article reported a clinically significant difference for body mass index; however, significant deficits in fat mass, fat-free mass, and muscle strength were observed in patients with Crohn’s disease compared with healthy controls, with more pronounced differences with increasing disease activity and length of diagnosis. Most research reported significantly lower serum micronutrients in patients with Crohn’s disease compared with healthy controls. Half of studies measuring micronutrient intake reported lower intakes in patients with Crohn’s disease compared with healthy controls. Fruit and vegetable intake was also lower in patients with Crohn’s disease. Difficulties characterizing the type and prevalence of malnutrition exist due to the heterogeneous nature of Crohn’s disease and warrants continued investigation. As a result of this review, we advocate that a nutrition assessment should include more parameters than weight and body mass index.
Keywords
Continuing Professional Education (CPE) Information
The Continuing Professional Education (CPE) quiz for this article is available for free to Academy members through the MyCDRGo app (available for iOS and Android devices) and through www.jandonline.org (click on “CPE” in the menu and then “Academy Journal CPE Articles”). Log in with your Academy of Nutrition and Dietetics or Commission on Dietetic Registration username and password, click “Journal Article Quiz” on the next page, then click the “Additional Journal CPE quizzes” button to view a list of available quizzes. Non-members may take CPE quizzes by sending a request to [email protected] . There is a fee of $45 per quiz (includes quiz and copy of article) for non-member Journal CPE. CPE quizzes are valid for 1 year after the issue date in which the articles are published.
Research Snapshot
Research Question: What is the existing evidence to inform a comprehensive nutrition assessment of patients with Crohn’s disease?
Key Findings: There were heterogeneous findings on nutritional status in Crohn’s disease. Significant deficits in fat mass, fat-free mass, and muscle strength were observed. Lower serum micronutrient levels, micronutrient intakes, and fruit and vegetable intakes were reported in patients with Crohn’s disease compared with healthy controls. The findings from this narrative review have informed the development of a practical clinical guide for comprehensive nutrition assessment of patients with Crohn’s disease.
1
It is associated with increased susceptibility to infections, gastrointestinal barrier dysfunction, postoperative complications, and reduced quality of life.2
, 3
, 4
Reasons for malnutrition in patients with Crohn’s disease are multifactorial. More than 80% of people with Crohn’s disease experience problems with food
5
and 72% alter their diet as a result,6
often leading to insufficient nutrient intakes.7
Active disease is associated with reduced appetite, low mood, and abdominal pain1
; furthermore, mucosal inflammation causes malabsorption due to damaged intestinal microvilli8
and increased diarrhea, leading to a loss of electrolytes and fluids.9
Systemic inflammation elevates nutrient requirements due to catabolism causing weight loss.1
The inflammatory response produces cell-damaging free radicals; micronutrients act as antioxidants to reduce damage; therefore, prolonged inflammation eliminates micronutrients via excessive utilization.10
Pharmacologic side effects also contribute to malnutrition. Corticosteroids increase adiposity and are associated with reduced bone mineral density.11
Micronutrient deficiencies in patients with Crohn’s disease are a further health care burden. Inflammation and suboptimal vitamin D levels are associated with impaired bone mineral density, making osteoporosis common in patients with Crohn’s disease.12
Dietary deficits in zinc reduce muscle mass and strength,13
which has deleterious consequences on functional ability and activities of daily living.14
Suboptimal circulating concentrations of folic acid, vitamin B-12, vitamin C, and selenium have also been reported.15
, 16
The risk of malnutrition persists during the remission phase of the disease; whilst 86% of patients with active disease avoid certain foods during flare-ups, 77% of patients continue to avoid certain foods during remission to prevent disease relapse.7
In clinical practice, nutrition assessment in patients with Crohn’s disease remains challenging and most frequently is measured using weight and body mass index (BMI).
17
Weight and BMI are inadequate measures of malnutrition in Crohn’s disease as systemic inflammation alters body composition, meaning BMI may mask deficits in lean mass due to increased fat mass.18
However, there are no guidelines on what components should be included in a comprehensive nutrition assessment for patients with Crohn’s disease.Accurate quantification of nutritional status in Crohn’s disease is essential to enable diet and nutrition therapy to be targeted to address specific deficits. However, in a study on nutrition assessment in patients with Crohn’s disease, body composition was measured in only 3%, handgrip strength in only 4%, and dietary micronutrient intake in 16% of patients, suggesting that current assessments are limited.
