The Microbiome and Antimicrobial Stewardship in Surgical Patients

Current State of Antimicrobial Resistance

Antimicrobial drugs are a critical component of patient care and are needed not only to treat active infections but to reduce the risk of surgical site infection. However, antimicrobial resistance (AMR) is a growing threat to our ability to care safely and effectively for acutely or critically ill or injured patients. The U.S. Centers for Disease Control and Prevention (CDC) reported in 2019 that more than 2.8 million infections with antibiotic-resistant pathogens occur in the United States, leading to more than 35,000 deaths. ¹

The prevalence of multi-drug–resistant organisms (MDROs) continues to grow. Key agents that have historically been the cornerstone for treatment of certain illnesses are now having their usefulness called into question. For example, fluoroquinolones are no longer recommended for the empiric treatment of gonococcal infections because of widespread resistance. ² Although many new antimicrobial agents have been developed in recent years to treat emerging MDROs, resistance to even these advanced agents has been reported. ³

Antimicrobial resistance negatively affects the care of surgical patients. A common example is methicillin-resistant Staphylococcus aureus (MRSA), a pathogen frequently associated with surgical site infections (SSIs). ⁴ The high prevalence of MRSA is a specific threat to the surgical patient. Indeed, SSIs due to MRSA are associated with higher morbidity, higher mortality, and increased lengths of stay compared with infections with methicillin-sensitive isolates. ⁵ The increasing prevalence of MDROs can lead to an antibiotic “arms race” in the hospital, with increased broad-spectrum medication use leading to increased AMR rates. This process has been aptly demonstrated with carbapenem exposure and the subsequent risk of carbapenem-resistant infection. ⁶ The increasing incidence and prevalence of MDROs in both the community and the hospital, and the management strategies utilized to address them, are now major considerations. One area of evolving interest is the human microbiome and its role in either enabling or mitigating against antimicrobial resistance.

The Role of the Human Microbiome

The term microbiota refers to the collective microbial community normally inhabiting the human body, with the microbiome referring to the collective genomic content of the microbiota. However, the terms microbiota and microbiome are often used interchangeably.^7,8 The human microbiota has many vital roles in human health. This importance is seen even early in human life, as neonates are born deficient in vitamin K because of poor diffusion through the placenta. ⁹ As the gut microbiota develops in the neonate, intestinal bacteria produce vitamin K2, which assists with oxidative phosphorylation. ⁷ The gut microbiota plays multiple important roles in human metabolism, such as degrading complex carbohydrates and plant polysaccharides into short-chain fatty acids with beneficial effects on gut motility and wound healing. ⁸ Because of its importance in metabolism, alterations in the gut microbiome have been associated with pathologic conditions related to metabolism such as obesity, diabetes mellitus, and cardiovascular disease. ⁸

Although the gut microbiota attracts a great deal of scientific attention, microbiota of other parts of the body are also of clinical relevance. The vaginal microbiota may influence the risk of urogenital diseases in women. ¹⁰ Relatedly, a predominance of certain bacteria dominating the vaginal microbiome could lead to an increased risk of acquiring human immunodeficiency virus (HIV) infection. ¹¹ The lung microbiome may play a role in the development of immunity through exposure to certain components of the microbiota. Finally, lack of the development of the pulmonary microbiota could lead to an increased risk of asthma and allergic diseases. ¹²

Gut Microbiome Populations

The specific species of bacteria that inhibit the human gut microbiota will vary between individuals based on factors such as country of origin and diet. ¹³ There are a few groups of bacteria, however, that appear to predominate. The nomenclature of these phyla undergoes periodic revision by microbial taxonomists, most recently in 2021. Major phyla in the human gut microbiota include Bacillota (formerly Firmicutes), Bacteroidota (formerly Cytophaga-Flavobacterium-Bacteroides, or CFB), and Pseudomonadota (formerly Proteobacteria). ¹⁴ Among these phyla, it appears that 90% of the bacteria are from the phyla Bacillota and Bacteroidota. The predominant species in the gut from phylum Bacteroidota include the genera Bacteroides and Prevotella, whereas predominant genera from Bacillota includes Clostridium, Eubacterium, and Ruminococcus. ¹⁵ Pseudomonadota include the eponymous genus Pseudomonas along with the order Enterobacterales, which contains numerous clinically relevant genera such as Escherichia, Klebsiella, and Enterobacter.

