AJCN EB Program 2010
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Clin Nutr 89: 240-247, 2009. First published December 3, 2008; doi:10.3945/ajcn.2008.26219
 American Journal of Clinical Nutrition, doi:10.3945/ajcn.2008.26219
Vol. 89, No. 1, 240-247, January 2009
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
89/1/240    most recent
ajcn.2008.26219v1
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Whelan, K.
Right arrow Articles by Taylor, M. A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Whelan, K.
Right arrow Articles by Taylor, M. A
Agricola
Right arrow Articles by Whelan, K.
Right arrow Articles by Taylor, M. A
© 2009 American Society for Clinical Nutrition

ORIGINAL RESEARCH COMMUNICATION

Fecal microbiota in patients receiving enteral feeding are highly variable and may be altered in those who develop diarrhea1,2,3

Kevin Whelan, Patricia A Judd, Kieran M Tuohy, Glenn R Gibson, Victor R Preedy and Moira A Taylor

1 From the Nutritional Sciences Division, Diet and Gastrointestinal Health, King's College London, London, United Kingdom (KW and VRP); the Lancashire School of Health and Postgraduate Medicine, University of Central Lancashire, Preston, United Kingdom (PAJ); the Food Microbial Sciences Unit, Department of Food Biosciences, University of Reading, Reading, United Kingdom (KMT and GRG); and the School of Biomedical Sciences, University of Nottingham, Nottingham, United Kingdom (MAT).

2 Supported by an unrestricted grant from King's College London and Nestlé UK.

3 Reprints not available. Address correspondence to K Whelan, Nutritional Sciences Division, King's College London, 150 Stamford Street, London SE1 9NH, United Kingdom. E-mail: kevin.whelan{at}kcl.ac.uk.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: The pathogenesis of diarrhea in patients receiving enteral feeding includes colonic water secretion, antibiotic prescription, and enteropathogenic colonization, each of which involves an interaction with the gastrointestinal microbiota.

Objective: The objective was to investigate temporal changes in the concentrations of fecal microbiota and short-chain fatty acids (SCFAs) in patients starting 14-d of enteral feeding and to compare these changes between patients who do and do not develop diarrhea.

Design: Twenty patients starting exclusive nasogastric enteral feeding were monitored for 14 d. Fecal samples were collected at the start, middle, and end of this period and were analyzed for major bacterial groups by using culture independent fluorescence in situ hybridization and for SCFAs by using gas-liquid chromatography.

Results: Although no significant changes in fecal microbiota or SCFAs were observed during enteral feeding, stark alterations occurred within individual patients. Ten patients (50%) developed diarrhea, and these patients had significantly higher concentrations of clostridia (P = 0.026) and lower concentrations (P = 0.069) and proportions (P = 0.029) of bifidobacteria. Patients with and without diarrhea had differences in the proportion of bifidobacteria (median: 0.4% and 3.7%; interquartile range: 0.8 compared with 4.3; P = 0.035) and clostridia (median: 10.4% and 3.7%; interquartile range: 14.7 compared with 7.0; P = 0.063), respectively, even at the start of enteral feeding. Patients who developed diarrhea had higher concentrations of total fecal SCFAs (P = 0.044), acetate (P = 0.029), and butyrate (P = 0.055).

Conclusion: Intestinal dysbiosis occurs in patients who develop diarrhea during enteral feeding and may be involved in its pathogenesis.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Enteral feeding is a common method of nutritional support for patients unable to achieve their nutritional requirements through oral diet alone. However, diarrhea is a potential consequence that has been reported in 2–95% of patients (1, 2) or on 2–26% of patient days (3). Diarrhea during enteral feeding may result in many problematic complications, including fluid and electrolyte abnormalities (4), fecal incontinence (5), and burdens for the patient and their nurse or caregiver (6).

The pathogenesis of diarrhea is multifactorial. Intragastric enteral feeding results in an abnormal secretion of water into the ascending colon (7). Meanwhile, the incidence of diarrhea is higher in those who are prescribed antibiotics (8), and enteral feeding is associated with an increased risk of Clostridium difficile–associated diarrhea (CDAD) (9). Each of these mechanisms involves an interaction with the gastrointestinal (GI) microbiota. For example, the microbiota may prevent infection through pathogen inhibition (10) and they ferment carbohydrates and proteins to produce short-chain fatty acids (SCFAs) that stimulate colonic water absorption (11). Meanwhile, antibiotics result in major alterations to both the microbiota (12) and SCFAs (13).

The GI microbiota is a highly complex and diverse microbial ecosystem consisting of >500 species of bacteria (14). Investigating their role in clinical nutrition has been impeded by the subjective limitations of bacterial culture for their analysis. However, nucleotide probes targeting the 16S rRNA of specific bacterial groups (14) have enabled identification of alterations in the luminal microbiota in inflammatory bowel disease (15) and CDAD (16).

