Characteristics of included studies [ordered by study ID]

Methods  RCT 

 Participants  CDAD 

 Interventions  teicoplanin dose study 100 mg bid (n=49) versus qid (n = 43) 

 Outcomes  cure bacteriologic resolution relapse 

 Notes  

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Abstract reports randomized, but not how 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  Unclear risk  Abstract reports double blind 
 Incomplete outcome data (attrition bias) All outcomes  High risk  47% dropouts, 20/43 in 100 mg bid group and 25/49 in 100 mg qid group 
 Other bias  Unclear risk  Criteria for 'improved' outcome not described  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin (n = 10) versus rifaximin (n =10) 

 Outcomes  combined symptomatic and bacteriologic resolution 

 Notes  

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Not described 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  High risk  No mention of blinding of patients or outcome assessors 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  No dropouts 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea
Unclear inclusion criteria and unclear cure criteria  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin (n = 24) versus teicoplanin (n = 27) 

 Outcomes  cure bacteriologic resolution relapse 

 Notes  

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  No description given 
 Allocation concealment (selection bias)  Unclear risk  No description given 
 Blinding (performance bias and detection bias) All outcomes  High risk  No mention of blinding of patients or outcome assessors 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  5/51 dropouts; 4/24 in vancomycin group and 1/27 in teicoplanin group 
 Other bias  Unclear risk  Not described  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin (n = 31) versus bacitracin (n = 31) 

 Outcomes  cure bacteriologic resolution relapse 

 Notes  

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Low risk  Computer generated 
 Allocation concealment (selection bias)  Low risk  A  Adequate 
 Blinding (performance bias and detection bias) All outcomes  Unclear risk  Unclear whether assessor was blinded 
 Incomplete outcome data (attrition bias) All outcomes  High risk  52% dropouts, groups not specified 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea
Good randomisation technique, but did not test for C. difficile in stool until after randomisation  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin dose study (125 (n = 28) versus 500 mg (n = 28) qid 

 Outcomes  cure bacteriologic resolution relapse 

 Notes  high dropout rate 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Low risk  Random numbers table 
 Allocation concealment (selection bias)  High risk  Not used 
 Blinding (performance bias and detection bias) All outcomes  High risk  Outcome assessor not blinded 
 Incomplete outcome data (attrition bias) All outcomes  High risk  18% dropouts, groups not specified 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea
Wide range of treatment duration  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin (n = 22) versus placebo (n = 22) 

 Outcomes  cure bacteriologic resolution 

 Notes  60% dropouts 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Not described 
 Allocation concealment (selection bias)  Low risk  Centralized randomization by pharmacy 
 Blinding (performance bias and detection bias) All outcomes  Unclear risk  Patients were blinded, but it was unclear whether the outcome assessor was too 
 Incomplete outcome data (attrition bias) All outcomes  High risk  52% dropouts; 10/22 in vancomycin group and 13 of 22 in placebo, with additional lost data 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea
Good grouping of patients according to stool testing.
Poor follow up procedures leading to only 4/12 patients having follow up data in treatment arm  


Methods  RCT 

 Participants  CDAD 

 Interventions  metronidazole (n = 20) versus metronidazole plus rifampin (n = 19) 39 randomized 

 Outcomes  resolution of diarrhea within 10 days and relapse within 40 days 

 Notes  0 dropouts 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Low risk  Quote: "Randomization was computer generated, and blinded study staff enrolled patients using numbered packages." 
 Allocation concealment (selection bias)  Low risk  Quote: "The sequence of randomization numbers was concealed until the end of the study"
Comment: As study staff were blinded before enrollment, there does not seem to be a source of bias 
 Blinding (performance bias and detection bias) All outcomes  High risk  Singleblinded
Quote: "A placebo was not used"
Comment: This could have been placebo controlled although good justification was given" 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  No dropouts 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea  


Methods  RCT 

 Participants  CDAD 

 Interventions  OPT 80 at 3 doses (100 mg, 200 mg and 400 mg) n = 16 in each group 

 Outcomes  Resolution of diarrhoea and abdominal discomfort within the 10 day treatment period without requiring any additional therapy and relapse within 6 weeks of end of treatment 

 Notes  

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  The trial used "interactive voice randomization system" 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  High risk  "the study was a dosefinding, randomized, openlabel study" 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  4% dropouts not included in analysis 
 Other bias  Unclear risk  Patients with severe disease were excluded and only patients with a primary episode or first relapse were included
Patients were excluded if they had > 24 hrs antibiotic (metronidazole or vancomycin) prior to enrolment  


