2/7/20181

Clostridium difficileSarah Doernberg, MD, MAS

Assistant professor and Medical Director of Antimicrobial Stewardship

Division of Infectious Diseases, UCSF

2.19,2018

Outline

Brief background and epidemiology

Diagnosis

Management—mild, uncomplicated disease

Management—moderate-severe disease

Management—recurrent/relapsed disease

Management—fulminant disease

Prevention

2/7/20182

One of CDC’s 3 “Urgent Threats”

https://www.cdc.gov/drugresistance/biggest_threats.html

500,000

3.8 billion

CDI Background Anaerobic, spore-forming gram-

positive bacillus

Toxins A + B

Multiple strains

• Epidemic strain ID’d 2004

• 078 strain

Fecal-oral spread

12% of all HAIs

Carriage of C. difficile

• < 3% for healthy adults in community

• 20% in hospitalized pts

• up to 50% in LTCF

Risk factors:

• Antibiotics

• Age

• Hospitalization

• Acid-suppression

• IBD

• Tube feeds

• Host immune factors

• Chemotherapy

• Female gender

• Domestic animals? Retail food?

Magill SS et al., NEJM 2014

2/7/20183

Epidemiology trends, inpatients

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6034a7.htm

Epidemic strain

Molecular testing era

Duration, number, and intensity of antibiotics affect risk for CDI

6Stevens V, et al. Clin Infect Dis 2011; 53: 42-48.

2/7/20184

Antibiotic use affects the population risk

Brown K et al. JAMA Intern Med. 2015 Apr;175(4):626-33

Spread of CDI in the hospital

Asymptomatic carriersSymptomatic cases

25-33%

30%

Walker AS et al. PLoS Med. 2012 Feb;9(2):e1001172.; Kamboj M et al. Infect Control Hosp Epidemiol. 2016 Jan; 37(1): 8–15; Curry SR et al. Clin Infect Dis. 2013 Oct 15; 57(8): 1094–1102; McDonald LC, Clin Infect Dis. 2013 Oct;57(8):1103-5

Endogenous carriage

?

2/7/20185

Diagnostic testing

Glutamate dehydrogenase Ag (GDH)• Bacterial detection• Sensitive but not specific

Polymerase chain reaction (PCR): • Toxin-producing gene• ↑Sensitivity

Enzyme immunoassay (EIA)• Protein detection• ↓Sensitivity• ↑Specificity for disease

CDI overdiagnosis

Polange CR et al., JAMA Intern Med. 2015 Nov;175(11):1792-801.

• 21% +PCR• Of these, 44% + toxin• Toxin-/PCR+

• ↓bacterial load• ↓abx• ↓diarrhea• No CDI-

complications

2/7/20186

Case

63 year old F s/p spinal fusion c/b hardware infection. She received a 6 week course of antibiotics for this and is admitted for redo spinal fusion. She has been constipated and has daily orders for senna, colace and miralax. 

On HD# 8, she develops 2 loose stools and tests positive for C. difficile.  She is afebrile with a normal WBC and is started on PO metronidazole. She has no further episodes of loose stools during the remainder of hospitalization. 

Overdiagnosis case

63 year old F s/p spinal fusion c/b hardware infection. She received a 6 week course of antibiotics and is admitted for redo spinal fusion. She has been constipated and has daily orders for senna, colace and miralax. 

On HD# 8, she develops 2 loose stools and tests positive for C. difficile.  She is afebrile with a normal WBC and is started on PO metronidazole. She has no further episodes of loose stools during the remainder of hospitalization. 

2/7/20187

MANAGEMENT

Treatment scenario #1. 63 y/o F recently treated for a UTI with levofloxacin, now having watery stools 4x/day, fever to 38.3, WBC 16K, Cr 1.7 (baseline 0.5). PCR positive for C. difficile toxin. With what should you treat her?