17
This narrative review comprehensively appraised the existing evidence for nutrition assessment of patients with Crohn’s disease in comparison to a healthy population. It aims to provide a comprehensive guide with relevant measures applicable to clinical practice.
Methods
Search Strategy and Study Selection
The population, intervention, comparison, outcomes, and type of study framework
19
was used to inform the criteria needed to answer the research question, What evidence exists on the nutritional status of patients with Crohn’s disease and how can this evidence inform nutrition assessment in clinical practice? The search strategy included studies of patients with Crohn’s disease aged 18 to 64 years using validated assessment methods available in clinical practice to establish nutritional status compared with a healthy age- and sex-matched control group (HC) sampled from the same population as those with Crohn’s disease. Studies that reported nutritional status outcomes, including body composition, muscle strength and function, micronutrient status, and/or dietary intake were included in the case that they were in the English language and primary research or systematic reviews.Limiting the search in this way allowed the literature review to establish a typical nutritional status in healthy people without Crohn’s disease and facilitated the comparative quantification of nutritional status in Crohn’s disease. Whilst anthropometric reference ranges for the healthy population have been developed, these vary depending on assessment methods used.
20
Recruiting a HC group ensured comparisons were made using identical methods to those used with patients with Crohn’s disease. Three databases were searched (Medline, Embase, and Scopus) on October 1, 2018. Multiple search terms were combined with the Boolean functions and and or to focus the search.21
The medical library subject heading terms or keywords included were [Crohn’s disease OR inflammatory bowel disease] AND [nutrition* assessment, body composition, body fat, fat mass, anthropometry, lean body weight, malnutrition, protein energy malnutrition, muscle strength, hand grip, grip strength, trace element, nutrition* status, nutrition* deficiency, vitamin deficiency, mineral deficiency, dietary intake, diet OR micronutrient]. Filters (English, human, and adults aged 18 to 64 years) were applied to target the search results.Following removal of duplicates, the titles, and where applicable abstracts, were screened for relevance. Abstracts of relevant titles were reviewed and in the case that an HC group was described the full text was examined against the inclusion and exclusion criteria.
Data Extraction and Synthesis
Eligible studies for data synthesis were critically appraised using the Critically Appraising Papers process in Hickson
22
and the Assessing Methodological Quality question checklist in Greenhalgh23
to assess quality of individual studies. Data were summarized in a data extraction spreadsheet according to anthropometric, biochemical, and dietary assessment techniques (per the Nutrition Care Process structure24
). The Nutrition Care Process was developed by the Academy of Nutrition and Dietetics and is used by members of the dietetics profession to ensure systematic, evidence-based nutrition care.24
Outcome data were only extracted when they were available and clinically relevant. Anthropometric outcomes included assessment of body composition using direct anthropometry, bioelectrical impedance analysis (BIA), dual energy x-ray absorptiometry (DEXA), computed tomography (CT) or magnetic resonance imaging (MRI), and muscle strength or function measurements. Biochemical outcomes included plasma or serum markers of nutritional status such as folic acid, vitamin B-12, vitamin C, vitamin D, zinc, copper, and selenium. Iron status and albumin were not collected because these are acute phase reactants and results are difficult to compare with an HC population. Dietary intake outcomes included macronutrient and micronutrient intake, food group intake, or exclusions of specific food groups. Where possible the anthropometric, biochemical, and dietary assessment methods and results were compared and critiqued across studies.Discussion
To our knowledge, this is the first review appraising the evidence for methods of nutrition assessment in patients with Crohn’s disease relevant to clinical practice. There were 41 eligible papers (Figure 1), including 2,370 patients with Crohn’s disease and 4,450 HC. All studies were cross-sectional in design. The Crohn’s disease cohorts included patients with active disease and/or disease in remission. Most studies included men and women with the exception of two studies that reported body composition data of only men
25
or only women.26
Nevertheless, compared with HC, there were significant differences in body composition and dietary intake as well as deficits in muscle strength and serum micronutrients. The findings follow the Nutrition Care Process (anthropometric, biochemical, and dietary assessment structure) and include recommendations for clinical practice (Figure 2).
