Central venous catheter-related bloodstream infections ... - CCIH

Journal of Hospital Infection (2009) 71, 117e122
Central venous catheter-related bloodstream
infections: improving post-insertion catheter
care *
I.M. Shapey a, *, M.A. Foster a , T. Whitehouse b , P. Jumaa c , J.F. Bion a,b
a
University of Birmingham, Edgbaston, Birmingham, UK
b
Division of Critical Care Medicine, University Hospital Birmingham NHS Foundation Trust,
Birmingham, UK
c
Department of Clinical Microbiology and Infection Control, University Hospital Birmingham NHS
Foundation Trust, Birmingham, UK
Received 11 September 2008; accepted 17 September 2008
Available online 14 November 2008
KEYWORDS
Catheter-related
bloodstream infection
(CRBSI); Central
venous catheter;
Healthcare-associated
infection; Intensive
care unit; Quality
improvement
Available online at www.sciencedirect.com
www.elsevierhealth.com/journals/jhin
Summary Patients with central venous catheters (CVCs) are at increased
risk of bloodstream infections and sepsis-related death. CVC-related
bloodstream infections (CRBSIs) are costly and account for
a significant proportion of hospital-acquired infections. The aim of this
audit was to assess current practice and staff knowledge of CVC post-insertion
care and therefore identify aspects of CVC care with potential
for improvement. We conducted a prospective audit over 28 consecutive
days at a university teaching hospital investigating current practice of
CVC post-insertion care in wards with high CVC usage. A multiple choice
questionnaire on best practice of CVC insertion and care was distributed
among clinical staff. Rates of breaches in catheter care and CRBSIs were
calculated and statistical significance assumed when P < 0.05. Data was
recorded from 151 CVCs in 106 patients giving a total of 721 catheter
days.Inall,323breachesincarewereidentifiedgivingafailurerate
of 44.8%, with significant differences between intensive care unit (ICU)
and non-ICU wards (P < 0.001). Dressings (not intact) and caps and taps
(incorrectly placed) were identified as the major lapses in CVC care with
158 and 156 breaches per 1000 catheter days, respectively. During the
study period four CRBSIs were identified, producing a CRBSI rate of
5.5 per 1000 catheter days (95% confidence interval: 0.12e10.97). There
* Previously presented at the Intensive Care Society State of the Art Meeting, London, December 2007.
* Corresponding author. Address: 15 Brownlow St, Leamington Spa, Warwickshire CV32 5XH, UK. Tel.: þ447988706153.
E-mail address: i.m.shapey@doctors.org.uk
0195-6701/$ - see front matter ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jhin.2008.09.016

118 I.M. Shapey et al.
Introduction
Central venous catheters (CVCs) are vascular infusion
devices used for monitoring haemodynamic
variables, renal replacement therapy and the
administration of medication. The invasive nature
of indwelling intravascular catheters predisposes
the patient to a host of possible complications.
Such complications can arise from the individual
characteristics of each patient (e.g. difficulty in
accessing subclavian or internal jugular veins) and
may include mechanical complications (e.g. pneumothorax,
haematoma, and arterial puncture),
thrombotic and infectious complications. 1e5
Healthcare-associated infection is a major cause
of morbidity and mortality. 6 Patients with CVCs are
at markedly increased risk of bloodstream infections
and sepsis-related death. 7 It is estimated
that up to 6000 patients in England per year may acquire
a catheter-related bloodstream infection
(CRBSI). 8 Treatment costs of such infections are estimated
to exceed £6000 per infection. 9
The National Audit Office (NAO) report in 2000
on the control of hospital-acquired infection highlighted
the need for improved education, training
and audit of compliance with infection control
guidelines. 10 The National Institute for Clinical Excellence
issued guidelines in 2003 for the prevention
of healthcare-associated infection. 11 Best
practice in the use of CVCs incorporates the use
of maximal sterile barrier precautions during catheter
placement; highest standards of hand hygiene;
2% chlorhexidine gluconate for skin
antisepsis; regular inspection of CVC insertion sites
(dressings); documentation of CVC insertion and
removal dates; maintaining closed systems (caps
and taps closed); removal of CVCs if signs of infection
are present and saline flushing of catheter
lumens.
