Denny-Venous Access

Venous
Access 2009

Coordinator:
Donald F Denny MD

Faculty:

Donald
F Denny MD

Joseph
Bonn MD

Brad
Glenn MD

Sohail
Contractor MD

Primary
contributors to workshop handout: Don Denny MD, Thomas Vesely MD,
David J. Eschelman MD, Jeffrey S. Pollak MD, Mel Rosenblatt, MD.
Edited by Howard Chrisman MD, Thomas Vesely MD and Don Denny MD

____________________________________________________________________________________________

Pearls:

1. “Physician
   extenders” such as nurse practitioners, physician assistants,
   and radiology assistants are assuming more responsibility for the
   insertion and management of central venous catheters and ports.

   
   

2. There
   are polyurethane PICCs specially approved for use with power
   injectors at flow rates up to 5 ml/s.

   
   

3. Chest
   and arm ports approved for use with contrast power injectors are
   also now available.

   
   

4. Skin
   glue, 2-octyl cyanoacrylate can be used to assist in skin closure,
   speeding closure and providing a watertight dressing.
   

   
   

5. A
   method for inserting a tunneled jugular catheter with a single
   incision and puncture from an infraclavicular approach has been
   reported.

   
   

6. Chlorhexidine
   gluconate has replaced betadine as the preferred cleansing agent for
   intact skin (CDC guidelines).

   
   

7. Location,
   location, location. The right internal jugular vein is the preferred
   site for the majority of central venous catheters (K/DOQI).
   

   
   

8. New
   developments in catheter coatings, including antibiotic-impregnated
   and antithrombotic coatings may reduce the incidence of
   catheter-related infection in longer term catheters, such as
   tunneled hemodialysis catheters. This is work in progress.
   

   
   

Suggested
use of this handout:

This
handout has grown by accretion with occasional deletional editing.
The easiest way to use this is to import into a word processor and
search for the term or topic in which you are interested. If you only
have the printed handout, email ddenny@prapa.com for an electronic
version.

Catheter
Choice: Clinical Perspective

Introduction:

In
patients requiring circulatory access for therapy the first step in
their treatment is choosing the appropriate access device. This
seemingly simple task is deceptively complex. Ideally, an access
device should provide reliable complication free circulatory access.
Since this ideal device does not exist it is left up to the clinician
to choose a device that will match the patient’s particular
clinical needs. To do this, one must be familiar with the variety of
access devices available and understand their clinical design
limitations. In this presentation, we review the available access
devices, the biomaterials used in their construction and highlight
the clinical issues related to device selection.

“The
Ideal Access Device”: Device Limitations

The
ideal access device would provide reliable complication free
circulatory access. Unfortunately, no such device exists.
Complications related to central venous access are frequent. These
complications include periprocedural complications, thrombotic
complications, infectious complications, and device malfunction ^1,
2. In designing devices to reduce complications and enhance
function, there is often a trade-off that must be made. As an
example, central venous thrombosis which is the result of the
long-term presence of a catheter in a vascular system can be reduced
by utilizing catheters of small diameter ^{3, 4}.
Unfortunately, small diameter catheters having a single lumen, cannot
sustain high flow rates and often fail prematurely due to catheter
occlusion. This same limitation applies to peripherally implanted
devices. These devices were designed to avoid complications such as
pneumothorax, inadvertent arterial puncture, vascular injury, and air
embolus, which are all associated with access into a central vein.
Peripheral veins, however, are small and only small diameter
catheters, with their inherent limitations, can be accommodated.
Another example involves infectious complications, which are, by far,
the most common complications associated with central venous access.
These complications are the result of inadvertent bacterial
introduction and subsequent bacterial adherence to the inert catheter
material which is present in the vascular system. The ideal access
device would be impervious to infection. In the absence of such a
device, manufactures have modified device designs to help reduce the
incidence of infectious complications⁵. Unfortunately,
these modifications have their disadvantages. Subcutaneously
implanted devices are an example of such a modification. These
devices have no external components and thus have a reduced incidence
of infection⁶. However, they are far more expensive than
devices that have external components, are more difficult to place,
and can not sustain high blood flow rates. Device infection, when it
occurs, may be masked until infection becomes severe. Many other such
trade-offs exist. Each device has its advantages and disadvantages. A
basic understanding of device construction and thorough knowledge of
the various available access devices is the critical first step in
selecting an appropriate access device.

Materials
Used in Access Device Construction :

Currently,
all chronic access catheters are constructed from one of two
different materials, polyurethane or silicone. Each of these
materials has different characteristics which have clinical impact.
Polyurethane is a block copolymer that contains high molecular weight
macroglycols linked together by a urethane group. This material is
produced by the rearrangement polymerization of diisocyanate and
macroglycols ⁷. Polyurethane has a high tensile strength
permitting catheters to be constructed with relatively thin walls.
This permits the catheter to have a high internal diameter to outer
diameter ratio. A variety of different biomedical polyurethanes,
with varying properties, are produced. Unfortunately, many of these
polyurethanes are unavailable for use in the production of chronic
venous access catheters. Two types, Chronoflex and Tecoflex, are the
most commonly encountered chronic catheter polyurethanes.

One
major disadvantages of all polyurethanes is Environmental Stress
Cracking (ESC). As its name suggests, ESC leads to mircocracks in
the material of the device as a result of the corrosive forces within
the environment of the living body. Once microcracks begin device
failure is inevitable. Different formulations of polyurethane are
somewhat resistant to this enzymatic degradation process ⁸ .
Chronoflex which is a carboxylated aliphatic nonether based
polyurethane is resistant to this process and therefore very
biostable ⁹ . Tecoflex, on the other hand, is an aliphatic
ether based polyurethane which is known to develop microcracks when
implanted for long periods.

Silicone
rubber is produced from polydimethylsiloxane (PDMS). PDMS is blended
with vinylmethylsiloxane and silica to produce medical grade
silicone. This material is extremely biostable, compliant and kink
resistant but mechanically weak. This mechanical weakness can be
overcome by increasing the silicone products’ thickness. Thus,
catheters composed of this material will have thicker walls and a
smaller internal to external diameter ratio.

For
subcutaneously implanted devices a reservoir is required. These
reservoirs have been constructed from stainless steel, titanium, and
plastic. Stainless steel is a non-reactive biostable metal which has
many biomedical applications. This material is strong, durable, and
inexpensive. Its disadvantages are that it is heavy, radiopaque, and
ferromagnetic. Its weight can result in device descent and catheter
pullback. Its ferromagnetic properties can result in MRI artifacts,
which can prevent the use of this diagnostic modality. Its increased
radiopacity can also prevent visualization of lesions that are
adjacent to the device. Titanium is a non ferromagnetic, biostable,
light metal with great strength. This material has been used in the
construction of many subcutaneously implanted devices. Its metallic
properties are far superior to that of stainless steel, however, it
is expensive and heavy as compared to non-metallic man-made
materials. Plastic has been used in the construction of many
biomedical devices. It is inexpensive, lightweight, non-ferromagnetic
and non-radiopaque. Unfortunately, plastics are not as hard as
metallic materials. This is potentially important in devices that are
being accessed with metal needles. The needles can gouge the plastic
and potentially cause premature device failure. Another problem
associated with the use of plastics, is material availability. Many
manufacturers are unwilling to accept the liability associated with
the biomedical use of their product. They have therefore restricted
sale of man-made materials and will only sell it to non-biomedical
manufacturers. Composite construction uses both metal and plastic in
a single device. This combines the positive features of each of the
two materials. Many devices are currently being constructed in this
fashion.

Devices:

Various
types of central venous access devices are available to satisfy the
particular clinical needs of the patient. These devices can be
categorized into one of two large groups: peripherally inserted
central venous access devices or centrally inserted central venous
access devices. Peripherally inserted venous access devices include
PICC lines as well as peripherally inserted subcutaneous reservoirs.
The PICC line can be, and is often, placed at the bedside by
specially trained personnel ¹⁰. In patients with a
paucity of obvious peripheral veins, image guidance is often required
for placement of these lines ¹¹. Peripherally implanted
subcutaneous reservoirs are small single lumen devices which are
implanted within the subcutaneous tissues of the forearm or arm and
connected to a small diameter peripherally inserted venous catheter
¹². The tip of this catheter is positioned in the SVC.
These devices are suitable for long-term therapy and often can remain
in place for several years.

Centrally
inserted central venous catheters can be subdivided into two groups:
high flow catheters and infusion catheters. High flow catheters are
designed to support blood flow rates as high as 400 ml per minute.
These high flow rates are required for specific uses such as dialysis
or plasmapheresis. For short-term use, non-tunneled, rigid catheters
are often utilized. In the chronic setting, large diameter,
tunneled, soft catheters constructed of polyurethane or silicone are
preferred ^{13, 14}.

Centrally
placed infusion catheters can be used in both the acute and chronic
setting. These devices are utilized for the administration of
various pharmacologic agents and blood sampling. Non-tunneled
catheters (e.g. triple lumen catheters), which are inserted at the
bedside, are suitable for short-term use. These devices have a high
infection rate and require daily maintenance, which limits their use
to the hospitalized patient.

Tunneled
catheters traverse a subcutaneous tunnel prior to entering the
central vein. They also possess a Dacron cuff, which surrounds the
catheter and is positioned within the subcutaneous tunnel. This
Dacron cuff permits tissue ingrowth and reduces the risk of infection
and dislodgement. These catheters are available in a variety of sizes
and can be single, double, or triple lumen. These catheters have a
variable life span. They often can remain in place, complication
free, for several months or even years ¹⁵. The major
disadvantage of this type of device is that it has an externalized
portion, which limits patient’s activities and requires daily
maintenance. For these reasons, tunneled catheters are often utilized
for patients requiring circulatory access from six weeks to three
months.

In
the patient who will require intermittent circulatory access over the
course of several months to years, centrally inserted subcutaneous
venous access devices (SVADs, or "ports") are ideal.
Similar to the peripherally implanted access devices, these devices
cannot be dislodged, do not limit patient activity, and have a
reduced infection rate as compared to externalized catheters ⁶.
However, centrally inserted SVADs have several advantages over their
peripheral cousins. These devices have larger access septums, can be
attached to larger diameter catheters, and are available in single
and double lumen varieties. The larger septum surface area eases
device access and reduces the chance of inadvertent administration of
pharmacologic agents within the subcutaneous tissues. The larger
diameter catheters permit more reliable blood draws and the
availability of double lumen devices permits administration of
complex multi- agent chemotherapeutic regimens.