Despite their potential role in the pathogenesis of diarrhea, few studies have used genotypic techniques to investigate the GI microbiota during enteral feeding (17). One study showed that enteral formula results in major reductions in total fecal bacteria in healthy subjects (18), whereas low concentrations of bifidobacteria were found in patients receiving short-term (19) and long-term (20) enteral feeding.

It is unclear whether such alterations in the GI microbiota are associated with diarrhea in patients receiving enteral feeding. Therefore, through collection and analysis of serial samples, the aim of this study was to investigate temporal changes in the concentrations of fecal microbiota and SCFAs in patients starting 14-d of enteral feeding and to compare these changes between those who did and did not subsequently develop diarrhea.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patients
Patients starting exclusive nasogastric enteral feeding with standard, fiber-free formula were recruited from St George's Hospital, London, United Kingdom. To reduce the diversity in the patient population recruited, the exclusion criteria included patients in the intensive care unit, who may have been receiving numerous diverse antibiotics at various doses. Other exclusion criteria included those who had received fiber, probiotic, or prebiotic supplements within the previous month and those with a GI disorder likely to result in altered microbiota or SCFAs (eg, irritable bowel syndrome, inflammatory bowel disease, and previous GI surgery).

The study was approved by the Local Research Ethics Committees of King's College London and the Merton, Sutton, and Wandsworth Health Authority. Before recruitment to the study, informed consent was obtained from the patient or from their next of kin and the medical team if unconscious at the time that assent was obtained.

Study design
In this prospective observational study, patients received enteral feeding via a nasogastric tube as clinically indicated. The volume of formula prescribed was based on each patient's total energy expenditure, which was calculated by adjusting basal metabolic rate [calculated by using modified Schofield equations (21)] for physical activity and disease (22). The energy contents of the formulas were 1.0 kcal/mL (Osmolite; Abbott UK, Maidenhead, Berks, United Kingdom), 1.2 kcal/mL (Osmolite Plus; Abbott UK), or 1.5 kcal/mL (Ensure Plus; Abbott UK), depending on each patient's calculated expenditure. Formula was delivered through a nasogastric tube via a pump infusion starting at a rate of 25 mL/h and increasing by 25 mL/h every 4 h until the full feeding rate was achieved (range:75–125 mL/h).

For the purposes of this study, patients were monitored for 14 d from passage of the first fecal sample after initiation of enteral feeding. Fecal output was recorded by nursing staff using a stool chart validated for use in this patient group (23). The chart facilitates accurate characterization of fecal frequency, consistency, and weight, which are then summarized into a daily stool score whereby a score of ≥15 is used to indicate diarrhea. Wards were visited 3 times/d on weekdays and were contacted by telephone on weekends to monitor data recording by the nursing staff. Demographic data (eg, age and sex) and clinical details were obtained from the patients’ medical notes, which were also monitored daily to obtain information on antibiotic prescription.

Whenever possible, a fecal sample was collected for analysis on days 1, 7, and 14. To ensure accurate measurement of microbiota and SCFAs, only fresh fecal samples were analyzed. However, in the absence of habitual or meal cues to stool output (24), feces were sometimes not voided on these days or were voided during the night, which precluded their use. Samples were therefore collected and analyzed from each patient during 4-d periods at the start (days 1–4), middle (days 6–9), and end (days 11–14) of the 14 d of enteral feeding.

Fecal microbiology
Fecal samples were collected within 1 h of voiding and were immediately transported to the laboratory for analysis of microbiota, SCFAs, pH, and C. difficile enterotoxins. Predominant components of the fecal microbiota were quantified by using fluorescent in situ hybridization (25). Briefly, fecal bacteria were harvested in phosphate-buffered saline (0.1 mol/L; pH 7·0) with the use of a stomacher and were then fixed overnight in 4% (wt:vol) paraformaldehyde. Bacteria were then washed and suspended in a 1:1 solution of phosphate-buffered saline and 96% (vol:vol) ethanol and stored at –20°C until hybridization. Total bacteria were enumerated by using the nucleic acid stain 4,6-diamidino-2-phenylindole for total cell counts (26). Meanwhile, individual species were enumerated through hybridization with oligonucleotide probes targeting specific regions of 16S rRNA (Table 1). In the absence of a probe targeting all clostridia, probes were used that would target C. lituseburense (cluster XI, including C. difficile), C. histolyticum (clusters I and II, including C. perfringens), and C. coccoides-E. rectale (cluster CIV). Probes were manufactured and labeled with the indocarbocyanin fluorescent marker Cy3. The bacteria were hybridized with the probe (4.5 ng/µL) in hybridization buffer (0.9 mol/L NaCl; 0.02 mol/L Tris/HCl; 0.01% wt:vol sodium dodecyl sulfate) and incubated overnight at the relevant temperature for that probe. The hybridization mix was then vacuum filtered, and the filter was mounted onto a microscope slide, which was then examined by using fluorescence microscopy on an Axioplan 2 microscope (Zeiss, Göttingen, Germany) equipped with an HBO-100 fluorescent lamp (Osram, Munich, Germany). Duplicate slides were enumerated manually through random selection of 15 fields. All hybridization and enumeration was conducted at the University of Reading by microbiologists blinded to all patient details, including whether the patient developed diarrhea.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Oligonucleotide probes used for fluorescence in situ hybridization

