The history of central venous cannulation began when a young surgical resident cannulated his antecubital vein in 1929. He watched the catheter progress 65 cm forward up his auricle, in a mirror held in front of a fluoroscope screen.1 Using this maneuver, the technique of vascular catheterization was born. Twenty-seven years later, this scientist, Werner Forssmann, was awarded the Nobel Prize, along with Andre Cournand and Dickson Richards, both of whom put this procedure into practice.1 Central venous catheters (Table 1) did not begin to gain widespread popularity until an article in 1962 by Wilson and his colleagues2 described the advantages of measurement of central venous pressure.3 The 1970s brought the description and employment of tunneled Silastic catheters described by both Broviac and Hickman and their colleagues.3

TABLE 1 Definition of Key Terms

Central Venous Catheters and Their Risks

Reliable vascular access is an essential tool when delivering modern day pediatric critical care therapeutics. Although these catheters can be lifesaving, they also carry several risks. These risks include local infections, catheter-related bloodstream infections (CRBSI), septic thrombophlebitis, endocarditis, metastatic abscesses, and mechanical complications during insertion. 4-6 The incidence of CRBSI varies considerably, based on site preparation, type of catheter, underlying patient diagnosis, and site of catheter placement. In the intensive care unit (ICU), rates of CRBSI are higher than in less acute or ambulatory settings. Some of these catheters are placed under suboptimal conditions due to emergent situations under which sterile conditions may have been omitted. In addition, hospitalized children may require indwelling catheters for extended periods of time and may become colonized with hospital-acquired organisms due to frequent catheter manipulations to draw laboratory specimens and to deliver intravenous medications, nutrition, and blood products.7-9

CRBSI are considered "silent" medical errors since they may be caused without operator knowledge during placement, site care, or line manipulation. Since there is a time delay before the infection presents, the "cause" and "effect" are often not correlated. Nosocomial bloodstream infections contribute to hospital-associated morbidity, mortality, and economic burden.10-13 Advanced practice nurses (APNs) are in a vital position to influence central venous catheter selection, site selection, insertion technique, and line care, as well as to educate other nursing and physician staff on the latest recommendations to promote best practices.

Epidemiology of Catheter-related Bloodstream Infections

The Centers for Disease Control and Prevention (CDC) and the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) recommend that the rate of CRBSI be expressed as the number of catheter-associated bloodstream infections per 1,000 catheter days. This parameter has more utility than expressing the number of catheter-related infections per 100 catheters (or percentage of catheters studied), as it accounts for CRBSI over time, adjusting risk for the number of days that the catheter is in place.8

Similar to adults, most bloodstream infections in children are associated with the use of intravascular devices.6,8,10,14 In the United States, CRBSIs occur with an estimated rate of 3%-7% and an estimated associated cost of $9,000-46,000 per episode.4,13,15,16 In adults, CRBSIs have been estimated at 2.1 bloodstream infections per 1,000 central catheter days in respiratory ICUs to 30.2 per 1,000 central catheter days in burn ICUs.7 In pooled pediatric intensive care unit (PICU) data from 1992-2003, obtained from the CDC's National Nosocomial Infection Surveillance System report in 2003,17 a CRBSI rate of 7.3 per 1,000 catheter days was reported. In PICUs, bloodstream infections (28%) are the most common nosocomial infection followed by ventilator-associated pneumonia (21%).9,10,11,18 Slonim et al13 found length of stay attributable to nosocomial bloodstream infections was 14.6 days for PICU and 21.1 days for the hospital ward patients. They13 also documented costs attributable to bloodstream infections to be $46,133. Excess length of stay was responsible for most of the associated costs.

