Transcript Title of Presentation Myriad Pro, Bold, Shadow, 28pt

Epidemiology and Prevention of
Catheter-Related Bloodstream Infections in
Outpatient Settings
Alice Guh, MD, MPH
Division of Healthcare Quality Promotion
Centers for Disease Control and Prevention
November 3, 2012
National Center for Emerging and Zoonotic Infectious Diseases
Division of Healthcare Quality Promotion
Overview

Epidemiology of catheter-related bloodstream
infections (CRBSI) in outpatient settings

Core and supplemental measures for CRBSI
prevention with focus on outpatient settings
Burden of Central Venous Catheter Use

>5 million CVCs inserted in US annually
 93% of patients receiving home infusion therapy, compared to
13% of hospitalized patients

Approximately 25-30% of all hemodialysis patients
 80% initiate hemodialysis with CVC

Estimated 2/3 of cancer patients use long-term CVC
Moureau N et al. J Vasc Interv Radiol 2002;13:1009-1016.
Herbst S et al. Infusion 1998;4(suppl):S1-132.
Kallen A et al. Clin Infect Dis 2010;51(3):335-341.
Van de Wetering MD et al. Cochrane Database Syst Rev 2007.
Outpatient Central Venous Catheter Use

Diverse indications






Hemodialysis
Chemotherapy administration
Intravenous antimicrobial therapy
Parenteral nutrition, intravenous fluids
Treatment of pulmonary hypertension
Various outpatient settings
 Physician offices, clinics, infusion centers, home settings (selfcare, home healthcare agencies)
Overall Burden of Catheter-Related
Bloodstream Infections (CRBSI)

~250,000 BSI cases in US annually
 Majority associated with central venous catheter
 Higher costs, crude mortality rates, and number of hospital-days

Acute care settings (2009 NHSN data)
 Medical-surgical wards: 1.2 cases / 1000 catheter-days
 Medical-surgical ICU: 1.5-2.1 cases / 1000 catheter-days

Outpatient settings
 Variable rates
 Multiple factors: patient comorbidities, catheter type, CVC
indication
Klevens RM et al. Public Health Rep 2007;122:160-166.
Edwards JR et al. Am J Infect Control 2009;37:783-805.
Outpatient CRBSI Rates

>50,000 patients receiving home infusion
 Retrospectively collected data from 37 US states
 0.19 cases per 1000 catheter-days
• Highest in tunneled (0.34) and nontunneled (0.22) catheters
• Lowest in midline catheters (0.09) and PICCs (0.11)
• Ports and midline had lowest combined local and BSI rates (0.3)

827 patients receiving outpatient/home infusion
 Two study sites, prospective evaluation
 0.99 cases per 1000 catheter-days
• Nonsignificantly higher risk in centrally inserted catheter versus
ports
Moureau et al. J Vasc Interv Radiol 2002;13:1009-1016.
Tokars J et al. Ann Intern Med. 1999;131:340-347.
Outpatient CRBSI Rates by Population Type

Outpatient hemodialysis facilities (NHSN data)
 4.2 BSI cases per 100 pt-months (1.4 cases per 1000 catheterdays)

Cancer patients
 Studies of mostly adults: 1.0 to 2.1 per 1000 catheter-days
 Outpatient pediatric studies: 0.1 to 7.4 per 1000 catheter-days

One study: home parenteral nutrition (n=53)
 2.5 cases of “line sepsis” per 1000 catheter-days
• Adults 0.8 /1000 catheter-days vs children 6.9 / 1000 catheter-days
Klevens RM et al. Semin Dial 2008;21:24-28.
Howell PB et al. Cancer 1995;75:1367-75.
Groeger Js et al. Ann Inter Med 1993;119:1168-1174.
Barrell C et al. AJIC 2012;40:434-439.
Gillanders L et al. Clin Nutr. 2012;31:30-34.
Pathogens Associated with Outpatient CRBSIs

Varies by patient population type

Gram-positive organisms most common
 Coagulase-negative Staphylococci – 28-60%

Increasing infections by gram-negative organisms
 Pediatric oncology and HSCT patients
• Nonendogenous organisms during summer months

