British Journal of Anaesthesia 1993; 70: 145-148
USE OF ULTRASOUND TO EVALUATE INTERNAL JUGULAR
VEIN ANATOMY AND TO FACILITATE CENTRAL VENOUS
CANNULATION IN PAEDIATRIC PATIENTSf
P. J. ALDERSON, F. A. BURROWS, L. I. STEMP AND H. M. HOLTBY
SUMMARY
Percutaneous cannulation of the internal jugular
vein in paediatric patients may be technically
difficult and is prone to complications. To investigate the possibility that anatomical factors
contribute to these difficulties, we used a twodimensional ultrasound scanner to examine venous
anatomy in children aged up to 6 yr. We found that
18% of our children had anomalous venous anatomy that may account for some of the difficulties
reported previously. The diameter of the internal
jugular vein was predicted poorly by the patient's
age (r2 = 0.259) or weight (x2 = 0.155). We also
evaluated the use of this ultrasound scanner during
percutaneous central venous cannulation in
neonates and infants. Determining the course of the
internal jugular vein with the scanner and then
marking it on the overlying skin reduced both the
time and number of needle insertions required to
aspirate jugular venous blood and increased the
chance of a complication-free cannulation. (Br. J.
Anaesth. 1993; 70: 145-148)
KEY WORDS
Anatomy: internal jugular vein. Anaesthesia: paediatric. Veins:
cannulation, ultrasound
The approach to central venous cannulation in
children depends on the experience of the operator
and the indication for catheterization. Situations that
require definitive intrathoracic central venous placement merit techniques in which the success rate is
high (cannulation of the internal jugular or
subclavian vein) [1]. Many paediatric anaesthetists
avoid a subclavian approach because of the report by
Groff and Ahmed [2] describing five major complications of this approach when used in 44 children
younger than 2 yr. Therefore, the internal jugular
vein provides the best means of reliably achieving
intrathoracic venous cannulation [3,4], although
cannulation by the internal jugular route is more
difficult in children younger than 3 months of age or
less than 4 kg in weight [4].
Denys and Uretsky [5] reported that anatomical
variation in the position of the internal jugular vein
in adults may complicate venous access by this route
when a landmark-guided approach was used. A
similar survey in children has not been conducted;
indeed, little has been published on the anatomy of
the internal jugular vein in paediatric patients [6].
Recent work has shown that two-dimensional
ultrasound may assist the anaethetist in cannulatjng
the internal jugular vein in adult patients [7—9]. This
technique has been shown to increase the chance of
successful cannulation, particularly at the first attempt. Significant reductions in the time taken to
aspirate venous blood from the internal jugular vein
and in the risk of accidental puncture of the carotid
artery have been reported.
We have examined internal jugular venous anatomy and the use of ultrasound to facilitate jugular
venous cannulation in neonates and infants.
PATIENTS AND METHODS
Approval was obtained from the Human Subjects
Review Committee of the hospital.
Part 1
The object of the first part of the study was to
determine if variations in the anatomy of the internal
jugular vein could account for some of the difficulty
experienced during landmark-guided percutaneous
jugular venous catheterization in children.
We studied 50 patients younger than 6 yr undergoing cardiac surgery. The age (days) and weight
(kg) of each patient were recorded. After induction
of anaesthesia and nasotracheal intubation, the
patient was positioned as if for cannulation of the
right internal jugular vein: a rolled towel was placed
under the shoulders, the head turned to the left and
the table tilted head-down by 15°. Various
characteristics of venous anatomy were examined
with a portable, battery-operated, two-dimensional
ultrasound
scanner
(Dymax
Corporation,
Pittsburgh, PA), which has a mechanical liquid-path
7.5-MHz transducer that allows high resolution
imaging of structures up to 40 mm deep to the skin.
We examined the relationship between the internal
P. J. ALDERSON, M.A., M.B., B.S.J F. A. BURROWS, M.D., F.R.C.P.C.J
L. I. STEMP*, M.D.; H. M. HOLTBY, M.B., B.S., F.R.C.P.C; De-
partment of Anaesthesia, Room 2303, The Hospital for Sick
Children and the Research Institute of the University of
Toronto, 555 University Avenue, Toronto, Ontario, Canada
M5G 1X8. Accepted for Publication: September 4, 1992.
