Nutrition 22 (2006) 855– 859
www.elsevier.com/locate/nut
Applied nutritional investigation
Malabsorption is a major contributor to underweight in Crohn’s disease
patients in remission
Nachum Vaisman, M.D.a,b,*, Iris Dotan, M.D.a,b, Aharon Halack, M.D.c, and
Eva Niv, M.D.a,b,c
a
Unit of Clinical Nutrition, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Inflammatory Bowel Disease Service, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
c
Department of Gastroenterology and Liver Diseases, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
b
Manuscript received October 5, 2005; accepted May 10, 2006.
Abstract
Objective: Undernutrition has been reported in 65–75% of patients with Crohn’s disease. The
present study aimed at identifying the relative contribution of malnutrition-causing factors in
patients with Crohn’s disease in remission.
Methods: Sixteen patients with Crohn’s disease (age 19 –57 y) in remission (Crohn’s Activity
Disease Index ⬍ 150) were included in the study. Their weight was stable for ⬎3 mo and they were
off steroids. They all completed 3-d food records and concomitantly collected stools. Self-reported
food records were analyzed and energy content in stools was determined by a direct bomb
calorimeter. Resting energy expenditure (REE) was studied by indirect calorimetry and body
composition by dual-energy X-ray absorptiometry. The study cohort was divided into two groups,
with a body mass index (BMI) equal to 18.5 kg/m2 serving as a cutoff point.
Results: Subjects with lower BMIs tended to have less lean body mass (P ⫽ 0.006), less bone
mineral density (P ⫽ 0.006), and lower REE (P ⫽ 0.003). No correlation was found between BMI
and energy intake but the percentage of malabsorption was negatively correlated with BMI (P ⫽
0.07). When dividing the study based on a BMI of 18.5 kg/m2, no difference was found in caloric
intake or REE between groups but subjects with lower BMIs had significant prominent malabsorption compared with the others (21.1 ⫾ 9.8% versus 11.7 ⫾ 3.5%, P ⫽ 0.015).
Conclusion: In the presence of similar energy intake, REE does not seem to contribute to lower BMI,
although nutrient malabsorption is higher in malnourished patients with Crohn’s disease in remission. We
suggest that malabsorption be evaluated in patients with Crohn’s disease who fail to gain weight during
disease remission to establish their extra caloric requirements. © 2006 Elsevier Inc. All rights reserved.
Keywords:
Crohn’s disease; Resting energy expenditure; Malabsorption
Introduction
Proper nutritional status is the outcome of a diet sufficient to
meet an individual’s needs and to keep body composition and
function within the normal range [1]. Disease-related malnutrition is a ubiquitous problem, with adverse effects on body
structure, function, and clinical outcome [2]. However, nutritional needs may differ according to other parameters, such as
energy expenditure, intestinal absorption, and nutrient losses
through different organs, mainly the gastrointestinal tract. One
* Corresponding author. Tel./fax: ⫹972-3-697-3973.
E-mail address: [email protected] (N. Vaisman).
0899-9007/06/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.nut.2006.05.013
of the most widely used indicators of malnutrition is the body
mass index (BMI), and a BMI ⬍18.5 kg/m2 has been frequently used as a cutoff point for undernutrition [3]. Increasing
nutritional intake may be one of the most effective methods of
treating disease-related malnutrition.
Crohn’s disease (CD) is a chronic intestinal disorder of
unknown etiology that may involve any part of the gastrointestinal tract. The small bowel is involved in 70% of CD
patients [4], and undernutrition has been reported in 65–
75% of these patients [5,6]. The pathogenic mechanisms
include poor dietary intake, increased energy expenditure,
nutrient malabsorption, and intestinal losses. The effect of
the interaction of these potential mechanisms has been only
856
N. Vaisman et al. / Nutrition 22 (2006) 855– 859
Materials and methods
tables. Patients were fed activated charcoal at the beginning
of the first and fourth days to delineate the stool collection
period. The collected stools were stored at ⫺20°C until
analysis. Fecal gross energy was determined on lyophilized
aliquots of stools using ballistic bomb calorimetry (Parr,
Moline, IL, USA). Fat content was determined gravimetrically after acid hydrolysis [9]. The percentage of malabsorption was calculated by dividing the caloric content of
the stools by the concomitant 3-d caloric intake.
