CRITICAL CARE NEPHROLOGY

Central Venous Dialysis Catheter Dysfunction

Martine Leblanc, Jean-Yves Bosc, Emil P. Paganini,
and Bernard Canaud
Central venous catheter dysfunction is a limiting factor in regard to renal replacement therapy
efficiency and can thus influence patient morbidity. Early catheter dysfunction is frequently due to
mechanical problems such as inadequate positioning, kinking, or constriction, but early fibrin
deposition can develop soon after insertion. Delayed dysfunction usually results from thrombus
formation, either within the lumen, around the catheter ("fibrin sleeve"l, or in the host vein. Catheter
dysfunction is suspected clinically or documented by simple imaging studies. It is usually evident and
manifested by failure to aspirate blood from the lumen(sl, inadequate blood flow and/or high
resistance pressures during hemodialysis. However, a more subtle dysfunction may lead to a high
recirculation of dialyzed blood and be overlooked if dialysis adequacy is not monitored regularly. Local
instillation of a fibrinolytic agent is usually successful in restoring catheter patency. Central venous
dialysis catheters present intrinsic limitations consequent to their composition and design, whereas
extrinsic limitations result from site of insertion, blood properties and anatomic particularities of a
given individual. These characteristics largely determine overall catheter performances. Performance
parameters to consider include maximal consistently achievable blood flow rate, resistance to blood
flow indicated by arterial and venous pressures during hemodialysis, and blood recirculation rate.
Catheter longevity is an important consideration for cuffed catheters implanted for long-term use. The
tolerated blood recirculation within central venous dialysis catheters should be below 10% to 15%, and
is ideally between 3% to 7% in most clinical settings. Several recent studies confirm that short femoral
catheters recirculate significantly more than is desirable. Well functioning and non reversed internal
jugular and subclavian venous catheters have, in general, recirculation rates less than 5%. With regard
to various performance criteria, the TwinCath (Medcomp, Harleysville, PAl appears particularly
advantageous. In any case, a good catheter maintenance program is of critical importance for the
prevention and the early detection of catheter dysfunction.
© 1997 by the National Kidney Foundation, Inc.
Index Words: Blood flow; catheter; dialysis; recirculation; thrombosis.

C atheter malfunction is one of the most

frequent and frustrating events in hemo-
dialysis that can increase the morbidity of
renal failure patients. In addition to the associ-
ated inconvenience and complications result-
ing from the extra manipulations frequently
required, the poor function of the dialysis
catheter can significantly compromise the effi-
cacy of renal replacement therapy. The follow-
ing discussion provides some information
about the causes of catheter malfunction, clues
to catheter dysfunction, available diagnostic
tools, and potential therapeutic and preven-
tive modalities.
Causes of Catheter Dysfunction
The causes of dialysis catheter dysfunction
may vary depending on the time elapsed since
placement. Immediate or early malfunction
usually results from mechanical problems in-
cluding malpositioning of the distal tip in the
vessel, catheter kinking in the subcutaneous
tunnel, or catheter constriction from a tight
ligature; these causes can be easily corrected.
Less frequently, early fibrin formation will
partially obstruct the lumen(s), reducing blood
flow availability, and causing a suction effect
on the venous wall on aspiration.
Delayed or late dysfunction of central ve-
nous dialysis catheters (arbitrarily defined here
as 15 days post-cannulation) has more com-
plex etiologies that generally relate to thrombo-
genesis, either as an obstruction inside the
catheter itself (thrombus that covers the inte-
rior of the catheter lumen, thus reducing its
diameter and obstructing its distal and/or
side holes), or as an external fibrin sleeve
surrounding the distal catheter and occluding
the side holes, 1 or finally, as a thrombosis of
the central host vein (described in the superior
vena cava, subclavian, internal jugular, brachio-
cephalic or femoral veins).2-6 Secondary dis-
From the Department of Nephrology, Lapeyronie Hospital,
Montpel/ier, France; and the Section of Dialysis and Extracorpo-
real Therapy, The Cleveland Clinic Foundation, Cleveland, OH.
Address correspondence to Bernard Canaud, MD, Division
of Nephrology, Lapeyronie Hospital, Montpel/ier 34295 France.
© 1997 by the National Kidney Foundation, Inc.
1073-4449/9710404-0010$3.00/0

Advances in Renal Replacement Therapy, Vol 4, No 4 (October), 1997: pp 377-389 377