 
Fecal SCFAs were analyzed by using gas-liquid chromatography. Samples were frozen at –80°C within 1 h of voiding and stored until analyzed. SCFAs were extracted in a buffer (1% wt:vol H3PO4; 0.1% wt:vol HgCl2) containing 2,2-dimethylbutyric acid as an internal standard. In duplicate, the extracted SCFAs were injected splitless into a Hewlett-Packard 6890 series gas-liquid chromatography system equipped with a 530-µm internal diameter and a 30-m fused silica capillary column with a film thickness of 1 µm (J&W Scientific, Koeniz, Switzerland). The initial oven temperature was 80°C, which increased by 10°C/min until 145°C and then to 100°C/min until 200°C to ensure complete elution. All chromatograms were automatically integrated on a Hewlett-Packard Chemstation.

Fecal bacteria and SCFA concentrations were expressed per gram of dry feces to standardize comparisons between samples with different water contents. Fecal water content was determined by measuring the weight loss of a 3–5-g sample after lyophilization. Duplicate preweighed samples were dried at –45°C in a lyophilizer for ≥7 d until constant weight was recorded.

Fecal pH was measured with a pH electrode and meter. Fresh fecal samples were diluted with buffer and homogenized in a stomacher; pH was measured by using an epoxy electrode designed for use in slurries (pHASE, BDH, London, United Kingdom), which was attached to a portable meter accurate to a pH of 0.01 (Hartman, London, United Kingdom).

C. difficile–enterotoxins A and B were detected with an enzyme-linked immunosorbent assay kit (ToxA/B II; TechLab, Blacksburg, VA) according to the manufacturer's instructions. A diagnosis of CDAD was given when a patient had diarrhea (stool score of ≥15) in conjunction with a positive enterotoxin test.

Statistical analysis
All data were analyzed by using SPSS for Windows (version 14.0; SPSS Inc, Chicago, IL). Data are presented as means ± SDs or medians (interquartile ranges; IQR) as indicated. A P value < 0.05 was considered statistically significant. Bacterial numbers were log transformed. Baseline samples were compared between age (≤75 y or >75 y), antibiotic prescription (yes or no), and diarrhea (yes or no) by using a Mann-Whitney test. A repeated-measures analysis of variance was conducted on data for fecal microbiota, SCFAs, and pH across the 3 time points (start, middle, and end). A within-subject factor analysis was used to investigate the effect of time, and a between-subject factor analysis was used to investigate the effect of age (≤75 y or >75 y), antibiotic prescription (yes or no), or diarrhea (yes or no) on these variables. Interactions between each of these variables and time were also investigated. Because of inadequate samples for analysis of SCFAs and pH at all 3 time points, mean values were calculated from each patient's pooled values across the 14-d period of enteral feeding. These mean values were then used to calculate the median and IQR for all 20 patients and for the 2 groups (no diarrhea and diarrhea), which were compared by using a Mann-Whitney test.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Patient cohort
Thirty-six patients (12 men and 24 women) were recruited to the study; however, 16 were subsequently excluded because of introduction of oral diet (n = 10), death (n = 4), transfer to another hospital (n = 1), or a decision to cease enteral feeding (n = 1) before study completion. The remaining 20 patients (5 men and 15 women) were aged 77 y and 5 mo ± 10 y and 7 mo.

Indications for enteral feeding were the risk of aspiration secondary to acute stroke (n = 15), muscular dystrophy (n = 1), multiple sclerosis (n = 1), tracheostomy placement (n = 1), disease-related anorexia secondary to fractured pelvis (n = 1), or pneumonia (n = 1). The energy prescription was 1525 ± 167 kcal/d. At baseline, the serum albumin concentration was 27.9 ± 6.4g/L, whereas 17 (85%) patients had values <35 g/L and 2 (10%) had values <20 g/L.

Eleven patients were already receiving antibiotics when the first sample was collected and had been receiving such for 5.0 ± SD d. Five patients were subsequently prescribed antibiotics during the 14 d of enteral feeding. These 16 patients received 1 (n = 3 patients), 2 (n = 4), 3 (n = 7), or 4 (n = 2) different courses of antibiotics. The antibiotics included a range of penicillins, cephalosporins, and quinolones delivered enterally or parenterally. Only 4 patients did not receive antibiotics before or during the 14 d of enteral feeding.

