Nephrol Dial Transplant (1998) 13: 978–981
Nephrology
Dialysis
Transplantation
Technical Note
The use of ultrasound for the placement of dialysis catheters
Conradin Nadig, Michael Leidig, Thomas Schmiedeke and Bernd Höffken
Department of Medicine IV of the University of Erlangen-Nuremberg, Germany
Abstract
Background. The jugular vein should be preferred to
the subclavian vein for the placement of dialysis catheters, since subclavian catheters result in a high incidence (up to 50%) of subclavian vein thromboses and
stenoses.
Method. We conducted a prospective, randomized
study between July 1996 and March 1997 to find out
whether through the use of ultrasound, the rate of
unsuccessful attempts in puncturing the internal jugular
vein could be reduced. Seventy-three internal jugular
vein cannulations were performed on 65 patients, using
the guide-wire technique (according to Seldinger). Two
groups were formed randomly by lot: in the first group
the position of the internal jugular vein was marked
on the skin by the use of ultrasound (Picker CS9100,
Convex 3.5 MHz) before disinfection and local anaesthesia took place. The puncture was performed according to this mark. In the second group, the internal
jugular vein was cannulated with real-time ultrasound
guidance on the monitor. Any withdrawal of the needle
with a consecutive forward movement was judged as
an unsuccessful attempt, whether or not a second skin
puncture was performed.
Result. Thirty-seven punctures of the internal jugular
vein with a skin mark determined by ultrasound yielded
87 unsuccessful attempts. Thirty-six punctures with
real-time ultrasound guidance resulted in 10 unsuccessful attempts (P<0.01). The time from the beginning
of the local anaesthesia to successful puncture was
4.8±2.2 min in the first group compared to
3.4±0.9 min in the second group (P<0.01). The crosssection of the internal jugular vein in the first group
was 1.7±0.8 cm2 versus 1.5±0.8 cm2 in the second
group (not significant). Neither of the two methods
caused any complications.
Conclusion. The puncture of the internal jugular vein
with real-time ultrasound guidance resulted in significantly fewer unsuccessful attempts of venepuncture
without requiring additional time.
Key words: central venous catheterization; jugular vein;
ultrasound
Correspondence and offprint requests to: Dr Conradin Nadig, alte
Landstrasse 30, CH-8803 Ruschlikon, Switzerland.
Introduction
Subclavian catheters result in a high incidence (up to
50%) of subclavian vein thromboses and stenoses
[1–4]. Jugular vein stenosis after catheterization is less
likely than subclavian vein stenosis, and even if it
occurs, it is often without clinical significance [1,2].
Therefore, the jugular vein should be preferred for
central venous catheterization [1–3]. Fewer difficulties
occur when using the right rather than the left jugular
vein because of the straighter course into the superior
vena cava and the right atrium. The external jugular
vein often offers itself for puncture, but it might be
troublesome to get the guide wire past the junction of
the external jugular vein to the brachiocephalic vein
and, in some cases, it is not possible to overcome this
obstacle even under fluoroscopic guidance. As a result
of these considerations, we picked the right internal
jugular vein as the first choice for central venous
catheterization. The internal jugular vein can easily be
found by ultrasound.
Subjects and methods
A prospective, randomized study was carried out between
July 1996 and March 1997 at the Department of Medicine
IV of the University of Erlangen-Nuremberg, Germany. The
following factors were evaluated: the number of unsuccessful
attempts of venepuncture was counted, defined as any withdrawal of the needle with a consecutive forward movement,
whether or not a second skin puncture was performed; the
time from the beginning of the local anaesthesia until successful venepuncture with the 18-gauge needle for introducing
the guide wire was recorded; the cross-section of the internal
jugular vein at the puncture site was measured (venepuncture
was not performed if the diameter of the vein was less than
5 mm); the compliance of the patient ( Valsalva manoeuvre,
keeping still during the procedure) was assessed (0 to 3, 3=
very good). A chest film was obtained after the whole
procedure and was evaluated for the presence of misplacement of the catheter, pneumothorax, haemothorax and mediastinal haematoma. The clinical experience of the five
physicians inserting the catheters ranged from 1 to 7 years.
The catheters were inserted under controlled, non-emergency
conditions.
© 1998 European Renal Association–European Dialysis and Transplant Association
Ultrasound for the placement of dialysis catheters
Preparation of the patient
Routine informed consent for central vein cannulation was
obtained from the patient. The patient was placed in a supine
Trendelenburg’s position, which was not changed until successful venepuncture was achieved. The face was turned away
from the site of the venepuncture. It was explained to the
patient that it was very important not to move the head
during the whole procedure. The Valsalva manoeuvre was
practiced with the patient.