A
traditional limitation of PICCs and ports is flow rate. With the
emergence of techniques using high injection rates for CT, such as CT
angiography, there is a need for higher flow devices in patients with
low flow venous access devices. There are now a number of PICCs which
are designed to accommodate power injectors for CT. Typically, these
can accommodate flows up to 5 ml/sec. At the time of this writing
(1/2008) there is one manufacturer with an approved chest port for
power injectors (Bard). The number of manufacturers offering these
devices is likely to expand in the near future.

An
earnest effort by device designers to improve function, ease
implantation, and reduce complications has lead to the development of
various unique access device designs. These unique designs, which
are found in each of the different access device categories, are
purported by their manufacturer to provide certain advantages over a
competing product. Often these modifications do provide a real
advantage where as on occasion, the advantage is only theoretical.
It is therefore left up to the implanting health care professional to
use a product with characteristics that match the patient’s
clinical needs.

Device
Selection:

In
selecting the appropriate access device several questions must be
asked. What is the access device needed for? How long will it be
needed? How many lumens are needed? How often will it be accessed?
Are there any lifestyle considerations that might be important? Would
a particular device design more appropriately match the patient’s
clinical needs? Answering these questions is key to choosing the
appropriate device. Following is an algorithm that highlights some of
these points.

Conclusions:

Rapidly
advancing technology has lead to the production of many new and
unique access devices. Each unique device may offer a particular
feature that may best suit a patient’s specific clinical need.
There are many other unique circumstances which may push for a
specific device. For example, a patient on antiplatelet therapy such
as Plavix, in whom holding the drug is contraindicated, may be better
treated with a non-tunneled catheter rather than a port, even if the
port would otherwise seem the better choice. A patient receiving
continuous infusion of a chemotherapeutic agent (e.g. 5 fluorouracil)
may find a chest device more convenient than an arm one. Through
familiarity with the wide variety of access devices combined with a
clear understanding of the patient’s access requirements is
essential in choosing the most appropriate access device.

PERIPHERAL
VENOUS ACCESS:

PICCs
AND ARM PORTS

PICCs
are peripherally inserted central catheters. PICCs were initially
developed for the neonatal and pediatric populations to deliver
hypertonic or sclerosing medications centrally, but their use has
expanded to adults, particularly with the increased use of outpatient
intravenous therapies. PICCs have replaced tunneled chest wall lines
for intravenous access for antibiotics, chemotherapy,
hyperalimentation, or administration of blood products. We are
experiencing an increasing number of requests for PICC placement in
hospitalized patients with poor venous access, in many cases
replacing temporary central line placements in these patients.
Indeed, the use of PICCs is one of the highest growth areas in
vascular access. An important recent development is the introduction
of high flow PICCs, ones capable of injecting contrast at up to 5
ml/s. the increasing use of CT angiography makes these devices
essential in the care of the hospitalized patient.

PICCs
are best suited for those patients requiring frequent or continuous
administrations of medications, rather than those with intermittent
needs (less frequently than every day or every other day) who would
be better suited for port placement. In most patients, continuous or
daily access of a port defeats the purpose of this device and
increases the infection rate.

Why
choose a PICC over a Hickman catheter? PICC insertion is quicker,
easier, and has fewer associated risks. The catheter is cheaper.
The average time of the procedural portion, excluding prepping and
suturing, is less than 10 minutes in 95% of patients. In many
patients only local anesthesia is required, though we offer conscious
sedation to those patients who are nervous or if access to the vein
is not achieved quickly. There is no risk of pneumothorax or
significant air embolism, and the puncture site is compressible. We
prefer that patients have platelet counts greater than 50,000 and no
coagulopathy, but exceptions are the norm. While it has not been
proven, we believe that the risk of central venous stenosis or
thrombosis is lower with these small caliber peripheral devices as
opposed to the larger tunneled catheters. Also, some papers have
suggested that a central venous stenosis is caused by the site of
venous puncture, so catheter placement through an arm vein would be
less likely to effect the central veins. The infection rate appears
to be lower for PICCs than tunneled catheters in our experience.
This is supported by research done by Maki, a prolific infectious
disease specialist at University of Wisconsin⁴⁰. He noted
that insertion site bacterial colonization rates were almost 100
times greater with central venous catheters than with peripheral
sites, adding that colonization of the insertion site is the single
most powerful predictor of increased risk for catheter related
infection. Rates of local catheter infections were 6.5 times greater
with central catheters than peripheral accesses, and bacteremia rates
were markedly increased with these central devices. Finally, many
patients fell "less violated" by having what they view as a
long, small caliber IV line through their arm rather than a tunneled
catheter on their chest.

Several
years ago, Joe Bonn, MD, of Thomas Jefferson University Hospital,
performed a clinical comparison of 106 surgically placed, tunneled
central venous catheters with 139 PICCs. The number of patients in
the study was limited by a precipitous decrease in the number of
tunneled chest wall catheters being placed at Jefferson. Surgically
tunneled catheters were placed, on average, three days after catheter
placement was requested whereas most PICCs were placed on the same
day or next day after the request. No insertion complications were
experienced with patients receiving PICCs, while four
surgically-tunneled catheters were malpositioned, two could not be
inserted, and one patient had significant bleeding complications.
Complications such as dislodgment, fracture, and migration were
somewhat more common with PICCs, but these problems often required
readmission for patients with tunneled catheters unlike those with
PICCs who were expeditiously treated as outpatients. Catheter
occlusion was seen in 9 PICCs as compared with one of the larger
surgically tunneled catheters. However, thrombophlebitis was seen in
six of the tunneled catheters and none of the PICCs in this series.

The
infection rate was 1.87/1000 catheter days for patients with tunneled
catheters as compared with 0.68/1000 catheter days for PICCs. In
summary, PICCs were inserted more promptly, had fewer insertion
complications and requirements for follow-up studies, and had lower
risks of thrombophlebitis and infection. PICCs were more likely to
occlude or dislodge, but were easier to manage without readmission.

PICCs
were designed for bedside placement through an antecubital or forearm
vein without fluoroscopic guidance. Initially, interventional
radiologists were involved with PICCs for correction of malpositions
during blind insertions, to negotiate across stenoses and occlusions,
and to gain access in those patients with depleted forearm veins
which is becoming more common in hospitalized patients with frequent
access needs. While mechanical fatigue is less concerning with the
newer PICC products, we believe that upper arm placement is
preferable over the forearm or antecubital space for both cosmetic
reasons and because the PICC is less obtrusive to daily activities.
Also, the device is less prone to kinking or migration when not
placed across the elbow joint.

Placement
of PICCs by the interventional radiology service offers the ability
to place the catheters in patients with non-palpable forearm veins
and to ensure appropriate catheter tip positioning. Experience with
catheter and guidewire techniques certainly facilitates placements,
particularly with some of the products which are available because
they are better suited for placement using the Seldinger technique.
Furthermore, when compared to surgical placement of tunneled chest
wall catheters, PICC placement in radiology offers the advantage of
being cheaper and much more accessible since the associated costs of
the operating room, recovery room, and anesthesia are avoided.
Likewise, since we control our own room schedules and want to
accommodate the referring physicians, we provide same day or at worst
next day service for venous access. We have tried to avoid doing
venous access procedures on the weekends. However, we do receive
occasional requests for assistance which includes inpatients with
poor access, patients going home over the weekend who, for some
reason, we weren't contacted about on Friday, and for patients
undergoing emergency chemotherapy who are usually leukemia patients.
John Cardella reported on the experience in Hershey comparing PICC
placement by physicians on the patient floor with placement in
interventional radiology. Both groups had about 150 patients. There
was a 69% success rate of PICC placement into the superior vena cava
from blind insertion. In 74% of cases, venous access was achieved
but the catheter tips were positioned improperly in this 5% group.
Interventional radiologists (including radiology residents) were able
to successfully place 99% of the PICCs. In two cases early in their
experience, they were unable to thread the PICC centrally, but a 4 Fr
angiographic catheter was correctly positioned.

Several
series of blind PICC insertions by nurses have noted success rates in
the 60-80% range. Because of the high success of PICC placement by
John Cardella and his group, they began placing all of the PICCs in
that hospital for almost two years. To save costs, the hospital
hired four nurses for bedside PICC placements. In a subsequent study
from Hershey, they noted that bedside placement by four specially
trained PICC nurses were technically successful in 327 (82.6%) cases.
Of the 69 patients in whom nurses could not place the PICCs, 45%
were due to an inability to cannulate a vein, 38% were an inability
to thread the catheter, and 17% were malpositioning of the catheter;
these patients were referred to interventional radiology. PICC
placement was not attempted by the nurses in another 63 patients, and
these patients were also referred to interventional radiology. The
hospital determined that in any inpatient in whom more than nine days
of intravenous access was anticipated, bedside PICC placement by a
nurse would be more cost-effective than multiple placement of
peripheral intravenous lines. More recently, there have been a number
of reports by burses using ultrasound guidance, with high success and
low complication rates.

Techniques
for PICC and arm port placement have been well described in the
literature³⁸. There have been only a few modifications in
technique since 1993.

PICCs
are available from a large number of manufacturers. Kits designed for
use in IR usually include some sort of micropuncture access kit,
which includes a 21 gauge needle, platinum tipped 0.018 inch
guidewire (steel or nitinol shaft), and a coaxial sheath/dilator
tapered to the wire which allows direct placement of the catheter
through the sheath without additional device exchanges. These access
kits are also available independently (e.g. Cook Denny kit).

PICCs
may be placed using blind access, ultrasound, or fluoroscopic
guidance for access and placement. The advantages and disadvantages
of each of these methods have been discussed elsewhere. In a recent
comparison of ultrasound versus venographic insertion, Chrisman et
al, concluded that the decision to use either method can be based on
clinical grounds and/or physician preference³⁹. There is
an occasional patient who exceeds the weight limit of our
angiographic tables. In these patients we place a PICC in our
holding area using ultrasound guidance to access a vein. The PICC is
inserted blindly by approximating the distance to the cavoatrial
junction using the paper ruler that comes with the kit, and then
checking the catheter tip position using a portable chest radiograph.

We
adhere to good sterile technique for PICC placement including a
surgical scrub and wearing caps, masks, gowns, and goggles. We do
not administer intravenous antibiotics prior to catheter placement A
tourniquet is placed around the upper arm.