Variability in microbiota and SCFAs
There were no within-subject changes in the concentration or proportion of any of the bacterial groups measured between the start, middle, and end of the 14 d of enteral feeding (Table 2). However, far from being stable, data from individual patients indicated that some had large reductions, some had large increases, and some had very stable microbiota (Figure 1).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Fecal microbiota, short-chain fatty acids (SCFAs), and fecal pH during enteral feeding

 

Figure 1
View larger version (44K):
[in this window]
[in a new window]

  		FIGURE 1. A–E: Concentration of bacteria measured by using fluorescence in situ hybridization at the start, middle, and end of 14 d of enteral feeding in 20 patients. The continuous black line represents mean ± SD values. No significant within-subject effect of time over this period was observed for any of the bacteria measured. However, the gray lines, which represent data for each patient, indicate a dynamic microbial ecosystem. The continuous gray lines represent data for patients who did not develop diarrhea (n = 10), and the dotted gray lines represent data for those who did develop diarrhea (n = 10). The within-subject effects of time and between-subject effects of diarrhea were calculated by using repeated-measures ANOVA. No significant interactions between time and diarrhea were observed.

 
Analysis of SCFA concentrations and fecal pH at all 3 time points was only possible in 12 and 13 patients, respectively, because of insufficient samples at one of the time points in the remainder. In these patients, there were no significant within-subject changes in the concentrations of any of the SCFAs or in fecal pH between the start, middle, and end of the 14-d enteral feeding period (Table 2). However, as for fecal bacteria, the raw data showed that individual patients had highly variable concentrations of SCFAs.

Microbiota, age, and antibiotics
The effect of age and antibiotics on microbiota, both at the start and across all 3 time points during enteral feeding, was investigated. In the first fecal sample collected, there was a trend toward lower numbers of bacteroides (median: 9.6; IQR: 1.4 compared with 10.3; 1.0 log10 cells/g dry feces; P = 0.054) and significantly lower Atopobium (median: 9.6; IQR: 1.0 compared with 10.1; 0.5 log10 cells/g dry feces; P = 0.007) in patients older than 75 y (n = 12) compared with those 75 y or younger (n = 8). However, there were no between-subject effects of age on the concentration or proportion of the microbiota between the start, middle, and end of the 14 d of enteral feeding (P values ranged from 0.1 to 0.841), nor were there any age or time interactions (P values ranged from 0.1 to 0.927).

In the first fecal sample collected, there were no significant differences in concentration or proportion of the microbiota between patients who were already receiving antibiotics (n = 11) and those who were not (n = 9) (P values ranged from 0.112 to 0.882). Interestingly, the fecal water content was greater in those who were already receiving antibiotics (median: 80.5%; IQR: 6.7) than in those who were not (70.9%; 15.5; P = 0.02). Similarly, there were no significant differences in the concentration or proportion of microbiota between those who did (n = 16) and did not (n = 4) receive antibiotics at any time point during the 14-d enteral feeding period (P values ranged from 0.19 to 0.984; repeated-measures ANOVA). There was, however, an interaction between antibiotic use and time for the proportion of clostridia, which decreased across the 3 time points in those not receiving antibiotics (27.3%, 9.0%, and 4.5%) relative to the stable proportions in those receiving antibiotics (9.3%, 9.0%, and 11.3%) (P for interaction = 0.043). There were no other antibiotic use or time interactions (P values ranged from 0.134 to 0.993).

Microbiota, SCFAs, and diarrhea
Ten patients (50%) experienced diarrhea at least once during the 14-d period, and its incidence was 14% of patient days. Four (20%) patients developed CDAD. The approaches to managing diarrhea included reducing or stopping antibiotics (n = 3), reducing the volume of formula delivered (n = 2) and, in those patients with CDAD, prescribing metronidazole (n = 4).

In total, serum albumin was measured 63 times, and C-reactive protein was measured 103 times throughout the study. Although it was not a primary objective of the current study, the association between serum albumin and C-reactive protein and the fecal score was briefly evaluated. Interestingly, the fecal score negatively correlated with serum albumin concentrations (Pearson's correlation coefficient = –0.363, P = 0.003) and positively correlated with C-reactive protein (Pearson's correlation coefficient = 0.264, P = 0.007) for the days on which they were measured. However, because some patients were represented numerous times, data were also analyzed by using an analysis of covariance, which indicated that neither albumin (B value = 0.146, P = 0.698) nor C-reactive protein (B value = –0.019, P value = 0.184) was significantly associated with fecal score between patients.

The first fecal sample collected at the start of enteral feeding contained significantly lower proportions of bifidobacteria in those who subsequently developed diarrhea (n = 10) than in those who did not (n = 10), respectively (median: 0.4% compared with 3.7%; IQR: 0.8 compared with 4.3; P = 0.035), and showed a trend toward higher proportions of clostridia in those who subsequently developed diarrhea (n = 10) than in those who did not (n = 10), respectively (median: 10.4% compared with 3.7%; IQR: 14.7 compared with 7.0; P = 0.063).