The aim of this study was to audit current
standards of practice in CVC maintenance in
high-use wards of a university teaching hospital,
assess knowledge of standards of CVC care
amongst staff in high-use wards and therefore
identify aspects of CVC care with the potential
for improvement.
are several opportunities to improve CVC post-insertion care. Future interventions
to improve reliability of care should focus on implementing
best practice rather than further education.
ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights
reserved.
Methods
Study setting and design
The study took place during one month in 2007 at
the Queen Elizabeth Hosptial, a tertiary referral
hospital in Birmingham. Clinical areas with the
greatest use of CVCs were identified using hospital
data on numbers of CVCs purchased.
A prospective audit of current standards of CVC
care was undertaken in these high-use clinical
areas. Staff knowledge of standards of best practice
in CVC care was assessed in the same areas
over the same period using a questionnaire.
Catheter care
The standard of post-insertion care of CVCs in
patients in high-use areas was audited over 28
consecutive days. Assessments were made once
daily, and based upon current evidence-based
guidelines and UK Department of Health recommendations;
they included inspection of CVC caps
and taps, dressings, catheter insertion sites and
lumens. 11,12 A breach in care was defined as a lapse
in optimal catheter care for any amount of time during
a 24 h period. Optimal care included: caps and
taps correctly placed, dressing intact, lumens
flushed and no signs of infection at the catheter site.
Results of all blood samples and catheter tips
sent for culture were recorded for patients with
a CVC during the 28 day study period. CRBSI was
defined as bacteraemia/fungaemia in a patient
with an intravascular catheter with the same
organism identified at both blood and CVC tip
culture, clinical manifestations of infection (i.e.
chills, fever, and/or hypotension), and no apparent
source for the bloodstream infection except
the catheter. Bloodstream infections were considered
to be associated with a central line if the line
had been in use during the 48 h period before the
development of the bloodstream infection. We
justify using this definition on the basis that, as
an observational study, we had no standardised
or mandated method for CVC tip and blood
cultures.

Central venous catheter post-insertion care 119
Staff survey
A questionnaire was distributed to staff working in
the high-use clinical areas on an opportunistic
basis. Knowledge of best practice of catheter
insertion and post-insertion care was assessed
using multiple choice questions. All grades of
nurses and doctors were approached during the
study period to complete the one-page
questionnaire.
Data and statistical analysis
We chose to restrict identifying opportunity for
failure (breaches) in care to a once daily visit.
Reliability rates (rates of breaches) for catheter
care and CRBSIs were calculated per 1000 catheter
days. Chi-squared analysis was used to compare
rate of breaches between clinical areas, according
to median duration ( 4 days and >4 days), and
between culture-positive and culture-negative
CVCs. P < 0.05 was considered to be statistically
significant.
Results
We identified five wards in our hospital with high
CVC use: two general surgical wards with designated
four-bedded post-operative areas (Surg 1
and Surg 2), an acute renal care ward (Renal) with
six high-care beds, a specialised hepatobiliary unit
with a nine bed high-dependency unit (HDU), and
a general intensive care unit (ICU) with 16 available
beds. We assumed that the annual hospital
usage of CVCs correlated with the number of CVCs
ordered, which was 3839 (3346 quad-lumen CVCs,
and 493 vascaths).
In all, 392 patients were admitted to the five
wards during the study period, of whom 106 (27%)
required a CVC. These 106 patients had an equal
sex distribution and a median age of 64 years
(range: 24e86). In total, 151 CVCs were identified,
of which 123 (81%) were standard quad-lumen
catheters and 28 (19%) were vascaths (temporary
haemodialysis catheters). The 151 CVCs were in
situ for a total of 721 catheter days, with a median
duration of four days (range: 1e20). Seventy-nine
(52%) of the CVCs were inserted in operative
theatres, 55 (36%) in the ICU, and 17 (12%) on
the ward. The most common site of insertion was
the internal jugular vein [120 CVCs (80%)]; 19 (12%)
were subclavian catheters, and 12 (8%) were
femoral. In total, 238 (91%) CVCs were removed
because there was no longer a clinical indication
for their use, nine (6%) removed because of
blockage, and four (3%) because of clinical suspicion
of infection.