Risk Factors for Catheter-related Bloodstream Infections

Risk factors for development of CRBSI in critically ill pediatric patients are different than those reflected in adult data.18 Critically ill children have varying degrees of immunocompromise arising from young age, underlying disease process, chronic stress, or immobility. Pediatric risk factors for CRBSI include process of care, specifically multiple central venous catheters, arterial catheter presence, procedures performed in the PICU, and transport out of the PICU.18

In very low birth-weight infants (those weighing < 1,000 grams), exposure to lipids has been identified as an independent risk factor for development of coagulase negative staphylococcocal bacteremia and candidemia in the neonatal ICU.19

Singh-Naz and colleagues11 noted that children receiving parenteral nutrition and antimicrobial therapy were associated with the highest risk of developing nosocomial infection (including lower respiratory tract, bloodstream, and urinary tract). After controlling for other factors, their study11 revealed that children receiving parenteral nutrition were 22 times more likely to develop nosocomial infection than those not receiving parenteral nutrition. Also, patients receiving more than 10 days of antimicrobial therapy were more than 5 times more likely to develop a nosocomial infection.11 This study was the first prospective cohort study investigating risk factors for nosocomial infection alone and in combination.11 In Singh-Naz's epidemiologic study,11 the identified significant risk factors for the development of pediatric ICU nosocomial infections included younger age, weighing less, higher admission day pediatric risk of mortality (PRISM) scores, and immune dysfunction. Children who require cannulation for extracorporeal life support have a 10-fold increased risk of developing CRBSI, and children requiring vascular access for renal replacement therapy had a 4-fold increase in CRBSI.7 Mortality associated with CRBSI has been estimated at 10%-25%.13,15,16,20-22

Microbiology of Pediatric Catheter-related Bloodstream Infections

Organisms most commonly associated with pediatric CRBSI change over time. Data published by the National Nosocomial Infections Surveillance System (NNIS) in 1999,10 revealed coagulase negative staphylococci (38%) bloodstream infections as most predominant in the PICU, followed by gram-negative rods (25%).16,18

Antibiotic resistance is becoming an ever increasing battle in critically ill children, especially among microorganisms that cause resistance.23 The predominant driving force for the rapidity of antimicrobial resistance emergence has been the gross overuse of antibiotics for conditions that do not require treatment or for which the selected antibacterial agent is ineffective.23,24 Modern medical therapeutics such as invasive surgical procedures, transplantation, oncologic therapies, and prolonged intensive care unit stays also create a large consortium of debilitated patients requiring central venous catheterization, courses of broad spectrum antibiotics, and prolonged hospitalizations. All of these factors, combined with ineffective hospital infection control practices have contributed to the emergence of vancomycin-resistent enterococci (VRE), methicillin-resistant Staphylococcus aureus (MRSA), Clostridium difficile, and an extended spectrum of multidrug-resistant, gram-negative bacilli, and Candida.23,24 Development of disseminated fungal infections, such as Candida species, have increased 11-fold in the United States in the last two decades.24 More restricted use of antimicrobial agents, particularly cephalosporins, in addition to improving strategies to decrease device-related infections, would likely reduce the emergence of all types of resistant organisms.23

The JCAHO now mandates hospitals to periodically audit their use of antimicrobial agents.23 Such audits evaluate the indications for antibiotic usage.

Pathogenesis of Pediatric Catheter-related Bloodstream Infections

For bacteria to cause CRBSI, they must gain access to the extraluminal or intraluminal surface of the catheter. Most commonly, skin organisms migrate from the insertion site through the cutaneous catheter tract, leading to colonization of the catheter tip at the time of insertion or days later.4,8,14,21 Additionally, catheter hub contamination can contribute significantly to intraluminal colonization of intravascular catheters.4,8 Occasionally, hematogenous spread from an alternate site of infection, such as pneumonia, may cause infection.21 Rarely, infusate contamination leads to CRBSI.8,21

After gaining access, the organisms can adhere to the device and become incorporated into the "biofilm" that permits sustained infection and hematogenous dissemination.4 Fibrin sheath and biofilm production, as a host reaction to the catheter material, is known to occur within 24 hours of catheter placement.7 This formation is known to increase with prolonged catheter placement.7