Figure 1Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram for studies included in the narrative review on nutrition assessment in patients with Crohn’s disease.
32209-3&id=gr1.jpg){kind=link}
Figure 2Components of a comprehensive nutrition assessment tool for patients with Crohn’s disease. Components are based on evidence from studies included in the narrative review.
93
, 94
, British Assocation for Parenteral and Enteral Nutrition
Malnutrition Universal Screening Tool (MUST).
Malnutrition Universal Screening Tool (MUST).
https://www.bapen.org.uk/pdfs/must/must_full.pdf
Date accessed: May 8, 2019
95
, 96
, 97
, 98
, 99
, 100
, 101
, 102
, 103
, 104
, 105
, 106
, 107
32209-3&id=gr2a.jpg){kind=link}
Figure 2Components of a comprehensive nutrition assessment tool for patients with Crohn’s disease. Components are based on evidence from studies included in the narrative review.
93
, 94
, British Assocation for Parenteral and Enteral Nutrition
Malnutrition Universal Screening Tool (MUST).
Malnutrition Universal Screening Tool (MUST).
https://www.bapen.org.uk/pdfs/must/must_full.pdf
Date accessed: May 8, 2019
95
, 96
, 97
, 98
, 99
, 100
, 101
, 102
, 103
, 104
, 105
, 106
, 107
32209-3&id=gr2b.jpg){kind=link}
Anthropometric Outcomes
Clinically relevant and commonly available anthropometric assessments methods were reviewed.
BMI
In the majority of studies (n=18), BMI was not significantly different between patients with Crohn’s disease and HC (Table 1),
11
, 13
, 15
, 16
, 25
, 27
, 28
, 29
, 30
, 31
, 32
, 33
, 34
, 35
, 36
, 37
, 38
, 39
but in eight of these studies, significant differences in body composition were observed.11
, 15
, 16
, 25
, 29
, 30
, 34
, 39
Where significant differences in BMI existed (n=12), it was always lower in patients with Crohn’s disease compared with HC.40
, 41
, 42
, 43
, 44
, 45
, 46
, 47
, 48
, 49
, 50
, 51
However, studies rarely assessed clinically significant differences in BMI, because BMI tended to be reported as a mean rather than as the proportion of patients who had a clinically underweight BMI (<18.5).52
Only one study assessed this, and the prevalence of underweight BMI was 21% in patients with Crohn’s disease and 2% to 4% in HC.38
Table 1Assessment of body composition in patients with Crohn’s disease (CD) compared with an age- and sex-matched healthy control group (HC)
| Author(s), year, country | Participants | Disease activity | Technique | BMI | Fat Mass (kg) | % Fat Mass | Fat-Free Mass (kg) | Visceral Adipose Tissue (cm2 or mL) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CD | HC | CD | HC | CD | HC | CD | HC | CD | HC | CD | HC | |||
| ←n→ | ←mean±standard deviation→ | |||||||||||||
| Capristo and colleagues, 46 1998, Italy | 43 | 60 | Remission | BIA | 21.5 | 23.7 | 12.2 | 17.0 | 20.4 | 25.5 | 49.2 | 50.4 | — | — |
| Capristo and colleagues, 47 1998, Italy | 18 | 20 | Remission | BIA | 20.5 | 23.6 | 12.6 | 17.4 | 22.0 | 26.4 | 45.6 | 49.5 | — | — |
| Mingrone and colleagues, 55 1999, Italy | 18 | 12 | Mixed | BIA | 21.6±2.9 | 23.8±1.8 | 13.8±5.6 | 19.0±3.5 | — | — | 48.0±7.1 | 47.7±6.1 | — | — |