When a comparison was made across all 3 time points, there were significantly higher concentrations of clostridia (P = 0.026), lower proportions of bifidobacteria (P = 0.029), and a trend to lower concentrations of bifidobacteria (P = 0.069) and higher bacteroides-prevotella (P = 0.07) in those who developed diarrhea than in those who did not develop diarrhea (Table 2, Table 3). There were no significant diarrhea or time interactions in the concentration or proportion of microbiota (Table 2).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Differences in the fecal microbiota and short-chain fatty acids (SCFAs) between patients with and without diarrhea

 
There were significantly higher concentrations of total fecal SCFA (P = 0.044), acetate (P = 0.029), and a trend toward higher concentrations of butyrate (P = 0.055) in patients who developed diarrhea (n = 5) compared with those who did not (n = 7; Table 2, Table 3). There were no significant interactions between diarrhea and time for SCFA concentrations (Table 2).

Of the 16 patients who received antibiotics, 9 (56%) developed diarrhea, whereas only 1 of those who did not receive antibiotics (25%) developed diarrhea (P = 0.582). Of the 11 patients receiving antibiotics at baseline, 7 (64%) developed diarrhea and 4 (36%) did not. This group of 11 patients was analyzed independently, and no significant differences in microbiota between those who did and did not develop diarrhea were detected in the baseline sample or across all 3 time points.

Individual values for microbiota, SCFAs, and fecal pH at the start, middle, and end of enteral feeding were pooled for each of the 20 patients, allowing inclusion of data from those who had insufficient sample for analysis of SCFAs and pH at all 3 time points. There were significantly higher concentrations (P = 0.019) and proportions (P = 0.041) of clostridia, higher proportions of bacteroides (P = 0.041), and higher concentrations of total SCFAs (P = 0.009) and acetate (P = 0.002) in patients with diarrhea (Table 4).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Pooled data for fecal microbiota, short-chain fatty acids (SCFAs), and pH during enteral feeding1

 
Interestingly, when patients were categorized into those without diarrhea (n = 10), those with CDAD (n = 4), and those with enteral feeding–associated diarrhea (diarrhea without CDAD; n = 6), there remained significantly higher concentrations of clostridia in patients with enteral feeding–associated diarrhea than in those without diarrhea (P = 0.003, Mann-Whitney plus Bonferroni post hoc correction).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Enteropathogenic infection and colonic water secretion are involved in the pathogenesis of diarrhea in patients receiving enteral feeding. The GI microbiota play an important role in preventing enteropathogenic infection, whereas the SCFAs they produce stimulate colonic water secretion. Despite the use of a relatively small sample of patients, this study showed that the fecal microbiota were highly variable and that significant differences existed between those that did and did not develop diarrhea.

At the start of the enteral feeding period, the concentration of total bacteria varied between patients by just over 10-fold, whereas for bifidobacteria it varied 1000-fold (Figure 1). This finding supports the view that the composition of the GI microbiota varies greatly between individuals.

The absence of a significant effect of time on the fecal microbiota was surprising, especially because enteral formula results in a reduction in fecal bacteria in healthy subjects over the same time period (18). However, there were sometimes large within-subject changes in fecal microbiota that occurred, albeit in nonsystematic directions (Figure 1). This instability was also observed for SCFAs and has been shown previously (31) and is likely to be caused by the absence of fermentable carbohydrate in the formula (17), the occurrence of pathological changes such as diarrhea (32), and antibiotic use.

The absence of a difference in microbiota between those prescribed and those not prescribed antibiotics was likely due to the low numbers of patients not prescribed them (n = 4), the variability in the timing, the route of delivery and the type of antibiotic prescribed (12), and the wide interindividual response of the microbiota to antibiotics (33). Antibiotics also have differential effects on fecal SCFAs. For example, vancomycin causes major reductions, erythromycin causes only moderate reductions, and doxycycline has no effect (34); each of these antibiotics were prescribed to at least one patient in the current study.

The current study was the first to use genotypic hybridization techniques to compare differences in the microbiota between a cohort of patients who did and did not develop diarrhea during enteral feeding. Those who developed diarrhea had higher fecal clostridia counts and a trend toward lower bifidobacteria counts. The latter may be the result, in part, of the use of antibiotics, such as amoxicillin, cefuroxime, erythromycin, and vancomycin (12), which reduce fecal bifidobacteria and were each prescribed to at least one patient during this study. However, lower proportions of bifidobacteria existed even in the first sample collected at the start of the 14-d period of enteral feeding, before the patients developed diarrhea and before many of these antibiotics were prescribed. This may therefore indicate that diarrhea that developed during enteral feeding was partly due to prior dysbiosis and not to antibiotic use.

