Abeloff's Clinical Oncology, 4th Edition

Part II – Problems Common to Cancer and its Therapy

Section E – Surgical Problems

Chapter 52 – Establishing and Maintaining Vascular Access

John C. Mansour,
John E. Niederhuber




More than 750,000 vascular access devices are used in the United States each year.



Placing the catheter tip in the superior vena cava or inferior vena cava provides large lumen and high flow.



Types of central access systems are traditional central line for short-term use, tunneled central lines for long-term use, surgically implanted infusion ports, and peripherally inserted central catheters (PICC lines).



Three questions to ask in selecting a catheter system are: (1) What device best meets the patient's therapy needs? (2) How is the device most safely inserted and maintained? (3) What are the likely immediate and long-term complication risks?



Vascular access devices can be placed using a number of anatomic sites to access the superior vena cava or inferior vena cava: subclavian vein, internal jugular vein, external jugular vein, and femoral vein.



Insertion can occur via Seldinger technique (closed) or by operative exposure of vein (open) technique.



Complications include vascular laceration, arterial puncture, pneumothorax (2%), hemothorax, and air embolus (overall placement complications should be <5%).



Long-term complications include catheter exit site or tract infection, catheter-associated sepsis, cardiac arrhythmias, catheter colonization, catheter thrombus (∼30%), fibrin sheath, extravasation, occluded catheter, and shearing of catheter.



Factors that increase the risk of catheter-associated infection include prolonged duration of indwelling time, multiple-lumen catheters, femoral or internal jugular vein locations, non-catheter-related bacteremia (neutropenic patient), number of times the system is accessed, difficult catheter placement, and poor technique in catheter or port-site care.


During the course of disease, many patients with cancer require intravenous chemotherapy, frequent blood sampling, transfusion of blood products, or total parenteral nutrition. Many of these intravenously administered therapies—especially chemotherapies—are inflammatory to small peripheral veins. Over the past three decades, techniques for obtaining and maintaining central vascular access have been developed and refined. Managing the treatment of patients with cancer requires a thorough familiarity with the special use of vascular access devices. This chapter provides a review of the pros and cons of various devices, insertion methods, and catheter maintenance techniques.

Although it would be ideal to base vascular access decisions on solid clinical information, few randomized controlled trials have examined the clinical controversies that are involved with chronic vascular access for the patient with cancer. In addition, many of the larger studies regarding central vascular access were performed in the inpatient intensive care unit (ICU) setting. Comparing ICU patients with patients with cancer who receive chemotherapy or weekly blood draws on an outpatient basis could lead to inaccurate conclusions. It is important, however, to review the available randomized trials, a number of carefully performed retrospective analyses, and pertinent ICU literature to address some of the questions concerning vascular access for the patient with cancer.

When managing a patient who requires vascular access for treatment of a malignancy, there are three important questions: (1) What device will best meet this patient's therapeutic needs? (2) How can we most safely insert and maintain central venous access? (3) What are the immediate and delayed complications of vascular access procedures that are unique to the oncology patient population?


Clinical oncologists use the vascular access devices that are discussed in this chapter to aspirate blood and centrally infuse agents into central veins, such as the superior vena cava (SVC) or the inferior vena cava. Many chemotherapeutics and parenteral nutrition formulations act as vesicants to the venous intima, causing inflammation and thrombosis of smaller veins. By infusing this type of product into the higher-flow, less thrombogenic cardinal veins, the durability and safety of vascular access can be extended, and patient comfort can be enhanced significantly.

Percutaneous Central Lines

Traditional central lines are placed by using the Seldinger technique, which is described in detail later in this chapter. The subclavian vein, internal jugular vein, or femoral vein is cannulated percutaneously, and the catheter is placed by using guide-wire assistance. There is a very short distance between the skin and the catheter's entry point into the vein. Theoretically, this proximity to skin flora increases the risk of subsequent central line infection.

These central lines are in common use throughout most hospitals for oncology patients, for critically ill patients who require central access, or for any patients who require infusions that are poorly tolerated via peripheral intravenous access. Such central lines provide excellent short-term access to the central venous system. In general, percutaneous central lines are considered only for short-term use, and after the first 7 to 10 days, percutaneous central lines have a markedly higher incidence of infection despite optimal skin entrance site dressing techniques. Obviously, prolonged patient neutropenia and episodes of bacteremia could result in shorter life spans of such catheters.

Patients could benefit from these traditional central venous catheters if the patients require a relatively short course of infusion or need a bridge to placement of a more long-term catheter. Meticulous sterile dressing changes are an absolute necessity for outpatients who need short-term therapy via these lines. Patients without the resources or dexterity to care for percutaneous central lines are at a prohibitively increased risk of line infection, bacteremia, or thrombotic event and should be provided with a long-term form of central access. Certain comorbid conditions (e.g., burns, open wounds near the line site, or tracheostomy) preclude placement of this type of vascular access.

Surgically Tunneled Central Lines

Most oncology patients need a long-term form of central venous access rather than a traditional central line. Surgically tunneled central lines were developed to increase the distance between the skin entrance site and the puncture of the vein. The hypothesis was that by increasing this distance, the life span of the central access would be increased by decreasing the incidence of infection and thrombosis. Tunneled central lines are commonly referred to by the name of the first brand marketed, Hickman. Other examples include Broviac, Quinton, and Groshong ( Fig. 52-1 ). These polymeric silicone rubber venous access catheters are placed via a subcutaneous tunnel that is described in detail later in this chapter. Clinical studies have supported the hypothesis that increasing the distance between the catheter exit site in the skin and the hole in the vein decreases the incidence of externally derived infection. [1] [2] Studies suggest that the incidence of bloodstream infections associated with tunneled catheters is approximately 1 to 2 per 1000 catheter days.[3] The frequency of bacteremia among nontunneled catheters has been reported at between 1.0 and 13.0 per 1000 catheter days.[4]


Figure 52-1  Examples of standard double- and triple-lumen Hickman catheters. The double-lumen catheter has both the standard Dacron cuff and an additional antibacterial barrier cuff.



A Dacron cuff 3 to 4 cm from the exit site encourages scar formation to fix the catheter in place. This exaggerated scarring eliminates the need for long-term sutures holding the catheter in place to the skin. Avoiding these fixation sutures can decrease the incidence of stitch reactivity and associated localized skin infections. One drawback of this type of central access device is the inconvenience of placing and removing the lines. In most instances, surgeons insert these lines in the operating room under local anesthetic and intravenous sedation. Therefore, these procedures require coordination of the patient, the surgeon, the anesthesia staff, and the operating room staff. They are also more uncomfortable to remove than are nontunneled lines owing to the Dacron cuff scar reaction. Removal requires intravenous sedation and local anesthesia but can be accomplished outside of the operating room.

