Clostridium Difficile Infection (CDI) and Antibiotic ......PB – Meta-analysis of data up to 2013 Allen S.J. et al Lancet 2013 ; 382 : 1249 Study or Subgroup Beniwal 2003 Beausoleil

Antibiotic – Associated Diarrhea (AAD) and Clostridium Difficile

Infection (CDI)

Dario Conte, M.D. Gastroenterology and Endoscopy Unit

Postgraduate School of Gastroenterology

Fondazione IRCCS Ca’ Granda

Ospedale Maggiore Policlinico – Milano

Università degli Studi – Milano

Gargnano - October 08 - 11, 2014

There is no potential conflict of interest relevant to this lecture

Clinical scenario A.C. 86 year old man Chronic ischemic cardiomyopathy Type 2 DM Chronic renal insufficiency Past ischemic stroke Ongoing enteral nutrition via n-g tube

PNEUMONIA (PMN)

Hospital admission

Empirical AB regimen for PMN High frequency of CDI

Antibiotic – Associated Diarrhea (AAD)

High incidence during or after AB regimen (5-40%)

Mechanisms %

__________________________________________________

Aspecific 70 – 80

CD-related (CDI) 20

Other pathogens 2 – 3

Specific mechanism(s) molecule related 3 – 5

CDI - Pathogenesis

AB regimen

Abnormalities in intestinal microflora

Exposition to environmental

CD spores

CD colonization Symptomless A and B toxins CDI

CDI – Key points

• Iatrogenic infection, whose incidence is increasing due to the widespread use of large spectrum AB

• Any type of AB can be responsible

• Symptoms can arise during, soon after or within 4 - 8 weeks after AB withdrawal (max after 4 – 10 days of therapy)

• Antifungal and antiviral products too can be involved

CDI – Risk factors

• Age

• Severity of underlying disease

• Comorbidities

• Enteral nutrition

• Concomitant use of PPI

• Concomitant use of H2 blockers

• PGFE type 1 (NAP 1) strain

CDI – clinical spectrum

Ileal / hypoalbuminemia / ascites

Pseudomembranes

Fever

Leucocytosis

Diarrhea

CD

3%

Fulminant colitis

Pseudo- membranous

colitis

Moderate to severe

colitis

Mild to moderate diarrhea

Healthy carrier

50- 60%

3-20%

C. Difficile

• G+ bacillus

• Sporigenic

• Toxigenic Tox A Tox B

CDI – Diagnostic test

• Reference standard : cytotoxic assay

• Second generation ELISA – Simpler and faster

– Sensitivity 90 % ( three samples)

– Specificity 100 % : this performance refers to the analytic one (i.e. concerns the presence of the toxin, not of the disease)

CDI – Treatment (I) • Stop responsible AB ! (whenever possible)

• Supporting therapy alone obtains spontaneous resolution in 25 %

• In moderate to severe forms :

– metronidazole 500 mg t.i.d. orally or

– vancomycin 500 mg f.i.d. orally

– fidaxomycin : AB promisingly more selective

(ongoing RCT)

Duration of treatment : 10 – 14 days

CDI – Treatment (II)

Metronidazole and Vancomycine :

same effectiveness

Vancomycine should be preferred in :

- pregnant women

- intolerants to metronidazole

- compromised patients

In case of impossibility of oral route :

- metronidazole 500 mg / t.i.d.

CDI – Treatment (III)

Healing rate: 95 %

Alternative regimens:

- bacitracin 25.000 U f.i.d. / 7 – 14 days

- rifampicin 600 mg b.i.d. / 14 days

- cholestyramine

- toxin adsorbent polymers ( synsorb, tolevamar )

CDI - Relapse (I)

• 10 – 20 % relapse within two weeks (clinical relapse), due to:

– persistence of spores

– reinfection (50 %)

– resistance to therapy.

• Further period of treatment with metro for

10 – 14 days.

CDI – Relapse (II)

Alternatives for multiple relapses (no RCT ! )

• Metro/vanco for 4 – 6 weeks with progressively decreasing doses.