Strategies to Reduce Risk of Catheter-related Bloodstream Infections

Site Selection

Site selection for catheter placement has been evaluated in adult patients. Limited data are available for pediatric patients. Density of local skin flora at the selected site is a primary offender in causing CRBSI.8 Adult studies recommend that central venous catheters (CVCs) be placed in the subclavian site instead of jugular or femoral sites to reduce infection risk.8 The CDC recommends avoiding femoral catheters in adults as they have been associated with high infection risk and are associated with higher risk of deep vein thrombosis. 8 In children, femoral catheters have demonstrated a lower incidence of mechanical complication and may have an equivalent incidence of infection than catheters in alternate sites.5,8,25,26 An advantage of using the femoral vessel cannulation includes easily identifiable landmarks and avoidance of potential life threatening complications including pneumothorax, hemothorax, carotid artery and cardiac perforation, and thoracic duct perforation.5,27,28 Using the femoral vein to obtain vascular access mitigates the need to obscure the head and neck for potential simultaneous airway manipulation.29 The most common complications of femoral cannulation are arterial puncture, hematoma, leg edema, and thrombosis, which are generally not associated with significant sequalae. 5,25 When selecting line placement site, patient comfort, patient specific factors (eg, preexisting catheters, irregularities in hemostasis, anatomic anomalies), risk of complication (eg, bleeding risk, pneumothorax), infection risk, potential for ambulation, and operator experience should all be used to guide selection.8,29Table 2 lists considerations in site selection.

TABLE 2 Site Selection Considerations

Type of Catheter Material and CRBSI

Both the material of the catheter and the intrinsic virulence factors of the offending organism affect the pathogenesis of CRBSI. In vitro studies have demonstrated that microorganisms are more adherent to catheters produced of polyvinyl chloride or polyethylenes than catheters composed of Teflon, silicon elastomer, or polyurethane.30,31 For this reason, most catheters sold in the United States are no longer made of polyvinyl chloride or polyethylene.8 Some catheters also have surface irregularities that promote pathogen adherence of certain species.8

Multilumen Catheters and CRBSI

The data are inconclusive in regards to single lumen versus multilumen central venous catheters and infectious rates. Investigations have yielded mixed results, and it is unlikely that there is a substantial increase in CRBSI or catheter colonization when multilumen catheters are utilized instead of single lumen catheters.5,15 Choice of single, double, or triple lumen catheters should be made by the APN according to the anticipated number of lumens required to adequately administer medications, blood products, and nutrition.8

Hand Hygiene and CRBSI

Hand hygiene also contributes significantly to CRBSI risk. Data are limited on which types of patient-care activities result in transmission of patient flora to the caregiver.32 Routine "clean" nursing tasks (including taking a pulse, holding a hand, and taking a temperature or blood pressure) can result in contamination of the nurse's hands with gram-negative bacilli, Staphylococcus aureus, enterococci, or Clostridium difficile.32 Other studies have documented that healthcare workers can contaminate their hands by touching inanimate objects in patients' rooms.32 Hand hygiene technique prior to venous cannulation should include washing hands with soap and water. This includes wetting the hands first, applying an amount of product recommended by the manufacturer, rubbing the hands together for at least 15 seconds, covering all surfaces of the hands, then rinsing the hands thoroughly, drying with a disposable towel, and using the towel to turn off the faucet, if using a non-automated faucet.8,32 The CDC recommends observing proper hand-hygiene techniques by either washing hands with conventional antiseptic containing soap and water or with waterless alcohol-based gels or foams. Hand hygiene should be performed before and after palpating catheter insertion sites, as well as before and after inserting, replacing, accessing, repairing, or dressing an intravascular catheter.8 Palpation of the catheter insertion site should not occur after application of skin antiseptic unless aseptic technique is maintained.8

Catheter Insertion Asepsis

The CDC recommendations incorporate maintaining aseptic technique for the insertion and care of intravascular catheters. Maximal barrier protection (cap, sterile gown, mask, sterile gloves, and large sterile drape) should be utilized during central venous cannulation.8,33

