Unformatted Attachment Preview
List of Approved Systematic Reviews
NR451 Capstone Project
Directions: Please choose ONE topic and its corresponding systematic review that is of most
interest to you, or most relevant to your practice. This systematic review will be the basis for
your capstone project. Please refer to the guidelines for each milestone for more details.
Promoting breastfeeding
Sinha, B., Chowdhury, R., Sankar, M. J., Martines, J., Taneja, S., Mazumder, S., ... Bhandari, N.
(2015). Interventions to improve breastfeeding outcomes: A systematic review and metaanalysis. Acta Paediatrica, 104, 114-134. doi:10.1111/apa.13127.
Chamberlain Library Permalink:
http://proxy.chamberlain.edu:8080/login?url=http://search.ebscohost.com/login.aspx?dire
ct=true&db=a2h&AN=110726890&site=eds-live&scope=site
Preventing central venous catheter-related infections
Lai, N. M., Lai, N. A., O’Riordan, E., Chaiyakunapruk, N., Taylor, J. E., & Tan, K. (2016). Skin
antisepsis for reducing central venous catheter-related infections. Cochrane Database of
Systematic Reviews, (7), CD010140. doi:10.1002/14651858.CD010140.pub2.
Chamberlain Library Permalink:
http://onlinelibrary.wiley.com.proxy.chamberlain.edu:8080/doi/10.1002/14651858.CD01
0140.pub2/full
Increasing health insurance coverage for vulnerable populations
Jia, L., Yuan, B., Huang, F., Lu, Y., Garner, P., & Meng, Q. (2014). Strategies for expanding
health insurance coverage in vulnerable populations. Cochrane Database of Systematic
Reviews, (11), 1-41. CD008194. doi:10.1002/14651858.CD008194.pub3.
Chamberlain Library Permalink:
http://proxy.chamberlain.edu:8080/login?url=http://search.ebscohost.com/login.aspx?dire
ct=true&db=mdc&AN=25425010&site=eds-live&scope=site
Preventing teen pregnancy and sexually-transmitted disease
Mason-Jones, A. J., Sinclair, D., Mathews, C., Kagee, A., Hillman, A., & Lombard, C. (2016).
School-based interventions for preventing HIV, sexually transmitted infections, and
pregnancy in adolescents. Cochrane Database of Systematic Reviews, (11), CD006417.
doi:10.1002/14651858.CD006417.pub3.
Chamberlain Library Permalink:
http://proxy.chamberlain.edu:8080/login?url=http://search.ebscohost.com/login.aspx?dire
ct=true&db=mdc&AN=27824221&site=eds-live&scope=site
Reducing hospital readmissions
Mistiaen, P., & Poot, E. (2006). Telephone follow-up, initiated by a hospital-based health
professional, for post discharge problems in patients discharged from hospital to home.
Cochrane Consumers and Communication Group. (4), CD004510.
doi:10.1002/14651858.CD004510.pub3.
Chamberlain Library Permalink:
http://onlinelibrary.wiley.com.proxy.chamberlain.edu:8080/doi/10.1002/14651858.CD00
4510.pub3/full
Cochrane Database of Systematic Reviews
Skin antisepsis for reducing central venous catheter-related
infections (Review)
Lai NM, Lai NA, O’Riordan E, Chaiyakunapruk N, Taylor JE, Tan K
Lai NM, Lai NA, O’Riordan E, Chaiyakunapruk N, Taylor JE, Tan K.
Skin antisepsis for reducing central venous catheter-related infections.
Cochrane Database of Systematic Reviews 2016, Issue 7. Art. No.: CD010140.
DOI: 10.1002/14651858.CD010140.pub2.
www.cochranelibrary.com
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 3.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 4.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 5.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 6.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Povidone-iodine (in aqueous solution) versus no skin antisepsis, Outcome 1 Catheter-related
BSI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.2. Comparison 1 Povidone-iodine (in aqueous solution) versus no skin antisepsis, Outcome 2 Catheter
colonisation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.1. Comparison 2 Chlorhexidine (in aqueous solution) versus no skin antisepsis, Outcome 1 Septicaemia. .
Analysis 2.2. Comparison 2 Chlorhexidine (in aqueous solution) versus no skin antisepsis, Outcome 2 Catheter
colonisation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.3. Comparison 2 Chlorhexidine (in aqueous solution) versus no skin antisepsis, Outcome 3 Number of patients
who required antibiotics during in-dwelling period of catheter. . . . . . . . . . . . . . . . . .
Analysis 3.1. Comparison 3 Alcohol versus no skin antisepsis, Outcome 1 Catheter colonisation. . . . . . . .
Analysis 4.1. Comparison 4 Chlorhexidine versus povidone-iodine, Outcome 1 Catheter-related BSI. . . . . . .
Analysis 4.2. Comparison 4 Chlorhexidine versus povidone-iodine, Outcome 2 Catheter-related BSI per 1000 catheterdays. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.3. Comparison 4 Chlorhexidine versus povidone-iodine, Outcome 3 All-cause mortality. . . . . . .
Analysis 4.4. Comparison 4 Chlorhexidine versus povidone-iodine, Outcome 4 Catheter colonisation. . . . . .
Analysis 4.5. Comparison 4 Chlorhexidine versus povidone-iodine, Outcome 5 Catheter colonisation per 1000 catheterdays. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.6. Comparison 4 Chlorhexidine versus povidone-iodine, Outcome 6 Insertion site infection. . . . . .
Analysis 5.1. Comparison 5 Chlorhexidine (in aqueous solution) versus alcohol, Outcome 1 Catheter-related BSI. .
Analysis 5.2. Comparison 5 Chlorhexidine (in aqueous solution) versus alcohol, Outcome 2 Catheter colonisation. .
Analysis 6.1. Comparison 6 Povidone-iodine (in aqueous solution) versus alcohol, Outcome 1 Catheter-related BSI.
Analysis 6.2. Comparison 6 Povidone-iodine (in aqueous solution) versus alcohol, Outcome 2 Catheter colonisation.
Analysis 7.1. Comparison 7 Alcohol versus octenidine in alcohol, Outcome 1 Catheter-related BSI. . . . . . .
Analysis 7.2. Comparison 7 Alcohol versus octenidine in alcohol, Outcome 2 Catheter-related BSI per 1000 catheterdays. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 7.3. Comparison 7 Alcohol versus octenidine in alcohol, Outcome 3 Catheter colonisation. . . . . . .
Analysis 7.4. Comparison 7 Alcohol versus octenidine in alcohol, Outcome 4 Catheter colonisation per 1000 catheterdays. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 7.5. Comparison 7 Alcohol versus octenidine in alcohol, Outcome 5 Skin colonisation. . . . . . . .
Analysis 7.6. Comparison 7 Alcohol versus octenidine in alcohol, Outcome 6 Adverse effects. . . . . . . . .
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1
1
2
4
7
8
8
13
14
17
18
21
22
23
26
28
28
28
35
66
70
70
71
71
72
72
73
74
75
76
77
78
78
79
79
80
81
81
82
82
83
83
i
Analysis 8.1. Comparison 8 Chlorhexidine (in alcohol) plus povidone-iodine (in aqueous solution) versus chlorhexidine (in
alcohol), Outcome 1 Catheter colonisation. . . . . . . . . . . . . . . . . . . . . . . .
Analysis 8.2. Comparison 8 Chlorhexidine (in alcohol) plus povidone-iodine (in aqueous solution) versus chlorhexidine (in
alcohol), Outcome 2 Catheter colonisation per 1000 catheter-days. . . . . . . . . . . . . . . .
Analysis 9.1. Comparison 9 Chlorhexidine (in alcohol) plus povidone-iodine (in aqueous solution) versus povidone-iodine
(in aqueous solution), Outcome 1 Catheter colonisation. . . . . . . . . . . . . . . . . . . .
Analysis 9.2. Comparison 9 Chlorhexidine (in alcohol) plus povidone-iodine (in aqueous solution) versus povidone-iodine
(in aqueous solution), Outcome 2 Catheter colonisation per 1000 catheter-days. . . . . . . . . . . .
Analysis 10.1. Comparison 10 Sanosil (hydrogen peroxide and silver) versus water as adjunct to chlorhexidine 2% aqueous
bath plus povidone-iodine aqueous 10% scrub, Outcome 1 Catheter colonisation. . . . . . . . . . .
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .
INDEX TERMS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
84
84
85
85
86
86
93
94
94
95
96
ii
[Intervention Review]
Skin antisepsis for reducing central venous catheter-related
infections
Nai Ming Lai1,2 , Nai An Lai3 , Elizabeth O’Riordan4 , Nathorn Chaiyakunapruk2 ,5,6 , Jacqueline E Taylor7 , Kenneth Tan8
1 School of Medicine, Taylor’s University, Subang Jaya, Malaysia. 2 School of Pharmacy, Monash University Malaysia, Selangor, Malaysia.
3
Intensive Care Unit, Queen Elizabeth II Jubilee Hospital, Coopers Plains, Australia. 4 Faculty of Nursing and Midwifery, The University
of Sydney and The Children’s Hospital at Westmead, Sydney, Australia. 5 Center of Pharmaceutical Outcomes Research, Department
of Pharmacy Practice, Faculty of Pharmaceutical Sciences, Phitsanulok, Thailand. 6 School of Population Health, The University of
Queensland, Brisbane, Australia. 7 Monash Newborn, Monash Medical Centre/Monash University, Clayton, Australia. 8 Department
of Paediatrics, Monash University, Melbourne, Australia
Contact address: Nai Ming Lai, School of Medicine, Taylor’s University, Subang Jaya, Malaysia. lainm@doctors.org.uk,
lainm123@yahoo.co.uk.
Editorial group: Cochrane Wounds Group.
Publication status and date: New, published in Issue 7, 2016.
Citation: Lai NM, Lai NA, O’Riordan E, Chaiyakunapruk N, Taylor JE, Tan K. Skin antisepsis for reducing central venous catheter-related infections. Cochrane Database of Systematic Reviews 2016, Issue 7. Art. No.: CD010140. DOI:
10.1002/14651858.CD010140.pub2.
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT
Background
The central venous catheter (CVC) is a device used for many functions, including monitoring haemodynamic indicators and administering intravenous medications, fluids, blood products and parenteral nutrition. However, as a foreign object, it is susceptible to
colonisation by micro-organisms, which may lead to catheter-related blood stream infection (BSI) and in turn, increased mortality,
morbidities and health care costs.
Objectives
To assess the effects of skin antisepsis as part of CVC care for reducing catheter-related BSIs, catheter colonisation, and patient mortality
and morbidities.