Preparation of the ultrasound probe
The exact middle of the ultrasound probe (Convex 3.5 MHz;
Picker CS9100) was marked on both sides with a waterresistant felt-tip pen (Edding 3300 permanent marker) using
a ruler.
Application of the skin mark
The ultrasound probe was moistened with a few drops of
alcoholic disinfection fluid, which was enough to enable the
ultrasound waves to penetrate into the tissue. The probe was
positioned at a right-angle to the internal jugular vein just
proximal to the clavicle. The compressibility was tested to
exclude venous thrombosis. After having positioned the
internal jugular vein exactly in the middle of the ultrasound
screen, a temporary mark was applied on the skin with a
pen according to the lateral mark in the middle of the probe.
This procedure was repeated twice at a distance of about
2 cm cranial, which yielded three temporary marks on the
skin showing the course of the internal jugular vein. After
the skin had dried, the temporary marks were connected
with a black line using the felt-tip pen. The site of the
venepuncture was determined by drawing a short transverse
mark crossing the line at a distance of 2 cm proximal to
the clavicle.
Puncture of the internal jugular vein with a skin mark
determined by ultrasound
979
orientation, it was helpful to identify the common carotid
artery, which normally lies more medial than the internal
jugular vein. With the other hand, local anaesthesia was
injected (22-gauge needle) according to the lateral mark in
the middle of the probe, which was visible through the sterile
plastic glove. Local anaesthesia was injected at a distance of
5 mm from the probe, and at a 70° angle (Figure 1). The
forward movement of the needle in a 70° angle from cranial
towards the internal jugular vein and the actual puncture of
the vein were followed on the ultrasound screen. The venepuncture was performed with the 22-gauge and the 18-gauge
needle. A nurse held the cable of the probe to prevent nonsterile contact with the site of puncture.
Statistical analysis
Statistical significance was determined according to the chisquare test. A P value less than 0.05 was considered to
indicate statistical significance.
Results
Seventy-three internal jugular vein cannulations were
performed on 65 patients. Puncture of the internal
jugular vein with real-time ultrasound guidance showed
significantly fewer unsuccessful attempts, and the
number of punctures without unsuccessful attempts
was significantly higher. Moreover, the procedure with
real-time ultrasound guidance required less time
( Table 1).
The venepuncture with a skin mark determined by
ultrasound was without success in 13 patients with a
total of 53 unsuccessful attempts at puncture (there
was a minimum of three attempts at puncture per
patient, maximum seven). In these patients, a puncture
with real-time ultrasound guidance followed, per-
After disinfection and draping the area with a sterile cloth,
a local anaesthesia was performed by injecting 5 ml
Scandicaine solution (1%) with a 22-gauge needle according
to the skin mark. The pulsation of the common carotid
artery was not used as a point of reference. The patient was
asked to perform the Valsalva manoeuvre. The internal
jugular vein was punctured from cranial with the 22-gauge
needle ( length 3.81 cm) according to the course of the vein
as marked on the skin, and at a 45° angle, to confirm access
to the vein before use of the 18-gauge needle. The return of
venous blood into a syringe attached to the needle showed
entry into the vein. In an analogous manner, venepuncture
with an 18-gauge needle ( length 7 cm) was performed.
Unsuccessful attempts were counted for venepuncture with
both the 18-gauge as well as with the 22-gauge needle.
Puncture of the internal jugular vein with real-time
ultrasound guidance
The ultrasound probe was placed in a sterile plastic glove
with 50 ml ultrasonic gel. The site of venepuncture was
moistened with a disinfecting spray. With one hand, the
probe was positioned at a right-angle to the internal jugular
vein just proximal to the clavicle and the vein was thus
shown in the middle of the ultrasound screen. For better
Fig. 1. Hands of the operators in relation to the probe and the needle
at the time of venepuncture by applying real-time ultrasound guidance. The vein was shown in the middle of the ultrasound screen.
Then the puncture was performed at a 70° angle according to the
mark in the middle of the probe, which is visible through the sterile
plastic glove. The skin mark applied before disinfection and local
anaesthesia shows the position of the internal jugular vein. In the
case of this patient, the puncture of the internal jugular vein with
skin mark determined by ultrasound would have been performed at
the same place as the puncture with real-time ultrasound guidance.
