If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
We conducted a systematic review of the literature and network meta-analysis (NMA) to compare the relative effectiveness of antibiotic treatments for Clostridioides (Clostridium) difficile infection (CDI) including vancomycin (VCM), metronidazole (MTZ) and fidaxomicin (FDX). Eligible studies were randomised controlled trials (RCTs) including adults with any severity of CDI that was treated with VCM, MTZ or FDX. The NMA was performed using a Bayesian framework, using a fixed-effects model.
The searches identified seven publications for inclusion, which provided five RCTs for VCM versus MTZ, and three RCTs for FDX versus VCM. The NMA showed that for clinical cure rate, there was no difference for FDX versus VCM, and there was a significant difference in favour of FDX versus MTZ (odds ratio [OR]: 1.77; 95% credible interval [CrI] 1.11, 2.83]). For recurrence rate, there was a significant difference in favour of FDX versus both VCM (OR: 0.50; 95% CrI: 0.37, 0.68) and MTZ (OR: 0.44; 95% CrI: 0.27, 0.72). For sustained cure (clinical cure without recurrence), there was a significant difference in favour of FDX versus VCM (OR: 1.61; 95% CrI: 1.27, 2.05) and MTZ (OR: 2.39; 95% CrI: 1.65, 3.47).
These findings suggest that FDX and VCM are effective first-line treatments for mild or moderate CDI, whereas MTZ is not, and FDX may be more effective at preventing CDI recurrence than VCM.
Clostridioides (Clostridium) difficile infection (CDI) is the leading cause of nosocomial bacterial diarrhoea in industrialised countries, and can be serious and life-threatening; its incidence and severity are rising, which impacts all aspects of healthcare delivery globally [
Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).
Treatment failure and recurrence of Clostridium difficile infection following treatment with vancomycin or metronidazole: a systematic review of the evidence.
The latest guidelines developed by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA), updated in 2017, recommend either VCM or fidaxomicin (FDX) for the treatment of an initial episode of mild or moderate CDI in adults [
Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).
]. A notable change from the previous IDSA/SHEA guidelines was that MTZ in mild or moderate CDI was downgraded from a first-line treatment to an alternative treatment option recommended when FDX or VCM is not available [
Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).
]. The guidance for VCM was based on evidence from head-to-head studies that reported significantly higher cure rates, as well as lower recurrence rates, for VCM versus MTZ [
]. The guidance for FDX was based on two head-to-head studies of FDX and VCM, which demonstrated the non-inferiority of FDX versus VCM for clinical cure, and showed that the recurrence rate was significantly lower for FDX than for VCM [
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
FDX is a narrow-spectrum macrocyclic antibiotic that was approved globally based on the results of two randomised, controlled trials (RCTs) of FDX versus VCM in patients with CDI in the USA, Canada and Europe [
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
]. More recently, a randomised study of nosocomial CDI in Japan showed that although non-inferiority was not demonstrated for FDX versus VCM, the 28-day recurrence rate was 19.5% and 25.3% for FDX and VCM, respectively [
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
]. To date, there are no comparative studies of FDX versus MTZ in CDI, although in a network meta-analysis (NMA) including data available up to July 2017, FDX was significantly better than MTZ for sustained symptomatic cure (odds ratio [OR]: 0.49; 95% CI: 0.35, 0.68) [
In this analysis, we conducted a systematic literature review to identify RCTs of FDX versus VCM and VCM versus MTZ; we then performed an NMA to assess the relative effectiveness of FDX versus MTZ in the treatment of CDI, using VCM as the common arm.
2. Patients and methods
2.1 Design and objectives
A systematic evidence-based approach was used to screen the literature and extract RCTs of FDX, VCM or MTZ for the treatment of CDI. An NMA was then conducted to assess the relative effectiveness of FDX versus VCM, and FDX versus MTZ using VCM as the common CDI treatment arm. The effectiveness outcomes of interest, calculated for each treatment, were clinical cure rates during the treatment period, recurrence rates during follow up, and sustained cure rate defined as clinical cure without recurrence during follow up.