Search methods
In May 2016 we searched: The Cochrane Wounds Specialised Register; The Cochrane Central Register of Controlled Trials (CENTRAL)
(The Cochrane Library); Ovid MEDLINE (including In-Process & Other Non-Indexed Citations and Epub Ahead of Print); Ovid
EMBASE and EBSCO CINAHL Plus. We also searched clinical trial registries for ongoing and unpublished studies. There were no
restrictions with respect to language, date of publication or study setting.
Selection criteria
We included randomised controlled trials (RCTs) that assessed any type of skin antiseptic agent used either alone or in combination,
compared with one or more other skin antiseptic agent(s), placebo or no skin antisepsis in patients with a CVC in place.
Data collection and analysis
Two authors independently assessed the studies for their eligibility, extracted data and assessed risk of bias. We expressed our results in
terms of risk ratio (RR), absolute risk reduction (ARR) and number need to treat for an additional beneficial outcome (NNTB) for
dichotomous data, and mean difference (MD) for continuous data, with 95% confidence intervals (CIs).
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1
Main results
Thirteen studies were eligible for inclusion, but only 12 studies contributed data, with a total of 3446 CVCs assessed. The total number
of participants enrolled was unclear as some studies did not provide such information. The participants were mainly adults admitted to
intensive care units, haematology oncology units or general wards. Most studies assessed skin antisepsis prior to insertion and regularly
thereafter during the in-dwelling period of the CVC, ranging from every 24 h to every 72 h. The methodological quality of the included
studies was mixed due to wide variation in their risk of bias. Most trials did not adequately blind the participants or personnel, and
four of the 12 studies had a high risk of bias for incomplete outcome data.
Three studies compared different antisepsis regimens with no antisepsis. There was no clear evidence of a difference in all outcomes
examined, including catheter-related BSI, septicaemia, catheter colonisation and number of patients who required systemic antibiotics
for any of the three comparisons involving three different antisepsis regimens (aqueous povidone-iodine, aqueous chlorhexidine and
alcohol compared with no skin antisepsis). However, there were great uncertainties in all estimates due to underpowered analyses and
the overall very low quality of evidence presented.There were multiple head-to-head comparisons between different skin antiseptic
agents, with different combinations of active substance and base solutions. The most frequent comparison was chlorhexidine solution
versus povidone-iodine solution (any base). There was very low quality evidence (downgraded for risk of bias and imprecision) that
chlorhexidine may reduce catheter-related BSI compared with povidone-iodine (RR of 0.64, 95% CI 0.41 to 0.99; ARR 2.30%, 95% CI
0.06 to 3.70%). This evidence came from four studies involving 1436 catheters. None of the individual subgroup comparisons of aqueous
chlorhexidine versus aqueous povidone-iodine, alcoholic chlorhexidine versus aqueous povidone-iodine and alcoholic chlorhexidine
versus alcoholic povidone-iodine showed clear differences for catheter-related BSI or mortality (and were generally underpowered).
Mortality was only reported in a single study.
There was very low quality evidence that skin antisepsis with chlorhexidine may also reduce catheter colonisation relative to povidoneiodine (RR of 0.68, 95% CI 0.56 to 0.84; ARR 8%, 95% CI 3% to 12%; ; five studies, 1533 catheters, downgraded for risk of bias,
indirectness and inconsistency).
Evaluations of other skin antiseptic agents were generally in single, small studies, many of which did not report the primary outcome
of catheter-related BSI. Trials also poorly reported other outcomes, such as skin infections and adverse events.
Authors’ conclusions
It is not clear whether cleaning the skin around CVC insertion sites with antiseptic reduces catheter related blood stream infection
compared with no skin cleansing. Skin cleansing with chlorhexidine solution may reduce rates of CRBSI and catheter colonisation
compared with cleaning with povidone iodine. These results are based on very low quality evidence, which means the true effects may
be very different. Moreover these results may be influenced by the nature of the antiseptic solution (i.e. aqueous or alcohol-based).
Further RCTs are needed to assess the effectiveness and safety of different skin antisepsis regimens in CVC care; these should measure
and report critical clinical outcomes such as sepsis, catheter-related BSI and mortality.
PLAIN LANGUAGE SUMMARY
Skin antisepsis for reducing central venous catheter-related infections
Review Question
We reviewed the evidence about whether using antiseptic treatments on people’s skin helps reduce infections related to central venous
catheters (CVCs).
Background
Central venous catheters (CVCs) are thin, flexible tubes that are inserted through the skin into a large vein, often in the arm or chest.
The tube can then be used to give fluids, medicine and nutrition to chronically and critically ill patients. However, CVCs pose a
significant risk of infection by providing a way for micro-organisms (germs) to spread into the body at the point where the catheter
is inserted. In order to try to reduce catheter-related infections, healthcare staff frequently use antiseptic solutions to clean the skin
around the catheter insertion site, both prior to insertion and whilst the catheter is in place. In this review, we summarise the evidence
of the benefits and harms of using antiseptics on the skin, and the effects of different antiseptic solutions.
Search date
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
2
We searched multiple medical databases in May 2016.
Study characteristics
In May 2016 we searched medical databases to find randomised controlled trials looking at the use of skin antiseptics in people with
CVCs. We included 13 studies in this review, although only 12 studies contributed data for a total of 3446 CVCs. The study participants
were mainly adults in intensive care units or other specialist hospital units. We reported our findings in terms of the number of catheters,
as some studies did not provide the number of patients assessed, and some patients had more than one CVC.One study was funded
by a national research body, five studies were funded in whole or in part by at least a pharmaceutical company, and in the remaining
seven studies funding sources were not stated.
Key results
Three studies examined the effect of cleansing versus no cleansing, and found no clear evidence of differences in blood infections,
infections in the catheter and need for antibiotics between patients who received cleansing compared to those who did not. Chlorhexidine
solution may reduce blood infections associated with the catheter compared with povidone-iodine solution (reducing the infection
rate from 64 cases per 1000 patients with a CVC with povidone iodine to 41 cases of infection per 1000 with chlorhexidine). This
translates into the need to treat 44 people to avoid one additional bloodstream infection. Chlorhexidine solution may (compared with
povidone iodine solution) also reduce the presence of infectious organisms within the catheter (reduced from 240 infected catheters
per 1000 people to 189 infected catheters per 1000 people). It is unclear whether antiseptic skin cleansing influences mortality rates as
only one study reported this and although similar death rates were observed with povidone iodine and chlorhexidine, small numbers
mean a difference cannot be ruled out.
Quality of evidence
The overall quality of evidence was poor due to flaws in the way the studies were designed, small study sizes, inconsistency of the results
between the included studies and the nature of the outcomes reported. These flaws have reduced our confidence in the results of the
studies. This means we cannot be certain whether cleaning the skin around CVC insertion sites with antiseptic reduces catheter-related
blood stream infection and other harmful effects, such as overall blood infections and mortality compared with no skin cleansing.
Cleansing with chlorhexidine solution may be more effective than povidone iodine but the quality of the evidence was very low.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
3
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Chlorhexidine compared to povidone- iodine for patients with a central venous catheter
Patient or population: patients with a central venous catheter
Settings: hospital inpatients
Intervention: chlorhexidine
Comparison: povidone-iodine
Outcomes
Illustrative comparative risks* (95% CI)
Assumed risk
Corresponding risk
Povidone- iodine
Chlorhexidine
Catheter-related BSI - over- Study population
all com parison between
chlorhexidine and povi- 64 per 1000
done-iodine
(during in-patient stay)
M oderate a
46 per 1000
Catheter-related BSI - sub- Study population
group: chlorhexidine in
aqueous solution versus 86 per 1000
povidone-iodine in aqueous
solution
M oderate
84 per 1000
Relative effect
(95% CI)
No. of Participants
(studies)
Quality of the evidence
(GRADE)
RR 0.64
(0.41 to 0.99)
1436
(4 RCTs)
⊕
Very low b,c
RR 0.64
(0.32 to 1.28)
452
(2 RCTs)
⊕
Very low c,d
41 per 1000
(26 to 63)
29 per 1000
(19 to 45)
55 per 1000
(28 to 110)
54 per 1000
(27 to 108)
4
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Catheter-related BSI - sub- Study population
group: chlorhexidine in alcohol versus povidone-iodine 70 per 1000
in aqueous solution
RR 0.77
(0.39 to 1.53)
503
(2 RCTs)
⊕
Very low c,d
RR 0.4
(0.13 to 1.24)
481
(1 RCT)
⊕⊕⊕
M oderate c
RR 1.15
(0.72 to 1.83)
213
(1 RCT)
⊕⊕
low c,e
RR 0.8
(0.48 to 1.34)
222
(1 RCT)
⊕⊕
low c,e
54 per 1000
(27 to 108)
M oderate
69 per 1000
Catheter-related BSI - sub- Study population
group: chlorhexidine in alcohol versus povidone-iodine 42 per 1000
in alcohol
53 per 1000
(27 to 106)
17 per 1000
(5 to 52)
M oderate
42 per 1000
17 per 1000
(5 to 52)
Prim ary BSI or clinical sep- No studies under this com parison assessed this outsis
com e.
All-cause
m ortal- Study population
ity - Chlorhexidine in aqueous solution versus povi- 236 per 1000
done-iodine in aqueous solution
Clinical assessm ent
M oderate
236 per 1000
All-cause
Study population
m ortality - Chlorhexidine in
alcohol versus povidone-iodine in aqueous solution
271 per 1000
(170 to 432)
271 per 1000
(170 to 432)
5
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Clinical assessm ent
236 per 1000
189 per 1000
(113 to 316)
M oderate
236 per 1000
189 per 1000
(113 to 316)
M ortality attributable the No studies under this com parison assessed this outCVC-related inf ections.
com e.
* The basis f or the assumed risk (e.g. the m edian control group risk across studies) is provided in f ootnotes. The corresponding risk (and its 95% conf idence interval) is
based on the assum ed risk in the com parison group and the relative effect of the intervention (and its 95% CI).
BSI: bloodstream inf ection; CI: Conf idence interval.
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our conf idence in the estim ate of ef f ect.
M oderate quality: Further research is likely to have an im portant im pact on our conf idence in the estim ate of ef f ect and m ay change the estim ate.
Low quality: Further research is very likely to have an im portant im pact on our conf idence in the estim ate of ef f ect and is likely to change the estim ate.