Higher risk for Candida infections in long-term
parental nutrition population

20-37% polymicrobial infections
Tokars J et al. Ann Intern Med. 1999;131:340-347.
Opilla M. AJIC. 2008;36(10):S173.e5-8.
Smith T et al. Infect Control Hosp Epidmiol 2002;23:239-243.
Barrell C et al. AJIC 2012:40:434-439.
Polymicrobial BSIs among Pediatric
Outpatients with CVCs
Downes KJ et al. Clin Infect Dis. 2008;46:387-394
General Risk Factors for Outpatient CRBSIs

Prospective study: 827 patients receiving home
infusions
Tokars J et al. Ann Intern Med. 1999;131:340-347.
CRBSI PREVENTION IN
OUTPATIENT SETTINGS
Limitations of Current Recommendations

Based on studies conducted in ICU settings

Prevention of outpatient CRBSIs largely focused on
hemodialysis patients
Surveillance

Outbreak detection

Staff feedback to improve performance

Collection of outcome and/or process measures
 CRBSI rates
 Adherence to hand hygiene
Value of Surveillance

Busy London dialysis unit: 112 patients

Implemented CDC dialysis surveillance; described
their experience over 18 months

After initial set up, required 2 hours per month

Outcomes: Reductions in
 Access-related bacteremia
 Antibiotic usage
 Hospital admissions
George A et al. BMJ 2006; 332:1435-1439
Slide courtesy of Dr. Priti Patel
Antimicrobial Starts
George A et al. BMJ 2006; 332:1435-1439
Slide courtesy of Dr. Priti Patel
Access-Related Bacteremia
George A et al. BMJ 2006; 332:1435-1439
Slide courtesy of Dr. Priti Patel
Observations

“Surveillance raised awareness and provided a
cornerstone for improved infection control and line
care involving all staff of the dialysis unit.”

“The data feedback generated unit led programmes
of risk reduction and infection control.”
George A et al. BMJ 2006; 332:1435-1439
Slide courtesy of Dr. Priti Patel
Challenges Related to Surveillance

Challenges of measuring outpatient CRBSI rates

No established surveillance system for all outpatient
settings

Determining infections originating in outpatient
facility or related to home infusion

Collecting appropriate denominator data, e.g.,
catheter days
Considerations for CRBSI Surveillance in
Certain Outpatient Settings

Hematology/oncology patients with long-term CVC
 Tracking positive blood culture results: laboratory notification,
ask patients at visits
 Denominator data: total number of line days/month

Building line days database
 Determine list of patients with CVCs
 Designate personnel to build and
maintain database
 Collect initial and subsequent data
• Nurses reporting, monthly surgery
list of lines placed/removed
Children’s Hospital Association Heme-Onc Collaborative for Prevention of CLABSI.
Pathogenesis: Mechanism for Colonization of
Longer-term CVC

Intraluminal pathway most common for CVC >1 week
 Contamination of the hub, catheter, or other administration
device

Presence of biofilm greater on luminal surface in
CVC >30 days

Emphasis on appropriate CVC maintenance and
access practices
Raad I et al. J Infect Dis 1993;168:400-407.
Safdar N et al. Intensive Care med 2004; 30:62-67.
Removal of Unnecessary CVC

Important component of interventions to decrease
CRBSIs

Multisite studies of ICU settings
 Implementation of multifaceted interventions led to significant
decrease in CRBSI rates
 Interventions included: asking providers daily whether catheters
can be removed
• Added to rounding form: “daily goals form”
Brenholz SM et al. Crit Care Med 2004;32:2014-2020.
Pronovost P et al. N Engl J Med 2006;355:2725-2732.
Hand Hygiene

Perform hand hygiene before and after:
 Palpating catheter insertion sites
 Changing dressing of catheter site
 Accessing catheter

Outpatient facilities
 Ensure easy access to alcohol-based hand rub and/or soap and
water
 Observation of practices and “just in time” feedback as needed
Skin Antisepsis for Cleansing Catheter SIte

Prospective, randomized trial (n=668 catheters: CVC,
arterial catheters)
Maki D et al. Lancet 1991;388:339-343.
Meta-analysis: Comparison of Chlorhexidine
and Povidone-Iodine Solution

Vascular
catheter-site
care

4143 catheters
(various types)

