The ICU Microbiome Project: Is There a Better Way to Treat Infections Than Antibiotics?

By Paul E. Wischmeyer M.D., Professor of Anesthesiology and Pediatrics, University of Colorado School of Medicine.   Contact:


Is Our Current Approach To Infection Working?

A great deal of time and effort are spent in eradicating bacteria and other microbial, fungal, and viral species in the intensive care unit (ICU). The U.S. Centers for Disease Control reports that 55% of all hospitalized patients receive an antibiotic during their stay, and in the ICU this number increases to 70% of patients. As recently described by Singer and Gynne, it is likely that this antibiotic use has in part contributed to an impressive 22-fold fall in crude mortality rates for infectious diseases in the US between 1900 and 1980 (1). Yet, it is troubling that mortality rates from infectious disease (up to 1996) increased—by 50%—with the septicemia rate nearly doubling(1). And in reality, it is unclear if the earlier reductions in mortality and increased life expectancy were due primarily to antibiotics innovations, or more likely, due to improved public health and education.

The massive global reliance on antibiotic use comes at great financial expense with antibiotics accounting for up to 30% of a hospital’s drug budget (2). Unfortunately, greater than the financial costs are the potential risks that inappropriate antibiotic use and overuse carry for patients. Evidence suggests that as many as 37% of antibiotic regimens are unnecessary or uncompliant with guidelines(3). This inappropriate antibiotic use leads to the emergence of multi-drug resistant bacterial infections; the incidence of these infections is rising rapidly both in the U.S. and worldwide (4).  A recent New England Journal of Medicine Article estimates antibiotic-resistant Clostridium difficile occurs now in >450,000 patients per year in the U.S. alone (4). Unfortunately, these multi-drug resistant infections are also becoming increasingly lethal. For example, Clostridium difficile is estimated to contribute to ~30,000 deaths/year in the U.S. (4, 5). Further, The U.S. Centers for Disease Control indicates death rates from sepsis following infections (like C. diff) have increased at a rate greater than any other common cause of mortality in the last year for which data was available (6). And, while age adjusted death rates are decreasing in the U.S., the death rate from sepsis continues to rise significantly (2). And as stated, this is punctuated by mortality rates from infectious disease in general (up to 1996) increasing by 50%, again with the septicemia rate nearly doubling (1). Thus, more advanced antibiotics do not appear to be translating to increased survival from infectious disease, but instead increasingly aggressive resistant organisms and emergence of newly lethal pathogens like C. diff.  Is it possible we need to rethink our strategy towards microbial therapy in the ICU?


Antibiotics Kill More Than Just Pathogens…

These concerns around antibiotics are compounded by the fact that antibiotics currently used to attempt treat infection not only kill pathogens, but also “health promoting” microbes. These adverse effects include the hypothesized loss of commensal gastrointestinal (GI) microbiota, which enables overgrowth of unwanted organisms (dysbiosis). This may have significant implications for organs far outside the GI tract as well. The gut has long been described as the “motor” of systemic inflammatory response syndrome (SIRS) and organ failure, regardless of the location of the initial infection (7). Thus, the effect of alterations in the gut microbiota and gut barrier homeostasis are thought to be transmitted to and propagated by downstream organs, such as the spleen and lung where large immune cell populations are harbored (7-9), leading to inflammation-induced organ failure in the ICU.

Further, it is important to realize that at the cellular level, multiple organ failure, which is a final common cause of death in the ICU, has long been attributed to mitochondrial failure. It has been long known that mitochondria trace their evolution from bacteria that produce energy for our cells. Recent literature not surprisingly reveals that mitochondria are known to be damaged by many of the antibiotics we commonly administer in the ICU (1). Thus, we and others hypothesize that antibiotics may be contributing to organ failure by not only leading to dysbiosis, but also by damaging the very core of our cells’ energy production (1).


Is There Another, Perhaps Better, Way To Prevent And Treat Infections?     

As described in a number of very recent and comprehensive review articles (7, 10) laboratory based studies in animals have shown that alterations in intestinal homeostasis and gut microbiota in experimental critical illness have been associated with increased inflammatory cytokine production, gut barrier dysfunction, and increased cellular apoptosis, all of which can contribute to multiple organ failure (MOF). To modulate this “motor” of systemic inflammation it has been hypothesized that repletion of health promoting bacteria via probiotics, prebiotics, stool transplantation or combination therapies may be a promising intervention to maintain gut integrity and prevent pathologic alterations in the gut (and other body sites) microbiota or “dysbiosis” (11-13). Early clinical trials of probiotic use in the ICU have demonstrated some promise in reducing overall infections(14). These studies use a range of probiotic strains and doses, but no ideal probiotic or probiotic mixture has been identified based on actual microbiome-based characterization of the effect of critical illness on patient microbiota.