It
is worth mentioning the choice of skin prep solution. Conventional
skin prep in the US has been betadine solution for many years.
Recently, FDA approved a chlorhexidine solution for use in the USA:
Chlora-Prep One-Step is a combination of 2% chlorhexidine gluconate
and 70% isopropyl alcohol delivered with a single-use applicator. CDC
guidelines recommend chlorhexidine solution preferentially.
Chlora-prep is best used on intact skin. For open wounds or
excoriated skin, the alcohol in the prep may be quite irritating to
the patient.

We
place PICCs in the non-dominant arm, in the event that
thrombophlebitis may develop. However, it may be preferable in a
disoriented patient to place the PICC in the dominant arm in attempts
to reduce the patient's ability to pull out the catheter. The site
should be sufficiently above the elbow so that the PICC hub will stay
above the elbow joint. Choose a site in the grove adjacent to the
biceps muscles, and palpate the location of the brachial artery. We
strongly favor routine real-time ultrasound guidance. The cephalic
vein is an excellent choice when it is of sufficient size for access.

Limit
local anesthesia to skin and subcutaneous tissue, as deeper
infiltration around the vein may cause venospasm. A superficial skin
knick is made, and the soft tissues spread.

After
gaining access by achieving free return of blood, gently advance the
0.018 inch wire, which often requires twiddling. The platinum tip of
this wire is much more gentle than the standard Cope mandril wire
which is prone to induce spasm. Apply gentle skin retraction and
back tension while passing the peel-away sheath. If difficulty is
encountered, pass the inner dilator portion first. If the peel-away
becomes damaged during the initial attempt at placement, just cut a
beveled tip on the end.

Advance
the .018 wire to the cavoatrial junction to determine the necessary
length of the catheter, and mark or bend the wire at the hub of the
peel-away. Flush the inner dilator of the peel-away sheath.

The
mark on the wire should be placed 1-1.5 cm behind the wings of the
PICC for the appropriate length. Next, advance the PICC through the
peel-away sheath with the hydrophilic obturator in place. Most of
the time the PICC will naturally be directed into the superior vena
cava.

If
not, use the obturator (with or without placing a gentle curve on the
tip) as a torquable wire, or try advancing the PICC off of the
obturator. If this doesn't work, use a .018 or .025 Glidewire.
Ideally, the catheter tip should be positioned at the cavoatrial
junction because the increased blood flow and cardiac motion
diminishes fibrin sheath formation at this level. Remove the
peel-away before taking out the hydrophilic obturator to prevent
catheter kinking at the skin.

The
PICC should be advanced such that this transition zone between the
catheter and hub is completely inserted through the skin to prevent
kinking of the thinner portion of the catheter and to allow more
secure suture placement at the skin entry site. Apply compression at
the puncture site for a few minutes if there is any bleeding. Fluoro
the length of the catheter again to ensure that the tip position is
appropriate, and that there are no kinks, especially at the skin
entry site. Suture the PICC to the skin.

Finally,
place a sterile clear or gauze dressing. If the hub of the PICC
dangles in the antecubital fossa, just tape it out of the way. The
PICC should be flushed per manufacturer recommendation or hospital
protocol. Most PICCs need heparin flush daily. Valved PICCs, e.g.
Groshong catheters, need no heparin.

Complications
during PICC placement are uncommon. We usually have one patient/yr
in whom a PICC cannot be successfully placed, and this is usually due
to venospasm. This is usually induced by needle punctures or
attempts at advancing the guidewire. We recommend using a single
wall puncture technique. Venospasm is more common with larger
needles which accompany some of the PICC kits, ranging from 14-18
gauge. These needles are satisfactory for access of antecubital
veins, but inappropriate for upper arm veins. The Cook kit has a 21
gauge needle. If placing another device, use a micropuncture kit for
initial access. Also, it's important to use a gentle guidewire and
twiddle the wire while exiting the needle. Make sure there is brisk,
free flow of blood from the needle before initial wire placement.
Injection of lidocaine too deeply and contrast extravasation from
errant punctures can also induce or exacerbate spasm. Use the
tourniquet to distend upper arm veins, and nitroglycerin may be
injected through a peripheral intravenous line in aliquots of 100 mcg
to help relieve spasm. A few minutes of patience is often most
helpful. Rarely, a patient may experience numbness, tingling, or
shooting pains down the arm during placement of the needle or
peel-away sheath. This indicates potential injury to the nerve;
select another puncture site! Current ultrasound units are able to
see the nerve in relation to the veins in may cases. This is usually
a problem only when trying to use one of the brachial veins,
paralleling the brachial artery.

Central
vein occlusion is another potential problem. Occasionally, a
Glidewire may be negotiated through the collaterals with catheter tip
placement at the cavoatrial junction. This is not always easy,
particularly with silicone catheters. Several companies make
polyurethane PICCs which offer improved trackability over a guide
wire. In cases where the central veins are occluded bilaterally, the
PICC tip may be placed peripherally if the infusate is not too
sclerosing if attempts at crossing the occlusion with a catheter and
guide wire are unsuccessful.

While
insertion complications with PICCs are minimal, a number of
complications during use have been reported in these two large series
[IV Nursing 1993; 16:92-99, and JVIR 1996; 7:5-13]. Few present any
real danger to patients and, as you can see, are mainly mechanical
problems related to the devices. The most common complication at
Hershey was premature or inadvertent catheter removal, although it is
worth noting that they did not sew their PICCs in place.
Thrombophlebitis and infection rate are other issues as with all
venous access devices.

Catheter
occlusion is a common complication, which is seen sometime during the
life of the PICC in 5-10 %. In most cases this is easily treated at
bedside with a brisk saline flush or low dose thrombolytic treatment
(e.g. tPA 1 mg). If these are unsuccessful, try clearing the
occlusion by passing a Glidewire through the PICC under fluoro. If
thrombosis of the surrounding vein develops, try treating with
heparin and symptomatic therapies. In patients with limited venous
access options, try to keep the catheter in as long as tolerated.
PICC tips can easily migrate with a brisk flush, but usually return
to the original location. Tip repositioning is rarely required and
can often be accomplished by a brisk saline flush during inspiration.
Otherwise try to reposition by just passing a Glidewire through the
catheter, or just change the catheter. Occasional inadvertent
catheter removal is unavoidable, but suture the PICC to the skin with
an added suture placed back around the wings for security in addition
to the sutures through the wings. We have not encountered infections
from the sutures. Catheter disruption or fracture usually requires a
PICC exchange, as few PICCs have repair kits (a notable exception is
the Groshong PICC). Damage is often caused by flushing with a small
syringe which generates a greater pressure than the catheter can
withstand. Don't use a 1 cc syringe, and be careful during
injections with a 3 cc syringe. As long as there are no signs of
local infection, we will exchange the PICC and maintain this same
site. We've encountered problems with fractures of the double lumen
PICCs due to incomplete reinforcement of the hub. Despite having a
reinforced segment which says "clamp here," multiple
clamping on the thinner part have resulted in PICC fracture.

If
a PICC is removed for sepsis, treat the patient through a temporary
access before placing another PICC. The need for intravenous
antibiotics is our most common indication for PICC placement. Even
though patients may have fevers, we will place PICCs as long as there
are no signs of sepsis. If the patient has previously had positive
blood cultures, we require a blood culture without growth for two
days or that the patient has been afebrile for a day while receiving
appropriate treatment before placing a PICC.

Ports:

Venous
arm ports are smaller single lumen versions of products that have
been available for placement on the chest wall. There are several
new arm ports recently available from a number of manufacturers (e.g.
Sims, Cook, Bard, Boston Scientific).

It's
useful to have a few surgical instruments such as forceps and
mosquitoes since this procedure is more involved. Port placement
requires strict adherence to sterile technique, though we do not give
antibiotics prior to placement of these devices. After vein access,
the incision is extended, and a small subcutaneous pocket is created
using blunt dissection. This needs to be about as big as the distal
1/3 of the pinky finger. The overlying skin should be thin enough
that a finger in the pocket or diaphragm of the port can be easily
palpated, but not so thin that the port will erode through the skin.
Irrigate copiously and make sure hemostasis is obtained. Check to
make sure that the port fits in the pocket. Securely attach the port
to the infusion catheter, check for flow and leaks, place the port in
the pocket, then close the incision. While ports commonly have
suture holes provided, it has become common not to suture the port in
position unless there is a specific need because of loose tissues.
Closure is commonly in 2 layers, a deep layer of resorbable suture to
close the potential space and provide mechanical strength, then a
running subcuticular suture of 4-0 coated vicryl. See below for
additional notes on incision closure. The port should be positioned
in the pocket such that the suture line is not over the diaphragm.
Place steri-strips, and access with a 1/2 inch non-coring needle if
the port is going to be used immediately. Ports should be flushed at
least once a month with heparinized saline if not in use.

Port
removal consists of cutdown though the placement scar, freeing the
catheter and removing it from the vein, then removing the port using
sharp and blunt dissection. Local anesthesia is complicated by the
presence of the scar, which delays time of onset of effective
anesthesia; leave extra time for lidocaine to take full effect. The
scar capsule around the port usually requires sharp dissection with a
scalpel; we usually use a #15. Removing the catheter first allows the
operator to use it to apply traction to the port during the removal.
Closure is typically single layer using resorbable subcuticular
technique with Vicryl 4-0. Steri-strips are applied to the incision
margins.

Implantation
of Tunneled and

Subcutaneous
Chest Wall Access Devices

It
is estimated that approximately 1.5 million chronic venous access
catheters are implanted in the United States each year. This number
is increasing at approximately 8% per year. Unfortunately, access
related complications are quite common. Acute and late complications
can cause significant morbidity and on occasion mortality. The cost
of managing these complications is enormous. Good image guidance as
well as proper implantation technique can prevent many of these
complications and reduce their associated morbidity and health care
costs.

The
involvement of the interventional radiologist in the implantation of
access devices has grown rapidly over the pass decade. Currently, a
majority of the peripheral devices are implanted by interventional
radiologists. This however is not the case with centrally implanted
devices, which in a majority of the institutions, are implanted by
the surgical staff. For the purpose of this review, we will focus on
the implantation of centrally inserted central venous access devices,
discuss insertion techniques, and highlight the merits of
radiologically guided placement.