2.2 Searches and data extraction
A systematic literature search performed on 3 July 2011 identified three articles that were eligible for inclusion; an updated literature search was performed to identify articles published up to 10 March 2014. A further literature search was conducted in September 2016. Search strings were based on ‘Clostridioides (Clostridium) difficile’, ‘CDI’ or ‘CDAD’, and ‘fidaxomicin’, ‘vancomycin’, ‘metronidazole’ or ‘probiotics’. The initial inclusion of probiotics was intended to broaden the subsequent network; however, no such articles were found fulfilling the search criteria. Research articles published in English were identified using the Ovid platform to search Medline, EMBASE and the Cochrane Central Register of Controlled Trials, and articles were identified using the Igaku Chuo Zasshi database, which includes more than 2500 biomedical journals and other serial publications published in Japan.
The Japanese Association for Infectious Diseases, the Japanese Society of Environmental Infections and the Japanese Association for Anaerobic Infection Research congress databases were also searched to obtain studies published only as abstracts (published in 2015 and 2016). The Clinicaltrials.gov database was used to identify closed clinical studies in adults from 2013 to 2016, using the search terms ‘Clostridioides (Clostridium) difficile’, ‘CDI’ or ‘CDAD’ with the interventions ‘fidaxomicin’, ‘vancomycin’, ‘metronidazole’ or ‘probiotics’.
Articles were eligible for inclusion if they reported RCTs of FDX, VCM or MTZ in adult patients with confirmed CDI of any severity. Articles were excluded if they were studies in patients aged <18 years or receiving treatment other than FDX, VCM, MTZ, or probiotics, or non-randomised studies including reviews, opinions, systematic reviews and editorials.
Articles were screened by title and abstract to identify those containing relevant data and the full text of the selected articles was fully assessed to ensure compliance with the inclusion and exclusion criteria. Data extracted into Excel included publication type, title, year of publication, trial name, trial number, sponsor/funding, country, study objective, study design, population, inclusion/exclusion criteria, sample size, details of intervention/comparator, outcomes of interest (clinical cure, recurrence rate and sustained cure rate), follow-up time and statistical methods (Table 1). Screening was performed by two independent reviewers, and a quality-control check of the data extraction was performed by a further independent reviewer.
Table 1Summary of randomised controlled trials included in the network meta-analysis.
Not included in the NMA. CDI, Clostridioides difficile infection; FA, fusidic acid; FDX, fidaxomicin; IM, intramuscularly; IV, intravenously; MTZ, metronidazole; TEC, teicoplanin; TLV, tolevamer; VCM, vancomycin. Follow-up periods are quoted from the end of treatment.
, 500 mg orally, three times daily for 10 days
Clinical cure, recurrence rate (follow up ≥30 days)
Not included in the NMA. CDI, Clostridioides difficile infection; FA, fusidic acid; FDX, fidaxomicin; IM, intramuscularly; IV, intravenously; MTZ, metronidazole; TEC, teicoplanin; TLV, tolevamer; VCM, vancomycin. Follow-up periods are quoted from the end of treatment.
Not included in the NMA. CDI, Clostridioides difficile infection; FA, fusidic acid; FDX, fidaxomicin; IM, intramuscularly; IV, intravenously; MTZ, metronidazole; TEC, teicoplanin; TLV, tolevamer; VCM, vancomycin. Follow-up periods are quoted from the end of treatment.
, 9 g + 3 g orally, three times daily for 14 days
Clinical cure, recurrence rate (follow up 28 days)
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
Clinical cure, recurrence rate (follow up 28 days)
III
Japan
74.0–75.0
102/110
VCM, 125 mg orally, four times daily for 10 days
a Not included in the NMA. CDI, Clostridioides difficile infection; FA, fusidic acid; FDX, fidaxomicin; IM, intramuscularly; IV, intravenously; MTZ, metronidazole; TEC, teicoplanin; TLV, tolevamer; VCM, vancomycin. Follow-up periods are quoted from the end of treatment.
All data used in the analysis were from the intention-to-treat (ITT) study populations. The endpoints of interest were: clinical cure rates during the treatment period; recurrence rates during follow up; and sustained cure rate during follow up. Definitions and endpoints from the published article were used when available, and were otherwise calculated as given in Table 2. Some studies reported recurrence as a proportion of patients who achieved initial clinical cure, while others reported recurrence as a proportion of patients who previously met the criteria for resolution of diarrhoea. Therefore, two definitions of sustained cure were calculated for the studies included in this paper: ‘sustained cure 1’ was defined as initial clinical cure with no recurrence and ‘sustained cure 2’ was defined as resolution of diarrhoea with no recurrence (Table 2).
