VOL.

95,

No.

Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

THE

By JOSEPH

‘if

HE

RELATION
OF ARTERIAL
TRAUMA
COMPLICATIONS
OF CEREBRAL
ANGIOGRAPHY*
H.

ALLEN,

M.D.,t

complications

raphy
workers.7

have
Most

CARLOS

of cerebral

been
investigated
of the studies

the

literature

have

been

this

has

it

impossible

made

PARERA,

3.

angiog-

to

serious

complications

termined.
postmortem

has

Crawford’
has
examinations

been

de-

demonstrated
may

show

that
ex-

phy and
associated

the artery
when
death
8 months
after
angiograsome of the cases were apparently
with
complications
during
or

following

angiography.

tensive
occurs

2

damage
hours

i.

Death.

2.

Neurologic
symptoms

angiographic

These

complications.
and
signs

indicating

of almost

may

be seen

are:

(a)

or

weakness,
seizures;
(c)
or paresthesias;
field defects
or
neuritis

(caused

damage

to

the

brachial

damage

to

the

recurrent

nerve

causing

From

Neurological

the

Department
Institute,

of the
(b)

mo-

needle

plexus);

(1)

laryngeal

hoarseness.
of Radiology,

Columbia-Presbyterian

Columbia

University,
Medical

Center,

College
New

t Special
ment
cTI

sub-

heto
or

disor-

certain

Records

should

of

the

a

careful
with
testing
to the drugs

also

be

kept

of

the

the quantity
of saline
solution
and contrast
medium
injected
(and
over
what
period).
The roentgenograms
taken
should
be carefully
examined
(in cases
where
there
are

focal
or
sensory(d) visual
blindness;
by

arm
or

pulse
and
blood
pressure,
needling
difficulties,
evidence
of arterial
damage
or extravascular
injection
of contrast
material,

These

coma;

the

(due
to anxiety),
hypotension,
ders of rhythm,
heart
failure.
Arteriovenous
fistula.

used.
in-

part

in
axillary,

causes
of most
complications,
prospective
study
is required,
of the patient
for sensitivity

Clinical
any

confusion

brachial

include

on occasions.

ischemia

brachial,

rupture
of aneurysm,
etc.
In order
to be reasonably

compli-

described.

brain

tor-hemiplegia,
generalized
numbness
scotomata,

*

of

of
the

M.D.

POTTS,

To this list one may
add some
less significant
or very rare complications,
such as
drug
sensitivity
to minimal
doses,
partial
or complete
occlusion
of the vessel
by subintimal
hemorrhage,
thrombosis
at the site
of puncture,
stimulation
of the carotid
sinus by palpation
or needling,
air embolism,
increase
of intra-arterial
pressure
causing

to

volvement

(e)

7.

to

A great
number
cations
have
been
the following:

GORDON

clavian
artery
is punctured).
Tracheal
compression
by a neck
matoma.
5. General
or local
allergic
reactions
the contrast
medium
or the local
general
anesthetic
agents.
6. Cardiac
complications:
Hypertension

determine

not

Evidence

D.

.

and

the causes
of many
of the complications.
Nearly
all of the authors
suggest
that some
of the complications
were
associated
with
a difficult
arterial
puncture
or the injection
of some or all of the contrast
material
outside the lumen
of the vessel,
but the frequency
with which
extraluminal
injections
or difficult
needle
punctures
are followed
by

and

(where

by many
reported
in

retrospective

M.D.,

TO

complications)
for evidence
of air
vessels
or rupture
of an aneurysm
the injection
of contrast
material.
The clinical
status
of the patient

is like-

wise

disease

important

of Physicians
York,

New

serious

and

Surgeons,

and

the

and

Blindness,

iFs

i

brain

Radiological

Service

of the

York.

Training
Fellowship,
NIH,
National
Institute
of Neurological
Diseases
of Radiology,
Vanderbilt
University,
School
of Medicine,
Nashville,
Tennessee.
Visiting
Fellow
in Radiology,
NIH,
National
Institute
of Neurological
Diseases
NB98-og.
Now
at Department
of Neuroradiology,
Institute
of Neurological

845

since

in the
during

and Blindness,
Sciences,
Madrid,

NBi

167-os.