378 Leblanc et al
lodging and kinking of the catheter can also
occur with long-term use.
A previous study concerning acute renal
failure patients showed that temporary (not
tunneled) dual-lumen subclavian dialysis cath-
eters present early mechanical dysfunction
more often than catheters inserted at other
sites, and this led to access failure as early as
the third hemodialysis session? A potential
explanation comes from the anatomic features
of the infraclavicular region that may cause
catheter kinking in the subcutaneous tunnel
near the clavicular bone.
Catheter Thrombogenicity
Catheter-related thrombosis remains a leading
cause of catheter dysfunction. Three broad
categories are recognized: (1) lumen throm-
bus, (2) fibrin sleeve gradually developing
over the exterior surface of the catheter from
the site of insertion down to the tip; and (3)
mural thrombus along the venous wall prob-
ably after an endothelial injury. Figure 1 shows
a fibrin sleeve found tightly attached to a
removed dialysis catheter; the pathological
analysis showed the exclusive presence of
fibrin.
These subtypes of thrombotic complica-
tions, alone or in combination, occur more
frequently than suspected on clinical grounds.
Early, partial, or complete thrombosis of femo-
ral, internal jugular, or subclavian veins with
indwelling catheters have been detected in
Figure 1. A fibrin sleeve that was tightly attached
to the external part of a removed Silastic dialysis
catheter.
20% to 70% of cases using various diagnostic
modalities. 4,8 In addition to the uncommon
finding of ipsilateral limb edema and dis-
tended veins, difficulty in cannulating the
vein, in achieving the desired blood flow
during dialysis, and/ or in aspirating blood
from the catheter all suggest that a thrombus
has formed. These are indeed expected find-
ings with poorly functioning catheters.
Whenever foreign material invades the
bloodstream, circulating platelets tend to ad-
here and aggregate on its surface, subse-
quently leading to the activation of the coagu-
lation cascade with the development of a fibrin
layer and a clot. There are a number of factors
involved in catheter thrombogenicity.
Catheter Properties
Catheters are manufactured from standard
polymers, including polyvinyl chloride, poly-
tetrafluoroethylene (Teflon), polyethylene,
polyurethane, and silicone elastomer (Silastic).
Most of these materials promote thrombus
formation, and the procoagulant effect of cath-
eter material is most likely mediated by plate-
lets and/ or activation of the contact phase. 9
Because polyurethane and particularly sili-
cone elastomer have been associated with less
thromboembolic complications, both are con-
sidered to be the most suitable materials for
catheterization. 1o,1l
Surface roughness and irregularities pro-
mote thrombogenicity because they serve as a
nidus for thrombus deposition; indeed, it has
been shown that fibrinogen is preferentially
adsorbed onto microscopic surface crevices. 12
Subtle changes in material manufacturing and
processing can also affect the blood response
and favor thrombogenesis.
Several surface treatments, such as heparin
immobilization onto the catheter surface, ion
implantation of protective substances offering
thrombus or infection-resistance, and protein
conditioning (with the purpose of inhibiting
fibrinogen adsorption or of specifically promot-
ing albumin adsorption which is inert in re-
gard to coagulation) have been attempted to
improve biocompatibility12-14; preliminary re-
sults appear promising15 and further clinical
evaluations are underway. However, heparin-
coated devices are costly and appear mostly
 Critical Care Nephrology: Catheter Dysfunction
effective in the short-term because the coating
tends to leach off or dissolve over time.!6
Catheter stiffness may be more important
than its chemical and surface composition in
regard to thrombogenicity. Stiff, inflexible cath-
eters may induce endothelial injury and vessel
wall abrasion, whereas their softer counter-
parts are more likely to prevent damage to the
endothelium. Compared to long and stiff poly-
ethylene catheters, either heparin-impreg-
nated or not, soft pliable silicone elastomer
catheters have been shown to cause less throm-
bosis.!1 Nevertheless, some polyurethane cath-
eters that appear rigid at room temperature
soften when located inside the body. When
considering all of these factors, polyurethane
appears to be a reasonable choice for short-
term catheterization, whereas Silastic is pre-
ferred for longer or "permanent" usey-2o In
most centers, extremely compliant silicone
catheters ordinarily require a surgical place-
ment and, consequently, are not readily avail-
able in the acute setting. Nevertheless, short-
term temporary silicone catheters that do not
necessitate subcutaneous tunneling and that
are easy to insert at the bedside over a
guidewire are now available (Hemo-Cath,
Medcomp, Harleysville, PA, and Uldall, Cook
Critical Care, Bloomington, IN). On the other
hand, percutaneous insertion of tunneled cath-
eters is possible at the bedside using the
Seldinger technique, with or without the aid of
a pull-apart introducer (TwinCath, Medcomp,
Harleysville, PA, and PermCath, Quinton In-
strument Co, Seattle, WA).18,21
Patient-Related Factors
Besides catheter material and stiffness, throm-
bus formation involves interactions among
several other factors: size of the host vessel
and state of its wall; local blood flow and
hemodynamics; and catheter residence time
within the vessel. In addition, certain blood
characteristics can further contribute to the
clotting tendency (eg, enhanced viscosity from
a high hematocrit value, reactive thrombocyto-
sis, and fibrinogen release in inflammatory
states).
There exists an inverse relationship be-
tween the incidence of catheter-related throm-
bus formation and the diameter of cannulated
vessels. 12 In large central veins, there is less
379
contact between the catheter itself and the
vessel wall; this factor is then probably not as
important for adults with dialysis catheters in
large veins. Nonetheless, central venous dialy-
sis catheters are located in low pressure ves-
sels, with diameters that are very sensitive to
changes in intravascular volume. Local hemo-
dynamic factors are important to consider
because they can be responsible for flow distur-
bances associated with thrombus formation.
To reduce these flow disturbances, the design
of the catheter should include a smooth sur-
face and contours avoiding bends, bifurca-
tions and sudden changes in diameter, thus
partially eliminating the resistance to fluid
flow that promotes thrombogenesis. Finally,
the distribution of catheter holes at the distal
end may confer greater or lesser stability
within the bulk flow of the vein, consequently
influencing local hemodynamics.
Expected Catheter Performance
Versus Dysfunction
Limitations of dialysis catheters inserted into
central veins depend in part upon their compo-
sition, geometry, and location. The catheter
material itself affects compliance-elastance pa-
rameters, whereas the catheter length (any-
where from 10 to 40 cm), its internal diameter
(0.5 to 2.0 mm), as well as its distal holes
(number and distribution) influence the resis-
tance to blood flow. Poiseuille's law states that
the pressure decrease or internal resistance
within the catheter is primarily related to its
internal diameter and secondarily to its length,
determining blood flow. Catheter length and
lumen diameter are the two key characteristics
to consider for optimizing resistance to flow.
Effective Blood Flow Rate