Table 2Summary of the datasets included and definitions (actual and calculated) used in the NMA.
Not explicitly reported; deduced for this analysis using the reported proportion of patients with relapse.
–
51/56
39/45
Recurrence (‘relapse’)
Recurrence of diarrhoea (≥4 watery stools per day for ≥48 h) in patients who had completed 10 days of treatment, had normal stools at the end of the treatment period and had not received additional antibiotics after the treatment period, plus a positive C. difficile stool culture/cytotoxin assay or visible pseudomembranes at endoscopy
–
6/51
2/39
Sustained cure 1 (‘cure’)
Resolution of diarrhoea within 6 treatment days, completion of 10 days of treatment and no relapse within the follow-up period
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
Resolution of diarrhoea (attainment of bowel movements with a hard/formed consistency on average, or ≤2 loose/watery bowel movements per day, on average) and absence of severe abdominal discomfort due to CDI, for >2 consecutive days including day 10
–
109/134
103/143
Recurrence
Confirmed diagnosis of CDI during the 4-week follow-up period in patients who previously had resolution of diarrhoea
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with resolution of diarrhoea.
Resolution of diarrhoea with no recurrence [calculated]
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with resolution of diarrhoea.
Resolution of diarrhoea (attainment of bowel movements with a hard/formed consistency on average, or ≤2 loose/watery bowel movements per day, on average) and absence of severe abdominal discomfort due to CDI, for >2 consecutive days including day 10
–
101/125
99/135
Recurrence
Confirmed diagnosis of CDI during the 4-week follow-up period in patients who previously had resolution of diarrhoea
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with resolution of diarrhoea.
Resolution of diarrhoea with no recurrence [calculated]
Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with resolution of diarrhoea.
Resolution of diarrhoea (≤3 UBMs for 2 consecutive days), maintenance of resolution until EOT and no requirement for further therapy at 2 days after EOT
253/287
265/309
–
Recurrence (‘clinical recurrence’)
Reappearance of >3 diarrhoeal stools per 24-h period within 4 weeks of end of treatment; presence of C. difficile toxin A/B in stool; and a requirement for further therapy
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
Resolution of diarrhoea (≤3 UBMs per day for 2 consecutive days, or a substantial number of UBMs at EOT) for the duration of treatment and no requirement for further therapy at 2 days after EOT
221/252
223/257
–
Recurrence
Reappearance of >3 UBMs per 24-h period within 30 days of EOT; positive C. difficile toxin stool test; and a requirement for further therapy
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
Resolution of diarrhoea (≤3 UBMs for 2 consecutive days), maintenance of resolution until EOT, and no requirement for further therapy treatment at EOT
87/104
95/108
–
Recurrence
More frequent UBMs then at EOT per 24-h period within 4 weeks of EOT, presence of C. difficile toxin A/B in stool; and a requirement for further therapy
17/87
24/95
–
Sustained cure 1 (‘global cure’)
Clinical cure at EOT with no recurrence during follow up
70/104
71/108
–
Sustained cure 2
[As for SCC 1]
70/104
71/108
–
CDI, Clostridioides difficile infection; EOT, end of treatment; FDX, fidaxomicin; MTZ, metronidazole; SCC, sustained clinical cure; UBM, unformed bowel movement; VCM, vancomycin. Terminology used in the published studies are given in parentheses in the ‘Outcome’ column, where it differs from terminology used in this report.
a Not explicitly reported; deduced for this analysis using the reported proportion of patients with relapse.
b Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with clinical cure.
c Not explicitly reported; calculated for this analysis by subtracting the number of patients with recurrence from the number of patients with resolution of diarrhoea.
d Johnson 2014 reports pooled data from Study 301 and 302 for the subgroups; for the total populations, numbers are provided for each study separately.
The NMA was performed using a Bayesian framework, using a fixed-effects model and a random-effects model. The fixed-effects and random-effects models were compared using the Deviance Information Criterion (DIC). Differences in DIC of <5 points were not considered meaningful; consequently, the simpler fixed-effects model was selected for the analysis because the difference in DIC between fixed-effects and random-effects models was negligible (see Results). A generalised linear model with a binomial error distribution and logit link function was used, and results were summarised as odds ratios (ORs) with 95% credible intervals (CrI). Forest plots presented the ORs for FDX versus VCM or MTZ: for clinical cure rate and sustained cure rate, an OR>1 favours FDX; for recurrence rate, an OR<1 favours FDX.
