Transcript March2012--CassandraSalgado

Clostridium Difficile: Epidemiology
and Clinical Spectrum
Cassandra D Salgado, MD, MS
Associate Professor of Medicine
Hospital Epidemiologist, MUSC
March 15, 2012
I have no Disclosures to report
relevant to the content of this
presentation.
Cassandra D Salgado, MD, MS
Associate Professor of Medicine
Hospital Epidemiologist, MUSC
March 15, 2012
Case Presentation
• 50 year old female was admitted for
COPD exacerbation. She was treated
with steroids, inhalers, and antibiotics.
• On hospital day 6 she develops fever of
102°F hypotension, diarrhea, and
increased WBC count with profound
bandemia (WBC= 48,000 with 30%
bands).
• Septic workup was done…
Case Presentation
• Patient continued to decompensate
with continued fevers and hypotension
requiring vasopressors and transfer to
MICU
• Bowel movements ceased on hospital
day 8 and patient was evaluated for
toxic megacolon…
Case Presentation: X-ray
Profoundly dilated bowel with wall thickening
Case Presentation
• The patient was taken to surgery…
Case Presentation
• Colectomy was performed
• Patient died on hospital day 9: autopsy revealed
severe disease and pseudomembranous colitis
Clostridium difficile (C. diff)
•C. diff is a spore
forming gram-positive
anaerobic bacillus that
produces at least two
exotoxins (A and B)
and a binary toxin
•The ability to produce
spores allows C. diff to
be acquired from
outside the host
C. difficile: Epidemiology
• “difficult” to isolate in laboratory
• 1978: cause of pseudomembranous colitis
• Most commonly recognized microbial cause of
hospital-acquired diarrhea
• 3-5% of healthy adults are colonized
• 20-40% of hospitalized patients are colonized
• Asymptomatic carriers outnumber patients with
disease by several fold
Bartlett JG. NEJM 1978;298:531. Viscidi R. Gastroenterol 1981;81:5.
McFarland LV. NEJM 1989;320:204.
Clostridium Difficile: Risk Factors
• Exposure to antibiotics the preeminent risk
factor for developing disease
• More than 90% of healthcare-associated CDI
occur while on antibiotics
– Agents active against anaerobic organisms are
considered to present the greatest risk (clindamycin)
– Most cases are associated with β-lactam therapy
– Growing number of reports of associations with
fluoroquinolone therapy
– Duration of antecedent therapy may be brief (surgical
prophylaxis)
C. Diff and Antibiotic Use
• Broad-spectrum agents with a greater effect on the normal
intestinal flora have been associated with CDAD
• Meta-analysis: 49 studies
Pooled Odds Ratio
95%CI
Penicillin
2.0
1.1-4.5
1st Gen Ceph
2.6
1.8-3.7
Quinolones
8.0
4.5-14.3
Clindamycin
9.0
6.3-12.9
Antipseudo PCN
9.6
4.2-21.8
Cefoxitin
10.6
3.0-37.5
Ceftazadime
28.8
12.7-65.1
Cefotaxime
36.2
19.0-68.9
Bignardi GE. J Hops Infect 1998;40:1.
Clostridium Difficile: Risk Factors
•
•
•
•
•
•
•
•
•
Cancer with or without chemotherapy
GI surgery and other types of GI manipulation
Older age
Severity of illness
Immunocompromising conditions
Longer duration of stay in hospital
ICU stay
Exposure to an infected roommate
Anti-ulcer medications
Bignardi GE. J Hops Infect 1998;40:1.
C. diff: Special Populations
• Most cases of CDI occur in healthcare
• Among hospitalized patients
– Medical patients are at increased risk compared to
surgical patients
• C. diff is the most common cause of acute
diarrheal illness in LTCF
– Population is older, receive more medications known
to increase risk of C. diff
• Neonates may also be colonized with C. diff
– Up to 70%
– Neonates colonized with toxogenic strains less likely
than adults to develop symptomatic disease
• Neonates may lack receptors for toxin A in their immature
enterocytes
Sunenshine R. Clev Clinic J Med 2006;73:187.