Surgically Implanted Infusion Ports

Implantable ports consist of a small injection reservoir with a self-sealing membrane; they are placed entirely beneath the skin. There is no external catheter. An internal catheter runs from the reservoir into the subclavian or internal jugular vein to provide central access. The internal catheter is essentially identical to those used for tunneled central lines. A noncoring (Huber) needle can pass directly through the skin into the reservoir for infusion or aspiration.[5] The self-sealing rubber cap on the reservoir prevents leakage from the reservoir after withdrawal of the Huber needle. The gauge of the noncoring needle—not the catheter—typically limits flow through the port system ( Fig. 52-2 ). The ports can also be used for continuous drug infusion, as shown in Figure 52-3 .


Figure 52-2  A, Examples of implantable infusion devices: Low-profile titanium ports (top center), peripheral access port used in the arm (bottom center), single and dual polysulfone-titanium ports (left), single- and double-lumen standard profile ports (right). B, Peripheral access catheter for placement of a small port in the forearm. This system utilizes a fluoro-free thermosensor (gray wire) initially inside the white silicon catheter. An external wand over the chest determines the site of catheter top placement.




Figure 52-3  Placement of a deflected point (Huber) needle in a port. The lower portion of the port has been cut away to show the needle tip in the reservoir. This example of a bent needle shows how the needle can be taped and secured at the skin surface for protracted infusion using an external pump.



The theoretical concern that bacteria more easily traverse the short distance between the skin and the “neo-vein” or reservoir and cause increased rates of infection does not prove true. [6] [7] [8] [9] [10] [11]Sterile technique and site care lead to an infection risk comparable to that of tunneled catheters. With adequate care, the rate of infection could be fourfold to fivefold less than that for tunneled catheters.[3]The low rates of infection and extravasation make infusion ports ideally suited for patients with cancer who need long-term single-lumen access and a low-maintenance catheter. Experience has shown, however, that patients with prolonged periods of neutropenia or significant risk of cutaneous eruptions might not be good candidates for port devices. The visible lump of the port could bother extremely thin patients; however, the port is hidden from plain view in most people.

Long-Line Central Access

Also known as a peripherally inserted central catheter (PICC line), this vascular access device is becoming increasingly popular. The catheter is inserted into a brachial, cephalic, or antecubital vein and advanced into the subclavian vein or higher. Successful placement rates of 75% might be improved by the addition of ultrasound guidance. [12] [13] [14] These lines can be placed simply and easily in the outpatient office setting and are well tolerated by patients, with minimal risk. Many oncology patients have poor arm veins after multiple peripheral infusions and are therefore poor candidates for this technique. The 50% risk of catheter-related thrombosis is the greatest drawback to more widespread use of PICC lines in oncology patients. This problem is secondary to the presence of a long length of catheter within the vein and to the catheter tip in the relatively low flow subclavian vein. Advancing the catheter tip into the SVC can reduce the incidence of thrombosis by more than half.[15] Accurate placement of the catheter tip in the SVC can be complicated by the large displacement of the catheter tip (up to 8 cm) with normal arm range of motion.[16]


Providing patients who are undergoing treatment for cancer with appropriate vascular access requires not only a thorough knowledge of available devices but also technical expertise in the procedures for placing these catheters. Although these procedures are sometimes viewed as routine, the risks of poor technique can be devastating for the patient. Before undertaking the procedure, one should carefully consider everything from choosing the site of insertion to selecting the dressing to be used at the end of the procedure ( Table 52-1 ).

Table 52-1   -- General Guidelines for Catheter and Port Use







Central line

Placed at bedside
Easy removal
One or multiple lumens
Can change over wire

Hospital use only

Sterile transparent dressing
Change dressing twice weekly

5 mL heparinized saline (10 units/mL) in each lumen daily or after each use

Antibiotic ointment may promote resistance

PICC catheter

Safe to insert and remove

1 week to 6 months of IV access

Transparent antimicrobial foam dressing
Change every 6 days

3 mL heparinized saline bid or after each use

Flush and draw blood slowly to avoid catheter migration

Hickman catheter

Multiple sizes and lumens
Tunneled under skin

Continuous infusion therapy
Long-term access

Newly placed: antimicrobial dressing for 7 to 10 days
Gauze or transparent dressing changed 2 times per week

5 mL heparinized saline daily or after each use

Clean with chlorhexidine during dressing change
After 4 weeks may clean with antimicrobial soap

Groshong catheter

Slit valve requires no heparin flushes
Smaller, more flexible catheter

Single- or double-lumen catheter for long-term use





Completely under skin
Minimal care

One lumen access and low maintenance


5 mL heparinized saline daily or after each use
20 mL saline after blood draw
Monthly flush with 5 mL heparinized saline (100 units/mL)

Clean with chlorhexidine before use, access with noncoring needle
Use no needle >20 gauge to access
EMLA cream helpful



Choosing Insertion Location

Vascular access devices can be placed by using a number of different access points, including the internal jugular, subclavian, external jugular, and femoral veins. The vast majority of catheters that are used for long-term vascular access in patients with cancer are placed in the internal jugular and subclavian veins. We will limit this discussion to these most common access sites.

The most frequent site for insertion of vascular access devices in the oncology population is the subclavian vein. Clavicular fracture or previous median sternotomy can alter the anatomy of this location. For a patient with such a history, an alternative site should be considered. Even for patients with standard venous anatomy, the acute angle at the confluence of the subclavian and internal jugular veins at the brachiocephalic vein can complicate the passing of the guide wire into the SVC. There is a higher incidence of pneumothorax, hemothorax, and catheter malposition among inexperienced operators, and there is a higher incidence of vein stenosis with subclavian vein placement than with internal jugular placement. [17] [18] [19] A subclavian artery puncture during line placement can be difficult to control, owing to the position of the artery posterior to the clavicle. A catheter placed on the anterior chest wall is more comfortable and easier to cover with clothing, however, than is a line in the neck. Additionally, some authors have concluded that with lines that are in place for more than 4 days, the risk of line infection is less for subclavian catheters than for internal jugular catheters. [17] [18]

In many ways, the internal jugular vein is the ideal location with regard to ease of placement of a central catheter. The path of the guide wire during placement is straight, thereby limiting many guide-wire complications and catheter malposition. In the event of carotid artery puncture, arterial bleeding can be controlled safely with the application of direct pressure. This advantage is especially important in the thrombocytopenic patient. The incidence of central venous occlusion is decreased, which could be important for patients who are likely to need an arteriovenous fistula for hemodialysis in the future. Unfortunately, patients often complain of pain with neck and shoulder movements after placement of an internal jugular catheter.

Preparing to Place the Vascular Access Device

Adequate preparation for placing a central venous catheter for oncology patients includes several steps before the actual line insertion. The surgeon must make several decisions that can limit the incidence of both immediate and delayed complications and ensure optimal line function.