A. Vancomycin 125 mg po qid

B. Vancomycin 500 mg po qid

C. Metronidazole 500 mg po tid

D. Fidaxomicin 200 mg po bid

2/7/20188

CDI treatment depends on severity

Mild to moderate: Does not meet criteria for severe

• Diarrhea ≥ 3 stools/24 hours

Severe

• Not well validated

• IDSA/SHEA guidelines: Severe disease = Peak WBC > 15K or Cr > 50% above baseline or “advanced age” (65? 75?)

Severe, complicated

• Severe plus hypotension, shock, ileus, and/or megacolon

Zar F A et al. Clin Infect Dis. 2007;45:302-307; Cohen et al., Infection Control and Hospital Epidemiology, 2010; 31: 431-455

Zar F A et al. Clin Infect Dis. 2007;45:302-307; Leffler DA and Lamont JT. NEJM 2015; 372:1539-1548; Johnson S et al., Clin Infect Dis 2014;59(3):345-54

RCTs metronidazole vs. vancomycin

• Similar findings for recent study of metronidazole vs vancomycin vs tolevamer• Cure not differential with regard to levels of severity• Higher recurrence across the board (20%)• Only vancomycin is FDA-approved

0

20

40

60

80

100

120

Cure, all Cure, mild-mod Cure, severe Recurrence

MTZ

Vanco

p = 0.005 p = 0.02 NSNS

2/7/20189

New evidence to support vancomycin

Stevens VW et al. JAMA Intern Med. 2017 Feb 6. doi: 10.1001/jamainternmed.2016.9045.

• aRR death vanco vs metronidazole• Any severity: 0.86; (0.74

to 0.98)• Severe: 0.79 (0.65 to

0.97)• NNT to prevent 1 death,

severe CDI: 25

What about fidaxomicin?

Cure Relapse

Strain

Epidemic Same Same

Non-epidemic Same

Concomitant abx

Prior CDI Same =/

Louie TJ, et al. NEJM 2011;364:422-431; Cornely et al, Lancet Infect Dis 2012;12:281-8 ; Petrella LA, et al. Clin Infect Dis 2012;55(3):351-7; Mullane et al., CID 2011;53(5):440-7; Corneley et al., CID 2012;55:s154-s161.; Bartsch SM et al., CID 2013; 57(4): 555-561; Konijeti GG et al., CID 2014; 58:1507-1514.

• Bottom line vs. vanco: Similar cure (~88%), lower recurrence (13-15% vs. 25-27% )

• Unclear role in multiply recurrent or severe disease

Fidaxomicin Vancomycin Metronidazole

$2800 $250-680 $22

2/7/201810

Real-world fidaxomicin experience

UK Trust Hospitals analyzed pre-and post-information s/p introduction of fidaxomicin

Each hospital had a different approach to rxwith fidaxomicin(e.g. all patients vs. selected populations)

Goldenberg SD et al. Euro J Clin Microbiol and Infect Dis 2016; 35L 251-9.

• A and B: Fidaxomicin used first-line for all• C, E, F, G: Selected episodes• D: Recurrences only

UCSF guidelines

http://idmp.ucsf.edu/news/updated-ucsfmcvasfzsfgh-guidelines-c-difficile-infection

Clinical definition Criteria Treatment

Initial, mild‐mod, Outpatient Not meeting criteria 

for severe

Metronidazole 500 mg po q8h x 

10‐14 days

If no response @ 5 days, switch 

to vancomycin 125 mg po q6h x 

10‐14 days

Initial, mild‐mod, Inpatient Not meeting criteria 

for severe

Vancomycin 125 mg po q6h x 10‐

14 days

If unable to obtain upon 

discharge, okay to complete the 

course with metronidazole 500 

mg po q8h

Initial, severe WBC ≥ 15 OR Cr ≥ 1.5x 

baseline without 

hypotension, shock, 

ileus, and/or 

megacolon

Vancomycin 125 mg po q6h x 10‐

14 days

2/7/201811

Additional considerations

Stop unnecessary antibiotics

Shorten antibiotic courses

Narrow antibiotic spectrum

Stop acid-suppressive medications when possible

• Esp PPI

Do not use anti-peristaltic agents until acute symptoms of CDI improve

Take-home

For mild-moderate disease, can choose metronidazole, more movement towards PO vancomycin in recent years

