View
5
Download
0
Category
Preview:
Citation preview
PREVENTION AND TREATMENT OF BACTERIAL INFECTIONS IN CIRRHOSIS
Dr. J. Fernández. Head of the Liver Unit
Hospital Clinic Barcelona, SpainAEEH Postgraduate Course,Madrid, February 15 2017
Prevalence of infection in the ICU
Fernández et al. Hepatology 2002 and 2012;; Gustot et al. Liver Int 2014
25-35%
PREVALENCE AT THE HOSPITAL ICU PREVALENCE
59%
Clinical impact of bacterial infections in cirrhosis
0
10
20
30
40
50
60
Before 2000
After 2000
MORTALITY AT 1 MONTH
Fernández et al. Hepatology 2002 and 2012;; Arvaniti et al. Gastroenterology 2010
Infections increase mortality 4-fold
%
Prevention of bacterial infections in cirrhosis
Antibiotic prophylaxis. Current indications in cirrhosis
Jalan, Fernandez. J Hepatol 2014
Antibiotic prophylaxis for the prevention of SBP in patients with cirrhosis and hemorrhage
Bernard et al. Hepatology 1999
Soriano 92
Blaise 94
Pauwels 96
Hsieh 98
Total
0.2 0.4Risk difference
0.0736, p=0.0060
-0.1 0.30.0 0.1
Antibiotic prophylaxis in cirrhotic patients with hemorrhage. Effects on survival
Bernard et al. Hepatology 1999
Rimola 85
Soriano 92
Blaise 94
Total
0.0 0.2 0.4Risk difference
0.091, p=0.004
-0.1 0.1 0.3
Hsieh 98
Pauwels 96
Bacterial infection and failure to control bleeding
0
20
40
60
80
100
Proven bacterial infection
Failure to controlbleeding
No
%
Goulis et al. Hepatology 1998
p<0.01
Antibiotic prophylaxisand rebleeding
0
10
20
30
40
50
Rebleeding
Prophylaxis
No
%
Hou et al. Hepatology 2004
p<0.01
Ceftriaxone vs. norfloxacin in the prevention of infections in UGB in advanced cirrhosis*
Norfloxacin
Ceftriaxone
p=0.03
Prob
abili
ty of
rem
aini
ng
free
of i
nfec
tion
days1086420
1,0
,9
,8
,7
,6
,5
,4
,3
,2
,1
Fernández et al. Gastroenterology 2006
*At least two of the following: ascites, severe malnutrition,encephalopathy or bilirubin > 3 mg/dl
Secondary prophylaxis of SBP
2420161284
1.0
0.8
0.6
0.4
0.2
0.0
Norfloxacin (n=40)
Placebo (n=40)
p=0.006
Months
Probability of SBP
0
Gines et al. Hepatology 1990
Primary prophylaxis of SBP in patients with advanced cirrhosis*
2420161284
1.0
0.8
0.6
0.4
0.2
0.0
Norfloxacin (n=35)
Placebo (n=33)
p=0.0007
Months
Probability of SBP
0
Fernández et al. Gastroenterology 2007*
*Ascitic fluid protein <15 g/L and serum creatinine ≥1.2 mg/dL or BUN ≥25 mg/dL or serum sodium ≤130 mEq/L or Child-Pugh score ≥ 9 points with serum bilirubin ≥ 3 mg/dL
Probability of type-1 HRS
2420161284
1.0
0.8
0.6
0.4
0.2
0.0Norfloxacin (n=35)
Placebo (n=33)
p=0.005
Months
Probability of type-1 HRS
0
Fernández et al. Gastroenterology 2007
Probability of short-term survival
p=0.001
321
1.0
0.8
0.6
0.4
0.2
0.0
Months
Probability
0
3-MONTH SURVIVAL
Norfloxacin (n=35)
Placebo (n=33)
Norfloxacin (n=35)
Placebo (n=33)
654321
1.0
0.8
0.6
0.4
0.2
0.0
Months
Probability
0
6-MONTH SURVIVAL
p=0.014
Fernández et al. Gastroenterology 2007
PREVENTION OF SBP 1-YEAR PROBABILITY OF SURVIVAL
Primary prophylaxis of SBP in cirrhotic patients with low protein ascites
Terg et al. J Hepatol 2008
Months
Placebo
Norfloxacin
Cum
ulative Incicenceof Death
Impact on survival of primary prophylaxis with norfloxacin in Child-Pugh C
Moreau et al. AASLD 2016
Probability of infection
Days of follow-up1 2 3-7 8-14 15-21 22-28
P-value < 0.001
Probability of bacterial infectionsin patients with ACLF and AD
Fernandez et al. Submitted
Outline
ü Current indications of antibiotic prophylaxis
ü Drawbacks of antibiotic prophylaxis
ü Potential alternatives
Definition of multiresistant (MDR) bacteria
ü MDR: Strains resistant to ≥ 3 of the main antibiotic families including β-lactams or to at least 1 agent in 3 or more families: difficult to treat bacteria.