Studies suggest that providing a single dose of prophylactic antibiotic to cover common skin flora before inserting the vascular access device reduces central line infection. It is difficult, however, to determine how this small benefit affects antibiotic resistance and subsequent infections. Using central venous catheters impregnated with antibiotics could be more effective in dealing with infectious complications and will be discussed later in this chapter. [20] [21]

A sterile surgical field with mask, gown, cap, gloves, and a large sterile drape should be used to minimize the risk of line infection.[22] The skin of the entire anterior neck and chest should be prepared with chlorhexidine, which is superior to povidone-iodine or alcohol in limiting line infections. [23] [24] [25] [26]

The choice to use ultrasound guidance to identify the vein during cannulation should be addressed before beginning the procedure. Less experienced operators will likely benefit from the use of ultrasound guidance as an adjunct to the anatomic landmarks technique. [13] [27] Ultrasound can decrease the incidence of arterial puncture and placement failure. Although a few reports in the literature exist regarding the advantage of these techniques, such superiority is often judged when compared with high rates of complications using the landmark approach as the control group.

The operator must also decide where to position the catheter tip within the central vein. Clearly, catheters that are positioned with the tip in the right atrium will function longer as a source for aspirating blood samples than will those with the tip positioned in the SVC. [28] [29] A case review of thrombosed catheters documents that the position of the tip of the catheter at the time of thrombosis seems to be the most important contributing factor. [30] [31] [32] [33] The closer the catheter tip is to the right atrium, the lower is the frequency of thrombosis and infection. The risk of a catheter tip in the right atrial position is primarily that of cardiac arrhythmias when the tip is near the tricuspid valve. An additional risk of right atrial catheter placement is right atrial thrombus or right atrial erosion.[32] These risks have led the Food and Drug Administration to publicly warn operators to avoid placement of the catheter tip within the right atrium. Instead, the catheter tip should sit at the junction of the SVC and the right atrium.

Insertion Technique

After informed consent is obtained, a rolled towel is placed directly under the vertebral column at the shoulders to extend the clavicles. A peripheral line is established, and the patient is connected to an EKG monitor and a pulse oximeter. Intravenous sedation is typically established by using small doses of benzodiazepines. The fluoroscopy operating table and patient are placed in the Trendelenburg position. The skin of the neck and entire upper anterior thorax is prepared, and sterile drapes are positioned. The skin and deep tissues are anesthetized with 1% lidocaine using a 25-gauge needle for the skin followed by a 22-gauge needle for the anticipated insertion tract ( Fig. 52-4A ).


Figure 52-4  A, Placement of a catheter using the Seldinger technique. The long insertion needle is shown entering the vein through skin and subcutaneous tissues. B, Demonstration of a catheter passing over the guide wire.



For subclavian vein puncture, the site of skin puncture is usually 1 cm below the angle of the lateral third of the clavicle. The long insertion needle, as depicted in Figure 52-4A , is slowly inserted below the clavicle, aiming for a point approximately one fingerbreadth above the sternal notch. During insertion, a small amount of negative pressure is maintained on the syringe. With experience, the physician develops a feel for the actual puncture of the vein, and when this occurs, the syringe fills easily with venous blood.

As the position of the needle is being carefully maintained, the syringe is removed, and the guide wire is inserted ( Fig. 52-4B ). If the patient experiences discomfort in the neck, the guide wire is partially withdrawn. Turning the patient's head away from the site of insertion and exerting a downward pull on the ipsilateral arm could facilitate entrance of the catheter or guide wire into the SVC. Cardiac ectopy indicates that the guide wire has entered the right atrium, and the wire should be withdrawn slightly.

Fluoroscopy may be used to pass the wire when difficulty is experienced. The correct position of the catheter is confirmed by fluoroscopy or chest x-ray before the catheter is secured with a 3–0-nylon suture at the skin exit site. The tip of the catheter is positioned 1 cm above the SVC-atrial junction when the patient is in the Trendelenburg position. A chest x-ray is obtained to document the absence of pneumothorax and accurate placement of the catheter within the SVC.

When it is necessary to use the veins in the neck, the right internal jugular provides more direct access to the SVC and right atrium. In this case, the patient's head is turned to the opposite side. The insertion site is located just lateral to the carotid artery and approximately two fingerbreadths above the head of the clavicle. Another useful landmark for the insertion site is the angle formed by the sternal and clavicular heads of the sternocleidomastoid muscle ( Fig. 52-5 ).


Figure 52-5  Demonstration of catheter insertion technique when utilizing a vein in the neck (right internal jugular).



Placement of a permanent Silastic catheter, such as a Hickman, is depicted in Figure 52-6 . Placement of the introducing needle and guide wire is as described in previous figures except that a 1-cm incision is made before insertion of the introduction needle. A prophylactic antibiotic is given before the procedure. It is often beneficial to provide the patient with intravenous sedation.


Figure 52-6  A, Placement of a permanent Silastic catheter (such as a Hickman). The first and second incision sites are shown, as is the path of the tunneler. B, The catheter is pulled through the tract by using a heavy suture. The catheter cuff is secured 3 to 4 cm from the skin exit site. C, The catheter is trimmed and positioned 1 cm below SVC-RA junction. The introducer and tear-away sheath are shown passing over the guide wire. The catheter is inserted into the sheath, and the sheath and introducer are removed.



A second 1-cm incision is placed lower on the anterior chest, usually at the level of the fourth or fifth interspace. The skin and subcutaneous tissues are first anesthetized with 1% lidocaine. The projected course of the tunneled catheter is also infiltrated with lidocaine. A tunneler is passed from the lower incision to the upper incision, where the guide wire is exiting. A heavy suture is tied to the end of the tunneler and brought down through the tract. The suture is tied to the end of the catheter and used to pull the catheter up through the tract, securing the cuff 3 to 4 cm from the skin exit site. The catheter is trimmed to the correct length that will position it about 1 cm above the SVC-RA junction with the patient in Trendelenburg position. This location is approximately four fingerbreadths below the sternal notch.

Care is taken to cut the catheter squarely and smoothly. An introducer and a tear-away sheath are passed over the guide wire into the vein ( Fig. 52-7 ). A slight rotary motion facilitates the introduction of the sheath and avoids crimping. To avoid developing a false passage, the guide wire should be withdrawn occasionally as the introducer is inserted. The guide wire is then removed, and a syringe is attached to the introducer to confirm that blood can be aspirated. The introducer is removed, and the thumb is used to control bleeding or air intake through the sheath. It is important to have the catheter tip poised to insert as the introducer is removed. The passage of the catheter could meet some resistance as it is passed between the clavicle and first rib.