Between-study heterogeneity was assessed using pairwise comparisons, with the inverse variance-weighted method for the fixed-effects model and either the restricted maximum likelihood or the DerSimonian-Laird method for the random-effects model. Odds ratios and 95% confidence intervals (CIs) were used to summarise the individual and pooled treatment effects (Figure A.1). An I2 score of <25 was considered to indicate low heterogeneity.
3. Results
3.1 Searches
In the initial literature search, 1261 publications were identified, of which 44 were selected for full-text evaluation and five publications (reporting data from five studies) met the inclusion criteria. In the updated literature search, 610 articles were identified, of which 118 were duplicates. After screening titles and abstracts of 492 articles, 30 were examined in detail; of these, two publications (reporting from three datasets) fulfilled the inclusion criteria (Fig. 1). From these seven publications, there were eight datasets, including five datasets for VCM versus MTZ and three datasets for FDX versus VCM (i.e. one article included two RCTs) (Fig. 2).
Fig. 1Flow chart of article selection. FDX, fidaxomicin; ICHUSHI, Igaku Chuo Zasshi database; MTZ, metronidazole; VCM, vancomycin.
Three randomised, controlled, double-blind phase III studies of FDX versus VCM were identified and met the inclusion criteria for the NMA, each comparing 10 days’ treatment of FDX and VCM in patients with CDI. Two studies were non-inferiority studies conducted in the USA, Canada and Europe, and the third study was a comparative study conducted in Japan (Table 1). The outcomes of these studies demonstrated comparable cure rates for the FDX and VCM treatment arms; however, in the FDX arm, recurrence rates were lower resulting in increased rates of sustained clinical cure compared with VCM [
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
In the studies conducted in the USA, Canada, and Europe, the primary endpoint was the non-inferiority of FDX versus VCM for clinical cure. In both RCTs, the primary outcome was clinical cure defined as the resolution of diarrhoea (≤3 unformed bowel movements [UBMs] for 2 consecutive days), maintenance of resolution for the duration of therapy, and no further requirement (in the investigator's opinion) for CDI therapy at 2 days after end of treatment (EOT) [
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
]. In one study, among 596 patients in the modified ITT population, FDX was non-inferior to VCM for initial clinical cure of CDI (88.2% versus 85.8%, respectively), and the 28-day recurrence rates were significantly lower for FDX vs VCM (15.4% versus 25.3%, respectively) [
]. In the other study, among 509 patients in the modified ITT population, FDX was non-inferior to VCM for clinical cure (91.7% versus 90.6%, respectively), and the 28-day recurrence rates were also significantly lower for FDX compared with VCM (12.7% versus 26.9%, respectively) [
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
A further randomised, double-blind, parallel-group study was conducted to assess the efficacy and safety of FDX versus VCM for the treatment of CDI in adult hospitalised patients in Japan [
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
]. The primary endpoint was the non-inferiority of FDX versus VCM for global cure rate of CDI, defined as the proportion of patients cured at end of treatment with no recurrence during 28-day follow up. Among 212 patients, the global cure rate was 67.3% for FDX and 65.7% for VCM: non-inferiority was not demonstrated. The 28-day recurrence rate was 19.5% for FDX compared with 25.3% for VCM [
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
Five RCTs of VCM versus MTZ were identified for inclusion in the NMA (Table 1). The earliest prospective, randomised study, conducted in the early 1980s, found that among 101 patients with CDI who received VCM or MTZ for 10 days, the clinical cure rate was about 35% in both treatment groups [
]. In a further trial conducted in the 1990s, 62 patients were randomised 1:1 to receive VCM or MTZ, and the clinical cure rate was 94% in both treatment groups [
Larger, more recent randomised trials report significant improvements in CDI clinical outcomes for VCM over MTZ. In a pooled analysis from two trials, reported by Johnson et al., the primary endpoint was the non-inferiority of VCM versus MTZ for clinical success, defined as the resolution of diarrhoea and absence of severe abdominal discomfort due to CDI for more than 2 consecutive days including day 10; resolution of diarrhoea was defined as attainment of bowel movements with a hard or formed consistency on average or 2 or fewer bowel movements/day with a loose or watery consistency on average [
]. Among 289 patients receiving MTZ and 266 patients receiving VCM, the clinical success rate for MTZ was inferior to VCM (72.7% and 81.1%, respectively).