• Vanco plus cholestyramine

• Vanco + Rifampicin

• Probiotics (PB)

• Fecal microbiota transplantation (FMT)

C.Difficile Asymptomatic carrier

• Infants and children 80%

• Adults 0-3%

• Hospitalized population 20%

• Minor risk of developing CDI

• Reservoir

• Treatment not indicated • Failure in eradicating CDI infection

• Possible prolongation of carrier state

C. Difficile – Infection control Guidelines

• Drastic AB reduction

• Hands washing

• Fecal “isolation”

• Mandatory use of disposable gloves

• Environmental disinfection with sodium hypochloride

• Specific educational training of nursing and medical staff

CDI – Incidence reduction Milestones

2007 : the peak incidence of CDI in the UK

• Hand wash policy • “Bare below the elbow” • Isolation of infected patients • Reduction of the use of high–risk AB Year CDI /100,000 bed / day 2007 136.7 2010 20.5 Society for Healthcare Epidemiology of America (SHEA) Infectious Disease Society of America (IDSA)

Infect Control Hosp Epidemiol 2010; 31 : 431

Two further topics to be

elucidated

• PB in AAD and CDI prevention

• FMT (Fecal Microbiota Transplantion) for CDI

Effectiveness of PB in preventing AAD Systematic review (up to 2005)

Active PB : S. boulardii , L rhamnosus , Different mix

Flaws : same Author of the largest study , low quality of the studies included ( adequate sample size < 10% ), significant heterogeneity

Mc Farland L.V. et al Amer J Gastroenterol 2006; 101 : 812

PB for prevention

and treatment of AAD:

a systematic review and

meta-analysis Hampel S et al JAMA 2012; 507: 1959

PB for prevention and treatment of AAD A

systematic review and meta-analysis

• 63 RCT (11, 811 participants)

• No evidence of publication bias

• Overall random effect model (I² i.e. heterogeneity inconsistent with chance: 54%)

• RR 0.58 (95% CI 0.50 – 0.68)

• Risk difference – 0.07 (95% CI – 0.10 – 0.05)

• NNT 13 (95% CI – 10 – 19)

• I² 54%

• Results were relatively insensitive to subgroup analyses for PB types, participants, setting, AB types

Overall : “More research is needed to determine which PB are associated with the greatest efficacy and for which patients receiving which specific AB”

Hampel S. et al JAMA 2012; 507 : 1959

PB for prevention and treatment of CDI A systematic review and meta-analysis (I)

• 20 RCT (3818 participants) • No evidence of publication bias • RR 0.34 (95% CI 0.24 – 0.49) • I² 0% (the heterogeneity is completely consistent with chance) • Optimal information size 5676 ↓evidence for imprecision→quality moderate (GRADE)

Johnston B.C. et al Ann Intern Med 2012 ; 157 : 878

PB for prevention and treatment of CDI A systematic review and meta-analysis (II)

Control risk : 0 – 40 % . Two studies possibly done during CD outbreaks

Johnston B.C. et al Ann Intern Med 2012 ; 157 : 878

AAD and CDI in patients randomized to Placebo or S. boulardii

Placebo (# 98)

S. Boulardii (# 106)

OR (95% CI)

P

# % # %

AAD 13 (13.3) 16 (15.1) 1.16 (0.53 – 2.56) NS

CDI 2 (2.0) 3 (2.8) 1.40 (0.23 – 8.55) NS

Control risk of CDI

Colli A. et al Amer J Gastroenterol 2012; 107: 922

AAD AND CDI IN PATIENTS RANDOMIZED TO PLACEBO OR ACTIVE REGIMEN (VSL3)

Control risk of CDI < 2%

Selinger C.P. et al J Hosp Infect 2013; 84 :159

PLACEBO ( # 112 ) ACTIVE ( # 117 ) P

# ( % ) # ( % )

AAD 10 (8.9 ) 5 (4.3) 0.19

CDI 0 (-) 0 ( - ) 1.0

30 D -

MORTALITY

4 (3.6) 2 (1.7) 0.44

..now, let us focus on the “most active” pb

i.e., lactobacilli and bifidobacteria

PB – Meta-analysis of data up to 2013

Allen S.J. et al Lancet 2013 ; 382 : 1249

Study or Subgroup

Beniwal 2003

Beausoleil 2007

Hickson 2007

Stockenhuber 2008

Total (95% CI)

Total events

Heterogeneity: Tau² = 0.00; Chi² = 2.95, df = 3 (P = 0.40); I² = 0%

Test for overall effect: Z = 6.08 (P < 0.00001)

Events

13

7

7

17

44

Total

105

44

57

340

546

Events

23

16

19

63

121

Total

97

45

56

338

536

Weight

26.9%

16.9%

16.9%

39.3%

100.0%

M-H, Random, 95% CI

0.52 [0.28, 0.97]

0.45 [0.20, 0.98]

0.36 [0.17, 0.79]

0.27 [0.16, 0.45]

0.37 [0.27, 0.51]

Year

2003

2007

2007

2008

Microbial preparation Placebo Risk Ratio Risk Ratio

M-H, Random, 95% CI

0.1 0.2 0.5 1 2 5 10

Favours Microbial preparation Favours Placebo

Lactobacilli and bifidobacteria to prevent AAD and CDI in older patients (PLACIDE) A randomized, double-blind, placebo - controlled, multicenter trial