Clostridium difficile: Clinical Disease
C. Diff exposure (fecal-oral) and colonization
Antibiotic therapy
Disruption of colonic microflora
Release of toxin A (enterotoxin) and toxin B (cytotoxin)
Mucosal injury and inflammation…
Can cause diarrhea, severe GI disease, and sometimes death
Clostridium difficile: Clinical Disease
•
•
•
•
•
Watery diarrhea (10-15 times per day)
Lower abdominal cramping
Low grade fever
Leukocytosis (WBC>15,000)
Symptoms generally occur during
antibiotic therapy
• Sometimes 5-10 days after completion
• Rarely weeks after completion
• Physical exam: abdominal tenderness
CDI vs. AAD
C. diff: Clinical Disease
Effect of Toxin
C. diff: Clinical Disease
• Pseudomembranous colitis: toxin induced shallow
ulcerations which release proteins, mucus, and
inflammatory cells
• Type 1- mildest changes confined to superficial layers
• Type 2- more severe with marked secretions
• Type 3- intense necrosis and full intestinal thickness
involvement produces a confluent membrane
C. diff: Clinical Disease
• Fulminant colitis (rare)
• Severe abdominal pain, abdominal
distention, diarrhea, fever,
hypovolemia, lactic acidosis. Marked
leukocytosis (WBC>40,000)
• Toxic Megacolon (rare)
• Severe abdominal pain, abdominal
distention, scant to no diarrhea, fever,
shock, marked leokocytosis
Clostridium difficile: Clinical Disease
• X-ray: distended
bowel and
colonic wall
thickening
• CT scan:
colonic wall
thickening
and/or
pericolonic
stranding
CDI: Diagnosis
• Cytotoxicity assay (tissue culture assay)
– Gold standard for diagnosis
– Add prepared stool to culture cells
• If toxin present will cause histopathologic changes
• Enzyme immunoassay (EIA)
–
–
–
Direct detection of toxin A or B
Dependent on toxin level
“three samples” rule
• Anaerobic Culture
– Often combined with other
diagnostic methods
• Endoscopy
– Helpful for atypical presentations
CDI: Diagnosis
Test
Time
Sensitivity Advantages Disadvantages
Endoscopy
2 hrs
51%
Anaerobic
culture
72 hrs 89-100%
Tissue
cytotoxin
assay
48 hrs 94-100%
ELISA toxin
A and B
2 hrs
Diagnostic
of p colitis
Low
sensitivity
Molecular
typing
No distinction
of toxin
production
Detects A
(most sensitive) and B
80-95%
Easy to use
False
positives
Decreased
sensitivity for
low level toxin
Decreases PPV and limits clinical management
Sunenshine R. Clev Clinic J Med 2006;73:187.
Planche T. Lancet ID 2008;8:7772008;8:777-784
CDI: Diagnosis
• PCR Based Testing
– Directed towards
toxin B or toxin A
and B
– Sensitivity and
specificity 97-100%
– No “three samples”
– Compares well with
toxinogenic culture
– Considered by
some to be the best
clinical diagnostic
tool
Clostridium Difficile: At What Cost?
• Estimated 3,000,000 new cases of C. diff
diarrhea and colitis in US hospitals
annually
– Affects as many as 10% of patients
hospitalized for more than 2 days
– Relapsing disease occurs in 20-25% of
adequately treated patients
• Further relapses occur in 65% of patients who
have suffered more than one relapse
Clostridium Difficile: At What Cost?
Study
Mortality LOS (excess) Costs
Miller MA. 2002
1.5%
13.6 days
$85,000 per yr
Kyne L. 2002
NR
10.2 (3.6) days
$3,669 per pt
McFarland LV. 1999
0%
NR
$10,970 per pt
Wilcox MH. 1996
NR
46.5 (21.3) days
£4,000 per pt
Riley TV. 1995
NR
24.5 (18) days
*$1,250,000 per yr
Macgowan A. 1995
NR
49.8 (30.2) days
NR
Olson MM. 1994
0.6%
NR
NR
Kofsky P. 1991
NR
NR
$2,000 to $5,000
per pt
*Australian dollars
• Clostridium Difficile
– Spore forming toxin producing bacillus
responsible for most nosocomial diarrhea.
Risks for CDI include antibiotic exposure,
other drugs, older age, and co-morbidities.
CDI has been associated with notable
morbidity, but traditionally low rates of
severe disease and mortality.