•Present address: The Cleveland Clinic Foundation,
9500 Euclid Avenue, Cleveland, Ohio, U.S.A.
Correspondence to F.A.B.
fPresented in part at the annual meeting of the American
Society of Anesthesiologists, New Orleans, Louisiana, U.S.A.,
October 1992.
146
BRITISH JOURNAL OF ANAESTHESIA
jugular vein and the carotid artery from the angle of
the mandible to the clavicle, and the diameter and
depth of the internal jugular vein at the level of the
cricoid cartilage. Diameter and depth were measured
to the nearest 0.5 mm with a pair of calipers applied
to the image on the scanner screen.
Part 2
The object of the second part of the study was to
determine if the use of such an ultrasound scanner
could facilitate percutaneous internal jugular vein
cannulation in neonates and infants.
We studied 40 patients younger than 2 yr who
would undergo percutaneous insertion of an internal
jugular cannula during cardiac surgery. None had
undergone prior cardiac surgery or jugular cannulation. They were allocated randomly to one of
two groups: group A = cannulation guided by
anatomical landmarks; group B = cannulation
guided by ultrasound.
All central venous cannulations were carried out
after induction of anaesthesia and nasotracheal
intubation. The patient was positioned for cannulation of the right internal jugular vein with a
rolled towel under the shoulders, the head turned to
the left and the table titled head-down by 15°. For
those in group A, the two heads of the sternomastoid
were palpated, the position of the carotid artery
verified and the cannula inserted at the junction of
the two heads of the sternomastoid angled toward
the ipsilateral nipple. For those in group B, the
ultrasound scanner was used to visualize the internal
jugular vein so that its projection could be marked
onto the overlying skin with a pen; this line then
served as a guide for cannulation.
In every patient, cannulation was carried out by an
experienced cardiac anaesthetist who used a 21gauge, 4-cm needle to locate the internal jugular
vein. A 16-gauge polyurethane double-lumen catheter was positioned using a Seldinger wire passed
through the 21-gauge needle. The time from the
initial skin puncture to the aspiration of venous
blood was recorded, together with the number of
needle insertions required to locate the vein. Details
of any complications were noted.
Data analysis
Data are presented as arithmetic means (SD). The
lines of best fit for the graphical figures were
determined by linear regression analysis. Parametric
data were analysed by one-way analysis of variance
and the Student-Newman-Keuls test. P < 0.05 was
considered to be significant.
RESULTS
Part 1
The patients ranged in age from 3 days to 5.5 yr
and in weight from 2.7 to 22 kg. None of the children
had been born prematurely.
In 82 % of patients, the internal jugular vein ran
antero-laterally to the carotid artery. In 4%, the
vein was in a similar position relative to the carotid
artery, but was of unusually small diameter (denned
as ^ 3 mm for neonates and infants and as ^ 5 mm
H
-i
12 -
10 -
8 ac
•a
6 -
4 -
2 -
0 -I
1
•
2
3
Age (yr)
I
'
[
4
FIG. 1. Relationship of the depth of the internal jugular vein (IJV)
to age. r1 = 0.290.
14 -i
12 -
10E
E
8
Q.
ID
642 0
J
1
0
•
1
5
•
i
10
•
i
1
15
20
'
1
2
Weight (kg)
FIG. 2. Relationship of the depth of the internal jugular vein (IJV)
to weight, r1 = 0.379.
for older children). In 10 %, the vein ran so medially
that it entirely overlay the carotid artery; in 2 % the
vein ran widely lateral to the carotid artery. In the
remaining 2%, the vein could not be seen.
At the level of the cricoid ring, the vein was seen
4-10 mm below the surface of the skin. This distance
tended to increase with both age and weight, the
better correlation being with weight (figs 1,2). At the
same site, the vein varied in diameter from 2.5 to
12 mm; again, vein size tended to increase with both
age and weight, but correlations were poor (figs 3,4).