Diarrhea during enteral feeding involves various mechanisms and primary alterations in absorptive physiology. Despite the small numbers involved, higher clostridia counts persisted in the 6 patients with enteral feeding–associated diarrhea (diarrhea without CDAD). Although such a dysbiosis has been reported in patients with CDAD (16), it has not been previously reported in patients with enteral feeding–associated diarrhea. Aging may be another contributor, with evidence that older persons have lower bifidobacteria and bacteroides counts than do healthy young adults (16, 35). Whether an association exists between aging and the risk of diarrhea during enteral feeding and whether this is mediated by GI dysbiosis warrants investigation.

Bifidobacteria exert antimicrobial activity against many enteropathogens (36), and their reduction in the present study may have impaired the ability of the GI microbiota to exert pathogen inhibition (10), which increased the risk of colonization with enteropathogens such as C. difficile. Although this dysbiosis is likely to be involved in the pathogenesis of diarrhea, the nature of the study design was such that a causal relation could not be proven. However, a major strength of this prospective cohort study was that it allowed collection of serial samples from unselected patients before diarrhea developed.

Whether the correction of dysbiosis will prevent diarrhea is debatable. Probiotic lactobacilli have been shown to prevent antibiotic-associated diarrhea and CDAD in hospitalized inpatients not receiving enteral feeding (37), whereas 2 controlled trials found contrasting results of their effect on the incidence of diarrhea during enteral feeding (8, 38). The results of the current study suggest that probiotic bifidobacteria may be a suitable candidate for future research. Prebiotic carbohydrates (eg, fructooligosaccharides) increase fecal bifidobacteria and reduce clostridia in healthy subjects (18), although the same effect was not confirmed in patients receiving long-term enteral feeding (20). To our knowledge, no controlled trials of the effects of prebiotics on diarrhea prevention in enteral feeding have been conducted (39).

SCFAs stimulate colonic water absorption in patients receiving enteral feeding (11). They have numerous other important roles, for example, colonocytes, the morphology and function of which may be impaired during malnutrition, use butyrate as a primary metabolic substrate (40). However, patients who developed diarrhea actually had higher concentrations of fecal SCFAs. This was likely due to faster small intestinal transit times resulting in transient carbohydrate malabsorption, which thus increases the substrate available for colonic fermentation, which is compounded by faster colonic transit that would reduce SCFA absorption (41, 42). Furthermore, exogenous and endogenous proteins undergo colonic fermentation, which results in production of ammonia and the branched-chain SCFAs isobutyrate and isovalerate (43). Therefore, the maintenance of neutral fecal pH despite elevated concentrations of SCFAs in those with diarrhea may be the result, in part, of protein fermentation and subsequent ammonia production. Interestingly, both SCFA absorption (44) and function (eg, maintenance of mucosal barrier function) (45) may be more effective at lowering the luminal pH. Therefore, research is required to investigate whether SCFAs are effective at maintaining colonic health in patients receiving concomitant enteral feeding and antibiotics.

In conclusion, hospitalized patients receiving enteral feeding have a variable GI microbiota. Higher clostridia and lower bifidobacteria counts occur in patients who develop diarrhea, and this dysbiosis may be involved in its pathogenesis. Future research to determine whether correction of this dysbiosis reduces the incidence of diarrhea in this patient group is required.


   ACKNOWLEDGMENTS
 
We thank the physicians, dietitians, and nurses of St George's Hospital (London, United Kingdom) for their assistance in this research as well as the patients for their participation. We also thank Catherine Murset-Mounoud for assistance with gas-liquid chromatography and Peter Milligan for statistical advice.