C. diff: New Epidemiologic Issues
• C. diff rates are increasing
• More severe disease with higher
mortality and higher rates of colectomy
• A common epidemic strain has been
found in North America and Europe
– Fluoroquinolone use may be driving its
emergence
CDI is Increasing
• Average rate of CDI among NNIS
hospitals from 1987-1998
– 12.2 cases per 10,000 pt-days
• Teaching hospitals: 13.0 per 10,000 pt days
• Non-teaching hospitals: 11.4 per 10,000 pt days
• CDC reports that hospitalizations with a
discharge diagnosis of CDI
– Was 31 per 100,000 in 1996
– Was 61 per 100,000 in 2003
www.CDC.gov
CDI is Increasing
CDI is Increasing
C. diff-Associated Infection
Recent data: Incidence of healthcare-associated C. diff colonization
and infection are 29.5 and 28.1 cases per 10,000 pt days, respectively.
Incidence of CDI per 10,000 patient days
NNIS 1987-2001 Archibald et al JID 2004;189:1585
C. diff: Increased Morbidity and Mortality
• A hospital in Pittsburgh reports 253
nosocomial CDI cases in 2 years with an
increase from 2.7 to 6.8 cases per 1,000
discharges (p<0.001)1
– 26 (10%) colectomies
– 18 (7%) mortality
• A Chicago hospital reports increased
severity of CDI and increased mortality in
MICU and oncology patients2
– high rates of shock
– 37% mortality
1Muto
et al. ICHE 2005;26:273., 2Patel et al. 2005 SHEA Annual Meeting #285.
C. diff: Increased Morbidity and Mortality
• Prospective study of 1719 CDI episodes
in 12 Quebec hospitals
– Determine incidence of CDI and
complications
• 22.5 per 1,000 admissions
– Incidence increased with age
• 30-day attributable mortality was 6.9%
– Attributable mortality increased with age
• 110 (6.5%) required ICU care
• 33 (1.9%) required colectomies
Loo VG et al. NEJM 2005;353(23):2442.
C. diff: Pathogenicity
• In general, C. diff strains either possess the
entire genetic element (PaLoc) responsible
for producing toxin (toxigenic strains) or lack
this genetic element and thus, do not cause
disease (non-toxigenic strains)
C. diff: Pathogenicity
• Mutation(s) in the tcd C gene can cause the C. diff strain to
produce increased amounts of toxin causing more severe
disease.
– 16 x more toxin A and 23 x more toxin B
• A binary toxin may also contribute, but clinical significance
is not yet known.
• This “epidemic strain” is PCR ribotype 027, PFGE type
NAP1, and restriction endonuclease type B1.
States with BI/NAP1/027 strain of C. difficile
Recent data: among patients colonized or infected with
C. diff, 36% and 63% have the NAP1 strain, respectively
DC
HI
AK
PR
CDI: Risk Factors
• Pittsburgh hospital outbreak (MVA)
OR
1.02
95%CI
1.01-1.04
Age (per year)
Diabetes
2.1
1.2-3.6
Organ Transplant
5.8
2.3-14.6
H2 blocker
2.0
1.1-3.5
PPI
2.4
1.3-4.4
Clindamycin
4.8
1.9-12.0
Ceftriaxone
5.4
1.8-15.8
*Levofloxacin
2.0
1.2-3.3
*Levofloxacin use significantly increased over the preceding
months leading up to the outbreak
Muto et al. ICHE 2005;26:273.
CDI: Risk Factors
• Quebec hospitals Outbreak
– Fluoroquinolone receipt OR 3.9 (2.3-6.6)
• All fluoroquinolones studies were independently
associated with CDI
– Cephalosporin receipt OR 3.8 (2.2-6.6)
• All cephalosporins studies were independently associated
with CDI
– PPI, H2 blockers, or chemotherapy were not
significantly associated with CDI
– All isolates with tcdC gene mutations also had the
binary toxin gene
• Severe CDI was observed in 22 of 132 patients with both
genetic elements
• Severe CDI was not observed in any pt that did not have
these genetic elements (p=0.03)
Loo VG et al. NEJM 2005;353(23):2442.
CDI: An Antibiotic-Resistance Issue?
