Journal Logo

Feature Articles

Probiotics in Critical Illness: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Sharif, Sameer MD1,2; Greer, Alisha MD1,2; Skorupski, Clarissa MD3; Hao, Qiukui MD4,5; Johnstone, Jennie MD, PhD6,7; Dionne, Joanna C. MD, PhD2,8; Lau, Vincent MD, MSc9; Manzanares, William MD, PhD10; Eltorki, Mohamed MBChB11; Duan, Erick MD2; Lauzier, Francois MD, MSc12,13; Marshall, John C. MD14,15; Heels-Ansdell, Diane MSc8; Thabane, Lehana PhD8; Cook, Deborah J. MD, MSc2,8; Rochwerg, Bram MD, MSc2,8

Author Information
doi: 10.1097/CCM.0000000000005580

Abstract

Critical illness disrupts the host microbiome, creating what has been referred to as the “pathobiome” (1) or dysbiosis through a combination of mechanisms including pathogenic bacterial overgrowth, increased gastrointestinal permeability, the inadvertent iatrogenic effects of agents such as narcotics, antibiotics, acid-suppressing agents, as well as host defense inflammatory response (2,3). The World Health Organization (WHO) defined probiotics as “live microorganisms, which when administered in adequate amounts confer a health benefit of the host” (4). Probiotics have been shown to suppress gastrointestinal cytokine production, stimulate a protective mucus layer, prevent gut apoptosis, and reduce pathogenic bacterial overgrowth (5–7).

Prebiotics are a nondigestible food ingredient that stimulate growth of bacteria in the colon. Synbiotics combine both prebiotics and probiotics and may act synergistically (8). A randomized controlled trial (RCT) of 4,556 healthy newborn infants in rural India found that synbiotics decreased the combined relative risk (RR) of sepsis and death by 40% and the risk of lower respiratory tract infections by 34% when compared with placebo (9). Whether these synergistic physiologic mechanisms lead to the prevention of nosocomial infections (10) in patients admitted to the ICU remains uncertain.

Previous RCTs have demonstrated variable effects of probiotics on patient important outcomes in this setting (11,12). A recent meta-analysis of 30 RCTs (n = 2,972 patients) in critically ill patients showed that probiotics were associated with an overall reduction of infections (risk ratio [RR], 0.80; 95% CI, 0.61–0.90), including ventilator-associated pneumonia (VAP) (RR, 0.74; 95% CI, 0.61–0.90); however, this conclusion was based on pooling of relatively small studies as well as statistically heterogeneous (inconsistent) results (11). Probiotics may also decrease the risk of antibiotic-associated diarrhea or Clostridioides difficile-associated diarrhea. In the United States alone, probiotic sales have reached $35 billion/yr (13). Although no utilization reviews document their prescription in the ICU setting, probiotics are suggested for selected critically ill patients (American Society for Parenteral and Enteral Nutrition) (14).

The recently completed PROSPECT trial (NCT 02462590) enrolled 2,650 mechanically ventilated patients in 44 ICUs, comparing the probiotic Lactobacillus rhamnosus GG versus placebo on VAP and other clinically important outcomes (15). As this study almost doubles the trial data on this topic, we conducted an updated systematic review and meta-analysis to determine the effect of probiotics or synbiotics on morbidity and mortality in critically ill patients.

METHODS

We registered the protocol for this systematic review on PROSPERO (CRD42020157080) on April 28, 2020. We have highlighted any deviations from the published protocol with an accompanying explanation. The full methods can be found in the Supplementary Appendix 10 (https://links.lww.com/CCM/H132); an abbreviated version is presented here.

Systematic Search

We conducted a comprehensive search of MEDLINE, EMBASE, CENTRAL, and unpublished sources including ClinicalTrials.gov, WHO International Clinical Trials Registry Platform, Latin-American and Caribbean System on Health Sciences from inception to May 4, 2021.

Study Selection

We included RCTs testing probiotics or synbiotics administered for any length of time compared with placebo or no treatment in critically ill adults and children admitted to an ICU, whether or not they were receiving life support. We excluded studies of prebiotics alone, as well as studies examining neonates (if the majority of study participants were under the age of 28 d). In keeping with the predefined protocol, we included children and adults in the analysis to explore the impact of the intervention across all populations and then examine for credible subgroup effects.

Data Extraction and Quality Assessment

Reviewers extracted data independently and in duplicate using prepiloted data abstraction forms. We assessed risk of bias (RoB) independently and in duplicate using a modified Cochrane RoB tool (16) for which each domain is rated as “low,” “probably low,” “high,” or “probably high.” We rated the overall RoB for an individual study as the highest risk attributed to any domain. We assessed the overall certainty of evidence for each outcome using the Grading Recommendations Assessment, Development, and Evaluation (GRADE) approach (17).

Statistical Analysis

We used DerSimonian and Laird (18) random-effects models to conduct meta-analyses using RevMan 5.3 (Cochrane Collaboration, Oxford, United Kingdom) software. We generated study weights using the inverse variance method and present results as RRs and risk difference (RD) for dichotomous outcomes and as mean differences (MDs) for continuous outcomes, both with 95% CIs. We converted medians to means when necessary using the method by Shi et al (19). We calculated absolute effects using the pooled baseline prevalence from the control arm of included trials.

For further details on heterogeneity, publication bias, study selection, subgroup analyses, and trial sequential analyses (TSA), please refer to the Supplementary Appendix (https://links.lww.com/CCM/H132).

RESULTS

Of 4,094 citations identified in the search (Supplementary Appendix 6, Supplement Fig. 1, https://links.lww.com/CCM/H132), we assessed 151 full texts and included 65 RCTs enrolling 8,483 patients. Baseline characteristics of included trials are summarized in Supplementary Appendix 9, Supplement Table 1 (https://links.lww.com/CCM/H132).