Figure 52-7  Example of an introducer kit. The kit includes a guide wire and introducer (dilator) with tear-away sheath (two different sizes shown). The upper, longer blunt object is a tunneler, which is included in some kits.



Fluoroscopy or a chest x-ray is obtained to confirm correct positioning and the absence of pneumothorax. The catheter is aspirated and flushed to confirm function. Care should be taken to avoid any angulations of the catheter through the subcutaneous tract, especially at the bend toward the subclavian vein. The incision below the clavicle is closed with a subcutaneous absorbable suture. The catheter is secured at the exit site with a 3-0 nylon suture, which is maintained for approximately 3 weeks to allow the fibrous tissue ingrowth into the subcutaneous cuff. The exit site is covered with a sterile gauze dressing.

Similar techniques apply to the use of neck veins. The catheter is tunneled over the clavicle, the exit site being the same as for subclavian vein placement. When other sites are required (e.g., the femoral vein with the catheter tip in the inferior vena cava), direct cut-down exposure of the saphenous vein or other large femoral branch is preferred.

Figure 52-8 illustrates the placement of an implanted injection port. A 1-cm incision is placed below the clavicle at the site planned for subclavian venipuncture. A second, 3-cm incision is placed lower on the chest in a position that provides a relatively flat surface and stability for the port chamber. Local anesthesia is 1% lidocaine with 1 : 200,000 epinephrine. A subcutaneous pocket just large enough to accommodate the port is dissected inferior to the incision. Ideally, the level of this pocket is over the underlying pectoral fascia. The skin coverage needs to be thick, but the port must be percutaneously accessible. A tunneler is passed subcutaneously from the pocket through the infraclavicular incision. A suture is tied to the tunneler and brought down through the tract. The suture is tied to the end of the port catheter and used to pull the catheter through the tract to the intraclavicular incision. Care should be taken to position with a gentle curve and to avoid catheter angulation. The port is secured in the pocket with three sutures of 0-prolene. It is necessary to anchor the port adequately to prevent flipping or rotation of the device.


Figure 52-8  A, Placement of an implanted injection port. First and second inclusion sites are shown, as is dissected pocket to accommodate port. B, Demonstration of placement of a port into a subcutaneous pocket.



The guide wire is inserted as described, and the dilator and sheath are passed over it. The catheter is cut to the desired length. The dilator and guide wire are removed, and the catheter is inserted through the sheath as described. The incision for the port pocket is closed with interrupted 3-0 absorbable sutures placed in the subcutaneous layer. The skin is approximated with a running subarticular 4-0 absorbable suture. A transparent nonpermeable dressing is used at the port incision.

Open insertion methods are quite safe in a skilled surgeon's hands. The major complication of the open technique is the possibility of air embolus during the actual catheter insertion. This is most likely to occur in patients who are hypovolemic, cachectic, or unable to tolerate positioning in Trendelenburg position. An air embolus happens most frequently when the internal jugular vein is used, and extreme care with the use of the purse-string suture around the insertion site and venous occlusion with vascular clamps should limit the possibility of introduction of air into the vascular system. With open direct surgical placement, although it takes considerably longer than the closed Seldinger technique, complications should be much lower than 5%. [33] [34] [35] In the Hopkins series, the open method results in a complication rate of less than 1%. Although this is a safer technique, it requires more training, more experienced operative personnel, and larger incisions. Any patient who has had repeated problems with closed insertions should be approached in an open fashion, however, as this is the most controlled and safest format for that patient. Sometimes, previous operations or radiation therapy in the region of the cardinal veins makes the open operative approach more difficult, but generally such problems are limited.


When complication rates are examined as a whole, certain patient factors, catheter factors, and operator factors seem to predict the occurrence of complications. Patient-related factors that predict higher complication rates include the presence of multiple comorbid conditions, atherosclerosis, abnormal anatomy, thrombocytopenia, immunocompromise, prior radiation therapy to insertion area, recent myocardial infarction, and patient restlessness.[36] In addition, multiple-lumen or stiffer catheters carry increased risks of complications. [37] [38] [39] [40] Factors related to the person performing the insertion of the catheter also influence risk of complications. Risk increases if the operator has inserted fewer than 50 central venous catheters, if more than two tries at cannulating the vein are required, or if the insertion of the catheter is difficult. [16] [41] [42] Risk factors for specific complications are discussed in the sections that follow.

Immediate Complications


Published complication rates for pneumothorax after jugular vein central line placement are approximately 0.5% and up to four times higher for subclavian vein procedures. [16] [37] [38] [43] The risk of this technical complication can be reduced dramatically among experienced operators or physicians who have experienced supervisors.[36]


Many patients with cancer are at increased risk of bleeding complications owing to thrombocytopenia, uremia, other platelet dysfunction, or anticoagulant therapy. Bleeding complications are associated most frequently with thrombocytopenia.[39] For patients with platelet counts less than 50 or International Normalized Ratio greater than 2.0, we consider administration of platelets or fresh-frozen plasma. Experienced physicians should perform these procedures with access to ultrasound guidance if necessary.

Cardiac Arrhythmias

Disruptions in the normal cardiac conduction pathway can be caused by contact between the catheter and the right atrium. Most of these arrhythmias are short-lived and self-limiting.[16] These problems are more common with insertion of pulmonary artery catheters than with catheters that are typically inserted for oncologic vascular access. They are associated with more significant sequelae in patients who have recently suffered a myocardial infarction or who have a history of left bundle branch block. [40] [44]

Delayed Complications

Infectious Complications

Infectious complications are the most common complications of long-term vascular access devices in the oncology patient population. Two factors make the interpretation of this relatively well-studied topic challenging. First, many of the large randomized controlled trials that have studied central line infections have concentrated on ICU patients. The ICU population has different risk factors and susceptibilities from those of the outpatient population of people with cancer. Nevertheless, by carefully reviewing the available data, we can make some conclusions regarding the pathogenesis, prevention, and treatment of line-related infections among oncology patients. In studies of ICU patients with central lines, factors that predispose to line infection have included malignancy, neutropenia, extended duration of indwelling time, and coincident parenteral nutrition. All of these factors can contribute to the incidence of infection among patients with vascular access for oncologic treatment. [45] [46]

Second, the confusing terminology regarding line-related infections can make the literature on this topic difficult to interpret. Local infection is a positive culture at the catheter insertion site. Catheter colonization or infection is the positive culture of a segment of removed catheter. Catheter-associated bacteremia is evidenced by a positive blood culture from a site other than the catheter and the positive culture of a segment of removed catheter with the same pathogen.