• Pittsburgh hospital outbreak
– No isolates R to metronidazole or vancomycin
– 85.7% R to clindamycin
• Clindamycin was a significant risk
– 94.5% R to levofloxacin
• Levofloxacin was a significant risk
• Quebec hospitals Outbreak
– No isolates R to metronidazole, vancomycin or
clindamycin
• Clindamycin was not a significant risk
– 100% of predominate type isolates R to
ciprofloxacin, moxifloxacin, gatifloxacin, and
levofloxacin
• All FQ were associated with increased risk
Muto et al. ICHE 2005;26:273., Loo VG et al. NEJM 2005;353(23):2442.
CDI and Fluoroquinolones
• Characterization of 187 C. diff isolates from 8
healthcare facilities from 2001-2003 and
comparison to historic isolates
– An epidemic strain was identified and was
responsible for 50% or more of the isolates from
most facilities (binary toxin and tcdC deletions in all)
– This epidemic strain was in existence since 1984
– Compared to similar historic strains where no R to
FQ was detected, all recent strains were R to
gatifloxacin and moxifloxacin
– Conclusion: A previously uncommon strain of C. diff
with variation in toxin genes has become more R to
fluoroquinolones and has emerged as a cause of
geographically dispersed outbreaks of CDI
McDonald LC et al. NEJM 2005;353(23):2433.
C. diff: Community Setting
• 31 yo woman 14 weeks pregnant with twins developed 3 weeks
of intermittent diarrhea- stool specimens positive for C. diff
– Only ABX exposure was TMP-SXT 3 mo prior
– Treated but ultimately developed severe disease hospitalized for
18 days
– Had recurrent disease 4 days after discharge, spontaneously
aborted her fetuses, developed sepsis and died
• CDC and Philadelphia Department of Public Health launched
investigation
C. diff: Community Setting
• 33 cases of CA-CDI were reported from NH, PA,
NJ, OH, all but 1 occurring in 2004-2005
– 15 (46%) required hospitalization or an ED visit
– 13 (39%) had a relapse requiring treatment
– 8 (24%) reported no ABX exposure in prior 3 months
• 3 of these cases had exposure to patient with “diarrheal
illness” (2 confirmed C. diff)
– Of those who had received ABX (25 patients)
• 3 had received <3 doses, 2 had received only 1 dose
• 10 reported exposure to clindamycin
– 10 among peripartum women
• Transmission to close contacts evident for 4 patients
– 23 among non-peripartum individuals
• Ages 6 mo to 72 years (mean 26 years)
C. diff: Community Setting
• The estimated minimum annual
incidence of CA-CDI in Philadelphia
and its surrounding four counties
– 7.6 per 100,000 population
– One case for every 5,549 outpatient ABX
prescriptions
• Twice as high as the <1 case per 10,000 cited in
earlier studies
C. diff: Community Setting
• Conclusions
– These cases of severe CDI among individuals
previously thought to be at low risk might reflect
changing epidemiology
– Certain features of CDI that have been uncommon
in the past might be changing
•
•
•
•
Close-contact transmission
High recurrence rate
Younger age patients
Lack of ABX exposure
– Because reporting was voluntary, the true
incidence of CDI is likely higher
• Clostridium Difficile
– Emerging issues related to CDI include an
increase in overall disease incidence associated
with an epidemic strain that contains a mutation
for increased toxin production and thus more
severe disease with higher risk of death. This
strain has an increased rate of resistance to
certain antimicrobials, particularly
fluoroquinolones, which may be driving the rates.
Additionally, the epidemiology of CDI in the
community, previously thought to be low risk,
may be changing.
CDI: Treatment
• Stop the inciting antibiotic
– Up to 25% will recover without further therapy
• Metronidazole vs. Vancomycin
– Several older retrospective studies comparing oral
metronidazole to oral vancomycin
• Metronidazole just as effective
– Response rates >95%
• Less expensive ($2 day vs. $70 day)
• Less risk for VRE emergence and spread
– A recent prospective observational study
• Response rate to metronidazole was only 78%
– Due to recent emergence of newer strain?
CDI: Treatment
• Metronidazole vs. Vancomycin
– Both drugs have good in vitro activity
• Median MICs <1.0
– Both drugs promote VRE overgrowth in
the stool during treatment
– Relapse rates after treatment are about the
same for both drugs
– Controversy regarding gut levels of
metronidazole
• Only present with active disease?