Neuroradiology
Spain.

Now

at Depart-

Training

Grant

Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

846

Joseph

lic. i.
ogram
(arrow).
another
needle

H.

that the point

tile

unusually

patient

vulnerable

D.

moderate

Gordon

Potts

object

of

how

DFcL1RR,

1965

tile

stud

present

of

punctures

tile

vessel

pared with tile
incidence
ill cases
wilere
the puncture

and
under

“clean”.
includes

of

to

trau-

extralu-

material
were
complications.
in cases
was

diffi-

completely
was
comsimple

An

and

retrograde

angiograms
anesthesia

oband

with

and

and hypaque
angiography.

50

intravenous

sitivity.
Cournand

After
puncturing
type
of needle,

of

contrast

shortl
an

medium

that

the
the

necessary,

made
and

were
was

correctl

vessel

(Fig.

i, zi

tile

position

of

center
of the
plane
technique
anteroposterior

for

the

the
drug

ansen-

the vessel
with
a 2 cc. injection

needle

was

cent

before
unusual

was

taneous
allterOposterior
oid
roentgenograms

per

cent
an-

of the contrast

test dose

was
given
to detect

needle
carotid,

carotid

medium
giograph\

within

complications

was

vertebral
local

or
or

wilere the needling
of tile vessels
cult or the ifljectiOn partl
or
outside
the lumen
of
tile
vessel

tile injection
The
study

was

difficult

6o per
vertebral

Renografin

the

for

giographies
brachial

S’I’UDV

frequently

sedation.

used

was

minal
injection
of
contrast
associated
wi til allgiograpilic
the incidence
of complications

tained

and

of

of angiograpil\.

determine

brachial,

Parera

of the

roentgenograms

showing

I’RESENT

lllatic

Carlos

neck after injection
of 2 cc. of contrast medium.
(A) Lateral roentgenthe needle lies against
the posterior
wall of the common
carotid
artery
There is a small quantity
of subintimal
contrast
medium.
(B) Anteroposterior
roentgenogram
of
case
showing
that the needle is pressing
against
the lateral wall of the artery.
The point
of the
was positioned
more
medially
before
the main injection
was made.

Polaroid

niav
nlake
to tile risks

lile

Allen,

a

and

simul-

lateral
taken
to

Polarprove

positioned
and

If found

point
of
it near

the

vessel.
A simultaneous
was
employed
so that
and lateral
roen tgenograms

bithe

altered

to

the
place

B).

the

VOL.

Arterial

No.

95,

Trauma

Complications

Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

with a single
injection
of
Ten cc. was used for cain adults
and 6 cc. for
angiography;
35-45
cc. was usufor
brachial
roentgenograms

angiography.
included

the

neck vessels
permitting,
in carotid
and vertebral
angiography,
the ready
recognition
of imperfect
injections.
It is important
to
note

that

only

the

placement
the blood

may
flows

be recognized
back
from
the

by the
needle.

ing

of the

injections

the

angiographer

had

performed

many

felt

confident

grossest

that

“clean”

puncture

flow of
revealed

blood;
that

errors

he

because

manual

tion.
The

Total

a

free

back

Polaroid
material

blood

injection

films
was

pressure

of saline

was

tervals
throughout
the
hand
strength
was measured
and after
the examination

way
sign

on
of

reduction

one side was
a developing

of the

solu-

recorded

at

in-

examination
and
before,
during,
with
a dynamo-

meter.
These
objective
strength
ments
were
found
helpful
in the
ognition
of certain
complications.
ness
of the hand
strength
measured
sometimes
hemiplegia.

strength

measureearly
recA weakin this
the

of both

of

No.