The RCT reported by Zar et al. compared VCM with MTZ, using the primary outcome of clinical cure defined as the resolution of diarrhoea by day 6 of treatment and a negative result of a C. difficile toxin A assay at days 6 and 10 of treatment [
]. Of 172 patients, 150 completed the study; while clinical cure rates with MTZ and VCM were similar for mild CDI (90% versus 98%, respectively), in patients with severe CDI, the clinical cure rates were significantly lower with MTZ than with VCM (76% versus 97%, respectively). None of the studies provided rates of sustained clinical cure, so these were calculated from clinical cure and recurrence/relapse rates as shown in Table 2.
3.4 Network meta-analysis (NMA)
The NMA included three FDX datasets and five MTZ datasets, each compared with VCM (Fig. 2). The total population and outcomes are shown in Table 2. Clinical cure rates were reported using the same methods of evaluation in all eligible studies, although the definition of clinical cure and recurrence differed across studies. For the purpose of evidence synthesis, clinical cure and recurrence definitions were considered similar and data were used as reported in the sources.
Heterogeneity of 23.10% between studies was observed for the FDX versus VCM comparison of sustained clinical cure (definitions 1 and 2) (Figures A.3 and A.4). For all other outcomes and for MTZ versus VCM, heterogeneity was 0–0.55% (Figures A.1–A.4). For all the endpoints of interest, there was a negligible difference in DIC between the fixed-effects and random-effects models (Table A.1), so the fixed-effects model was used.
3.5 Fixed-effects model
There was no difference between FDX and VCM in clinical cure rate, and a significant difference in favour of FDX for clinical cure rate compared with MTZ (OR: 1.77 [95% CrI: 1.11, 2.83]) (Fig. 3). For recurrence (defined as clinical cure without recurrence or diarrhoea resolution without recurrence), there was a significant difference in favour of FDX versus both VCM (OR: 0.50 [95% CrI: 0.37, 0.68]) and MTZ (OR: 0.44 [95% CrI: 0.27, 0.72]) (Fig. 4).
Fig. 3Clinical cure for fidaxomicin versus metronidazole and vancomycin. FDX, fidaxomicin; MTZ, metronidazole; OR, odds ratio; VCM, vancomycin. Data are shown as median log-ORs with 95% credible intervals.
Fig. 4Recurrence for fidaxomicin versus metronidazole and vancomycin. FDX, fidaxomicin; MTZ, metronidazole; OR, odds ratio; VCM, vancomycin. Data are shown as median log-ORs with 95% credible intervals.
There were significant differences in favour of FDX for sustained cure 1 (clinical cure without recurrence) versus VCM (OR: 1.61 [95% CrI: 1.27, 2.05]) and MTZ (OR: 2.39 [95% CrI: 1.65, 3.47]) (Fig. 5). There were also significant differences in favour of FDX for sustained cure 2 (diarrhoea resolution without recurrence) versus VCM (OR: 1.61 [95% CrI: 1.27, 2.05]) and MTZ (OR 2.12 [95% CrI: 1.46, 3.08]) (Fig. 6).
Fig. 5Sustained cure 1 (clinical cure without recurrence) for fidaxomicin versus metronidazole and vancomycin. FDX, fidaxomicin; MTZ, metronidazole; OR, odds ratio; VCM, vancomycin. Data are shown as median log-ORs with 95% credible intervals.
Fig. 6Sustained cure 2 (diarrhoea resolution without recurrence) for fidaxomicin versus metronidazole and vancomycin. FDX, fidaxomicin; MTZ, metronidazole; OR, odds ratio; VCM, vancomycin. Data are shown as median log-ORs with 95% credible intervals.
VCM and MTZ have been the mainstays of CDI treatment for decades, with MTZ used in mild or moderate episodes and VCM in severe CDI. MTZ is recommended as a first-line treatment for mild or moderate CDI in Europe [
Australasian Society of Infectious Diseases updated guidelines for the management of Clostridium difficile infection in adults and children in Australia and New Zealand.
Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).
]. In this NMA comparing the effectiveness of FDX, VCM and MTZ for the treatment of CDI, FDX was more effective than MTZ but comparable to VCM for clinical cure rate, while FDX was more effective than both VCM and MTZ with regard to CDI recurrence and sustained cure rates.