For severe disease, choose vancomycin

• Higher cure, but same relapse

Role of fidaxomicin unclear

• Consider if high risk of relapse or need CA

• ? Use in multiply recurrent disease

• ? Role in severe disease

2/7/201812

Treatment scenario #2: You are seeing a 62 y/o F who has takes chronic amoxicillin/clavulanic acid for suppression of Enterococcal osteomyelitis and has developed her second bout of C. difficile colitis. Her WBC count is 9 and Cr is 0.3. What should you treat her with?

A. Metronidazole 500 mg po TID

B. Vancomycin 125 mg PO QID

C. Vancomycin taper

Risk for recurrent CDI

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1st episode 2nd episode 3rd episode

No recurrence

Recurrence

Johnson S. J Infect 2009;58(6):403-10; Pepin J et al. Clin Infect Dis. 2005 Jun 1;40(11):1591-7

2/7/201813

Treatment scenario #3. This patient returns one month after you have treated her with a 14-day course of PO metronidazole complaining of ongoing diarrhea. A repeat stool toxin is positive. What do you do?

A. Metronidazole 500 mg po TID x 14 days

B. Vancomycin 125 mg PO QID x 14 days

C. Vancomycin taper

D. Fidaxomicin 200 mg PO BID x 10 days

E. Other

Kelly and LaMont, N Engl J Med. 2008;359(18):1932-40.

Vancomycin taper

125 mg po 4x daily x 14 days

125 mg po 2x daily x 7 days

125 mg po 1x daily x 7 days

125 mg po every other day x 8 days (4 doses)

125 mg po every 3 days x 15 days (5 doses)

2/7/201814

↓↓↓Fecal diversity with rCDI

FMT basics

Colonization resistance

Related donors or banked stool

• Need to screen for transmissible diseases

Multiple RCTs have now been done

Guidance document available (Bakken et al)

Chang JY et al. JID 2008; 197: 435-8; Kassam et al., Arch Intern Med. 2012;172(2):191-3. Gough et al., CID 2011;53(10):994-1002; Bakken JS et al Clin Gastroenterol Hepatol 2011; 9: 1044-49

FMT trial trends

6 published

• 3 vs. abx management

• 3 vs. FMT refinements

Over time, ↓efficacy in RCTs

↑response to comparator abx

Might matter whether active recurrence vs. prior recurrence?

Might need multiple FMTs

Vanco taper might be better than we thought?

Commercially-prepared FMT in development

Johnson S and Gerding DN. Clin Infect Dis (2016) 64 (3): 272-274; van Nood E et al. N Engl J Med 2013; 368:407-415; Orenstein R et al. Clin Infect Dis (2015) 62 (5): 596-602.

2/7/201815

The latest on FMT

Hota SS et al. Clin Infect Dis (2016) 64 (3): 265-271

• Multiple previous trials supported FMT…

• But comparator group not standard of care

• Phase 2/3 open-label RCT• Stopped early for futility• FMT by enema• Recurrence: 9/16 (56%) FMT

vs. 5/12 (42%) taper group• 95% CI for ∆ CDI with FMT =

-2.8% to +47.3%

FMT meta-analysis

Overall response:

• Multiple infusions: 92% (89-94%)

• Single infusion: 84% (79-89%)

Just RCTs:

• Multiple: 91% (88-94%)

• Single: 77% (56-93%)

Quarashi MN et al. Aliment Pharmacol Ther. 2017 Sep;46(5):479-493. doi: 10.1111/apt.14201. Epub 2017 Jul 14.

2/7/201816

Kelly CR et al. Ann Int Med 2016; Van Nood E, et al. NEJM 2013; 368: 407-15; Cammarota et al, Alim Pharm Ther 2015:41:835; Youngster I et al., CID 2014;58:1515-1522

FMT leads to better cure rates

• RCT of rCDI treated with autologous vs donor FMT

• ≥ 3 episodes• Via colonoscopy• Note regional

differences• 1 donor had a 9.1 KG

wt gain• Good results for RCT

of FMT via NGT vs colo (N = 20)

FMT for abx resistance?