Examinations

Right Carotid
Left Carotid
Right Brachial
Left Brachial
Right Vertebral
Left Vertebral

way
Dur-

partially
subintimal.
Almost
all injections
were made
with
a Taveras
mechanical
injector
and
between
injections
the tubing
and needle
were
kept
clear
of clots by intermittent

of

No.

Artery

Injections

157

219

240

341

6

96

38
14

46
17

24

28

538

747

of needle

of the

however,
the
the
contrast

Ii

TABLE

could
be obtained
contrast
medium.
rotid
angiography
vertebral
ally
injected
The
lateral

847

earliest
Gradual

hands

seen when
the sedation
was increased.
This
series
includes
data
compiled
from
400
patient
visits
for angiography:
538
examinations
were
performed
and 41 patients
made
more
than
one visit.
Unsuccessful
examinations
are included
even
if
no contrast
material
was injected,
but at
least an attempt
was made
to puncture
the
artery.
The
angiographic
examinations
were
April
from

done
between
1964
and
the
the series
were

September
1963
and
only
cases
eliminated
those
whose
records

were

incomplete.

281

In

gle vessel
and
were
injected.
are shown
aminations

the
cc.

in Table
performed

number
injections

one

The
number
of exon each
vessel
and

i.

were

sin-

two vessels
diagnoses

of injections
(excluding
made
before
the

roentgenograms
Table

instances

in Jo! instances
The
angiographic

taken)

are

the
Polaroid

2

shown

in

ii.

was
COMPLICATIONS

TABLE

Angiographic

Diagnosis

I

In
were

.

No.of

Cases

Normal
Vascular
occlusive
disease
Aneurysms,
arteriovenous
malformations
or follow-up
studies
on these

intracerebral

Total

i death,

ab14
400

538

and 19 complications
recovered
after

time.

The

findings

two

described

sented

I.

with

communicating
right
internal

neck
showed

in the

are

examinations

there

(3.9 per
i permanent

plegia
patient

CASE
9

dilatations,
diagnosis)

included

The
104

or

series
of
complications

complications
55

cases

Tumors
Subdural,
epidural
hematomas
Others
(junctional
scesses, questionable

i6i
57

this
2!

from
a variable
cases

summarized

most

serious

in more

detail

J.O.,

a female

a hemorrhage

cent)
right

which
hemi-

which
period
that

showed

in Table

complications

the
of
III.

are

below.
aged

59

from

a right

years,
preposterior

aneurysm
which
was treated
by
carotid
artery
occlusion
in the
in
1955.
Postoperative
angiography
no filling
of the aneurysm
from either

Joseph

848

H.

Allen,

Carlos

Parera

Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

TABLE

Case
No.

Acre

i

5

2

6

Patient
became
unresponsive
after
main
injection
of
contrast
medium;
death
36 hours
later
Right
hemiparesis
and
patient
became
obtunded
hours
after
ar-

teriography;

permanent

6

52

7

52

8

33

Weakness

3

6
6
68

57
io

of left

arm

after

attempted

left vertebral
injection;
complete
recovery
in 4 days
Aphasia,
apraxia
and confusion;
recovered
in 2 days
Loss
of vision
in both
eyes;
gradual
recovery
over 2 days

i

Gordon

Potts

1965

DECEMBER,

III
Subintimal

Contrast

Angiographic
Diagnosis
and Comments

LCd

No

LCd

No

LV1I

No

No

Normal

LCd

LCd

Yes
No

Yes
No

Normal
Cerebral

atheroma

RBII

No

No

Cerebral

atheroma

RCA
RBA

No

Yes

LVII

Yes

Yes

Atherosclerotic
ease
Left cerebellar
toma

No

Yes

No

No

Yes

(total
extravasation
of
contrast
medium)
Yes (wall of vessel
irregular
due
to
needle
trauma)

hemi-

paresis and aphasia
Transient
right
hemiparesis;
complete
recovery
in 6 hours
Aphasia,
cleared
in 24 hours
Severe
blood
pressure
drop
during
examination;
recovered
in a few
hours
Numbness
of
left
side;
complete
clearing
in 2 days
Hives
(lasted
2 days)

D.