980
C. Nadig et al.
Table 1. Comparison of puncture of the internal jugular vein with a skin mark determined by ultrasound versus puncture with real-time
ultrasound guidance
Variable
IJV puncture with
skin mark determined
by ultrasound (n=37)
IJV puncture with
real-time ultrasound
guidance (n=36)
P
Unsuccessful attempts, total
Puncture without unsuccessful attempts
Time anaesthesia-puncture (min)
IJV cross-section (cm2)
Patient compliance 0–3 (3=v. good)
Age (years)
87
13 (35%)
4.8±2.2
1.7±0.8
2.0±1.1
66±12
10
30 (83%)
3.4±0.9
1.5±0.8
2.4±0.9
69±11
<0.01
<0.01
<0.01
NS
NS
NS
NS, not significant; IJV, internal jugular vein.
formed by the same physician. In 10 of these 13
patients (77%), the puncture with real-time ultrasound
guidance was achieved without unsuccessful attempts,
in each of the remaining three patients, two unsuccessful attempts were noted. In 10 of these 13 cases, the
skin mark had been displaced by 0.5–1.2 cm because
the patient had moved, and the mark therefore no
longer corresponded to the position of the vein. In two
cases the internal jugular vein was deeply positioned
and could not be reached with the 3.81 cm long 22gauge needle at an angle of 45°. In another case, the
skin mark was displaced by 0.8 cm and, in addition,
the internal jugular vein could only be punctured with
a longer needle because of its deep position.
Sixty-one catheters (84%) were inserted in the right
internal jugular vein, 12 catheters (16%) in the left
one. The reason for choosing the left side was an
already existing catheter on the right side in four cases,
a diameter of the internal jugular vein of less than
5 mm on the right side in three cases, a removed
catheter as a result of an infection on the right side in
two cases, a thrombosis of the right internal jugular
vein in one case and of the right brachiocephalic vein
in another case and finally a stent positioned in the
right brachiocephalic vein in the last case (Figure 2).
Punctures performed on high risk patients and on
patients with no compliance
In each group one patient suffered from thrombocytopenia (platelet count 29 000 and 33 000/ml respectively).
One patient was punctured with real-time ultrasound
guidance with a prolongation of the prothrombin and
partial thromboplastin times (PT 36%, normal >60%;
PTT 56 s, normal <40 s).
The patient compliance scored 0 in 11 patients; six
of these patients were punctured with a skin mark
determined by ultrasound.
Fig. 2A,B. (A) shows a small diameter of 2.5 mm of the right
internal jugular vein (I.J.V.), whereas the diameters on the left side
(B) measure 15×11 mm, suitable for catheterization. The carotid
artery (C.A.) can be seen next to the vein.
chance respectively 54 and 93 days after the insertion
of the catheter.
Discussion
Complications
Neither of the two methods caused any complications
related to the venepuncture. In two patients, a thrombosis of the internal jugular vein was discovered by
A significant improvement of the venepuncture success
could be demonstrated by applying real-time ultrasound guidance. This procedure causes fewer complications because of the smaller number of needle passes
Ultrasound for the placement of dialysis catheters
and it also reduces the discomfort for the patient [5–8].
The time it takes to wrap the transducer in a sterile
sheath is compensated by a more rapid localization of
the vein [6,7,9,10]. By using ultrasound, anomalous
venous anatomy such as a small diameter and a deep
position of the internal jugular vein or a thrombosis
can be recognized early [5,10–12]. The diameter of the
internal jugular vein is not reduced through palpation
of the pulsation of the common carotid artery, since
palpation is not performed. In our experience the main
reason for unsuccessful punctures with a skin mark
determined by ultrasound is a displacement of the skin
mark due to movements of the patient’s head.
In the study of Denys and colleagues [9], first
attempt success for internal jugular vein cannulation
with real-time ultrasound guidance reached 78%, and
in the study of Gallieni and Cozzolino [5], the internal
jugular vein was accessed with real-time ultrasound
guidance on the first attempt 87% of the time. Although
the benefit associated with real-time ultrasound guidance for internal jugular vein puncture has been shown,
this method is too rarely used in nephrology. It might
be that a needle guide strapped onto the ultrasound
probe is too complicated to be widely used, or that
there are problems with practical details of the procedure. For this reason, we deliberately used a simple
method and put considerable effort into describing it
carefully. The results in the study of Denys et al. [9]
indicated a very short and steep learning curve for
venepuncture with real-time ultrasound guidance.
