In this systematic review, 15 randomized controlled trials were reviewed to determine the potential of pre, pro, or synbiotics administration to improve enteral feeding tolerance in tube-fed critically ill patients. Gut microbiota is a key regulator of gut function, host metabolism, and appetite. Microbial metabolites, including SCFAs, bile acids, and various neuroactive agents, interact with the GI tract and peripheral tissue through affecting the enteric nervous system and central appetite pathways or altering bile acid signaling (34). These effects result in changes in gastric motility and emptying (35, 36), which may reduce enteral feeding intolerance. Besides, gut microbiota can influence intestinal barrier function and modulate the immune system, thus indirectly affect metabolism and eating behavior (16).
5.1 Effect on energy intake or feed volume
We found six studies that evaluated the effect of pre, pro or synbiotics on enteral feeding volume or energy intake. Considering the application of probiotics or synbiotics, no significant effect was reported. Only 2 of 4 studies, that used prebiotics (one soluble guar gum for 4 days and the other FOS for 21 days) in the intervention group, found significant beneficial effects (21, 32). It should be noted that in both of these studies, patients in the intervention group received significantly more volume and energy on the first day. Therefore, it seems that the significant difference between the two groups in terms of received feed volume and energy at the end of the study may not be merely attributed to the effect of prebiotics.
5.2 Effect on target calorie achievement
Four trials evaluated the effect of pre or probiotics on frequency or time to achieve the target calorie. All studies but one found no significant effect. In this study, probiotic administration for seven consecutive days was associated with a significantly faster return of the gut function (29). The included studies were heterogeneous in population features, intervention, duration, eligibility criteria, and EN protocol. Thus, the conflicting results may be attributed to these factors. It is also believed that the beneficial effect of probiotics or synbiotics could be highly strain-specific.
5.3 Effect on diarrhea
In the critical care setting, diarrhea is the most common gastrointestinal complication of EN (37), which may result in several unfavorable clinical conditions including enteral nutrition cessation and exacerbation of undernutrition (32). Factors that contribute to the pathogenesis of diarrhea include altered physiological responses due to EN, antibiotics administration, and altered gut microbiota function (38). Therefore, gut microbiota manipulation may be an approach for the prevention and management of diarrhea in the critical care setting. For example, gut microbiota manipulation can reverse abnormal colonic water secretion by SCFAs production (39), alter colonic motor activity (40), and interfere with pathogen colonization in the gut, which protects against diarrhea (38).
The effect of prebiotics on diarrhea was evaluated in five clinical trials (19-21, 28, 32). Four studies investigated the effect of prebiotic on the prevalence of diarrhea. While two studies found a significant (20, 32) and one a non-significant decrease (19), the other reported a non-significant increase (28). The number of days of diarrhea was also investigated in two studies, one of which reported a significant decrease (20), while the other found a non-significant increase(28). The number of liquid stools was also reported to be lower in the prebiotic group in one trial (21).
It should be noticed that water-soluble fiber like pectin or guar gum exhibits antidiarrheal effect by two mechanisms:1) production of SCFAs and maintaining gut microbiota homeostasis or 2) reuptake of water and electrolytes (41). The beneficial effect of water-soluble fibers on SCFAs production is well documented in non- critically ill patients and healthy subjects, but it is not clearly observed in critically ill patients(41). So, the positive effect of water-soluble fibers in the mentioned studies may be attributed to the increased reuptake of water and electrolytes, not necessarily acting as prebiotics.
Regarding the effect of probiotics on the incidence of diarrhea, two studies reported a trend towards reduced diarrhea incidence in the probiotic group (18, 25), and one reported a non-significant increase (24). The effect of probiotic administration on diarrhea days was demonstrated in three of the included trials. Two of them reported a significant decrease in diarrhea duration (18, 25), while one reported a non-significant increase (26). In the probiotic group, the number of liquid stools per patient per day was reported to be significantly lower in one study (23) but, loose stools were non-significantly more in another study (26).
A non-significant decrease in the prevalence of diarrhea (22) and a significant decrease in the incidence of enteritis (31) was reported to be associated with synbiotic administration.
5.4 Effect on length of stay
All but one study found no beneficial effects for gut microbiota manipulation on clinical endpoints, including LOS in hospital and ICU. A recent systematic review and meta-analysis by Manzanares et al. also showed that despite the beneficial effects of probiotic and synbiotic administration on overall infections and ventilator-associated pneumonia, these agents had no significant effect on LOS in hospital or ICU (42).
To the best of our knowledge, this systematic review was the first study to review the effect of pre, pro, and synbiotics on feeding tolerance in enterally- fed critically ill patients. As we assessed relevant outcomes in a heterogeneous ICU population, our results could be attributed to a broad spectrum of critically ill patients with sepsis, trauma, or other medical conditions. Although, the inclusion of diverse patient groups in this systematic review may be considered as a limitation for interpretation of the results. There was also great diversity in the type of administered prebiotic or probiotic strains, duration of treatment, and dose. This heterogeneity also made it impossible to quantitatively evaluate the results. Furthermore, most of the included studies reported the energy intake or feeding tolerance as a secondary outcome, not mentioning the EN protocols, while the reported EN protocols were heterogeneous in other studies.