Technique

Preferred
Access Sites:

Several
central veins are available for insertion of access catheters. The
subclavian veins, because of their location on the chest wall and the
broad familiarity with subclavian access techniques, are the most
frequently utilized veins. Access into these veins, however, is
associated with significant long and short term complications ¹⁶.
Pneumothorax and symptomatic central venous thrombosis with
resultant arm swelling are more likely to occur with subclavian vein
access. In our view, jugular vein access is the preferred access
site. This matches K/DOQI recommendations as well. This access, in
the short term, avoids pneumothorax and in the long term reduces the
rate of symptomatic central vein thrombosis. Access of the internal
jugular veins for the implantation of chronic central venous access
devices has been avoided because of the torturous course the catheter
must take as it comes across the neck down to the chest wall. The
use of the low posterior approach to the right internal jugular vein
is an alternative to solve this particular problem. The posterior
approach involves accessing the internal jugular vein posterior to
the posterior belly of the sternocleidomastoid muscle just above the
level of the clavicle. From this location, the needle is directed
from lateral to medial across the neck parallel to the operating
table. When the vein is accessed in this fashion the catheter can
assume a gentle curve as it courses through its subcutaneous tunnel.
Routine use of real-time ultrasound guidance contributes to higher
success and lower complication rates for central venous access.

Recently,
there have been reports of a single puncture/incision technique for
accessing the internal jugular vein from an infraclavicular approach.
This technique avoids the need for separate creation of a tunnel for
long-term catheters. First reported by Glenn⁴¹, the
technique uses a curved small gauge access needle such as the #22
needle found in normal micropuncture kits. Ultrasound is necessary,
as it is used to follow the curved needle as it is advanced
subcutaneously from the infraclavicular skin access to the lateral
margin of the jugular vein at the base of the neck. Glenn⁴¹
and Contractor⁴² have reported using this technique for
placement of tunneled central catheters, ports, and dialysis
catheters, from both right and left sides.

In
patients with occluded internal jugular and subclavian veins
alternate approaches are necessary. Placement of a chronic catheter
in the common femoral vein is an option, however, this approach is
associated with a high incidence of symptomatic lower extremity DVT
and an increased infection rate. With good radiologic guidance
access into the inferior vena cava can be obtained through a
translumbar or transhepatic approach providing an alternate route for
chronic circulatory access ^{17, 18}

Guidance
Modalities:

Ultrasound
can be used to assess venous patency as well as direct the needle
into a vein without injuring surrounding structures ^{19, 20}.
High frequency transducers, 7 megahertz or higher, are preferred for
visualizing superficial vascular structures. The use of a scored
needle tip can enhance ultrasound visualization facilitating access
of the vein.

Contrast
material can also be used to guide venous access. Iodinated contrast
as well as Co₂ gas injected through a peripherally
inserted IV permits radiographic visualization of the subclavian vein
^{21, 22}. Once opacified, a needle can be directed into the
subclavian vein under direct fluoroscopic vision.

Venous
Entry

Once
visualized, the internal jugular vein (or other) is accessed with a
21 gauge needle (e.g. “microstick”). The use of a small
gauge needle reduces the trauma associated with inadvertent puncture
of surrounding structures. Once entry into the vein has been gained,
an 0.018 inch guide wire is advanced. The wire is advanced down to
the level of the right atrium to ensure that entry into the venous
system has been accomplished. Once the guidewire is in place, a 3
French/ 5 French coaxial dilator is inserted. Through this 5 French
dilator a stiffer guidewire can be inserted permitting single step
insertion of the large diameter peel away sheaths through which the
access catheters are advanced.

Tunneled
Catheters

Once
venous access is obtained, an appropriate site on the
infra-clavicular chest wall is chosen for the catheter exit site.
After adequate local anesthesia a small incision is made at that
location and a tunneling device is used to create a subcutaneous
tunnel to the venous entry site. Prior to tunneling the catheter it
is important to determine the precise length of catheter required so
that the Dacron cuff can be positioned 1 to 3 cms from the catheter
exit site and the catheter tip can be positioned at the RA/SVC
junction. It is helpful to take a guidewire and position its tip at
the RA/SVC junction and then make a bend or otherwise mark the back
end of the wire where the catheter exit site should be. This wire is
then removed and is used as a template to cut the catheter to the
appropriate length. Once cut to length, the catheter is attached to
the tunneling device and pulled through the subcutaneous tunnel.
After tunneling the catheter, attention is diverted to the guide wire
in the central vein. Over this stiff guidewire the appropriate size
peel away sheath is inserted. The dilator and guidewire are removed
and the soft access catheter is advance through the peel away sheath
and positioned in the SVC. Once the catheter tip is at the
appropriate location, the sheath is peeled back and removed. Two
important points must be made about this step. First, excessive
advancement of the peel-away sheath into the vein should be avoided
as this can often result in a kink in the sheath when the dilator and
guidewire are removed prohibiting passage of the soft access
catheter. Second, when the dilator of the peel away sheath is
removed, great care should be taken to avoid air embolization. This
can be accomplished by pinching the sheath closed, instructing the
patient not to inhale, and rapidly inserting the access catheter into
the sheath. For large bore Hemodialysis catheters, sheaths with
hemostatic valves are now available. These considerably simplify the
procedure.

Subcutaneously
Implanted Chest Wall Venous access Devices (SVADs or ports)

The
implantation of subcutaneous chest wall infusion devices is similar
to that for tunneled catheters, however, a subcutaneous pocket must
be created. A site over the anterior medial aspect of the second rib
is chosen for reservoir placement. High medial placement is
essential to avoid the descent of the device into the breast tissue
and retraction of the venous catheter. The site is infiltrated with
1% lidocaine and an incision large enough to accommodate the access
reservoir is made. Blunt dissection is used to fashion a
subcutaneous pocket just large enough to accommodate the infusion
device. Overlying subcutaneous fat thickness should be a minimum of
0.5 centimeters to prevent skin necrosis and a maximum of 1 cm to
ease the transcutaneous access of the device. The pocket is
irrigated with saline to remove tissue debris and hemostasis is
achieved. The rear of the previously placed access catheter is cut
to length and connected to the infusion reservoir. The reservoir is
then inserted into the pocket and the pocket is closed in two layers:
the subcutaneous layer with interrupted 3-0 or 4-0 resorbable sutures
(e.g. Vicryl, Monopril) and the dermal layer with a running or
interrupted 4-0 resorbable subcuticular stitch. Removal is identical
to the process for arm ports, described above.

Some
operators are using skin adhesive, 2-octyl cyanoacrylate (Dermabond,
Johnson and Johnson) to assist in skin closure. While there is little
data for this specific application, there is ample reported
experience with this agent in surgical skin closure for other
procedures. In addition to potentially speeding the time of closure,
this agent can be used without a dressing, and patients can shower in
24 hours.

Complications

The
incidence of complications is largely dependent on the type of device
implanted [14]. Complications from most common to least include
infection, catheter malfunction, venous thrombosis, pneumothorax,
wound complications, and catheter fragmentation. In a review of our
first 154 chest wall SVADs insertions, we noted certain advantages
over classical surgical implantation. These included a high success
rate, a reduced complication rate, a reduced request to implantation
time, improved complication management, and reduced overall charges.
In our series, technical success was 100% compared to the 5-8%
failure rate among surgically implanted devices. In addition to our
high success rate, our complication rate was lower then that reported
in most surgical series (Table 1). Although these devices were all
implanted outside of an operative setting, no acute local infections
were noted within the first 30 days after implantation. When a
complication did occur usually some form of intervention was
required. In our series 65% of the complications required
interventions outside of just device removal. These interventions
included placement of small chest tubes, fibrin sheath disruption,
catheter repositioning, and venous recanalization with thrombolytic
therapy and intravascular stent placement. Many, if not all of these
interventions, required extensive radiologic imaging which could not
be performed by the surgical service.

Conclusions

Proper
implantation technique can dramatically influence the incidence of
both acute and late access complications. The use of good image
guidance and thorough understanding of catheter and guidewire
techniques can improve the overall success rate, reduce the
complication rate, and permit treatment of complex access problems.
By accomplishing this, costs related to circulatory access can be
significantly reduced.

CENTRAL
VENOUS CATHETERS FOR HEMODIALYSIS

Hemodialysis
central venous catheters (CVCs) are large caliber lines designed to
permit venovenous hemodialysis. These may be temporary, which should
be used for less than 3 weeks, or long-term, tunneled hemodialysis
catheters (THCs) which are intended for longer periods. The temporary
CVCs are typically made from polyurethane and directly enter the
vein, while THCs are made from softer polyurethanes or silicone.
These have a subcutaneous course prior to entering the vein with a
polyester cuff in this region to promote fibrosis – this helps
secure the catheter and acts as a barrier to infection migrating
intravascularly. Current catheters typically have two lumens, one for
outflow to the hemodialysis machine and the other for return,
although at least one THC system achieves this by way of two separate
catheters. The large size of these catheters is due to the need for
high flow rates to meet the requirements of relatively short dialysis
sessions. Desired flows are typically at least 250 to 300 ml/minute,
and preferably faster. Catheter sizes are typically 13.5 to 15.5 F.

The
choice of device and the route of access are addressed in the K/DOQI
standards (http://www.kidney.org/professionals/KDOQI/).

Patient
preparation involves obtaining preliminary clinical and laboratory
data (primarily coagulation parameters and platelets). After
obtaining informed consent, a very limited ultrasound examination of
the proposed access vein(s) can be done to assess patency. Given the
known risk of developing post catheterization stenoses in the
subclavian veins, these should be avoided so as not to interfere with
the outflow from a potential hemodialysis arteriovenous shunt in the
arm. Additionally, right-sided veins appear to be less prone to
central vein occlusive problems than left ones. Therefore, the
optimal access is the right internal jugular vein (IJV). If this is
obliterated, the left IJV is the next choice. If the external jugular
veins are open, it may be preferable to use. The access site is then
sterilely prepped. For temporary catheters, three povidone-iodine
washes can be used. When placing a THC, a more vigorous scrub is
needed, typically two chlorhexidine scrubs followed by 3
povidone-iodine washes, with the field to include the upper chest,
where the catheter will exit. Additionally, the operator needs to
scrub his/her hands, all personnel in the room should be capped and
masked, and the room should receive frequent, thorough cleansing if
it is being used for placing tunneled lines. The procedure room
should adhere to hospital operating room standards.

We
prefer to puncture the right IJV using a 21 gauge needle and a low
posterior approach, using ultrasound guidance. This transversely
oriented venipuncture means that a temporary catheter will lie over
the shoulder rather than under the chin and permits a gentler curve
to the infraclavicular region for tunneled catheters. Large veins can
usually be easily entered using surface anatomic landmarks for
guidance, but ultrasound guidance should be available. A small
footprint 7.5 MHz probe is optimal for this. Once the vein is
entered, a 0.018 inch mandril wire is advanced under fluoroscopic
guidance into the right atrium, or even the inferior vena cava, to
confirm proper venous positioning, and the needle exchanged for a
dilator.