Description of Included Studies

We included an additional 35 studies (n = 5,511) since the last major systematic review and meta-analysis addressing this topic (11). Fifteen studies (n = 4,564) were multicenter and 50 were single-center (n = 3,919). Of eight studies (n = 670) enrolling pediatric patients (20–27), five of these (n = 574) included patients younger than 14 years, whereas two (n = 76) included patients younger than 17 years; one study (n = 20) included patients younger than 22 years old and had an average age of participants of 7 years (26). Of 65 studies, 57 (n = 7,813) included only critically ill adults.

Thirteen trials assessed synbiotics (n = 1,190) and 52 assessed probiotics (n = 7,293). Of note, 42 studies enrolled less than 100 patients. The following microorganisms were used at various doses and combinations throughout the studies (Supplementary Appendix 9, Supplement Table 1, https://links.lww.com/CCM/H132): Lacticaseibacillus casei, Lactiplantibacillus plantarum, L. acidophilus, L. delbrueckii, L. bulgaricus, Bifidobacterium longum, B. breve, B. salivarius, Pediococcus pentosaceus, Lactococcus raffinolactis, B. infantis, B. bifidum, Streptococcus thermophilus, Ligilactobacillus salivarius, L. lactis, B. lactis, Saccharomyces boulardii, L. rhamnosus GG, L. johnsonii, L. casei, S. faecalis, Clostridium butyricum, Bacillus mesentericus, L. sporogenes, S. boulardii, L. paracasei, B. subtilis, and Enterococcus faecium. The most common probiotic organisms used in isolation were L. rhamnosus GG (15,22,26–33) (n = 3,152), L. plantarum (34–39) (n = 506), and S. boulardii (40–42) (n = 186).

Of 65 trials, close to half were at probably high RoB (21,26,30,32,33,35,39,43–67) (n = 2,674). Of the remaining, two were judged to be at high RoB (62,68) (n = 129), 26 trials at probably low RoB (22,24,27–29,31,34,36,40–42,69–83) (n = 2,300), and seven trials were judged to be at low RoB (15,20,23,25,38,84,85) (n = 3,380) (Supplementary Appendix 9, Supplement Table 2, https://links.lww.com/CCM/H132, for all RoB judgments).

Efficacy Outcomes

Infections.

Table 1 shows the summary of findings for the outcomes including the certainty of evidence. Based on the pooled results, probiotics may reduce VAP (RR, 0.72; 95% CI, 0.59–0.89 and RD, 6.9% reduction; 95% CI, 2.7% fewer to 10.2% fewer); however, this is based on low certainty evidence limited by RoB and inconsistency (Fig. 1 and Table 1). As per the TSA analysis, the optimal information size was reached for VAP (n = 2,212) (Supplementary Appendix 8, Supplement Fig. 44, https://links.lww.com/CCM/H132). Statistical testing for VAP suggested a possibility of publication bias; however, visual inspection of the funnel plot did not suggest important publication bias, as such we did not lower our certainty further (Supplementary Appendix 9, Supplement Table 5, https://links.lww.com/CCM/H132). Similarly, probiotics may reduce healthcare-associated pneumonia (HAP) (RR, 0.70; 95% CI, 0.55–0.89 and RD, 5.5% reduction; 95% CI, 2.0% fewer to 8.2% fewer; low certainty) (Supplementary Appendix 7, Supplement Fig. 2, https://links.lww.com/CCM/H132), catheter-related bloodstream infections (CRBSIs) (RR, 0.57; 95% CI, 0.27–1.19 and RD, 1.5% reduction; 95% CI, 2.5% reduction to 0.6% increase; low certainty) (Supplementary Appendix 7, Supplement Fig. 3, https://links.lww.com/CCM/H132), and the risk of other healthcare-associated infections (RR, 0.66; 95% CI, 0.55–0.80 and RD, 12.2% reduction; 95% CI, 16.2% reduction to 7.2% reduction; low certainty) (Supplementary Appendix 7, Supplement Fig. 4, https://links.lww.com/CCM/H132). Probiotics have an uncertain effect on Clostridioides difficile infection (RR, 0.43; 95% CI, 0.15–1.17 and RD, 1.9% reduction; 95% CI, 2.8% reduction to 0.6% increase; very low certainty) (Supplementary Appendix 7, Supplement Fig. 5, https://links.lww.com/CCM/H132) and probably have no effect on urinary tract infections (RR, 0.94; 95% CI, 0.78–1.12 and RD, 0.6% reduction; 95% CI, 2.3% reduction to 1.3% increase; moderate certainty) (Supplementary Appendix 7, Supplement Fig. 6, https://links.lww.com/CCM/H132).