All hospital
settings
Chaiyakunapruk N et al. Ann Intern Med 2002;136:792-801.
Meta-analysis: Results
Chlorhexidine gluconate reduced catheter-related BSI by approximately
50% (summary risk ratio 0.49 [95% CI, 0.28-0.88])
Similar findings when only CVCs were included in the
analysis (summary risk ratio, 0.51 [95% CI, 0.27-0.97])
Chaiyakunapruk N et al. Ann Intern Med 2002;136:792-801.
Why Chlorhexidine Gluconate May be Better
Antisepsis Than Povidone-Iodine

Microbicidal effect might not be affected by proteinrich biomaterials (e.g., blood, serum)

Prolonged residual effect (at least 6 hours)

Superior bactericidal effect against coagulasenegative staphylococci
 Disinfection of peritoneal dialysis catheter sites
Chaiyakunapruk N et al. Ann Intern Med 2002;136:792-801.
Shelton DM. Adv Perit Dial. 1991;7:120-4.
Catheter Site Dressing: Gauze and Tape vs
Transparent Polyurethane Dressing

Study of peripheral catheters (n=2000 catheters)
 No difference in rate of catheter colonization or phlebitis

Systematic review and meta-analysis:
 8 of 23 studies included; data available from only 6 studies
• No difference in incidence of infectious complications (catheterrelated sepsis, exit site infection)
 Updated review in 2011
• higher CRBSI rate with polyurethane dressing, but small sample
size with low quality evidence

Use either sterile gauze or sterile, transparent,
semipermeable dressing to cover catheter site
Maki DG et al. JAMA 1987;258:2396-403.
Gillies D et al. J Adv Nurs 2003;44:623-32.
Gillies D et al. Cochrane Database Syst Rev 2011;9:CD003827.
Catheter Site Dressing Changes

Wear clean or sterile gloves

Replace dressing if becomes damp, loosened, or
visibly soiled

Remove dressing to allow examination if:
 Tenderness at insertion site
 Other symptoms suggestion of local infection or BSI

Do not use topical antibiotic ointment or creams on
insertion site (except for dialysis catheters)
 Potential for fungal infections
 Antimicrobial resistance
Needleless Connectors

Catheter hub is important portal of entry

Needleless connectors evolved from split septum to
mechanical valves

Potential decreased microbial contamination rate
compared to stopcocks/caps

Randomized controlled trial in ICU
 243 patients, mean CVC duration 9.9 days
 CVC with needleless connectors vs 3-way stopcock/cap
 CRBSI incidence significantly reduced with needleless
connectors
1000
catheter days vs 5.0 / 1000 catheter days)
Caeey AL et al.(0.7
J Hosp/Infect
2003l54:288-93.
Bouza E et al. J Hosp Infect 2003;54:279-87.
Yebenes JC et al. Am J Infect Control 2004;32:291-5.
Importance of Access Port / Connector
Disinfection

Appropriate disinfection must be performed

Experimental model evaluating barrier effect of 3
different needleless connectors
 Peripheral catheter with connector inserted in blood culture
bottle
 Contaminated external surfaces of connectors with different
concentrations of S. epidermidis
 Assigned to “correct cleaning group” (70% alcohol before
handling) vs. control group (no disinfection before handling)
 Incorrect handling reduced sterility from 94.4 to 66.7% (p=0.001)
Yebenes JC et al. Crit Care Med 2008;36:2558-61.
Disinfection Procedure for Connectors

Mixed findings regarding alcohol vs chlorhexidine
disinfectants
 Earlier study showing ethanol-based disinfectants most effective
 Recent studies: higher microbial contamination following alcohol
(69%) than chlorhexidine (30.8%) or povidone-iodine (25%)
 Role of antimicrobial impregnated connector

Wiping with 70% alcohol for 3-5 sec not effective

No difference when vigorously scrubbing 15 sec with
alcohol or chlorhexidine
 In vitro study of various mechanical valves
Salzman MB et al. J Clin Microbil 1993;31:475.
Casey AL et al. J Hosp Infect 2003;54:288-293
Menyhay SZ et al. Infect Control Hosp Epidemiol 2006;27:23-7.
Kaler W et al.JAVA 2007;12:3-9.
CRBSI Associated with Mechanical Valves?