Very low quality: We are very uncertain about the estim ate.
a
’M oderate risk’ was calculated f rom the m edian control event rate f or each outcom e.
of the f our included studies had unclear risks of bias in allocation concealm ent, and all had high risks of bias in
blinding of participants and personnel.
c The 95% CI was wide.
d There was an overall very serious concern on risk of bias that resulted in downgrading of two levels: both studies had unclear
risk of bias under allocation concealm ent and high risk of bias under blinding of participants and personnel, and one study
had serious unit of analysis issue as the outcom e was reported using catheters as the unit, and the num ber of catheters
analysed exceeded the num ber of participants by over 50%, ref lecting that f act that som e patients received m ultiple catheters
during the study, which could have seriously af f ected the ef f ect estim ate.
e The single study had unclear risk in allocation concealm ent, high risk in blinding of patients and personnel which m ight give
rise to perf orm ance bias, which in turn m ight af f ect the risk of m ortality, as well as high risk of attrition bias.
b Three
6
BACKGROUND
Description of the intervention
Please refer to Appendix 1 for a glossary of terms (lay definitions
in the context of this review only).
A number of evidence-based guidelines have been developed in
recent years aimed at reducing CVC-associated BSIs. Important
measures recommended by two of the major guidelines include
the following (CDC 2011; Pratt 2007):
• Staff education
• Quality assurance: systematically monitoring compliance to
the established guidelines and evaluating issues relating to
compliance
• Hand hygiene
• The use of aseptic technique during insertion and use of
CVCs
• Effective skin antisepsis at the insertion site
• Maximum sterile barrier precautions (i.e. wearing sterile
gloves, sterile gown, a cap and a mask and using a large sterile
drape)
• Use of subclavian vein as the preferred site of insertion
rather than the internal jugular or femoral veins, as this has been
shown to reduce infectious, mechanical and thrombotic
complications (Hamilton 2007)
• The use of antimicrobial or antiseptic impregnated CVCs.
Description of the condition
The concept of central venous catheterisation was first introduced
in the early part of the last century by Bleichroder, Forssmann,
Duffy and Authaniac, after Bleichroder reportedly inserted the
first central venous catheter (CVC) in a human in 1905 (Puri
2009). In the past four decades, the use of the CVC has become
important in the management of many critically and chronically ill
patients. Insertion of a CVC provides secure vascular access for the
administration of intravenous medications, fluids, blood products
and parenteral nutrition. It also serves as an essential conduit for
blood sampling, haemodynamic monitoring, renal replacement
therapy and plasmapheresis.
It is estimated that 5 million CVCs are inserted every year in the
United States and 200,000 each year in the UK (Worthington
2005). One of the major problems associated with the use of
CVCs is colonisation by micro-organisms that could result in local
or systemic infection. Research has shown that infectious complications associated with CVCs cause significant morbidity and
mortality, with considerable costs to the healthcare system (CDC
2011; Cicalini 2004). In the USA, approximately 80,000 reported
cases of CVC-associated blood stream infections (BSIs) occur in
intensive care units (ICUs) every year; this number more than
triples when considering the entire hospital system (CDC 2011).
Although the exact mortality attributable to these BSIs remains
unclear, reports have cited figures up to 35% (CDC 2011). The
associated cost incurred due to BSIs is considerable, including
costs of additional medication, nursing time and increased length
of hospital stay. The total annual cost of caring for patients with
CVC-associated BSIs in the USA alone is estimated to range anywhere from USD 296 million to USD 2.3 billion (CDC 2011).
Micro-organisms colonise the CVCs and gain access to the blood
stream of the patients via three main routes (CDC 2011; Cicalini
2004; Pagani 2008):
• External surface of CVC through contaminated insertion
site
• Internal surface of CVC through contamination of catheter
hubs, injection ports and lines; usually by the hands of
healthcare workers or patients
• Contaminated intravenous drugs, infusates and nutritional
preparations.
For short-term CVCs, investigators have proposed colonisation
from the skin to the external surface of the CVCs as the major
route of infection, while for long-term CVCs, the internal surface route becomes increasingly important, as the micro-organisms gain access to the internal surface as a result of contamination
from repeated handling of the CVCs (Cicalini 2004).
Effective skin antisepsis throughout the in-dwelling period of the
catheter may prevent microbial contamination of the insertion site,
thus delaying or reducing the risk of catheter colonisation and the
subsequent development of infective complications. Given that
insertion site contamination leads to colonisation on the external
catheter surface and infection, one would expect skin antisepsis
to have some impact on reducing BSIs, especially with short-term
CVCs.
Pioneering work by Pasteur, Semmelweis and Lister laid the foundation for the practice of antisepsis in medicine (Bankston 2005;
Bynum 2008; Nuland 2003). Antisepsis is defined as the prevention of infection by inhibiting the growth of causative micro-organisms, while antiseptics are antimicrobial substances capable of
producing antisepsis (Taber 2016). An ideal antiseptic agent would
need to be immediately and persistently effective when applied to
living tissues, including when a small amount of blood is present,
and to be effective against all pathogenic bacteria, viruses, fungi,
protozoa, tubercle bacilli and bacterial spores (Taber 2016). At the
same time it should be non-toxic to living tissue, hypoallergenic
and safe to use repetitively on all parts of the body (Edwards 2008;
Hardin 1997). Human skin naturally has abundant microbiological flora which include resident (i.e. colonising) flora and transient (i.e. contaminating or non-colonising) flora. Resident flora
tend to inhabit deeper layers of the skin and therefore are not
readily removed by the mechanical action of washing with soap
and water. In contrast, transient flora are not consistently present
in most people and can usually be removed by mechanical action
(Larson 1995; Ryan 2004). Both resident and transient flora are
implicated in the pathogenesis of CVC-associated infections, thus
effective skin antisepsis may require not only mechanical removal
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
7
but also the chemical killing and inhibition of both the resident
and transient flora of the human skin (Edwards 2008).
How the intervention might work
There is a large number of antiseptic agents available and three
are considered particularly important in skin antisepsis: chlorhexidine, iodine and alcohol. All three agents have a broad spectrum of
activity against gram positive, gram negative, aerobic and anaerobic bacteria, enveloped viruses such as human immunodeficiency
virus (HIV), herpes simplex virus (HSV) and cytomegalovirus
(CMV), as well as fungi, although they differ in their effects against
tubercle bacilli and bacterial spores. We summarise their characteristics here:
• Chlorhexidine, which is available mostly as chlorhexidine
gluconate and less commonly as chlorhexidine acetate or
hydrochloride (Martindale 2016), exercises its antimicrobial
action chiefly by causing a disruption of microbial cell
membranes. Its activity against tubercle bacilli and bacterial
spores is limited (Larson 1995; Russell 1986). Chlorhexidine
gluconate has an intermediate onset of effect, which is reported
to be minimally affected by organic materials such as blood, pus
or sputum. It also appears to cause relatively low level of skin
irritation and has little allergenic potential. However, its activity
is pH dependent, and its effect is known to be compromised by
many substances, including those used in natural soaps (Larson
1995; Martindale 2016).
• Iodine and iodophors exert their antimicrobial effects
through chemical destruction of the microbial cell wall and
cellular contents. They are effective against tubercle bacilli and
bacterial spores. They kill bacteria within seconds to minutes but
are rapidly inactivated in the presence of organic materials such as
blood, pus or sputum. There have been reports of frequent skin
irritation, allergic reactions and systemic toxicity in susceptible
individuals (Edwards 2008; Hardin 1997; Larson 1995).
• Alcohols are available as either ethyl (ethanol), normalpropyl (n-propyl) or isopropyl alcohol for use as antiseptic
agents. Alcohols derive their antimicrobial activity from
denaturation of cellular proteins. They are effective against
tubercle bacilli but less so against bacterial spores. Alcohols have
a rapid onset of action, but they lose their antimicrobial effects
very quickly. Importantly for this review, they are often
combined with other agents such as chlorhexidine gluconate or
iodine to achieve optimal antisepsis. Alcohols are also poor
cleaning agents, and their use is usually not recommended when
significant amounts of blood or dirt are present. There have been
reports of excessive skin drying and discomfort following
application (Larson 1995; Martindale 2016).
Other antiseptic agents include the following (Larson 1995;
Martindale 2016):
• Triclosan
• Hexachlorophene
• Chloroxylenol
• Quarternary ammonium compounds such as cetrimide and
benzalkonium chloride
• Octenidine dihydrochloride
• Phenolic or carbolic acid compounds
• Hydrogen peroxide.
Why it is important to do this review
A meta-analysis showed that using chlorhexidine gluconate for
catheter site care reduced the risk of catheter-related BSIs by
49% when compared with povidone iodine (Chaiyakunapruk
2002). However, the meta-analysis only evaluated chlorhexidine
gluconate and povidone-iodine as skin antiseptics, and some studies within it assessed a combination of arterial catheters as well
as central and peripheral venous catheters. Some uncertainties remain regarding the best agent, or combination of agents, for use
as skin antisepsis for CVCs alone; the optimal interval between
application of antiseptics as well as the best method for applying
these agents. Examination of the latest National Healthcare Safety Network report, which superseded the National Nosocomial
Infections Surveillance (NNIS 2004), revealed that the CVC-associated BSI rate in different ICUs in the USA ranges from 1.0
to 5.6 BSI per 1000 CVC-days (Edwards 2008). These figures
compare favourably with the previous NNIS figures of 2.7 to 7.4
BSI per 1000 CVC-days (NNIS 2004). The observed improvement in CVC-associated BSI rate is probably multifactorial in nature, but the recent educational and awareness campaigns about
nosocomial infections and the implementation of infection control measures in many hospitals in the USA may have played a role.
The impact of different skin antisepsis regimens in the presence
of comprehensive infection control measures and lower baseline
BSI rates remains unclear. Furthermore, the availability of new
studies using different skin antiseptic preparations and the continuing emergence of drug resistant micro-organisms necessitates a
systematic review to aid clinical decision-making and to highlight
future research needs (O’Grady 2002; Parienti 2004; Pratt 2007).
OBJECTIVES
To assess the effects of skin antisepsis around central venous
catheter sites, on rates of catheter-related BSIs, catheter colonisation, and patient mortality and morbidities.
METHODS
Criteria for considering studies for this review
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
8
Types of studies
We included randomised controlled trials (RCTs) and cluster
RCTs comparing one skin antiseptic regimen (a single agent or
a combination of agents) with another regimen (a single agent
or a combination of agents, placebo or no antisepsis). We excluded cross-over studies due to the possible contaminating effect of one intervention over another. We also excluded studies
assessing CVCs for haemodialysis, as this is covered by another
Cochrane review (McCann 2010).
Types of participants
We included studies involving adults and children cared for in a
hospital setting (in adult or paediatric wards or ICUs) with any
underlying illness and a CVC inserted for any reason during the
study period. Studies that enrolled a patient more than once were
acceptable provided that the enrolment took place in separate hospital admissions. We excluded studies conducted in neonatal settings, for example in a neonatal intensive care unit (NICU), as the
types of catheters used, the insertion site and techniques, the possible complications as well as the risk factors for sepsis are different
compared with those in older children and adults (Trieschmann
2007).
