Effect of dietary fibre on the gastrointestinal microbiota during critical illness: A scoping review

Abstract

The systemic effects of gastrointestinal (GI) microbiota in health and during chronic diseases is increasingly recognised. Dietary strategies to modulate the GI microbiota during chronic diseases have demonstrated promise. While changes in dietary intake can rapidly change the GI microbiota, the impact of dietary changes during acute critical illness on the microbiota remain uncertain. Dietary fibre is metabolised by carbohydrate-active enzymes and, in health, can alter GI microbiota. The aim of this scoping review was to describe the effects of dietary fibre supplementation in health and disease states, specifically during critical illness. Randomised controlled trials and prospective cohort studies that include adults (> 18 years age) and reported changes to GI microbiota as one of the study outcomes using non-culture methods, were identified. Studies show dietary fibres have an impact on faecal microbiota in health and disease. The fibre, inulin, has a marked and specific effect on increasing the abundance of faecal Bifidobacteria. Short chain fatty acids produced by Bifidobacteria have been shown to be beneficial in other patient populations. Very few trials have evaluated the effect of dietary fibre on the GI microbiota during critical illness. More research is necessary to establish optimal fibre type, doses, duration of intervention in critical illness.

Key Words: Gastrointestinal microbiota; Dietary fibre; Health; Critical illness; Short chain fatty acids

Core Tip: This review explores the influence of dietary fibre on the gastrointestinal microbiota, emphasizing its role in both health and critical illness. Dietary fibres promote the production of beneficial short-chain fatty acids that enhance health through multiple immunological and anti-inflammatory effects. While fibres like inulin and arabinoxylan oligosaccharides positively affect the microbiota in healthy individuals, their impact on critically ill patients remains uncertain. Limited studies suggest potential benefits but highlight the need for further research in critically ill populations.

INTRODUCTION

The term microbiome refers to “a collection of microbial genomes in any habitat such as an organ, and include bacteria, viruses, fungi and protozoa” and includes both the microbiota [the term used to define all microbes in a certain environment, such as the gastrointestinal (GI) microbiome] and their activity (structural elements, metabolites/signal molecules, and the surrounding environmental conditions)[1].

The importance of the GI microbiota to maintaining health, and changes in the GI microbiota observed during chronic diseases, is increasingly recognised[2-6]. Because of this, dietary strategies to modulate the GI microbiota to maintain health or to treat chronic diseases are appealing[7]. It is established that a dietary intervention can rapidly change the GI microbiota[8], and fibre is one such dietary intervention[9]. Dietary fibre provides a substrate for the GI microbiota to produce short chain fatty acids (SCFAs). This may be beneficial as SCFAs are key compounds in promoting intestinal integrity, occurring through the protective effect of SCFA on barrier function reducing bacterial translocation. SCFAs also improve mucous production and gut motility and have local and systemic anti-inflammatory actions[10,11].

The definition of what constitutes a dietary fibre remains controversial[12]. The most widely accepted definition is that of the International Codex Alimentarius commission, which defines dietary fibre as edible carbohydrate polymers, naturally occurring in food, or extracted from raw food or synthetically produced, with three or more monomeric units “which are not hydrolysed by the endogenous enzymes in the small intestine of humans” and not absorbed. Prebiotics are defined by the International Association for probiotics and Prebiotics (ISAPP)[13] as substrates that are “selectively utilized by host microorganisms conferring a health benefit”. Therefore, certain dietary fibres, or what has been termed “microbiota accessible carbohydrates”[14], that confer a health benefit are categorised as prebiotics.

The taxonomy of dietary fibres is complex. One classification is to separate into (1) naturally occurring edible carbohydrate polymers (e.g., legumes, cereals, fruits, and vegetables); (2) raw food derived edible carbohydrate polymers; and (3) synthetic carbohydrate polymers. Based on solubility, dietary fibres can also be classed as insoluble, which assist laxation, or water soluble, which are completely fermented in the large intestine producing SCFAs[15].

Because of their resistance to digestion, dietary fibres reach the colon where they are metabolised by carbohydrate active enzymes and affect the GI microbial ecology. GI microbes have the ability to ferment undigested dietary fibre to produce active metabolites[16], forming the link between dietary fibre and host health.