For optimal function, a central venous dialysis
catheter should be able to deliver an extracor-
poreal blood flow consistently between 250
and 400 mL / min during intermittent sessions,
and ideally greater than 300 mL/min. This
requisite is especially important for end-stage
renal disease patients with catheters in place
for long-term use. Unfortunately, most thin
dual-lumen venous catheters cannot provide
consistently maximal blood flow rates greater
than 300 mL / min and this becomes a limiting
380 Leblanc et al
factor for any high-flux modality.22 Currently,
the maximal blood flow rates achieved with
venovenous central dialysis catheters (with an
acceptable venous resistance) reach 400 to 450
mL / min, as provided by the TwinCath. This
catheter, for which a large experience is avail-
able in Europe, uses two free floating single 10
Fr lumen catheters generally inserted within
the same vein and having two distinct subcuta-
neous tunnels of 8 to 10 cm. 2l
An interesting study by Athirakul et aP3
reported the maximal achievable blood flow
with three different cuffed permanent cath-
eters (in 30 treatments per catheter); blood
flow reached 396 mL / min with TwinCath, 384
mL/min with PermCath, and a significantly
lower mean value of 320 mL / min with VasCath
SoftCell (Bard Instrument Co, Toronto, On-
tario). The maximal blood flow rate was lim-
ited by the generated negative pressure (lower
than -300 mm Hg).
On the other hand, if the treatment option is
continuous or slow intermittent renal replace-
ment therapy, as often used in critcally ill acute
renal failure patients, a blood flow of 200 to
250 mL / min usually suffices, and most tempo-
rary catheters available on the market are
suited to deliver such flows. However, it should
be realized that the effective blood flow rate is
not indicated accurately by most dialysis ma-
chines, and, in general, an overestimation of
500T----------------------------------------------------,
400
300
200
100
o locity method (~). A total
Recorded Qb versus measured Qb (mUmin)
10% to 15% is expected, particularly at higher
flows. 22,24 In a recent survey of 27 patients
dialyzed in Montpellier and having the Twin-
Cath, various levels of blood flow rate as
indicated by the dialysis machines were com-
pared to the measured blood flow rates using a
noninvasive method relying on ultrasound
velocity dilution (Transonic Systems Inc, Ithaca,
NY, and Meditor, France).2s,26 A mean differ-
ence of 7.5 ± 9.0% was found in a total of 114
measurements. Overall, the discrepancies did
not appear significant from a clinical perspec-
tive, although they tended to be greater with
increasing blood flows (Fig 2).
Catheter Tip and Distal Holes
The position of the catheter tip may influence
the blood flow achieved. Nevertheless, correct
positioning of the tip into the distal innomi-
nate vein or proximal superior vena cava is
essential to minimize complications; the tip
should lie above the superior vena cava-right
atrial junction to avoid damage to the heart
(perforation of the right chambers, valvular
tear, tamponade). To reach this goal, the inser-
tion depth should be kept at 20 cm or less in
most adults. Although placing the catheter tip
lower, ie, into the right atrium, may reduce
malfunction and allow higher blood flow
rates,27 this practice is potentially dangerous
and not advocated. 28-30
400 368±30
Figure 2. Blood flow rate
indicated by the blood
pump (_) compared with
the blood flow rate mea-
sured by the ultrasound ve-
of 114 measures were per-
formed.
 Critical Care Nephrology: Catheter Dysfunction
Various catheter types vary significantly
when considering their distal end (Fig 3).
Numerous distal holes should better preserve
blood flow and prevent any partial luminal
obstruction from causing an extreme resis-
tance to blood flow. However, these lateral
holes can be occluded by a fibrin sleeve devel-
oping around the catheter. On theoretical
grounds, it then appears advantageous to
ensure two free-floating lumens within the
vessel for several reasons: the larger indi-
vidual lumens and the more numerous holes
can provide higher blood flows at lower resis-
tive pressures; the longitudinal distance be-
tween inflow and outflow sides can be rela-
tively longer than in dual-lumen single
catheters; and the lateral distance taking place
between the two catheters could potentially
reduce blood recirculation (as discussed be-
low).
Pressure Recordings
Pressures recorded by the dialysis machine on
both afferent and efferent blood sides provide
useful information and are also indicative of
catheter performance. 31 During hemodialysis,
the negative pressure generated by the blood
pump on the" arterial side" of the catheter to
ensure blood inflow into the extracorporeal
circuit, can potentially induce vein collapse,
especially in conditions of hypovolemia. Ide-
ally, the pressure on the arterial side should
not be more negative than 300 mmHg to
prevent vessel collapse with damage to its
.I
r_-_-_-_-_-_--_-_-_-_-_-_~_~~~r~~--------~~
• r--""'""
Dual·lumen
Double-catheter
C insertion. A few studies have now confirmed
.,..J ;J e ~t"
.......... '!.-+
I'
Double cathe ter
Figure 3. Different catheter types: (A) dual-lumen
with co-axial side holes and a distal hole; (B)
dual-lumen with distal holes; (C) twin or double
catheter with distal and side holes.
381
wall and possibly induction of hemolysis. At
blood flow rates of 350 to 400 mL/min, the
positive pressure on the venous or efferent
catheter side should be less than 250 mmHg.
The ratio of the venous pressure to the blood
flow rate should usually be around 1:2.32
Increased pressures on both sides (more nega-
tive arterial pressure and more positive ve-
nous pressure) indicate a greater resistance to
flow, strongly suggesting catheter dysfunc-
tion. For example, with the TwinCath, a signifi-
cant relationship is observed between achieved
blood flows and arterial and venous pressures
as monitored by the dialysis machine (Fig 4).
Blood Recirculation
Recirculation of dialyzed blood within the
catheter itself is another performance param-
eter to monitor since this phenomenon has a
negative impact on treatment efficiency. Unfor-
tunately, there is probably an obligatory mini-
mal blood recirculation rate in almost all dual-
lumen catheters (or double catheters when
inserted into the same central vein). To mini-
mize or abolish central catheter blood recircu-
lation, either two different catheters would
have to be inserted in two distant veins or one
each into an artery and a vein, as for continu-
ous arteriovenous modalities performed in
critically ill patients. Obviously, such options
are not suitable in day-to-day practice. One
should also remember that cardiopulmonary
recirculation does not occur with venovenous
catheters because, by definition, this phenom-
enon relates exclusively to arteriovenous ac-
cesses. 33
e The information on blood recirculation rates
with indwelling catheters is relatively recent.
Twardowski et ap4 reported a negligible recir-
culation of 2% in long-term well-functioning
B PermCath and VasCath (VasCath Inc, Missis-
sauga, Ontario) catheters implanted at the
upper thorax level. Blood recirculation was
shown to depend in part on the site of catheter
0
that femoral catheters, particularly the shorter
0 ones, are associated with higher blood recircu-
lation rates. 35-38 Kelber et ap5 showed in 17
patients having temporary catheters (Flexxi-
con, VasCath Inc, Mississauga, Ontario) recir-
culation rates of 4%, 2%, and 10% in internal
jugular, subclavian, and 24-cm femoral vein
382 Leblanc et al