Methods  RCT 

 Participants  CDAD 

 Interventions  Nitazoxanide in two doses (n = 44) versus metronidazole (n = 98) 

 Outcomes  Resolution of diarrhea at 7 and 31 days and time to resolution 

 Notes  32 dropouts 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Quote: "patients were randomized to 1 of 3 groups, in doubleblinded fashion"
Comment: No mention is made of how randomization was established 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  Low risk  Double blind: tablets made to look identical 
 Incomplete outcome data (attrition bias) All outcomes  High risk  23% dropouts, groups not specified 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea  


Methods  RCT 

 Participants  CDAD 

 Interventions  Vancomycin 125 mg 6 hourly (n = 27) versus nitazoxanide 500 mg 12 hourly (n = 22) 

 Outcomes  Complete resolution of symptoms and signs attributable to C difficile within 3 days after completion of therapy
Relapse 

 Notes  

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Not described 
 Allocation concealment (selection bias)  Low risk  Centralized randomization (randomization codes held by study sponsor) 
 Blinding (performance bias and detection bias) All outcomes  Low risk  Doubleblind, doubledummy: all patients had active and placebo tablets for the duration of the study 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  2% dropout not included in analysis 
 Other bias  Low risk  The study appears to be free of other sources of bias  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin (n = 56) versus metronidazole (n = 45) 

 Outcomes  cure bacteriologic resolution 

 Notes  no relapse data 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Not described 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  Unclear risk  Not described 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  7% dropouts; 4 from vancomycin group and 3 from metronidazole group 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea
Also unclear as to which patients had original antibiotic removed  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin versus metronidazole versus teicoplanin versus fusidic acid. N for each group a bit unclear as 126 randomized and 7 dropped out, leaving the specified numbers for analysis. 

 Outcomes  cure bacteriologic resolution relapse 

 Notes  cost data presented 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Low risk  Quote: "patients were randomized according to a table of random numbers" 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  High risk  No mention of blinding of patients or outcome assessors 
 Incomplete outcome data (attrition bias) All outcomes  Unclear risk  5.5% dropouts, groups not specified 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea
Did exclude patients with no WBCs in stool sample "to insure inclusion of patients with significant disease due to C. difficile"
Did however explicitly report that original offending antibiotic had been stopped  


Methods  RCT 

 Participants  CDAD 

 Interventions  fusidic acid (n = 67) versus metronidazole (n = 64) 131 randomized 

 Outcomes  cessation of diarrhea and conversion to toxin () 

 Notes  17 dropouts 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Low risk  Quote: "An independent statistician provided a computergenerated list of random set numbers" 
 Allocation concealment (selection bias)  Low risk  Quote: "the investigator teams were unaware of the treatment allocation" 
 Blinding (performance bias and detection bias) All outcomes  Low risk  All medication packs were coded and contained identicallooking pills. 
 Incomplete outcome data (attrition bias) All outcomes  High risk  26% dropouts; 20/67 in fusidic acid group and 14/64 from metronidazole group 
 Other bias  High risk  Did not exclude other pathogens in the stool as causes of diarrhoea  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin versus bacitracin; 21 in each group 

 Outcomes  cure bacteriologic resolution relapse 

 Notes  cost data presented 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Unclear risk  Not described 
 Allocation concealment (selection bias)  Unclear risk  Not described 
 Blinding (performance bias and detection bias) All outcomes  Unclear risk  Patients blinded but unclear whether assessors were too  although the abstract reports it was doubleblind 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  no dropouts 
 Other bias  Low risk  Controls for diarrhoea resolution on removal of offending antibiotic  


Methods  RCT 

 Participants  CDAD 

 Interventions  vancomycin (n = 82) versus metronidazole (n = 90) 

 Outcomes  cessation of diarrhoea
conversion to C. difficile toxin A negative stool
relapse at 21 days post cure 

 Notes  12.5% dropout, but achieved power 

 Risk of bias 
 Bias  Authors' judgement  Support for judgement 
 Random sequence generation (selection bias)  Low risk  Quote: "a member of pharmacy staff randomized participants by selecting a card from a sealed envelope..." 
 Allocation concealment (selection bias)  Low risk  Centralized randomization by pharmacy using sealed envelope 
 Blinding (performance bias and detection bias) All outcomes  Low risk  Quote: "patients received either vancomycin liquid and a placebo tablet that was similar in appearance to metronidazole or a metronidazole tablet and an unpleasantlyflavored placebo liquid" 
 Incomplete outcome data (attrition bias) All outcomes  Low risk  13% dropouts, missing outcome data balanced in numbers across intervention groups with similar reasons for missing data across groups 
 Other bias  Unclear risk  No patients with suspected or lifethreatening intraabdominal complications (perforation or obstruction) were included
No patients in ITU or with pseudomembranous colitis were included
Unclear whether prior antibiotics had been stopped  