The authors’ responsibilities were as follows—KW, PAJ, VRP, and MAT: conception and design of the study and data interpretation; KW: patient recruitment, data collection, assembly, and analysis; KW, KMT, and GRG: microbiological analysis; and KW, PAJ, KMT, GRG, VRP, and MAT: drafting, revision, and approval of the manuscript. No conflicts of interest were declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Cataldi-Betcher, EL, Seltzer, MH, Slocum, BA & Jones, KW. Complications occurring during enteral nutrition support: a prospective study. JPEN J Parenter Enteral Nutr 1983;7:546–52..[Abstract/Free Full Text]
  2. DeMeo, M, Kolli, S, Keshavarzian, A, et al.. Beneficial effect of a bile acid resin binder on enteral feeding induced diarrhea. Am J Gastroenterol 1998;93:967–71..[CrossRef][Medline]
  3. Bliss, DZ, Guenter, PA & Settle, RG. Defining and reporting diarrhoea in tube-fed patients–what a mess!. Am J Clin Nutr 1992;55:753–9..[Abstract/Free Full Text]
  4. Stroud, M, Duncan, H & Nightingale, J. Guidelines for enteral feeding in adult hospital patients. Gut 2003;52(supp 7):vii1–vii12..[Free Full Text]
  5. Bliss, DZ, Johnson, S, Savik, K, Clabots, CR & Gerding, DN. Fecal incontinence in hospitalized patients who are acutely ill. Nurs Res 2000;49:101–8..[CrossRef][Medline]
  6. Majid, HA, Emery, PW & Whelan, K. Attitudes of patients and nurses towards diarrhoea during enteral tube feeding. J Hum Nutr Diet 2008;21:395 (abstr)..
  7. Bowling, TE, Raimundo, AH, Grimble, GK & Silk, DB. Colonic secretory effect in response to enteral feeding in humans. Gut 1994;35:1734–41..[Abstract/Free Full Text]
  8. Bleichner, G, Blehaut, H, Mentec, H & Moyse, D. Saccharomyces boulardii prevents diarrhoea in critically ill tube fed patients. Intensive Care Med 1997;23:517–23..[CrossRef][Medline]
  9. Bliss, DZ, Johnson, S, Savik, K, Clabots, CR, Willard, K & Gerding, DN. Acquisition of Clostridium difficile and Clostridium difficile–associated diarrhoea in hospitalised patients receiving tube feeding. Ann Intern Med 1998;129:1012–9..[Abstract/Free Full Text]
  10. Vollaard, EJ & Clasener, HA. Colonization resistance. Antimicrob Agents Chemother 1994;38:409–14..[Free Full Text]
  11. Bowling, TE, Raimundo, AH, Grimble, GK & Silk, DB. Reversal by short-chain fatty acids of colonic fluid secretion induced by enteral feeding. Lancet 1993;342:1266–8..[CrossRef][Medline]
  12. Sullivan, A, Edlund, C & Nord, CE. Effect of antimicrobial agents on the ecological balance of human microflora. Lancet Infect Dis 2001;1:101–14..[CrossRef][Medline]
  13. Clausen, MR, Bonnen, H, Tvede, M & Mortensen, PB. Colonic fermentation to short chain fatty acids is decreased in antibiotic-associated diarrhoea. Gastroenterology 1991;101:1497–504..[Medline]
  14. Furrie, E. A molecular revolution in the study of intestinal microflora. Gut 2006;55:141–3..[Abstract/Free Full Text]
  15. Seksik, P, Rigottier-Gois, L, Gramet, G, et al.. Alterations of the dominant faecal bacterial groups in patients with Crohn's disease of the colon. Gut 2003;52:237–42..[Abstract/Free Full Text]
  16. Hopkins, MJ, Sharp, R & Macfarlane, GT. Age and disease related changes in intestinal bacterial populations assessed by cell culture, 16S rRNA abundance, and community cellular fatty acid profiles. Gut 2001;48:198–205..[Abstract/Free Full Text]
  17. Whelan, K, Judd, PA, Preedy, VR & Taylor, MA. Enteral feeding: the effect on faecal output, the faecal microflora and SCFA concentrations. Proc Nutr Soc 2004;63:105–13..[CrossRef][Medline]
  18. Whelan, K, Judd, PA, Preedy, VR, Simmering, R, Jann, A & Taylor, MA. Fructo-oligosaccharides and fiber partially prevent the alterations in fecal microbiota and short-chain fatty acid concentrations caused by standard enteral formula in healthy humans. J Nutr 2005;135:1896–902..[Abstract/Free Full Text]
  19. Majid, HA, Emery, PW & Whelan, K. Low concentrations of faecal bifidobacteria in patients receiving enteral tube feeding. J Hum Nutr Diet 2007;20:364, (abstr)..
  20. Schneider, SM, Girard-Pipau, F, Anty, R, et al.. Effects of total enteral nutrition supplemented with a multi-fibre mix on faecal short-chain fatty acids and microbiota. Clin Nutr 2006;25:82–90..[CrossRef][Medline]
  21. Department of Health. Dietary reference values for food energy and nutrients for the United Kingdom. London, United Kingdom: HMSO. 1991..
  22. Elia, M. Artificial nutritional support. Med Int 1990;82:3392–6..
  23. Whelan, K, Judd, PA & Taylor, MA. Covert assessment of concurrent and construct validity of a chart to characterize fecal output and diarrhea in patients receiving enteral nutrition. JPEN Parenter Enteral Nutr 2008;32:160–8..[Abstract/Free Full Text]
  24. Heaton, KW, Radvan, J, Cripps, H, Mountford, RA, Braddon, FEM & Hughes, AO. Defecation frequency and timing, and stool form in the general population: a prospective study. Gut 1992;33:818–24..[Abstract/Free Full Text]