TABLE 1. - Abbreviated Grading Recommendations Assessment, Development, and Evaluation Summary of Findings
Outcomes No. of Patients Effect Certainty Importance
No. of Studies Probiotics/Synbiotics Placebo Relative (95% CI) Absolute (95% CI)
Ventilator-associated pneumonia
 17 501/2,367 (21.2%) 588/2,371 (24.8%) RR 0.72 (0.59–0.89) 69 fewer per 1,000 (from 102 fewer to 27 fewer) ⊕⊕OO Critical
Low a
Healthcare-associated pneumonia
 15 80/662 (12.1%) 130/712 (18.3%) RR 0.70 (0.55–0.89) 55 fewer per 1,000 (from 82 fewer to 20 fewer) ⊕⊕OO Critical
Low
Mortality
 47 725/3,513 (20.6%) 794/3,739 (21.2%) RR 0.95 (0.87–1.04) 11 fewer per 1,000 (from 28 fewer to 8 more) ⊕⊕⊕OO Critical
Moderate
Serious adverse events
 18 11/2,197 (0.5%) 0/2,193 (0.0%) RR 9.96 (1.25–79.09) 0 fewer per 1,000 (from 0 fewer to 0 fewer) ⊕⊕OO Critical
Low
ICU length of stay
 31 2,806 2,811 MD 1.38 lower (2.19 lower to 0.57 lower) ⊕⊕OO Important
Low
Duration of mechanical ventilation
 12 785 801 MD 2.53 lower (3.74 lower to 1.31 lower) ⊕⊕OO Important
Low
Hospital length of stay
 27 2,824 2,842 MD 2.21 lower (3.24 lower to 1.18 lower) ⊕⊕OO Important
Low a
MD = mean difference, RR = risk ratio.
aAlthough statistical testing suggested small study effects (publication bias), we have already lowered certainty for risk of bias, and the funnel plot did not appear asymmetric based on visual inspection. The combination of borderline risk of bias, borderline inconsistency, and borderline publication bias did not add up to lowering by three levels but rather by two.
Dashes indicate continuous outcomes do not have relative effects estimates.

F1
Figure 1.:
Forest plot comparing probiotics/synbiotics and placebo for the outcome of ventilator-associated pneumonia. df = degrees of freedom.

Mortality.

Probiotics probably have no effect on mortality (RR, 0.95; 95% CI, 0.87–1.04 and RD, 1.1% reduction; 95% CI, 2.8% reduction to 0.8% increase; moderate certainty) (Fig. 2; and Supplementary Appendix 9, Supplement Table 2, https://links.lww.com/CCM/H132).

F2
Figure 2.:
Forest plot comparing probiotics/synbiotic and placebo for the outcome of mortality at the longest point of follow-up. df = degrees of freedom.

Morbidity.

Probiotics may reduce duration of invasive mechanical ventilation (IMV) (MD, 2.53 d fewer; 95% CI, 1.31 d fewer to 3.74 d fewer; low certainty) (Supplementary Appendix 7, Supplement Fig. 7, https://links.lww.com/CCM/H132), hospital length of stay (LOS) (MD, 2.21 d fewer; 95% CI, 1.18 d fewer to 3.24 d fewer; low certainty) (Supplementary Appendix 7, Supplement Fig. 8, https://links.lww.com/CCM/H132), and ICU LOS (MD, 1.38 d fewer; 95% CI, 0.57 d fewer to 2.19 d fewer; low certainty) (Supplementary Appendix 7, Supplement Fig. 9, https://links.lww.com/CCM/H132). Statistical testing for hospital LOS suggested a possibility of publication bias; however, visual inspection of the funnel plot did not suggest important publication bias, and therefore we did not further lower our certainty for this outcome (Supplementary Appendix 9, Supplement Table 5, https://links.lww.com/CCM/H132). Probiotics have an uncertain effect on duration of antibiotics (MD, 1.77 d fewer; 95% CI, 0.17 d fewer to 3.36 d fewer; very low certainty) (Supplementary Appendix 7, Supplement Fig. 10, https://links.lww.com/CCM/H132).

Probiotics have an uncertain effect on organ dysfunction (standardized MD, –0.22 points; 95% CI, –0.78 to 0.35; very low certainty) (Supplementary Appendix 7, Supplement Fig. 11, https://links.lww.com/CCM/H132). With respect to advanced life support, probiotics likely have no effect on the initiation of IMV (RR, 1.04; 95% CI, 0.85–1.27 and RD, 2.1% increase; 95% CI, 8.7% fewer to 15.7% increase; moderate certainty) (Supplementary Appendix 7, Supplement Fig. 12, https://links.lww.com/CCM/H132) or inotropic/vasopressor therapy (RR, 1.08; 95% CI, 0.79–1.48 and RD, 2.1% increase; 95% CI, 5.5% reduction to 12.1% increase; low certainty) (Supplementary Appendix 7, Supplement Fig. 13, https://links.lww.com/CCM/H132). No trials reported the effect of probiotics on the initiation of renal replacement therapy.

Probiotics likely have no effect on the frequency of diarrhea (RR, 0.98; 95% CI, 0.85–1.12 and RD, 1.0% reduction; 95% CI, 7.2% fewer to 5.8% increase; moderate certainty) (Supplementary Appendix 7, Supplement Fig. 14, https://links.lww.com/CCM/H132) and have an uncertain effect on the duration of diarrhea (MD, 2.59 d fewer; 95% CI, 5.59 d fewer to 0.41 d increase; very low certainty) (Supplementary Appendix 7, Supplement Fig. 15, https://links.lww.com/CCM/H132).

Possible Harms

Probiotics may increase serious adverse events (SAEs), albeit they were defined differently across trials (RR, 9.96; 95% CI, 1.25–79.09; RD, 0%; low certainty) (Fig. 3; and Supplementary Appendix 9, Table 3 [https://links.lww.com/CCM/H132] ). Among 18 trials documenting SAEs, only two trials reported any events (15,72); nine patients developed mesenteric ischemia and two had Lactobacillus species isolated in a culture from a sterile site or as the sole or predominant organism cultured from a nonsterile site, associated with persistent or significant disability or incapacity that was life-threatening or resulted in death. Probiotics may also increase the frequency of probiotic organism isolates from sterile sites or as the sole or predominant organism in a nonsterile site (RR, 15.16; 95% CI, 2.01–114.60 and RD, 0.8% increase; 95% CI, 0.01% increase to 6.7% increase; low certainty) with 15 events in the probiotics arm and 1 in the placebo arm (Supplementary Appendix 7, Supplement Fig. 16, https://links.lww.com/CCM/H132). Among nine trials documenting this outcome, only one trial reported any events (15).