Patients
Statistical analysis
The study cohort was recruited from the outpatient inflammatory bowel disease clinic at the Tel Aviv Sourasky
Medical Center (Tel Aviv, Israel). The diagnosis of CD was
based on clinical, morphologic, and histopathologic criteria.
The study inclusion criteria were (1) age ⱖ18 years, (2)
disease remission ⬎3 mo as determined by a Crohn’s Disease Activity Index score ⬍150 [7], (3) no steroid treatment
for ⬎3 mo, (4) no other chronic diseases, and (5) stable
body weight during the 3 mo preceding the study.
Patients with fistulae, ileostomy, or colostomy were excluded. A detailed medical history was taken regarding the
disease course, previous surgeries, disease location, clinical
symptoms, and medication. BMI was calculated based on
weight and height measurements, and a BMI of 18.5 kg/m2
was taken as the cutoff point between undernutrition and a
normal nutritional state [3].
Pearson’s correlation coefficients were calculated to examine the association between BMI and other continuous
parameters. Subjects were also assigned to one of two
groups based on BMI (BMI ⭵18.5 kg/m2) and comparison
between groups of patients in terms of demographic and
clinical factors was performed with the Mann-Whitney and
Fisher’s exact tests, as applicable. The difference between
groups with respect to all clinical parameters adjusted for
gender was examined using one-way analysis of covariance,
with gender as a covariate based on ranks. Pearson’s partial
correlation, controlled for gender, was also computed for
BMI and energy expenditure measurements. The statistical
significance level was set at 0.05. SPSS software (SPSS,
Inc., Chicago, IL, USA) was used for statistical analysis.
partly investigated. Moreover, most studies were performed
on hospitalized patients with active disease; this is especially important because inflammatory processes may affect
different metabolic features. The present study aimed at
defining the relative contribution of undernutrition-causing
factors in CD patients in remission.
Results
Nutritional assessment
Resting energy expenditure measurements.
Resting energy expenditure (REE) was measured by an
open-circuit, indirect calorimeter (Deltatrac, Helsinki, Finland). Patients fasted (water drinking was allowed) from
2000 h the night before the test until the next morning.
Patients lied supine for 30 min before commencing the
study at 0800 h. After calibration with standardized oxygen
and carbon dioxide gas concentrations (95% O2 5% CO2), a
plastic canopy was placed over the patient’s head and REE
was measured for 1 h. There was a 10-min washout period
before starting data collection. The interindividual coefficient of variation in our laboratory is ⬍3%. The results were
compared with the Harris-Benedict prediction equations [8].
The normative values of our laboratory are within 90% to
110% of that predicted by the above equations.
Body composition measurements.
Lean body mass (LBM) and body fat mass were measured by dual-energy X-ray absorptiometry (Lunar, Madison, WI, USA). These measurements and those of REE
were taken after an overnight fast and after voiding.
Food intake records, stool collection, and caloric intake.
Patients were instructed to keep 3-d food records and to
collect stools throughout the same 3 d. The food records
were analyzed according to the Israeli food composition
Sixteen patients who met the predefined inclusion criteria were recruited. The group consisted of 6 women and 10
men, with an age range of 21–50 y and a BMI range of
14.7–24.8 kg/m2. Subjects were also categorized into two
groups based on BMI (⬍18.5 versus ⬎18.5 kg/m2). Each
group consisted of eight patients. The group with normal
BMI levels included two women compared with four in the
underweight group. Seven patients in each group had ileocecal and ileocolonic involvement and one patient in each
group had Crohn’s colitis. Four patients in the underweight
group and one patient in the normal BMI group underwent
an ileocecal resection.
Patients’ characteristics, including body composition parameters, are presented in Table 1.
In the entire group, BMI was positively correlated with
LBM (P ⫽ 0.006), bone mineral density (P ⫽ 0.006), REE
(P ⫽ 0.003), and REE as a percentage of that predicted by
the Harris-Benedict equations (REEPP; P ⫽ 0.017). No
correlation was found between daily energy intake and
BMI, but percentage of malabsorption was negatively correlated with BMI (P ⫽ 0.07). As expected, REE was positively correlated with LBM (P ⫽ 0.0001), which itself was
positively correlated with protein intake (P ⫽ 0.05). A
significant negative correlation was found between malabsorption and REEPP.
After dividing the group by a BMI of 18.5 kg/m2 and
after adjusting for gender based on ranks, it emerged that
N. Vaisman et al. / Nutrition 22 (2006) 855– 859
Table 1
Patients’ characteristics*
Female/male
Age (y)
BMI (kg/m2)
Body fat (%)
Lean body mass (kg)
BMD (g/cm2)
T score
BMI ⱖ 18.5
kg/m2
BMI ⬍ 18.5
kg/m2
2/6
29.6 ⫾ 7.8
21.1 ⫾ 1.8
21.5 ⫾ 10.0
49.6 ⫾ 7.9
1.16 ⫾ 0.12
⫺0.12 ⫾ 1.23
4/4
36.7 ⫾ 12.4
17.0 ⫾ 1.2
17.5 ⫾ 7.2
39.6 ⫾ 9.8
1.06 ⫾ 0.08
⫺1.27 ⫾ 0.62
P†
0.36
⬍0.0001
0.021
0.038
0.15
0.096
BMD, bone mineral density; BMI, body mass index
* Mean ⫾ SD.
†
Analysis of covariance adjusted for gender based on ranks.
patients with lower BMI levels had significant lower body
fat percentages and lower LBM (Table 1).
Table 2 lists the nutritional parameters of the two groups.
There was no difference in caloric intake between groups or
in the relative contribution of the different macronutrients
composing their diet (carbohydrates, fats, and proteins).
REE tended to be higher in the normal-weight group (BMI
⬎ 18.5 kg/m2), but it was not different after adjusting for
gender based on ranks (Fig. 1). When REE was expressed as
REEPP, it was in the normal range in both groups (90 –
110%) but was statistically higher in the better nourished
subjects (P ⬍ 0.036). Four patients in the higher BMI group
(⬎18.5 kg/m2) and two in the malnourished group had
REEPP values above the normal range (REEPP⬎ 110%).
There was no difference in the respiratory quotient.
Energy malabsorption was significantly higher in the
lower BMI group (21.1 ⫾ 9.8% versus 11.7 ⫾ 3.5%, P ⫽
0.015; Table 2; Fig. 1). Fat malabsorption as a percentage of
fat intake was also higher in this group but did not reach
statistical significance (16.48 ⫾ 12.05% versus 9.01 ⫾
4.05%).
Discussion
Our results indicate that nutrient malabsorption seemed
to be the most important contributor to underweight in our
patients. Different quantities of caloric intake did not seem
to cause the difference in BMI and neither did an increased
REE because REE was lower in patients with lower BMI.
Underweight is a frequent finding in patients with CD
and is usually attributed to the above-mentioned factors.
Decreased caloric intake of patients with CD may be attributed to several factors, such as anorexia, abdominal pain,
nausea, vomiting, and intestinal obstruction. Anorexia associated with inflammation appears to be a major contributor [10] and decreased caloric intake is frequently observed
in patients with CD with high disease activity [11]. Our
patients were all in remission for ⬎3 mo before study entry
and had a Crohn’s Disease Activity Index score ⬍150. This
may explain why we found no significant differences in
caloric intake between the low and normal BMI groups. Our
857
results are in accordance with previous studies, such as
those by Geerling et al. [12] and Lanfranchi et al. [13] who
found no differences in caloric intake in patients with CD
during remission.
It has long been hypothesized that increased energy expenditure contributes to weight loss in patients with CD. However,
studies on energy metabolism in these patients have been
contradictory [14]; some have indicated that energy expenditure is not significantly different between patients and healthy
controls [15,16], whereas others demonstrated an increased
[17–19] or even decreased [20] REE. Our study suggests that
REE is within normal range when expressed as REEPP. Moreover, a significantly positive correlation was found between
BMI and REE and between BMI and REEPP, meaning that the
more malnourished subjects have lower REE. After categorizing two groups based on BMI, REEPP in the better nourished
subjects was in the upper limits of the normal range and four
of eight subjects had an REEPP1 ⬎110%. In the undernourished group, only two subjects had an REEPP above normal,
and as a group undernourished subjects tended to have a
significantly lower expense. Nevertheless, we have demonstrated a significant difference in REEPP between patients with
normal BMI and those who are underweight. This difference
may explain some of the contradictory findings in the literature
with respect to energy expenditure of these patients.