Understanding the pathophysiology of catheter-associated bacteremia can help to limit the incidence of this complication. Up to 50% of catheter-associated bacteremia is caused by coagulase-negative staphylococcus. This common skin flora can colonize the catheter during insertion or later. A thrombus that forms at or along the catheter tip can become a nidus for bacterial proliferation, with resultant bacteremia. Thrombosis significantly increases the risks of colonization and infection.[47] Bacteria can also be introduced into the bloodstream by hub contamination, by hematogenous seeding from another focus of infection, and, rarely, by the infusate itself.

Factors that increase the risk of catheter infection include prolonged indwelling time, multiple-lumen catheters, femoral or internal jugular vein location, difficult catheter placement, and non-catheter-related bacteremia. [45] [48] As was mentioned previously, nontunneled catheters are at increased risk of catheter infection compared with tunneled catheters, and totally implantable devices are even less susceptible than tunneled catheters. [49] [50]

Multiple-lumen catheters have demonstrated higher infection rates than single-lumen catheters. The addition of each lumen exponentially increases incidence of infection. [51] [52] [53] [54] There are two possible explanations for this observation. First, the increased internal lumen diameter of the line is associated with a higher thrombosis rate and accumulation of loose thrombus at the catheter tip. Thrombosis causes more breaking of the line and more line manipulation when attempting to declot, leading to subsequent infection. Second, the multiple ports invite multiple interruptions of the line for access and result in a greater likelihood of introduction of bacteria. Despite their greater risk of iatrogenic infection, multiple-lumen catheters have great appeal for patients who need multiple simultaneous infusions of incompatible drugs. Very strict nursing guidelines must be followed in the management of multiple-lumen catheters to prevent iatrogenic infections and thrombosis. [55] [56]

Diagnosis and management of catheter-related infection differs between nontunneled and tunneled catheters. However, some diagnostic principles apply to both tunneled and nontunneled catheters. Routine surveillance blood cultures should not be performed. When a catheter-related infection is suspected due to fever, chills, or purulence around the catheter site, percutaneous and catheter blood samples should be submitted for culture. [57] [58] [59] [60] Qualitative culture with continuously monitored differential time to positivity compares the time to positivity for catheter blood cultures to percutaneous peripheral-blood cultures. This technique has demonstrated excellent specificity and sensitivity for detecting catheter-related infection in tunneled catheters. [61] [62]

In most patients with nontunneled central venous catheters, the line should be removed if the patient demonstrates signs of site infection or sepsis or if blood culture results from the catheter and percutaneous blood samples are positive. [61] [62] Seven to 10 days of narrow-spectrum antibiotic therapy is generally recommended. In certain cases of clinically stable patients with a single episode of coagulase-negative staphylococcus, a trial of antibiotic therapy might salvage the catheter.[53] For any patient with persistent bacteremia despite antibiotic therapy and removal of the infected catheter, a thorough investigation of possible septic sources, such as endocarditis or septic thrombus, is warranted.

Tunneled catheters or infusion ports should be removed in cases of sepsis, complicated infections, tunnel tract infections, or port abscesses.[63] A thorough evaluation confirming the surgically implanted catheter as the source of infection should precede the removal of any of these vascular access devices. In the absence of complicated infection, catheter salvage could be indicated. Antibiotic lock therapy is a reasonable approach for attempting to salvage lines with common coagulase-negative staphylococcus, Staphylococcus aureus, or gram-negative bacilli intraluminal infections. This therapy consists of instilling the catheter lumen with high concentrations of antibiotics and leaving them there for several days. [64] [65] [66] [67] If salvage therapy fails, a new tunneled catheter can be placed after removing the infected catheter, treating with an appropriate course of antimicrobial therapy, and repeating blood cultures with negative results.

Catheter Thrombus

Thrombus of central vein access devices predisposes to both line malfunction and line infection. The most common site of thrombus formation with prolonged indwelling central catheters is where the catheter enters the vein. At this point, a fibrin sleeve progresses distally toward the tip of the catheter. The precipitating event is likely local trauma from line insertion with subsequent endothelial damage and disruption of intraluminal laminar flow. The venous intima exposed to blood flow activates the coagulation cascade. Understanding this pathophysiology helps to explain why difficult line placements often lead to shorter catheter survival and increased rates of infection.

The incidence of central venous catheter-related thrombus as demonstrated by ultrasound is greater than 30% for catheters that are in place longer than 7 days.[68] This problem typically presents as progressive difficulty in flushing the catheter. A change in posture or the Valsalva maneuver could allow aspiration of blood. Occasionally, thrombosis can present as extremity edema. This can be especially devastating after axillary nodal resection or axillary irradiation. Management of a nonfunctioning catheter is discussed later.

Acute line obstruction can also be caused by precipitation of incompatible medications. Common offenders include total parenteral nutrition, etoposide salts, lipid emulsions, calcium salts, antibiotics, and sodium bicarbonate. Infusion of a solution specifically matched to the precipitated material might flush the line.[69]


Extravasation is defined here as the leaking of infusate into the subcutaneous tissue surrounding a central venous catheter. This complication can be caused by needle displacement from an implanted port, a defect in the catheter tubing, or withdrawal of the catheter from the vein due to inadequate fixation. Additionally, “backtracking” can occur when the catheter is partially occluded and infusate tracks up along the fibrin sleeve and into the subcutaneous tissue. Findings that are suggestive of extravasation include sudden swelling at the line site, increased patient discomfort during infusion, and sudden loss of blood return.[70]

Clavicular–First Rib Compression

This infrequently discussed complication may occur up to 1% of the time in long-term indwelling catheters.[71] When the subclavian vein is cannulated more medially than usual in the narrow space between the clavicle and first rib, the line can be compressed between the first rib and the clavicle. Patients who report difficulty infusing while in the sitting position or when the ipsilateral arm is elevated or abducted should be suspected of this compression. This malposition can be observed on chest x-ray as kinking of the line over the first rib. The catheter can break free and embolize if the line is not promptly removed. An interventional radiologist using a percutaneous retrograde femoral catheterization approach can retrieve an embolized section of catheter.


When a central venous catheter does not return blood or infuse solution, the malfunction should be assessed systematically to maximize catheter durability and to minimize the incidence of catheter-related complications. The first step in this assessment is reviewing the most recent chest x-ray to confirm appropriate placement. The next step is checking for gravity flow. Connecting a bag of normal saline to gravity flow and asking the patient to change position, cough, and breathe deeply will demonstrate whether flow is positional. If the catheter is patent to gravity flow, lower the bag below the catheter and assess for blood return. With this information, one can make a decision as to whether to remove, reposition, or declot the catheter (Table 52-2 and 52-3 [2] [3]).

Table 52-2   -- Management of a Suspected Clot



Gravity Flow Observation

Corrective Action

Fibrin sheath[*]

Fibrin collects on the tip or encases the catheter

Excellent in all positions but no blood return in any position

May be declotted following the declotting procedure

Occluded catheter[*]

Fibrin and platelets collect inside catheter

Inability to infuse solutions and draw blood

May be declotted following the declotting procedure; drug precipitate cannot be corrected with tPA or heparin


These are the only two instances in which it is safe to declot after checking the chest x-ray for proper placement.