Despite the predominance of certain bacteria, the specific composition of bacterial populations differs throughout the human gastrointestinal tract. In the small intestine, Veillonella and Streptococcus species (particularly Streptococcus gallolyticus, Streptococcus mitis, and Streptococcus salivarius) are most common. ¹⁶ In the large intestine, Bacillota and Bacteroidota still predominate, but potential pathogens may also be present in small numbers, including Campylobacter jejuni, Salmonella enterica, and Vibrio cholerae, collectively comprising less than 0.1% of the microbiota. Escherichia coli (part of phylum Pseudomonadota) and Bacteroides fragilis (from phylum Bacteroidota) play significant roles in human disease and are generally thought of as major contributors to the microbiome. Although readily recovered from cultured human stool specimens, they too account for a similarly small portion of the gut microbiota (around 0.1%). ¹⁷

Changes in the Microbiome in Acute Illness

The human microbiota can be altered by critical illness. In 1969, Johansen et al. ¹⁸ described an increased percentage of gram-negative bacilli growing in pharyngeal cultures from critically ill patients compared with healthy or minimally ill patients. Subsequent work has reinforced this early concept: critical illness may have a profound effect on the microbiome, including the replacement of the healthy microbiota. This alteration of the microbiota is thought to be the result of three factors: introduction of new bacteria, elimination of current bacteria, and altered rates of bacterial reproduction. ¹⁹ The introduction of new bacteria is generally via the oropharynx, and alterations in the oropharyngeal microbiota during critical illness can promote the introduction of oropharyngeally acquired bacteria into the gut. ¹⁸ Critically ill patients often have a decrease in oral intake, causing decreased introduction of food-associated bacteria, an event of uncertain significance that may be influenced by the presence or absence of gastric acidity. ²⁰

Healthcare interventions can also have an impact on the influx of new bacteria into the microbiota. Topical oral decontamination has been shown to affect the contents of the gut microbiota by reducing the migration of oropharyngeal bacteria into the gastrointestinal tract, which can therefore cause an increase in the population of other, possibly pathogenic, bacteria such as Enterococcus faecalis. ²¹ Proton pump inhibitors have been shown to support bacterial overgrowth in the small intestine, further altering the gut microbiota in a maladaptive fashion. ²²

Elimination of bacteria from the human gut microbiota is also greatly affected by critical illness. In the healthy host, gut bacteria are primarily eliminated from the gastrointestinal tract via defecation. ¹⁹ In critical illness, intestinal motility is altered by factors such as hyperglycemia, electrolyte abnormalities, and therapeutic agents including opioids.^23,24 The dysmotility may favor gut colonization with pathogenic organisms, such as Pseudomonas, and subsequently increase the risk of nosocomial infection related to gastrointestinal tract pathogens. ²⁵

In addition to altered introduction of organisms and impaired elimination, the microenvironment of the gastrointestinal tract may be altered in critical illness, allowing certain bacteria to thrive compared with others. Inflammation and ischemia of the gut epithelium occurs because of inconsistent or inadequate perfusion in states of hypovolemia and shock, and are likely exacerbated when vasopressor support is required, especially prior to restoration of effective plasma volume. This inflammation can lead to increased nitrate concentrations, which promotes the growth of Escherichia coli. ²⁶ Inflammation of the intestinal mucosa can also increase the overgrowth of Enterobacterales and enterococci. ²⁷ Therefore, the gut microenvironment may play a modifiable role in the development of an unbalanced microbiome, typically denoted as a dysbiome. Clinical interventions used in the intensive care unit (ICU) can also have major effects on the human microbiota. In addition to enteral feeding, proton-pump inhibitors, and vasopressors, the administration of antibiotic agents has effects that may be profound and long-lasting. ¹⁹