The
lengths given for double lumen temporary catheters are the distances
from the tip to the region where the two lumens split apart. Two
lengths are given for tunneled hemodialysis catheters: from the tip
to the polyester cuff and from the tip to the hub. Given their
stiffer nature, temporary catheters should have their tips at the
superior vena cava-right atrial junction or high right atrium. Softer
THCs should have their tips in the mid right atrium, to take
advantage of the maximal blood flow available here and to limit the
potential of developing stenoses of the superior vena cava related to
chronic catheter tip irritation. Consideration should be given to the
possibility of catheter pullback from the initial placement in a
supine patient, especially in those with large amounts of chest
tissue. It may be advisable to position the catheter in the lower
right atrium in such patients, anticipating that it will retract to
the mid right atrium. Soft catheters have not been known to cause
right atrial perforation. The proper catheter length to use in a
patient can be determined by using a bent wire. While the wire should
be bent at the venipuncture site for temporary catheters (taking into
account the length of the hub of the dilator), it should be bent at
the expected location of the cuff along the subcutaneous route for
THCs.

Placing
a temporary CVC is a simple matter of exchanging the dilator for the
catheter and suturing it in place. Placing a THC involves first
tunneling from an infraclavicular entry site to the venipunture site,
after which, the dilator is exchanged for a peel-away sheath. The THC
is then advanced through this, taking care to avoid air from entering
the system or excessive blood loss from these large caliber conduits.
Recently, large sheaths with hemostatic valves have become available;
these are specifically designed for Hemodialysis catheter placement,
and minimize the risks of serious back bleeding and air embolism. An
additional technical aid is the use of a glidewire within the
catheter, which may serve to stiffen it and allow easer advancement
into the vein. After confirming brisk aspiration of blood from each
lumen, they are primed with appropriate volumes of heparinized
saline. We use 1000 units/ml heparin for THCs. The THC is secured at
its entry/exit site with a 2-0 nylon stitch, taking care not to
significantly constrict the lumen of these soft catheters, the
venipuncture incision closed with a 4-0 absorbable stitch, and
sterile dressing is applied. The retention stitch may be removed
after the subcutaneous Dacron cuff s securely embedded in the track,
usually 3 weeks or more. A post-procedure chest X-ray is obtained to
assess for proper catheter positioning and pneumothorax.

A
variety of problems are possible with hemodialysis CVCs.
Procedure-related ones include pneumothorax, internal or
transcatheter hemorrhage, and air emboli. The incidence of these
events can be kept to nearly zero by using a low posterior approach
to the IJV, a 21 gauge needle for the puncture, ultrasound guidance
as needed for the venipuncture, fluoroscopic guidance for wire and
catheter manipulations, introducing a THC into the peel-away sheath
during a Valsalva maneuver or end-inspiration, and making sure that
the clamps on the lumens are closed.

Catheter
infection results in the failure of 11% to 28% of THCs. These may be
local infections, involving the exit site or subcutaneous tunnel, or
intravascular catheter infection, with septicemia. While most of
these occur late, and are therefore related to contamination during
dressing changes, dialysis, or hematogenous seeding from another
source, a smaller number result from contamination at the time of
placement. By convention, any infection occurring within 30 days of
catheter placement is considered a procedural complication.
Antibiotics may control infection without catheter removal. If this
is unsuccessful, one can try exchanging a catheter over a wire for
non-purulent processes. If pus is present, exchanging the catheter
with creation of a new tunnel may suffice, but persistent or
life-threatening infections should be treated with removal of the
catheter and placement of a temporary one through a new route. Some
centers believe that trying to maintain a catheter with infections
from certain organisms, such as Staphylococcal aureus, runs the risk
of potentially devastating recurrences, such as epidural abscess, and
therefore recommend removing such infected catheters immediately.

Prevention
of infection is best accomplished by proper attention to sterile
technique during catheter placement and during subsequent care.
Dressing changes should be performed at least three times per week.
The value of using a prophylactic antibiotic, such as vancomycin, at
the time of catheter placement is unclear, as is the use of topical
antibiotics or a povidone-iodine solution. We use the latter but not
the former. Similarly, the issue of using transparent, semipermeable
membranes as opposed to dry gauze dressings is unsettled. Some
catheters have an additional silver-impregnated cuff of collagen that
has antiseptic properties. Interestingly, some studies have shown
lower rates of infection with IJV catheters than subclavian vein
ones.

Antithrombotic
hemodialysis catheters are now available from several manufacturers.
These products have a heparin surface coating that resists the
deposition of plasma proteins and platelets for an extended period of
time. As demonstrated by in vitro studies this antithrombotic
activity is reported to persist for 30-90 days. However, there have
been no prospective clinical trials that have yet demonstrated the
clinical effectiveness of these surface coatings. At this point in
time the value of these products is based upon our understanding of
the biological processes the lead to catheter-related thrombosis and
infection, and the ability of these surface coatings to resist these
processes, but not upon any published clinical studies.

Catheter
malfunction results in the failure of 17% to 53% of THCs. This
consists of any mechanical problem interfering with its use,
including external catheter dislodgment and perforation as well as
the more usual issue of poor flow. When poor flow is encountered,
dialysis centers will typically try to switch the inflow and outflow
lines, or to fill each lumen with low dose tPA, 1 mg per lumen.

When
empiric low dose thrombolytic therapy fails, patients are generally
sent for radiological evaluation. Our routine first consists of
visual inspection for external kinks, holes, an excessively
constrictive stitch, or an exposed cuff. This is followed by
fluoroscopic inspection for internal kinks and proper positioning of
the tip, generally in the mid right atrium. Vigorous aspiration is
then performed to confirm poor flow. The first 5 to 10 ml of fluid
aspirated should be discarded as this contains the highly
concentrated heparin used to prime the lumens. Contrast is then
injected. If no mechanical problem is found, and if thrombolytic
therapy fails, we recommend catheter exchange.

Over
an 18 month period, we determined that the most common problems with
malfunctioning THCs were malpositioning and pericatheter thrombus,
followed by dislodgment and kinking. Malpositioning is managed by
placing the new catheter in a proper position. Fibrin sheaths and
small clots at the tips of catheters can be treated by disrupting it
with a large angioplasty balloon (e.g. 10 to 14 mm diameter x 4 cm
length), although immature ones can even be disrupted by simple
manipulations with a curled heavy duty wire. More extensive
pericatheter thrombus may require tPA infusions and patients who
rapidly develop thrombus should probably be anticoagulated. We no
longer use a transfemoral snare to strip fibrin sheaths. Even
tunneled catheters can be exchanged easily over a guidewire, which is
cheaper and faster than transfemoral catheter stripping.

THC
malfunction appears to be less likely with IJV access as compared to
the subclavian vein and with proper tip positioning in the right
atrium.

Reported
one year survival rates for long-term hemodialysis catheters range
from 25% to 74%, although these do not include the results of
radiological treatments. In the subgroup of patients with
malfunctioning catheters, we were able to achieve a 1 year primary
assisted patency of 50% using the above methods, although many
patients had more than one treatment.

Alternative
Central Vein Access

Translumbar
IVC catheters

Since
the initial report by Kenney ²³ in 1985, there have been
a series of reports on the success of percutaneous translumbar IVC
catheter placement. This technique is recommended when routine access
via the jugular or subclavian veins or SVC is unavailable. In
children, catheters are 6 to 8 French. In adults, single and
multilumen catheters ranging in size from 6 to 18 French can be used.
Both external catheters and implantable ports may be placed using
this approach. These catheter sizes support the full range of central
venous device applications, including dialysis. Anticoagulation
should be reversed and any coagulopathy corrected prior to device
placement. Prophylactic antibiotics are given if that is the general
protocol used for venous access device placement; a regimen providing
coverage for staphylococcus such as cephazoline 1 gram IV is
in common use. Conscious sedation and local anesthesia are given.
General anesthesia is rarely needed.

Technique

The
vein access site in adults is over the right iliac crest 8 to 10 cm
(4 finger breadths) lateral to the midline. Since the skin entry site
is in the back, the catheter is tunneled to a right flank exit
position for better patient comfort and easier use. Prior to skin
preparation, the proposed access and exit or port implantation sites
are chosen. A large sterile field is established covering the right
side of the back, right flank, and right anterior abdomen and chest.
The patient is positioned in a prone oblique position with the right
side sufficiently elevated to allow access to the planned exit site.

The
three procedure steps are establishment of venous access, tunneling,
and creation of the exit site or port pocket. The order is arbitrary
but access is usually first. Percutaneous catheterization of the IVC
using a translumbar approach is performed using C-arm fluoroscopy.
Direct IVC catheterization is analogous to translumbar aortography.
The IVC is generally more lateral than the aorta and is more oval in
shape ²⁴ . A preoperative CT scan confirms normal IVC
anatomy and helps to plan the best position and angle for needle
placement. Intra-operative sonographic guidance may help in pediatric
cases. Placement of a guiding catheter from a femoral vein approach
is used in the occasional case of difficult access but is not
routinely necessary. A short 1 to 1.5 cm transverse incision is made
at the access site. A 21-g needle is advanced at a cephalad oblique
angle entering the IVC below the renal veins at the L2-L3 level. The
needle may first be directed towards the anterior edge of the
vertebral body, then redirected just anterior to this to enter the
IVC. The deep tissues around the spine and IVC are often sensitive to
pain; deep anesthetic can be given through the access needle. A
palpable “pop” may be felt as the vein is entered. There
should be free aspiration of blood through the needle. A very soft,
platinum tipped 0.018 inch (0.46 mm) guidewire (e.g. Neff access kit,
Cook Inc., Bloomington, IN or AccuStick, MediTech Inc., Natick, MA)
is advanced through the needle and the coaxial dilator is then placed
over the wire. After exchange for a heavy duty guidewire (e.g.
Amplatz guidewire, Cook), the track is dilated and a peel-away sheath
is placed with its tip in the IVC. The sheath chosen should be 1- or
2- French larger than the desired catheter.

The
catheter exit site or port pocket may then be created. This is
usually in the right flank or below the right breast but can be
modified as needed. For ports, the pocket is made over the lower ribs
so that there is a secure base for palpating and accessing the port.
Special considerations apply for high flow catheters such as those
used for hemodialysis. Because these catheters are shorter than those
used for other purposes, a more lateral or posterolateral exit site
may be used.