Difficult
Arterial
Puncture

Vessels
Injected*

Complication

and

LBA
RCA

aneurysm;
precarotid
ligation

Left
middle
rysm

cerebral

the
may

left
have

aneu-

vascular

dis-

hemangioblas-

Left
sided
atherosclerosis
Arteriovenous

LCd

LCA

Right
carotid
vious
right

aneurysm

and

malformation;
vertebral
artery
been accidentally

punctured
69

II

Aphasia
2 days

and

confusion;

12

68

Dysphasia,
returned
in 2 days

13

51

Blood
pressure
drop
ing to normal
in
Confusion,
dysphasia

SI

14

sided

i

66

i6

g

17

7

i8

o

19

38
o

20

21

in

worse
after examination;
to pre-angiographic
state

weakness;

i

to 8o/6o,
returnhour
and mild right

cleared

in 6 hours

Right
arm weakness;
complete
clearing in 2 days
Right
arm weakness
and confusion;
cleared
in 24 hours
Low blood
pressure
at intervals
for
24 hours
Intermittent
fall of blood
pressure
for
hours
Hives
(lasted
12 hours)
Blind
spots,
scotoma,
nausea
and
vomiting
lasting
4 hours
Loss of voice,
and hoarseness
due to
recurrent
laryngeal
involvement;
recovered
in 12 hours

g

#{149}
LCA

recovered

Left

carotid

No

LCd

Yes

LCd

No

RCI

Yes

No

Normal

RBA

No

Yes

Atherosclerosis

No

Yes

Normal

LCd

Yes

Yes

Normal

RCI

Yes

Yes

Unsatisfactory

LCd

Yes

Yes

Normal

LCd
RCA
LCA

No
Yes

No
Yes

Normal
Right
sphenoid
ingioma
Normal

one

Temporo-thalamic

tumor

LBA
Lcd
(retrograde
injection)

RCA
LCd

(both

sides)
No

LVII

RCA
LVA
RVA

to be the

Glioblastoma

showed
irregularity
of artery
near
needle)
No

artery

Right
carotid
artery
Left vertebral
artery
Right
vertebral
artery
LBA
Left
brachial
artery
RBA
Right
brachial
artery
The vessel
italicized
is presumed

(roentgenograms

concerned

in the

complication.

(both

sides)
No

angiogram

ridge

men-

\OL.

Arterial

No.

95,

Trauma

Complications

849

:
Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

9
k

‘4

-I

2.

2.

!I1i

Tn1L

.1
Vlc\\

111

th

fl

I

(h1.tJ

nt.
tii

Sj?1
1 tCT’M

1

I

H.

side

and

the right
internal
carotid
be completely
occluded.
developed
headaches
and
a right
shown

to

nerve

lesion

ization

and

angiography

artery

In
third

showed

1959

I
jr.

!t[Td
1 ltd

\

(:

1 1

11111

( )!w

Immediately

c

was

injected

she

The
right leg was more spastic
than the left and the
left eye was deviated to the right.
Fifteen
miiiutes
later tile patient
was very restless
but she
could not
move
iler
rigilt
limbs
as well as tile
left. The right pupil was contracted
to pinpoint
and she was markedly
ilypotensive.
She gradually deteriorated
and death
occurred
36 hours
later.

cranial
recanal-

of the internal
carotid
artery.
Because
was no evidence
of a further
subarachnoid
hemorrhage
on
this occasion,
the
patient
was
treated conservatively.
One day before her final admission tile patient
had a subarachnoid
hemorrhage.
zingiograp/zy.
Two injections
of 2 cc. of contrast
material
were
Illade
into
tile
left common
carotid
artery
in order
to check
the needle
position with
Polaroid
films and then the main
injection
of io cc. was made.
Most
of the contrast
material
during
this
last
injection
was
there

subintimally.