Skin Antisepsis

The skin is the largest body organ as well as the first line of defense against a wide spectrum of bacterial pathogens.20 When the integrity of the skin is compromised, this line of defense is weakened. Antiseptics, or disinfectants, are chemical agents used to decrease the risk of infections. Chlorhexidine gluconate is active against a wide range of gram-positive and gram-negative bacteria, yeasts, and molds.20,32 It acts by disrupting cytoplasmic membranes and remains active hours after its application.20 Chlorhexidine is superior to povidone-iodine and other antiseptics in reducing colonizing flora immediately and has substantial residual activity after application.4,20,32,34 A large meta-analysis by Chaiyakunapruk et al,35 involving 4,143 catheters in adults, concluded that chlorhexidine gluconate reduced the risk for CRBSI by 49% when compared to povidone-iodine. A follow-up analysis by this same group calculated that the slightly higher cost of chlorhexidine compared with the decreased morbidity and mortality resulted in a $113 savings per catheter.36 The CDC recommends that a 2% chlorhexidine solution is utilized, although tincture of iodine, an iodophor, or 70% alcohol may be used.8 If povidone iodine is used, allow it to remain on the skin for at least 2 minutes or until dry prior to line insertion.8

There has been limited research in neonates and children regarding the use of chlorhexidine impregnated sponges (Bio-patch(TM)). One randomized neonatal study37 reported a significant decrease in colonized central line catheter tips in the Biopatch(TM) group compared with those neonates with standard dressings, but no difference in CRBSI. They37 also found an associated localized contact dermatitis in very low birth-weight infants. There is currently no recommendation for use of the chlorhexidine for infants aged <2 months.8

Catheter Site Dressing Changes

Local inflammation at the insertion site of a central venous catheter is uncommon with CRBSI. Most likely, this is because most CRBSIs are caused by coagulase negative staphylococci that provoke little local inflammation. Signs of inflammation at the catheter site should raise suspicion for CRBSI caused by Staphylococcus aureus or gram-negative bacilli, particularly if the patient is demonstrating fever or other signs of infection.22 Purulence, while rare, has been found to be highly predictive of CRBSI.22

Recommendations for catheter site dressing changes are extrapolated to pediatrics from adult studies. Central catheter site dressing changes should be performed utilizing sterile gauze or sterile transparent semipermeable dressing to cover the catheter site.8 Gauze dressings are preferable to transparent, semipermeable dressings in patients who are diaphoretic or who have oozing or bleeding catheter insertion sites. Catheter dressings should be changed at least weekly or if the dressing becomes damp, loosened, or visibly soiled.8

Catheter Securement Devices

Since inadvertent catheter dislodgement has serious implications, securing catheters reliably is imperative. Dislodgement can predispose the child to phlebitis, infiltration, and additional needle sticks. Various devices have been created to address this issue. Most recently, sutureless securement devices have been created. These devices include an adhesive anchor device (StatLock(R)) (see Figure 1) that has an adhesive pad to attach to the child's skin and a mechanism to which the catheter and tubing are attached. The anchor device is available in variety of sizes for neonatal and pediatric catheters. The neonatal securing device uses a custom-formulated adhesive on a breathable anchor pad that performs in both low and high humidity environments. It is also skin-safe for the fragile and underdeveloped neonate skin. These devices are latex-free and hypoallergenic. These features can be advantageous in preventing catheter-related bloodstream infections, as they obviate the need for sutures, which create a portal of entry through which bacteria can gain access to the vascular device. 8 A study by Yamamoto38 evaluating sutureless securement in peripherally inserted central catheters (PICCs) demonstrated a shorter average securement time compared to sutures, as well as a reduced number of catheter-related bloodstream infections, from 9.4% to 1.1%. They found no difference in catheter dislodgement between traditional sutures and the sutureless device.38 To date, there have been no studies evaluating the use of sutureless securement in pediatric patients.

Figure 1. Statlock(R) Sutureless Securing Device.