F3
Figure 3.:
Forest plot comparing probiotics/synbiotics and placebo for the outcome of serious adverse events. df = degrees of freedom.

Subgroup Analysis

The majority of prespecified subgroup analyses did not demonstrate credible subgroup effects for any of the outcomes of interest. Analysis based on RoB demonstrated subgroup effects for the outcomes of VAP (p interaction = 0.01) (Fig. 4) and hospital LOS (p interaction = 0.004) (Supplementary Appendix 7, Supplement Fig. 18, https://links.lww.com/CCM/H132) resulting in rating down of their certainty using the GRADE approach. There were no credible subgroup effects found when comparing L. rhamnosus GG to other probiotics (Supplementary Appendix 7, Supplement Figs. 19–23, https://links.lww.com/CCM/H132). Post hoc subgroup analyses at the request of reviewers examining the probiotic L. plantarum found no credible subgroup effects for any of the outcomes (Supplementary Appendix 7, Supplement Figs. 24–28, https://links.lww.com/CCM/H132). Of note, there was no evidence of effect modification for any of the outcomes when comparing probiotics and synbiotics (Supplementary Appendix 7, Supplement Figs. 29–33, https://links.lww.com/CCM/H132). Trial reporting precluded the subgroup analysis comparing patients in shock with those without shock.

F4
Figure 4.:
Forest plot comparing probiotics/synbiotics and placebo for the outcome of ventilator-associated pneumonia by quality of study. df = degrees of freedom, RoB = risk of bias, VAP = ventilator-associated pneumonia.

Sensitivity Analysis

Post hoc sensitivity analyses without the pediatric data did not change the outcomes (Supplementary Appendix 7, Supplement Figs. 34–38, https://links.lww.com/CCM/H132). Further post hoc sensitivity analyses without the high RoB studies found that the use of probiotics or synbiotics had no effect on VAP (RR, 0.91; 95% CI, 0.73–1.13) (Supplementary Appendix 7, Supplement Fig. 39, https://links.lww.com/CCM/H132), mortality (RR, 0.98; 95% CI, 0.89–1.08) (Supplementary Appendix 7, Supplement Fig. 40, https://links.lww.com/CCM/H132), HAP (RR, 0.89; 95% CI, 0.45–1.76) (Supplementary Appendix 7, Supplement Fig. 41, https://links.lww.com/CCM/H132), ICU LOS, (MD, 1.00 d fewer; 95% CI, 1.84 d fewer to 0.16 d more) (Supplementary Appendix 7, Supplement Fig. 42, https://links.lww.com/CCM/H132), and hospital LOS (MD, 0.57 d fewer; 95% CI, 1.92 d fewer to 0.78 d more) (Supplementary Appendix 7, Supplement Fig. 43, https://links.lww.com/CCM/H132).

DISCUSSION

This systematic review and meta-analysis suggest that probiotics or synbiotics may reduce VAP, HAP, and other healthcare infections when used in critically ill patients; however, this is based on low certainty evidence. Despite this possible reduction in nosocomial infections, probiotics do not influence the risk of mortality, inotrope/vasopressor therapy, initiation of IMV, or frequency of diarrhea, and have an uncertain effect on rates of CRBSI, CDI, duration of diarrhea, and organ dysfunction.

One concerning finding is the risk of the administered probiotic being isolated from sterile sites or as the sole or predominant organism cultured from a nonsterile site; of note, all of the events for this outcome were from one trial (15). Probiotics may also increase SAEs; all such reported outcomes were from two studies; one RCT documenting mesenteric ischemia in nine patients and a second RCT isolating Lactobacillus species in two patients in a culture from a sterile site associated with prolongation of ICU admission (15,72). Of note, the study that reported an increased risk of mesenteric ischemia examined the role of post-pyloric synbiotic administration in critically ill patients with pancreatitis (72); however, there were concerns with this study and how serious adverse events were captured (86). Specifically, 31 of the 33 deaths were not reported directly to the data and ethics committees and were reported later; thus, the evaluation of these incidents could not take place during the conduct of the study (86). Furthermore, the post-pyloric administration of probiotics in that study raises caution about that route of administration. The SAEs summarized in this review were observed in the strict clinical research environment; generalized use of probiotics among unselected patients in practice may increase this risk.

This updated review was prompted by the publication of Probiotics to Prevent Severe Pneumonia and Endotracheal Colonization Trial (PROSPECT) (NCT 02462590), which concluded that probiotics (L. rhamnosus GG) conferred no health benefits among mechanically ventilated critically ill patients (15). This meta-analysis more than doubles the number of trials and almost triples the number of patients included in the most recent meta-analysis (11). Results overall reflect low certainty evidence for most outcomes, as opposed to PROSPECT (NCT 02462590), the largest trial to date, which did not find that probiotics reduce VAP, CDI, ICU LOS, or mortality. Reasons for the discordant findings between some previous trials and PROSPECT may reflect the fact that prior trials suggesting benefit were at higher RoB compared with more recent research. For instance, subgroup effects were found for the outcomes of VAP (p interaction = 0.01) and hospital LOS (p interaction = 0.004) when comparing high versus low RoB studies; in particular, studies at high RoB were more likely to have an effect in showing reduced rates of VAP (Fig. 4) and hospital LOS (Supplementary Appendix 7, Supplement Fig. 18, https://links.lww.com/CCM/H132). Furthermore, sensitivity analyses without the high RoB studies negated the effects found for VAP, HAP, and hospital LOS. Further, 42 of the trials included in this systematic review enrolled fewer than 100 patients, which may increase the risk of type I error, which is mitigated by a larger trial, all else being equal. Taken together, the limitations of some of the included trials suggest the overall results of this meta-analysis need to be interpreted with caution (Supplementary Appendix 9, Supplement Table 1, https://links.lww.com/CCM/H132).