The aim of this scoping review was to describe in general the effects of dietary fibre supplementation in health and chronic disease states and specifically during critical illness. Randomised controlled trials and prospective cohort studies that include adults (> 18 years age) and reported changes to GI microbiota as one of the study outcomes, as assessed by non-culture methods, were identified.

IDENTIFYING RELEVANT STUDIES

A literature search using the key words “microbiota”, “dietary fibre”, “critical illness”, “disease” using the databases Ovid MEDLINE, EMBASE, CENTRAL was conducted. Data extracted included study design, interventions and outcome relating to GI microbiota (using next-generation sequencing) and faecal SCFA concentration.

DESCRIPTORS OF DIVERSITY

A brief summary of descriptors used to assess microbiota is required to understand existing evidence. Microbiomes are often described in terms of alpha diversity and beta diversity[17].

Alpha diversity relates to the measure of richness and distribution of a microbial community and is applicable to a single sample (within-sample diversity). It is akin to the summary statistic of a single population. Alpha diversity is quantified by various measures such as the ‘Shannon index’ or the ‘Simpson index’ which are equations used in ecological studies[18].

Beta diversity is a measure of the similarity or dissimilarity of two microbial communities (between sample diversity). Beta diversity can be measured by taxa overlap or quantified by ‘Bray-Curtis dissimilarity’ which is a statistic of dissimilarity of two ecological communities, with a value of zero representing no dissimilarity and one indicating no similarity. These measures do not give information on changes to abundance of specific taxa but can allow assessment of differences in composition of microbes. UniFrac is another relatively new method specifically designed to measure of dissimilarity between communities using phylogenetic data[19,20].

Operational taxonomic units (OTU) is the term used to describe groups of closely related individuals in an ecological system and replaces the term “species” in many microbiome diversity analyses. OTU refers a cluster of sequences based on a similarity threshold (e.g., 97% is commonly used in 16S studies), as a proxy for assignment to a taxonomic group. The field of metagenomic analysis has been expanding rapidly[21]. Most modern studies use Amplicon Sequence Variants[22] or sub OTUs sequencing data where technical errors are corrected (denoising) to obtain the “true” biological sequence. This is then compared at the sequence level to differentiate between sequences where a single nucleotide is different, instead of clustering at a threshold of 97% identity. These clusters and the respective number of reads within are an estimation of the abundance[17,23].

BIFIDOGENESIS AS A TRIAL OUTCOME

Bifidobacteria belong to the Actinobacteria, one of the largest bacterial phyla (Figure 1).

Figure 1  A schematic of the phylogenetics species Bifidobacteria that belongs to the phylum Actinobacteria.

Bifidobacteria are Y shaped cells that dominate the GI tract of breast-fed infants. They are an important taxon in early life and removal is associated with collapse of the ecological community, hence they are regarded as a ‘keystone’ taxon. The bifidobacterial population stabilizes in adulthood to make up 3%–6% of the total gut microbial population.

Bifidobacteria bestow health benefits to the host through the production of active metabolites[24] and by maintenance of a low intraluminal pH, which is non-conducive to growth of pathogenic organisms. Levels lower than healthy controls have been reported in obese[25] and diabetic adults[26] and childhood allergic disorders and infections[27]. Hence, effect on bifidogenesis diversity and/or numbers as an outcome for smaller trials is justified, as it is a surrogate marker of beneficial effect to the host.

EFFECT OF DIETARY FIBRE ON GI MICROBIOTA IN HEALTH

Various dietary fibres have been evaluated in healthy humans. These include inulin, arabinogalactans, arabinoxylan oligosaccharides, polydextrose, human milk oligosaccharides, galacto-oligosaccharides (GOSs), oligofructose, resistant dextrin, resistant starch, xylo-oligosaccharides, partially hydrolysed guar gum and various mixtures of fibres. One of the most frequently studied dietary fibres is inulin, which is a water-soluble indigestible fibre polymer of fructose that has no effect on blood glucose. The prebiotic effect of inulin is discussed in further detail later.