Catheter care
During the 721 CVC days, 323 breaches were
recorded, giving a failure rate of 44.8% (95% CI:
44.2e44.8). The mean failure rates for catheters
in place up to, and for more than, the median
duration were 44.2% and 41.6% respectively. Aspects
of CVC care most commonly breached were
keeping dressings intact and caps and taps in place
(158 and 156 breaches per 1000 catheter days
respectively) (Figure 1). Statistical analysis
showed that the ICU had a significantly lower
breach rate than non-ICU wards (P < 0.001), and
that there was no significant increase in breach
rate of CVC care with duration of placement
(P > 0.05).
Infection rates
Of the 151 CVCs in the study, only 57 (37.7%)
were sent for culture, 24 of which (42.1%) had
tips that produced positive cultures. CVC tips
were much more likely to be sent for culture
from the ICU and renal ward (Table I). During the
Breach rate per 1000 catheter days
800
700
600
500
400
300
200
100
0
Lumens
Signs of infection
Renal Surg 1 Surg 2 HDU ICU
Wards
Dressings
Caps and taps
Figure 1 Breach rate according to ward and aspect of
care. HDU, high-dependency unit; ICU, intensive care
unit.

120 I.M. Shapey et al.
Table I Central venous catheter (CVC) statistics according to ward
Renal Surg 1 Surg 2 HDU ICU Total
No. of patients (male/female) 6 (1/5) 4 (4/3) 7 (1/3) 43 (19/24) 46 (28/18) 106 (53/53)
Admissions/month 24 112 118 67 71 392
Patients requiring CVC/month (%) 25 3.5 6 64 65 27
No. of CVCs 9 7 4 44 86 151
Catheter days 23 46 32 185 435 721
Median duration of CVCs (range) 2 (1e6) 6 (1e7) 4 (1e7) 4 (1e9) 5 (1e20) 4 (1e20)
Total no. of breaches 15 41 23 107 137 323
Breach rate per 1000 catheter days 652 674 470 583 336 448
CVC tips sent for culture, N (%) 3 (33) 1 (14.3) 0 (0) 9 (20) 44 (51.1) 57 (37.7)
Positive colonisation
of CVC tips, N (%)
2 (66) 1 (100) 0 (0) 4 (44) 18 (40.9) 24 (42.1)
Blood samples sent for culture, N (%) 3 (50) 4 (57) 1 (25) 16 (37.2) 25 (54.3) 49 (46.2)
Positive blood
samples, N (%)
3 (100) 2 (50) 0 (0) 2 (12.5) 11 (44) 18 (36.7)
CRBSI (rate per 1000 catheter days) 0 0 0 0 4 (5.5) 4 (5.5)
HDU, high-dependency unit; ICU, intensive care unit; CRBSI, catheter-related bloodstream infection.
same period, blood samples from 50 (47%) of the
106 patients were sent for culture, and 18 (36%)
were positive. Four patients contracted a CRBSI,
all of whom were located in the ICU, which
gave a CRBSI rate of 5.5 per 1000 catheter days
(95% CI: 0.12e10.97). Three of these patients’
specimens grew coagulase-negative staphylococci
and one grew enterococci from concurrent CVC
tip and blood culture. All culture-positive CVCs
had been sited in the internal jugular vein. Statistical
analysis showed no difference in breach rate
between infected and non-infected CVCs
(P ¼ 0.9). Table II shows the organisms identified
at CVC tip and blood culture.
Table II Organisms identified at central venous
catheter (CVC) tip and blood culture
CVC tip culture (N ) Blood culture (N )
Coagulase-negative Coagulase-negative
staphylococci (8) staphylococci (11)
Candida spp. (7) Enterococci (3)
Mixed skin
organisms (6)
E. coli (2)
Klebsiella
Klebsiella
pneumoniae (1) pneumoniae (1)
Enterobacter sp. (1) Clostridium
perfringens (1)
E. coli (1) Vancomycin-resistant
enterococci (1)
Enterococci (1) b-Haemolytic
streptococci (1)
Vancomycin-resistant Corynebacterium
enterococci (1)
Staphylococcus
aureus (1)
sp. (1)
Staff survey
Fifty-six members of staff completed the questionnaire:
4 (7%) student nurses, 28 (50%) staff
nurses, 12 (21%) sisters/charge nurses, 7 (13%)
foundation doctors, 5 (9%) specialist registrars
and no consultants. The proportions of correct
answers to questions on insertion technique were
as follows: washing of hands (100%); sterile gown
(98%); sterile drapes (89%); use of chlorhexidine
preparation (89%); use of iodine preparation (38%);
use of face mask (20%); avoidance of the femoral
route (73%). Table III shows the results of the staff
survey on CVC post-insertion care. There is considerable
agreement among staff on most aspects of
CVC care, with the exception of daily flushing of
lumens and the use of guidewires to replace CVCs.