Allen SJ et al Lancet 2013; 382 : 1249 - 57

Microbial preparation

Placebo OR (95% CI) P

# % # %

AAD 159/1470 (10.8) 153/1471 (10.4) 1.04 (0.83 - 1.32) NS

CDI 12/1470 (0.8) 17/1471 (1.2) 0.70(0.34 – 1.48) NS

Meta-analysis of studies, including Allen 2013

Lancet 2013; 382 : 1249

Study or Subgroup

Beniwal 2003

Beausoleil 2007

Hickson 2007

Stockenhuber 2008

Allen 2013

Total (95% CI)

Total events

Heterogeneity: Tau² = 0.45; Chi² = 31.09, df = 4 (P < 0.00001); I² = 87%

Test for overall effect: Z = 2.18 (P = 0.03)

Events

13

7

7

17

159

203

Total

105

44

57

340

1470

2016

Events

23

16

19

63

153

274

Total

97

45

56

338

1471

2007

Weight

19.8%

17.8%

17.9%

21.0%

23.6%

100.0%

M-H, Random, 95% CI

0.52 [0.28, 0.97]

0.45 [0.20, 0.98]

0.36 [0.17, 0.79]

0.27 [0.16, 0.45]

1.04 [0.84, 1.28]

0.49 [0.25, 0.93]

Year

2003

2007

2007

2008

2013

Microbial preparation Placebo Risk Ratio Risk Ratio

M-H, Random, 95% CI

0.1 0.2 0.5 1 2 5 10

Favours Microbial preparation Favours Placebo

Funding may influence

trial results

Funding influence on trial results

Kolber MR et al Am J Gastroentrol 2014

PB - Overall considerations

ARE PB EFFICACIOUS IN PREVENTING

AAD and CDI ?

PB SHOULD NOT BE RECOMMENDED

If it is possible, implementing simple hygienic rules,

to obtain a CDI incidence lower than 2%,

even a magic “very effective” probiotic,

able to reduce the basal risk of 50% ,

would be not so useful:

to avoid one case of CDI the number of patients needed to treat

NNT would be > 100

Colli A et al Am J Gastroenterol 2014

«THE MAGIC BULLET»

FMT (Fecal Microbiota Transplantation) for CDI

Fecal bacteriotherapy, also called FMT, may be a useful treatment for CDI

through restoration of the intestinal microbiota

Eiseman B. et al Surgery 1958; 44 : 854

Systematic review of FMT for recurrent CDI

347 pts 92 % disease resolution

(only case series and reports are included )

Sough E. et al Can Infect Dis 2011; 53 : 994

Duodenal infusion of donor feces for recurrent CDI ( I )

42 patients randomly assigned to FMT, Vancomycin or Vancomycin plus bowel lavage

Planned number of patients per group : 40

Interim efficacy analysis (at predefined time ?)

Termination of the trial

Due to early termination, rate ratios for the primary end point (overall cure) were calculated with the exact 99.9 % CI (Haybitte – Peto rule, i.e. p < 0.001 for the primary end point)

Van Nood E. et al NEJM 2013 ; 360

Duodenal infusion of donor feces for recurrent CDI (II)

Cure (%) without relapse

Overall cure rate ratio of donor-feces infusion: 3.05 (99.9 % CI 1.08 – 290.05). QUITE AN IMPRECISE RESULT

van Nood E. et al NEJM 2013 ; 360

22 case series, 15 case reports and one RCT: 536 patients

467 cases (87%) experienced resolution of diarrhea

“NO severe adverse events were reported with FMT”

BUT

to detect a rare adverse event, i.e. < 1 : 1000 at least 3000 participants

should have been enrolled

FMT to treat CDI. A systematic review

Cammarota G et al. J Clin Gastroenterol 2014

Efficacy of combined jejunal and colonic FMT for recurrent CDI

• Very hot topic: be careful in drawing conclusions

Joannidis J.P. Why most published research findings are false?

PLoS Med 2005 ; 2 : 124

• > 30 studies registered and active on clin. Gov • e.g., Dutta S.K. : efficacy = 27/27 (!)

Dutta S. K. et al Clin Gastroenterol Hepatol 2014

Final considerations

A tendency in human thinking is to believe rather than disbelieve

Type 1 processing occurs by viewing something as more predictable and coherent than is really the case

Be skeptical

Kahneman D. Thinking, fast and slow. New York: Doubleday Canada, 2011 Schermer M. The believing brain. New York: Times Books, Henry Holt and Company, 2011

Acknowledgements

• To Christian Gluud, Dimitrinka Nikolova, Agostino Colli and Mirelli Fraquelli for their invaluable help

• To Sara Comparetti for her terrific work

• To all my Colleagues for continuos cooperation

• To all of you for patience in attending this lecture