Arterial Pressure (mmHg) Venous Pressure (mmHg)

y = -O.68x + 14.39 r2 = 0.97 y = O.83x - 98.99 r2 = = 0.92
o ~~-----~,..,....,..., 350------------~~

300
250
200
150
100
50
O+-~--~--~~--;
o 100 200 300 400 500 o 100 200 300 400 500
Blood Flow (mL/min) Blood Flow (mL/min)
Figure 4. Arterial and venous pressures as monitored by the dialysis machine in relation to blood flow rate for
well-functioning TwinCaths.

catheters at a set blood flow rate of 250
mL/min. Those rates did not significantly
increase when the blood flow was increased
up to 400 mL / min. However, 15-cm femoral
catheters had much higher recirculation rates
(18 :::!: 7% and 38 :::!: 14% at respective blood
flows of 250 and 400 mL / min).
In a study of 38 dual-lumen temporary
dialysis catheters (Quinton), recently inserted
for acute dialytic support (mean delay of 2
days), and well-functioning, we also found
much higher recirculation rates in short (13.5-
cm) femoral catheters when compared to
19.5-cm femoral or 13.5- and 19.5-cm subcla-
vian catheters (Table 1; Fig 5).36 There was no
relationship between recirculation rates and
arterial or venous resistances recorded during
dialysis.
Kraus et al also obtained similar results. 37
They studied various temporary catheter
brands (silicone and polyurethane) inserted in
femoral and subclavian veins for acute dialytic
support and reported that femoral catheters
recirculated significantly more at all blood

Table 1. Temporary Central Catheter Blood Recirculation Rates

Recirculation (% )
Qb Time AR VR
Mean SD SEM Range (mUmin) (h) (mmHg) (mmHg) URR(%) N
All 10.3 9.2 1.3 0-33.9 298::!:: 9 3.8::!:: 0.4 248::!:: 38 133::!:: 19 57.8::!:: 13.0 50
Femoral
All 16.1* 9.1 1.8 1.9-33.9 3.8 ::!::0.4 54.5::!:: 14.2 26
19.5 em 12.6 6.9 1.7 1.9-29.4 296::!:: 11 3.6 ::!:: 0.3 239::!:: 44 130::!:: 18 54.4::!:: 14.6 17
13.5 em 22.8t 9.1 3.0 9.3-33.9 300::!:: 6 3.9::!:: 0.2 259::!:: 24 136::!:: 13 54.8::!:: 14.3 9
Subclavian
All 4.1 3.6 0.7 0-12.9 3.9::!:: 0.4 62.5 ::!:: 10.9:\: 24
19.5 em 5.4 4.3 1.3 0-12.9 300::!:: 7 3.9::!:: 0.4 264::!:: 30 138::!:: 22 60.8::!:: 10.2 11
13.5 em 3.0 2.6 0.7 0-7.1 298::!:: 8 3.9 ::!:: 0.5 240::!:: 38 132::!:: 22 63.9::!:: 11.6 13
NOTE. AR, VR, Qb, and URR are expressed as means:!: SD.
Abbreviations: AR I VR, arterial and venous resistances; Qb, blood flow rate; URR, urea reduction ratio.
"Femoral versus subclavian catheter blood recirculations, P = .OOOL
t13.S cm versus 19.5 cm femoral catheter blood recirculations, P = .004.
:j:Subclavian versus femoral catheter urea reduction ratios, P = .03.
Reprinted with permission from Leblanc et al. 36
 Critical Care Nephrology: Catheter Dysfunction 383

12

10
...
I/)

.S!
Q) 8
.c
cou
6
'0
...
Q)
.D 4
E
:::l
Z
2

0
0-3.0 3.1-6.0 6.1-10.0 10.1-15.0 15.1-20.0 20.1-25.0 25.1-30.0 30.1-35.0

Catheter blood recirculation rate (%)

Figure 5. Distribution of recirculation rates (50 measurements) for subclavian (E21) and for femoral temporary
dialysis catheters (lwl) well functioning and recently inserted. Reprinted with permission from Leblanc et aP6

flows tested. Mean blood recirculation rates We recently evaluated 26 well-functioning

for femoral catheters were, respectively 12.8 ±
13.2%,24.5 ± 18.0%, and 24.7 ± 19.0% at blood
flows of 200, 250, and 300 mL/min, whereas
for subclavian catheters recirculation was only
4.2 ± 3.5% at a blood flow rate of 300 mL / min.
The explanation for this phenomenon may
be anatomic; short (13- to 15-cm) femoral
catheters probably do not reach the inferior
vena cava in most adult patients, and their tip
may thus remain in a smaller vein providing a
reduced bulk flow and a potential for higher
blood recirculation.
Finally, Athirakul et aP3 compared three
types of cuffed permanent catheters inserted
mainly in the right jugular vein, and reported
mean recirculation rates of 3.7% for the
PermCath, 3.9% for the TwinCath, and 4% for
the VasCath SoftCell.
It is well known that reversing the lines of
the catheter (using the venous port for inflow
and vice-versa) leads to much higher recircula-
tion rates and lower treatment efficacy34-36; it is
particularly important for the nursing staff to
be aware of this phenomenon.
Despite the useful information they pro-
vide, the above studies could be criticized
because of the method used for recirculation
measurement. All relied on blood-side solute
concentrations, which are not necessarily reli-
able. 39 Nonetheless, they all used a similar
sampling technique, ie, low or stop-flow two-
needle method, and their results are in excel-
lent agreement. 40-42
TwinCath catheters of variable residence time
(24 hours to 24 months) for blood recirculation
rates at increasing blood flow rates (200 to 400
mL/ min) using a different method. All these
cuffed and tunneled catheters were inserted
into the right internal jugular vein, except two
(one in the subclavian and one in the femoral
vein). A total of 110 measures were performed
with the ultrasound velocity dilution tech-
nique (Transonic Systems Inc, and Meditor),
thus avoiding the pitfalls of blood sampling.
The results are presented in Fig 6. A large
variability was observed in recirculation rates
at the different blood flows tested, and the
correlation between recirculation and blood
flow, although slightly positive, was rather
weak. Figure 7 is an example of the informa-
tion available on the bedside computer after
an injection of saline into the venous drip
chamber with this method.
Consequences of Catheter Dysfunction
Central venous catheter dysfunction will com-
promise dialysis efficiency and the overall
quality of renal replacement therapy- There-
fore, any significant dysfunction should be
detected and corrected as soon as possible to
prevent further therapy inadequacy and fur-
ther complications (mainly vessel damage).
The well-recognized stenosis and thrombosis
of subclavian veins in dialysis patients has
notably reduced the indications for subclavian
384 Leblanc et al