Malabsorption in patients with CD was mainly found in
patients after ileal resection, affecting fat and vitamin B12
absorption [21,22] and other vitamins and trace elements
[23]. Filipsson et al. [21] studied fecal excretion in patients
with CD before intestinal resection and found predominantly mild steatorrhea in 24% of patients with ileal disease,
26% of those with ileocolonic involvement, and 17% of
those with Crohn’s colitis. After resection of the ileum and
ileocecal valve, fecal fat excretion increased. The frequency
and severity of steatorrhea correlated with the extent of ileal
resection and was infrequent and mild when resections were
⬍30 cm in length. In our study, a marginally negative
significant correlation was found between BMI and malabsorption (P ⬍ 0.07). After categorizing the study population
Table 2
Energy intake, resting energy expenditure, and energy malabsorption in
16 patients with Crohn’s disease
Daily caloric intake (kcal/d)
Contribution (%)
Carbohydrate
Fat
Protein
REE (kcal/d)
REEPP (%)
RQ
Malabsorption (%)
Fat excretion/intake (%)
BMI ⱖ 18.5
kg/m2
BMI ⬍ 18.5
kg/m2
P*
2097 ⫾ 575
2197 ⫾ 845
0.70
54 ⫾ 9
28 ⫾ 7
18 ⫾ 9
1750 ⫾ 224
110.6 ⫾ 5.2
0.83 ⫾ 0.04
11.7 ⫾ 3.5
9.01 ⫾ 4.05
51 ⫾ 10
32 ⫾ 6
16 ⫾ 4
1432 ⫾ 297
104.2 ⫾ 5.6
0.82 ⫾ 0.04
21.1 ⫾ 9.8
6.48 ⫾ 12.05
0.067
0.036
0.86
0.007
0.46
BMI, body mass index; REEPP, resting energy expenditure based on
Harris-Benedict equations; RQ, respiratory quotient
* Analysis of covariance adjusted for gender based on ranks.
858
N. Vaisman et al. / Nutrition 22 (2006) 855– 859
Fig. 1. The different components of energy balance in 16 patients with Crohn’s disease in remission based on a body mass index ⬎18.5 kg/m2 (lighter bars)
versus ⬍18.5 kg/m2 (darker bars). *P ⫽ 0.067, **P ⫽ 0.007. REE, resting energy expenditure. BMI, body mass index; REE, resting energy expenditure.
into two groups, both groups had increased malabsorption,
but underweight subjects had a statistically significant increased malabsorption compared to the well-nourished subjects (P ⬍ 0.007). Fat malabsorption was increased but not
statistically different between groups. Increased protein
losses may have also contributed to the difference in total
energy malabsorption between groups.
Very few studies have investigated concomitantly energy
intake, energy absorption, and energy expenditure in patients with CD. Rigaud et al. [11] studied these parameters
in two groups of patients. One group kept a stable weight
over the 3 mo before being studied, whereas the others lost
weight during this period. These groups were mixed with
regard to disease activity and patients with malabsorption
were excluded. No differences were observed between
groups regarding REE and energy malabsorption. Therefore, the researchers concluded that decreased energy intake
is the main contributor to weight loss. Our study is different
because all our patients had a stable weight. We also asked
a different clinical question and thus chose a different set of
patients. Our question, which is frequently being encountered in the clinic, is: Why are some patients in remission
underweight despite our efforts to rehabilitate them? Therefore, our patients who were had a low disease index, and
malabsorption was not an exclusion criterion. Moreover, in
our setting, malabsorption was found to be the main contributor to the difference between groups.
Stable weight is achieved when net digestible energy
(energy intake minus energy malabsorption) equals energy
expenditure. Our patients kept a stable weight for the last 3
mo before the study and were, therefore, in energy balance.