Gallieni and Cozzolino [5] pointed out that for a
relatively inexperienced operator, cannulation of the
internal jugular vein was greatly facilitated by using
ultrasound guidance and that this method was easy to
learn. We, too, believe, that ultrasound guidance is
beneficial precisely for operators with little experience.
In our department the time required for the venepuncture of the internal jugular vein has been drastically
reduced with the availability of the ultrasound device.
We did not include a comparison between venepunctures without ultrasound and punctures with ultrasound because of already existing studies and due to
the patient’s discomfort [5–7,10].
The 7.3-cm wide convex probe used in the study is
also suitable for abdominal ultrasonography. We did
not notice a difference in the handling of other types
of ultrasound probes, and we believe that the shape of
the probe is not of great importance. The success rates
in venepuncture can probably be improved if an experienced operator uses a probe with a groove for the
advancing needle. However, these probes are rarely
available in the hospital and are more complicated
to handle.
We used a disinfectant spray to moisten the skin to
enable the penetration of the ultrasound waves into
the tissue. We found a spray practical, and a further
disinfection of the skin took place. However, compared
981
to saline, the evaporation of the alcoholic solution is
faster, and the disinfectant spray had to be reapplied.
As to the image resolution we found no difference
between using a disinfectant spray or saline.
While this study was being conducted a patient
developed a severe bleeding due to arterial puncture
during an attempt of central venous catheterization.
The puncture was performed by an operator who was
not taking part in the study and who did not use
ultrasound. As a result of this event the methods for
venepuncture with the use of ultrasound as described
above became quickly known in our clinic. Experienced
operators who had not used ultrasound up to that time
told us that they were astonished at the success of using
ultrasound for venepuncture. Referring to previous
studies, we believe that the puncture of the internal
jugular vein with real-time ultrasound guidance on the
monitor is the gold standard for high success rates in
central venous catheterization [5–7,9,13,14].
References
1. Schillinger F, Schillinger D, Montagnac R et al. Postcatheterization vein stenosis in hemodialysis: comparative radiographic study of 50 subclavian and 50 internal jugular accesses.
Nephrol Dial Transplant 1991; 6: 722–724
2. Agraharkar M, Isaacson S, Mendelssohn D et al. Percutaneously
inserted silastic jugular hemodialysis catheters seldom cause
jugular vein thrombosis. ASAIO J 1995; 41: 169–172
3. Konner K. Subclavian haemodialysis access: is it still justified
in 1995? Nephrol Dial Transplant 1995; 10: 1988–1991
4. Clark DD, Albina JE, Chazan JA et al. Subclavian vein stenosis
and thrombosis: a potential serious complication in hemodialysis
patients. Am J Kidney Dis 1991; 15: 265–268
5. Gallieni M, Cozzolino M. Uncomplicated central vein catheterization of high risk patients with real time ultrasound guidance.
Int J Artif Organs 1995; 18: 117–121
6. Mallory DL, McGee WT, Shawker TH et al. Ultrasound
guidance improves the success rate of internal jugular vein
cannulation. A prospective, randomized trial. Chest 1990; 98:
157–160
7. Skolnick ML. The role of sonography in the placement and
management of jugular and subclavian central venous catheters.
Am J Roentgenol 1994; 163: 291–295
8. Patel KR, Chan FA, Kerr A et al. Subclavian artery to innominate vein fistula after insertion of a hemodialysis catheter. J Vasc
Surgery 1991; 13: 382–384
9. Denys BG, Uretsky BF, Reddy PS et al. Ultrasound assisted
cannulation of the internal jugular vein. A prospective comparison to the external landmark guided technique. Circulation
1993; 87: 1557–1562
10. Koski EM, Suhonen M, Mattila MA. Ultrasound-facilitated
central venous cannulation. Crit Care Med 1992; 20: 424–426
11. Denys BG, Uretsky BF. Anatomical variations of internal
jugular vein location: impact on central venous access. Crit Care
Med 1991; 19: 1516–1519
12. Cavatorta F, Zollo A, Campisi S, et al. Internal jugular vein
catheterization under echographic guidance. Int J Artif Organs
1993; 16: 820–822
13. Troianos CA, Jobes DR, Ellison N. Ultrasound-guided cannulation of the internal jugular vein. A prospective, randomized
study. Anesth Analg 1991; 72: 823–826
14. Brancaccio D. Real time ultrasound for venous catheter placement. Int J Artif Organs 1995; 18: 115–116
Received for publication: 25.6.97
Accepted in revised form: 20.11.97