Thus, before large trials of “dysbiosis therapy” can be meaningfully undertaken, confirmation and characterization of the hypothesized “dysbiosis” of critical illness is urgently needed.  Major funding bodies and experts in the field are indicating more generalizable clinical studies of the ICU microbiome changes are required to better diagnose dysbiosis, and a few pilot (<15 patients) microbiome analysis-based studies have begun to assess these changes (15, 16). The results of these early studies show signals of concerning dysbiosis and loss of microbial diversity. These initial findings have led experts and major funding bodies in the field to conclude there is urgent need for larger, more generalizable, prospective studies that characterize the microbiome in a larger critical care population to confirm and characterize this potential dysbiosis and move towards therapeutic interventions using microbiome signatures (17).


The ICU Microbiome Project: Can We Characterize The Dysbiosis Of Critical Illness?

To attempt to address this question, a collaboration between The Knight Lab and Paul Wischmeyer, M.D.-a researcher and critical care physician at the University of Colorado-was formed a few years ago. This collaboration, which could not have happened without the essential efforts of Daniel McDonald and Gail Ackerman in the Knight Lab, sought to collect fecal, oral, and skin microbiome samples at two timepoints, within 48 hours of ICU admission, and at ICU discharge or ICU day 10. Four hospital centers in the United States and Canada participated using the existing International Nutrition Survey framework developed by Dr. Wischmeyer’s close collaborator, Dr. Daren Heyland at the Clinical Evaluation and Research Unit in Canada. Using this structure we collected samples from 115 patients using standard Earth Microbiome Project protocols for collection and processing as was utilized in the American Gut Project. ICU patient results were compared to a range of existing microbiome databases including a “healthy” cohort of American Gut Subjects.

Our initial results demonstrate that, when compared to healthy American Gut Subjects, critical illness shows rapid and distinct changes from a “healthy” fecal and oral microbiome. (Figure 1) Fecal ICU samples tend to have a lower relative abundance of Firmicutes (Figure 2), and increased relative abundance of Proteobacteria.


Figure 1. A PCoA plot of ICU samples (large spheres) together with a subset of American Gut samples (small spheres). Spheres are colored by body site.



Figure 2. A PCoA plot of ICU samples (large spheres) together with a subset of American Gut samples (small spheres). Spheres are colored by abundance of Firmicutes in the sample (red=higher abundance).


Large depletions were observed in organisms shown to confer anti-inflammatory benefits, such as Faecalibacterium (18), which produces short chain fatty acids that are vital to the gut. Conversely, many of the taxa that increased in ICU patients contain well-recognized pathogens such as Enterobacter and Staphylococcus. Ongoing analysis (and perhaps the subject of future blogs) by Daniel, Rob, and Paul will utilize SourceTracker to assess source composition of ICU samples and will use Qiita to examine the effects of critical illness on microbial diversity. Additionally, examination of potential relationships between changes in the ICU microbiome and clinical outcome will be examined. In summary, our initial data from the ICU Microbiome Project confirms that severe dysbiosis occurs in a broad, larger population of critically ill subjects. This data may help guide creation of targeted microbial therapies, focused on correcting potentially “illness-promoting” dysbioses using specific probiotics or targeted, multi-microbe “stool pills” to restore a healthy microbiome and improve outcomes in critical illness. And in the end, perhaps this is the beginning of a road to a better way to treat and prevent infection than the ubiquitous (and maybe questionably effective…in the long run) antibiotics universally given to most all patients in hospitals and ICUs today!