Several reports of increased CRBSI when switching
from split septum to mechanical valves
 Acute care settings
 Large, multicenter study across 5 hospitals (16 ICUs, 1 entire
hospital, 1 oncology unit)

Pediatric hematology/oncology patients receiving
home infusion
 182 patients, >75,000 catheter days
 CRBSI significantly increased when mechanical valves
introduced (0.8 to 1.4 / 1000 catheter days)
Mechanical Valves in Long-Term Acute
Care Setting
CRBSI increased from 1.79 to 5.9 / 1000 catheter days
Salgado CD et al. Infect Control Hosp Epidemiol 2007;28:684-688.
Potential Explanations for Increased CRBSI
with Mechanical Valves

Device-specific vs all mechanical valves?

Improper cleaning of connector surface (difficulty in
adequate disinfection)
 Recommendations may differ by device type

Fluid flow properties and inadequate flushing (poor
visualization in opaque devices)

Exposure to blood/nutritional fluids enable biofilm
formation

Presence of internal corrugations could harbor
organisms
Recommendations for Disinfecting
Access Port / Connectors

Scrub access port / connects with appropriate
antiseptic
 Chlorhexidine, povidione-iodine, 70% alchohol

Access port with only sterile devices

Split septum may be preferred over some
mechanical valves
 Must follow manufacturer recommendations for disinfection
when using mechanical valves
Outbreak Related to Unsafe Injection Practices

Outpatient pediatric bone marrow transplant clinic

September 2007: Initially 6 patients with CVC had
BSI, some polymicrobial

Surveillance blood cultures during outbreak period
(n=30 patients)
 13 patients with BSI, 17 without BSI

Cohort study looking at risk factors

Infection control assessment, including saline flush
preparation
Wiersma P et al. Infect Control Hosp Epidemiol 2010;31:522-27.
Cohort Study Results
Wiersma P et al. Infect Control Hosp Epidemiol 2010;31:522-27.
Cohort Study Results
Wiersma P et al. Infect Control Hosp Epidemiol 2010;31:522-27.
Infection Control Assessment:
Saline Flush Preparation

Outside of automated medication supply

Prepared predrawn saline and heparin syringes:
 Preservative-free, single-dose 50-mL saline vial
 Multidose 10-mL heparin vials

Vials accessed multiple times

Predrawn syringes and vials not dated

Outbreak likely due to extrinsic
contamination of saline vials
Wiersma P et al. Infect Hosp Control Epidemiol 2010;31:522-27.
Recommendations for Safe Injection Practices

Injection safety refers to proper use and handling of
supplies for administering injections and infusions
 Syringes, needles, IV tubing, vials and parenteral solutions

Key injection safety recommendations include:
 Dedicate single dose vials for single patient use
 Always use new syringe and needle to access medication vials
 Avoid prefilling and storing batch-prepared syringes (outside of
pharmacy setting)
 Whenever possible, use commercially manufactured or
pharmacy-prepared prefilled syringes (saline, heparin)
Education

Education of healthcare personnel
 Proper care/maintenance of catheter
 Periodically assess adherence to recommended practices

Education of patients
 Do not submerge catheter or catheter site in water
 Report any changes in catheter site or new discomfort
SUPPLEMENTAL CRBSI
PREVENTION MEASURES
Supplemental Measures

Chlorhexidine-impregnated sponge dressings

Antimicrobial / antiseptic impregnated catheters

Antimicrobial / antiseptic catheter locks
Chlorhexidine-Impregnated Sponge Dressings
vs Standard Dressings

Largest multicenter, randomized controlled trial

7 ICUs (mix of medical and surgical) across
academic and community hospitals

Included 1636 adult patients (n=3778 catheters)

Interventions included:
 CHGIS dressing applied to entire insertion site under
semitransparent dressing (controls: only semitransparent
dressing)
 Of note: Alchohol based povodone-iodine was used for
antisepsis

Outcomes included catheter-related infection (BSIs)
rates, catheter colonization
Timsit JF et al. JAMA 2009;301:1231-41.
Results of CHGIS vs Standard Dressings:
Cumulative Risk of Catheter-Related Infections
Major catheter-related
infection rate:
1.4 / 1000 catheter-days to
0.6 / 1000 catheter-days
Catheter-related BSI rate*:
1.3 / 1000 catheter-days to
0.4 / 1000 catheter-days
Catheter colonization rate:
15.8 / 1000 catheter-days to
6.8 / 1000 catheter-days
Timsit JF et al. JAMA 2009;301:1231-41.
Additional Findings From Same Study