ü The most frequent are: extended-spectrum β-lactamase-producing Enterobacteriaceae (ESBL), Pseudomonas aeruginosa, Acinetobacter baumanii, MRSA, VSE and VRE.
Fernandez and Berg et al. J Hepatol 2016
Increasing prevalence of MDR bacteria over time
0
20
40
1998-2000 2005-2007 2010-2011
%1998-2000
2005-2007
2010-2011
Fernandez et al. Hepatology 2002 and 2012
* p<0.05 vs. other periods
10%
18%23%
*
0
20
40
60
80
100
Community-
acquired
Health care-
associated
Nosocomial
%
Fernandez et al. Hepatology 2012
N=507p=0.001
Prevalence of MDR bacteria according tothe site of acquisition of infection
2005-2007
4%14%
35%
0
20
40
60
80
100
Community-
acquired
Health care-
associated
Nosocomial
% Community-acquired
Health care-associated
Nosocomial
2010-2011
N=162p=0.002
20%
39%
§ Nosocomial infection 4.43 <0.0001§ Long-term norfloxacin prophylaxis 2.69 0.004§ Use of beta-lactams (last 3 months) 2.39 0.02§ Infection by MDR bacteria (last 6 months) 2.45 0.04
HR p
Risk factors for infections caused by MDR bacteria
Fernández et al., Hepatology 2012
Ariza et al., J Hepatol 2012
§ Nosocomial infection 2.54 § Previous use of cephalosporins 2.98§ Upper gastrointestinal bleeding 8.15§ Diabetes mellitus 2.52
HR*
* 3rd generation cephalosporin-resistant bacteria in SBP
Geographic distribution of antibiotic-resistant bacteria in cirrhosis
QUINOLONE-RESISTANCE TGC-RESISTANCE
Fernandez and Berg et al. J Hepatol 2016
SBP prophylaxis and CDI
Bajaj et al. Am J Gastroenterol 2010
0
5
10
15
20
25
30
35
40
SBP
prophylaxis
No
Prevalence of Clostridium difficile infection
P=0.01
Potential alternatives in the prevention of SBP in cirrhosis
Lutz et al. Plos One 2014, Kedarisetty et al. Gastroenterology 2015Ubeda et al. J Hepatol 2016
RIFAXIMIN COLONY-STIMULATING FACTORS OBETICHOLIC ACID
New antibiotic strategiesfor bacterial infections in cirrhosis
Severe baseline circulatorydysfunction
Rapid deterioration of cardiovascular function and organ blood perfusion/higher organ damage
Kidneys Other organs and systems
Renal failure
Relative adrenal insufficiency
Liver
Gut
Brain
Increased translocation of viable bacteria and endotoxin
Encephalopathy
Jaundice, coagulopathy, encephalopathy
Adrenal glands
Excessive inflammatory response to SBP
Pathophysiological basis ofmultiorgan failure in cirrhosis
Arroyo et al. Nat Rev Nephrol 2011
Impact of cirrhosis on the outcome of infected patients in the ICU
Gustot et al. Liver Int 2014
Mortality higher than that observed in the general population: 30-57% vs. 50-100%
Early antibiotic treatment“The golden hour concept” in the cirrhotic population
In patients with cirrhosis and septic shock, inappropriate and delayed appropriate initial empiric antimicrobialtherapy is associated with increased mortality.