Characteristics of excluded studies [ordered by study ID]

Study  Reason for exclusion 

 Johnson 1992  Participants were asymptomatic carriers of Clostridium difficile without diarrhea 
 Louie 2006  RCT with a non antibiotic arm (tolevamer) 
 Lowy 2010  Does not compare two antibiotics and focuses on the recurrence of Clostridium difficile rather than treatment of the existing infection 
 Mattila 2008  RCT with a non antibiotic arm (Clostridium difficile immune whey) 
 McFarland 2002  Nonrandomized study 
 Noren 2006  Publication of antibiotic resistance development from a group previously reported by Wullt2004. 
 Numan 2007  Assessment of immune whey efficacy  

Comparison 1. Vancomycin versus Placebo
Comparison 2. Metronidazole versus Vancomycin

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  3  335  Risk Ratio (MH, Fixed, 95% CI)  0.93 [0.86, 1.02] 
 2 Bacteriologic Initial Cure  1  62  Risk Ratio (MH, Fixed, 95% CI)  0.96 [0.70, 1.30] 
 3 Bacteriologic Cure  2  163  Risk Ratio (MH, Fixed, 95% CI)  0.85 [0.62, 1.17] 
 4 Cure (symptomatic and bacteriologic cure)  1  172  Risk Ratio (MH, Fixed, 95% CI)  0.88 [0.75, 1.02] 
  1  90  Risk Ratio (MH, Fixed, 95% CI)  0.91 [0.76, 1.08] 
  1  82  Risk Ratio (MH, Fixed, 95% CI)  0.83 [0.64, 1.09] 
 5 Symptomatic Cure minus recurrences including Zar with all exclusions treated as failures  3  335  Risk Ratio (MH, Fixed, 95% CI)  0.91 [0.81, 1.03]  

Comparison 3. Bacitracin versus Vancomycin

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  2  104  Risk Ratio (MH, Fixed, 95% CI)  0.82 [0.60, 1.12] 
 2 Symptomatic Cure  2  104  Risk Ratio (MH, Fixed, 95% CI)  0.58 [0.34, 0.99] 
 3 Bacteriologic Initial Response  2  104  Risk Ratio (MH, Fixed, 95% CI)  0.52 [0.31, 0.86] 
 4 Symptomatic Recurrence  2  104  Risk Ratio (MH, Fixed, 95% CI)  1.11 [0.50, 2.48]  

Comparison 4. Teicoplanin versus Vancomycin

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  2  110  Risk Ratio (MH, Fixed, 95% CI)  1.07 [0.95, 1.19] 
 2 Symptomatic Cure  2  110  Risk Ratio (MH, Fixed, 95% CI)  1.21 [1.00, 1.46] 
 3 Bacteriologic Initial Response  2  110  Risk Ratio (MH, Fixed, 95% CI)  1.43 [1.14, 1.81] 
 4 Bacteriologic Cure  1  59  Risk Ratio (MH, Fixed, 95% CI)  1.82 [1.19, 2.78] 
 5 Symptomatic Recurrence  2  110  Risk Ratio (MH, Fixed, 95% CI)  0.44 [0.14, 1.36] 
 6 Bacteriologic Recurrence  1  59  Risk Ratio (MH, Fixed, 95% CI)  0.37 [0.11, 1.23]  

Comparison 5. Fusidic Acid versus Vancomycin

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  2  191  Risk Ratio (MH, Fixed, 95% CI)  0.94 [0.83, 1.08] 
 2 Symptomatic Cure  1  60  Risk Ratio (MH, Fixed, 95% CI)  0.85 [0.61, 1.17] 
 3 Bacteriologic Initial Response  2  191  Risk Ratio (MH, Fixed, 95% CI)  0.96 [0.78, 1.18] 
 4 Bacteriologic Cure  1  60  Risk Ratio (MH, Fixed, 95% CI)  0.68 [0.44, 1.06] 
 5 Symptomatic Recurrence  2  191  Risk Ratio (MH, Fixed, 95% CI)  1.09 [0.64, 1.86] 
 6 Bacteriologic Recurrence  1  60  Risk Ratio (MH, Fixed, 95% CI)  1.87 [0.61, 5.73]  