Nevertheless, undernourished patients had significant malabsorption. What are the possible compensating mechanisms that enable them to keep a stable weight? In the past,
undernutrition was found to decrease REE [24]. Because
our underweight patients with CD suffer most probably
from undernutrition, they may experience a relative reduction in REE. This is supported by the negative correlation
between REEPP and malabsorption in our study (P ⬍
0.037). This may also explain why the REEPP of our undernourished patients was significantly lower than that of
the well-nourished subjects. Reduced REE may be a compensatory response for the caloric deficit due to malabsorption and may help to maintain a stable weight. In contrast to
undernourished subjects, patients with normal BMI levels
consume, most probably, enough calories and absorb
enough to keep a stable weight and REE within the upper
limit of the normal range. To study this hypothesis, we
classified our patients into those who had a normal REE
(REEPP ⬍ 110%) and those who had above normal expense
(⬎110%). Subjects with the lower REEPP had more malabsorption than did those with above normal values (443 ⫾
360 versus 232 ⫾ 136 kcal/d, respectively; P ⫽ 0.069).
Another possible compensating mechanism for malabsorption is an increased energy intake. The difference between
groups did not reach a significant level.
How can we then increase weight in our patients? Nutritional rehabilitation was found to increase REE in healthy
and diseased undernourished patients [25]. Moreover, in
undernourished patients with cystic fibrosis and significant
malabsorption, nutritional rehabilitation increased REE beyond the expected increase that was attributed to the increase in LBM [25]. This increase was suggested to indicate
a loss of the compensating mechanism. If this is the case, we
may speculate that, when nutrition intervention overcomes
malabsorption in our patients, an increase in their REE will
take place. This should be taken into account when planning
nutritional support.
N. Vaisman et al. / Nutrition 22 (2006) 855– 859
There are a few limitations to our study. The study
population was rather small due to difficulties in recruiting
subjects, mainly due to the inconvenience of the 72-h fecal
collections. The second limitation of this study is the fact
that other components of daily energy expenditure, i.e.,
physical activity and diet-induced thermogenesis, were not
measured, whereas REE comprises only 70% of daily energy expenditure. Because none of our patients had active
disease during the 3 mo (at least) before study entry, and
they were all working and maintaining a normal lifestyle, a
significant difference between groups in terms of this energy component is not very likely. The third limitation has
to do with the fact that the undernourished group had a
larger number of individuals who underwent intestinal resection, and this created some unbalance between groups.
However, this most probably reflects the higher percentage
of undernutrition in patients who had resection of the ileocecal valve. Whether the results would be the same if
individuals with intestinal resection had been excluded from
the study cannot be determined due to the small size of the
groups. We are also aware that we did not have a formal
healthy control group because the main purpose of the study
was to compare well-nourished with undernourished patients with CD in remission. The REE of healthy subjects
measured in our energy laboratory was within the normal
range based on the Harris-Benedict equation (90 –110%)
and the norms of our laboratory for energy and fat malabsorption are 11% and 8%, respectively.
In conclusion, we have demonstrated that malabsorption
of nutrients is most probably the most significant contributor to underweight in patients with CD in remission because
REE was found to be positively correlated with BMI and
energy intake did not correlate significantly. We suggest,
therefore, measuring intestinal absorption in patients with
CD who fail to gain weight during disease remission in
addition to REE whenever possible and, accordingly, to
look for alternative routes to increase net caloric intake.
Acknowledgments
Esther Eshkol is thanked for editorial assistance.
References
[1] Jeejeebhoy KN, Detsky AS, Baker JP. Assessment of the nutritional
status. JPEN 1990;14(suppl):193S– 6.
[2] Baxter JP. Problems of nutritional assessment in the acute setting.
Proc Nutr Soc 1999;58:39 – 46.
[3] Physical status: the use and interpretation of anthropometry. Technical Report Series 854. Geneva: World Health Organization; 1995.
[4] Munkholm P. Crohn’s disease— occurrence, course and prognosis.
An epidemiologic cohort study. Dan Med Bull 1997;44:287–302.
859
[5] Fleming CR. Nutrition consideration in patients with Crohn’s disease.