The main outcome in our analysis was clinical cure rate, which in the fixed-effects model was significantly higher for FDX versus MTZ, and similar for FDX versus VCM. The result of the comparison with VCM is consistent with a previous meta-analysis that did not include the most recent randomised trial in Japan, and which reported that clinical cure rates were similar for FDX and VCM (OR: 1.17; 95% CI: 0.82, 1.66) [
Clinical efficacy of fidaxomicin compared with vancomycin and metronidazole in Clostridium difficile infections: a meta-analysis and indirect treatment comparison.
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
]. Our analysis calculated recurrence rate to be significantly lower, and sustained cure rates to be correspondingly higher, with FDX versus MTZ. With the exception of the Japanese phase III trial [
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
], FDX was also associated with significant improvements in recurrence rates and sustained cure rates compared with VCM. Our findings are consistent with a recent NMA based on published data available up to 2017 that compared 13 treatments for CDI, including FDX, MTZ and VCM [
]. The estimates reported in the study suggest that for sustained symptomatic cure, defined as the number of patients with resolution of diarrhoea minus the number with recurrence or death, FDX was better than MTZ (OR: 0.49 [95% CI: 0.35, 0.68]), and better than VCM (OR: 0.67 [95% CI: 0.55, 0.82]) [
]. The authors of the study concluded that MTZ should not be recommended for treatment of CDI, and that FDX should be considered for initial therapy for mild or moderate CDI [
The approach in our analysis allowed the effectiveness of FDX and MTZ to be compared across clinical trials using VCM as the common arm. Our study is primarily limited by variability between the original studies in multiple aspects, including treatment dosage and duration; the definitions of clinical cure, recurrence rate and sustained cure; and the duration of follow up. In particular, sustained cure rates were provided in the FDX versus VCM studies, but were calculated from using clinical cure and recurrence rates for VCM versus MTZ. Lastly, the therapeutic environment or intestinal flora in the studies conducted in the 1980s and 90s may have differed from that in contemporary studies.
In summary, our meta-analysis suggests that FDX and VCM, but not MTZ, are effective first-line treatments for CDI, and that FDX may be more effective at preventing CDI recurrence than VCM.
Conflicts of interest
A. Fukushima was an employee of Creativ-Ceutical at the time of the study, and V. Taieb is an employee of Creativ-Ceutical, working on behalf of Astellas Pharma, Inc. H. Okumura and S. Shoji are employees of Astellas Pharma, Inc. and M. English is an employee of Astellas Pharma Global Development, Inc.
Data sharing statement
Access to anonymised individual participant level data will not be provided for this trial as it meets one or more of the exceptions described on www.clinicalstudydatarequest.com under “Sponsor Specific Details for Astellas”.
Acknowledgements
The study was funded by Astellas Pharma, Inc., Japan, and Astellas Pharma Global Development, Inc., USA. Medical writing support was provided by Annick Moon, PhD, for Cello Health MedErgy, funded by Astellas Pharma, Inc.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).
Treatment failure and recurrence of Clostridium difficile infection following treatment with vancomycin or metronidazole: a systematic review of the evidence.
Fidaxomicin versus vancomycin for infection with Clostridium difficile in Europe, Canada, and the USA: a double-blind, non-inferiority, randomised controlled trial.
Efficacy and safety of fidaxomicin for the treatment of Clostridioides (Clostridium) difficile infection in a randomized, double-blind, comparative Phase III study in Japan.
Australasian Society of Infectious Diseases updated guidelines for the management of Clostridium difficile infection in adults and children in Australia and New Zealand.
Clinical efficacy of fidaxomicin compared with vancomycin and metronidazole in Clostridium difficile infections: a meta-analysis and indirect treatment comparison.
☆All authors meet the ICMJE authorship criteria and have made substantial contributions to all the following: the conception and design of the study, or acquisition of data, or analysis and interpretation of data; drafting the article or revising it critically for important intellectual content; and final approval of the version to be submitted.
30207-7&id=gr1.jpg){kind=link}
30207-7&id=gr2.jpg){kind=link}
30207-7&id=gr3.jpg){kind=link}
30207-7&id=gr4.jpg){kind=link}
30207-7&id=gr5.jpg){kind=link}
30207-7&id=gr6.jpg){kind=link}