ULTRASOUND AND JUGULAR CANNULATION IN CHILDREN
than in those in group B in whom ultrasound was
used to mark the course of the vein. In addition,
significantly more needle insertions (P = 0.018) were
needed to locate the vein in group A patients.
The most common complication during
cannulation was inability to pass the Seldinger wire
into the superior vena cava after venous blood had
been aspirated. This occurred eight times in group A
and three times in group B. Carotid artery puncture
occurred in two patients in group A and in one in
group B. Overall, eight cannulations in group A and
16 in group B were uncomplicated {P < 0.05).
14 -,
12- 1 0 E
~
147
8H
CD
I e
'S
B 42 -
DISCUSSION
0
1
2
3
4
5
6
Age (yr)
FIG. 3. Relationship of the diameter of the internal jugular vein
(IJV) to age. r1 = 0.259.
14 12 10-
E
£
8 -
a
E
CD
•D
^
420
J
10
15
Weight (kg)
20
25
FIG. 4. Relationship of the diameter of the internal jugular vein
(IJV) to weight, r1 = 0.155.
TABLE I. Comparison of landmark-guided and ultrasound-guided
cwmulation techniques (range or mean (so)).* P < 0.05 compared
with group B
Access data
Success rate
Time taken (s)
Time taken (s)
No. needle insertions
Group A
Group B
Landmark-guided Ultrasound-guided
16/20
6-170
56.4 (48.9)*
2.0(1.0)*
20/20
2-90
23.0 (27.4)
1.35(0.7)
Part 2
The two groups were similar in age and weight
(group A: 281 (218) days, 6.8 (2.5) kg; group B: 258
(170) days, 6.6 (2.5) kg). Each group contained three
neonates and 17 infants.
Table I compares the success of the two
cannulation techniques. In four of the children
assigned to group A, the internal jugular vein was
not located. In the other 16 patients, it took
significantly longer (P = 0.011) to locate the vein
Eighteen percent (nine of 50) of the patients younger
than 6 yr in our study had anatomical factors that
may complicate a landmark-guided approach to the
internal jugular vein.
In a previous prospective study [10] of landmarkguided percutaneous internal jugular venous
catheterization in children aged up to 14 yr, the right
internal jugular vein was cannulated successfully in
31 of 52 patients (59.6 %), and the left in nine of 52.
Greater success rates were obtained with children
older than 2 yr, although this finding was not
statistically significant. The complications noted
included 12 (23%) carotid artery punctures. In a
retrospective study [11] of percutaneous jugular
venous catheterization in 200 children, 45% of
whom were 5 yr or younger, the internal jugular vein
was cannulated successfully in 109 of 128 patients
(85.2 %). The recent Osaka study [4] of 106 children
aged up to 7 yr reported that the rate of successful
cannulation on the first attempt, defined so that it
included multiple needle insertions at a single site,
was 78.3%. Complications included 12 (11.3%)
arterial punctures.
From these studies it appears that the proportion
of patients in whom percutaneous internal jugular
cannulation is difficult is of the same magnitude as
the proportion of patients in whom the relationship
between the vein and the carotid artery is anomalous
in some way.
Venous blood is often identified during
cannulation as the needle is drawn back through the
vein. As we have found that the carotid artery runs
posterior to the vein in 10% of the children in our
study, we suggest that the risk of carotid puncture is
of this magnitude. This risk appears to be similar to
the rates of arterial puncture in the research cited
above.
There is little information on the diameter of the
internal jugular vein in childhood. One anatomical
survey [6] included 11 children aged from 13 months
to 16 yr. This subset showed a poor correlation
between the diameter of the internal jugular vein and
age, height, weight or body surface area. The
smallest vein seen in these children was 6.6 mm in
diameter when the child was in the Trendelenburg
position. Their anatomical findings are similar to
those reported here.