--
0~
40
35 y = 0,024x

-...
R2= 0,0279
Q)
(U 30

• •
-
c
.-
0
25
•
- 20 • • ••
('Q
:J
... • •
I•
0
15
• I
•
IIi
0
...
Q)
10

- I ••
"C
0 5
-m
0

0
0 50 100 150 200 250 300 350 400 450

Blood flow rate (mL/min)

Figure 6. Blood recirculation rates in TwinCath of various residence times at different blood flows, as
measured by the dilution ultrasound velocity method (110 measures).

catheter insertion,38,43 while internal jugular
cannulation, associated with a lesser incidence
of thrombosis, is gaining more and more in
popularity.44,45 In fact, subclavian dialysis cath-
eters should be avoided except for short-term
placement in acute situations not associated
with a need for long-term dialytic support. 46,47
If simple urea kinetic evaluation is used to
monitor dialysis adequacy, one has to be care-
ful for the patient with catheter dysfunction.

Dialyzer Blood Flow (Ob) =295 mUmln Recirculation = 10%

1.20
Volt. .90
.60
.30
.oo+--____ L_C.~~ _ _ __ _ _ _ _ _ __ Figure 7. Example of the
information available on
the bedside computer when
-10 .00 10. 20. 30. 40. evaluating access recircula-
Time (sec)
tion with the Transonic He-
modialysis Monitor (Tran-
M.REe 3 19:06 Ob = 295 mUmln Recirculation =10% sonic Systems Inc, Ithaca,
M.REe 2 19:05 Ob = 337 mUmln
= 376 mUmln
Recirculation =13% NY, and Meditor, France).
M.REe 1 19:04 Ob Recirculation = 19%
 Critical Care Nephrology: Catheter Dysfunction
The urea reduction ratio (URR) can be mislead-
ing, being falsely higher with blood recircula-
tion, and if the sample is taken immediately
post-session. In our previous study, however,
URR was lower at high blood recirculation
probably because there was a slight delay in
urea sampling after the session, thus allowing
the recirculation effect to dissipate. 36
For the same reason, using a urea sample
drawn immediately post-dialysis to calculate
Kt/V may overestimate the effective dialysis
dose in patients with significant blood recircu-
lation. More precise and sophisticated measure-
ments of blood recirculation and double-pool
Kt/V calculations may help to detect access-
related problems. As a broad guideline, the
tolerated blood recirculation rate within cen-
tral venous dialysis catheter should be less
than 10% to 15% in most settings, and ideally
between 3% to 7%, especially in situations
mandating an extremely efficient dialysis ses-
sion (low dialysis dose, high catabolism, intoxi-
cations).
As a permanent access for patients without
a functioning peripheral access, a good com-
promise has been to use the TwinCath, prefer-
ably implanted into the right internal jugular
vein, and to keep a length difference of 3 to 5
cm between both tips (Fig 8).21,48 During the
last decade, 1,176 permanent TwinCath cath-
eters have been implanted in 997 chronic renal
Figure 8. TwinCath inserted into the right internal
jugular vein. Note the distance between the two
catheters at the distal end and the subcutaneous
tunnel.
385
failure patients with a mean duration of use of
218 days (ranging from 30 days to 8 years). A
total of 620 patients have been on high-
efficiency dialytic modalities with such cath-
eters for more than 3 months with satisfactory
results in terms of dialysis adequacy. The
actuarial TwinCath survival (unassisted pa-
tency) for this subgroup is shown in Figure 9.
In another series of 64 PermCath catheters in
51 patients, the mean duration of use was
316 ± 337 days, leading to actuarial 1-, 2-, 3-,
and 4-year survival rates of 74%, 43%, 25%,
and 12%,17 Another survey of 210 patients also
using PermCath catheters reported mean cath-
eter longevity rates between 12 and 85 weeks. 49
However, in this last study, flow and clotting
problems were frequent with respective inci-
dences of 46% and 24%.
Diagnostic Approach and
Therapeutic Options
Catheter dysfunction is manifested clinically
when the desired blood flows cannot be
reached or when abnormal venous and/ or
arterial resistance pressures are indicated on
the dialysis machine. A complete catheter oc-
clusion will generally be detected by the inabil-
ity to aspirate blood from the catheter port(s)
with a syringe as performed routinely at the
beginning of the session. An attempt at flush-
ing an obstructed lumen might be helpful but
is not without risk and should not be a routine
procedure. One has to be aware of the small
risk for clot embolization into the vasculature
and the lung.l o However, clinical experience
shows that the consequences of this manipula-
tion are usually minimal. 50
A catheter obstruction or a complete occlu-
sion may be confirmed by local intracatheter
instillation of contrast media (catheterogram)
(Fig 10). To evaluate the host or proximal/
distal veins for possible thrombosis and/or
stenosis, a phlebography is the most accurate
diagnostic test, whereas Doppler ultrasonogra-
phy and 1311 fibrinogen scans are useful but
less reliable alternatives. 51 If catheter occlusion
is diagnosed, intracatheter instillation of a
fibrinolytic agent with a sufficient dwelling
time will restore patency in the majority of
cases.9,52-55 Urokinase is most widely used and
has been associated with relatively good re-
386 Leblanc et al