Another reason for discrepant results between PROSPECT (NCT 02462590) and other trials relate to patient illness severity and the probiotic evaluated. Patients in PROSPECT (NCT 02462590) appear to be sicker (overall mortality of 28.1%) than those enrolled in the other trials (pooled mortality for trials without PROSPECT was 17.4%), reflecting enrollment criteria requiring anticipated duration of invasive ventilation of at least 72 hours. Furthermore, to our knowledge, only one of the included trials ensured the integrity of probiotic dosing by conducting capsule quality control measures throughout (PROSPECT [NCT 02462590]); this trial showed no effect of L. rhamnosus GG on any clinical outcomes.

These findings represent the most current, comprehensive summary of evidence to guide practice for critically ill patients. While PROSPECT (NCT 02462590) did not demonstrate an improvement in clinical outcomes, this review found a potential reduction in VAP, albeit based on the totality of randomized trial data representing low certainty evidence that is further dampened by the sensitivity analyses without high RoB studies. Specifically, for the outcome of VAP, nine of the 17 trials were low RoB (n = 3,914 patients) were judged to be at low RoB, whereas 11 RCTs (n = 824 patients) were judged to be at high RoB. Subgroup analysis suggested studies judged to be at high RoB demonstrated a larger reduction in VAP compared with those judged to be low RoB, and this analysis contributed to the overall low certainty in pooled estimates. As such, despite the large number of trials and a pooled effect estimate suggesting benefit, the effect of probiotics on VAP in ICU patients remains somewhat uncertain. Given the heterogeneity of patients in the included studies, further research may explore whether there are specific subgroups of patients who may benefit from probiotics in reducing VAP. We hope that this systematic review and meta-analysis may inform future guidelines, which will consider the balance of benefits, harms, values, preferences, and costs in developing clinical recommendations.

This systematic review and meta-analysis has several strengths including a preregistered protocol, a comprehensive literature search including unpublished sources, duplicate and independent screening and data abstraction, and GRADE assessment of certainty of evidence. Inclusion of pediatric literature allows for a more robust and generalizable understanding of the effect of probiotics in critical illness. Given the sensitivity analysis excluding studies done in children did not change the results, and subgroup analysis comparing children versus adults did not demonstrate evidence of credible subgroup effect, this evidence should be considered to be generalizable to both the adult and pediatric critical care communities. Furthermore, with an additional 35 studies (n = 5,511 patients) since the last major systematic review, this body of evidence is the largest to date. This report also has limitations. Foreign language papers were only abstracted by one reviewer and infectious outcome definitions varied across trials. We were unable to perform one preplanned subgroup analysis (shock compared with no shock) due to lack of sufficient granularity in published trials. Many of the outcomes were based on low or very low certainty of evidence, rendering definitive conclusions challenging. Furthermore, we were unable to study the impact of probiotic dose and the role of enteral nutrition due to significant heterogeneity and lack of data.

CONCLUSIONS

Probiotics or synbiotics may reduce VAP, HAP, and other healthcare infections when used in critically ill patients, however, this is based on low certainty evidence due in large part to high RoB studies. Despite this possible reduction in nosocomial infections, probiotics do not appear to have an effect on mortality, use of vasopressors, initiating IMV or its duration, or the frequency or duration of diarrhea. When administered to critically ill patients, probiotic organisms may be identified in sterile sites or as the sole or predominant organism in a nonsterile site.

ACKNOWLEDGMENTS

We would like to thank Karin Dearness, Director of library services, St. Joseph’s Healthcare, Hamilton, for her assistance in performing the comprehensive search of the databases.