Human milk oligosaccharides have a very specific effect of increasing Bifidobacteria. In a parallel, double-blind, randomised, placebo-controlled trial involving one hundred adult participants divided into ten groups and randomised to the following: 2′-O-fucosyllactose (2′FL), lacto-N- neotetraose (LNnT) or 2:1 “mass ratio” combination of 2’FL:LNnT, each intervention group receiving three daily doses of either 5, 10 or 20 g of 2’FL or LNnT or 2:1 combination and the placebo group receiving 2 g for > 12 days[28]. Using 16S rRNA amplicon sequencing, the authors concluded that human milk oligosaccharide supplementation modified gut microbiota by significantly increasing the relative abundance of Actinobacteria compared to placebo, particularly Bifidobacterium, and reducing the relative abundance of Firmicutes and Proteobacteria and this was dose dependent (P < 0.05, R² = 28%).

Arabinoxylan oligosaccharide is another dietary fibre that has been studied extensively; delivered as enriched bread[29,30] or simply as powder[31-33] and, similar to inulin, the common theme is bifidogenesis (an increase in population of Bifidobacteria and a relative reduction in other species) but additionally an increase in faecal SCFA concentrations (results detailed in Table 1)[29-33].

Table 1 Randomised placebo controlled double blinded trials of arabinoxylan oligosaccharides.

Ref.	Intervention	Method of analysis	Microbiota	Faecal SCFA concentration
Damen et al[29], 2012	Wheat/Rye bread with or without AXO (n = 27)	FISH	Increased Bifidobacterium spp.a	Increased
Walton et al[33], 2012	Bread with or without AXO (n = 27)	FISH	No significant difference	Increased
François et al[30], 2012	Wheat bran derived AXO or placebo drink (n = 57)	FISH	Increased Bifidobacterium with 10 g/day dose	Increased total SCFA with 10 g/day dose
Müller et al[32], 2020	AXO powder or Maltodextrin (n = 48)	16S rRNA amplicon sequencing	Increased bifidobacterium, decreased alpha diversity	No changes
Chung et al[31], 2020	AXO powder or Maltodextrin (n = 21)	16S rRNA sequencing	Significant increase in Bifidobacterium speciesb; Decrease in Firmicutesc	Significantly increased

^aP = 0.06.

^bP ≤ 0.01.

^cP < 0.001.

AXO: Arabinoxylan oligosaccarides; FISH: Fluorescent in situ hybridisation; SCFA: Short chain fatty acid.

Trials on arabinogalactans, polydextrose, GOSs, oligofructose, resistant dextrin, resistant starch, xylo-oligosaccharides, partially hydrolysed guar gum show similar results on bifidogenesis but minimal effect on faecal SCFA concentrations (Supplementary Table 1 and Supplementary Table 2).

Conversely, some trials[34-37] did not report any effect of dietary fibres on the GI microbiota or faecal SCFA concentrations (Table 2). Inadequate sample size or shorter duration of exposure have been suggested by these authors to explain the lack of effect. Brandl et al[36] in two parallel randomised trials evaluated 10 g/day of extrinsic wheat fibre enriched baked/cooked solid foods (n = 10) or in flavoured drinks (n = 19) compared to the same products without enrichment for five days did not find any alteration in alpha (P value = 1.0) or beta-diversity (P value = 0.996) of the GI microbiota compared to placebo. It was proposed the short exposure time was the reason for lack of modulation of the microbiota, although other studies have shown to have an effect with similar duration of dietary interventions[38]. Pectin supplementation (7.5 g twice daily), in the form of powder extracts of sugar beet, for four weeks did not change the GI microbiota compared to maltodextrin in the same dose[35]. The authors suggested the dose might have been inadequate, compared to similar studies of pectin. Vuholm et al[37] confirmed the lack of effect of whole grain wheat (WGW) and whole grain rye (WGR) cereals on the diversity of GI microbiota compared to refined wheat: Between week 0 and week 6 of intervention the Shannon indices were 8.3 ± 1.2 vs 8.2 ± 1.2 for WGW, 8.5 ± 1.1 vs 8.5 ± 1.0 for WGR and 7.9 ± 1.3 vs 8.1 ± 0.8 for refined wheat (P value = 0.50) and the Chao1 indices were 2652 ± 503 vs 2604 ± 565 for WGW, 2658 ± 487 vs 2711 ± 429 for WGR and 2499 ± 640 vs 2624 ± 397 for refined wheat (P value = 0.39) in line with a similar 6 week trial comparing WGW supplemented diet (> 80 g/day of WGW) to no supplementation (< 16 g/day WGW)[34] but cautioned that the inadequate sample size could have increased the risk of a beta error.