Permanent TwinCath (used 3 months or more)

Actuarial Survival (%)

100 N =620
80

60

40

20 '.~''''''
.'~~~ ....
, ..
.
".
"

o L,------.---~--,_~--_.------.,.-'~-.->~-~-~)~-=--=--=-:--~ Figure 9. Actuarial sur-

o 20 40 60 80 100 120 vival of 620 TwinCaths in-
serted for 3 months or
Time (months) longer.

sults. Tissue plasminogen activator has also
been used successfully. Various proposed pro-
tocols are presented in Table 2. We recommend
instillating at least 5,000 IV of urokinase in
each catheter lumen with a volume sufficient
to fill the lumen completely (to the tip) and
leave it in situ 1 hour. After flow restoration
with a thrombolytic agent, many centers now
advocate low dose oral anticoagulants to main-
tain catheter patency.9
If subclavian stenosis is found, percutane-
ous transvenous angioplasty of the vessel with
or without fibrinolytic administration and with
or without stent placement can be performed
with an encouraging initial success. However,
patency rates decrease over time because of
restenosis. 56,57 The management of central vein
thrombosis should include the removal of the
catheter. In anecdotal cases, this has led to a
spontaneous resolution of the thrombus. Fi-
nally, surgical thrombectomy may be neces-
sary in cases of failure of less aggressive
approaches.
Pulmonary emboli and obstruction to blood
return to the right side of the heart are the
most lethal or feared complications; thus, pre-