Millan B et al. Clin Infect Dis 2016;62:1479-1486

2/7/201817

FMT via pill?

Unblinded noninferiority (15%) multicenter RCT of 117 adults with ≥ 3 episodes of CDI

• Stratified by age (65) and immunosuppressed status (15%)

• Median duration of rCDI rx prior to transplant: 2.3-2.4 mths

Absence of rCDI @ 12 weeks: 51/53 (96%) in capsule group versus 50/52 (96%) colonoscopy group (per-protocol)

• Difference: 0% (95% CI, -6.1% to infinity)

• Pts with recurrence were retreated with same modality

• Sensitivity analyses including LTFU as recurrences still met noninferiority

• Both groups had increased microbial diversity post-transplant

Kao D et al. JAMA. 2017;318(20):1985-1993. doi:10.1001/jama.2017.17077

FMT adverse events

CommonDiarrhea

Cramping

Belching

Nausea

Bloating

Rare/seriousProcedure-related harms

• Perforation

• Aspiration

Norovirus

Bacteremia

IBD flare

Unknown long-term effects

• Weight changes

• Chronic disease exacerbation

Drekonja D et al. Ann Intern Med 2015;162(9):630-8.

2/7/201818

FMT durability

Single-center retrospective f/u of all patients receiving FMT

• 137/191 (72%) response rate (additional 26 deceased and 15 without contact information)

• 2/26 (7.7%) of deceased died of rCDI

• Median time to f/u: 22 months (range, 3-51)

• Most (97%) got PO vancomycin, 53 (39%) also fidaxomicin

24/137 (18%) had rCDI post-FMT

61/137 (45%) got additional antibiotics

• 43/113 (38%) w/o rCDI vs. 18/24 (75%) w/ rCDI (p < 0.01)

2/7/2018Mamo Y et al. Clin Infect Dis. 2017 Dec 19. doi: 10.1093/cid/cix1097. [Epub ahead of print]

UCSF guidelinesClinical definition Criteria Treatment

1st recurrence Except special populations below

Same as for initial therapy, stratified by illness severity

1st recurrence, special population Hematologic cancer with neutropenia expected > 30 daysRecent bone-marrow transplant or treatment for GVHDSolid-organ transplant < 3 mthsOtherwise not an FMT candidate

Fidaxomicin 200 mg po q12h x 10 days(be sure to check insurance coverage before prescribing for outpatients; if insurance does not cover can try the MERCK pt assistance program at www. merckhelps.com)

≥ 2nd recurrence Vancomycin tapered and/or pulsed PLUSEvaluate for FMTConsult ID, GI

http://idmp.ucsf.edu/news/updated-ucsfmcvasfzsfgh-guidelines-c-difficile-infection

2/7/201819

Take-home

Recurrent CDI is a challenge

Treat first episode with same agent, adjust for severity

Subsequently, use vanco taper

Primary FMT indications

• Recurrent or relapsing FMT (usu > 2 episodes)

• Moderate CDI not responding to Rx

• Possibly severe/complicated

Treatment scenario #4: 63 y/o F recently treated for a UTI with levofloxacin, now with profuse diarrhea, T 38.7, BP 79/50, HR 140, WBC 30K, Cr 3.2, and lactate 3.7. What do you treat her with?

A. Vancomycin 125 mg po qid

B. Vancomycin 500 mg po qid

C. Vancomycin 500 mg PR qid

D. Metronidazole 500 mg iv tid

E. Fidaxomicin 200 mg po bid

F. A+C+D

G. B+C+D

2/7/201820

Total colectomy with end ileostomy

Retrospective cohort pts in ICU for CDI

• N = 161 (38 surgery, 123 medical rx)

Indications: Shock (40%), megacolon (29%), no response to med rx (26%), perforation (5%)

aOR death 0.2 (0.1-0.7) colectomy vs. medical rx

• WBC > 50K and lactate > 5 conferred very poor prognosis

• More beneficial in age ≥ 65, immunocompetent, WBC ≥ 20, lactate 2.2-4.9

• 53% died (58% medical rx, 34% surgical)

• Selection bias likelyLamontagne et al., Ann Surg 2007;245(2):267-72.