Table 52-3   -- Management of an Unknown Clot



Gravity Flow Observation

Corrective Action

Do not attempt to declot any of the following conditions.


Catheter cut short and abutting vessel wall

Positional; no blood return in any position

Must be removed; do not attempt declot, which may erode vessel wall

Pinch-off syndrome

Insertion site too close to first rib and clavicle

Excellent when lying down but absent when standing

Must be removed to avoid catheter breakage

Transverse catheter

Catheter crosses into opposite subclavian

Positional flow, slow blood return, position noted on chest x-ray

Interventional radiology can reposition via femoral vein access


Catheter in right atrium

Positional; position noted on chest x-ray

Catheter must be pulled back into SVC or removed

Flipped up

Catheter tip in jugular vein; can occur in association with vomiting or coughing

Positional, blood return may be positional, patient complaint of tinnitus or headache with catheter flushing, position noted on chest X-ray

Interventional radiology may be able to reposition if catheter not cut too short

Major vessel thrombus

External thrombus within vessel due to insertion trauma

No change; good blood return; arm, hand, or neck swelling

Documentation of thrombus with ultrasound or venogram, treatment with thrombolytics after consultation with hematology



Two conditions are appropriate for thrombolysis of the catheter with alteplase or tissue plasminogen activator (TPA): fibrin sheaths and intraluminal thrombus occlusions. Contraindications for declotting procedures include high bleeding risk and known recent or current episode of bleeding.

Catheter malfunction caused by fibrin sheaths usually demonstrates good gravity flow and no blood return because the sheath acts as a one-way valve, allowing outflow but not inflow through the catheter. When intraluminal fibrin and platelets occlude the catheter, the catheter will demonstrate neither gravity flow nor return, regardless of any positional changes. If the caregiver is convinced that the catheter malfunction fits one of these categories and appropriate catheter position is confirmed on chest x-ray, catheter thrombolysis is indicated. Contrast venography is not necessary before using TPA to open nonmechanical catheter occlusions.

Techniques for declotting catheters vary from institution to institution but follow the same general principles. We recommend instilling 500 μg of TPA and allowing the infusate to dwell for 1 hour. If patency is not restored after 1 hour, instill an additional milligram of TPA and allow the infusate to dwell for another hour. Using a very similar technique, the authors of the Cardiovascular Thrombolytic to Open Occluded Lines Trial (COOL-2) studied nearly 1000 patients with occluded central venous catheters. These patients in a predominantly oncology population did not undergo prethrombolysis contrast studies. No deaths or major bleeding episodes were directly attributable to the thrombolysis. More than 87% of the catheters were opened with the TPA. At 3-month follow-up, nearly 75% of the catheters were still patent.[72] A similar study in the oncologic population reported a success rate greater than 80%.[73]


The risk of complications with a long-term vascular access device does not end with insertion of the device. Proper care and maintenance of each type of central catheter can dramatically reduce the incidence of catheter-related bacteremia, catheter thrombosis, and line failure.[74] Maintenance is performed not only by qualified staff of oncology centers but also by patients and family members in the outpatient setting. In Table 52-1 and Box 52-1 , we have detailed our recommendations for central catheter care after a thorough review of the best available literature.

Box 52-1 


Troubleshooting Guidelines

If a catheter does not have a blood return or will not infuse solution, perform the following troubleshooting techniques to diagnose the problem before proceeding with the declotting procedure:



Connect a flush bag of NS or D5W to the catheter and open the roller clamp to allow the fluid to flow to gravity.



Have the patient change position, take deep breaths, and cough while the drip rate is observed.



When the fluid infuses at the fastest rate, lower the bag and observe for a blood return.



If a blood return is observed, use the catheter as indicated, making a note that it is positional.



If no blood return is observed in any position, if gravity flows freely in all positions, and if the last chest x-ray verifies that the catheter tip is in the SVC, proceed to declot the catheter with a likely fibrin sheath obstruction.



If the x-ray shows that the catheter tip position is questionable, obtain a dye study to verify the exact location of the catheter tip, integrity of the catheter, and flow of solution. Management algorithm for patients with isolated pulmonary metastasis from extrathoracic malignancy. CXR, chest radiography; FNA, fine-needle aspiration; Hx, history; VATS, video-assisted thoracoscopy.

Declotting Procedure

See the preceding guidelines before proceeding to declot, as follows:



Using a 1-mL syringe, create a vacuum by aspirating back on the plunger (at the hub) and clamp. Repeat once.



Instill 2 mL 1 : 1000 U heparin solution (in each lumen) and let dwell for 1 hour. Assess for a blood return.



If the line remains clotted, instill the appropriate amount of alteplase to fill the catheter (listed following) using the same technique as with the heparin.



Let dwell for 30 minutes to 1 hour and then assess for a blood return.

Catheter Type

Alteplase (mL)

Central line, 14 gauge


Central line, 16 gauge


Hickman, 12 F


Hickman, 10 F










Patients and physicians have a large variety of vascular access devices from which to choose, and no individual catheter is the ideal solution for all clinical situations. By understanding the strengths and weaknesses of each catheter system, insertion site, and maintenance technique, we hope to minimize the morbidity associated with these commonly used devices. Clearly, having a great deal of experience in placing vascular access devices in the oncology population minimizes the number of problems. Line care is the joint responsibility of the health care team and can be a major source of patient morbidity if each and every member of the oncology team does not maintain the highest quality of care.

More than 750,000 vascular access devices are used in the United States each year. We must continue to conduct clinical studies to make the safest, most cost-effective choice for each cancer patient.