Figure 10. Contrast media

injected into a TwinCath
(catheterogram). The lu-
men on the left is patent
while a distal occlusion is
seen on the right.
 Critical Care Nephrology: Catheter Dysfunction
Table 2. Thrombolytic Agents to Reopen Clotted Hemodialysis Catheters
Investigator Agent
Athirakul and Schwab54 Urokinase 5,000 IU /mL
Shrivastara et al53 Urokinase 5,000 IU / mL
(can be diluted 1:2
with saline)
Glynn et al63 Urokinase diluted in
saline
Lawson et al55 Urokinase 5,000 IU / mL
Powelson JA (personal Urokinase 5,000 IU /mL
communication)
Paulsen et al52 TIssue plasminogen acti-
vator 2 mg/2 cm3
vention and early detection are essential. If
suspected, transesophagal ultrasonography is
useful to document a thrombus extending into
the right atrium. Because it prevents an exten-
sion of a subclavian, internal jugular, innomi-
nate, or superior vena cava thrombus, sys-
temic anticoagulation is recommended for a
period of 6 weeks to 3 months.
In the absence of vessel thrombus and when
the catheter lumen appears relatively patent, it
seems reasonable to replace a dysfunctional
catheter over a guidewire, although this prac-
tice is controversial from an infectious compli-
cation standpoint because it may increase the
risk of bloodstream infection. 58 One has to
measure this inherent risk against the poten-
tial complications resulting from catheter inser-
tion at a different site.
Prevention of Catheter Dysfunction
Prevention of catheter malfunction involves
considering the characteristics of the catheter
chosen, the site and technique of insertion, the
residence time of the catheter, and the fre-
quency of previous catheterization( s) at a given
site. Several precautions taken at the time of
implantation prevent some mechanical prob-
387
Proposed Regimen
For PermCath: 1 mL per branch plus 0.2 or 0.3 mL
of saline in proximal or distal ports, respectively.
Aspirate after 30 minutes; 0.3 mL of saline may
be instilled every 10 minutes (X3).
Inject an adequate filling volume in each line for 1
hour; if still occluded, a similar dose in both
ports can be injected and left in situ for 24 hours.
2.5 mL in occluded branch left for 3 hours then
aspirate and flush the line with heparin.
0.2 to 0.4 mL or volume sufficient to fill catheter
lumen. Attempt to aspirate every 5 minutes
X 15 minutes. If not patent after 1 hour, a second
similar instillation is recommended.
1.2 mL in each port left in situ for 20 minutes. If not
successful, systemic urokinase can be consid-
ered: 125,000 IU (one 250,000 IU vial is reconsti-
tuted with 5 mL sterile water and its total con-
tent is added to 150 mL of D5W) infused over 5
minutes in each port (250,000 IU over 10 min-
utes). Pretreat with Benadryl and continue with
intravenous heparin.
For PermCath: volume sufficient to fill the lumen.
Dwell time over 60 minutes or up to next
dialysis.
lems (avoiding catheter bending or kinking,
appropriate suturing, verifying immediate
flows, flushing the lumen[s] with saline fol-
lowed by heparin immediately after vessel
insertion, localizing of the catheter tip and
position on x-ray). The creation of a subcutane-
ous tunnel allows a better stability of the
catheter for longer use and is more comfort-
able for patients; indeed tunneled catheters
have been called "permanent."17,48,49,59 The rela-
tive decrease in infectious risk attributed to
subcutaneous tunnels remains controversial
and is probably more in relation to the inser-
tion technique and catheter care. 9,60,61
Care of the catheter after its insertion is of
utmost importance, particularly with regard to
infectious complications. 62 Dialysis catheters
must be reserved exclusively for hemodialysis
sessions and the exit site covered after the
procedure. To maintain patency between treat-
ments, each lumen should be flushed with
saline and heparin should be left in situ. The
heparin concentration is anywhere between
5,000 to 10,000 IU / mL and the amount to
infuse should be determined carefully to fill
each catheter lumen totally and to reach its
distal end. The local heparin is usually re-
388 Leblanc et al
moved before tubing connection at the next
treatment.
References
1. Hoshal VL, Ause RG, Hoskins PA: Fibrin sleeve
formation on indwelling subclavian central venous
catheters. Arch Surg 102:353-358,1971
2. Madden B, Huissoon Ap, Colhoun EN, et al: Superior
vena caval obstruction: A rare complication of Quin-
ton double-lumen haemodialysis catheters. Nephrol
Dial TranspI4:586-587, 1989
3. Domoto DT, Garvin PJ: Recurrent venous thrombosis
associated with permanent internal jugular vein hemo-
dialysis catheters. Trans Am Soc Artif Intern Organs
33:140-143,1987
4. Trottier SJ, Veremakis C, O'Brien J, et al: Femoral deep
vein thrombosis associated with central venous cath-
eterization: Results from a prospective, randomized
trial. Crit Care Med 23:52-59,1995
5. Clark DD, Albina JE, Chazan JA: Subclavian vein
stenosis and thrombosis: A potential serious complica-
tion in chronic hemodialysis patients. Am J Kidney
Dis 15:265-268, 1990
6. Spinowitz BS, Galler M, Golden RA, et al: Subclavian
vein stenosis as a complication of subclavian catheter-
ization for hemodialysis. Arch Intern Med 147:305-
307,1987
7. Paganini EP, Pudelski B, Bednarz D: Dialysis delivery
in the ICU-Are patients receiving the prescribed
dialysis dose? J Am Soc NephroI3:384, 1992 (abstr)
8. Rahn NH, Rubin E, Koehler RE: Thrombophlebitis of
the internal jugular vein: Noninvasive imaging. South
Med J 77:1308-1312, 1984
9. Bander SJ, Schwab SJ: Central venous angioaccess for
hemodialysis and its complications. Semin Dial 5:121-
128,1992
10. MacLean-Ross AH, Griffith CDM, Anderson JR, et al:
Thromboembolic complications with silicone elas-
tomer subclavian catheters. J Parenter Enteral Nutr
6:61-63,1982
11. Curelaru I, Gustavsson B, Hansson AH, et al: Material
thrombogenicity in central venous catheterization. II.
A comparison between plain silicone elastomer and
plain polyethylene, long, antebrachial catheters. Acta
Anaesthesiol Scand 27:158-164,1983
12. Eberhart RC, Clagett CP: Catheter coatings, blood
flow, and biocompatibility. Semin Hematol 28:42-48,
1991
13. Sioshansi P: New processes for surface treatment of
catheters. Artif Organs 18:266-271, 1994
14. Nichols AB, Owen J, Grossman B, et al: Effect of
heparin bonding on catheter-induced fibrin formation
and platelet activation. Circulation 70:843-850, 1984
15. Krafte-Jacobs B, Sivit CJ, Mejia R, et al: Catheter-
related thrombosis in critically ill children: Compari-
son of catheters with and without heparin bonding. J
Pediatr 126:50-54,1995
16. Vanholder R, Ringoir 5: Vascular access for hemodialy-
sis. Artif Organs 18:263-265, 1994
17. Gibson Sp, Mosquera D: Five years experience with
the Quinton Permcath for vascular access. Nephrol
Dial Transplant 6:269-274, 1991
18. Canaud B, Leray-Moragues H, Garred LJ, et al: Perma-
nent central vein access. Semin Dial 9:397-400, 1996
19. Uldall PR, DeBruyne M, Besley M, et al: A new
vascular access catheter for haemodialysis. Am J
Kidney Dis 3:270-277,1993