| Wiroth and colleagues, 13 2005, France | 41 17 M 24 W | 25 10 M 15 W | Remission | BIA | 22.1±3.5 22.1±3.7 | 24.0±2.4 21.4±1.6 | 13.0±5.1 15.3±4.8 | 16.4±5.4 16.0±4.2 | 18.3±6.0 25.8±4.5 | 21.7±5.6 27.5±5.5 | 56.2±6.4 42.9±6.5 | 58.0±4.5 41.0±4.4 | — | — |
| Filippi and colleagues, 15 2006, France | 54 | 25 | Remission | BIA | 22.1 | 22.1 | 14.4 | 16.6 | — | — | 49.2 | 46.7 | — | — |
| Valentini and colleagues, 16 2008, Austria, Germany, and Italy | 94 33 M 61 W | 61 20 M 41 W | Remission | BIA | 22.3 22.1 | 23.7 21.8 | 12.7 18.1 | 15.2 16.6 | — | — | 58.5 43.9 | 67.4 44.1 | — | — |
| Benjamin and colleagues, 45 2011, India | 80 | 100 | Remission | BIA | 21.6±5.0 | 23.9±4.0 | 13.4±10.6 | 14.1±7.5 | 21.9±11.3 | 21.5±9.1 | 43.3±10.4 | 48.9±7.4 | — | — |
| 43 | Active | 18.8±3.6 | 23.9±4.0 | 8.2±5.9 | 14.1±7.5 | 15.7±9.3 | 21.5±9.1 | 40.7±8.5 | 48.9±7.4 | — | — | |||
| Rizzi and colleagues, 39 2012, Italy | 78 42 M 36 W | 75 41 M 34 W | Mixed | BIA | 22±2 21±3 | 22±1 22±2 | 12±6 15±4 | 22±4 21±4 | — | — | 53±4 37±3 | 49±7 40±3 | — | — |
| Lu and colleagues, 44 2016, China | 150 109 M 41 W | 256 115 M 139 W | Mixed | BIA | 19.8±3.6 19.1±3.2 | 23.9±3.0 22.1±2.9 | 9.9±6.0 12.7±5.0 | 16.8±6.4 17.2±5.5 | — | — | — | — | — | — |
| Katznelson and colleagues, 25 2003, United States | 20 M | 20 M | Mixed | BIA and CT | 24.2 | 23.3 | — | — | 21.0 | 17.7 | — | — | 115±11 | 69±7 |
| Buning and colleagues, 26 2015, Germany | 31 W | 19 W | Mixed | MRI | 25.9 | 23.8 | — | — | — | — | — | — | 1,185±1,403 | 941±988 |
| ←n→ | ←mean±standard deviation→ | |||||||||||||
| Geerling and colleagues, 28 1998, The Netherlands | 32 14 M 18 W | 32 14 M 18 W | Remission | DEXA | 23.2±3.7 22.8±4.1 23.4±3.5 | 24.6±3.6 26.4±3.5 23.3±3.2 | 17.6±7.9 13.2±7.0 20.9±7.1 | 19.7±6.3 18.4±5.0 20.7±7.2 | 26.1±9.4 18.4±6.3 32.1±6.5 | 28.7±7.5 23.5±5.3 32.7±16.45 | 48.6±9.3 56.4±7.1 42.6±5.5 | 49.7±11.6 60.5±8.9 41.2±2.9 | — | — |
| Tjellesen and colleagues, 34 1998, Denmark | 31 13 M 18 W | 88 19 M 69 W | Remission | DEXA | 23.5 21.1 | 23.9 22.0 | 20.3 21.6 | 19.2 21.3 | 27.8 38.8 | 23.1 32.8 | 51.8 34.9 | 62.2 42.4 | — | — |
| Geerling and colleagues, 30 1999, The Netherlands | 20 40 | 20 40 | Mixed Mixed | DEXA | 22.7±2.5 22.8±4.0 | 23.0±2.8 24.0±3.3 | 19.4±6.3 17.7±8.7 | 19.5±7.0 18.9±4.9 | 28.3±8.3 26.7±10.6 | 29.2±8.6 27.7±6.3 | 49.2±9.9 47.1±8.7 | 46.8±8.9 49.9±11.0 | — | — |
| Geerling and colleagues, 29 2000, The Netherlands | 23 | 23 | Mixed | DEXA | 22.2±2.7 | 22.7±2.7 | 18.5±6.5 | 19.0±6.8 | 27.5±8.4 | 28.7±8.7 | 48.9±9.8 | 46.9±8.9 | — | — |
| Jahnsen and colleagues, 11 2003, Norway | 60 24 M 36 W | 60 24 M 36 W | Mixed | DEXA | 23.3 23.2 23.4 | 23.4 24.8 22.5 | 20.8 16.7 23.5 | 20.0 18.1 21.3 | 31.4 23.1 37.0 | 29.2 22.6 33.6 | 44.5 54.2 38.0 | 48.8 61.0 40.7 | — | — |
| Cuoco and colleagues, 48 2008, Italy | 13 | 20 | Active | DEXA | 19.8±1.2 | 23.4±1.1 | 21.1±1.4 | 19.6±2.4 | 35.8±4.1 | 49.6±4.4 | ||||
| Schneider and colleagues, 43 2008, France | 82 | 50 | Remission | DEXA | 21.1±3.4 | 22.2±2.5 | 16.2±8.9 | 16.1±4.9 | 25.7±10.0 | 25.9±7.7 | 43.8±9.4 | 46.7±10.1 | ||
a BMI=body mass index.