Diverting loop ileostomy + colonic lavage

3/42 (7%) converted to total colectomy (2 for abd compartment sx)

79% had ileostomy reverted

VS historical colectomy controls, OR for death = 0.24 (0.09-0.63)

• 19% died w/i 30 days

• 14% more died afterwards, all deemed due to underlying illness

RCT recruiting (projected end date 2018)

Neal et al. Ann Surg. 2011 Sep;254(3):423-7; discussion 427-9.

2/7/201821

Multicenter loop ileostomy study

10 centers, 98 patients

• 21% LI patients ; trend towards lower APACHE, less vasopressor use, later surgery

Overall mortality 32% (unadjusted, 34% TC vs 24% LI, p = 0.4)

• Mortality adjusted for confounders, LI 17% vs TC 40%; p = 0.002

• Less blood loss for the LI group

• LI group did worse if reoperation needed

Ferrada P et al. J Trauma Acute Care Surg. 2017 Jul;83(1):36-40

FMT for severe disease

Study Population Intervention Outcome

Cammarota et al, Aliment PharmacolTher 2015

Subgroup of RCT w/ recurrent CDI, N = 7 w/pseudomembranesSingle-center

RCT FMT via colovs vancoInitial 2 pts 1 FMT via colo; remainder FMT q3 days prn

Mortality: 29% (1 FMT)Cure: 71% (≥ 2 FMT)

Fischer et al, Aliment Pharmacol Ther2015

Cohort, N = 29Severe (10) +/-complicated (19)Single-center

FMT via colo ~qwkwith intermittent vanco

Mortality: 7% (both severe/comp)Success: 93% (≥ 2 FMT in 55%)

Zainah H et al. Dig Dis Sci 2015

Cohort, N = 14 with severe, refractory CDI (43% in ICU)Single-center

FMT via NGT, rptat 48-72hr if not response

Mortality: None d/tCDI (29% at 100 dd2/2 underlying dz)Success: 79% (≥ 2 FMT in 21%)

Aroniadis et al. J Clin Gastroenterol2015

Multicenter cohortN = 1776% severe/complicated

FMT mostly via colo

Success: 94% (≥ 2 FMT in 6%)

2/7/201822

FMT for severe disease

Single center retrospective cohort analysis of 111 patients

• FMT group treated with vanco 2-4 days pre- & 4 days post-FMT

• Clinician discretion whether to offer FMT

• 66 (59%) underwent FMT; 45 (41%) did not (incl 26 due to clinical improvement on medical rx and 13 due to instability)

3 month mortality: 12.1% (8/66) in the transplant group vs 42.2% (19/45) in the antibiotic group (OR 0.19 [95% CI, .073–.49])

‒ Multivariable analysis: FMT OR, 0.13 (95% CI, .04–.44)

‒ However, risk of confounding by indication high

‒ No control for systemic antibiotic receipt

2/7/2018Hocquart M et al. Clin Infect Dis. 2017 Aug 24. doi: 10.1093/cid/cix762. [Epub ahead of print]

Take-home for severe, complicated CDI

Use high-dose oral +/- rectal vancomycin

Use IV metronidazole

Consider surgical intervention early

• Consider diverting loop ileostomy

FMT is promising but need more data, multiple FMTs may be needed

• Make sure medical therapy has been optimized

Additional therapies (IVIG, other antibiotics) lack data

2/7/201823

Treatment scenario #5. You are starting your 70 y/o M patient on 4 weeks of ciprofloxacin for prostatitis. He asks you whether he should take probiotics. How do you counsel him?