Semin Colon Rectum Surg 1994;5:167–73.
[6] Zurita VF, Rawls DE, Dyck WP. Nutritional support in inflammatory
bowel disease. Dig Dis 1995;503:92–107.
[7] Best WR, Becktel JM, Singleton JW, Kern F Jr. Development of a
Crohn’s disease study. Gastroenterology 1976;70:439 – 44.
[8] Harris JA, Benedict FG. A biometric study of basal metabolism.
Publication 279. Washington, DC: Carnegie Institution of Washington; 1919.
[9] Sklan D, Arieli A, Chalupa W, Krolfelz DS. Digestion and absorption
of lipid and bile acids in sheep fed stearic acid, oleic acid or tristearil.
J Dairy Sci 1985;68:1667–75.
[10] Tache Y. Experimental colitis and reduction of food intake: what are
the mechanisms? Gastroenterology 2000:118:625– 6.
[11] Rigaud D, Angel LA, Cerf M, Carduner MJ, Melchior JC, Santier C,
et al. Mechanisms of decreased food intake during weight loss in
adult Crohn’s disease patients without obvious malabsorption. Am J
Clin Nutr 1994;60:775– 81.
[12] Geerling BJ, Badart-Smook A, Stockbrugger RW, Brummer RJ.
Comprehensive nutritional status in patients with long-standing
Crohn’s disease currently in remission. Am J Clin Nutr 1998;67:
919 –26.
[13] Lanfranchi GA, Brignola C, Campieri M, Bazzocchi G, Pasquali R,
Bassein L. Assessment of nutritional status in Crohn’s disease in
remission or low activity. Hepatogastroenterology 1984;31:129 –32.
[14] Gassull MA, Cabre E. Nutrition in inflammatory bowel disease. Curr
Opin Clin Nutr Metab 2001;4:561–9.
[15] Stokes MA, Hill GL. Total energy expenditure in patients with
Crohn’s disease: measurement by the combined body scan technique.
JPEN 1993;17:3–7.
[16] Schneeweiss B, Lochs H, Zauner C, Fischer M, Wyatt J, MaierDobersberger T, et al. Energy and substrate metabolism in patients
with active Crohn’s disease. J Nutr 1999;129:844 – 8.
[17] Klien S, Meyers S, O’Sullivan P, Barton D, Leleiko N, Janowitz HD.
The metabolic impact of active ulcerative colitis. Energy expenditure
and nitrogen balance. J Clin Gastroenterol 1988;10:34 – 40.
[18] Kushner RF, Schoeller DA. Resting and total energy expenditure in
patients with inflammatory bowel disease. Am J Clin Nutr 1991;53:
161–5.
[19] Azcue M, Rashid M, Griffiths A, Pencharz PB. Energy expenditure
and body composition in children with Crohn’s disease: effect of
enteral nutrition and treatment with prednisolone. Gut
1997;41:203– 8.
[20] Chan ATH, Fleming R, O Fallon WM, Huizenga KA. Estimated
versus measured basal energy requirements in patients with Crohn’s
disease. Gastroenterology 1986;91:75– 8.
[21] Filipsson S, Hulten L, Lindstedt. G. Malabsorption of fat and vitamin
B12 before and after intestinal resection for Crohn’s disease. Scand J
Gastroenterol 1978;13:529 –36.
[22] Papazian A, Minaire Y, Descos L, Andre C, Melange M, Vignal J.
Relationships between the extent of ileal lesion of resection and
vitamin B12, bile acids and fat absorption. Hepatogastroenterology
1981;28:106 –9.
[23] Geerling BJ, Stockbrugger RW, Brummer RJM. Nutrition and inflammatory bowel disease: an update. Scand J Gastroenterol 1999;
34(suppl 230):95–105.
[24] Vaisman N, Rossi MF, Goldberg E, Dibden LJ, Wykes LJ, Pencharz
PB. Energy expenditure and body composition in patients with anorexia nervosa. J Pediatr 1988;113:919 –24.
[25] Vaisman N, Clarke R, Pencharz PB. Nutritional rehabilitation increases resting energy expenditure without affecting protein turnover
in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 1991;13:
383–90.