b Only mean is provided where standard deviation was absent in the original reference.
c BIA=bioelectrical impedance analysis.
d M=men.
e W=women.
f CT=computed tomography.
g MRI=magnetic resonance imaging.
h DEXA=dual energy x-ray absorptiometry.
i Weighted mean reported.
j Newly diagnosed.
k Longstanding disease >5 years.
∗ P<0.05 for CD vs HC.
∗∗ P<0.01 for CD vs HC.
∗∗∗ P<0.001 for CD vs HC.
DEXA
Seven studies used DEXA to determine body composition (Table 1).
11
, 28
, 29
, 30
, 34
, 43
, 48
DEXA studies most frequently found no difference in fat mass (FM) between patients with Crohn’s disease and HC,28
, 29
, 34
, 43
, 48
but a trend of fat-free mass (FFM) depletion in patients with Crohn’s disease.11
, 34
, 48
Superior FFM was observed in patients with newly diagnosed Crohn’s disease compared with HC.28
, 29
The only study to include patients with longstanding Crohn’s disease (>5 years) found they had significantly lower FFM compared with HC.30
These findings suggest that lean mass depletion in patients with Crohn’s disease occurs over time. One study recruited patients with active Crohn’s disease and showed that BMI was significantly lower in the active disease group as was FFM, and FM was nonsignificantly different when compared with HC.48
DEXA scans have ethical and practical limitations. Small amounts of radiation are absorbed by bone and tissue and increasing exposure to radiation is linked to an increased cancer risk.53
In addition, whole-body DEXA scans are conducted by specialist radiographers,54
which presents a practical barrier for routine clinical use.BIA
Eleven studies used BIA to determine body composition (Table 1).
13
, 15
, 16
, 25
, 39
, 41
, 42
, 44
, 45
, 46
, 47
In contrast to DEXA, the majority of BIA studies observed a lower FM in patients with Crohn’s disease compared with HC15
, 39
, 45
, 46
, 47
, 55
but, as shown in Table 1, the results were not consistent.13
, 16
, 25
For FFM, there were no consistent differences between groups.A study from India in patients with active Crohn’s disease detected significant deficits in FM and FFM.
45
However, it lacks external validity to non-Indian populations because recent data demonstrate significant ethnic disparities in body composition, especially in South Asians.56
Another study in patients with active Crohn’s disease reported lower FM compared with HC.39
Thus, there are body composition deficits in patients with active Crohn’s disease, highlighting the importance of considering disease activity in the clinical assessment section of the Nutrition Care Process.CT and MRI
Three studies used medical imaging techniques to further explore body composition.
25
, 26
, 37
One study undertook umbilicus CT scanning to determine body fat distribution alongside BIA.25
The researchers found intra-abdominal fat was significantly higher in patients with Crohn’s disease vs HC. Furthermore, using MRI, visceral adipose tissue was significantly higher in patients with Crohn’s disease in remission compared with HC.26
In another study, CT scans were used to characterize muscle size.37
Quadricep muscle cross-sectional area was 14% lower in patients with Crohn’s disease compared with HC; however, this was not statistically significant.Muscle Strength and Function
Eight studies assessed muscle strength and function.
13
, 16
, 28
, 29
, 36
, 39
, 44
, 57
Limited studies have reported the potential effect of disease duration on muscle strength or function.16
, 36
In patients with newly diagnosed Crohn’s disease, muscle strength is similar to HC,29
whereas at least 5 years after diagnosis, the literature suggests a reduction in muscle strength and increased muscle fatigue in active disease or disease in remission.16
, 28
, 36
, 57
However, disease activity may influence muscle strength.39
, 44
There are no reports of change in muscle strength over time in patients with Crohn’s disease compared with HC. It is unknown whether reduced muscle strength during active disease is a temporary reduction in strength associated with a disease flare and if, or how quickly, muscle strength improves once the disease is in remission. One study found no difference in handgrip strength but reduced muscle endurance between patients with Crohn’s disease in remission for at least 3 months compared with HC.