Arabi et al. Hepatology 2012
Merli et al. Plos One 2015
MDR-XDR-PDR bacterial infections in cirrhosis
N=124 bacterial infections (51% resistant-strains)
0
20
40
60
80
100
MDR XDR PDR
%
Fernández et al., Hepatology 2012
Infections by MDR bacteria.Resolution of infection and outcome
Resolution* (%)100
80
60
40
20
0
92%
YesMDR bacteriaNo
p<0.0001
70%
* After modification of antibiotic theraphy
Septic shock (%)100
80
60
40
20
0YesNo
MDR bacteria
p<0.0001
26%10%
Fernández et al., Hepatology 2012
Infections by MDR bacteria.Clinical outcome
Hospital mortality (%)100
80
60
40
20
0YesNo
MDR bacteria
p=0.001
25%12%
Fernández et al., Hepatology 2012
Nosocomial infections in cirrhosis.Failure of the guidelines based on the use
of third-generation cephalosporins
Resolution withouttreatment modification (%)
100
80
60
40
20
0NosocomialCommunity
-acquired
40%
83%
SBPUTIS. bacteremiaCellulitisPneumoniaOther
78%90%67%82%67%91%
CA
26%29%18%50%44%65%
Nosocomial
<0.001<0.0010.05nsns0.005
p
Fernández et al., Hepatology 2012
Healthcare-associated infections in cirrhosis.Efficacy of third-generation cephalosporins
Resolution withouttreatment modification (%)
100
80
60
40
20
0Health
care-associatedCommunity-acquired
73%83%
SBPUTIS. bacteremiaCellulitisPneumoniaOther
78%90%67%82%68%91%
CA
71%59%60%81%33%91%
HCA
ns0.02nsnsnsns
pp=0.05
Problems and solutions in the history ofSBP and other infections in cirrhosis
- Low efficacy of treatment
- High rate of recurrence
- High prevalence of HRS
- High prevalence ofMDR bacteria
Problem Solution
3rd-generation cephalosporins
Antibiotic prophylaxis
Albumin infusionPreventive measures & modification of antibiotic guidelines
Fernández and Gustot, J Hepatol 2012
Spontaneous bacterial infection(SBP, SBE or SB)
Community-‐acquired infection
Health-‐care associated infection
Nosocomial infection
Presence of ≥ 2 risk factors for multiresistant bacteria* or severe sepsis or shock?
NO YES
Third-‐generation cephalosporins
Carbapenem + glycopeptide**
Acevedo et al., Hepatol Int 2012
Proposed algorithm for the empirical treatment ofspontaneous infections in cirrhosis
Recommended empirical antibiotic therapy
Jalan and Fernández et al. J Hepatol 2013
Impact of new empirical antibiotic strategies
HCA INFECTIONSB-lactams vs Imipenem+GPC
NOSOCOMIAL SBPCefepime vs meropenem+daptomycin
RESOLUTION RATE: Cefepime: 25%
Meropenem+daptomycin: 87%Piano et al, Hepatogy 2015, Jindal et al. Liver int 2015
0
20
40
60
80
100
SMT Broad-spectrum
%
Treatment failure
Hospitalmortaility
p< 0.05
Algorithm of empirical antibiotic treatment in ICU Severe sepsis or shock? orAPACHE II > 15 or SOFA score > 8?