  25. Harmsen, HJ, Gibson, GR, Elfferich, P, et al.. Comparison of viable cell counts and fluorescence in situ hybridization using specific rRNA-based probes for the quantification of human fecal bacteria. FEMS Microbiol Lett 2000;183:125–9..[CrossRef][Medline]
  26. Kapuscinski, J. DAPI: a DNA-specific fluorescent probe. Biotech Histochem 1995;70:220–33..[Medline]
  27. Langendijk, PS, Schut, F, Jansen, GJ, et al.. Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genus-specific 16S rRNA-targeted probes and its application in fecal samples. Appl Environ Microbiol 1995;61:3069–75..[Abstract]
  28. Manz, W, Amann, R, Ludwig, W, Vancanneyt, M & Schleifer, KH. Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiology 1996;142:1097–106..[Abstract/Free Full Text]
  29. Franks, AH, Harmsen, HJ, Raangs, GC, Jansen, GJ, Schut, F & Welling, GW. Variations of bacterial populations in human feces measured by fluorescent in situ hybridization with group-specific 16S rRNA-targeted oligonucleotide probes. Appl Environ Microbiol 1998;64:3336–45..[Abstract/Free Full Text]
  30. Harmsen, HJ, Wildeboer-Veloo, AC, Grijpstra, J, Knol, J, Degener, JE & Welling, GW. Development of 16S rRNA-based probes for the Coriobacterium group and the Atopobium cluster and their application for enumeration of Coriobacteriaceae in human feces from volunteers of different age groups. Appl Environ Microbiol 2000;66:4523–7..[Abstract/Free Full Text]
  31. Sobotka, L, Bratova, M, Slemrova, M, Manak, J, Vizda, J & Zadak, Z. Inulin as the soluble fibre in liquid enteral nutrition. Nutrition 1997;13:21–5..[Medline]
  32. Nakao, M, Ogura, Y, Satake, S, et al.. Usefulness of soluble dietary fiber for the treatment of diarrhea during enteral nutrition in elderly patients. Nutrition 2002;18:35–9..[CrossRef][Medline]
  33. De La Cochetière, MF, Durand, T, Lepage, P, Bourreille, A, Galmiche, JP & Doré, J. Resilience of the dominant human fecal microbiota upon short-course antibiotic challenge. J Clin Microbiol 2005;43:5588–92..[Abstract/Free Full Text]
  34. Hoverstad, T, Carlstedt-Duke, B, Lingaas, E, et al.. Influence of oral intake of seven different antibiotics on faecal short-chain fatty acid excretion in healthy subjects. Scand J Gastroenterol 1986;21:997–1003..[Medline]
  35. Woodmansey, EJ. Intestinal bacteria and ageing. J Appl Microbiol 2007;102:1178–86..[Medline]
  36. Lievin, V, Peiffer, I, Hudault, S, et al.. Bifidobacterium strains from resident infant human gastrointestinal microflora exert antimicrobial activity. Gut 2000;47:646–52..[Abstract/Free Full Text]
  37. Hickson, M, D'Souza, AL, Muthu, N, et al.. Use of probiotic Lactobacillus preparation to prevent diarrhoea associated with antibiotics: randomised double blind placebo controlled trial. BMJ 2007;335:80 (abstr)..[Abstract/Free Full Text]
  38. Heimburger, DC, Sockwell, DG & Geels, WJ. Diarrhoea with enteral feeding: prospective reappraisal of putative causes. Nutrition 1994;10:392–6..[Medline]
  39. Whelan, K. Enteral-tube-feeding diarrhoea: manipulating the colonic microbiota with probiotics and prebiotics. Proc Nutr Soc 2007;66:299–306..[CrossRef][Medline]
  40. Wong, JM, de Souza, R, Kendall, CW, Emam, A & Jenkins, DJ. Colonic health: fermentation and short chain fatty acids. J Clin Gastroenterol 2006;40:235–43..[CrossRef][Medline]
  41. Vogt, JA & Wolever, TM. Fecal acetate is inversely related to acetate absorption from the human rectum and distal colon. J Nutr 2003;133:3145–8..[Abstract/Free Full Text]
  42. Lewis, SJ & Heaton, KW. Increasing butyrate concentration in the distal colon by accelerating intestinal transit. Gut 1997;41:245–51..[Abstract/Free Full Text]
  43. Tuohy, KM, Hinton, DJ, Davies, SJ, Crabbe, MJ, Gibson, GR & Ames, JM. Metabolism of Maillard reaction products by the human gut microbiota: implications for health. Mol Nutr Food Res 2006;50:847–57..[CrossRef][Medline]
  44. Sellin, JH & DeSoignie, R. Short-chain fatty acid absorption in rabbit colon in vitro. Gastroenterology 1990;99:676–83..[Medline]
  45. Suzuki, T, Yoshida, S & Hara, H. Physiological concentrations of short-chain fatty acids immediately suppress colonic epithelial permeability. Br J Nutr 2008;100:297–305..[Medline]
Received for publication March 28, 2008. Accepted for publication September 5, 2008.





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
89/1/240    most recent
ajcn.2008.26219v1
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Whelan, K.
Right arrow Articles by Taylor, M. A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Whelan, K.
Right arrow Articles by Taylor, M. A
Agricola
Right arrow Articles by Whelan, K.
Right arrow Articles by Taylor, M. A

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS