Reducing Central Line-Associated Blood Stream Infections

Literature Review: Reducing Incidences of Central Line-Associated Blood Stream Infections

A bundle is a group of interventions related to a disease process, that when executed together, produce better outcomes than when implemented individually. Numerous studies done in the developed countries have shown that proper implementation of evidence based practices grouped together as central venous catheter bundle had brought a dramatic reduction in the incidence of CLABSI. Studies in developing countries had also shown high incidence of CLABSI and reduction in CLABSI rate albeit lesser than that of developed countries.

Studies from India have a shown a higher incidence of CLABSI and poor adherence to central line catheter bundle. Morbidity and mortality due to CLABSI is considerably high despite underreporting of such events. The development and publication of guidelines often does not lead to changes in clinical behavior and guidelines are rarely if ever, integrated into bedside practice in a timely fashion. The most effective means for achieving knowledge transfer remains an unanswered question across all medical disciplines. Our study aims to determine the compliance with CVC bundle in management of patients in medicine wards and ICU at All India Institute of Medical Sciences New Delhi and the impact of intervention in the form of periodic physician education and feedback in compliance with CVC bundle and central line catheter related complications.

Review of literature

Ever since the introduction of central venous catheters in the early 1950s, it had varied uses and later numerous studies revealed that it was associated with a varied number of complications. Although mechanical complications were common in the early years of CVC use CLABSIs quickly became recognized as a serious complication associated with their usage.

Central venous catheter

is defined as a catheter whose tip terminates in the great vessels. The great vessels are the aorta, pulmonary artery, superior vena cava,inferior vena cava, brachiocephalic veins, internal jugular veins,subclavian veins, external iliac veins, common iliac veins,femoral veins, and, in neonates, the umbilical artery/vein.

Types :

Common types of central venous catheters are Non tunneled catheters, Tunneled catheters, Implantable ports, PICC.

A peripherally inserted central catheter (PICC) is peripherally placed, but is considered a central catheter because its tip terminates in the central circulation. These venous catheters can also have single, double or triple lumens although single lumens are frequently used and are for intermediate to long term therapy for blood draws or infusions.

Complications of CVC:

Mechanical complications



Risk of complication at catheterization site

Internal jugular




<0.1 to 0.2

1.5 to 3.1




0.4 to 0.6


Arterial punctures (%)





Low risk (into inferior vena cava,

passing through right atrium)

High risk (crossing to

ascending internal jugular vein ,contralateral subclavian


Low risk (lumbar

venous plexus)

Pneumothorax is a common complication with subclavian and IJV cannulations without the use of ultrasound. The use of real time ultrasound reduces the number of attempts and associated with a significantly lower failure rate with internal jugular vein ( Relative risk 0.14, 95% confidence interval 0.06 to 0.33).

Limited evidence also exists for sublclavian and femoral routes in this metanalysis. Thus the chances of pneumothorax will be greatly reduced.

Pneumothorax is usually apparent immediately on Chest X rays and management may vary from simple observation to ICD placement with needle drainage needed for tension pneumothorax as emergency(11). Delayed pneumothorax is also known to occur with an incidence of 0.4%, more common with subclavian and with multiple attempts(12). Bedside ultrasonography allows diagnosis of pneumothorax to be made immediately with high sensitivity by clinician but is operator dependent(13).

Misplacement of catheters occur commonly such as tip malposition or rarely such as within artery. It is common practice to assess tip position lying above carina for right sided catheters assuming pericardial reflection below carina and below carina for left sided catheters in view of acute angulation to superior venacava(14). Management varies depending on the complication such as repositioning of tip for tip malposition lying below carina or when lying with an artery, interventional radiologist or vascular surgeon opinion is sought and removed accordingly(15).

Arterial injuries are more common with femoral and internal jugular rather than subclavian approach. A systematic review of complications of central venous catheters revealed significantly more arterial punctures (3.0% vs 0.5%) and less malpositions(5.3% vs 9.3%) with jugular access(16). It leads to hematoma in approximately 40% of patients. The best way to prevent arterial injury is by ultrasound assistance during cannulation(17). Other rarer complications are local hematomas,cerebrovascular accidents mostly seen with arterial injuries via internal jugular access, arrhythmias, perforation of the vein or right atrium, chylothorax, pseudo aneurysm, AV fistulas, cardiac tamponade, guidewire loss and catheter embolisation etc. have been reported. These complications largely depend on the site of insertion and on operator experience. Such complications can be prevented by ultrasound guidance and proper techniques.

Infectious complications are most dreaded as it is associated with mortality rates upto 25% and in developing countries even up to 60% and prolong the duration of hospital stay and are largely preventable. Evidence based guidelines have been developed as the central venous catheter bundle which significantly decreases the incidence of infections as shown in below studies.CVC use in non ICU settings is associated with at least a 2 fold rise in infection rate than in ICU settings. However studies are very limited on the infection rates as well as on the preventive measures in non ICU settings.

Thrombotic complications

range from 1.2 to 3 % in subclavian veins to up to 8 to 34% in femoral cannulations. Merrer et al in a randomized control trial found significantly increased incidence of thrombotic complications(21.5% vs1.9%,p<0.001)(18).This can be avoided by judicious site selection and proper flushing techniques (9).

Every day the central venous catheters are accessed for a variety of purposes which may include frequent CVP monitoring to IV infusions. Every time the catheter is accessed it should be done in a sterile way after the port is scrubbed with antiseptics else the infection rate increases. Dressings should be changed regularly depending on the type and as indicated.

Risk factors for CLABSI :

Intrinsic factors

Extrinsic risk factors

Age – children more likely

Prolonged hospitalization before CVC insertion

Underlying diseases or conditions—hematological

and immunological deficiencies, cardiovascular disease,

and gastrointestinal diseases

CVC duration, with the risk increasing with CVC dwell time

Male gender

Parenteral nutrition administration

Femoral or internal jugular access site

Multilumen CVCs

Lack of maximal sterile barriers

CVC insertion in an ICU or emergency department

Heavy microbial colonization at insertion site

The semi quantitative analysis of culture of


proved to be an effective and cost effective measure for the diagnosis of central venous catheter infections(19).

The study conducted by

Pronovost et al

in US which five basic measures hand washing, using full-barrier precautions during the insertion of central venous catheters, cleaning the skin with Chlorhexidine, avoiding the femoral site if possible and removal of unnecessary catheters significantly reduced the incidence of central venous line infections and served as a landmark study in the development of central venous catheter bundle(5). The study was conducted across 103 ICUs in Michigan in US and the rates of infection per 1000 catheter days were measured at 3 monthly intervals and the mean rate of infections decreased from 7.7 to 1.4 after 18 months.

After this study numerous studies were conducted which reinforced the effectiveness of the central venous catheter bundle.

Parra et al demonstrated that even a simple educational program like 15 min lecture given to ICU personnel highlighting 10 evidence based strategies can result in a reduction in CLABSI rates. In his study CLABSI rates decreased from 4.22 to 2.94 infections per 1000 catheter days(20).


conducted the

Bacteremia zero project

(21) to assess the applicability and effectiveness of the Michigan keystone ICU project in Spanish ICUs. This multifactorial nationwide intervention project was implemented between April 2008 and June 2010, with data collected at regular intervals to evaluate the progress of the project. A total of 192 ICUs (68% of all Spanish ICUs) participated in the project. The intervention was effective in reducing the incidence of CRBSI by approximately 50%in hospitals.

Burrel et al

demonstrated the benefit of having a checklist incorporating both the clinician and patient bundle(22). During this study they found that the compliance with the checklist improved significantly and the infection rate decreased by 50% when the compliance was good.

A nationwide study was conducted in US by

Furuya et al

(23) to find out the central line bundle implementation in ICU and its impact on Bloodstream infections. They found that CL bundle is associated with lower infection rates only when compliance is high. Complying with any one of three CL Bundle elements resulted in decreased CLABSI rates of 38%. This study clearly demonstrated the compliance should be very high to show a demonstrable decrease in the central line infection rate and that on-going evaluation was necessary.

Previously it was uniformly thought that femoral line insertion were associated with a greater incidence of infections .However recent studies have shown that the difference to be insignificant. A study was conducted by

Kedar S Deshpande

(24) in New York where they found that there was no stastically significant difference in the incidence of infections between the three routes in case of major infections when catheter is inserted optimally, catheter care is performed by trained by trained intensive care unit staff although there was a higher case of infections in the femoral group(0.881/1000 catheter days subclavian, 2.0/1000 catheter days internal jugular vein, 5.96/1000 catheter days femoral catheter p=0.1338)

Blood stream infections :

NNIS (National Nosocomial Infection Surveillance system) now renamed as


(National Healthcare Safety Network) defines blood stream infections as presence of a recognized pathogen cultured from one or more blood cultures and organism cultured from blood not related to infection at another site or presence of at least one of fever, chills and hypotension with signs and symptoms and positive results not related to infection at another site and presence of at least one of the following: Common skin contaminant (e.g.diphtheroids, bacillus species, propionibacterium species, coagulase negative staphylococci or micrococci) cultured from two or more blood samples drawn on separate occasions or Common skin contaminant cultured from at least one blood culture in a sample from a patient with an intravascular catheter or Positive antigen test on blood (e.g.,



Streptococcus pneumoniae


Neisseria meningitidis

, or group B streptococcus)(25).


Further 2 distinct terminologies are used in relation to central line infections these are used interchangeably usually though they are different.



I ( central line associated blood stream infections) is defined as BSI if a CVC was present at the time of or within 48 hrs before the defining blood culture was obtained. There is no requirement to identify the organism on the catheter. This definition was developed for surveillance, not for diagnosis.

In contrast, a


(catheter related blood stream infection) requires that the CVC be in place at the time the positive blood culture was obtained and a positive quantitative or semi quantitative culture of the same organism from the catheter or time to positivity.

The CLABSI definition is more practical than the CRBSI definition for surveillance. However, it may overestimate the true rate of CVC–related infections, as it can sometimes be difficult to determine infections related to the central line rather than remote unrecognized infections (for example, urinary tract infections, pneumonia, intra-abdominal abscess). Interobserver variability and a lack of standardization in CLABSI surveillance are other important limitations(26).

Usually the organism grown on culture is likely to be significant only when atleast 15 CFU or 10 *3 colonies are isolated of the same organism.

Catheter infection and colonization:

Catheter infection and colonization can occur by 3 routes –

  • Extra luminal from organisms migrating along the catheter surface,
  • Intraluminal through the ports or through contaminated fluids rarely,
  • Hematogenous from a source of infection elsewhere in the body.

Terminologies commonly used with regard to catheter infection and colonization are defined as follows (24)

1. Catheter infection

A. 15 CFU on catheter tip with the same bacteria and sensitivities (one different sensitivity included) in one or more blood cultures.

B. Resolution of clinical signs and symptoms of infection in 24–48 hrs after catheter removal regardless of bacterial growth either in blood or on catheter tip.

C. 15 CFU on catheter tip with local signs of infection regardless of blood culture bacterial growth.

2. Catheter colonization

A. Catheter tip with 15 CFU without growth in blood cultures and another source of infection found.

B. Catheter tip with 15 CFU and bacteria in peripheral blood cultures from another source of infection.

3. Contamination: <

15 CFU on catheter tip without bacterial growth in blood cultures or bacteria in blood cultures from another source of infection.

Till date most of the studies have emphasized in catheter related infections as proposed by NHSN but as of date due to the aggressive use of broad spectrum antibiotics culture negativity is more common. So to remove this confounding factor any fever in a patient with central line which has no proven focus of infection elsewhere and resolves within 2 days after removal of central line can be taken as CLRI. This is based on the fact that catheter infection is “cured” only with removal or a prolonged course of intravenous antibiotics; the relatively short courses of antibiotics used for ICU infections could not truly affect the catheter infection end point. However antibiotic use can result in negative cultures results even in the presence of catheter infection.

This definition was implemented in a study by

Kedar S Deshpande et al

(24)in New York where they found that the overall incidence of central line infection is low whatever the route may be and there was no stastically significant increase in infections in the femoral route provided proper strict aseptic insertion and maintenance techniques are followed. Thus the dangerous mechanical complications of pneumothorax can be avoided.

Developing countries scenario:

Rosenthal et al

(27)showed that the neonatal blood stream infections to be 5 times higher in developing countries than the developed countries. INICC conducted a surveillance study in 36 countries involving 422 ICUs and found a overall increased incidence of HAI. CLABSI rates were found to be 3 times higher when compared with USA (6.8 per 1000 CL days vs 2 per 1000 CL days). There was increased resistance to multiple gram negative organisms and MRSA was also more frequent in developing countries(28). Unadjusted crude excess mortality rate was also higher in developing countries both for CLABSI and VAP in the range of 7.3% to 15.2%

Reasons for the differences in the degree of burden of HAIs in developing countries include the following: Limited knowledge and training in basic infection prevention and control , Limited awareness of the dangers associated with HAIs , Inadequate infrastructure and limited resources, Poor adherence to routine hand hygiene, Reuse of equipment (for example, needles, gloves) , Poor environmental hygiene and overcrowding, Understaffing, Inappropriate and prolonged use of antimicrobials and invasive devices , Limited local and national policies and guidelines, Variable adherence to official regulations or legal frameworks, where they exist and Insufficient administrative support.

A quasi experimental study was conducted by Apisarnthanarak et al in Thailand where he found improved adherence to central line bundle particularly hand hygiene improved significantly after intervention from 8% to 54% and CA-BSI decreased significantly from 14 per 1000 catheter days to 1.4 per 1000 catheter days(29) .

Indian scenario:

Indian studies have shown that the incidence if central venous catheter infections are 3 to 4 times higher in our settings and that gram negative bacteria and fungi are more common in our settings compared to gram positive infections in western settings(30)(31), The Study conducted by

Pawar et al

in Escort hospital revealed that gram negative bacilli had a higher prevalence ( 71% ) as against western settings where gram positive organisms account for 70 % . They also found that the duration of catheterization, coexistent infections and increased temperature were the important predictors of CLABSI.

A study was conducted in 12 Indian ICUs by


which revealed a higher incidence in our settings and a overall poor adoption of the catheter bundle practices and the incidence gradually and progressively decreased after proper practices (6,18). Prospective surveillance in Indian ICUs yielded a central venous catheter-related bloodstream infection (CVC-BSI) rate of 7.92 per 1000 catheter-days(18). The incidence of central venous catheter infections then decreased subsequently after the adoption of proper practices. The study also revealed that proper education; performance feedback and outcome and process surveillance of CLABSI rates significantly improved infection control practices and brought a 54% decline in CLABSI incidence(6)



conducted a prospective before and after cohort study in eight Indian cities where they found implementation of six components of INICC simultaneously resulted in a significant reduction in CLABSI from 6.4 CLABSIs per 1000 central line days to 3.9 CLABSIs per 1000 central line days resulting in 53% CLABSI rate reduction which was highly significant(31).

Some studies had shown that empowering the nurses with the central venous bundle and to interfere if the proper practices are not followed will go a long way in the further reduction of CLABSIs. However in spite much of the reported successes in ICU population the non ICU population are still at a significant risk of infection. In developing countries empowering the nurses is still a long way to go as understaffing and proper techniques are still not adopted(15,6). Still application of inexpensive and practical infection prevention efforts, such as improved hand hygiene and removal of CVCs when they are no longer needed, can have a major impact on CLABSI rates.

IHI central venous catheter bundle was implemented as a project in 5 million lives campaign(32). Essential features are

Hand Hygiene


The cornerstone of WHO’s “

Clean Care Is Safer Care

” campaign, the “

My 5 Moments for Hand Hygiene

” approach, has resulted in the development of resources, including localized country-specific tools, to facilitate adherence to hand hygiene guidelines(33).

To minimize the risk of CLABSI associated with direct contact of the hands of health care personnel, the 2011 USCDC guideline recommends that hand hygiene be performed at the following times: before and after palpating the site of catheter insertion, before and after inserting the catheter, before and after accessing, replacing, repairing, or dressing the catheter. In addition, after the antiseptic has been applied to the site, further palpation of the insertion site should be avoided, unless aseptic technique is maintained (34).

Adherence to hand hygiene is generally suboptimal with rates under 40 % in multiple studies. Improving hand hygiene can be achieved through multiple educational interventions.

Maximum sterile barrier precautions


Raad et al

conducted a prospective randomized control study to determine the effectiveness of the maximum sterile precautions .they found that the control group had 6 times higher infection rates and the infections occurred early and mostly were caused by skin microorganisms(35).

Skin preparation:

Reducing colonization at the insertion site is a crucial part of CLABSI prevention. It can be done with aqueous povidone iodine, aqueous chlorhexidine, alcoholic chlorhexidine or alcoholic povidone iodine. A recent meta analysis revealed that the 2 %chlorhexidine is associated with a 50% decrease in the CLABSI rates compared to povidone iodine(36). However, a recent study by

Furuya et al

(23) identified the importance of allowing chlorhexidine to dry fully before CVC insertion in order to optimize the use of this agent. An economic analysis suggested that using chlorhexidine rather than povidone iodine would result in a 1.6% decrease in CLABSIs and a 0.23% decrease in mortality, as well as save $113 per catheter used(37). The proposed mechanism is believed to be prolonged antimicrobial effect, its lack of inactivation to blood and serum and synergistic effect with alcohol. However whether chlorhexidine alcohol combination is superior to alcoholic povidone iodine is still unresolved. The role of alcohol as antiseptic agent is often forgotten while describing chlorhexidine(38). CDC recommends >0.5% chlorhexidine alcohol to be used in case of central line insertion. However a recent Indian study conducted in 2013 by Kulkarni et al found that both povidone iodine and chlorhexidine had zero colony counts in skin preparation(39).


Catheter site dressings can be done by transparent to semitransparent dressings or gauze dressings. The advantage with sterile dressings is that visual inspection can be done daily and they need to be changed once in 7 days unless gauze dressings which needs to be changed every 2 days. However if patient is in DIC or there is bleeding at the site of insertion gauze dressings are preferred(40).

The current recommendations are to assess the continued need for the catheter every day, perform catheter site care with Chlorhexidine at dressing changes, replace administration sets and add-on devices no more frequently than every 72 hours, unless contamination occurs, replace tubing used to administer blood, blood products, or lipids within 24 hours of start of infusion, change caps no more often than 72 hours or according to manufacturer’s recommendations and whenever the administration set is changed(41).

Shapey et al conducted a prospective audit in a university teaching hospital and found that several breaches were there in CVC post insertion care with a failure rate of 44.8% mostly in caps, dressings and proposed that focus should be shifted to best practice implementation rather than further teaching(42)

The significance of catheter maintenance or post insertion bundle was demonstrated by Guerin et al who showed that implementation of a post insertion bundle in hospitals with good compliance to insertion bundle resulted in decrease in CLABSIs from 5.7 to 1.1 CLABSIs per catheter days. He emphasized that insertion bundle by itself was not sufficient. The interventions done were assessing the catheter site daily, changing dressings if necessary, application of chlorhexidine sponge at the catheter site, assessing the need for catheter daily, performing hand hygiene and alcohol scrub before accessing hub each time(43).

Using closed container systems than open container infusion systems had significantly reduced the incidence of CLABSI in many studies. This was demonstrated by Maki et al who conducted a study in 4 countries and found that switching to closed infusion containers decreased CLABSIs from 10.1 to 3.3 per 1000 catheter days RR 0.33 p<0.001(44).

Prompt removal of catheters:

Zingg et al

in his study found that in several site visits, neither the nurse nor the treating physician knew why the patient had a CVC particularly in non ICU settings. They also showed that catheters in non ICU settings had a longer dwell time and lesser utilization(45) .

Trick et al

in his study found that 4.6% of catheter-days were not justified. Both of these research groups also found differences in CVC use between ICU and non-ICU settings: unjustified CVC–days were more common in the non-ICU settings. The median duration of CVC days were more common for catheters that were unjustified rather than justified. They also proposed as to consider removal of CVC if needed when patient is getting transferred out of ICU (46).