Significant decrease in bacterial skin colonization
with CHGIS dressings
 Not associated with greater resistance of bacteria

Severe contact dermatitis leading to removal of
CHGIS: 8 patients (10.4 / 1000 patients)
 No systemic adverse reactions to chlorhexidine occurred

No difference in catheter colonization between
dressing change at 3 days vs 7 days
Timsit JF et al. JAMA 2009;301:1231-41.
Chlorhexidine-Impregnated Sponge Dressing
in Cancer Patients

Randomized controlled trial at a single hopsital

601 patients receiving chemotherapy (>9000 catheter
days)
 Used chlorhexidine and silver sulfadiazine-impregnated CVC
 CVC for ≥5 days, removed when not needed or patient
discharged
 Intervention: CHGIS dressing (controls: sterile transparent
dressing), changed regularly after 1 wk

CRBSI was 46% less in the CHGIS dressing group

Catheter tip bacteria similar in both groups (>50% S.
epidermidis)
Ruschulte H et al. Ann Hematol 2009;88:267-72.
Other CHGIS Studies

Meta-analysis of RCTs: CHGIS vs standard dressing
(7 studies) or povidone-iodine dressing (1 study)
 Associated with reduction of vascular and epidural catheter exit
site colonization
 Trend towards reduction in CRBSI
 Local cutaneous reactions in 5.6% patients in 3 studies, 96% of
these in neonatal patients

Study involving 2 outpatient dialysis centers
 Prospective, crossover intervention trial over 1-year period: 121
patients with tunneled catheters received CHGIS dressing
 Use of CHGIS did not decrease CRBSI incidence
Ho KM et al. J Antimicrob Chemother 2006;58:281-7.
Camins BC et al. Infect Control Hosp Epidemiol 2010;31:1118-23.
Current Recommendations for ChlorhexidineImpregnated Sponge Dressing

Limited evidence indicating CHGIS use may
decrease CRBSI rates

Mainly studied short-term CVCs

May consider if CRBSI rate not decreasing despite
implementation of core measures
 In patients >2 months of age
 ?Applicability to long-term CVC use
Antimicrobial / Antiseptic Impregnated
Catheters

Several randomized studies in 1990s
 Chlorhexidine / silver sulfadiazine coated on external luminal
surface vs standard uncoated catheters

Meta-analysis:
 11 studies for catheter
colonization
 12 studies for CRBSI
 Mostly ICU patients
 Median CVC duration
5.1-11.2 days
Meta-Analysis Results
CVCs impregnated with chlorhexidine/silver sulfadizine
effective in reducing catheter colonization and CRBSI
Summary OR 0.44 (95% CI, 0.36-0.54)
Veenstra DL et al. JAMA 1999;281:261-7.
Summary OR 0.56 (95% CI, 0.37-0.84)
Chlorhexidine / Silver Sulfadiazine
Coated Catheters

Second generation catheters
 Chlorhexidine on internal surface extending to hubs
 Higher concentration of chlorhexidine/silver sulfadizine on
external luminal surface

Prospective, randomized studies of 2nd generation
catheters vs standard uncoated catheters:
 Significant reduction in catheter colonization
 Underpowered to detect difference in CRBSI

Rare reports of anaphylaxis
Brun-Buisson C et al. Intensive Care Med 2004;30:837-43.
Ostendorf T et al. Support Care Cancer 2005;13:993-1000.
Rupp ME Ann Intern Med 2005;143:570-80.
Minocycline / Rifampin Impregnated Catheters

To assess long-term catheters impregnated with
minocycline/rifampin in reducing CRBSI:
 Prospective, randomized trial in oncology hospital
 Mean CVC duration >60 days
M-R catheter:
0.25/1000 catheter
days
Uncoated catheters:
1.28/1000 catheter
days
Hanna H et al. J Clin Oncol. 2004;22:3163-71.
Catheters Impregnated with Minocycline /
Rifampin vs Chlorhexidine / Silver Sulfadiazine