Effects of Antibiotic Agents on the Microbiome

Much has been published on the effects of antimicrobial drugs on the human microbiome. Although the mechanisms by which antibiotic agents affect the microbiota are complex and are still not understood perfectly, many plausible explanations have been advanced. Antibiotic agents cause increased gut translocation of bacteria across the colonic epithelium, promoting inflammatory responses leading to dysbiosis and further altering of the microbiota through a feedback loop. ²⁸ This dysbiosis can also lead to the translocation of potential pathogens, including MDROs, to extraintestinal organs as well, including the lungs. ²⁹ Systemic antibiotic agents greatly modify, and generally reduce, the diversity of the bacteria present within the gut microbiota. Importantly, the specific changes in a given host depend greatly on the agents used because their impact is not uniform. The use of cephalosporins decreases the amount of Enterobacterales as a whole while increasing the specific abundance of Citrobacter, Klebsiella, and Pseudomonas species. ³⁰ Carbapenems may decrease the abundance of Enterobacterales as well as obligate anaerobes (such as Bacteroides) while selecting for less-susceptible opportunistic pathogens including Enterococcus. ³¹ The use of older (e.g., cefoxitin, cefpodoxime, and cefuroxime) and newer (e.g., cefepime, ceftaroline, and ceftazidime-avibactam) cephalosporins as well as carbapenems can promote the colonization of Clostridioides (previously Clostridium) difficile throughout the intestinal microbiota.^31–34 Currently, Clostridioides difficile infections occur in approximately 4% of ICU patients, with well-known consequences of prolonged hospitalization, increased resource utilization, morbidity, and mortality. ³⁵

Antibiotic agents decrease the bacterial diversity and also promote increasing levels of drug resistance within the microbiota. High rates of broad-spectrum antibiotic use in critical care settings will dramatically increase the prevalence of extended-spectrum β-lactamase producing Enterobacterales (ESBL-E).^36,37 Carbapenems are the typical drugs of choices to treat infections due to ESBL-E, but bacterial isolates with resistance to even these nearly last-line agents is an emerging threat to patients in the ICU. The risk of carbapenem-resistant pathogens increases after drug exposure and may have an ecological impact on not only the individual patient, but to the ICU, the hospital, long-term care facilities, and the community more generally. ³⁸ The use of oral glycopeptides (e.g., vancomycin) can similarly increase risk of the intestinal colonization with vancomycin-resistant Enterococcus faecium (VRE). ³⁹ Given the prevalence of MRSA in many ICUs, vancomycin is a commonly utilized empiric agent when infection is suspected in the critically ill or injured.

Although acute changes in the microbiota from antibiotic use are well-described, the duration for which these changes last after completing antimicrobial therapy will vary between hosts. One study demonstrated the return of normal microflora 35 days after the administration of ertapenem and ceftriaxone to healthy volunteers, ⁴⁰ whereas a similar study using meropenem displayed return of normal microflora only 14 days after the treatment. ³¹ However, these studies were performed on healthy individuals, so the return of the normal microbiota in a critically ill patient may be difficult to predict from these data. Regardless, the acute alteration of the gut microbiota by antibiotic agents has been shown to be substantial and clinically relevant. Even a 24-hour period of prophylactic antibiotic therapy for SSI prophylaxis can substantially alter the host microbiome, and post-prophylaxis Clostridioides difficile infection and disease is well documented. This effect is likely to be more pronounced in patients who receive non-guideline–conforming extended prophylaxis, a persistent practice by some surgical services.

Antimicrobial Stewardship in Critically Ill Patients

As the role of the human microbiota and its acute disruption by antibacterial drugs have been described increasingly, many efforts have been made to either use narrower spectrum antibiotic agents or to decrease the duration of therapy to help minimize the effects to the microbiota without losing treatment efficacy. Increasing evidence supports the use of shorter courses of therapy for a wide variety of serious infections, including ventilator-associated pneumonia (VAP), intra-abdominal infections, soft tissue infections, and bacteremia. ⁴¹ Similarly, there is a balance between appropriate empiric therapy, where broader spectrum regimens are both reasonable and the standard of care in the unstable patient with septic shock and organ failure, and specific therapy, where narrowing to pathogen-specific agents is preferred once culture results are available.

