Randomised Clinical Trial

A 12-Strain Bacterial Mixture Versus Faecal Microbiota Transplantation Versus Vancomycin for Recurrent Clostridioides difficile Infections

Anne Abildtrup Rode; Mahtab Chehri; Laura Rindom Krogsgaard; Kristine Klysner Heno; Anna Tølbøll Svendsen; Iben Ribberholt; Morten Helms; Jørgen Engberg; Kristian Schønning; Michael Tvede; Christian Østergaard Andersen; Ulrich Stab Jensen; Andreas Munk Petersen; Peter Bytzer


Aliment Pharmacol Ther. 2021;53(9):999-1009. 

In This Article


We performed the largest randomised, controlled trial to date on the treatment of recurrence of C difficile infection comparing FMT with standard-of-care treatment with oral vancomycin and additionally with a well-characterised 12-strain bacterial mixture, rectal bacteriotherapy. The study focussed on minimising selection bias by recruiting participants from primary to tertiary care, involved tapering of vancomycin for the control group according to guidelines and we applied treatments (rectal bacteriotherapy and FMT) by enema to minimise invasiveness and costs.

We confirmed that FMT has a higher efficacy in achieving clinical cure of recurrent C difficile infection than oral vancomycin—also when administered by enema, yet two or three infusions were sometimes needed. Notably, we saw a more convincing efficacy of FMT for participants with a first recurrence of C difficile infection as opposed to multiple recurrences.

A cure rate of 76% after FMT (1-3 infusions) is in keeping with previous findings from randomised controlled trials reporting cure rates of 81%[7] and 92%[29] for single naso-intestinal or colonoscopic infusions and 90% with 1–4 colonoscopic infusions.[8] Nevertheless, in our trial only 56% were cured after a single FMT by enema. The limited effect after a single infusion is, however, in line with some other trials.[8,12]

There are several possible reasons for the discordance between the efficacies of FMT. Firstly, the delivery method is probably a key factor, since single FMT by colonoscopy or naso-intestinal tube have yielded higher efficacy rates in former trials.[7,29] Two meta-analyses showed that administrating FMT by colonoscopy was superior to enemas.[30,31] Our results and Hota et al[12] support this conclusion, yet as in our study, the meta-analyses also found an increased effect after multiple infusions by enema, yielding an efficacy rate similar to FMT by colonoscopy. Nevertheless, the compared studies are very heterogenic and the delivery methods have not been compared head-to-head.[30–32] Furthermore, we do not know if bowel preparation as used before colonoscopy and in some cases also prior to naso-intestinal administration can improve the effect of FMT by enema.

Rectal bacteriotherapy with the 12-strain bacterial mixture was not as effective as expected based on previous results from case series with reported cure rates from 64% to 88%.[20,21] The current study is the first randomised controlled trial investigating the effect of rectal bacteriotherapy and of well-characterised bacterial mixtures in general. Here rectal bacteriotherapy seemed similar in efficacy to oral vancomycin and not equally effective to FMT, when used according to our protocol.

Petrof et al introduced a mix of 33 bacteria species isolated from a healthy donor and described the successful treatment of two patients with multiple recurrences of C difficile infection.[33] Cammarota et al introduced a 15-strain consortium (bacteria derived from successful engraftment in patients cured by FMT) and treated 10 patients in a single-group pilot study curing all after one to two infusions.[34] Thus, both these bacterial mixtures have so far only been investigated in small single-group studies.

In the FMT group, the protocol allowed for repeating treatment within the first 14 days of follow-up, if necessary. The results clearly show the benefit of multiple infusions of donor stool (with a different donor). We did not include a possibility to repeat rectal bacteriotherapy beyond the three planned consecutive infusions. Since nine of the 15 participants failing this treatment had their recurrence of C difficile infection within the first 14 days, repetition similar to what was used for FMT might have been favourable. Further research into dosing and timing of rectal bacteriotherapy is necessary.