13
Longitudinal research on muscle strength during periods of active disease and disease remission would provide further understanding on the effects of acute and chronic inflammation on muscle strength and function. Muscle wasting and weakness results in fatigue and reduced quality of life,13
both of which are prevalent in people living with Crohn’s disease.58
, 59
Direct Anthropometry
Five studies report the use of direct anthropometry in their methods
16
, 28
, 40
, 46
, 47
; however, three do not report their data.16
, 46
, 47
The authors cite strong correlations between their direct anthropometry results and BIA/DEXA as a justification for presenting only the results of the latter. However, critics may argue this preferential inclusion of BIA/DEXA results at the expense of omitting anthropometric data represents reporting bias.60
Direct anthropometry is the most frequently used body composition assessment method in clinical practice because of its low cost and feasibility.61
Therefore, there is a missed opportunity for these unreported anthropometric data to be available to clinicians.One study calculated body FM percentage using composite measures of skinfold thickness from the bicep, tricep, subscapular, and suprailiac.
28
FM percentage and muscle mass did not differ significantly between patients with Crohn’s disease and HC. This finding that the body composition of Crohn’s disease patients is not inferior to HC is surprising; especially considering 47% of the group had active disease (Crohn's disease activity index [CDAI] >150). In another study, lower tricep skinfold thickness was reported in men with Crohn’s disease compared with HC men, whilst there was no difference between women, suggesting there may be sex differences.40
Summary for Anthropometric Outcomes
The majority of studies found no significant difference in BMI between Crohn’s disease and HC groups,
11
, 13
, 15
, 16
, 17
, 25
, 27
, 28
, 29
, 30
, 31
, 32
, 33
, 34
, 36
, 37
, 39
, 41
confirming that using BMI alone provide limited data for an optimal nutrition assessment. Only 14 studies examined FFM and FM, half of which suggest that FFM is decreased in Crohn’s disease11
, 16
, 29
, 30
, 34
, 45
, 48
and two studies suggests that intra-abdominal FM is greater in Crohn’s disease than HC.25
, 26
A reduction in muscle endurance in Crohn’s disease, and reduced muscle strength during active or longstanding Crohn’s disease has been reported.13
, 16
, 28
, 36
, 39
, 44
, 57
BIA is a more feasible and less invasive measure of body composition than CT or DEXA scans.61
However, the routine use of BIA in clinical practice may be time intensive and financially challenging; thus, mid-arm anthropometry and handgrip strength (HGS) are measures that can be readily and cheaply assimilated into clinical practice61
(Figure 2). Because body composition fluctuates over the disease course, anthropometric assessments should be repeated to monitor change.Biochemical Outcomes
Comprehensive plasma micronutrient studies are arguably lacking, with most articles only quantifying two or three micronutrients.