A. Probiotics will prevent antibiotic-associated diarrhea, including CDI

B. Probiotics will prevent antibiotic-associated diarrhea but not CDI

C. Probiotics are useless

RCT of probiotics for CDI

Diarrhea class Probiotic Placebo OR

AAD 159/1470 (11%) 153/1471 (10%) 1.04 (0.83–1.32)

CDI 12/1470 (0.9%) 17/1471 (1.2%) 0.70 (0.34–1.48)

Allen SJ et al., Lancet 2013 Oct 12;382(9900):1249-57

• No benefit for probiotic• Very low rates of CDI in this population• Majority of patients were receiving

amoxicillin/ampicillin or second-generation cephalosoporins (UK study)

• Likely underpowered for the CDI outcome

2/7/201824

Meta-analysis

Very rigorous

NNT: 43 (95% CI, 36−58)

Remained significant even when missing data was excluded

Studies limiting people to initiation within first 1-2 days on abx had stronger effect size

• Limits UK study (Allen)

Shen NT et al. Gastroenterology. 2017 Jun;152(8):1889-1900.

Approaches to prevent CDI

Gerding D N , Johnson S. CID 2010;51:1306-1313

2/7/201825

Infection control

Gloves + gowns for duration of diarrhea

Wash with soap and water

Private rooms

• Dedicated commode

Bleach cleaning

Antimicrobial stewardship

Cohen et al., Infection Control and Hospital Epidemiology, 2010; 31: 431-455

Identification and isolation of carriers

Longtin Y et al. JAMA Intern Med. 2016;176(6):796-804.

2/7/201826

Non-toxigenic C. diff for secondary prevention

173 patients with 1st or 2nd episode of CDI w/i 28 days (phase II)

• 1-2 days after stopping CDI treatment randomized to non-toxigenic C diff (NTCD-M3) vs. placebo

Recurrence: • OR 0.3; 95% CI, 0.1-0.7• Of NTCD-M3 group, 2% for

those colonized vs. 31% if not colonized

Gerding DN et al., JAMA 2015; 313(17):1719-1727

Secondary prophylaxis?1. Retrospective cohort at two hospitals in Quebec

2. Retrospective cohort St. Louis

Carignan A et al. Am J Gastroenterol. 2016 Dec;111(12):1834-1840. Van Hise NW et al. CID 2016; 63 (5): 651-3

Adult w/ CDIRx’d non-CDI abx within 90 d

(in or outpt)

Recurrence w/i 6 mo

aHR 0.59 (0.43-0.80)

aHR 1st CDI 0.91 (0.57-1.45)

aHR recurrence 0.47 (0.32-0.69)

Adult w/ CDIRx’d non-CDI abx within 90 d

(inpt)

Recurrence w/i 4 weeks

OR 0.12 (0.04-0.4) No multivariate analysis

2/7/201827

Host protection

Kyne et al., NEJM 2000;342(6):390-7. Lowy et al. NEJM 2010 Jan 21;362(3):197-205.

Actoxumab Bezlotoxumab

Monoclonal Abs for secondary preventionMODIFY I and II trials

Wilcox MH et al. N Engl J Med 2017; 376:305-317

• NNT = 10• No clear

subgroup benefited

• Cost may be an issue

• ∆ sustained cure Bezlotuxvs. SOC: 9.7% (4.8-14.5)

2/7/201828

C diff vaccine?

CDI prevention summary

Remember infection control basics

Role of isolation of carriers evolving

Unclear role for probiotics, unlikely to be a game-changer

Non-toxigenic C. diff is promising

Passive immunity is effective but costly

There may be a role for vaccine in the future

Do not forget good infection control and antimicrobial stewardship practices!

2/7/201829

CDI take-home Vancomycin preferred, metronidazole okay for mild-moderate disease

Vancomycin for severe disease

Fidaxomicin may be appropriate for those at high risk for relapse and/or those requiring CA

• But, might not be cost effective

Fulminant disease: PO/PR vancomycin and IV metronidazole

• Consider surgery

• Maybe FMT

1st recurrencesame agent (or fidaxomicin)

2nd and beyondvancomycin pulse and taper

FMT may be the best option for recurrent CDI

Prevention is important

THANK YOU!

2/7/201830