The Detrimental Effects of Oral Vancomycin

TO THE EDITOR—In the 1 September 2020 issue of Clinical Infectious Diseases, Johnson and colleagues [1] evaluated oral vancomycin prophylaxis in a randomized trial for prevention of Clostridioides difficile infection (CDI). We agree with and wish to elaborate on 2 of Professor Garey’s editorial comments that (1) one important concern for oral vancomycin as a prophylactic agent is its “profound effect on the microbiome that by itself decreases colonization resistance to C. difficile, vancomycin-resistant Enterococcus, and other multidrug-resistant organisms” and (2) that “CDI may develop at the same rate or higher in patients given oral vancomycin as primary prophylaxis but be simply delayed during the vancomycin administration time period” [2].

The degree of microbial-community disruption caused by oral vancomycin is substantial and should not be discounted. The potent anti-anaerobic activity of oral vancomycin parallels or exceeds that of many other conventional antibiotics in terms of magnitude and duration [3–5]. This helps to explain why CDI recurrences following treatment are so common, and why early attempts at treating asymptomatic carriage of C. difficile with oral vancomycin demonstrated a temporary suppression that was followed by paradoxically increased susceptibility to both colonization and true infection [6]. This is consistent with our clinical experience, in which we have observed CDIs immediately following the cessation of prophylactic courses of oral vancomycin.

Given this risk of this rebound effect, risks of oral vancomycin should be carefully considered before adopting a routine prophylaxis strategy, with an emphasis on the period after cessation of prophylaxis. Johnson and colleagues attempted to evaluate subjects after discharge, though it was unclear if this corresponded well to treatment cessation. Unfortunately, follow-up was only completed in less than half of the trial participants (48% vancomycin prophylaxis, 44% no prophylaxis); a true CDI difference of 15%, for example, would not have been detectable at this power.

The article by Johnson and colleagues did not report on the effects of vancomycin on fecal microbial composition, and we caution that this unexamined aspect of the study may belie potentially ominous consequences to the intestinal microbiome. Besides fecal microbiota transplantation (FMT) from a healthy donor, several other nonantibiotic approaches to CDI are in development, including banked autologous FMT [7], spore-enriched fractions of healthy-donor stool (ClinicalTrials.gov number, NCT03183128), rationally designed probiotics (ClinicalTrials.gov number, NCT03788434), a charcoal-based antibiotic adsorbent [8], and intraluminal enzymatic digestion of antibiotics [9]. We welcome clinical trials of any approach against CDI that aims to spare or improve the commensal community and thereby avoid the risk that oral vancomycin might throw out the baby with the bathwater.

Note

Potential conflicts of interest. Y. T. reports grants from the National Institutes of Health (NIH) (grant numbers U01 AI124275 and R01 AI137269), during the conduct of the study. J. U. P. reports a Cancer Center Core Facility Grant from the National Cancer Institute (P30 CA008748), Principal Investigator: Peled (National Heart, Lung, and Blood Institute K08HL143189), and a research grant from the Parker Institute for Cancer Immunotherapy at Memorial Sloan Kettering during the conduct of the study; consulting for Da Volterra outside the submitted work; meal at an education session from Jazz Pharmaceuticals; research support donation from Theodore and Laura Hromadka; a Memorial Sloan Kettering Cancer Center Cancer Systems Immunology Pilot Grant; and royalties, research funding, and travel reimbursement from Seres Therapeutics, outside the submitted work. Additionally, J. U. P. reports patents pending for methods and compositions for detecting risk of cancer relapse (licensed by Seres Therapeutics, 15/756845, filed 1 March 2018); bacterial compositions and methods for cancer survival (62/977908, filed 18 February 2020), and β-lactamase compositions for treatment of graft-vs-host disease (62/843849, filed 6 May 2019). Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.