50
, 62
, 63
, 64
, 65
, 66
, 67
Geerling and colleagues28
, 29
are the only research group to measure an extensive range of micronutrients.28
, 29
A major limitation of the 18 micronutrient studies (Table 2 and Table 3)27
, 28
, 29
, 31
, 32
, 33
, 49
, 50
, 51
, 57
, 62
, 63
, 64
, 65
, 66
, 67
, 68
, 69
, 70
is that only two15
, 16
report deficiency prevalence for micronutrients other than vitamin D. In clinical practice, patients are not treated for low micronutrient levels unless they are deficient,71
thus it would be more clinically relevant to report the prevalence of micronutrient deficiency rather than mean micronutrient levels.Table 2Blood markers of nutritional status in patients with Crohn’s disease (CD) compared with an age- and sex-matched healthy control group (HC)
| Author(s), year, country | Participants | Disease activity | Folic Acid (nmol/L) | Vitamin B-12 (pmol/L) | Vitamin C (μmol/L) | Zinc (μmol/L) | Copper (μmol/L) | Selenium (μmol/L) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CD | HC | CD | HC | CD | HC | CD | HC | CD | HC | CD | HC | CD | HC | ||
| ←n→ | ←mean±standard deviation→ | ||||||||||||||
| Kallel and colleagues, 50 2011, Tunisia | 89 | 103 | Mixed | 19.3±6.9 | 18.4±7.0 | 218±118 | 279±125 | — | — | — | — | — | — | — | — |
| Yakut and colleagues, 67 2010, Turkey | 45 | 53 | Mixed | 17.4±12.0 | 22.4±7.5 | 207±122 | 252±132 | — | — | — | — | — | — | — | — |
| Geerling and colleagues, 62 1999, The Netherlands | 20 32 | 20 32 | Remission Mixed | — | — | — | — | — | — | 12.4 12.0 | 13.0 13.1 | — | — | — | — |
| Geerling and colleagues, 29 2000, The Netherlands | 23 | 23 | Mixed | 10.7±9.1 | 12.4±5.6 | 225±60.7 | 270±88.2 | 47.6±17.7 | 54.5±22.9 | 12.3±3.0 | 12.9±1.3 | 23.6±8.9 | 22.2±7.4 | 0.92±0.16 | 0.99±0.16 |
| Geerling and colleagues, 28 1998, The Netherlands | 32 | 32 | Remission | 14.4±13.4 | 13.4±5.88 | 403±282 | 263±91.5 | 35.3±25.8 | 57.8±22.3 | 12.0±1.7 | 13.4±2.2 | 19.1±4.6 | 20.1±6.9 | 0.86±0.14 | 1.30±0.15 |
| Hinks and colleagues, 64 1988, United Kingdom | 11 | 22 | Active | — | — | — | — | — | — | 12.7±1.8 | 12.9±1.7 | 17.3±3.3 | 16.3±2.6 | — | — |
| Ringstad and colleagues, 65 1993, Norway | 47 27 20 | 123 76 47 | Not stated | — | — | — | — | — | — | 14.4 13.5 | 12.7 12.9 | 20.8 23.8 | 15.8 18.1 | 1.31 1.24 | 1.45 1.37 |
| Gentschew and colleagues, 63 2012, New Zealand | 351 | 853 | Not stated | — | — | — | — | — | — | — | — | — | — | 1.37±0.01 | 1.41±0.01 |
| Wendland and colleagues, 66 2001, Canada | 37 | 37 | Mixed | — | — | — | — | 64.0±4.6 | 78.4±2.9 | — | — | — | — | 0.81±0.04 | 0.80±0.04 |
a Only mean is provided where standard deviation was absent in the original reference.
b Newly diagnosed Crohn’s disease.
c Diagnosed Crohn’s disease for more than 5 years.
d Men.
e Women.
∗ P<0.05 CD vs HC.
∗∗ P<0.01 CD vs HC.
∗∗∗ P<0.001 CD vs HC.
Table 3Vitamin D concentration and prevalence of deficiency in patients with Crohn’s disease (CD) compared with an age- and sex-matched healthy control group (HC)
| Author(s), year, country | Participants | Disease activity | Vitamin D (nmol/L) | Vitamin D 25(OH)D3 (ng/mL) | Suboptimal micronutrient level | Suboptimal criteria | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| CD | HC | CD | HC | CD | HC | CD | HC | |||
| ←n→ | ←mean±standard deviation→ | ←n (%)→ | ||||||||
| Geerling and colleagues, 28 1998, The Netherlands | 32 | 32 | Remission | — | — | — | — | 18 (56.0) | 9 (28.0) | <70 nmol/L (summer and autumn) or < 25 nmol/L (winter) |
| Ardizzone and colleagues, 68 2000, Italy | 51 | 30 | Mixed | — | — | 19.5±7.5 | 18.1±7.9 | — | — | — |
| Duggan and colleagues, 27 2004, Ireland | 44 | 44 | Remission | 75.0±28.7 | 105.3±55.5 | — | — | 3 (6.8) | 2 (4.5) | <40 nmol/L |
| Tajika and colleagues 51 2004, Japan | 33 | 15 | Mixed | — | — | 15.2±6.5 | 16.9±5.2 | 9 (27.3) | 1 (6.7) | <10 ng/mL |