Table 2 Randomised controlled trials that failed to change the gastrointestinal microbiota.

Ref.	Intervention/placebo	Method of analysis
Ampatzoglou et al[34], 2015	high (> 80 g/day) or low (< 16 g/day) whole grain (n = 33)	FISH analysis
An et al[35], 2019	15 g/day sugar beet pectin or maltodextrin	16S rRNA amplicon sequencing
Brandl et al[36], 2020	Commercial fibre product enriched with 10 g wheat or without enrichment (n = 48)	16S rRNA amplicon
Vuholm et al[37], 2017	Whole grain or refined wheat group (n = 70)	16S rRNA amplicon sequencing

FISH: Fluorescent in situ hybridization.

PREBIOTIC POTENTIAL OF INULIN IN HEALTHY ADULTS

The ISAPP accepts inulin-type fructans, galacto oligosaccharides and lactulose as prebiotics[13]. In 2018, the Federal Drug administration reviewed the scientific evidence of the physiological effects of non-digestible carbohydrates, including inulin and confirmed their prebiotic effect[39].

Inulin is a naturally occurring polysaccharide that belongs to a class of carbohydrates known as fructans, which are polymers of fructose[40]. Inulin is one of the six types of fructans. Inulin is not simply one molecule; it is heterogenous collection of fructose polymers (polydisperse compound) linked by beta 2,1 linkages. The degree of polymerisation refers to the number of fructosyl fructose beta (2,1) linkages, with inulin having anywhere between 3 to 60 degrees of polymerisation. The beta configuration of the linkages render inulin indigestible by humans as the human hydrolysing enzymes specifically break down alpha bonds.

Naturally occurring inulin and oligofructose are found in artichokes, asparagus, bananas, chicory root, garlic, onions, leeks, and wheat[41]. Inulin is commercially extracted from agave, Jerusalem artichokes or chicory roots. It is also synthesised from sucrose. The process of extraction produces a powder extract. Commercial “high performance” inulin is a highly refined form of inulin, consisting of an average of 25 monosaccharide units. This form of inulin has many advantages including reduced solubility, increasing its creamy texture and therefore being utilised as fat replacements in spreads, dairy products, and frozen desserts. Shorter chain inulin has increased solubility and properties similar to sugar and 30% sweetness compared to sugar and used in a variety of food products such as cereals, frozen yogurts, and cookies.

The chain length of inulin affects fermentation characteristics. An in vitro study of inulin fermentation by human faecal inoculum reported that inulin with the greatest degrees of polymerisation (> 20) had the least effect on SCFA production by microbes[42]. The mean degrees of polymerisation of inulin used in the previously mentioned review was 15 (SD 9.4)[43]. Inulin was reported to have a more prominent prebiotic effect than oligofructose when fermentation activity and bacterial community composition were compared in a laboratory model (Simulator of the Human Intestinal Microbial Ecosystem)[44].

There have been at least 25 trials exploring the prebiotic effect of inulin on faecal microbiota, quantified using next-generation sequencing and reviewed in detail elsewhere[43] and a summary is provided here. The results of these trials demonstrated that ingested inulin markedly stimulated bifidogenesis. Ten trials measured faecal SCFA concentration, with only one reporting an increase in SCFA concentration. Inulin used in these trials varied from natural sources (ITF rich vegetables such as artichoke, garlic, salsify, shallot, leek, scorzonera, onion and celery root) to refined proprietary inulin in the form of chocolate chews and snack bars[45], powdered form[46] in milk or extracted from globe artichoke[47]. The median (IQR) dose used in these trials was 14.5 g/day (10 g, 16 g) and duration of intervention 3.5 weeks (2 weeks, 6 weeks). It was reported that a dose of 34 g/day increased flatulence and bloating but not diarrhoea[48].