The
tunnel is created between the sheath entry point and the exit or port
site. Lidocaine 1% is given subcutaneously along the course of the
tunnel. Any standard tunneling tool can be used. The tunnel is
usually longer than that used for chest wall catheters, and in
addition has to make a curved path around the flank. The Davol
tunneler (Bard, Salt Lake City, UT) is a 12 inch semi-rigid nylon
tunneler which can be used to pull a catheter from its nose or tail;
it is very effective for the long, flank tunnel used in this
procedure. In some cases it may be necessary to make an additional
flank incision and create the tunnel in two parts because of the
length of the tunnel and the curvature of the flank. This will also
be necessary in patients who have had prior surgery, injury, or skin
disease along the planned tunnel path. The catheter is pulled through
the tunnel and trimmed to the desired length. It is introduced
through the sheath and positioned at the cavoatrial junction; the
sheath is then peeled away. Depending on the type of catheter used,
the order and technique for tunneling and catheter length trimming
may need to be adjusted. For example, valved catheters such as the
Groshong (Bard) are placed in the vein prior to tunneling the back
end to the exit site.

The
catheter should be flushed immediately with heparin. The back
incision is closed using a subcuticular suture or butterfly adhesive
strip. For external catheters, an anchoring silk suture is tied
around the catheter at the exit site. The anchor suture should be
left in for 3 weeks while the subcutaneous Dacron cuff is fixed by
in-growth of connective tissue.

Ports
may be placed using this route. The port reservoir should be placed
over the ribs to provide a stable platform for access.

One
difficulty associated with the procedure is a tendency of the sheath
to kink at the point of entry into the IVC. The chance of this
happening can be decreased by making an oblique track to minimize the
angle at the junction between the sheath and the IVC. If a kink
prevents passage of the catheter, the operator can pass a guidewire
that has a low coefficient of friction (e.g. Terumo Glidewire)
through the catheter to stiffen it. The catheter and the wire can be
advanced as a unit into and through the sheath. If this is still not
successful, the guidewire may be advanced in front of the catheter
through the kink; the sheath is then retracted so that it is just
entering the IVC and is no longer bent, then the catheter and
guidewire can be advanced together.

Potential
procedure complications specific to this route include
retroperitoneal hemorrhage, injury to the ureter, and puncture
of an abdominal viscus such as the duodenum or colon. Despite the
catheter size, bleeding at the site of IVC entry has not been a
problem either at the time of placement or removal. Laceration of the
ureter has not been reported; establishing an access track close to
the spine should avoid the ureter. Viscus perforation has not been
described.

Results

Denny
reported successful use of the translumbar IVC route in seven
procedures in six patients with 6- to 12-French catheters, with no
case of subsequent catheter malposition or IVC thrombus ²⁵.
Lund placed 46 catheters in 40 patients ¹⁷. The catheters
used were 37 single lumen 14.4 French apheresis catheters, seven dual
lumen 12 French catheters, and two single lumen 9.6 French catheters.
IVC thrombus occurred in eight patients, two of which were occlusive.
All thrombi were successfully lysed with urokinase. Catheter
malposition was found at follow-up in five patients: four catheters
were repositioned angiographically, while one resumed its position
spontaneously. The patient's position, respiratory motion, coughing,
sneezing and vomiting all may cause the catheter to move. CT scans
were performed in 31 patients at 1 to 10 weeks (mean 13 days)
following the procedure and in 13 patients at 1 to 9 weeks (mean 19
days) following catheter removal. None showed retroperitoneal
hemorrhage. In three patients, the catheter tip migrated out of the
IVC. This was felt to be related to patient obesity and a short
intravascular catheter length. An additional cause of catheter
dislodgement in my personal experience has been premature removal of
the external catheter anchor stitch. Two patients in whom the anchor
stitch was removed at two weeks found the catheter in bed beside them
the next morning.

Hemodialysis
can be done using the translumbar route. A single case was reported
by Gupta et al ²⁶ . Lund reported placement of
17 catheters in 12 patients ²⁷ . All placements
were successful. Thrombotic complications occurred 7 times (0.33/100
patient days). Infectious complications occurred 6 times (0.28/100
patients days). Cumulative patency was 52% at 6 months and 17% at 12
months. These rates are comparable to those reported using
hemodialysis catheters at other sites.

The
use of the translumbar IVC technique in children has been reported by
several authors. Denny reported a two year old child with short bowel
syndrome in whom a single lumen 7 French Broviac catheter was placed
⁴. Robertson et al reported three children ²⁸ .
Azizkhan et al subsequently reported excellent results in seven
children using the translumbar approach for four catheters and the
transhepatic approach for seven; all but one patient were younger
than 2.5 years ²⁹. Recently, Malmgren et al reported 12
catheters placed in 4 children. All procedures were successful and
uncomplicated. The median catheter patency was 4.8 months (range 1 to
10 months).

Transhepatic
IVC catheters

Placement
of a long-term central catheter using a transhepatic approach to the
IVC was described in a case report by Crummy et al ³⁰ .
Only a small number of cases have been reported, although anecdotal
reports suggest that its use is more common. Transhepatic IVC access
is a good alternative when the translumbar approach is not feasible.
Catheters and ports both large and small can be placed using this
method. Indications for catheterization in the reported patients have
mostly centered on parenteral nutrition. I am unaware of hemodialysis
access being placed this way.

Technique

The approach is
analogous to that of transhepatic cholangiography and biliary
drainage. Access is from the right flank intercostal or anterior
subcostal approach. Ultrasound and CT are used in procedure planning.
Ultrasound is often used for intra-procedural guidance; ultrasound
helps to avoid major intrahepatic vascular structures and large
biliary radicles. Access to the IVC can be planned either through an
hepatic vein or direct to the IVC. For the former, the middle hepatic
vein is the largest and is the primary choice.

After
choosing the access and exit or port sites, the skin is cleansed
widely. The use of broad spectrum antibiotics as prophylaxis is
suggested for this route although data are lacking. Local anesthesia
is given subcutaneously using lidocaine 1% and is carried down to the
liver capsule. A 21 gauge needle is advanced with fluoroscopy or
ultrasound guidance. When blood is freely aspirated, a soft-tipped,
platinum 0.018 inch (0.46 mm) guidewire is advanced. A coaxial
dilator (e.g. Neff set, Cook or AccuStick, MediTech) is placed and a
heavy duty guidewire (Amplatz guidewire, Cook) is advanced to the
right atrium or the SVC. A peel-away sheath large enough to
accommodate the desired catheter is then placed.

The
exit site is then prepared. Ports should be placed over the ribs for
a secure base of access. External catheters can be brought out where
convenient. The catheter is passed between exit and access site in
whichever direction is appropriate for the device being used. Similar
to the translumbar route (see above), a lengthy, complex tunnel may
be needed. The catheter is cut to length; due to the relatively short
intravascular distance, the catheter tip is usually left in the right
atrium. Excessive length in the atrium may lead to atrial arrhythmias
and bothersome palpitations, or prolapse across the tricuspid valve
with ventricular irritability. Care should be taken to leave a
sufficient length of intravascular catheter to avoid inadvertent
malposition during respiratory excursion of the liver. Some degree of
slack at the entry point into the liver may be advisable. In the
pediatric age group, patient growth may greatly shorten the
intravascular portion. Hepatic vein rather than direct IVC access may
be preferred in these patients so that intravascular length is
maximized ^{31 11}.

Results

Kaufman
et al reported placement of a 9 French catheter for long-term total
parenteral nutrition with five month followup in a patient with
infrarenal IVC thrombus ³². In the report by Azizkhan 11
transhepatic catheters were successfully placed in seven children ^31
11. Hepatic vein entry was chosen to maximize the intravascular
length.

No
procedure complications attributable to the transhepatic approach
have been reported, although the number of reported cases is quite
small. There are anecdotal reports of catheter tip dislodgment which
are probably due to the excursion of the liver during respiration.

Access using
Collateral Veins

Even
in cases of extensive central vein occlusion, including the IVC,
alternative routes of percutaneous catheterization often can be
found. The modern angiography suite with state-of-the-art
fluoroscopy, digital angiography, and imaging aids such as road
mapping and last image hold assists in localization and puncture of
small collateral veins. Intra-operative sonography provides added
guidance in many cases. Collateral veins will often be hypertrophied
from high venous flow secondary to central vein obstruction. Small
vessel catheterization systems including 21 gauge needles, 0.018 in.
(0.46 mm) platinum or gold tipped guidewires, and small sheaths are
used for cannulation and passage of catheters through collateral
networks to the central veins. Due to the smaller size of these veins
and their more circuitous course, the catheter size may be restricted
depending on the access used. This will limit the usefulness of these
approaches in high flow applications such as dialysis.

Azygos
and Hemizygos veins

Denny
described the placement of a 6 F Broviac catheter into the right
atrium using the hemizygos system in a patient with SVC occlusion
above the azygos vein and IVC occlusion between the renal and hepatic
veins ³¹ . Venous return was via enlarged
azygos and hemizygos veins. The hemizygos vein was punctured under
direct fluoroscopic control during contrast injection using a
catheter in the left renal vein. A guidewire, catheter, and sheath
were advanced through the hemizygos and azygos veins to the SVC and
right atrium. A 6 French Broviac catheter was tunneled from the left
flank. Its tip was introduced through the sheath and positioned at
the junction of the SVC and right atrium.

Intercostal veins

Percutaneous
use of the intercostal veins for alternative access has been the
subject of several reports. Prior to this, there were reports of
intercostal cannulation using surgical cutdown. Meranze et al
reported a patient with SVC occlusion between the right atrium and
the azygos vein, and occlusion of the left subclavian vein; venous
return was through the azygos vein to the IVC ³³. The
right subclavian vein had been thrombosed previously and was now
recanalized but narrow. A combined surgical and radiological approach
was used. A hypertrophied intercostal vein was catheterized through
the narrowed right subclavian vein to the SVC and azygos vein. A
stone basket was exchanged for the catheter. After surgical cutdown
of the intercostal vein over the stone basket, a silicon rubber
catheter was grasped and pulled from the intercostal vein to the high
flow azygos system. Direct puncture and catheterization of an
intercostal vein was described by Andrews ³⁴ . A 19 year
old man requiring long-term central access for total parenteral
nutrition had occlusion of both subclavian and brachiocephalic veins.
Femoral venogram showed occlusion of the external and common iliac
veins and the infrarenal IVC. Hypertrophied intercostal venous
collaterals were present. A 20 gauge needle and 0.018 inch guidewire
were used to puncture a lower left posterior intercostal vein under
fluoroscopic control with contrast injection from the femoral vein
catheter. A 6.5 French catheter was tunneled from the side and
introduced through the intercostal vein, to the hemizygos and azygos
veins to the SVC. The procedure was uncomplicated and there was
satisfactory catheter function during the six week followup period.