Risk factors of MDR bacteria?1
Meropenem6
+daptomycin, linezolid orvancomycin5
+(ciprofloxacin, amikacin and/orcolistin)7
+(echinocandine)8
NO
YES
1. Risk factors of MDR bacteria : a) previous colonization; b) Antibiotic treatment > 5 days in the last 3 months; c) Hospitalization > 5 days in the last 3 months, and d) Nursing-home
NO YES
Ertapenem+
daptomycinlinezolid orvancomycin6
Cefotaxime or ceftriaxone4
4. Plus azitromicin 500 mg/day, 3 days in pneumonia or change to fluoroquinolones
6. If previous treatment with carbapenem (6 weeks) start with piperacillin-tazobactam or ceftazidime + tigecyclin
5. Daptomycin in infections with high-risk of bacteremia (catheter, IE). Linezolid in pneumonía, cellulitis or CNS infection. Vancomicyn insteed of daptomycin or linezolid if GFR >60 mL/min, no other nephrotoxic drug and no use in the previous month
7. Consider to add depending on local epidemiology, recent antibiotic treatments and source of infection8. Add an echinocandin if 2 or more of the following criteria: a/ colonization by Candida spp (candiduria or rectal swab) and antibiotic treatment or steroids, b/ parenteral nutrition, c/ Gastro-duodenal surgery or necrohemorragic pancreatitis, d/ renal replacement therapy
High risk source of infection?2
or CRP > 10 mg/dL3?
2. High-risk infection: pneumonia, peritonitis (high bacterial load) or with high-risk of severe complications (meningitis).
NO
3. CRP levels correlate with bacterial load
Fernandez et al, Hepatogy 2015
cefotaxime 6-8 g / d in continuous infusion2 g 1-2 g / 8 h
Initial dose1 Doses and way of administration In the first 48 h
ceftazidimemeropenem 2 g 1-2 g / 8 h6 g / d in continuous infusion
“de-escalation”at 72h
ceftriaxone 2 g 1 g / 12-24 h1 g / 12 h
piperacilin-(tazobactam) 4-(0,5) g 4,5 g / 6-8 h16 g / d in continuous infusion
levofloxacin 500 mg / 12 h 1000 mg 500 mg / 24 h
ciprofloxacin 600 mg 400 mg / 8-12 h400 mg / 8 h
tigecycline 100 mg / 12 h 200 mg 50-100 mg / 12 h
fosfomycin 200-300 mg / kg / d in continuous infusion4 g 2 g / 6 h
metronidazole 500 mg / 6 h 1000-1500 mg 500 mg / 6-8 h
1.The first dose is independent of the renal function
Doses and patterns of IV administrationof the main antibiotics in severe sepsis or shock
Fernandez et al, Hepatogy 2015
Re-evaluation of treatment at 48-72 h
Positive cultures Negative or no cultures
Adjust antibiotic treatment to the results of the cultures:monotherapy if possible
(decrease in CRP levels)
Maintain treatment for 5 days if no source of infection and for 7 days in the rest
(increase in CRP levels)
New samples, change of catheters, image techniques, new antibiotic schedules
Good evolution Bad evolution
Fernandez et al, Hepatogy 2015
Key points and conclusions
ü Antibiotic prophylaxis must be restricted to high-risk groups to decrease the likelihood of emergence of antibiotic resistance.
ü New RCT are needed to assess potential alternatives to classical antibiotics, mainly rifaximin, obeticholic acid and G-CSF
Key points and conclusions
ü MDR bacteria are frequently isolated in nosocomial and to a lesser extend HCA infections in cirrhosis.
ü They are associated with poorer prognosis and higher mortality rate.
ü Third-generation cephalosporins continue to be the gold-standard antibiotic treatment of many of the infections acquired in the community.
Key points and conclusions
ü Empirical treatment of nosocomial and possibly some health care-associated infections should be adapted to the local epidemiological pattern of antibiotic resistance.
ü Patients with severe sepsis or shock should receive broad spectrum antibiotic regimens, with optimized IV administration (high doses/continuous infusion)
ü Empirical antibiotic regimen should be rapidly narrowed and shortened to decrease the risk of emergence of antibiotic resistance
Recommended