REFERENCES

1. Oami T, Chihade DB, Coopersmith CM, et al.: The microbiome and nutrition in critical illness. Curr Opin Crit Care. 2019; 25:145–149
2. Mittal R, Coopersmith CM, et al.: Redefining the gut as the motor of critical illness. Trends Mol Med. 2014; 20:214–223
3. Jacobs MC, Haak BW, Hugenholtz F, et al.: Gut microbiota and host defense in critical illness. Curr Opin Crit Care. 2017; 23:257–263
4. Food and Agriculture Organization of the United Nations/World Health Organization: Evaluation of Health and Nutritional Properties of Powder Milk and Live Lactic Acid Bacteria. 2001, pp 1–4
5. Wischmeyer PE, McDonald D, Knight R, et al.: Role of the microbiome, probiotics, and ‘dysbiosis therapy’ in critical illness. Curr Opin Crit Care. 2016; 22:347–353
6. Khailova L, Petrie B, Baird CH, et al.: Lactobacillus rhamnosus GG and Bifidobacterium longum attenuate lung injury and inflammatory response in experimental sepsis. PLoS One. 2014; 9:e97861
7. Klingensmith NJ, Coopersmith CM, et al.: The gut as the motor of multiple organ dysfunction in critical illness. Crit Care Clin. 2016; 32:203–212
8. Schrezenmeir J, de Vrese M, et al.: Probiotics, prebiotics, and synbiotics–approaching a definition. Am J Clin Nutr. 2001; 73:361S–364S
9. Panigrahi P, Parida S, Nanda NC, et al.: A randomized synbiotic trial to prevent sepsis among infants in rural India. Nature. 2017; 548:407–412
10. Vincent JL, Rello J, Marshall J, et al.; EPIC II Group of Investigators: International study of the prevalence and outcomes of infection in intensive care units. JAMA. 2009; 302:2323–2329
11. Manzanares W, Lemieux M, Langlois PL, et al.: Probiotic and synbiotic therapy in critical illness: A systematic review and meta-analysis. Crit Care. 2016; 19:262
12. Besselink MG, Timmerman HM, Buskens E, et al.; Dutch Acute Pancreatitis Study Group: Probiotic prophylaxis in patients with predicted severe acute pancreatitis (PROPATRIA): Design and rationale of a double-blind, placebo-controlled randomised multicenter trial [ISRCTN38327949]. BMC Surg. 2004; 4:12
13. Sniffen JC, McFarland LV, Evans CT, et al.: Choosing an appropriate probiotic product for your patient: An evidence-based practical guide. PLoS One. 2018; 13:e0209205
14. Hempel S, Newberry SJ, Maher AR, et al.: Probiotics for the prevention and treatment of antibiotic-associated diarrhea: A systematic review and meta-analysis. JAMA. 2012; 307:1959–1969
15. Johnstone J, Meade M, Lauzier F, et al.; Prevention of Severe Pneumonia and Endotracheal Colonization Trial (PROSPECT) Investigators and the Canadian Critical Care Trials Group: Effect of probiotics on incident ventilator-associated pneumonia in critically ill patients: A randomized clinical trial. JAMA. 2021; 326:1024–1033
16. Higgins JP, Altman DG, Gøtzsche PC, et al.; Cochrane Bias Methods Group; Cochrane Statistical Methods Group: The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011; 343:d5928
17. Guyatt GH, Oxman AD, Vist GE, et al.; GRADE Working Group: GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008; 336:924–926
18. DerSimonian R, Laird N, et al.: Meta-analysis in clinical trials. Control Clin Trials. 1986; 7:177–188
19. Shi J, Luo D, Weng H, et al.: Optimally estimating the sample mean and standard deviation from the five-number summary. arXiv:2003.02130
20. Angurana SK, Bansal A, Singhi S, et al.: Evaluation of effect of probiotics on cytokine levels in critically ill children with severe sepsis: A double-blind, placebo-controlled trial. Crit Care Med. 2018; 46:1656–1664
21. Banupriya B, Biswal N, Srinivasaraghavan R, et al.: Probiotic prophylaxis to prevent ventilator associated pneumonia (VAP) in children on mechanical ventilation: An open-label randomized controlled trial. Intensive Care Med. 2015; 41:677–685
22. Abstract translations. Pediatr Crit Care Med. 2011; 12:A69-A70
23. Kumar S, Bansal A, Chakrabarti A, et al.: Evaluation of efficacy of probiotics in prevention of candida colonization in a PICU-a randomized controlled trial. Crit Care Med. 2013; 41:565–572
24. Simakachorn N, Bibiloni R, Yimyaem P, et al.: Tolerance, safety, and effect on the faecal microbiota of an enteral formula supplemented with pre- and probiotics in critically ill children. J Pediatr Gastroenterol Nutr. 2011; 53:174–181
25. Wang Y, Gao L, Yang Z, et al.: Effects of probiotics on ghrelin and lungs in children with acute lung injury: A double-blind randomized, controlled trial. Pediatr Pulmonol. 2018; 53:197–203
26. Mayes T, Gottschlich MM, James LE, et al.: Clinical safety and efficacy of probiotic administration following burn injury. J Burn Care Res. 2015; 36:92–99
27. Honeycutt TC, El Khashab M, Wardrop RM 3rd, et al.: Probiotic administration and the incidence of nosocomial infection in pediatric intensive care: A randomized placebo-controlled trial. Pediatr Crit Care Med. 2007; 8:452–458; quiz 464
28. Ferrie S, Daley M, et al.: Lactobacillus GG as treatment for diarrhea during enteral feeding in critical illness: Randomized controlled trial. JPEN J Parenter Enteral Nutr. 2011; 35:43–49
29. Forestier C, Guelon D, Cluytens V, et al.: Oral probiotic and prevention of Pseudomonas aeruginosa infections: A randomized, double-blind, placebo-controlled pilot study in intensive care unit patients. Crit Care. 2008; 12:R69
30. Kwon JH, Bommarito KM, Reske KA, et al.; Centers for Disease Control and Prevention (CDC) Prevention Epicenters: Randomized controlled trial to determine the impact of probiotic administration on colonization with multidrug-resistant organisms in critically ill patients. Infect Control Hosp Epidemiol. 2015; 36:1451–1454
31. Morrow LE, Kollef MH, Casale TB, et al.: Probiotic prophylaxis of ventilator-associated pneumonia: A blinded, randomized, controlled trial. Am J Respir Crit Care Med. 2010; 182:1058–1064