| Gilman and colleagues, 31 2006, Ireland | 47 | 47 | Remission | 71.6±33.0 | 113±69.2 | — | — | 9 (19.1) | 2 (4.3) | <40 nmol/L |
| Joseph and colleagues, 49 2009, India | 34 | 34 | Mixed | — | — | 16.3±10.8 | 22.8±11.9 | 27 (79.0) | 17 (50.0) | <20 ng/mL |
| Suibhne and colleagues, 33 2012, Ireland | 81 | 70 | Mixed | 47.8±27.3 | 51.9±24.5 | — | — | 51 (63.0) | 36 (51.0) | <50 nmol/L |
| Grunbaum and colleagues, 32 2013, Canada | 34 | 48 | Remission | 71.1±31.1 | 68.3±26.2 | — | — | 10 (29.4) | 11 (22.9) | <50 nmol/L |
| Salacinski and colleagues, 57 2013, United States | 19 | 19 | Remission | — | — | 32.0±9.1 | 35.3±11.1 | 2 (10.5) | 1 (5.3) | <20 ng/mL |
| Dumitrescu and colleagues, 69 2014, Romania | 14 | 94 | Mixed | — | — | 23.0±10.0 | 31.0±13.0 | 5 (36.0) | 19 (20.0) | <20 ng/mL |
| Tan and colleagues, 70 2014, China | 107 | 122 | Mixed | — | — | 11.6±5.0 | 12.9±4.4 | — | — | — |
a To convert nmol/L vitamin D to ng/mL, multiply nmol/L by 0.4006. To convert ng/mL vitamin D to nmol/L, multiply ng/mL by 2.496. Vitamin D of 70 nmol/L=28.042 ng/mL.
b No statistical test reported comparing CD and HC.
∗ P<0.05 for CD vs HC.
∗∗ P<0.01 for CD vs HC.
∗∗∗ P<0.001 for CD vs HC.
Disease activity was reported in all but three of the studies.
63
, 65
, 66
The remaining studies reported micronutrient concentrations in either patients with Crohn’s disease in remission28
, 62
or in a heterogeneous patient group.29
, 50
, 62
, 63
, 65
, 66
, 67
There were no studies comparing micronutrient differences between active and remission Crohn’s disease, although the validity of measuring micronutrients in active disease is questionable. In clinical practice, and in the included studies, micronutrients are quantified in the plasma fraction of blood. However, inflammatory responses in patients with active Crohn’s disease have been found to decrease plasma micronutrient concentrations by decreasing albumin, independent of their actual body stores.72
Micronutrients on circulating erythrocytes provide a more accurate marker of micronutrient stores, particularly for zinc, copper, selenium, riboflavin and vitamin B-6, but this analysis is not available in routine clinical practice. Indeed, the transport protein for copper increases in the acute phase response, which may explain one study’s finding of significantly higher serum levels of copper in patients with Crohn’s disease compared with HC.65
In summary, the majority of studies reported lower mean levels of circulating micronutrients in patients with Crohn’s disease compared with HC, including folic acid, vitamin B-12, vitamin C, vitamin D, zinc, and selenium.
28
, 29
, 50
, 62
, 63
, 65
, 66
The majority of studies reported higher prevalence of vitamin D deficiency in patients with Crohn’s disease compared with HC.28
, 31
, 33
, 49
, 51
, 69
Whilst the review findings do not support the routine measurement of vitamin B-6 and thiamine in all Crohn’s disease patients, consideration must be given to their jejunal absorption site. For patients with small bowel disease or previous resection, it is common practice to measure micronutrients absorbed at the jejunum every 3 to 6 months.73
See Figure 2 for key micronutrients that should be measured in Crohn’s disease in clinical practice, and their accuracy in reflecting body stores during the acute phase response.Dietary Assessment Outcomes
Eleven studies assessed dietary intake and the main findings are summarized in Table 4.
15
, 16
, 25
, 27
, 28
, 29
, 38
, 40
, 45
, 46
, 62
Energy intake was similar between patients with Crohn’s disease and HC in eight studies15
, 16
, 17
, 25
, 27
, 28
, 29
, 62
and lower in the other three studies,40
, 45
, 46
especially in patients with a lower BMI.40
, 45
Although nine studies15
, 25
, 28
, 29
, 38
, 40
, 45
, 46
, 62
measured protein intake, seven of these found no significant differences in intakes between groups (Table 4).15
, 28
, 29