Antimicrobial/Antiseptic-impregnated Catheters and Cuffs

Certain antibiotic or antiseptic-impregnated catheters can reduce the risk for CRBSI. Table 3 summarizes some studies that have been completed in adults evaluating the utility of antibiotic-impregnated catheters.39-43 Studies released so far have been completed in double and triple lumen catheters with a dwell time of <30 days.39-43 Impregnated catheters are generally more expensive than non-impregnated catheters, but overall can lead to decreased hospital costs.44 One decision analysis model study, by Veenstra45 suggested that the use of antiseptic-impregnated catheters could return a savings of $196 per catheter. Although all of the studies on antibiotic-impregnated catheters have been completed in adult subjects, the Food and Drug Administration (FDA) has approved these catheters for use in infants and children >3 kg. No antiseptic/antibiotic-impregnated catheters are available for use in infants weighing <3 kg.8

TABLE 3 Selected Studies Evaluating Antiseptic-impregnated Catheters in Adults

TABLE 3 Selected Studies Evaluating Antiseptic-impregnated Catheters in Adults (Continued)

Two meta-analyses have demonstrated that the use of catheters with external impregnation with chlorhexidine/silver sulfadiazine reduced risk for CRBSI.45,46 These catheters are considered to be first-generation catheters. Second-generation catheters with chlorhexidine coating on both the internal and external catheter surfaces are now available.34 Theoretically, this would seem to increase reduction in CRBSI, yet studies have yet to be done that support this hypothesis.

Although there is a theoretical concern that antiseptic-impregnated catheters may have the potential to select organisms resistant to the antiseptic, this has yet to be demonstrated.34 Continued surveillance will be required as use of these coated catheters continues to rise. Some institutions are considering antiseptic-impregnated catheters only when CRBSI rates remain high, despite aggressive attempts at implementing the additional CDC recommendations to decrease the rate of infection.

Antibiotic/Antiseptic Ointments

Intermittent application of a polyantibiotic ointment was perhaps one of the first strategies used to decrease incidence of CRBSI.4 Four studies47-50 conducted from 1960-1980, did not show benefit for prevention of CRBSI. They did, however, demonstrate a 5-fold increase in catheter colonization with Candida species.4,47-50 Application of antibiotic ointments to catheter insertion sites increases the rate of catheter colonization by fungi, promotes emergence of antibiotic-resistant bacteria, but has not been shown to decrease the incidence of CRBSI.4,47-50 The CDC does not recommend routine application of polyantibiotic ointments or lotions at catheter insertion sites.8

Replacement of Catheters

Removal of intravascular devices with suspected or documented CRBSI in pediatric patients is often problematic due to limited venous access. In one pediatric ICU study,25 no relationship was found between duration of catheterization and the daily probability of infection, suggesting that routine replacement of central venous catheters likely does not reduce the incidence of CRBSI. No trials, pediatric or adult, have demonstrated catheter replacement at a new site or via guidewire exchange at scheduled intervals to lower infection rates.21,34 Labor and material costs dedicated to this practice do not appear to be justifiable, and mechanical complications associated with additional insertion is not defensible.34 This practice should be avoided. With extracorporeal membrane support therapy and CRBSI, the patient is anticoagulated, and the device should ideally only be removed upon discontinuation of extracorporeal therapy.7

Tunneling of Catheters

In critically ill children, central venous catheters are placed with the intent that they will be used as short-term vascular access. Residents, fellows, APNs, or attending physicians generally place these lines at the bedside. The primary site of bacterial entrance is the cutaneous insertion point.21,28,51 Tunneling, or burying the catheter under the skin, is often done for long-term vascular access in chronically ill children. The tunneled catheter provides a greater distance between the potentially contaminated skin entry site and the catheter tip.51 The practice of tunneling a catheter is based on the assumption that this method provides greater stability and reduced infection risk.51 Nahum et al28 found that tunneling femoral vessels in critically ill children is a safe procedure and is effective in reducing bacterial colonization compared with nontunneled catheters, but it did not significantly reduce bloodstream infection.28 At this time, there are no data to support routine tunneling of central venous access in pediatric patients.