  1. Mayhall CG: Diagnosis and management of infections of implantable devices used for prolonged venous access.  Curr Clin Top Infect Dis1992; 12:83-110.
  2. Dryden MS, Samson A, Ludlam HA, et al: Infective complications associated with the use of the Quinton ‘Permcath’ for long-term central vascular access in haemodialysis.  J Hosp Infect1991; 19:257-262.
  3. Pegues D, Axelrod P, McClarren C, et al: Comparison of infections in Hickman and implanted port catheters in adult solid tumor patients.  J Surg Oncol1992; 49:156-162.
  4. Elliott TS: Can antimicrobial central venous catheters prevent associated infection?.  Br J Haematol1999; 107:235-241.
  5. Muller H, Zierski J: The Huber needle as a special cannula for the puncture of implanted ports and pumps: a mistake in multiple variations [in German].  Klin Wochenschr1988; 66:963-969.
  6. Niederhuber JE, Ensminger W, Gyves JW, et al: Totally implanted venous and arterial access system to replace external catheters in cancer treatment.  Surgery1982; 92:706-712.
  7. Brothers TE, Von Moll LK, Niederhuber JE, et al: Experience with subcutaneous infusion ports in three hundred patients.  Surg Gynecol Obstet1988; 166:295-301.
  8. Brincker H, Saeter G: Fifty-five patient years' experience with a totally implanted system for intravenous chemotherapy.  Cancer1986; 57:1124-1129.
  9. Khoury MD, Lloyd LR, Burrows J, et al: A totally implanted venous access system for the delivery of chemotherapy.  Cancer1985; 56:1231-1234.
  10. McGovern B, Solenberger R, Reed K: A totally implantable venous access system for long-term chemotherapy in children.  J Pediatr Surg1985; 20:725-727.
  11. Bothe Jr A, Piccione W, Ambrosino JJ, et al: Implantable central venous access system.  Am J Surg1984; 147:565-569.
  12. Robinson MK, Morgensen KM, Grudinskas GF, et al: Improved care and reduced costs for patients requiring peripherally inserted central catheters: the role of bedside ultrasound and a dedicated team.  JPEN J Parenter Enteral Nutr2005; 29:374-379.
  13. Schummer W, Schummer C, Tuppatsch H, et al: Ultrasound-guided central venous cannulation: is there a difference between Doppler and B-mode ultrasound?.  J Clin Anesth2006; 18:167-172.
  14. Parkinson R, Ghandhi M, Harper J, et al: Establishing an ultrasound guided peripherally inserted central catheter (PICC) insertion service.  Clin Radiol1998; 53:33-36.
  15. Kearns PJ, Coleman S, Wehner JH: Complications of long arm-catheters: a randomized trial of central vs peripheral tip location.  JPEN J Parenter Enteral Nutr1996; 20:20-24.
  16. Rosen M, Latto IP, Ng WS: Handbook of Percutaneous Central Venous Catheterization,  London, Saunders, 1992.
  17. Collignon P, Soni N, Pearson I, et al: Sepsis associated with central vein catheters in critically ill patients.  Intensive Care Med1988; 14:227-231.
  18. Pearson ML: Guideline for prevention of intravascular device-related infections: I. Intravascular device-related infections: an overview. The Hospital Infection Control Practices Advisory Committee.  Am J Infect Control1996; 24:262-277.
  19. Ruesch S, Walder B, Tramer MR: Complications of central venous catheters: internal jugular versus subclavian access—a systematic review [comment].  Crit Care Med2002; 30:454-460.
  20. Darouiche RO, Raad II, Heard SO, et al: A comparison of two antimicrobial-impregnated central venous catheters. Catheter Study Group [comment].  N Engl J Med1999; 340:1-8.
  21. Raad I, Darouiche R, Dupuis J, et al: Central venous catheters coated with minocycline and rifampin for the prevention of catheter-related colonization and bloodstream infections: a randomized, double-blind trial. The Texas Medical Center Catheter Study Group [comment].  Ann Intern Med1997; 127:267-274.
  22. Raad II, Hohn DC, Gilbreath BJ, et al: Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. [comment].  Infect Control Hosp Epidemiol1994; 15(Pt 1):231-238.
  23. Vassilomanolakis M, Plataniotis G, Koumakis G, et al: Central venous catheter-related infections after bone marrow transplantation in patients with malignancies: a prospective study with short-course vancomycin prophylaxis [comment].  Bone Marrow Transplant1995; 15:77-80.
  24. Spafford PS, Sinkin RA, Cox C, et al: Prevention of central venous catheter-related coagulase-negative staphylococcal sepsis in neonates.  J Pediatr1994; 125:259-263.
  25. Shaul DB, Scheer B, Rokhsar S, et al: Risk factors for early infection of central venous catheters in pediatric patients.  J Am Coll Surg1998; 186:654-658.
  26. Duggan J, O'Connell D, Heller R, Ghosh H: Causes of hospital-acquired septicaemia: A case control study.  Q J Med1993; 86:479-483.
  27. Keenan SP: Use of ultrasound to place central lines.  J Crit Care2002; 17:126-137.
  28. Shivnan JC, McGuire D, Freedman S, et al: A comparison of transparent adherent and dry sterile gauze dressings for long-term central catheters in patients undergoing bone marrow transplant.  Oncol Nurs Forum1991; 18:1349-1356.
  29. Bosserman G, McGuire DB, McGuire WP, Nicholls D: Multidisciplinary management of vascular access devices.  Oncol Nurs Forum1990; 17:879-886.
  30. Holcombe B: Restoring patency of long-term central venous access devices.  J Intravenous Nurs1993; 16:195.
  31. Greenberg S, Kosinski R, Daniels J: Treatment of superior vena cava thrombosis with recombinant tissue type plasminogen activator.  Chest1991; 99:1298-1301.
  32. Gilon D, Schechter D, Rein AJ, et al: Right atrial thrombi are related to indwelling central venous catheter position: insights into time course and possible mechanism of formation.  Am Heart J1998; 135:457-462.
  33. Gauderer MW: Vascular access techniques and devices in the pediatric patient.  Surg Clin North Am1992; 72:1267-1284.
  34. Lameris JS, Post PV, Zonderland HM, et al: Percutaneous placement of Hickman catheters: comparison of sonographically guided and blind techniques.  AJR Am J Roentgenol1990; 155:1097-1099.
  35. Stacey RG, Filshie J, Skewes D: Percutaneous insertion of Hickman-type catheters [comment].  Br J Hosp Med1991; 46:396-398.
  36. Polderman KH, Girbes AJ: Central venous catheter use. 1: Mechanical complications.  Intensive Care Med2002; 28:1-17.
  37. Hagley MT, Martin B, Gast P, Traeger SM: Infectious and mechanical complications of central venous catheters placed by percutaneous venipuncture and over guidewires.  Crit Care Med1992; 20:1426-1430.
  38. Lucey B, Varghese JC, Haslam P, Lee MJ: Routine chest radiographs after central line insertion: mandatory postprocedural evaluation or unnecessary waste of resources?.  Cardiovasc Intervent Radiol1999; 22:381-384.