32. Sharafkhah M, Zarinfar N, Amiri M, et al.: Probiotic effects in prevention from ventilator-associated pneumonia. Koomesh. 2016; 17:803–813
33. Malian M, Reichenbach R, Peck Arika, et al: 2013 Abstract. Crit Care Med. 2012; 40:275
34. Oláh A, Belágyi T, Issekutz A, et al.: Randomized clinical trial of specific lactobacillus and fibre supplement to early enteral nutrition in patients with acute pancreatitis. Br J Surg. 2002; 89:1103–1107
35. Qin HL, Zheng JJ, Tong DN, et al.: Effect of Lactobacillus plantarum enteral feeding on the gut permeability and septic complications in the patients with acute pancreatitis. Eur J Clin Nutr. 2008; 62:923–930
36. Klarin B, Johansson ML, Molin G, et al.: Adhesion of the probiotic bacterium Lactobacillus plantarum 299v onto the gut mucosa in critically ill patients: A randomised open trial. Crit Care. 2005; 9:R285–R293
37. Klarin B, Molin G, Jeppsson B, et al.: Use of the probiotic Lactobacillus plantarum 299 to reduce pathogenic bacteria in the oropharynx of intubated patients: A randomised controlled open pilot study. Crit Care. 2008; 12:R136
38. Litton E, Anstey M, Broadhurst D, et al.: Early and sustained Lactobacillus plantarum probiotic therapy in critical illness: The randomised, placebo-controlled, restoration of gut microflora in critical illness trial (ROCIT). Intensive Care Med. 2021; 47:307–315
39. McNaught CE, Woodcock NP, Anderson AD, et al.: A prospective randomised trial of probiotics in critically ill patients. Clin Nutr. 2005; 24:211–219
40. Schlotterer M, Bernasconi P, Lebreton F, et al.: Intérêt de Saccharomyces boulardii dans la tolérance digestive de la nutrition entérale à débit continu chez le brûlé. Nutr Clin Metab. 1987; 1:31–34
41. Tempé JD, Steidel AL, Blehaut H, et al.: Prevention of diarrhea administering Saccharomyces boulardii during continuous enteral feeding. Sem Hop. 1983; 59:1409–1412
42. Bleichner G, Bléhaut H, Mentec H, et al.: Saccharomyces boulardii prevents diarrhea in critically ill tube-fed patients. A multicenter, randomized, double-blind placebo-controlled trial. Intensive Care Med. 1997; 23:517–523
43. Cui LH, Wang XH, Peng LH, et al.: The effects of early enteral nutrition with addition of probiotics on the prognosis of patients suffering from severe acute pancreatitis. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2013; 25:224–228
44. Kate V, Sureshkumar S, Rohith G, et al.: 836 Effect of synbiotics in reducing the systemic inflammatory response and septic complications in moderately severe and severe acute pancreatitis- a prospective parallel arm double-blind randomized trial. Gastroenterology. 2020; 158:S-166–S-167
45. Mallick S, Kathirvel M, Thillai M, et al.: PrePro Trial: Randomized double-blind placebo controlled trial to analyze the effect of synbiotics on infectious complications following living donor liver transplantion. [CTRI no. - CTRI/2017/09/009869]. HPB. 2018; 20:S283–S284
46. Abstracts of the Society of Critical Care Medicine 40th Critical Care Congress. San Diego, California, USA. January 15-19, 2011. Crit Care Med. 2010; 38(12 Suppl):A1–A306
47. Yongmei L, et al.: Adjuvant treatment of 14 cases of severe acute pancreatitis with probiotics. World Chin J Dig. 2007; 15:302–304
48. Oláh A, Belágyi T, Issekutz A, et al.: Combination of early nasojejunal feeding with modern synbiotic therapy in the treatment of severe acute pancreatitis (prospective, randomized, double-blind study). Magy Seb. 2005; 58:173–178
49. Rammohan A, Sathyanesan J, Rajendran K, et al.: Synbiotics in surgery for chronic pancreatitis: Are they truly effective? A single-blind prospective randomized control trial. Ann Surg. 2015; 262:31–37
50. Randomized controlled study of probiotics containing Lactobacillus casei (Shirota strain) for prevention of ventilator-associated pneumonia. J Med Assoc Thai. 2015; 98:253–259
51. Shimizu K, Yamada T, Ogura H, et al.: Synbiotics modulate gut microbiota and reduce enteritis and ventilator-associated pneumonia in patients with sepsis: A randomized controlled trial. Crit Care. 2018; 22:239
52. Šrámek V, Dadak L, Štouračová M, et al.: Impact of addition of synbiotics (Synbiotic 2000 Forte) to enteral nutrition on the course of MODS, occurrence of sepsis, immune status and gut function in long-term critically ill patients. Anesteziol Intenziv Med. 2007; 18:157–163
53. Shimizu K, Hirose T, Ogura H, et al.: PP019MON synbiotics reduced the incidence of diarrhea in ventilated patients with sirs. Clin Nutr Suppl. 2011; 6:122–122
54. Tan M, Lu XL, Duan JW, et al.: Effects of probiotics on blood glucose levels and clinical outcomes in patients with severe craniocerebral trauma. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2013; 25:627–630
55. Tan M, Zhu JC, Du J, et al.: Effects of probiotics on serum levels of Th1/Th2 cytokine and clinical outcomes in severe traumatic brain-injured patients: A prospective randomized pilot study. Crit Care. 2011; 15:R290
56. Thoma G, Stavrou G, Malliou P, et al.: OR06: The use of probiotics as a preventative measure against surgical site infections in multiple trauma ICU patients: A preliminary study. Clin Nutr. 2019; 38:S6
57. Wan G, Wang L, Zhang G, et al.: Effects of probiotics combined with early enteral nutrition on endothelin-1 and C-reactive protein levels and prognosis in patients with severe traumatic brain injury. J Int Med Res. 2020; 48:300060519888112
58. Wang G, Wen J, Xu L, et al.: Effect of enteral nutrition and ecoimmunonutrition on bacterial translocation and cytokine production in patients with severe acute pancreatitis. J Surg Res. 2013; 183:592–597
59. Wang J, Ke H, Liu K-X, et al.: Effects of exogenous probiotics on the gut microbiota and clinical outcomes in critically ill patients: A randomized controlled trial. Ann Palliat Med. 2021; 10:1180–1190