Multicenter randomized trial (n=12 hospitals):
 M-R CVC vs first generation Chlorhexidine/silver sulfadizine
CVC
M-R catheters
• 12 times less
likely to have
CRBSI
• 3 times less likely
to be colonized
Darouiche RO et al. N Engl J Med 1999;340:1-8.
Additional Considerations for
Minocycline/Rifampin Impregnated Catheters

No comparison with 2nd generation chlorhexidine /
silver sulfadiazine catheters

Concern for increased antimicrobial resistance but
not shown in clinical settings
 Prospective, 7-year follow-up study in cancer center (>500,000
catheter days)
Ramos ER et al. Crit Care Med 2011;39:245-51.
Current Recommendations for Antimicrobial /
Antiseptic Impregnated Catheters

May consider use if CRBSI rates not decreasing:
 Either chlorhexidine / silver sulfadiazine or minocycline / rifampin
impregnated CVC
 If catheter is expected to remain in place >5 days

Additional data needed on other new catheters
 Platinum/silver
 Miconazole/rifampin
 M-R / Chlorhexidine / silver sulfadiazine
Antimicrobial / Antiseptic Catheter Lock

Filling catheter lumen with antimicrobial solution
when not in use

Various concentrations and combinations
 Antibiotics: vancomycin, gentamicin, ciprofloxacin, minocycline,
amikacin, cefazolin, cefotaxime, ceftazidime
 Antiseptics: alcohol, taurolidine*, trisodium citrate*

Usually combined with anticoagulant
 Heparin, EDTA
*Not approved for this use in the US
Antimicrobial Lock Solutions in Select
Patient Populations

Several studies in higher-risk patients, longer-term
CVC use
 Hemodialysis patients
• Study of 291 patients – significantly lower CRBSI rate using 30%
trisodium citrate (1.1/1000 CVC days) vs heparin (4.1/1000 CVC
days)
 Oncology patients
• Study of 126 patients – median CVC days 200-247 days:
o Vanc/Cipro/Heparin (0.55/1000 CVC days) vs Vanc/Heparin
(0.37/1000 CVC days) vs Heparin (1.72/1000 CVC days)
 Patients receiving long-term parenteral nutrition

Generally found reduction in CRBSI rates
Weijmer MC et al. J Am Soc Nephrol 2005;16:2769-77.
Henrickson KJ et al. J Clin Oncol 2000;18:1269-78.
Use of 70% Ethanol Lock

Prospective, randomized trial of patients with
hematological disease (long-term CVC):
 70% ethanol lock vs heparinized saline
0.6 CRBSI/1000
days
3.11/1000 CVC
days
Sanders J et al. J Antimicrob Chemother 2008;62:809-15.
Additional Considerations Regarding
Catheter Locks

Need to balance benefits with potential side effects
 Toxicity, allergic reactions
 Emergence of antimicrobial resistance

Limitations of studies
 Small sample size
 Heterogeneity of patient populations

Wide variety of compounds for use
 No FDA approved formulations
Recommendations for Catheter Locks

Not recommended for general use

May consider in patients with long-term CVC and
recurrent CRBSI despite adherence to aseptic
technique
Summary

CRBSI in outpatient settings is an emerging issue
 Impacting diverse patient populations
 Various outpatient settings

Limited surveillance and prevention data for
outpatient settings
 Additional research warranted for long-term CVC use, novel
technologies
Summary: Key Recommendations

Similar to inpatient settings with emphasis on line
care/maintenance practices:





Remove unnecessary CVC
Perform hand hygiene
Use >0.5% chlorhexidine/alcohol for skin / CVC site antisepsis
Appropriate disinfection of connectors/ports prior to access
Consideration of supplemental measures as needed
Thank you
For more information please contact Centers for Disease Control and
Prevention
1600 Clifton Road NE, Atlanta, GA 30333
Telephone, 1-800-CDC-INFO (232-4636)/TTY: 1-888-232-6348
E-mail: [email protected] Web: www.cdc.gov
The findings and conclusions in this report are those of the authors and do not necessarily represent the official
position of the Centers for Disease Control and Prevention.
National Center for Emerging and Zoonotic Infectious Diseases
Division of Healthcare Quality Promotion