Wafa N. AAC 2008;52(7):2403-2406.
CDI: Treatment
• Metronidazole vs. Vancomycin
Cure by SOI
Metronidazole Vancomycin P-value
mild
37/41 (90)
39/40 (98)
0.36
severe
29/38 (76)
30/31 (97)
0.02
5/69 (7)
0.27
Relapse Rate 9/66 (14)
Zar FA. CID 2007;45:302-307.
CDI: Treatment
• Metronidazole vs. Vancomycin
Cure by SOI
Metronidazole Vancomycin P-value
mild
26/33 (79)
23/27 (85)
NS
moderate
40/53 (75)
58/73 (80)
NS
severe
37/58 (65)
28/33 (85)
<0.05
27/103 (23)
NS
Relapse Rate 29/100 (29)
Louie T. ICAAC 2007. Abstract K-4259.
CDI: Treatment
• Metronidazole vs. Vancomycin
Metronidazole Vancomycin P-value
LOS (days)
12.8
11.5
<0.001
Mortality
7.9%
6.8%
0.02
ICU stay
23.2
17.7
<0.001
Drug cost
$2439
$2492
NS
$14718
<0.001
Hospital cost $16953
Lahue B. ECCMID 2007. Abstract 1732.
CDI: Treatment
• Appropriate antibiotics directed towards C. diff should
be given for 10 days for mild to moderate disease
– Metronidazole (PO or IV), Vancomycin (PO)
• Most experts recommend starting with
metronidazole
• Clinicians should be vigilant about monitoring
response
– If disease does not progress, should not consider
treatment failure before 6-7 days
– Vancomycin may be more appropriate for severe
disease
• Intraluminal or oral
– Consider surgery if CDI progresses or if severe
CDI: Treatment
• Nitazoxanide vs. Vancomycin (severe disease)
Nitazoxanide
Vancomycin
P-value
EOT
response
8/10 (80)
7/10 (70)
NS
Relapse
1/10 (10)
1/10 (10)
NS
Sustained
response
7/10 (70)
6/10 (60)
NS
Musher D. CID 2009;48:e41-e46.
CDI: Treatment
• Fidaxomicin vs. Vancomycin
Clinical Response EOT
Sustained Response
Dificid
Vanco
Difference
(95% CI)
Dificid
Vanco
Difference
(95% CI)
Trial 1
88%
86%
2.6%
(-2.9-8.0)
70%
57%
12.7%
(4.4-20.9)
Trail 2
88%
87%
1.0%
72%
(-4.8-6.8%)
57%
14.8%
(5.9-23.3%)
Some evidence to suggest that Dificid may be better for non-B1 isolates of
C Diff when compared to vancomycin (redcued recurrence).
Equivalent for B1 isolates.
Package Insert DIFICID 2011.
CDI: Treatment
• Tolevamer (polymer): phase 3- inferior
to vancomycin
• Ramoplanin (antibiotic): phase 3
• Rifaximin (antibiotic): effective when
used with vancomycin for recurrent
disease
• Monoclonal antibody: phase 2
• Vaccine: phase 2
CDI: Recurrent Disease
• Meta-analysis of 12 studies to identify
risks for recurrent disease
– Continued use of non-CDI antibiotics
• OR 4.23 95%CI 2.10-8.55, p<0.001
– Concomitant receipt of antacid medications
• OR 2.15 95%CI 1.13-4.08, p=0.02
– Older age
• OR 1.62 95%CI 1.11-2.36, p=001
Garey KW. JHI 2008;70:298-304.
CDI: Treatment
• No consensus on recurrent disease
– Alternative agents, pulsed or tapering
vancomycin courses
– Ion-exchange resins and polymers
– IVIG: some positive results in case reports
• Do not treat asymptomatic colonization
– Current therapies not effective for this purpose
• Vancomycin effects not sustained and may place pts at
risk for prolonged carriage after treatment
• Can only obtain significant intraluminal levels of
metronidazole in presence of diarrhea
CDI: Alternative Regimens
• Meta-analysis of RR by Type of Probiotic for
Prevention of Antibiotic Associated Diarrhea
#RCT
Saccharomyces
6
Pooled
RR
0.37
Lactobaccillus
6
0.31
95%CI, p-value
0.13-0.72, 0.006
0.26-0.52, <0.0001
• Meta-analysis of 6 RCT for Prevention of CDI
– RR 0.59, 95%CI 0.41- 0.85, p=0.005
– Only Saccharomyces showed significant reduction
for recurrent CDI
McFarland LV. Am J Gastro 2006;101:812.