.LU\t11
.

injection

tile

patient

became

after

this

decerebrate.

male
aged
6 years,
was
had
hypertension
(up
to
260/150)
for
i
years.
Seven
days
before
admission
the patient
had
a subarachnoid
hemorrhage which was followed
by aphasia and a right
CASE

known

2.

L.\V.,

to

have

a

850

Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

hemiparesis,
Angiography.

raphy
contrast

both

Joseph

H.

Allen,

of which

gradually

A left common

Carlos

Parera

angiog-

was

performed
using
a total
of 22 cc. of
medium.
Frontal,
lateral
and oblique
views
showed
a middle
cerebral
aneurysm
with
marked
associated
vascular
spasm
and a low
convexity
subdural
hematoma
(Fig. 2, zI, B and
C). Four hours
after
the study
the patient
became
obtunded,
there
was a more
dense
hemiplegia
and the patient
was more
aphasic.
Sixteen months
later the patient
still had some left
sided
weakness
and
severe
aphasia
but
was
ambulatory.
DIFFICULT

Where
with the
tured,

ARTERIAL

more
needle
the

puncture

attempts
the artery
was

were
was

recorded

way

we

have

trauma

inflicted

jacent

tissues.

a crude

on
One

artery

hundred

its

ad-

fourteen

(21

per cent)
of the 538 vessels
injected,
or
where
puncture
was attempted,
were
classified
as difficult;
of these
i ij.
difficult
punctures
10
(9.6 per cent) were associated
with complications.
Of the
2!
cases
showing
angiographic
complications,

7 (33

per

cent)

were

difficult

to puncture.
SUBINTIMAL

INJECTIONS

In 38 of the
538 examinations
(7.1
per
cent)
some subintimal
or extraluminal
contrast
medium
injection
occurred
and
of
these
9 showed
all of the contrast
medium
outside
the lumen
of the vessel.
Of the 2!
cases

showing

cent)

were

intimal

complications,

recorded

contrast

as

(62

13

having

some

per

sub-

medium.

Of

the

ture

or

cases

2!
14

with

had

subintimal

injection

medium
or both.
It
that,
in some
cases,
was

caused

by

angiographic

either

the

com-

a difficult
of

punccontrast

seems
almost
certain
an intimal
dissection
injection

of saline

occurred

taken,

it

corded

were

associated

lumen

by

subintimal

only

death

in the

series

a subintimal
injection
carotid
angiography
patient

who

had

the

usually

it is probable
than
those
re-

with

the

after

would

not be recognized.
Therefore,
that
more
complications

as

narrowing

of

dissection.
was

The

associated

when
left
was
performed

previously

solu-

much

damage

with

a Cournand

puncture

but any narrowing
is not
usually
seen

had

with
common
on
the

a

right

to

the

vessel

type

caused
unless

as

by the puncture
the
catheter

into the
to opacify

vessel

wall

at

the

tured.
Serious

point

where

allergic

intravenous
dium
is

test
given.

the

complications
in

dose
of
Hypertension

is

vesthe

is puncmay

if the

ly always
be avoided
about
drug
sensitivity

a

of needle

only inserted
a short
distance
sel or the reflux
is sufficient

near-

patient

is asked

advance

and

the

an

contrast
meis generally

caused
by apprehension
be controlled
by increased
tension
may
be induced

and

may
sedation.
by painful

generally
Hypostimuli

or

needling,

particularly

of

and
may
ephedrine.

be corrected
Most
of the

traumatic

carotid
intravenous

sinus,

serious
neurologic
ies appear
to

difficult
sel or
tions.
tions

have

to

the
with
more

complications
in this serbeen
associated
with

or traumatic
subintimal
Presumably
are caused

ondary
DISCUSSION

plications,

damage

were

least

of the

and

if this

1965

DECEMBER,

diffi-

as

indication

the

and

angiograms

Potts

made
punc-

cult.
It is recognized
that
what
may
be a
difficult
puncture
for one operator
may
be
easy
for another,
but
by classifying
it in
this

Gordon

internal
carotid
artery
ligated.
It has sometimes
been
suggested
that
damage
to the vessel
does not occur
when
catheter
techniques
are used.
However,
it
seems
likely
that
the percutaneous
insertion of a catheter
into
a vessel
causes
at

PUNCTURE

than
before

D.

tion

improved.

carotid

and

or
by

punctures
extravascular

most
of
cerebral

a traumatic

of the

vesinjec-

these
complicaischemia
sec-

examination.