Kaufman
et al reported a case of central catheterization from the arm using a
chest wall collateral and an intercostal vein ³⁵ . A 27
year old woman with Hodgkin’s disease had occlusion of the
subclavian and brachiocephalic veins and supra-azygos SVC due to
compression from adenopathy. The basilic vein was punctured. Using a
hydrophilic-coated guidewire and catheter (Terumo, Tokyo, Japan), a
lateral chest wall collateral was catheterized. From there, the third
right intercostal vein, azygos vein, and superior vena cava were
catheterized in sequence. An arm port with a 5.8 French catheter (PAS
Port, Sims Deltec, St. Paul, MN) was placed using this technique.

Access
Using Recanalization of Central Veins

The
fourth category of alternative access is the use of occluded veins
using guidewire recanalization, angioplasty, and stent placement. The
use of an occluded vein may aggravate symptoms of venous blockade
such as arm, head, or neck edema. Thrombosis may progress after
catheter placement. On the other hand, there are many anecdotal
reports of successful recanalization and catheter placement using the
jugular and subclavian veins. Torosian et al reported the results of
a combined surgical and radiological approach in three patients with
SVC occlusion with excellent outcome ³⁶ . More recently,
Ferral et al reported their results using recanalization of occluded
central veins from a femoral approach ³⁷ .
Successful access was achieved using the right axillary vein in two,
thyrocervical collaterals in two, posterior vertebral vein in one,
and the external jugular vein in one. Catheters were used for
hemodialysis in four and for antibiotics in two. In their technique,
a catheter and guidewire were advanced through the occluded jugular
or subclavian vein from the femoral approach. The vein access site
was then punctured using a micropuncture kit. The guidewire was
grasped using a snare and pulled through the occluded vein. In this
manner, through and through access can be obtained, which simplifies
vein dilation and catheter placement. This is a technically appealing
approach whose long-term success and complications are as yet
undefined.

Angioplasty
and stent placement has been used to treat venous occlusions and
stenosis of various etiologies. This can be applied to venous access
as well. Venous strictures can form at the tip of a catheter as a
result of local thrombosis or vein irritation. Occlusion of the vein
may obstruct the catheter and cause edema. There are many anecdotal
reports of treatment of such obstructions using interventional
techniques.

Surgical
Needles:

There
are several types of surgical needles that are available for
performing various surgical tasks. A detailed discussion of all the
different types of needles is beyond the scope of this summary.
However, a basic understanding of the different types of needles is
important for anyone performing wound closure. Regardless of the
ultimate intended use, all surgical needles have three basic
components: the eye, the body, and the point. The eye of the needle
is the back end of the needle, which contains the suture material.
Most sutures today come with the material pre-attached to the suture
needle. The body or shaft is the portion that is usually used for
grasping. The cross-sectional configuration of the body may be
round, oval, side-flattened, rectangular, triangular, or trapezoidal.
The longitudinal shape of the body may be straight, half curved,
curved, or compound curved. The most commonly used needles are
curved. Curvatures may be 1/4, 3/8, 1/2, or 5/8ths circle. The most
commonly used curved needle is the 3/8 circle.

The
point of the needle is defined as the extreme tip of the needle to
the maximum cross-section of the body. Sharpness of needle point,
shape and size are important characteristics. Each specific point is
designed and produced to the required degree of sharpness to smoothly
penetrate the types of tissues to be sutured. The basic needle
shapes are cutting, tapered, and blunt. Cutting needles have at
least two opposing cutting edges. These edges are sharpened so that
they will cut through tissue that is tough and difficult to
penetrate. A conventional cutting needle has two opposing cutting
edges with a third cutting edge on the apex of the triangular
configuration. The cross sectional shape changes from triangular to
flat as it approaches the body. Reverse cutting needles differ from
the conventional cutting configuration in that the third cutting edge
is located on the outer convex curvature as oppose to the inside
concave curvature. This design offers the advantage of having the
flat surface closest to the edges of the incision or wound. Thus,
the hole left by the needle leaves a wide wall of tissue for the
suture to be tied against and reduces the danger of tissue cut out.
The taper needle is used primarily on soft easily penetrated tissues,
such as peritoneum, abdominal viscera and subcutaneous tissue. This
needle is rounded at its tip and tapers to a fine point. These
needles are not well suited for penetrating skin or dermal tissue.

Needles
vary in size and wire gauge. The appropriate needle diameter as well
as needle radius should be chosen to match the size of the wound that
requires closure. For closing subcutaneous tissue during the
implantation of subcutaneous venous access devices a 26 mm taper
needle is ideal. For closure of the dermal layer a 19mm cutting
needle is preferred.

Removing
Tunneled Catheters

There
is an amazing absence of information on this topic, despite the known
difficulties. This can be a simple process if thought goes into the
placement of the cuff at the time of the initial device procedure. By
placing the cuff within 2-3 cms of the skin exit, the cuff and
surrounding tissues are accessible to dissection through the skin
opening.

If
the catheter has been in place for less than a month, it will often
be possible to remove it using direct traction, with or without local
lidocaine around the cuff. Otherwise, removal using sharp dissection
will be required.

A
subcutaneous cuff which is within 1 to 3 cms of the skin exit site
can be freed by sharp dissection through the exit site opening in the
skin. Lidocaine is infiltrated around the cuff through the skin
opening. The catheter is retracted, pulling the cuff closer to the
exit. Using a pair of small sharp-sharp scissors, the cuff is
progressively freed by cutting carefully around the catheter,
reaching through the skin opening. The operator must be careful to
open and close the scissors parallel to the catheter to avoid
inadvertent transection of the catheter itself, which will tend to
make removal more difficult. With cutting and steady retraction, the
catheter can be freed in 5 minutes or less. This is an appealing
alternative to a number of other Rube Goldberg-ian proposals
involving pulleys, weights, and abrupt door closures.

If
the cuff is placed deeper in the tunnel than 4 cms, then a small
incision will usually need to be made over the cuff; the cuff is then
freed using direct exposure and sharp dissection with scalpel and
scissors.

Suture
Material and Wound Closure

The
mainstay of wound closure is suture material. Adhesive strips (e.g.
Steristrips) are also extremely useful for skin surface closure and
wound edge alignment.

Skin
glue, 2-octyl cyanoacrylate, (Dermabond) has been available for a
number of years for wound closure, and has found use in a number of
applications, including emergency laceration treatment and in plastic
surgery. While there is little direct data for the use of skin glue
in venous access procedures, many practitioners have begun to use
skin glue for wound edge closure during port placement, and for
closure of the vein access incision for tunneled catheters. In
addition to potentially simplifying the closure process, and saving
time, skin glue provides an immediate sterile dressing which is
waterproof. Patients may be able to shower or clean within 24 hours
after device placement with closure assisted by skin glue. Skin glue
may replace subcuticular sutures, or augment them, following port
placement. They do not replace deep sutures.

Sutures
can be conveniently divided into two broad groups: absorbable and
non-absorbable. These two broad categories can be further subdivided
into monofilament and multifilament. The monofilament suture is made
of a single strand whereas the multifilament suture consists of
several filaments twisted or braided together. The monofilament has
the advantage of not harboring micro- organisms within the braid of
the material, however, its handling and tying characteristics are not
as good as the multifilament suture. The size and tensile strength
for all suture material are standardized by specific regulations.
The diameter of the material is designated by “0”. The
more “0”s in the numbers the smaller the size of the
strand. Thus 4-0 suture is thinner then 3-0 suture. The accepted
surgical axiom is that the tensile strength of any suture need never
exceed the tensile strength of the tissue it holds. Thus both the
size and the tensile strength of that particular material is
important in choosing the correct suture.

Absorbable
Sutures

There
are several different types of absorbable sutures, which have varying
absorption times. These types of suture include surgical gut,
collagen, polyglactin, polyglycolic acid, and polydioxanone.
Polyglactin is the most commonly used absorbable suture for the
closure of skin wounds. This braided material undergoes hydrolysis
and is completely absorbed between the sixtieth and ninetieth day.
Enzymes are not required to break down this polymer. Only water is
required. Thus, synthetic absorbable sutures exhibit a very low
degree of tissue reaction when compared to surgical gut or collagen.

Non
Absorbable Sutures

A
host of non-absorbable sutures have been used for a variety of
different surgical closures. These materials include silk, cotton,
linen, stainless steel, nylon, polyester, and polypropolene. Each of
these sutures have different characteristics which make their use
more appropriate in one situation over another. The advantage of
non-absorbable sutures is that they maintain their structural
integrity for a long period of time and cause very little tissue
reaction. The one disadvantage of this material when used to close
skin is that they must be removed in time.

Wound
Closure

In
implanting subcutaneous venous access devices one must be familiar
with the closure of subcutaneous tissue and skin. Since the device
sits below several mm of subcutaneous fat it is often necessary to
close this tissue over the device. There is some debate as to
whether this layer of fat should indeed be closed. One argument
maintains that the presence of even a single suture, since it is a
foreign body, can increase the risk of infection. On the other hand,
if this layer is not closed, a dead space can be left. This dead
space can allow the accumulation of fluid, which can delay healing
and predispose to the development of infection. In our practice, we
recommend closing this layer with an absorbable suture material. The
suture material that we most commonly use is a 4-0 suture material
which will reabsorb such as Vicryl or Monocryl. The subcutaneous
layer is closed to eliminate dead space and also to approximate the
wound edges. It is important when closing this layer to ensure that
the knots lie deep to the subcutaneous tissues so that they do not
interfere with skin healing or cause patient discomfort. To
accomplish this, the suture must be inserted into the tissues in a
specific fashion. On one side of the wound the needle is initially
placed in the tissue from deep to superficial. As it goes across to
the other side of the wound, the needle is placed from superficial to
deep. Thus, when the knot is tied, it lies facing the deeper
subcutaneous tissue.