The effective dose of inulin for a consistent bifidogenic effect is 3-20 g/day based on a meta-analysis of fifty trials (n = 2525)[49]. These results were consistent across various subgroups including healthy adults, patients with diabetes and non-alcoholic fatty liver but not in those with diarrhoeal illness, Crohn’s, or coeliac disease.

A review of human trials evaluated the response to inulin over time[50]. The authors reported that the number of Bifidobacteria (expressed as colony forming units/g of faeces) reaches a maximum by a week and wanes within a week of stopping inulin. Such a time course is consistent with the hypothesis that inulin acts as a selective substrate for the fermentation by Bifidobacteria that are stimulated to grow.

It is likely the bifidogenic effect of inulin is dependent on the pre-existing GI flora. A randomised controlled trial suggested that the bifidogenic effect was dependent on the Bacteroides/Bifidobacterium ratio at baseline, those low ratios benefiting the most[51].

EFFECT OF DIETARY FIBRE ON GI MICROBIOTA DURING CHRONIC ILLNESS

Dietary fibres have been trialled as interventions in chronic illness such as type I and II diabetes[52-54], hypertension[55], asthma[56], metabolic syndrome[57], obesity, hypercholesterolemia, and non-alcoholic fatty liver disease.

A recent meta-analysis of nine randomised controlled trials on participants with type II diabetes summarised that dietary fibre significantly increased the relative abundance of Bifidobacterium, decreased plasma lipopolysaccharide and total cholesterol levels, and reduced the body mass index of participants[58].

There is evidence that gut microbiota and their metabolites, such as trimethylamine N-oxide and SCFAs, play an important preventative role in the onset, progression, and exacerbation of heart disease including coronary artery disease, hypertension, heart failure, atrial fibrillation, and myocardial fibrosis[59,60]. It has been proposed that dietary fibre has the capacity to modulate these responses[61].

The effect of dietary fibre on GI microbiota and potential downstream impact is illustrated by studies involving patients with end stage renal failure. It is possible that the progression of chronic kidney disease may be altered by modulation of the GI microbiota. A trial of beta-glucan as a prebiotic for fourteen weeks significantly altered the GI microbiota and reduced uraemia without altering biochemical markers of renal function[62]. A proprietary mixture of dietary fibre mainly constituted of galactomannan (10 g/day) evaluated in a parallel group, placebo-controlled randomised clinical trial of 384 patients who were haemodialysis dependent[63] reported significantly greater abundance of Bifidobacterium adolescentis, Lactobacillus and Lactobacillaceae in the intervention group. However, a randomised blinded crossover pilot trial on twelve haemodialysis dependent participants did not report any effect of 10-15 g/day of dietary supplementation with inulin compared with maltodextrin (6-9 g/day), an inert comparator, on the GI microbiota or faecal SCFA concentrations[64]. It is likely the small numbers in each arm could explain the lack of a difference.

In summary, dietary fibre can be utilised to manipulate the GI microbiota in various chronic disease states, with potential for beneficial ‘down-stream’ effects on patient centred outcomes. Further studies are required to establish optimum sample size, type of fibre, dose, and duration of exposure. Mechanistic trials are required to assess reliable trial outcome measures.

EFFECT OF DIETARY FIBRE ON GI MICROBIOTA DURING CRITICAL ILLNESS

Very few studies in critical illness have explored the effect of dietary fibre on GI microbiota during critical illness. A randomised, parallel-group, single-centre, open-label, pilot trial of twenty two adult patients admitted to the intensive care unit (ICU) reported the effect of dietary fibre on the GI microbiome in critically ill patients[65]. Patients were eligible if presenting with sepsis and it was expected that they would receive three or more days of IV antibiotic but patients who had recent surgery involving the intestinal lumen were excluded. Patients were randomised to either a proprietary mix of soy and oat derived feed containing 14.3 g fibre/L, or a calorie and micronutrient equivalent formula with no fibre content. The intervention and comparator were given via nasogastric tube for as long as possible at the discretion of the treating physician. Rectal swabs were collected at admission, day 3, 7, 14, and 30. Bacteria were quantified using 16S rRNA sequencing. The primary outcome, which was within-individual change from baseline (day 0 to day 3) in the abundance of SCFA-producing bacteria, was not significantly different between groups. However, the mean change in relative abundance of SCFA producers was greater in the fibre-fed group than the non-fibre-fed group. OTUs corresponding to Faecalibacterium and Odoribacter became more abundant in the intervention group over the course of the intervention. There was no statistical difference in diversity, with a decline observed in both groups. It should be noted that overall dose of fibre was relatively modest with the fibre group receiving a median of 10.7 g/day (IQR, 5.9-18.2). A subsequent parallel group blinded dose-finding study by the same authors using supplementation with inulin (32 and 16 g/day) compared to placebo and continued for seven days is ongoing (NCT03865706).