One example of major relevance to surgical practice is the Study to Optimize Peritoneal Infection Therapy (STOP-IT) trial, showing no difference in clinical outcomes when using a shorter course of antibiotic agents (mean, four days) versus a longer course (mean, eight days) in patients with complicated intra-abdominal infections after source control. ⁴² Source control to decrease the host bacterial bioburden is a key aspect of shortening treatment duration. Among critically ill patients with post-operative intra-abdominal infections, patients randomized to eight days versus 15 days of antibiotic therapy showed no differences in mortality rate or length of hospital stay. ⁴³ Prolonged courses of therapy for VAP similarly lack evidence for benefit, with shorter (eight days) versus longer (15 days) courses of antibiotic therapy having similar cure rates and no difference in mortality.^44–46

Although there is broad expert consensus that antimicrobial stewardship programs (ASPs) are both desirable and necessary, the direct impact of these strategies on the microbiome remain to be proven. Antimicrobial stewardship programs have clear associations with decreases in excessive antimicrobial usage and in the prevalence of AMR within institutions. ⁴⁷ As the gut microbiome is a known reservoir (and transport mechanism) for AMR-associated bacterial genes, ⁴⁸ ASPs likely exert a favorable impact on the microbiome, and should remain a cornerstone of antibacterial prescriptive practice and guidance.

Repair of the altered microbiota remains an active area of investigation. Direct reconstitution of the gut microbiome with probiotic preparations may be an appealing strategy, but data supporting the use of probiotics for the treatment and prevention of Clostridioides difficile infection have been inconsistent. Probiotics are not recommended in the 2021 American College of Gastroenterology (ACG) Clostridioides difficile management guidelines. ⁴⁹ Conversely, fecal microbiota transplantation (FMT) has increasing evidence supporting its benefit in the treatment of Clostridioides difficile and is recommended for refractory infection by both the ACG and Infectious Diseases Society of America (IDSA) guidelines.^49–52 Interestingly, FMT may also reverse fecal colonization with MDROs, restoring drug susceptibility to the patient microbiota. ⁵³

Future Considerations

In addition to the current strategies to overcome resistant pathogens and preserve the microbiota, newer therapies for infection may have favorable impacts on the microbiome. For example, omadacycline is a novel tetracycline derivative with in vitro activity against MRSA, VRE, and ESBL-E. Omadacycline appears to have a less deleterious effect on the microbiome than comparator agents and does not appear to provoke Clostridioides difficile proliferation or toxin production in vitro. ⁵⁴ It has received approval from the Food and Drug Administration (FDA) for the treatment of community-acquired bacterial pneumonia and acute bacterial skin and skin structure infections, but data regarding its utility in critical illness are limited. ⁵⁵

Rapid diagnostic assays may also optimize therapy by reducing the time to pathogen identification. Commercial multiplex syndromic panels utilize nucleic acid amplification testing (NAAT) help identify specific pathogens in sputum, cerebrospinal fluid, synovial fluid, and blood cultures more rapidly than traditional culture methods, and permit more-rapid antimicrobial de-escalation. ⁵⁶ These panels often also identify common resistance genes with high positive predictive values, although the absence of such resistance markers on NAAT is insufficient to exclude AMR (especially with gram-negative pathogens). As highly sensitive assays, they may have difficulty distinguishing between infection and colonization rendering interpretation reliant upon clinical judgement. Accordingly, infectious diseases consultation may be useful in ambiguous cases. ⁵⁷

Conclusions

The human microbiome plays a role in both health and disease. As clinicians, our ability to do harm to the microbiota is substantial, with uncertain short- and long-term impacts. Antimicrobial stewardship programs have the potential to minimize this impact, align antibiotic usage practices with rational strategies underpinned by data, and ultimately improve patient outcomes.