Subcuticular
closure

The
subcuticular layer is a layer of tough connected tissue just below
the skin. If sutured, this layer will hold the skin edges in close
approximation, resulting in a good cosmetic closure. With the
subcuticular closure there can be scar expansion over time. However,
a simple interrupted skin closure often results in additional
scarring as the needle passes through the skin on each side of the
wound being closed. Each of these needle passes can result in a
small scar and gives rise to the typical railroad track appearance.
Since the subcuticular stitch never exits the epidermis no such
scarring is noted. Absorbable synthetic sutures whose tensile
strength releases in 28 days are ideally suited for continuous dermal
closures. In contrast, non-absorbable dermal continuous sutures can
be utilized however the end must exit the skin and be removed after
wound healing has taken place. In our practice either 4.0 or 5.0
suture absorbable (polyglactin) filament is used. The continuous
dermal suture is begun as an interrupted anchoring dermal suture by
implanting the first stitch at one apex of the wound. After tying a
secure knot, the non-needle suture end is cut short. The next stitch
is passed horizontally from the apex of the wound, where the knot was
just tied, through the superficial dermis. After exiting the dermis,
the position of next bite is identified by pulling the suture across
at right angles to the wounds. The needle should enter the dermis
and exit the dermis just below the epidermis. Each bite should be
less than 3 mm to avoid skin puckering and malposition of the wound
edges. When the suture reaches the opposite apex the last suture
bite is not pulled through but kept as a loop to form the closure
knot. While not essential, it is helpful to pass a stitch back from
the apex to the region of the suture loop on the opposite side of the
wound so that when the knot is tied the end are directly opposite one
another as to avoid skin puckering. Once the knot is tied, the loop
is cut away and the remaining portion of the suture containing the
needle is passed through the middle of the closed wound and brought
out away from the wound through the skin. By retracting on the
suture, the knot will pull underneath the dermal layer. Once the
skin edges have been approximated in this fashion, one can apply
steri-strips to add additional tensile strength and better close the
epidermal layer.

Conclusion

In
the implantation of subcutaneous venous access devices, the closure
of subcutaneous fat as well as skin is important. While skin can be
closed with interrupted non-absorbable sutures, this often leads to
excessive scarring which can be avoided by closing the wound in a
subcuticular fashion. While the subcuticular closure is technically
more challenging, it is worthwhile to master this technique as it
results in the most cosmetically pleasing result.

References

1. Eastridge
   BJ, Lefor AT. Complications of indwelling venous access devices in
   cancer patients. J Clin Oncol 1995; 13(1):233-8.

   
   

2. Simmons
   JR, Buzdar AU, Ota DM, Marts K, Hortobagyi GN. Complications
   associated with indwelling catheters. Med Pediatr Oncol 1992;
   20(1):22-5.

   
   

3. Haire
   WD, Lieberman RP, Lund GB, Edney JA, Kessinger A, Armitage JO.
   Thrombotic complications of silicone rubber catheters during
   autologous marrow and peripheral stem cell transplantation:
   prospective comparison of Hickman and Groshong catheters. Bone
   Marrow Transplant 1991; 7(1):57-9.

   
   

4. Lowell
   JA, Bothe A, Jr. Central venous catheter related thrombosis. Surg
   Oncol Clin N Am 1995; 4(3):479-92.

   
   

5. Groeger JS,
Lucas AB, Coit D, et al. A prospective, randomized evaluation of the
effect of silver impregnated subcutaneous cuffs for preventing
tunneled chronic venous access catheter infections in cancer patients
[see comments]. Ann Surg 1993; 218(2):206-10.

6. Groeger JS,
Lucas AB, Thaler HT, et al. Infectious morbidity associated with
long-term use of venous access devices in patients with cancer [see
comments]. Ann Intern Med 1993; 119(12):1168-74.

7. Szycher M,
Siciliano A, Reed A. Polyurethanes in Medical Devices. Medical
Design and Material 1991.

8. Phillips RE,
Smith MC, Thoma RJ. Biomedical applications of polyurethanes:
Implications of failure mechanisms. J Biomater Appl 1988;
3(2):207-27.

9. Reed AM, Potter
J, Szycher M. A solution grade biostable polyurethane elastomer:
ChronoFlex AR. J Biomater Appl 1994; 8(3):210-36.

10. Lam S, Scannell
R, Roessler D, Smith MA. Peripherally inserted central catheters in
an acute-care hospital. Arch Intern Med 1994; 154(16):1833-7.

11. Cardella JF, Fox
PS, Lawler JB. Interventional radiologic placement of peripherally
inserted central catheters. J Vasc Interv Radiol 1993;
4(5):653-60.

12. Salem RR, Ward
BA, Ravikumar TS. A new peripherally implanted subcutaneous permanent
central venous access device for patients requiring chemotherapy. J
Clin Oncol 1993; 11(11):2181-5.

13. Vanholder R,
Ringoir S. Vascular access for hemodialysis. Artif Organs
1994; 18(4):263-5.

14. Mauro MA, Jaques
PF. Radiologic placement of long-term central venous catheters: a
review. J Vasc Interv Radiol 1993; 4(1):127-37.

15. Abrahm JL,
Mullen JL. A prospective study of prolonged central venous access in
leukemia. Jama 1982; 248(21):2868-73.

16. Yerdel MA,
Karayalcin K, Aras N, Bozatli L, Yildirim E, Anadol E. Mechanical
complications of subclavian vein catheterization. A prospective
study. Int Surg 1991; 76(1):18-22.

17. Lund GB,
Lieberman RP, Haire WD, Martin VA, Kessinger A, Armitage JO.
Translumbar inferior vena cava catheters for long-term venous access.
Radiology 1990; 174(1):31-5.

18. Denny DF, Jr.
Placement and management of long-term central venous access catheters
and ports. AJR Am J Roentgenol 1993; 161(2):385-93.

19. Denys BG,
Uretsky BF, Reddy PS. Ultrasound-assisted cannulation of the internal
jugular vein. A prospective comparison to the external
landmark-guided technique. Circulation 1993; 87(5):1557-62.

20. Gualtieri E,
Deppe SA, Sipperly ME, Thompson DR. Subclavian venous
catheterization: greater success rate for less experienced operators
using ultrasound guidance [see comments]. Crit Care Med 1995;
23(4):692-7.

21. Higano ST, Hayes
DL, Spittell PC. Facilitation of the subclavian-introducer technique
with contrast venography. Pacing Clin Electrophysiol 1990;
13(5):681-4.

22. Hahn ST,
Pfammatter T, Cho KJ. Carbon dioxide gas as a venous contrast agent
to guide upper-arm insertion of central venous catheters [see
comments]. Cardiovasc Intervent Radiol 1995; 18(3):146-9.

23. Kenney PR,
Dorfman GS, Denny DF, Jr. Percutaneous inferior vena cava cannulation
for long-term parenteral nutrition. Surgery 1985; 97(5):602-5.

24. Cazenave FL,
Glass-Royal MC, Teitelbaum GP, Zuurbier R, Zeman RK, Silverman PM. CT
analysis of a safe approach for translumbar access to the aorta and
inferior vena cava. AJR Am J Roentgenol 1991; 156(2):395-6.

25. Denny DF, Jr.,
Greenwood LH, Morse SS, Lee GK, Baquero J. Inferior vena cava:
translumbar catheterization for central venous access. Radiology
1989; 172(3 Pt 2):1013-4.

26.
Gupta A, Karak PK, Saddekni S. Translumbar inferior vena cava
catheter for long-term hemodialysis. J Am Soc Nephrol 1995;
5(12):2094-7.

27.
Lund GB, Trerotola SO, Scheel PJ, Jr. Percutaneous translumbar
inferior vena cava cannulation for hemodialysis. Am J Kidney Dis
1995; 25(5):732-7.

28.
Robertson LJ, Jaques PF, Mauro MA, Azizkhan RG, Robards J.
Percutaneous inferior vena cava placement of tunneled silastic
catheters for prolonged vascular access in infants. J Pediatr Surg
1990; 25(6):596-8.

29.
Azizkhan RG, Taylor LA, Jaques PF, Mauro MA, Lacey SR. Percutaneous
translumbar and transhepatic inferior vena caval catheters for
prolonged vascular access in children. J Pediatr Surg 1992;
27(2):165-9.

30.
Crummy AB, Carlson P, McDermott JC, Andrews D. Percutaneous
transhepatic placement of a Hickman catheter [letter]. AJR Am J
Roentgenol 1989; 153(6):1317-8.

31.
Denny DF. Central venous access via the hemizygous vein. In:
Trerotola SO SS, ed. Venous Interventions: Society of Vascular and
Interventional Radiology, 1995; 507-510.

32.
Kaufman JA, Greenfield AJ, Fitzpatrick GF. Transhepatic cannulation
of the inferior vena cava. J Vasc Interv Radiol 1991;
2(3):331-4.

33.
Meranze SG, McLean GK, Stein EJ, Jordan HA. Catheter placement in the
azygos system: an unusual approach to venous access. AJR Am J
Roentgenol 1985; 144(5):1075-6.

34.
Andrews JC. Percutaneous placement of a Hickman catheter with use of
an intercostal vein for access. J Vasc Interv Radiol 1994;
5(6):859-61.

35.
Kaufman JA, Crenshaw WB, Kuter I, Geller SC. Percutaneous placement
of a central venous access device via an intercostal vein. AJR Am
J Roentgenol 1995; 164(2):459-60.

36.
Torosian MH, Meranze S, McLean G, Mullen JL. Central venous access
with occlusive superior central venous thrombosis. Ann Surg
1986; 203(1):30-3.

37.
Ferral H, Bjarnason H, Wholey M, Lopera J, Maynar M, Castaneda-Zuniga
WR. Recanalization of occluded veins to provide access for central
catheter placement. J Vasc Interv Radiol 1996; 7(5):681-5.

38.
Hovsepian DM, Bonn J, Eschelman DJ. Techniques for peripheral
insertion of central venous catheters. J Vasc Interv Radiol 1993;
4: 795-803.

39.
Chrisman HB, Omary RA, Nemcek AA, et al. Peripherally Inserted
Catheters: Guidance with use of US versus Venography in 2,650
Patients. J Vasc Interv Radiol 1999; 10-473-75.

40.
Maki D, Kluger DM, Crnich CJ . The Risk of Bloodstream Infection in
Adults With Different Intravascular Devices:

A
Systematic Review of 200 Published Prospective Studies. Mayo Clin
Proc. 2006;81:1159-1171

41.
Glenn B. Single-Incision method for the Placement of an Implantable
Chest port or a Tunneled catheter. J Vasc Interv Radiol 2007; 18:
137-140.

42.
Contractor SG, Phatak TD, Bhagat N. De Novo Placement of a Tunneled
Dialysis Catheter over a Wire with the Single-Incision Technique. J
Vasc Interv Radiol 2008;19: 1255-1259.

Page
1 of 19