Types of interventions
Intervention
The use of any skin antiseptic regimen (a single agent or a combination of agents) used for cleansing the skin around CVC insertion sites.
Comparisons
A different skin antisepsis regimen (a single agent or a combination
of agents), placebo or no skin antisepsis for CVC insertion sites.
We required that the selection, insertion, use, maintenance and
removal of CVCs in the intervention and comparison groups followed the standard protocol of the hospital setting in the study.
The skin antisepsis regimen had to be the only systematic difference between comparison groups (i.e., not catheter material or
concurrent CVC-related antiseptic measures).
We accepted the duration of the studies as variously specified by
the authors. We did not place any limit on the minimum and
maximum duration of the follow-up period for each study.
Number of patients with CVC-related blood stream infection
(BSI)
• Catheter-related BSI confirmed by laboratory
• Primary BSI or clinical sepsis.
We present the criteria for the diagnosis of CVC-related BSI in
Appendix 2 (Pagani 2008).
Mortality
• All-cause mortality
• Mortality attributable to CVC-related infections.
We included suitable studies using other definitions of CVC-related and associated infections, provided the authors justified their
definitions with valid sources.
Secondary outcomes
• Number of patients with insertion site infection, either
microbiologically documented (i.e. exudates at catheter insertion
site yield a micro-organism with or without concomitant BSI) or
clinically documented (i.e. erythema or induration within 2 cm
of the catheter insertion site in the absence of associated BSI and
without accompanying purulence) (Pagani 2008)
• Number of patients with catheter colonisation, as defined
by the study authors using well-accepted definitions such as a
significant growth of micro-organism (more than 15 colonyforming units (CFU)) from the catheter tip, subcutaneous
segment or catheter hub in the absence of clinical signs of
infection (Pagani 2008)
• Number of drug-resistant organisms from cultures,
including insertion site cultures, catheter cultures and blood
cultures
• Number of adverse events associated with the use of
antiseptic agents, including skin irritation, contact dermatitis,
systemic allergic reaction and anaphylaxis
• Antibiotic usage during hospitalisation
• Length of hospitalisation, either ICU stay or overall
hospital stay
• Cost of care, including cost of the antiseptic agent and the
cost of treating any adverse effects
• Quality of life, measured using validated tools.
Search methods for identification of studies
Electronic searches
Types of outcome measures
Primary outcomes
We searched the following databases for relevant RCTs:
• The Cochrane Wounds Specialised Register (searched 23
May 2016);
• The Cochrane Central Register of Controlled Trials
(CENTRAL) (The Cochrane Library) (2016, Issue 4);
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
9
• Ovid MEDLINE (including In-Process & Other NonIndexed Citations and Epub Ahead of Print) (1946 to 23 May
2016);
• Ovid EMBASE (1974 to 23 May 2016);
• EBSCO CINAHL Plus (1937 to 23 May 2016).
We used the search strategy in Appendix 3 to search the Cochrane
Central Register of Controlled Trials (CENTRAL). We adapted
this strategy for Ovid MEDLINE, Ovid EMBASE and EBSCO
CINAHL Plus which can be found in Appendix 4, Appendix 5
and Appendix 6, respectively. We combined the MEDLINE search
with the Cochrane Highly Sensitive Search Strategy for identifying
randomised trials in MEDLINE: sensitivity- and precision-maximising version (2011 revision) (Lefebvre 2011). We combined
the EMBASE search with the Ovid EMBASE filter developed
by the UK Cochrane Centre (Lefebvre 2011). We combined the
CINAHL searches with the trial filters developed by the Scottish
Intercollegiate Guidelines Network (SIGN 2015).
We searched the following trial registries for details of ongoing
clinical trials and unpublished studies.
• ClinicalTrials.gov (http://www.clinicaltrials.gov/).
• WHO International Clinical Trials Registry Platform (
http://apps.who.int/trialsearch/Default.aspx).
• EU Clinical Trials Register (https://
www.clinicaltrialsregister.eu/).
Searching other resources
We checked for further reports of eligible studies using the citation
lists of papers identified by the above strategies. We also scanned
references lists of relevant Cochrane reviews and guidelines and
contacted experts in the field.
Data collection and analysis
Selection of studies
Two review authors (NML, EOR) independently assessed the first
round of search results for potentially relevant studies. We retrieved
in full those that appeared to meet the inclusion criteria, or where
this could not be determined, for further assessment. Two review
authors independently assessed the full papers retrieved, resolving
any disagreement with input from a third review author (NC). We
included the studies if they fulfilled the criteria for inclusion as
outlined above and if the amount of information contained in the
article enabled the extraction of outcome data for meta-analysis.
We screened publications for duplicate reports of the same trial
and contacted the trial authors for clarification when necessary.
If we confirmed a duplicate publication, we identified a primary
reference, but extracted unique data from all versions.
Data extraction and management
Two pairs of review authors (NAL and NML, PL and EOR) independently extracted and coded all data for each included study using a pro forma designed specifically for this review. Each pair was
responsible for half of the total number of included studies. We
extracted the following information on each study: study design,
participants, setting, sample size, nature of intervention, comparison, outcomes, methods (unit of allocation and analysis) and results. We screened for duplicate entries of patients, where possible,
by matching the initial number of patients recruited against the
total number along each step in the conduct of the study.
We found a discrepancy between the number of catheter and the
number of patients in most studies. This was due to multiple
catheters being inserted in some patients who were enrolled after
each insertion. We were unable to limit our analysis to one catheter
per participant as none of the studies provided the data in this
format.
We resolved any disagreement among the review authors by discussion and formulation of a consensus acceptable to all members
of the review team.
Assessment of risk of bias in included studies
Two authors (NAL and NML) independently assessed each included study using the Cochrane tool for ’Risk of bias’ assessment
(Higgins 2011a). This tool addresses six specific domains.
1. Sequence generation
2. Allocation concealment
3. Blinding
4. Incomplete outcome data
5. Selective outcome reporting
6. Other issues (e.g. extreme baseline imbalance, designspecific risks of bias such as recruitment in cluster for clusterRCT, block randomisation of unblinded trials or fraud).
We present detailed criteria on which we based our judgement in
Appendix 7. We assessed blinding and completeness of outcome
data for each outcome separately. We completed a ’Risk of bias’
table for each eligible study. We resolved any disagreement among
the review authors by discussion to achieve a consensus. We presented an overall assessment of the risk of bias using a ’Risk of bias
summary figure’, which presented all of the judgement in a crosstabulation of study by entry. This display of internal validity indicated the weight the reader may give to the results of each study.
In addition, we assessed whether trials followed a standard protocol for all groups under study with regard to the insertion, use,
maintenance and removal of CVC, and regarding the concurrent
use of other antiseptic measures such as antimicrobial impregnated
CVCs, antiseptic-soaked dressing and prophylactic antibiotics. We
referred to the study protocol, where available, for further details
if necessary. We made relevant remarks in the corresponding ’Risk
of bias’ table for each study if there were significant concerns in
this aspect.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
10
Measures of treatment effect
For dichotomous data, we used risk ratio (RR) to measure outcome estimates of the same scale. We estimated the number needed
to treat for an additional beneficial outcome (NNTB) from the
pooled risk difference (RD) using an online NNTB calculator (
http://nntonline.net/visualrx/). For continuous data, we pooled
measures at a similar time point using the mean difference (MD).
Two studies reported the measure of variance as a standard error
(SE) or 95% confidence intervals (CI) (Humar 2000; Dettenkofer
2010). We obtained standard deviations (SD) for the above-mentioned studies from the SE using the formula SD = SE x square
root of the number of participants, and from the 95% CI using the
formula SD = square root of the number of participants x (upper
limit or CI − lower limit of CI)/3.92.
Unit of analysis issues
One potential unit of analysis issue that we had anticipated was
the issue that arose as a result of the studies using catheters, rather
than patients, as the unit of analysis in catheter-related outcomes
such as catheter-related BSI and catheter colonisation. Ideally, if
the study performed randomisation and analysis based on the participants, and each participant had only one catheter evaluated,
adjustment for clustering would not have been necessary. However, if a study included multiple catheters per patient and clearly
stated so, we would have assessed whether the authors had undertaken statistical adjustment to account for the effects of clustering
by using appropriate analysis models such as the ’generalised estimating equation’ (GEE) model (Higgins 2011b). If investigators
had made adjustments for clustering, we would have combined the
study with other studies in the meta-analysis. If they had not, or
if it was unclear whether there were adjustments made, we would
have assessed the number of catheters as well as participants in the
study. If the studies had also reported the number of participants
with events and the total number analysed, we would have only
reported the outcomes using the participants, rather than catheters
as the unit of analysis. However, if the study did not provide participant-level data, we would not have been able to avoid the unit
of analysis issues. We would have acknowledged this as a major
limitation of the review in our discussion and undertaken sensitivity analysis to assess the pooled results after excluding studies
with no adjustments for clustering.
However, in this review, none of the included studies provided
participant-level data for catheter-specific outcomes. As a result, we
could not adjust for the unit of analysis issue, nor could we perform
sensitivity analysis to assess the results with and without studies
with unadjusted unit of analysis issues. We have acknowledged
this in our discussion, as planned.
Another possible unit of analysis issue that could have arisen was
the effects of clustering that arose in cluster-RCTs in which randomisation was performed at the unit, rather than the participant
level. However, we did not include any cluster-RCTs in this review.
Had we identified an eligible cluster-RCT (e.g. trial in which the
assignment to intervention or control group was made at the level
of the unit or ward rather than the individual), we would have
addressed the possible unit of analysis issues as follows.
First, we would have assessed whether the authors had made adjustments for the effects of clustering to account for non-independence among the participants by using appropriate analysis
models such as the ’generalised estimating equation’ (GEE) model
(Higgins 2011b).
If investigators did not make adjustments for the effects of clustering, we would have performed adjustment by multiplying the
SEs of the final effect estimates by the square root of the ’design
effect’, represented by the formula ’1 + (m − 1) x ICC’, where m
is the average cluster size (number of participants per cluster) and
ICC is the intracluster correlation. We would have determined the
average cluster size m by dividing the total number of participants
by the total number of clusters. We would have used an assumed
ICC of 0.10, which has been proposed to be a realistic general
estimate based on previous similar studies (Campbell 2001). We
would also have combined the adjusted final effect estimates from
each trial with their SEs in our meta-analysis using the generic
inverse-variance methods, as stated in the Cochrane Handbook for
Systematic Reviews of Interventions (Higgins 2011b).
If it were impossible to find out whether trialists made adjustments
on the effect of clustering, we would still have included the studies
concerned in our meta-analysis using the effect estimates reported
by the authors, and performed sensitivity analyses to assess how
excluding those studies would affect the overall pooled estimates.
Dealing with missing data
We assessed whether there was a high attrition rate and whether an
intention-to-treat analysis was performed. To assess whether the
dropout rate was important, we inspected the absolute attrition
rate and the attrition rate in relation to the event rates for the intervention and the comparison groups. If the absolute dropout rate
was 20% or more, we judged the study to be at high risk of bias
due to incomplete outcome data. If the dropout rate was lower
than 20%, we used a ’worst-case-scenario’ method for the primary
outcomes (Guyatt 1993). For instance, for an unfavourable outcome such as catheter-related BSI or mortality, if the results of
a trial favoured the intervention group, we assumed all dropouts
from the intervention group to have developed the outcome, and
all dropouts from the comparison group to have not developed the
outcome. We then analysed to see if such an assumption changed
the direction of the results (e.g. from favouring the intervention
group to favouring the comparison group). If so, we considered
the dropout rate to be significant. We made the reverse assumption when a trial favoured the comparison group, or when the
outcomes examined were favourable, such as survival or treatment
success.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
11
Assessment of heterogeneity
We assessed all the included studies in terms of their clinical and
methodological characteristics.
1. Baseline characteristics of the participants
2. Clinical settings of the studies (e.g. intensive care units,
oncology wards, renal units)
3. Co-interventions
4. Methodological quality (as detailed in the ’Risk of bias’
assessment, for example studies at high risk of bias are defined as
studies with unclear or no allocation concealment, and studies
where participants, caregivers or investigators are not blinded, or
where blinding is unclear)
5. Nature of intervention (comparison between one skin
antiseptic regimen and placebo as opposed to comparison of two
active regimens)
6. Outcome assessment and unit of analysis.
We visually inspected the forest plots for any evidence of heterogeneity of treatment effects. We used the I2 statistic (Higgins 2003)
to measure inconsistency in the results, with a value of 50% or
greater indicating moderate to substantial statistical heterogeneity.
We found significant statistical heterogeneity in one analysis (
Analysis 4.4) and provided a plausible explanation the possible
reason for heterogeneity in the form of risk of attrition bias in some
included studies. We decided to still provide the pooled estimate
for this analysis and separated the studies based on the risk of
attrition bias in our pre-specified sensitivity analysis.
Assessment of reporting biases
We planned to screen for publication bias in our review using a
funnel plot if there were more than 10 studies included in the
analysis. If publication bias was implied by a significant asymmetry
of the funnel plot, we would have included a statement in our
results with a corresponding note of caution in our discussion. We
did not generate any funnel plot in this review as there were fewer
than 10 studies included in the analysis across all the comparisons
and outcomes.
Data synthesis
We used Review Manager software to perform meta-analysis of the
included studies (RevMan 2014). We used a fixed-effect model for
most of our analyses, as there was no substantial clinical and statistical heterogeneity. For the outcomes with substantial clinical and
statistical heterogeneity that was not satisfactorily explained or reduced by subgroup analyses, we used a random-effects model that
took into account between-study variability within the analysis
and lessened the possibility of spurious inferences of significance
compared to the fixed-effect model. We used the Mantel-Haenszel method to analyse all the dichotomous outcomes, as we anticipated relatively frequent events for most of our outcomes. For
continuous outcomes, we employed the inverse variance methods
using the effect measure of mean differences. In our assessment
of the effects of missing data, we compared our adjusted analysis
using the best- and worst-case scenarios to the completer analysis
as reported by the study authors.
When there were more than two arms evaluated in a study, for example, aqueous chlorhexidine versus alcoholic chlorhexidine versus aqueous povidone-iodine, we set up separate pairwise comparisons as subgroups under the major comparison of chlorhexidine versus povidone-iodine, as follows: aqueous chlorhexidine
versus aqueous povidone-iodine; and alcoholic chlorhexidine versus aqueous povidone-iodine. In so doing, we halved the total
number of participants and events in the povidone-iodine group
to avoid double-counting.
Had we identified studies that assessed cost-effectiveness, we
planned to provide only a narrative review of their findings and
not directly compare costs in studies using different units of measurement, due to the complexity of analysing cost-effectiveness if
different price-years were used.
Subgroup analysis and investigation of heterogeneity
In this review, we created subgroups of comparisons based on
the solution used, for example, a subgroup for chlorhexidine in
aqueous solution versus povidone iodine in aqueous solution, and
another subgroup for chlorhexidine in alcohol versus povidoneiodine in aqueous solution.
Had data been available, we would have carried out the following
subgroup analyses:
1. Short term CVCs (less than 10 days) versus longer term
CVCs (10 days or more)
2. CVCs with antimicrobial modifications (antimicrobial
impregnation, cuffs, hubs) versus CVCs with no antimicrobial
modifications
3. Studies undertaken in paediatric patients versus adult
patients
4. Studies undertaken in different patient populations with
different levels of care (intensive care patients, oncology patients,
renal patients and patients in general medical or surgical wards)
5. Studies undertaken with co-interventions (e.g. sepsis
prevention bundle) versus studies done without co-interventions
6. Studies that used rigorous criteria (e.g. as outlined in Pagani
2008) for determining catheter-related infections versus studies
that used more liberal criteria.
Sensitivity analysis
We performed the following sensitivity analyses.
1. Best- and worst-case scenarios to assess the impact of
missing data, as described in the section ’Dealing with missing
data’.
2. Including and excluding studies with unclear and high risks
of selection bias, namely, studies with unclear or high risk for
random sequence generation, allocation concealment or both.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
12
Had sufficient data been available, we would have performed additional sensitivity analyses to include and exclude studies with
methodological issues other than selection bias, such as a lack of
blinding to the participants, caregivers or investigators, or where
blinding was unclear.
’Summary of findings’ table
We created a ’Summary of findings’ table, which displayed seven
major outcomes in our review, using the web-based GRADEpro
software (http://gdt.guidelinedevelopment.org) (Schünemann
2011a). We used the eight GRADE considerations (study limitations, consistency of effect, imprecision, indirectness and publication bias, large effect, plausible confounding and dose response
relationship) to assess the overall quality of the body of evidence
(Schünemann 2011b). In generating the ’Summary of findings’
table, we interpreted the median control group event rate for the
outcome as ’moderate risk’.
RESULTS
Description of studies
Results of the search
We identified 609 records from the initial search of the Cochrane
Wounds Group Specialised Register, CENTRAL, MEDLINE,
EMBASE and CINAHL. We performed additional searches from
relevant published studies and identified two further studies that
appeared to be relevant. After removing duplicates, there were 574
records. Of these, 107 articles appeared to be relevant after we inspected the titles. We evaluated the abstracts and if necessary, the
full text of the articles, excluding 84 of the 107 records, including
one duplicate publication of another excluded study. Of the remaining 23 articles, one was an ongoing study, and we could not
fully assess six as we are still awaiting their full texts or further information from the authors. Ultimately, 16 articles describing 13
studies were available and met our inclusion criteria. Among these
16 articles, three were additional publications relating to three included studies. The flow diagram of the studies from the initial
search to the meta-analysis is shown in Figure 1. We describe all
the included studies in the Characteristics of included studies table
and note the reasons for excluding the others in the Characteristics
of excluded studies table.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
13
Figure 1. Study flow diagram.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
14
Included studies
We included 13 RCTs, conducted in eight countries, including the
USA (four studies), France (two studies), and Canada, Germany,
Iran, Japan, Spain, Switzerland and Finland (1 study each). Ten
trials were single centre RCTs and three were multicentre RCTs
(Dettenkofer 2010; Humar 2000; Yasuda 2013) The number of
patients recruited ranged from 50 (with 50 CVCs) in Sadowski
1988 to 420 (with 998 CVCs) in Vallés 2008. Mimoz 1996,
Mimoz 2007 and Yasuda 2013 did not report the number of
participants. Prager 1984 recruited children (n = 3) in addition
to adults (in this case, n = 159), while Sadowski 1988 recruited
children and adolescent from 10 weeks to 15 years of age. All
studies included participants of both sexes.
Six studies recruited patients from the medical/surgical ICUs
(Maki 1991; Mimoz 1996; Mimoz 2007; Vallés 2008; Tuominen
1981; Yasuda 2013), two studies recruited patients who were either pre- or post-cardiac surgery (Levy 1988; Yousefshahi 2013),
one study enrolled patients from a burns unit (Sadowski 1988),
one from haematology and surgical units (Dettenkofer 2010) and
the remaining three studies were conducted hospital-wide, which
included intensive-care and non intensive-care patients (Humar
2000; Langgartner 2004; Prager 1984). The average duration of
catheterisation, where reported, varied from 2 to 21.1 days (range
1 to > 30 days).
There were ten basic comparisons between two or three arms in
the included studies, with subgroups based on type of solution in
two comparisons.
• Comparison 1: povidone-iodine (in aqueous solution)
versus no skin antisepsis (Prager 1984).
• Comparison 2: chlorhexidine (in aqueous solution) versus
no skin antisepsis (Tuominen 1981).
• Comparison 3: alcohol versus no skin antisepsis (Sadowski
1988).
• Comparison 4: chlorhexidine versus povidone-iodine
(Humar 2000; Maki 1991; Mimoz 2007; Vallés 2008; Yasuda
2013). The specific subgroups for this comparison are listed
below based on the different preparations of chlorhexidine and/
or povidone-iodine:
◦ Chlorhexidine in aqueous solution versus povidoneiodine in aqueous solution (Maki 1991; Vallés 2008).
◦ Chlorhexidine in alcohol versus povidone-iodine in
aqueous solution (Humar 2000; Vallés 2008).
◦ Chlorhexidine in alcohol versus povidone-iodine in
alcohol (Mimoz 2007).
◦ Chlorhexidine in alcohol versus povidone-iodine (base
solution unknown) (Yasuda 2013).
Among the studies included in this comparison, two (Vallés 2008;
Yasuda 2013) carried out three-arm comparison. Vallés 2008 compared 2% chlorhexidine in aqueous solution (group 1), 0.5%
chlorhexidine in alcohol (group 2) and 10% povidone-iodine in
aqueous solution (group 3), while Yasuda 2013 compared 1%
chlorhexidine in alcohol (group 1), 0.5% chlorhexidine in alcohol
(group 2) and 10% povidone-iodine (base solution unknown).
Because the authors of Yasuda 2013 did not specify the base solution for the povidone-iodine group, we could not include this
study in any subgroup in our meta-analysis.
• Comparison 5: chlorhexidine (aqueous) versus alcohol
(Maki 1991).
• Comparison 6: povidone-iodine versus alcohol.
◦ Povidone-iodine in aqueous solution versus alcohol
(Maki 1991).
◦ Povidone-iodine-impregnated adherent film versus
alcohol (Levy 1988).
• Comparison 7: alcohol versus octenidine in alcohol
(Dettenkofer 2010).
• Comparison 8: chlorhexidine (in alcohol) plus povidoneiodine (in aqueous solution) versus chlorhexidine in alcohol
(Langgartner 2004).
• Comparison 9: chlorhexidine (in alcohol) plus povidoneiodine (in aqueous solution) versus povidone-iodine (in aqueous
solution) (Langgartner 2004).
• Comparison 10: Sanosil (hydrogen peroxide and silver)
versus water as adjunct to chlorhexidine 2% aqueous bath plus
povidone-iodine 10% aqueous scrub (Yousefshahi 2013).
In terms of the timing of intervention, most studies assessed skin
antisepsis prior to insertion and regularly thereafter during the indwelling period of the catheters, ranging from every 24 h to every
72 h. Three studies evaluated the skin antisepsis intervention only
prior to catheter insertion (Levy 1988; Yasuda 2013; Yousefshahi
2013), and one study examined skin antisepsis prior to removal
of the catheters (Sadowski 1988). Maki 1991 and Mimoz 1996
evaluated central venous as well as arterial catheters, although only
Maki 1991 provided a separate report of patients receiving CVCs
for the outcomes of catheter-related BSI and catheter colonisation,
while only Mimoz 1996 provided CVC-specific reports for both
outcomes per 1000 catheter-days.
The concentration of chlorhexidine-based solution used in the
studies ranged from 0.05% to 2%, with three studies using a combination of chlorhexidine plus alcohol. The concentration of povidone-iodine was 10% in all studies except Mimoz 2007, which
used 5% povidone-iodine together with 70% ethanol. All of the
studies that evaluated alcohol used 70% isopropyl alcohol except
Dettenkofer 2010, which used a combination of 45% 2-propanol
or 74% ethanol with 10% 2-propanol.
In terms of concomitant CVC-related infection control measures,
six studies clearly described the use of maximal sterile barrier
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
15
precaution (Dettenkofer 2010; Humar 2000; Langgartner 2004;
Mimoz 1996; Mimoz 2007; Vallés 2008), three studies described
part of the maximal sterile precaution (such as the use of sterile
gloves, gown or dressing) without explicitly mentioning maximal
sterile precaution (Levy 1988; Maki 1991; Yousefshahi 2013), and
four studies did not provide any clear description (Prager 1984;
Sadowski 1988; Tuominen 1981; Yasuda 2013).
The included studies assessed almost exclusively two major outcomes, namely, catheter colonisation or equivalent (all 13 studies) and catheter-related BSI or equivalent (8 studies). The other
outcomes assessed were sepsis, skin colonisation, insertion site infection, number of patients who required antibiotics during the
period of catheter use and adverse effects (only evaluated in one
study). Only one study reported mortality (Vallés 2008), and no
study reported cost of care or quality of life.
Control group risk of infection varied from 6.0% to 32.0% for
catheter colonisation, and from 4.1% to 9.8% for catheter-related
BSI.
Of the eight studies that evaluated the primary outcome of
catheter-related BSI, all except Yasuda 2013 clearly defined this
outcome in line with our definitions, detailed in Appendix 2. The
exact wording varied among the studies, but the definitions involved a positive blood culture in the presence of catheter with
clinical evidence of sepsis, improvement of the clinical signs following removal of the catheters or both. One study (Yousefshahi
2013) used the Centers for Disease Control and Prevention (CDC)
definitions of catheter-related BSI (CDC 2011), which were
also consistent with the definitions adopted in this review. Most
studies used previously validated laboratory methods to perform
catheter and blood cultures, adopting microbiological definitions
for colonisation and bloodstream infection that were consistent
with published literature in the evaluation of catheter-related infections, including the use of molecular subtyping. In Yasuda 2013,
the published abstract did not contain the definition of catheterrelated BSI.
All studies reported catheter-related outcomes such as catheterrelated BSI and catheter colonisation using the catheter as the unit
of analysis. Ten of the 13 included studies provided the number of
participants alongside the number of catheters, although none provided separate reports of the catheter-related outcomes using participants as the unit of analysis. The number of catheters matched
the number of participants in six studies (Dettenkofer 2010;
Levy 1988;Humar 2000; Maki 1991; Sadowski 1988; Yousefshahi
2013); in three studies, the number of catheters exceeded the number of participants: by 10% in Prager 1984, 18% in Langgartner
2004 and 50% in Vallés 2008. In Tuominen 1981, there were
fewer catheters analysed than participants enrolled, with no reason
provided.
We did not incorporate the outcome data of Yasuda 2013 into
our meta-analysis, as it was published only as an abstract and did
not state the base solution used (either aqueous or alcohol) for the
povidone-iodine group. We are awaiting further information from
the authors.
In terms of funding source, one study (Dettenkofer 2010) received
funding from a national research agency, five studies (Humar 2000;
Maki 1991; Mimoz 1996; Mimoz 2007; Prager 1984) were funded
in whole or in part by a pharmaceutical company, and in the
remaining seven studies (Langgartner 2004; Levy 1988; Sadowski
1988; Tuominen 1981; Vallés 2008; Yasuda 2013; Yousefshahi
2013), the sources of funding were not stated.
Excluded studies
We excluded a total of 83 articles based on one or more of the
following reasons.
1. Study design or article type (54 studies): the studies were
either retrospective or prospective cohort studies, cross-over
study, before-and-after intervention studies, prospective nonrandomised intervention studies, meta-analyses, economic
analyses with no original trial data, in vitro experiments, studies
with research questions or outcomes that did not match our
review, commentaries or an abstract of an included study,
excluded study or a study awaiting classification.
2. Population (17 studies): the participants in the studies were
either neonates, people undergoing haemodialysis or all patients
in ICU, not only those with CVCs in place.
3. Intervention (25 studies): the studies either assessed
antimicrobial-impregnated dressing or cerebral ventricular
catheter.
4. Insufficient information (four studies): the studies either
reported combined outcome data for arterial, venous or Swan
Gantz catheters (or a combination of these), with no separate
reporting for venous catheter and little possibility of contacting
the authors for further information, or they reported outcome
data that were unsuitable for meta-analysis.
Among the excluded articles, three articles were merged with other
articles as their secondary references on the basis of duplication
of information as stated under reason number 1 above, including
two included studies (Maki 1991; Mimoz 1996) and one excluded
study (Garland 2009b).
A description of each study is available in the ’Characteristics of
excluded studies’ table.
Risk of bias in included studies
There was a wide variation in the risk of bias of the included
studies. Overall, there was approximately a one-third split in the
domains that were judged to be low risk, unclear risk and high risk.
There was at least one high-risk domain in each of the included
studies. All studies were judged to be at high risk for blinding of
participants, except Dettenkofer 2010 (low risk) and Yousefshahi
2013 (unclear risk). Yasuda 2013 had unclear risks of bias in all
domains, as there was insufficient information in the published
abstract. The proportions of included studies with low, high and
unclear risks of bias in each domain is illustrated in Figure 2, and
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
16
the risk of bias judgment of each included study in each domain
is depicted in Figure 3. Additionally, we have provided a detailed
description of the risk of bias of each study in the ’Characteristics of
included studies’ table. We summarise our risk of bias assessments
for each domain below.
Figure 2. Risk of bias graph: review authors’ judgements about each risk of bias item presented as
percentages across all included studies.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
17
Figure 3. Risk of bias summary: review authors’ judgements about each risk of bias item for each included
study.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
18
Allocation
For random sequence generation, we judged 6 of the 13 included studies to have low risk of bias (Dettenkofer 2010; Humar
2000; Mimoz 1996; Mimoz 2007; Tuominen 1981; Vallés 2008).
For allocation concealment, three studies had low risk of bias
(Dettenkofer 2010; Langgartner 2004; Mimoz 2007). In these
studies, the authors clearly stated the method of sequence generation, which involved some form of random number scheme,
mostly by computers. There were also clear statements in the
’Methods’ that reassured the readers of the independence between
sequence generation and allocation. Two studies were judged to
be at high risk in sequence generation as well as allocation concealment, as they allocated participants either using an alternate
sequence or based on their hospital registration numbers (Prager
1984; Yousefshahi 2013). There was an unclear risk of bias in one
or both domains for 8 of the 13 included studies due to insufficient information provided in the articles.
Blinding
All of the studies except Dettenkofer 2010, Yasuda 2013 and
Yousefshahi 2013 had a high risk of bias with regard to blinding of
participants. Maki 1991, Mimoz 1996 and Mimoz 2007 clearly
stated that they did not blind participants, while other studies
did not specify. However, blinding was considered very unlikely
in these studies because they compared either a skin antisepsis
regimen against no regimen, one skin antisepsis solution against
another with a different appearance, or a skin antisepsis regimen
against a different and clearly distinguishable infection control
measure with no documented attempt to mask the participants.
Eight studies did not report blinding of outcome assessors (Humar
2000; Langgartner 2004; Levy 1988; Prager 1984; Sadowski 1988;
Tuominen 1981; Yasuda 2013; Yousefshahi 2013), while the other
five did not make any clear statements one way or the other (
Dettenkofer 2010; Maki 1991; Mimoz 1996; Mimoz 2007; Vallés
2008). Although investigators objectively measured the outcome
of catheter colonisation, catheter-related BSI required some degree
of clinical judgment, which might have been affected by lack of
blinding.
Incomplete outcome data
We judged studies to have a high risk of attrition bias for the
following three reasons, alone or in combination:
1. High absolute attrition rates (≥ 20% attrition) or an
attrition rate that was higher than the event rates in the control
group
2. Vulnerability of the pooled estimates to best- and worstcase scenarios using the dropouts in the assigned groups
3. Marked imbalance in the attrition rates between the
assigned groups.
Four studies had high risk of bias in this domain either because
they had more than 20% withdrawals (Dettenkofer 2010; Humar
2000; Langgartner 2004) or because their results changed significantly with best- and worst-case scenarios (Vallés 2008). Six studies had low risk of bias (Levy 1988; Maki 1991; Mimoz 2007;
Prager 1984; Sadowski 1988; Yousefshahi 2013), and the information on withdrawal was not sufficient in the remaining three
studies (Mimoz 1996; Tuominen 1981; Yasuda 2013).
Selective reporting
Nine studies had low risk of reporting bias (Dettenkofer 2010;
Humar 2000; Maki 1991; Mimoz 1996; Mimoz 2007; Prager
1984; Sadowski 1988; Tuominen 1981; Yousefshahi 2013), and
three studies carried a high risk (Langgartner 2004; Levy 1988;
Sadowski 1988). The three studies that were judged to have high
risk of reporting bias did not report key outcomes that would be
expected in such types of studies, such as catheter-related BSI,
clinical sepsis or mortality.
Other potential sources of bias
We screened for other potential sources of bias including extreme
baseline imbalance, block randomisation of unblinded trials, unit
of analysis issues and any evidence of fraud. As blinding was highly
unlikely in most included studies, the use of block randomisation
posed an additional risk of bias due to the possibility of disrupting
the integrity of the random sequence with educated guess on the
likely allocation of the future participants (Higgins 2011a). Two
studies (Humar 2000; Vallés 2008) were judged to have high risk
under ’other potential sources of bias’ as they used block randomisation, and the authors did not state whether they used varying
block sizes in either trial.
Unit of analysis issues were a particular concern in three studies
(Langgartner 2004; Prager 1984; Vallés 2008), in which the number of catheters analysed exceeded the total number of participants.
This meant that some participants had multiple catheters analysed
in the study as the authors of the three studies did not limit one
catheter per participants in the analyses. The results might have
been affected as the outcomes data from multiple catheters from
the same participants were most likely not independent from each
other. A more detailed description of the risk of bias of the trials
is provided in ’Assessment of risk of bias in included studies’.
Effects of interventions
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
19
See: Summary of findings for the main comparison
Chlorhexidine compared to povidone-iodine in reducing catheter
related infections
In this review, we assessed outcomes for a total of 3446 catheters in
our meta-analysis of 12 studies. The total number of participants
was unclear as some studies did not report this detail. Overall,
we carried out 10 comparisons, with variations related to the base
solution in comparisons 4 and 6.
• Comparison 1: povidone-iodine (in aqueous solution)
versus no skin antisepsis (Prager 1984).
• Comparison 2: chlorhexidine (in aqueous solution) versus
no skin antisepsis (Tuominen 1981).
• Comparison 3: alcohol versus no skin antisepsis (Sadowski
1988).
• Comparison 4: chlorhexidine versus povidone-iodine.
◦ Chlorhexidine in aqueous solution versus povidoneiodine in aqueous solution (Maki 1991; Vallés 2008).
◦ Chlorhexidine in alcohol versus povidone-iodine in
aqueous solution (Humar 2000; Vallés 2008).
◦ Chlorhexidine in alcohol versus povidone-iodine in
alcohol (Mimoz 2007).
• Comparison 5: chlorhexidine (in aqueous solution) versus
alcohol (Maki 1991).
• Comparison 6: povidone-iodine versus alcohol.
◦ Povidone-iodine in aqueous solution versus alcohol
(Maki 1991).
◦ Povidone-iodine-impregnated adherent film versus
alcohol (Levy 1988).
• Comparison 7: alcohol versus octenidine in alcohol
(Dettenkofer 2010).
• Comparison 8: chlorhexidine (in alcohol) plus povidoneiodine (in aqueous solution) versus chlorhexidine in alcohol
(Langgartner 2004).
• Comparison 9: chlorhexidine (in alcohol) plus povidoneiodine (in aqueous solution) versus povidone-iodine (in aqueous
solution) (Langgartner 2004).
• Comparison 10: Sanosil (hydrogen peroxide and silver)
versus water as adjunct to chlorhexidine 2% aqueous bath plus
povidone-iodine 10% aqueous scrub (Yousefshahi 2013).
Below, we report on our outcomes of interest in order of the comparisons that examined them.
Primary outcomes
Catheter-related BSI
Comparison 1: aqueous povidone iodine versus no skin
antisepsis (1 RCT, 179 catheters)
Prager 1984 was the only study that compared povidone iodine
in aqueous solution versus with no skin antisepsis (dry dressing).
There was no clear evidence of a difference in the rate of catheterrelated BSI (RR 0.99, 95% CI 0.37 to 2.61; 179 catheters; Analysis
1.1). The estimate is very uncertain as the comparison was underpowered to detect important differences in the outcome. The
quality of evidence for this outcome was rated as very low due
to very serious risk of bias issues (random sequence generation,
allocation concealment, non-blinding of participants and unit of
analysis issue) as well as imprecision.
Comparisons 2: aqueous chlorhexidine versus no skin antisepsis and comparison 3: alcohol versus no skin antisepsis
No study reported this outcome for these comparisons.
Comparison 4: chlorhexidine versus povidone-iodine (4
RCTs, 1436 catheters)
Overall, chlorhexidine (any solution) was associated with a lower
rate of catheter-related BSI than povidone-iodine (any solution)
(absolute risk reduction (ARR) of 2.30%, 95% confidence interval
(CI) 0.06% to 3.70%; risk ratio (RR) 0.64, 95% CI 0.41 to 0.99;
NNTB 44, 95% CI 27 to 1563; four studies, 1436 catheters, I2
= 0%; Analysis 4.1; Figure 4). This evidence was very low quality,
downgraded for imprecision (one level) and risks of bias (two levels) in allocation concealment, blinding of participants and unit
of analysis issues under “other sources of bias”. Analyses of subgroups according to the base solution used showed no clear differences between chlorhexidine and povidone-iodine in the rates
of catheter-related BSI: chlorhexidine in aqueous solution versus
povidone-iodine in aqueous solution (RR 0.64, 95% CI 0.32 to
1.28, 2 studies, 452 catheters, I2 = 15%), chlorhexidine in alcohol
versus povidone-iodine in aqueous solution (RR 0.77, 95% CI
0.39 to 1.53; 2 studies, 503 catheters, I2 = 0%), chlorhexidine in
alcohol versus povidone-iodine in alcohol (RR 0.40, 95% CI 0.13
to 1.24; 1 study, 481 catheters). The small number of trials in
each subgroup means that the comparisons were underpowered,
and the results are uncertain. We considered the evidence from
the data to be of very low overall quality (downgraded for imprecision (one level) and risks of bias (two levels) in allocation concealment, blinding of participants and unit of analysis issues. We
have highlighted the results for these outcomes from the overall
comparison of chlorhexidine versus povidone-iodine as well as the
three subgroup comparisons in our Summary of findings for the
main comparison.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
20
Figure 4. Forest plot of comparison: 1 Chlorhexidine versus povidone-iodine, outcome: 1.1 Catheterrelated BSI.
For the outcome of catheter-related BSI per 1000 catheter-days,
chlorhexidine was associated with an apparent lower BSI rate compared with povidone-iodine (RR 0.53, 95% CI 0.30 to 0.94; 4
studies, 1450 catheters, I2 = 0%; Analysis 4.2). Analyses of subgroups according to the base solution used found evidence of a
possible difference between chlorhexidine in alcohol versus povidone-iodine in aqueous solution (RR 0.49, 95% CI 0.25 to 0.95;
3 studies, 661 catheters, I2 = 31%), but relative effects were unclear for the other base solutions in comparison (chlorhexidine in
aqueous solution versus povidone-iodine in aqueous solution (RR
0.82, 95% CI 0.23 to 2.93; 1 study, 308 catheters), and chlorhexidine in alcohol versus povidone-iodine in alcohol (RR 0.41, 95%
CI 0.06 to 2.92; 1 study, 481 catheters). All subgroup comparisons were underpowered and the overall quality of evidence for
this outcome was very low due to very serious risk of bias issues
(non-blinding of participants, incomplete outcome data and unit
of analysis issues).
Comparison 5: aqueous chlorhexidine versus alcohol (1 RCT,
99 catheters)
A single small study compared chlorhexidine in aqueous solution
with alcohol (Maki 1991) and found no clear difference in the
absolute rate of catheter-related BSI between the alcohol-based
solution and the chlorhexidine-based solution (RR 0.24, 95% CI
0.02 to 2.54; 99 catheters; Analysis 5.1). The comparison was
underpowered and the quality of evidence for this outcome was
low due to risk of bias of the study (non-blinding) and imprecision.
Comparison 6: aqueous povidone-iodine versus alcohol (1
RCT, 109 catheters)
Maki 1991, the only study that compared povidone-iodine in
aqueous solution with alcohol did not find a clear difference in the
rate of catheter-related BSI between the two groups (RR 1.04, 95%
CI 0.24 to 5.08; 109 catheters; Analysis 6.1). The comparison was
underpowered and the quality of evidence for this outcome was
low due to risk of bias issue (non-blinding of the participants) and
imprecision.
Comparison 7: alcohol versus octenidine in alcohol (1 RCT,
387 catheters)
Dettenkofer 2010 was the only study to compare alcohol versus octenidine in alcohol, and found no clear difference between
groups in the absolute rate of catheter-related BSI (RR 2.01, 95%
CI 0.88 to 4.59; 387 catheters; Analysis 7.1) or catheter-related
BSI per 1000 catheter-days (RR 2.18, 95% CI 0.54 to 8.77; 387
catheters; Analysis 7.2). The comparison was underpowered and
the quality of evidence for both outcomes was low due to risk of
bias issue (incomplete outcome data) and imprecision.
Septicaemia (whether or not CVC-related)
Comparison 2: chlorhexidine versus no skin antisepsis (1 RCT,
136 participants)
The only study that reported the outcome of septicaemia (irrespective of its relationship with CVC) was Tuominen 1981, which
compared chlorhexidine with no skin antisepsis. This study of 136
participants compared the use of 0.05% chlorhexidine in aqueous
solution with no skin antisepsis and found no clear difference in
the rate of septicaemia between the two groups, but the result was
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
21
inconclusive due to imprecision (RR 2.91, 95% CI 0.31 to 27.31;
Analysis 2.1). The quality of evidence for this outcome was low
due to risk of bias issue (non-blinding of participants) and imprecision, as stated above.
Mortality (all-cause or CVC-related)
Comparison 4: chlorhexidine versus povidone-iodine (1 RCT,
329 participants analysed, 106 participants in povidone-iodine group were included in both subgroup comparisons below)
A single study (Vallés 2008) reported mortality. The study divided
the participants into three groups: chlorhexidine in aqueous solution, chlorhexidine in alcohol and povidone-iodine in aqueous
solution. Analyses according to subgroups showed no clear differences in the rates of mortality between chlorhexidine in aqueous solution and povidone-iodine in aqueous solution (RR 1.15,
95% CI 0.72 to 1.83; 213 participants) (Analysis 4.3), or between
chlorhexidine in alcohol and povidone-iodine in aqueous solution (RR 0.80, 95% CI 0.48 to 1.34; 222 participants) (Analysis
4.3)(Figure 5). However, the comparison was underpowered to
detect important differences in the outcome, and the quality of
evidence for both analyses was low due to a combination of risk of
bias issues and imprecision in the outcome estimates (Summary of
findings for the main comparison). Consequently true differences
in the mortality associated with use of chlorhexidine or povidone
iodine cannot be ruled out.
Figure 5. Forest plot of comparison: 1 Chlorhexidine versus povidone-iodine, outcome: 1.3 All-cause
mortality.
Secondary outcomes
Catheter colonisation
Comparison 1: aqueous povidone-iodine versus no skin
antisepsis (1 RCT, 179 catheters)
Based on Prager 1984, the only study in this underpowered comparison, it is unclear whether there is any difference in the effect
on catheter colonisation of aqueous povidone iodine and no skin
antisepsis (RR 0.93, 95% CI 0.53 to 1.60; 179 catheters; Analysis
1.2). There was very low quality evidence due to serious risk of
bias (random sequence generation, allocation concealment, nonblinding of participants and unit of analysis issue) and indirectness
of the outcome.
Comparison 2: aqueous chlorhexidine versus no skin
antisepsis (1 RCT, 124 catheters)
Based on Tuominen 1981, the only study to compare chlorhexidine in aqueous solution with no skin antisepsis, there was no
clear difference in the rate of catheter colonisation and therefore
uncertainty as to their relative effects remains (RR 1.26, 95% CI
0.61 to 2.59; 124 catheters; Analysis 2.2). The quality of evidence
was very low due to risk of bias (non-blinding of participants),
indirectness of the outcome and imprecise estimate from an underpowered analysis.
Comparison 3: alcohol versus no skin antisepsis (1 RCT, 50
catheters)
Based on a single study in this underpowered analysis (Sadowski
1988), it remains unclear whether there is a difference between
cleansing the skin with alcohol and no skin antisepsis prior to
catheter removal (RR 0.75, 95% CI 0.30 to 1.85; 50 catheters;
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
22
Analysis 3.1). The quality of evidence was very low due to risk of
bias (non-blinding of the participants), indirectness and imprecision.
Comparison 4: chlorhexidine versus povidone-iodine (5
RCTs, 1533 catheters)
Pooled analysis of five studies that compared chlorhexidine with
povidone iodine showed an overall reduction in the risk of catheter
colonisation with chlorhexidine (RR 0.68, 95% CI 0.56 to 0.84;
ARR 8%, 95% CI 3 to 12%; NNTB 13, 95% CI 9 to 34; 5 studies, 1533 catheters, I2 = 55%; Analysis 4.4; Figure 6). Analysing
subgroups according to the solution, there appeared to be reductions in rates of catheter colonisation favouring chlorhexidine in
the following comparisons:
Figure 6. Forest plot of comparison: 1 Chlorhexidine versus povidone-iodine, outcome: 1.4 Catheter
colonisation.
• Chlorhexidine in aqueous solution versus povidone-iodine
in aqueous solution (RR 0.60, 95% CI 0.40 to 0.91; 2 studies,
442 catheters, I2 = 56%).
• Chlorhexidine in alcohol versus povidone-iodine in alcohol
(RR 0.52, 95% CI 0.34 to 0.80; 1 study, 481 catheters).
However, the rate of catheter colonisation between chlorhexidine
in alcohol versus povidone-iodine in aqueous solution appeared to
be similar (RR 0.86, 95% CI 0.64 to 1.14; 3 studies, 600 catheters,
I2 = 58%).
There was moderate heterogeneity present for the overall pooled
analysis, as indicated by the I2 of 55%. The extent of heterogeneity
remained even with the studies separated into subgroups according
to the solution used, as shown above. We investigated other possible sources of heterogeneity by exploring factors that were present
in the population, intervention, comparison, outcome definitions
and risk of bias among the included studies. We noted that although there were some differences in the characteristics of the
included studies in terms of population (surgical versus cardiac
versus general ICUs) and intervention (different concentrations of
chlorhexidine used, duration of catheterisation and the concurrent
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
23
use of other antiseptic substances alongside chlorhexidine-based
solution), these differences did not plausibly explain the degree of
heterogeneity, as separating the studies into subgroups according
to these factors did not reduce the degree of heterogeneity.
However, we identified one plausible source of heterogeneity under the risk of bias criterion. We found that only two out of five included studies (Maki 1991; Mimoz 1996) had low risk of attrition
bias, while the other three were at high risk of bias in this domain.
The two studies with low risk of attrition bias showed significant
benefits of chlorhexidine compared with povidone-iodine, whilst
the remaining studies showed no significant difference between
the two groups. Grouping studies with low risk and high risk of
attrition bias separately reduced the I2 statistic to 0% and 41%,
respectively.
We undertook best- and worst-case scenarios to determine the impact of missing data from these three studies and found that the
overall pooled analysis was substantially altered, with the best-case
scenario moving the direction of the pooled estimate to significantly and substantially favour the chlorhexidine group, and the
worst-case scenario moving the pooled estimate to significantly
favour the povidone-iodine group (see ’Sensitivity analysis’ for details).
Having identified a plausible explanation for the observed heterogeneity, we still decided to combine all five studies under three
different subgroups according to the type of solution used (either
aqueous or alcohol). Taking all considerations, the overall quality
of evidence for this outcome was very low, as there were very serious concerns regarding risk of bias (non-blinding of participants,
incomplete outcome data and unit of analysis issue), indirectness
of the outcome and inconsistency among the study results.
Comparison 5: aqueous chlorhexidine versus alcohol (1 RCT,
99 catheters)
According to a single study (Maki 1991), it remains unclear
whether there is a difference in the rates of catheter colonisation
between chlorhexidine in aqueous solution and alcohol (RR 0.38,
95% CI 0.11 to 1.33; 99 catheters; Analysis 5.2), but the comparison was underpowered. The quality of evidence for this outcome
was very low due to risk of bias (non-blinding of participants),
indirectness and imprecision.
Comparison 6: aqueous povidone-iodine versus alcohol (3
RCTs, 169 catheters)
It is unclear whether there is a difference in the rates of catheter
colonisation between patients who received CVC cleansing with
povidone-iodine and those who receive cleansing with alcohol,
either overall (RR 1.76, 95% CI 0.76 to 4.09; 2 studies, 169
catheters, I2 = 43%), or in subgroups comparing povidone-iodine
in aqueous solution versus alcohol (RR 1.25, 95% CI 0.49 to 3.14;
1 study, 109 catheters) or povidone-iodine-impregnated adherent
film versus alcohol (RR 9.00, 95% CI 0.51 to 160.17; 1 study,
60 catheters; Analysis 6.2). The comparisons were underpowered,
and the overall quality of evidence for this outcome was very low
due to risk of bias (non-blinding of participants), indirectness of
the outcome and imprecision.
Comparison 7: alcohol versus octenidine in alcohol (1 RCT,
322 catheters)
Dettenkofer 2010, the only study to compare alcohol versus
octenidine in alcohol, showed that alcohol alone is probably associated with a higher rate of catheter colonisation compared to
octenidine (RR 2.26, 95% CI 1.22 to 4.21; 322 catheters; Analysis
7.3). However, there appeared to be no clear difference between
the two groups in terms of catheter colonisation per 1000 catheterdays (RR 2.23, 95% CI 0.79 to 6.29; 322 catheters; Analysis 7.4).
The quality of evidence for both outcomes was low, due to concerns in risk of bias (non-blinding of participants) and indirectness
of the outcomes.
Comparison 8: chlorhexidine in alcohol plus povidoneiodine in aqueous solution versus chlorhexidine in alcohol (1
RCT, 88 catheters)
In an underpowered analysis from a single study (Langgartner
2004), a combination of chlorhexidine plus povidone-iodine appeared to be associated with lower rate of catheter colonisation
(RR 0.19, 95% CI 0.04 to 0.81; 88 catheters; Analysis 8.1) as well
as catheter colonisation per 1000 catheter-days (RR 0.19, 95% CI
0.06 to 0.59; 88 catheters; Analysis 8.2) compared with chlorhexidine alone, although the effects were uncertain due to the very
low quality of evidence, which was reduced by risk of bias (nonblinding of participants, incomplete outcome data, unit of analysis issue), indirectness and imprecision.
Comparison 9: chlorhexidine in alcohol plus povidoneiodine in aqueous solution versus povidone-iodine in
aqueous solution (1 RCT, 95 catheters)
In another single-study, underpowered analysis based on
Langgartner 2004, there appeared to be lower rate of catheter
colonisation (RR 0.15, 95% CI 0.04 to 0.62; 95 catheters; Analysis
9.1) as well as catheter colonisation per 1000 catheter-days (RR
0.17, 95% CI 0.05 to 0.52; 95 catheters; Analysis 9.2) using a combination of chlorhexidine and povidone-iodine compared with using povidone-iodine alone, but the effects were very uncertain due
to the very low quality of evidence, which was reduced by risk of
bias (non-blinding of participants, incomplete outcome data, unit
of analysis issue), indirectness and imprecision.
Skin antisepsis for reducing central venous catheter-related infections (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
24
Comparison 10: Sanosil (hydrogen peroxide and silver)
versus water as adjunct to chlorhexidine 2% aqueous bath
plus povidone-iodine 10% aqueous scrub (1 RCT, 249
catheters)
From the single study in this underpowered comparison (
Yousefshahi 2013), it is uncertain whether there is any clear difference between the two groups in the rate of catheter colonisation
(RR 1.08, 95% CI 0.68 to 1.72; 249 catheters; Analysis 10.1) due
to the very low quality of evidence, which was reduced by risk of
bias (random sequence generation, allocation concealment), indirectness and imprecision.
under adverse effects. To avoid duplication, we included only the
most commonly reported adverse effect, namely, skin irritation.
For this outcome, there was moderate quality evidence showing
no clear difference between in adverse effect rates between patients
whose CVC sites were cleansed with alcohol and those who were
cleansed with octenidine in alcohol (RR 0.85, 95% CI 0.60 to
1.20; 398 participants; Analysis 7.6). The quality of evidence was
reduced by imprecision of the effect estimates from an underpowered analysis.
Number of patients who were on antibiotics during the
period of catheter use
Insertion site infection
Comparison 4: Chlorhexidine versus povidone-iodine (1
RCT, 242 catheters)
Based on the result of a single study (Humar 2000) in an underpowered analysis, it is uncertain whether there is any clear difference between chlorhexidine (in alcohol) and povidone-iodine (in
aqueous solution) with regard to insertion site infection, as the
quality of evidence was very low due to risk of bias (non-blinding of the participants, incomplete outcome data), indirectness
and imprecision. The authors reported this outcome as the mean
CFU count (MD − 2.80, 95% CI − 9.10 to 3.50; 242 catheters;
Analysis 4.6).
Comparison 2: Chlorhexidine in aqueous solution versus no
skin antisepsis ( 1 RCT, 136 participants)
The only study that evaluated this outcome, Tuominen 1981
found no clear difference between the two groups with regard to
the number...