A multicentre retrospective study[66] was nested within a longitudinal, prospective adult ICU-based cohort[67]. In this retrospective analysis of changes to GI microbiota the authors stratified fibre intake in the first 72 hours after admission into tertiles (high-, low- and no-fibre). Rectal swabs were collected within four hours of admission and another at 72 (± 4) hours after and analysed using 16S rRNA gene sequencing. Overall, the median fibre intake was 13.4 g (IQR 0-35.4 g), with 39.3 g (IQR 34.7-50.2 g) in the high fibre group (n = 47) and 11.2 g (IQR 3.8-18.2 g) in the low fibre group (n = 46). Thirty-six patients received no fibre. The median abundance of SCFA producers decreased by 0.33% within 72 hours of admission and so did the measures of intra-individual and inter-individual diversity, however the high fibre group had significantly higher levels of SCFA producing taxa. Higher fibre intake was associated with greater faecal microbial diversity at 72 hours (P = 0.04) but not richness (P value = 0.50). High fibre intake was also associated with greater abundance of SCFA producing bacteria after adjusting for antibiotics, admission APACHE IV score, and exposure to mechanical ventilation or use of vasopressors. There was a 0.3% median increase in SCFA producers for every additional 10 g of fibre intake (P value < 0.01). The dietary fibres administered in this observational study varied considerably and the lack of randomisation was a major limitation.

In a multicentre, blinded, placebo controlled trial of adult patients admitted to ICU who were expected to receive enteral nutrition (containing 0.7 g/100 mL soluble fibre and 0.8 g/100 mL insoluble fibre)for at least three days were randomised to 7 g/d of proprietary supplemental oligofructose/inulin powder mixed in 50 mL water, or a matched inert comparator (maltodextrin), via nasogastric tube for a maximum of 14 days[68]. Severity of illness or frequency of mechanical ventilation were not reported. 22 subjects completed at least 7 days of the trial (10 in intervention and 12 in the placebo arm). Fresh faecal samples were collected immediately after recruitment and between 7 to 14 days after intervention/placebo commenced. Faecal microbiota was analysed using fluorescent in situ hybridisation (FISH) methodology within an hour of collection. At baseline, there was no difference in GI microbiota, measured as total cells and expressed as log10 cells/g dry faeces, between the intervention (mean 10.3, SD 0.3) and placebo (mean 10.3, SD 0.3) group (P value > 0.05, t test). After adjusting for baseline values there were significantly lower mean concentrations of Faecalibacterium prausnitzii in the intervention group vs the placebo group (7.0, SD 1.0 vs 8.4, SD 1.3, P value < 0.01) and Bacteroides-Prevotella (9.1, SD 1.0 vs 9.9, SD 0.9, P value < 0.05) but no significant differences in the concentrations of Clostridium coccoides-Eubacterium rectale, Bifidobacteria and Lactobacillus enterococci. These results are in contrast to results from inulin trials in healthy participants, which reported a marked increase in faecal SCFA concentrations and Bifidobacteria[45-47,69]. Factors that could possibly explain the difference between this trial in the critically ill and the effects observed in health, are the trial was underpowered (a sample size of 40 was required based on trials in a healthy population[70]) and the duration of exposure was limited (mean of 12 days) and subjects received enteral nutrition for an average of 7.6 days before the first day of intervention owing to the need to collect fresh faecal samples.