Discussion
Catheter care
The failure rate of 44.8% demonstrates very low
reliability of CVC post-insertion care. Significantly
lower breach rates were recorded in the ICU
compared with other wards. Reasons for this
difference may include: (i) the 1:1 ICU nurse-topatient
ratio facilitates optimal care; (ii) greater
volume of CVC use in the ICU with more experience
and confidence in CVC care; (iii) ICU staff are more
aware of the need for rigorous infection control.
Similar breach rates regardless of duration of CVC
placement indicate that reliability of CVC care
does not deteriorate with time, and therefore that
constant vigilance is required at all times.

Central venous catheter post-insertion care 121
Table III Results of the staff survey on central
venous catheter (CVC) post-insertion care
Daily flushing of
CVC lumens
Taps should be
alcohol-wiped
or -sprayed
Caps should be
on at all times
A transparent dressing
should be used
Skin sites should be
inspected daily
for infection
Blood cultures should
be taken if infection
is suspected
Replacement or removal
of CVC after a fixed
amount of time
CVCs should not be
replaced over a
guide wire
Aspects of catheter care most commonly
breached were keeping caps and taps in place and
dressings intact. Caps and taps are the most used
component of the CVC and represent a likely portal
for pathogen entry into the vascular system. Dressings
are subjected to movement and abrasion which
affect their integrity and increase the risk of
breach. A study by Trick et al. suggests that an association
exists between poorly maintained dressings
and obese patients, patients with vascaths, and
patients outside the ICU. 13 Though we did not routinely
record patients’ body mass index, we have
identified a similar relationship between the integrity
of dressings, vascaths and non-ICU wards.
Infection rates
There is a large variation in the number of CVC tips
sent for culture from the five clinical locations.
Several reasons exist for this pattern which we
discuss below. Although a weak relationship exists
between organisms identified from blood and catheter
tip culture, this concurrence does not prove
causation. These organisms were, however, consistent
with results from other published reports. 14
Staff survey
Agree Undecided Disagree
38 13 5
55 1 0
53 1 2
55 1 0
55 1 0
54 1 1
51 3 2
21 23 12
The survey of staff knowledge showed considerable,
and accurate, agreement among staff on
most aspects of CVC care. However, two areas of
catheter care causing uncertainty are the daily
flushing of lumens and replacement of CVCs using
a guidewire. This concern did not vary according
to clinical area, occupation, or staff grade (data
not shown), and it is likely that it represents
genuine clinical uncertainty; replacement over
a guidewire may be a clinical necessity in some
circumstances. The main conclusion we draw
from the survey is that the discrepancy between
staff knowledge of CVC care and current practice
indicates implementation failure, not lack of
knowledge.
Limitations of the study
Our study specifically focused on post-insertion
CVC management, and in the absence of information
about standards of insertion we are
unable to link breaches in care to CRBSIs. The
small numbers of CRBSIs during the study, the
absence of a mandated policy on blood cultures
and catheter tip cultures, and the large sampling
variation between ordinary wards and the highcare
areas (ICU and renal) further limit interpretation.
CVC tips are cultured on the basis of
clinical suspicion of infection which creates
a sampling bias. Positive cultures were represented
as the proportion of CVCs or blood samples
sent for culture and not of the total number of
CVCs or patients. Thus, 5.5 is the likely minimum
rate of CRBSIs at our hospital; the true rate may
well be higher than this.
We conclude that there are multiple opportunities
for improvement in CVC post-insertion care, in
particular maintaining closed systems and intact
dressings. Future interventions to improve the
reliability of CVC care need to focus on methods
of implementing best practice rather than increasing
education on best practice and should target
clinical areas with high CVC use. Our rate of
CRBSIs, though comparable with other centres, is
too high considering recent large-scale reductions
in CRBSIs. 14e20
Conflict of interest statement
None declared.