Comparison 6. Nitazoxanide versus Vancomycin

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  1   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  49  Risk Ratio (MH, Fixed, 95% CI)  1.04 [0.76, 1.43] 
  1  29  Risk Ratio (MH, Fixed, 95% CI)  0.98 [0.64, 1.49] 
  1  20  Risk Ratio (MH, Fixed, 95% CI)  1.14 [0.69, 1.90] 
 2 Recurrence of diarrhea within 31 days  1   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  49  Risk Ratio (MH, Fixed, 95% CI)  0.61 [0.06, 6.33] 
  1  29  Risk Ratio (MH, Fixed, 95% CI)  0.46 [0.02, 10.45] 
  1  20  Risk Ratio (MH, Fixed, 95% CI)  1.0 [0.07, 13.87] 
 3 Sustained response  1   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  49  Risk Ratio (MH, Fixed, 95% CI)  1.09 [0.75, 1.58] 
  1  29  Risk Ratio (MH, Fixed, 95% CI)  1.06 [0.68, 1.66] 
  1  20  Risk Ratio (MH, Fixed, 95% CI)  1.17 [0.61, 2.23]  

Comparison 7. Rifaximin versus Vancomycin
Comparison 8. Metronidazole versus Nitazoxanide
Comparison 9. Metronidazole versus Metronidazole + Rifampin
Comparison 10. Metronidazole versus Teicoplanin

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  1  59  Risk Ratio (MH, Fixed, 95% CI)  0.97 [0.86, 1.09] 
 2 Symptomatic Cure  1  59  Risk Ratio (MH, Fixed, 95% CI)  0.87 [0.69, 1.09] 
 3 Bacteriologic Initial Cure  1  59  Risk Ratio (MH, Fixed, 95% CI)  0.76 [0.60, 0.98] 
 4 Bacteriologic Cure  1  59  Risk Ratio (MH, Fixed, 95% CI)  0.76 [0.58, 1.00] 
 5 Symptomatic Recurrence  1  59  Risk Ratio (MH, Fixed, 95% CI)  2.26 [0.48, 10.73]  

Comparison 11. Metronidazole versus Fusidic Acid

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  2  190  Risk Ratio (MH, Random, 95% CI)  1.09 [0.60, 1.99] 
 2 Symptomiatic Cure  2  190  Risk Ratio (MH, Fixed, 95% CI)  1.00 [0.78, 1.30] 
 3 Bacteriologic initial Cure  2  190  Risk Ratio (MH, Fixed, 95% CI)  1.02 [0.83, 1.25] 
 4 Bacteriologic Cure  1  59  Risk Ratio (MH, Fixed, 95% CI)  1.35 [0.87, 2.11] 
 5 Symptomatic Recurrence  2  190  Risk Ratio (MH, Fixed, 95% CI)  0.91 [0.53, 1.55]  

Comparison 12. Teicoplanin versus Fusidic Acid

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  1  57  Risk Ratio (MH, Fixed, 95% CI)  1.04 [0.92, 1.17] 
 2 Symptomatic Cure  1  57  Risk Ratio (MH, Fixed, 95% CI)  1.36 [1.02, 1.83] 
 3 Bacteriologic Initial Cure  1  57  Risk Ratio (MH, Fixed, 95% CI)  1.68 [1.19, 2.37] 
 4 Bacteriologic Cure  1  57  Risk Ratio (MH, Fixed, 95% CI)  1.73 [1.19, 2.51] 
 5 Symptomatic Recurrence  1  57  Risk Ratio (MH, Fixed, 95% CI)  0.26 [0.06, 1.11]  

Comparison 13. Dose finding studies

Outcome or subgroup title  No. of studies  No. of participants  Statistical method  Effect size 

 1 Symptomatic Initial Response  2   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  56  Risk Ratio (MH, Fixed, 95% CI)  0.92 [0.72, 1.17] 
  1  92  Risk Ratio (MH, Fixed, 95% CI)  0.79 [0.49, 1.29] 
 2 Symptomatic Recurrence  3   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  56  Risk Ratio (MH, Fixed, 95% CI)  0.8 [0.24, 2.67] 
  1  92  Risk Ratio (MH, Fixed, 95% CI)  1.14 [0.47, 2.78] 
  1  48  Risk Ratio (MH, Fixed, 95% CI)  0.2 [0.03, 1.43] 
 3 Symptomatic Cure  3   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  56  Risk Ratio (MH, Fixed, 95% CI)  0.95 [0.65, 1.38] 
  1  92  Risk Ratio (MH, Fixed, 95% CI)  0.61 [0.29, 1.29] 
  1  48  Risk Ratio (MH, Fixed, 95% CI)  1.26 [1.03, 1.54] 
 4 Bacteriologic Cure  2   Risk Ratio (MH, Fixed, 95% CI)  Subtotals only 
  1  56  Risk Ratio (MH, Fixed, 95% CI)  0.8 [0.24, 2.67] 
  1  92  Risk Ratio (MH, Fixed, 95% CI)  0.98 [0.51, 1.88]  