20. Mustata 5, Less P, Agraharar M, et al: Further experi-
ence with a percutaneously inserted double-lumen
Silastic catheter for end-stage renal failure patients
with refractory vascular access problems. ASAIO J
42:244-245,1996
21. Canaud B, Beraud JJ, Joyeux H, et al: Internal jugular
vein cannulation with two silicone rubber catheters: A
new and safe temporary vascular access for hemodialy-
sis. Thirty month's experience. Artif Organs 10:397-
403, 1986
22. Collins AJ: High-flux, high-efficiency procedures, in
Henrich WL (ed): Principles and Practice of Dialysis.
Baltimore, MD, Williams & Wilkins, 1994, pp 76-88
23. Athirakul K, Conlon P, Schwab 5: Cuffed central
venous hemodialysis catheters and adequacy of dialy-
sis. J Am Soc NephroI7:1402-1403, 1996
24. Schmidt DF, Schniepp BJ, Kurtz SB, et al: Inaccurate
blood flow rate during rapid hemodialysis. Am J
Kidney Dis 17:34-37, 1991
25. Krivitski NM: Theory and validation of access flow
measurement by dilution technique during hemodialy-
sis. Kidney Int 48:244-250, 1995
26. Krivitski NM: Novel method to measure access flow
during hemodialysis by ultrasound velocity dilution
technique. ASAIO J 41:M741-M745, 1995
27. Feldman HI, Kobrin 5, Wasserstein A: Hemodialysis
vascular access morbidity. J Am Soc Nephrol 7:523-
535, 1996
28. Defalque RJ, Campbell C: Cardiac tamponade from
central venous catheter. Anesthesiology 50:249-252,
1979
29. Food and Drug Administration: Precautions neces-
sary with central venous catheters. Washington, DC,
US Government Printing Office, 1989, p 15
30. McGee WT, Ackerman BL, Rouben LR, et al: Accurate
placement of central venous catheters: A prospective,
randomized, multicenter trial. Crit Care Med 21:1118-
1123,1993
31. Stroud CC, Meyer SL, Bawkon MC, et al: Vascular
access for extracorporeal circulation. Resistance in
double lumen cannulas. ASAIO J 37:M418-M419, 1991
32. Raja MR: Vascular access for hemodialysis, in Ing TS,
Daugirdas JT, (eds): Handbook of Dialysis. Boston,
MA, Little, Brown, 1994, pp 53-77
33. Schneditz D, Kaufman AM, Polaschegg H, et al:
Cardiopulmonary recirculation during hemodialysis.
Kidney Int 42: 1450-1456, 1992
34. Twardowski ZJ, Van Stone Jc, Jones ME, et al: Blood
recirculation in intravenous catheters for hemodialy-
sis. J Am Soc NephroI3:1978-1981, 1993
35. Kelber J, Delmez JA, Windus DW: Factors affecting
delivery of high-efficiency dialysis using temporary
vascular access. Am J Kidney Dis 22:24-29, 1993
36. Leblanc M, Fedak 5, Mokris G, et al: Blood recircula-
 Critical Care Nephrology: Catheter Dysfunction
tion in temporary central catheters for acute hemodi-
alysis. Clin NephroI45:315-319, 1996
37. Kraus M, McMcusky M, Clark W: Temporary hemodi-
alysis catheter recirculation studies: A comparison of
design and site selection. Blood Purif 13:390, 1995
(abstr)
38. Uldall R: Hemodialysis access-Part A: Temporary, in
Jacobs C, Kjellstrand CM, Koch KM, et al (eds):
Replacement of Renal Function by Dialysis. Kluwer
Academic Publishers, Dordrecht, 1996, pp 277-292
39. Hester RL, Curry E, Bower J: The determination of
hemodialysis blood recirculation using blood urea
nitrogen measurements. Am J Kidney Dis 6:598-602,
1992
40. Pederson JA, Dunlay R, Williams C, et al: Two-needle
calculation of recirculation compared with the stan-
dard three-needle method. Clin Nephrol 33:203-206,
1990
41. Sherman RA: The measurement of dialysis access
recirculation. Am J Kidney Dis 22:616-621,1993
42. Sherman RA, Matera JJ, Novik L, et al: Recirculation
reassessed: The impact of blood flow rate and the
low-flow method reevaluated. Am J Kidney Dis 23:846-
848,1994
43. Cimochowski GE, Worley E, Rutherford WE, et al:
Superiority of the internal jugular over the subclavian
access for temporary dialysis. Nephron 54:154-161,
1990
44. Schillinger F, Schillinger D, Montagnac R, et al: Post-
catheterisation vein stenosis in hemodialysis: Com-
parative angiographic study of 50 subclavian and 50
internal jugular access. Nephrol Dial Transplant 6:722-
724,1991
45. Agraharkar M, Isaacson S, Mendelssohn D, et al:
Percutaneously inserted silastic jugular hemodialysis
catheters seldom cause jugular vein thrombosis.
ASAIO J 41:169-172, 1995
46. De Moor B, Vanholder R, Ringoir S: Subclavian vein
hemodialysis catheters: Advantages and disadvan-
tages. Artif Organs 18:293-297, 1994
47. Konner K: Subclavian haemodialysis access: Is it still
justified in 1995? Nephrol Dial Transplant 10:1988-
1991,1995
48. Canaud B, Cristol JP, Klouche K, et al: Silastic twin
catheters: A flexible solution for temporary or long-
term vascular access problems in hemodialysis pa-
tients, in Man NK, Botella J, Zucchelli P (eds): Blood
Purification in Perspective. Cleveland, OH, ICAOT
Press, 1992, pp 25-29
49. Dunea G, Domenico L, Gunnerson P, et al: A survey of
permanent double lumen catheters in hemodialysis
patients. ASAIO Trans 37:M276-277, 1991
389
50. Anderson AJ, Krasnow SH, Boyer MW, et al: Hickman
catheter clots: A common occurrence despite daily
heparin flushing. Cancer Treat Rep 71 :651-653, 1987
51. Caruana RJ, Raja RM, Zeit RM, et al: Thrombotic
complications of indwelling central catheters used for
chronic hemodialysis. Am J Kidney Dis 6:497-501,
1987
52. Paulsen D, Reisoether A, Aasen M, et al: Use of tissue
plasminogen activator for reopening of clotted dialy-
sis catheters. Nephron 64:468-470,1993
53. Shrivastava D, Lundin Ap, Dosunmu B, et al: Salvage
of clotted jugular vein hemodialysis catheters. Neph-
ron 68:77-79, 1994
54. Athirakul K, Schwab SJ: Opening thrombosed silastic
hemodialysis catheters. Semin Dial 9:289, 1996
55. Lawson M, Bottino JC, Hurtubise MR, et al: The use of
urokinase to restore the patency of occluded central
venous catheters. Am J Intr Ther Clin Nutr 1:29-32,
1982
56. Beathard GA: Percutaneous transvenous angioplasty
in the treatment of vascular access stenosis. Kidney Int
42:1390-1397,1992
57. Newman GE, Saeed M, Himmelstein S, et al: Total
central vein obstruction. Resolution with angioplasty
and fibrinolysis. Kidney Int 39:761-764,1991
58. Cobb DK, High KP, Sawyer RG, et al: A controlled trial
of scheduled replacement of central venous and pul-
monary-artery catheters. N Engl J Med 327:1062-1068,
1992
59. Kinnaert P, Hooghe L, De Pauw L, et al: Use of the
Hickman catheter as permanent vascular access for
hemodialysis. Trans Am Soc Artif Intern Organs 36:104-
106,1990
60. Moss AH, Vasilakis C, Holley JL, et al: Use of a silicone
dual-lumen catheter with a dacron cuff as a long-term
vascular access for hemodialysis patients. Am J Kid-
ney Dis 16:211-215, 1990
61. De Meester J, Vanholder R, De Roose J, et al: Factors
and complications affecting catheter and technique
survival with permanent single-lumen dialysis cath-
eters. Nephrol Dial Transplant 9:678-683, 1994
62. Vanherweghem JL, Dhaene M, Goldman M, et al:
Infections associated with subclavian dialysis cath-
eters: The key role of nurse training. Nephron 42:116-
119,1986
63. Glynn MFX, Langer B, Jeejeebhoy KN: Therapy for
thrombotic occlusion of long-term intravenous alimen-
tation catheters. J Parenter Enteral Nutr 4:387-390,
1980