60. Wu P, Yu Y, Li L, et al.: Effect and safety of probiotics combined early enteral nutrition on severe acute pancreatitis patients. Biomed Res. 2017; 28:1403–1407
61. Xie CX, Ye CN, et al.: Effect of early enteral nutrition combined with synbiotics agents on normal intestinal flora and SIgA in patients with hypertensive intracerebral hemorrhage. Chin J Clin Nutr. 2013; 21:13–16
62. Zhuang CX, Zhang G, et al.: Probiotics prevent enteral nutrition-associated diarrhea in critically ill patients. Chin J Clin Nutr. 2012; 20:359–362
63. Zeng J, Wang CT, Zhang FS, et al.: Effect of probiotics on the incidence of ventilator-associated pneumonia in critically ill patients: A randomized controlled multicenter trial. Intensive Care Med. 2016; 42:1018–1028
64. Zhang H, Huang H-T, et al.: Bifidobacterium to prevent intestinal flora imbalance in patients with severe acute pancreatitis. China J. 2012; 24:73–75
65. Caixia X, Jingci Z, Guangfu H, et al.: Influence of early enteral nutrition combined with synbiotics agents on intestinal flora and SIgA of patients with severe craniocerebral injury [J]. Chin Nurs Res. 2010; 28
66. Xiaowei X, Yiyan Z, Li D, et al.: Probiotics as supplement for early enteral nutrition decreases infection in severe brain injury. J Third Mil Med Univ. 2013; 35:536–539
67. Yu CZ, Tuerhun A, Wang ZQ, et al.: Clinical application of homogenate diet and yogurt in patients with severe head injury. Chin J Clin Nutr. 2007; 15:232–236
68. Shinotsuka CR, Alexandre MR, David CM, et al.: Enteral nutritional therapy with pre, pro and symbiotic and gastrointestinal tract and inferior airway colonization in mechanically ventilated patients. Rev Bras Ter Intensiva. 2008; 20:241–248
69. Alberda C, Gramlich L, Meddings J, et al.: Effects of probiotic therapy in critically ill patients: A randomized, double-blind, placebo-controlled trial. Am J Clin Nutr. 2007; 85:816–823
70. Spindler-Vesel A, Bengmark S, Vovk I, et al.: Synbiotics, prebiotics, glutamine, or peptide in early enteral nutrition: A randomized study in trauma patients. JPEN J Parenter Enteral Nutr. 2007; 31:119–126
71. Barraud D, Blard C, Hein F, et al.: Probiotics in the critically ill patient: A double blind, randomized, placebo-controlled trial. Intensive Care Med. 2010; 36:1540–1547
72. Besselink MG, van Santvoort HC, Buskens E, et al.; Dutch Acute Pancreatitis Study Group: Probiotic prophylaxis in predicted severe acute pancreatitis: A randomised, double-blind, placebo-controlled trial. Lancet. 2008; 371:651–659
73. Falcão de Arruda IS, de Aguilar-Nascimento JE, et al.: Benefits of early enteral nutrition with glutamine and probiotics in brain injury patients. Clin Sci (Lond). 2004; 106:287–292
74. Frohmader TJ, Chaboyer WP, Robertson IK, et al.: Decrease in frequency of liquid stool in enterally fed critically ill patients given the multispecies probiotic VSL#3: A pilot trial. Am J Crit Care. 2010; 19:e1–e11
75. Giamarellos-Bourboulis EJ, Bengmark S, Kanellakopoulou K, et al.: Pro- and synbiotics to control inflammation and infection in patients with multiple injuries. J Trauma. 2009; 67:815–821
76. Jain PK, McNaught CE, Anderson AD, et al.: Influence of synbiotic containing Lactobacillus acidophilus La5, Bifidobacterium lactis Bb 12, Streptococcus thermophilus, Lactobacillus bulgaricus and oligofructose on gut barrier function and sepsis in critically ill patients: A randomised controlled trial. Clin Nutr. 2004; 23:467–475
77. Klarin B, Wullt M, Palmquist I, et al.: Lactobacillus plantarum 299v reduces colonisation of Clostridium difficile in critically ill patients treated with antibiotics. Acta Anaesthesiol Scand. 2008; 52:1096–1102
78. Knight DJ, Gardiner D, Banks A, et al.: Effect of synbiotic therapy on the incidence of ventilator associated pneumonia in critically ill patients: A randomised, double-blind, placebo-controlled trial. Intensive Care Med. 2009; 35:854–861
79. López de Toro Martín-Consuegra I, Sanchez-Casado M, Pérez-Pedrero Sánchez-Belmonte MJ, et al.: The influence of symbiotics in multi-organ failure: Randomised trial. Med Clin (Barc). 2014; 143:143–149
80. Mahmoodpoor A, Hamishehkar H, Asghari R, et al.: Effect of a probiotic preparation on ventilator-associated pneumonia in critically ill patients admitted to the intensive care unit: A prospective double-blind randomized controlled trial. Nutr Clin Pract. 2019; 34:156–162
81. Malik AA, Rajandram R, Tah PC, et al.: Microbial cell preparation in enteral feeding in critically ill patients: A randomized, double-blind, placebo-controlled clinical trial. J Crit Care. 2016; 32:182–188
82. Oláh A, Belágyi T, Pótó L, et al.: Synbiotic control of inflammation and infection in severe acute pancreatitis: A prospective, randomized, double blind study. Hepatogastroenterology. 2007; 54:590–594
83. Rayes N, Seehofer D, Theruvath T, et al.: Supply of pre- and probiotics reduces bacterial infection rates after liver transplantation–a randomized, double-blind trial. Am J Transplant. 2005; 5:125–130
84. Muscedere J, O'Callaghan N, Maslove D, et al.: ESICM LIVES 2017: 30th ESICM Annual Congress. September 23-27, 2017. Intensive Care Med Exp. 2017; 5(Suppl 2):44
85. Sanaie S, Ebrahimi-Mameghani M, Hamishehkasr H, et al.: Effect of a multispecies probiotic on inflammatory markers in critically ill patients: A randomized, double-blind, placebo-controlled trial. J Res Med Sci. 2014; 19:827–833
86. Sheldon T, et al.: Dutch probiotics study is criticised for its “design, approval, and conduct.” BMJ. 2010; 340:c77
Keywords:

clinical trial; critical illness; probiotics; randomized; synbiotics

Supplemental Digital Content

Copyright © 2022 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.