Systemic Antibiotic Prophylaxis

No studies have documented benefit of enteral or parenteral antibacterial or antifungal agents in reducing incidence of CRBSI in adults.8 Two studies52,53 have evaluated vancomycin prophylaxis among low birth-weight infants. Both52,53 demonstrated reduction in CRBSI, but no reduction in mortality. Since prophylactic administration of vancomycin has been associated with VRE, the risk of developing VRE likely outweighs the benefit of vancomycin prophylaxis.8

Antibiotic Lock Prophylaxis

Antibiotic lock prophylaxis has been attempted by filling the lumen of the catheter with an antibiotic solution and leaving it to dwell in the lumen of the catheter as a way to reduce CRBSI. Three studies54-56 using a vancomycin containing solution have been done in neutropenic patients with long-term catheters. These studies54-56 demonstrated a significantly lower rate of CRBSI and length of time to first onset of vancomycin susceptible bacteremia. One pediatric study compared heparin flush alone to heparin and vancomycin flush and found no difference in CRBSI.8,57 Since the use of vancomycin is an independent risk factor to the development of VRE, this practice has not been endorsed by the CDC.8 Minocycline and ethylenediaminetetraacetic acid (EDTA), an antimicrobial/anticoagulant combination, has been proposed as a lock solution; however, no controlled or randomized controlled trials (RCTs) have demonstrated its efficacy.8

Replacement of Intravenous Administration Sets

CDC's recommendations support the safety and cost-effectiveness of replacing crystalloid intravenous (IV) administration sets no more frequently than every 72 hours.8,58-60 Data from a more recent adult study61 including adult cancer patients supports delaying the replacement of IV tubing up to 7 days if the patient is not receiving total parenteral nutrition, blood transfusions, or interleukin-2 though the IV tubing.

Needleless Vascular Access Devices

Needleless vascular access devices were introduced in an attempt to decrease needle-stick injury to healthcare workers.43,62 Conflicting data on associated CRBSI have been published.4,8,62-65 Inappropriate use of these devices may have been responsible for some instances of increased CRBSI.4 One small PICU study demonstrated an increased risk of infection when utilizing a first-generation needleless device when replaced every 6 days.64 The CDC reports that when used per manufacturer recommendations, needleless vascular access devices do not affect rates of CRBSI.8

Quality Assurance and Education

Although all of these recommendations may help the decline in CRBSI, they will not be successful unless there is adequate education for the healthcare workers who insert and care for patients with CVCs (see Table 4). Well organized educational programs that enable healthcare workers to provide, monitor, and evaluate outcomes are critical to the actualization of this effort. Reports8 over the last 20 years demonstrate infection rate decline following the implementation of standardized aseptic care and that insertion and maintenance of vascular devices by experienced staff may decrease risk of catheter colonization and CRBSI.

TABLE 4 Summary of Strategies to Reduce Risk of Catheter-related Bloodstream Infection

ICU groups have begun to evaluate the impact of education and bedside behaviors on decreasing catheter-related bloodstream infection.66,67 A focused intervention directed primarily at nursing staff can lead to a remarkable decrease in incidence of bloodstream infections.66 Although initial educational programs have led to modest improvements in bedside behavior practices and a decrease trend in CRBSI, continued educational initiatives and monitoring of bedside behaviors are required to identify deviations from recommended practice.67

Conclusion

Central venous catheters are frequently required to provide advanced therapeutics to critically ill children. Acquired nosocomial CRBSI remain a significant source of morbidity and mortality. The majority of literature available on CRBSI involves adult patient data, and some of this information may not be directly extrapolated to infants and children due to differences in age, underlying medical condition, process of care, and predominance of pathogen strains. Education should include evidence-based strategies to decrease incidence of infections. These strategies should be reviewed regularly with staff who insert and manage patients with CVCs in order to minimize risk. Regular surveillance of CRBSI, knowledge of local pathogens and susceptibility patterns, and implementation of evidence-based guidelines promote safety and decrease preventable infections in critically ill infants and children. Additional data are needed to estimate the attributable morbidity, mortality, and financial burden of CRBSI in critically ill children and provide optimal strategies for prevention.

Note: The reader is encouraged to read the 2002 guidelines published by the CDC8 for further information. Recommendations from the 2002 guidelines for the prevention of intravascular catheter-related infections published by the National Institute of Health, replacing previous guidelines from 1996.

References