These

complications
should
be greatly
reduced
by
an atraumatic
puncture.
The
direct
toxic
effects
of the contrast
medium
undoubtedly
cause
some complications
but these
can be
reduced

by

quantity

of contrast

ing

repeated

injecting
injections

the

medium
at

smallest

possible

and

by avoid-

short

intervals.

VOL.

When

Downloaded from www.ajronline.org by 24.231.218.146 on 06/21/16 from IP address 24.231.218.146. Copyright ARRS. For personal use only; all rights reserved

Arterial

No.

95,

the

simultaneous

biplane

Trauma
technique

of filming
is not used
and each projection
requires
a further
injection
of contrast
medium,
larger
quantities
of the medium
are
employed

toxic

and

effects

The
gest

findings
that,

technique
raphy

it is to

of the

be

expected

that

medium

will

increase.

in
when

the
a

series
simultaneous

is employed
with

the

of

sugbiplane

for cerebral
injection

cases

the

of

Complications
dium
rare.
D.

851

(other

Gordon

than

reactions)

Medical

Center

are

M.D.

Potts,

Department

allergic

of Radiology

Neurological
Institute
Columbia-Presbyterian
622

West

i68th Street

New

York,

New

York

10032

angiogrelatively

small
quantities
of contrast
medium,
over
one-half
of the complications
are caused
by
trauma
to the vessel
or subintimal
injection of contrast
material.
Presumably,
as
larger
quantities
of contrast
material
are
injected
during
angiography,
the importance
of the toxicity
of the contrast
medium
increases
in relation
to arterial
trauma.
SUMMARY

In a series of 538 angiographic
studies
on
400
patients
there
were
21
complications
including
I death
and I permanent
hemiplegia.
All of the remaining
complications
cleared
within
days.
The
evidence
suggests
that when
simultaneous
biplane
angiography
is used,
and
small
quantities
of contrast
medium
are injected,
trauma
to the artery
and subintimal
injection
of the contrast
material
are the
most common
causes
of complications,
and
that
true
toxic effects
of the contrast
me-

The
authors
tance
of Dr.

tion

of this

wish

to acknowledge

M. Taveras

Juan

the

in the

assis-

prepara-

paper.
REFERENCES

I.

2.

T. Pathological
arteriography.
7.
Neurol.,
Psychiat., 1956, 19, 217-221.

CRAWFORD,

J. R.,

FIELD,

SURE,
R.
angiography
Neurosurg.,
3.

cerebral

Complications

in

2,000

1962,

19,

A. D., and

KAPLAN,

of

Jr.

cerebral

of

Neurosurg.,

J. T., and

ROBERTSON,

L.,

effects

&

DE
of

consecutive
775-78!.

SAUScerebral

7.

cases.

A. E. Complications

WALKER,

angiography.

Neurology,

4,

643-656.
4.

.

B. Cerebral

KENDALL,

Brit. 7. Radiol.,
D. W.,

LINDNER,

and

GURDJIAN,

tions in patients
Neurosurg.,
1962,

angiography
using Conray.
1964,37,
8 1-589.
HARDY,
W. G., THOMAS,
L. M.,
E. S. Angiographic
complicawith cerebrovasculardisease.
7.
19,

179-185.

6. NEWTON,

T. H., and

COUCH,

errors

in arteriographic

diagnosis

carotid

artery

Radiology,

occlusion.

R.

S.

C.

Possible

of internal
1960,

75,

766-773.
NING,

R. H., JR., GOODELL,
H., and DUNH. S. Complications
of carotid
arteriog-

raphy.

Arch.

7. PATTERSON,

Neurol.,

1964,

JO,

5 13-520.