  39. Doerfler ME, Kaufman B, Goldenberg AS: Central venous catheter placement in patients with disorders of hemostasis.  Chest1996; 110:185-188.
  40. Thomson IR, Dalton BC, Lappas DG, Lowenstein E: Right bundle-branch block and complete heart block caused by the Swan-Ganz catheter.  Anesthesiology1979; 51:359-362.
  41. Maki DG, Parillo JE, Bone RC: Nosocomial infections in the intensive care unit.   In: Parillo JE, Dillinger RP, ed. Critical Care Medicine: Principles of Diagnosis and Management,  St. Louis: Mosby; 1995:893-954.
  42. Koksoy C, Kuzu A, Erden I, Akkaya A: The risk factors in central venous catheter-related thrombosis.  Aust N Z J Surg1995; 65:796-798.
  43. Kincaid EH, Davis PW, Chang MC, et al: “Blind” placement of long-term central venous access devices: report of 589 consecutive procedures.  Am Surg1999; 65:520-523.
  44. Morris D, Mulvihill D, Lew WY: Risk of developing complete heart block during bedside pulmonary artery catheterization in patients with left bundle-branch block.  Arch Intern Med1987; 147:2005-2010.
  45. Howell PB, Walters PE, Donowitz GR, Farr BM: Risk factors for infection of adult patients with cancer who have tunnelled central venous catheters.  Cancer1995; 75:1367-1375.
  46. Fuchs PC, Gustafson ME, King JT, Goodall PT: Assessment of catheter-associated infection risk with the Hickman right atrial catheter.  Infect Control1984; 5:226-230.
  47. Barzaghi A, Dell'Orto M, Rovelli A, et al: Central venous catheter clots: Incidence, clinical significance and catheter care in patients with hematologic malignancies.  Pediatr Hematol Oncol1995; 12:243-250.
  48. Band JD: Pathogenesis of and risk factors for central venous catheter-related infections. [2003/04/01; cited 2003/05/28].  Available from: http://www.uptodate.com
  49. Groeger JS, Lucas AB, Thaler HT, et al: Infectious morbidity associated with long-term use of venous access devices in patients with cancer [comment].  Ann Intern Med1993; 119:1168-1174.
  50. Carde P, Cosset-Delaigue MF, Laplanche A, Chareau I: Classical external indwelling central venous catheter versus totally implanted venous access systems for chemotherapy administration: a randomized trial in 100 patients with solid tumors.  Eur J Cancer Clin Oncol1989; 25:939-944.
  51. Eastridge BJ, Lefor AT: Complications of indwelling venous access devices in cancer patients.  J Clin Oncol1995; 13:233-238.
  52. Franson TR, Zak O, van den Broek P: Evaluation of new anti-infective drugs for the treatment of vascular access device-associated bacteremia and fungemia. The European Working Party of the European Society of Clinical Microbiology and Infectious Diseases.  Clin Infect Dis1993; 17:789-793.
  53. La Quaglia MP, Lucas A, Thaler HT, et al: A prospective analysis of vascular access device-related infections in children.  J Pediatr Surg1992; 27:840-842.
  54. Benezra D, Kiehn TE, Gold HW, et al: Prospective study of infections in indwelling central venous catheters using quantitative blood cultures.  Am J Med1988; 85:495-498.
  55. Smith A: The effect of a nursing staff education pro-gram on compliance with central line care policy in the cardiac intensive care unit.  Pediatr Nurs2006; 32:95.
  56. Render ML, Brungs S, Kotagal U, et al: Evidence-based practice to reduce central line infections.  Jt Comm J Qual Patient Saf2006; 32:253-260.
  57. Schmitt SK, Knapp C, Hall GS, et al: Impact of chlorhexidine-silver sulfadiazine-impregnated central venous catheters on in vitro quantitation of catheter-associated bacteria [comment].  J Clin Microbiol1996; 34:508-511.
  58. Irwig L, Tosteson AN, Gatsonis C, et al: Guidelines for meta-analyses evaluating diagnostic tests [comment].  Ann Intern Med1994; 120:667-676.
  59. Siegman-Igra Y, Anglim AM, Shapiro DE, et al: Diagnosis of vascular catheter-related bloodstream infection: a meta-analysis.  J Clin Microbiol1997; 35:928-936.
  60. DesJardin JA, Falagas ME, Ruthazer R, et al: Clinical utility of blood cultures drawn from indwelling central venous catheters in hospitalized patients with cancer [comment].  Ann Intern Med1999; 131:641-647.
  61. Blot F, Nitenberg G, Chachaty E, et al: Diagnosis of catheter-related bacteraemia: a prospective comparison of the time to positivity of hub-blood versus peripheral-blood cultures.  Lancet1999; 54:1071-1077.
  62. Blot F, Schmidt E, Nitenberg G, et al: Earlier positivity of central-venous-versus peripheral-blood cultures is highly predictive of catheter-related sepsis.  J Clin Microbiol1998; 36:105-109.
  63. Peacock SJ, Eddleston M, Emptage A, et al: Positive intravenous line tip cultures as predictors of bacteraemia.  J Hosp Infect1998; 40:35-38.
  64. Messing B, Peitra-Cohen S, Debure A, et al: Antibiotic-lock technique: a new approach to optimal therapy for catheter-related sepsis in home-parenteral nutrition patients.  JPEN J Parenter Enteral Nutr1988; 12:185-189.
  65. Douard MC, Arlet G, Leverger G, et al: Quantitative blood cultures for diagnosis and management of catheter-related sepsis in pediatric hematology and oncology patients.  Intensive Care Med1991; 17:30-35.
  66. Krzywda EA, Andris DA, Edmiston Jr CE, Quebbeman EJ: Treatment of Hickman catheter sepsis using antibiotic lock technique.  Infect Control Hosp Epidemiol1995; 16:596-598.
  67. Benoit JL, Carandang G, Sitrin M, Arnow PM: Intraluminal antibiotic treatment of central venous catheter infections in patients receiving parenteral nutrition at home [comment].  Clin Infect Dis1995; 21:1286-1288.
  68. Randolph AG, Cook DJ, Gonzales CA, Andrew M: Benefit of heparin in central venous and pulmonary artery catheters: a meta-analysis of randomized controlled trials [comment].  Chest1998; 113:165-171.
  69. Ulz L, Petersen FB, Ford R, et al: A prospective study of complications in Hickman right-atrial catheters in marrow transplant patients.  JPEN J Parenter Enteral Nutr1990; 14:27-30.
  70. Wickham R, Purl S, Welker D: Long-term central venous catheters: issues for care.  Semin Oncol Nurs1992; 8:133-147.
  71. Andris DA, Krzywda EA, Schulte W, et al: Pinch-off syndrome: a rare etiology for central venous catheter occlusion.  JPEN J Parenter Enteral Nutr1994; 18:531-533.
  72. Deitcher SR, Fesen MR, Kiproff PM, et al: Safety and efficacy of alteplase for restoring function in occluded central venous catheters: results of the cardiovascular thrombolytic to open occluded lines trial.  J Clin Oncol2002; 20:317-324.
  73. Timoney JP, Malkin MG, Leone DM, et al: Safe and cost effective use of alteplase for the clearance of occluded central venous access devices.  J Clin Oncol2002; 20:1918-1922.
  74. Polderman KH, Girbes AR: Central venous catheter use. 2: Infectious complications.  Intensive Care Med2002; 28:18-28.