The safety profiles of FMT and rectal bacteriotherapy were alike. Yet, in the case of a potential iatrogenic infection, rectal bacteriotherapy has the benefit that we know the exact content and that all strains are sensitive to commonly used antibiotics. Furthermore, it has the potential of large-scale production with a larger capacity than donor stool, even when donor stool banks are established. Nevertheless, potential long-term effects of rectal bacteriotherapy are unknown. Yet, these are also largely unknown for FMT.[35]

It is possible that a bacterial mixture is an oversimplification of the treatment or that bacterial species with other metabolic abilities (or other essential compounds of stool) than the ones included in rectal bacteriotherapy are necessary to regain colonisation resistance to C difficile. Proposed mechanisms behind the effect of FMT include increased resistance with a diverse microbiota or competition of nutrients, re-establishment of a normal bile acid or short-chain fatty acid composition or possibly an introduction of microbiota-based bacteriocins.[36] Knowledge of mechanisms of action of FMT and possible active components should be included in designing or optimising bacterial mixtures for treatment of recurrent C difficile infection.

For now, we do not have data on possible microbiota changes after rectal bacteriotherapy, yet this knowledge could be crucial in further developing the treatment.

A major strength of our study is the fact that we—in accordance with available European and American guidelines when initiating the trial[10,11]—treated participants with ≥2 recurrences with a long tapered course of vancomycin as the standard-of-care treatment. This is in contrast to former studies comparing FMT with vancomycin, where multiple recurrences were treated with vancomycin for only 10–14 days[7,29] or for 10 days followed by a pulse regimen for 3 weeks.[8] Thus, these studies could have underestimated the effect of oral vancomycin. Yet, our study mostly included participants with a first or second recurrence as opposed to these former trials, which might partly explain our reported relatively high cure rate in the vancomycin group. Nevertheless, the use of the long tapering regimen could also explain this, concurring with Hota et al.[12]

Importantly, our results indicated that FMT might reduce the all-cause mortality at 6 months compared with standard-of-care treatment with oral vancomycin. These findings should, however, be interpreted with caution because of the small number of deaths (n = 13). Albeit, it corresponds to a number needed to treat of only six persons to prevent one death. To our knowledge, this is the first randomised controlled trial to suggest an effect of FMT on mortality for individuals with recurrent C difficile infection, yet a recent propensity-matched cohort study had similar findings.[37] This may be explained by our larger population size than former trials or by the inclusion of a different study population, for example including a high percentage of individuals with a first recurrence, which have been shown to have a higher mortality than individuals with multiple recurrences, probably due to survival bias.[38,39] The finding of a potential reduction in all-cause mortality after FMT might serve as an argument for using this treatment earlier on, for example already at the first recurrence of C difficile infection.

There are several limitations to our study. The study was conducted as an open-label trial since it was not practically possible to blind the large differences between the interventions. Furthermore, we only managed to include 98 participants. Although we present the largest randomised controlled trial to date on this subject, the rather low number of participants hampers conclusions, especially in subgroup analyses. Furthermore, the regimens for rectal bacteriotherapy and FMT differed somewhat, including number and timing of infusions and in the vancomycin-free interval before infusion (ie 12 vs 36 hours). The treatment regimens were based on existing litterature[7,8,20] and clinical experience with the two treatments when designing the study.

In addition, the majority of participants in our study were individuals with one to two recurrences of C difficile infection, making these results less applicable to patients with multiple recurrences. We screened for participants widely by identifying them from positive C difficile stool tests and used an active outreaching recruitment method, thus identifying the majority of the participants early on. Lastly, only six participants were infected with CD027. There was some indication that the rate of failure and mortality was higher for these participants, but due to the low number, we cannot explore this further.

In conclusion, rectal bacteriotherapy with three infusions of a 12-strain bacterial mixture seems equal in effect to vancomycin for treating recurrent C difficile infection and a potential role for this treatment should be further investigated. However, FMT applied by enema one to three times is superior in efficacy compared with both rectal bacteriotherapy and with oral vancomycin for these patients. However, enema application sometimes calls for two infusions to yield the warranted effect. Importantly, FMT might reduce the mortality in this patient group and seems to be most beneficial at the first recurrences. These findings suggest that faecal microbiota transplantation should be a first-line treatment for individuals with recurrent C difficile infection.