CVCs are divided into two categories: non-tunneled and tunneled. Each catheter type has specific line care needs, advantages, disadvantages and complications (Table 17.2). Selection of the optimal type of CVC for use in a specific disease or treatment protocol is not standardized. Factors to consider in catheter selection include the age and weight of the child, the length and intensity of therapy, frequency of blood sampling, anticipated supportive care interventions including transfusions, infusions and nutrition, level of patient activity, body image, and family ability to understand teaching and properly care for the line.

Though rare, complications during insertion can arise and cause significant morbidity and include pneumothorax, hemothorax, chylothorax, malpositioning, arterial puncture and failure to place. Factors associated with complications include physician inexperience, multiple insertion attempts, prior catheterizations, patient anatomy, prior surgery or radiation in the area and a high body mass index (Mansfield et al. 1994; Lefrant et al. 2002; Kusminsky 2007). An insertion failure rate of up to 43 % and complication rate up to 24 % occurs with ≥3 insertion attempts leading to a recommendation of limiting each operator to a maximum of two unsuccessful attempts (Mansfield et al. 1994; Eisen et al. 2006). The definition of an “attempt” varies among studies ranging from one puncture to multiple punctures by one operator at one site making comparisons difficult (Eisen et al. 2006; Balls et al. 2010).

Several studies have been completed evaluating the advantage of using real-time ultrasound-guided assistance (UGA) rather than the anatomic landmark technique for placement of a CVC (Augoustides and Cheung 2009; Pittiruti et al. 2009; Balls et al. 2010). A meta-analysis by Randolph et al. (1996) concluded that this technique led to an improved insertion success rate and a decrease in complications in both internal jugular and subclavian vein insertions. McGee and Gould (2003) found that UGA is effective in catheterization of the internal jugular vein with a decreased incidence of mechanical complications and placement failure. However, they found no benefit using this technique with subclavian vein insertions as the clavicle lies directly over the vessel, impeding visualization. In their studies, Mansfield et al. (1994) and Troianos et al. (2011) reached a similar conclusion. A retrospective observational study by Balls et al. (2010) assessed 1,222 CVC placement attempts concluding that the use of UGA did not improve the success of placement on the first attempt but overall saw a reduced number of total attempts. Further study is required to determine whether the routine use of UGA is feasible due to the high cost of equipment, required personnel training and equipment maintenance (Randolph et al. 1996).

Catheter insertion in the subclavian vein carries a higher risk of pneumothorax, malpositioning and failure to place compared to internal jugular insertion, while internal jugular catheterization is associated with a higher incidence of arterial puncture and hematoma (McGee and Gould 2003; Eisen et al. 2006). Although the subclavian vein carries the greater risk of insertion complications, it remains the preferred approach due to a lower rate of infection noted in some studies (McGee and Gould 2003). A prospective, observational study by Deshpande et al. (2005) found no difference in CVC infection rates for subclavian, internal jugular or femoral vein insertion sites in adult patients. The 2011 Centers for Disease Control (CDC) Guidelines for the Prevention of Intravascular Catheter-Related Infections declined to make a recommendation for the preferred CVC insertion site leaving the issue unresolved (O’Grady et al. 2011).

Children with acute lymphoblastic leukemia (ALL) often present with neutropenia and thrombocytopenia theoretically putting them at increased risk for complications with CVC placement. However, two separate studies evaluating 172 and 98 children, respectively, found no increased rate of complication with early CVC placement in newly diagnosed ALL patients (Handrup et al. 2010; Gonzalez et al. 2012). Platelet thresholds for CVC placement are undefined (see Chap.​ 2). Handrup et al. (2010) also concluded that the nonelective removal rate was similar between early and later placed CVCs. A retrospective analysis of 362 patients with ALL assessed complication rates between timing of insertion (early, ≤day 15 of induction, vs. late, >day 15 of induction) and type of CVC (ports vs. external CVCs) and found that early placement was associated with an increased risk of a positive blood culture and external CVCs were associated with an increased risk of positive blood cultures, thrombotic complications, and early removal (McLean et al. 2005). Due to the conflicting evidence, institutions providing initial care of newly diagnosed oncology patients must decide on the benefit of CVC placement timing, with ongoing monitoring for early complications and of line care in this setting.

