The Role of Fecal Microbiota Transplantation in the Decolonization of Antibiotic-Resistant Bacteria
Christopher Saddler, MD and Nasia Safdar, MD, PhD
Antibiotic resistance is a serious worldwide threat, causing more than 2 million infections annually and 23,000 deaths in the United States alone.1 Furthermore, these infections lead to a significant economic impact, with a median increase in total hospital costs of $38,121 compared with those with antibiotic-susceptible infection.2 To date, several strategies have been employed to combat antibiotic resistance, primarily based in infection control and antimicrobial stewardship. For those already colonized, the risk of infection and associated morbidity and mortality is increased,3 yet no optimal strategy exists to address this clinical challenge. Decolonization protocols using antibiotics have had inconsistent outcomes and raise the risk of developing further resistance. Accordingly, this practice cannot be recommended for routine use.4–9 Given that several studies have demonstrated the efficacy of FMT for preventing recurrence of Clostridium difficile infection (CDI; previously Clostridium difficile infection) via restoration of the gut microbiome, it has been postulated that FMT may have a role in antibiotic-resistant bacteria (ARB) decolonization. This section highlights the data on decolonization of ARB by FMT.
The widespread early adoption of FMT for rCDI and the overall increased understanding of the microbiome has sparked significant interest in FMT for ARB decolonization. Initially, as FMT for rCDI became more widely used, case reports and case series began to emerge reporting on concomitant non–C. difficile ARB decolonization. Notably, in 2016 Millan et al10 reported a sustained reduction in antibiotic resistance genes following FMT for rCDI in those with symptom resolution, adding to the proof-of-concept data. Overall, these preliminary results were favorable; however, these studies were limited due to small sample size, study design limitations, and reporting bias. These results are summarized in Table 10.5-1.11–22
Patients with rCDI are frequently colonized with vancomycin-resistant Enterococcus (VRE), with prevalence nearing 20%.23 Accordingly, several studies were conducted examining the role of FMT of decolonizing VRE. Dubberke et al15 reported a 72.7% VRE decolonization rate in a subanalysis of 11 patients from a randomized clinical trial evaluating a microbiome-based therapy for rCDI. Additionally, Santiago et al22 reported a statistically significant difference between decolonization of VRE comparing allogeneic FMT (FMT from a healthy donor) with autologous FMT (FMT from one’s own stool; used as a placebo). After 6 weeks, 92% (11/12) of those initially colonized were decolonized in the group receiving allogenic FMT, whereas only 43% (3/7) of those initially colonized in the group receiving autologous FMT were decolonized.
Other clinical applications regarding ARB have been explored. Tariq et al19 reported decreased rates of recurrent urinary tract infections (UTIs) and improved antibiotic susceptibility profiles among those who underwent FMT for rCDI and had a history of recurrent UTIs. These data opens the door for other potential studies investigating the role FMT in reestablishing antibiotic susceptibility in locations outside the gut.
With these promising results in the co-colonized CDI population, FMT for the primary purpose of ARB decolonization has been increasingly pursued. Case reports and small case series among patients without concurrent CDI began to emerge, reporting 100% efficacy for decolonization of extended-spectrum beta-lactamase (ESBL)–producing organisms and carbapenem-resistant Enterobacteriaceae (CRE).24–29 Subsequently, larger case series and pilot prospective, uncontrolled studies followed with mixed, but generally promising, results suggesting 37.5% to 87.5% efficacy for ARB decolonization without any safety concerns.30–36 However, only 1 randomized clinical trial (N = 39) by Huttner et al37 has been reported to date, which showed more modest potential benefit (41% vs 29% decolonization rate of extended-spectrum beta-lactamase Enterobacteriaceae and CRE) not meeting statistical significance. This study was unfortunately hampered by not meeting goal sample size and thus lacked statistical power. It also used 5 days of oral antibiotics for intestinal decolonization. Table 10.5-2 summarizes the current data of non-CDI trials.24–38 Fortunately, numerous studies as reported on clinicaltrials.gov are in progress, which may add to our knowledge base.
Abbreviations: CR = case report; CRE = carbapenem-resistant Enterobacteriaceae; CS = case series; GNR = gram-negative rod; MCT = multicenter trial; MDR = multidrug-resistant; MRSA = methicillin-resistant Staphylococcus aureus; NR = not reported; PA = Pseudomonas aeruginosa.
aDecline in enterococcal relative fecal abundance from 24% to 0.2%; no recurrence of infection.
bClinical resolution; no recurrent UTI after 25-year history with negative urine cultures in 25 months intervening.
The Huttner et al37 study perfectly demonstrates some of the greater issues in the study of FMT for ARB decolonization. First, the definition of colonization has been widely variable and requires uniformity. Second, the methodology of assessing decolonization must also be determined, with clarification regarding whether use of rectal swabs or stool in selective culture, polymerase chain reaction arrays for antibiotic resistance mechanisms, or more elaborate testing such as stool for metagenomics evaluation of antibiotic resistance genes is most appropriate. Third, the methodology of studies should be standardized, including whether antibiotics are given for intestinal decolonization; whether bowel lavage is performed; and what dose, what frequency, and via what route FMT should be delivered. These require more systematic study moving forward. Further study is also needed to establish factors that predict success or failure of ARB decolonization. Importantly, future studies must be placebo controlled to assess efficacy given that spontaneous decolonization of ARBs does occur, at highly variable rates. In addition, studies must focus on individual organisms as opposed to an aggregation of multiple ARBs, given that there may be variability in effect size. Finally, although FMT appears safe, the long-term safety and efficacy will require comprehensive continued monitoring.
Overall, FMT for ARB decolonization is an intervention that is in its infancy and requires significant further study to identify the optimal target population. FMT should remain an investigational agent at this point for ARB decolonization until established interventional strategies and randomized controlled trials have determined the benefit.
Abbreviations: ACTB = Acinetobacter; AMG = aminoglycoside; CRPA = carbapenem resistant Pseudomonas aeruginosa; CR-ACTB = carbapenem resistant Acinetobacter baumannii; MIU = million international units; SMALT = Stenotrophomonas maltophilia.
a41% vs 29% decolonization with placebo, non-statistically significant