  1.      Singer M, Glynne P. Treating critical illness: the importance of first doing no harm. PLoS Med. 2005;2(6):e167.
  2.      Ruttimann S, Keck B, Hartmeier C, Maetzel A, Bucher HC. Long-term antibiotic cost savings from a comprehensive intervention program in a medical department of a university-affiliated teaching hospital. Clin Infect Dis. 2004;38(3):348-56.
  3.      Fridkin S, Baggs J, Fagan R, Magill S, Pollack LA, Malpiedi P, et al. Vital signs: improving antibiotic use among hospitalized patients. MMWR Morb Mortal Wkly Rep. 2014;63(9):194-200.
  4.      Lessa FC, Mu Y, Bamberg WM, Beldavs ZG, Dumyati GK, Dunn JR, et al. Burden of Clostridium difficile infection in the United States. N Engl J Med. 2015;372(9):825-34.
  5.      Rello J, Quintana E, Ausina V, Net A, Prats G. A three-year study of severe community-acquired pneumonia with emphasis on outcome. Chest. 1993;103(1):232-5.
  6.      Milbrandt EB, Kersten A, Rahim MT, Dremsizov TT, Clermont G, Cooper LM, et al. Growth of intensive care unit resource use and its estimated cost in Medicare. Crit Care Med. 2008;36(9):2504-10.
  7.      Klingensmith NJ, Coopersmith CM. The Gut as the Motor of Multiple Organ Dysfunction in Critical Illness. Crit Care Clin. 2016;32(2):203-12.
  8.      Broquet A, Roquilly A, Jacqueline C, Potel G, Caillon J, Asehnoune K. Depletion of natural killer cells increases mice susceptibility in a Pseudomonas aeruginosa pneumonia model. Crit Care Med. 2014;42(6):e441-50.
  9.      Khailova L, Baird CH, Rush AA, McNamee EN, Wischmeyer PE. Lactobacillus rhamnosus GG improves outcome in experimental pseudomonas aeruginosa pneumonia: potential role of regulatory T cells. Shock. 2013;40(6):496-503.
  10.    Dickson RP. The microbiome and critical illness. Lancet Respir Med. 2016;4(1):59-72.
  11.    Marini JJ, Gattinoni L, Ince C, Kozek-Langenecker S, Mehta RL, Pichard C, et al. A few of our favorite unconfirmed ideas. Critical care. 2015;19 Suppl 3:S1.
  12.    Krezalek MA, DeFazio J, Zaborina O, Zaborin A, Alverdy JC. The Shift of an Intestinal “Microbiome” to a “Pathobiome” Governs the Course and Outcome of Sepsis Following Surgical Injury. Shock. 2016;45(5):475-82.
  13.    Andrade ME, Araujo RS, de Barros PA, Soares AD, Abrantes FA, Generoso Sde V, et al. The role of immunomodulators on intestinal barrier homeostasis in experimental models. Clinical nutrition. 2015;34(6):1080-7.
  14.    Petrof EO, Dhaliwal R, Manzanares W, Johnstone J, Cook D, Heyland DK. Probiotics in the critically ill: a systematic review of the randomized trial evidence. Crit Care Med. 2012;40(12):3290-302.
  15.    Ojima M, Motooka D, Shimizu K, Gotoh K, Shintani A, Yoshiya K, et al. Metagenomic Analysis Reveals Dynamic Changes of Whole Gut Microbiota in the Acute Phase of Intensive Care Unit Patients. Dig Dis Sci. 2015.
  16.    Zaborin A, Smith D, Garfield K, Quensen J, Shakhsheer B, Kade M, et al. Membership and behavior of ultra-low-diversity pathogen communities present in the gut of humans during prolonged critical illness. MBio. 2014;5(5):e01361-14.
  17.    Lyons JD, Ford ML, Coopersmith CM. The Microbiome in Critical Illness: Firm Conclusions or Bact to Square One? Dig Dis Sci. 2016.
  18.    Sokol H, Pigneur B, Watterlot L, Lakhdari O, Bermudez-Humaran LG, Gratadoux JJ, et al. Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identified by gut microbiota analysis of Crohn disease patients. Proc Natl Acad Sci U S A. 2008;105(43):16731-6.


Paul Wischmeyer M.D., EDIC, is a Professor of Anesthesiology and Pediatrics (Nutrition Section) at the University of Colorado School of Medicine. My current research focuses on the role of the microbiome and dybiosis in the pathobiology and treatment of critical illness, surgery, and trauma. We also have long-standing interests in the role of nutrition and probiotics to improve outcome from critical illness, surgery, cancer, and other acute/chronic illnesses. Our past and current projects span the range of translational research from basic mechanistic pathways in cellular systems and in vivo proof–of–concept modeling to human pilot trials and large multi-center randomized controlled clinical trials.