Infection remains the major complication of an indwelling CVC, with bloodstream infection causing the most significant risk of morbidity and mortality. Terms used to describe intravascular catheter-related infection are confusing with central line-associated bloodstream infection (CLABSI) and catheter-related bloodstream infection (CRBSI) often used interchangeably. CLABSI is defined as an infection occurring in the patient with a CVC and not related to an infection at another site and is the term used by the CDC National Healthcare Safety Network (NHSN). CRBSI is a clinical definition requiring specific laboratory testing, quantitative blood cultures, differential time to positivity or culture of a segment of the removed catheter (O’Grady et al. 2011). Common organisms causing CLABSI include Staphylococcus epidermis, Staphylococcus aureus, Enterococcus faecalis, Klebsiella pneumoniae, Pseudomonas aeruginosa and Candida albicans. See Chaps.​ 1 and 14 for prevention, recognition and treatment of suspected infection or sepsis.

Most CVC-related infections are thought to occur by one of two methods: colonization at the exit site with pathogen migration along the external catheter surface or hub contamination leading to intraluminal colonization with spread into the circulation (McGee and Gould 2003). Within hours of CVC placement, a protein-rich sheath begins developing, covering the external and internal surfaces of the catheter. The protein sheath allows adherence of microbes which then produce a slimy substance (biofilm) becoming embedded in the matrix (Raad et al. 1993). Pathogens within a biofilm behave differently with an increased rate of reproduction and a greater resistance to antimicrobial therapy (Raad et al. 1993; Donlan 2011).

Antiseptic-impregnated catheters (AIC) coated with either chlorhexidine and silver sulfadiazine (CSS) or minocycline-rifampin (MR) have been studied in an effort to determine their effectiveness in decreasing the rate of CLABSI. Randomized clinical trials have generally not shown these catheters to be beneficial (McGee and Gould 2003). In a randomized clinical trial evaluating 232 catheters inserted in 180 critically ill hospitalized adult patients in use <10 days, there was no significant difference in the rates of colonization between antiseptic-impregnated and non-impregnated catheters (Theaker et al. 2002). Separate meta-analyses reviewing randomized controlled trials comparing AICs and non-AICs with a median insertion duration of 7–12 days concluded the efficacy of CSS catheters to be <2 weeks with MR catheters being effective somewhat longer (Mermel 2000; Walder et al. 2002). The results of numerous studies are difficult to compare with no type of catheter showing a definitive advantage. The CDC recommends that institutions develop strategies to provide education of personnel who insert and maintain catheters, with use of maximal sterile barrier precautions (i.e., cap, mask, sterile gown, sterile gloves and a sterile full body drape for line insertion) and skin antisepsis with >0.5 % chlorhexidine with alcohol for insertion. The use of antimicrobial-coated catheters is recommended at institutions where implementation of these CDC strategies fails to decrease CLABSI rates (O’Grady et al. 2011). Maki et al. (2006) determined that institutions with a baseline CLABSI rate of >2 % would benefit from use of AICs as this was the threshold at which AICs would decrease overall costs.

A functioning CVC is a catheter that flushes easily, infuses without difficulty and has brisk blood return (Baskin et al. 2009). Occlusion is the most common noninfectious complication of CVCs with an occurrence rate of 25 % and resulting in delays in the administration of chemotherapy and supportive care (Stephens et al. 1995). Rapid assessment is needed to determine both the cause of the obstruction and the appropriate interventions. Causes of CVC occlusion include mechanical complications, drug precipitate or lipid residue and thrombosis. Each of these problems can result in either partial or complete obstruction of the catheter. A partial occlusion allows for fluid infusion but either a sluggish or complete inability to withdraw blood (ball-valve effect). A complete occlusion allows neither fluid infusion nor blood withdrawal. An unusual problem may occur with implanted ports allowing blood withdrawal but not fluid administration due to a thrombus inside the reservoir at the outlet port (a reverse ball-valve effect).