Funding sources
None.
References
1. Merrer J, De Jonghe B, Golliot F, et al. Complications of
femoral and subclavian venous catheterization in critically

122 I.M. Shapey et al.
ill patients: a randomised control trial. J Am Med Assoc
2001;286:700e707.
2. Sznajder JI, Zveibil FR, Bitterman H, Weiner P, Bursztein S.
Central vein catheterization: failure and complication rates
by three percutaneous approaches. Arch Intern Med 1986;
146:259e261.
3. Mansfield PF, Hohn DC, Fornage BD, Gregurich MA, Ota DM.
Complications and failures of subclavian-vein catheterisation.
N Engl J Med 1994;331:1735e1738.
4. Veenstra DL, Saint S, Saha S, Lumley T, Sullivan SD. Efficacy
of antiseptic-impregnated central venous catheters in preventing
catheter-related bloodstream infection: a metaanalysis.
J Am Med Assoc 1999;281:261e267.
5. Raad I, Darouiche R, Dupuis J, et al. Central venous catheters
coated with minocycline and rifampin for the prevention
of catheter-related colonization and bloodstream
infections: a randomized, double-blind trial. Ann Intern
Med 1997;127:267e274.
6. Mermel LA. Prevention of intravascular catheter-related infections.
Ann Intern Med 2000;132:391e402.
7. Appelgren P, Hellström I, Weitzberg E, Söderlund V,
Bindslev L, Ransjö U. Risk factors for nosocomial intensive
care infection: a long-term prospective analysis. Acta
Anaesthesiol Scand 2001;45:710e719.
8. Waghorn DJ. Intravascular device associated systemic infections:
a two year analysis of cases in a district general hospital.
J Hosp Infect 1994;28:98e101.
9. National Audit Office. Improving patient care by reducing
the risk of healthcare associated infection: a progress report.
London: The Stationery Office; 2004.
10. National Audit Office. The management and control of
healthcare associated infection in Acute NHS Trusts in England.
London: Stationery Office; 2000.
11. National Institute for Clinical Excellence. Infection control,
prevention of healthcare-associated infection in primary
and community care. London: NICE; 2003.
12. Department of Health. Saving lives: reducing infection,
delivering clean and safe care. High impact intervention
No. 1: central venous catheter care bundle. London:
DoH; 2007.
13. Trick WE, Miranda J, Evans A, Charles-Damte M, Reilly BM,
Clarke P. Prospective cohort study of central venous
catheters among internal medicine ward patients. Am J
Infect Control 2006;34:636e641.
14. Paragioudaki M, Stamouli V, Kolonitsiou F, Anastassiou ED,
Dimitracopoulos G, Spiliopoulou I. Intravenous catheter infections
associated with bacteraemia: a 2-year study in
a university hospital. Clin Microbiol Infect 2004;10:
431e435.
15. National Nosocomial Infections Surveillance System. National
Nosocomial Infections Surveillance (NNIS) System
report, data summary from January 1992 through June
2004, issued October 2004. Am J Infect Control 2004;32:
470e485.
16. Pronovost P, Needham D, Berenholtz S, et al. An intervention
to decrease catheter-related bloodstream infections
in the ICU. N Engl J Med 2006;355:2723e2732.
17. Sherertz RJ, Ely EW, Westbrook DM, et al. Education of
physicians-in-training can decrease the risk for vascular
catheter infection. Ann Intern Med 2000;132:641e648.
18. Warren DK, Zack JE, Mayfield JL, et al. The effect of an
education program on the incidence of central venous catheter-associated
bloodstream infection in a medical ICU.
Chest 2004;126:1612e1618.
19. Eggimann P, Harbarth S, Constantin MN, Touveneau S,
Chevrolet JC, Pittet D. Impact of a prevention strategy
targeted at vascular-access care on incidence of infections
acquired in intensive care. Lancet 2000;355:
1864e1868.
20. Berenholtz SM, Pronovost PJ, Lipsett PA, et al. Eliminating
catheter-related bloodstream infections in the intensive
care unit. Crit Care Med 2004;10:2014e2020.