Summary of findings for the main comparison. Metronidazole versus Vancomycin for Clostridium difficileassociated diarrhea in adults

Metronidazole versus Vancomycin for Clostridium difficileassociated diarrhea in adults 
 Patient or population: patients with Clostridium difficileassociated diarrhea in adults Settings: Intervention: Metronidazole versus Vancomycin 
 Outcomes  Illustrative comparative risks* (95% CI)  Relative effect (95% CI)  No of Participants (studies)  Quality of the evidence (GRADE)  Comments 
 Assumed risk  Corresponding risk 
 Control  Metronidazole versus Vancomycin 
 Symptomatic Cure minus recurrences including Zar with all exclusions treated as failures  Study population  RR 0.91 (0.81 to 1.03)  335 (3 studies)  ⊕⊝⊝⊝ very low^{1,2,3}  
 782 per 1000  712 per 1000 (626 to 805) 
 Medium risk population 
 781 per 1000  711 per 1000 (625 to 804) 
 *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; 
 GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.  

^{1} Sparse data ^{2} The Wenisch 1996 study was not blinded ^{3} Both the Teasley 1983 and Wenisch 1996 studies did not exclude other pathogens in stool as causes of diarrhea

Summary of findings 2. Teicoplanin versus Vancomycin for Clostridium difficileassociated diarrhea in adults

Teicoplanin versus Vancomycin for Clostridium difficileassociated diarrhea in adults 
 Patient or population: patients with Clostridium difficileassociated diarrhea in adults Settings: Intervention: Teicoplanin versus Vancomycin 
 Outcomes  Illustrative comparative risks* (95% CI)  Relative effect (95% CI)  No of Participants (studies)  Quality of the evidence (GRADE)  Comments 
 Assumed risk  Corresponding risk 
 Control  Teicoplanin versus Vancomycin 
 Symptomatic Cure  Study population  RR 1.21 (1.00 to 1.46)  110 (2 studies)  ⊕⊝⊝⊝ very low^{1}  
 714 per 1000  843 per 1000 (693 to 1000) 
 Medium risk population 
 708 per 1000  835 per 1000 (687 to 1000) 
 *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; 
 GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.  

^{1} Sparse data ^{2} The Wenisch 1996 and de Lalla 1992 studies were not blinded ^{3} The Wenisch 1996 study did not exclude other pathogens in stool as causes of diarrhea

Summary of findings 3. Metronidazole versus Teicoplanin for Clostridium difficileassociated diarrhea in adults

Metronidazole versus Teicoplanin for Clostridium difficileassociated diarrhea in adults 
 Patient or population: patients with Clostridium difficileassociated diarrhea in adults Settings: Intervention: Metronidazole versus Teicoplanin 
 Outcomes  Illustrative comparative risks* (95% CI)  Relative effect (95% CI)  No of Participants (studies)  Quality of the evidence (GRADE)  Comments 
 Assumed risk  Corresponding risk 
 Control  Metronidazole versus Teicoplanin 
 Symptomatic Cure  Study population  RR 0.87 (0.69 to 1.09)  59 (1 study)  ⊕⊕⊝⊝ low^{1}  
 833 per 1000  750 per 1000 (583 to 966) 
 Medium risk population 
 833 per 1000  750 per 1000 (583 to 966) 
 *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; 
 GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.  

^{1} Sparse data ^{2} The Wenisch 1996 study was not blinded ^{3} The Wenisch 1996 study did not exclude other pathogens in stool as causes of diarrhea

Table 1. Guidelines for Antibiotic Treatment of CDAD

 HPA^{1}  ESCMID^{2}  SHEA & IDSA^{3} 
 MILD CDAD  Stop inciting antibiotic and observe, or metronidazole 500 mg po tid. Alternate dosing also recommended and change to vancomycin if no better in 4 days.  Stop inciting antibiotic and observe, or Metronidazole 500 mg po tid  Stop inciting antibiotic, or metronidazole 500 mg po tid for 10 days. 
 SEVERE CDAD  Vancomycin 500 mg po qid with tapering  Vancomycin 125 mg po qid for 10 days  Vancomycin 125 mg po qid for 10 days 
 SEVERE AND CANNOT TOLERATE ORAL MEDS  Intravenous metronidazole and vancomycin via nasogastric tube or enemas qid.  same  same 
 SURGERY  For toxic megacolon or lactate > 5  For perforation, toxic megacolon, Ileus, lactate > 5  
 RECURRENCE   First as Primary
2nd: oral vancomycin with taper  same  