Inability to demonstrate an internal jugular vein in
2% of our patients was surprising: congenital
absence of the internal jugular vein is extremely rare,
148
particularly in the absence of other cerebrovascular
malformations. There are two possible explanations
for our result: the ultrasound scanner could not
identify the internal jugular vein because the diameter was so small, or the ultrasound probe
occluded the vein because, despite our efforts, the
probe exerted excessive pressure on the neck.
In comparison with group A in our study, both the
time required to locate the internal jugular vein and
the number of needle insertions required were less in
group B. These findings concur with those of earlier
reports in adult patients [7-9], indicating that
ultrasound guidance improved the speed and safety
of central venous access. In previous studies the
ultrasound probe was wrapped in a sterile sheath so
that it could be manipulated in the sterile field by the
operator during cannulation. Although this technique allows the operator to guide the needle into the
vein visually, it requires the extra expense of the
sterile sheath and a considerable degree of manual
dexterity in paediatric practice. We found that our
simplified approach, without the sheath, was also
effective.
Using the ultrasound scanner, we found that our
complication rate was 20%, compared with 60%
when using the landmark approach. Our most
common complication was inability to insert the
Seldinger wire into the superior vena cava. Other
studies have not reported this as a complication,
although it can greatly increase the overall time
required to place the central catheter. The reason for
the high incidence of difficulty inserting the
Seldinger wire in neonates and infants is probably
the ease with which the needle tip becomes displaced
from the lumen of the vein as the syringe is
disconnected from the needle. We were unable to
show that the rate of carotid puncture was reduced
with the use of ultrasound because of our small
sample size.
BRITISH JOURNAL OF ANAESTHESIA
ACKNOWLEDGEMENTS
We thank Dymax Corporation for loan of the Site Rite ultrasound
scanner for the duration of these studies.
REFERENCES
1. Motoyama EK, ed. Smith's Anesthesia for Infants and
Children, 5th Edn. St Louis: CV Mosby Co., 1990; 242-243.
2. Groff DB, Ahmed N. Subclavian vein catheterization in the
infant. Journal of Pediatric Surgery 1974; 9: 171-174.
3. Postel JP, Quintard JM, Ricard J, Delaplace R, Bernard F,
Canarelli JP. Elaboration d'une technique sure d'acces
veineux central en pediatrie. Notre experience de 700
catheters centraux percutanes. Chirurgie Pidiatrique 1990;
31: 219-222.
4. Hayashi Y, Uchida O, Takaki O, Ohnishi Y, Nakajima T,
Kataoka H, Kuro M. Internal jugular vein catheterization in
infants undergoing cardiovascular surgery: An analysis of the
factors influencing successful catheterization. Anesthesia and
Analgesia 1992; 74: 688-693.
5. Denys BG, Uretsky BF. Anatomical variation of internal
jugular vein position: Impact on venous access. Circulation
1990; 82 (Suppl.): 111-68.
6. Mortensen JD, Talbot S, Burkart JA. Cross-sectional internal
diameters of human cervical and femoral blood vessels:
Relationship to subject's sex, age, body size. Anatomical
Record 1990; 22S: 115-124.
7. Denys BG, Uretsky BF, Reddy PS, Rufmer RJ, Sandhu JS,
Breishlatt WM. An ultrasound method for safe and rapid
central venous access. New England Journal of Medicine 1991;
324: 566.
8. Troianos CA, Jobes DR, Ellison N. Ultrasound-guided
cannulation of the internal jugular vein. A prospective,
randomized study. Anesthesia and Analgesia 1991; 72:
823-826.
9. Koski EMJ, Suhonen M, Mattila MAK. Ultrasoundfacilitated central venous cannulation. Critical Care Medicine
1992; 20: 424-426.
10. Prince SR, Sullivan RL, Hackel A. Percutaneous catheterization of the internal jugular vein in infants and children.
Anesthesiology 1976; 44: 170-174.
11. Nicolson SC, Sweeney MF, Moore RA, Jobes DR. Comparison of internal and external jugular cannulation of the
central circulation in the pediatric patient. Critical Care
Medicine 1985; 13: 747-749.