CDI: Alternative Regimens
• 2 patients died of unrelated causes
• 1 patient had recurrent disease (6.3%)
• 15 had no recurrent disease at 90 day follow-up
C. Diff: Transmission
• Two major potential reservoirs of C. diff in
hospitals are patients and inanimate objects
• Continued pressure from indiscriminate
antibiotic use
– Failure to stop antibiotics when they are no longer
indicated
• Patients with symptomatic intestinal infection
probably contribute most
– Contamination of the environment due to persistence
of highly resistant spores
• Can persist for many weeks or months after the patient has
left the environment
– Personnel hand carriage probably accounts for the
majority of hospital transmission
Original (Incorrect) Hypothesis
for C. diff Hospital Infection
C. Diff acquisition
Hospitalization
Antimicrobials
C. Diff disease
Alternative Hypothesis for C. diff
Hospital Infection
C. Diff acquisition
C. Diff acquisition
Antimicrobials
Asymptomatic
colonization
Many factors associated with
development of disease
Hospitalization
C. Diff disease
Asymptomatic C. diff Colonization
• Data from four prospective studies in
which rectal swabs were obtained
weekly from hospitalized patients:
– Rates of CDI in non-colonized patients was
3.6% compared to 1.3% in colonized
patients (p=0.02)
– Rate also significantly decreased when
only patients who received antibiotics
were analyzed (p=0.02)
Lancet 1998;351:633-636
Clostridium difficile: Control
• Consider two distinct but related
approaches
– Efforts directed at interrupting horizontal
spread (keep patient from acquiring
organism)
– Efforts to minimize the possibility that
exposure will result in infection (reduce
individual risk of disease)
Control Measures: Prevention of CDI
•
•
•
•
•
•
•
Follow antimicrobial usage restrictions
Avoid electronic thermometers
Dedicate patient care items
Use full barrier precautions for CDI pts
Place CDI pts in private room
Perform appropriate hand hygiene
Perform adequate environmental
decontamination (bleach)
• Educate HCWs regarding CDI
SHEA/IDSA Practice Recommendations. ICHE 2008;29:Suppl 1.
Hand Hygiene and C. diff Risk
• Hand washing with soap or chlorhexidine is
equally effective in removing spores of C. diff
from the seeded hands of volunteers
• Alcohol hand hygiene products are not
sporicidal and not likely to reduce the burden
of spores on hands when used without hand
washing
• CDI outbreaks have NOT been directly
related to use of alcohol hand hygiene
products
Control Measures: Prevention of CDI
• Approaches that should not be
considered for routine prevention
– Testing of asymptomatic patients
– Repeat testing of CDI pts after successful
therapy
• Unresolved issues
– Use of contact precautions by family members
– Standing nursing orders for testing
– Restricting gastric acid lowering medications
SHEA/IDSA Practice Recommendations. ICHE 2008;29:Suppl 1.
Prevention of C. diff Hospital
Infection
Antimicrobials
C. Diff acquisition
Asymptomatic
colonization
Give non-toxigenic C. diff
Hospitalization
C. Diff disease
• CDI: Treatment and Control
– Effective treatment regimens have been
developed for mild to moderate disease which
include metronidazole, vancomycin, and
fidaxomicin. Severe disease may require
alternative uses of traditional drug regimens and
possibly adjunctive agents. The transmissibility of
C. diff has been well described in healthcare
facilities and control measures have been
recommended. Novel prevention methods are
being considered.
Summary
• Clostridium difficile remains an important
nosocomial pathogen
• The epidemiology of Clostridium difficile is
changing
– Increased rates in acute care and long term care facilities
– Increased morbidity and mortality
– Possibly due to a genetically altered strain with increased
virulence
• Effective treatment regimens exist
– Study of new treatment regimens should continue
• Effective control measures have been
described
– Study of new or enhanced control measures should continue
Questions…