Critical appraisal 2

This critical appraisal is about the article written by E. van Nood et al. in 2013. This study is an RCT and was performed to determine the advantage of donor-feces infusion compared with vancomycin treatment, both with and without bowel lavage. Since this article states an important improvement in therapy for recurrent Clostridium difficile infection (CDI) and the fact that we used this article as backbone in this weblog article, we want to critically evaluate it and check whether it was performed rightly.

Relevance research question
The Research question of this article is: is donor-feces infusion compared with vancomycin treatment, both with and without bowel lavage, a more effective treatment for recurrent Clostridium Difficile infections. Feces donor infusion has already been described as effective in many patients. However, there is a lack of good randomized studies and therefore a gab in supporting data for this treatment. To fill in those gabs randomized studies like these need to be done. And thus is this a relevant research question.

Type of research question
This research questions falls into the category of the 'questions about the effectiveness of treatment'. In this case the effectiveness of the FMT compared to vancomycin therapy, with or without bowel lavage. This states that the study uses 1 intervention group: FMT with bowel lavage. And 2 control groups: vancomycin with bowel lavage and vancomycin without bowel lavage. The study population is mentioned: patients with recurrent CDI, but it is not mentioned how many relapses the patients have. The clinical outcome is not mentioned in the research question, but later in the article described as their primary outcome: cure without recurrent CDI within 10 weeks after the start of therapy.

Study design
The study was a not blinded randomized control trial in which 43 patients were randomly assigned to received one of three therapies: 17 in the intervention group and 13 in both control groups. Study participants were recruited in a period of 2 years form the Amsterdam medical center (AMC) and were randomly assigned by the study physicians. Patients needed to have a relapse of CDI and needed to be treated for at least one course or antibiotic therapy. The CDI was defined by diarrhea and a positive stool test for C dif. Toxin. Patients who were immunocompromised, administered to the ICU or using antibiotics for other infections than C. difficile were excluded from the study. All patients gave written informed consent and the study was approved by the ethics committee of the AMC. As mentioned before the study was not blinded, it may not be possible to make a placebo for the FMT, since this is an fluid infusion method and antibiotics need to be taken orally.

Bias
The randomization was performed by the study physicians, but in the article treatment allocation concealment was not mentioned. This could have influenced the study results, since it is not sure if the study physicians knew which patients could be addressed for which study group.

The characteristics of the tree different study groups were not significantly different as shown in table below.

In the flow chart, you can see that 1 patient was excluded from the study after randomization. This particular patient required high-dose prednisolone because of a fast decrease in renal-graft function. Following the exclusion criteria the patient needed to be excluded. This happened before the treatment was initiated, so this had no influence on the study results.

In the vancomycin without bowel lavage group one patient died before the first stool sample could be tested. In the intention-to-treat analysis the vancomycin therapy in this patients was considered a failure.

Original protocol?
The original idea was to recruit 40 patients per study group. This number was calculated by a power of 80% and one sided significance level of 0,025. They needed to include way less patients because most patients in the vancomycin control groups showed recurrent CDI. Therefor the data and safety monitoring board recommended closure of the trail. The method stayed the same during the trial.

Study hypothesis
The hypothesis stated in the article was a cure rate of 90% in the donor-feces infusion group and a cure rate of 60% in both control groups.

Statistical analysis
Differences in cure rates between the study groups were determined with Fisher's exact probability test. The statistical significance of a change in microbiota abundance was calculated with the use of a paired-samples Student t-test. Besides they used a Wilcoxon signed-rank test.

Conclusion of the article
The conclusion of the article was: 'The infusion of donor feces was significantly more effective for the treatment of recurrent C. difficile infection than the use of vancomycin.' This is in line with the results. In the FMT group 81% was cured after one infusion and 94% overall compared to a cure rate of 31% in the vancomycin without bowel lavage group and 23% in the vancomycin group. Donor-feces infusion was compared to both vancomycin groups statistically better. (P<0.01 after first infusion, P<0.001 for overall cure rates). See figure below.

Conflicts of interest
There are no conflicts of interest. This study was supported by The Netherlands Organization for Health Research and Development and performed in the Amsterdam Medical Center.

My conclusion
Because of the missing information about the treatment allocation concealment the differences between the intervention and control group can be bigger than in reality. This is an important fact that needs to be taken in account when reading this article. Moreover, the number of participants in this study is very small. But the recruitment of participants needed to be canceled because of ethical reasons. All this taken in account I still think this is a good study to see that the FMT indeed is a better therapy than vancomycin. 


                                                                                                                                           Written by Jodie Dekker
                                                                                                                                                        Posted on 14 oct 2018

Article of pictures and critical appraisal: 
[1] Kelly, C. LaMont J. Clostridium difficile - More difficult than ever. In: The New England Journal of Medicine. N Engl J Med 2008; 359:1932-1940