O’Keefe et al[71] conducted a case-control study of four adult critically ill patients with impaired gastric emptying (not otherwise defined) referred to a nutrition service for jejunal feeding. Five age matched healthy volunteers served as controls. Three of the four patients had severe acute pancreatitis, APACHE II scores > 20, and were started on proprietary semi-elemental feeds containing 4 g/L of soluble fibre as oligofructose and inulin. The four patients were followed up for a mean duration of 27 days. Results were reported as proportions. Prior to supplementation the Bacteroidetes composition made up 35% of the microbiota in healthy subjects and 60% in the critically ill (measured as copies/mg dry faecal weight in millions by qPCR). After 2-5 weeks of supplementation the proportion of Firmicutes (considered to be a major SCFA producer and the most abundant phylum in the human GI tract) increased “six-fold” with an accompanied increase in faecal SCFA concentrations and restored the proportion of Bifidobacterium towards that of normal healthy controls (1.5 × 10⁵ to 5.9 × 10⁵, control 8.4 × 10⁵ copies/mg dry faecal weight in millions). This trial used a method of microgenomic analysis and reported actual abundance instead of sequencing methods that report relative abundance. The overall microbial load in four patients were several orders of magnitude lower than the healthy controls. This case series showed the potential for enteral Inulin supplementation to be bifidogenic in critical illness.

THE RELATION BETWEEN DIETARY FIBRE, FAECAL SCFA CONCENTRATION, AND GI MICROBIOTA COMPOSITION

Studies describing a causal relationship between GI microbiota, SCFA production and host metabolic regulation have been conducted in animals[72-76]. However, the extent to which a dietary fibre intervention changes the GI microbiota and affects SCFA production is not well described in human trials[77].

Not all fibres increase SCFA-producing bacteria. A randomised double-blind cross-over trial comparing wheat grain and wheat bran cereal showed that despite a significant increase in Bifidobacterium species in the stool samples of both groups there was no change in faecal SCFA concentration[78]. In this trial, wheat bran diet was not found to be bifidogenic. A trial of a diet supplemented with GOS for twelve weeks in participants who were obese and had pre-diabetes reported a significant increase in abundance of faecal Bifidobacterium species without any change in faecal or plasma SCFA (acetate) concentrations[79].

Several hypotheses have been generated to explain the observed lack of effect on SCFA concentrations despite significant bifidogenesis: Hepatic clearance of acetate, metabolism of acetate into butyrate, metabolism of butyrate locally by colonocytes and absorption of butyrate by the colon[80-82]. An in vitro study showed that slowly fermenting dietary fibre increased faecal SCFA levels by bypassing proximal colonic absorption[83].

A systematic review of randomised controlled trials reported a paucity of trials exploring the relation between dietary fibre, SCFA concentration and GI microbiota[77]. The few trials that could be analysed (n = 16) lacked methodological homogeneity in terms of types of fibre used as intervention, subjects chosen (healthy adults, overweight individuals and those with metabolic syndromes) and methodology of microbial analysis (16S rRNA analysis, qPCR, human intestinal tract chip, FISH and metagenomics).

FUTURE DIRECTIONS

Despite a substantial improvement in knowledge of the human microbiome in healthy adults, our understanding of the effect of critical illness on the GI microbiome remains sparse.

Currently there is convincing observational trial evidence for the beneficial health effects of non-digestible fermentable carbohydrates. However, well designed interventional trials are required to elucidate the clinical efficacy of specific prebiotics, the role of GI microbiota in bringing about these effects, their mechanism of action and the variability of individual differences in GI microbiota that influence effects. Randomised trials are required to test the hypothesis that modulating the GI microbiome has beneficial effects in critical illness.

The effect of SCFAs on cellular function is becoming clearer in animal studies; mechanistic trials on humans are required to elucidate the effect of SCFAs on health. The evolving field of metabolomics will potentially help identify signaling mechanisms between the microbiome and host.

CONCLUSION

Dietary fibres have an impact on faecal microbiota in health and chronic disease. Inulin has a marked specific effect on increasing the abundance of faecal Bifidobacteria. SCFA produced by Bifidobacteria appear to have beneficial ‘down-stream’ effects that could lead to improvement in patient-reported outcomes. Very few trials of the effect of dietary fibre have been conducted in the critically ill population and the effect of dietary fibre on GI microbiota is unknown.