Alan
M. Cohen,
MD
#{149} Carl
Bronchial Hemoptysis
F. Doershuk,
(55%).
The
#{149}Robert
MD
C. Stern,
Artery Embolizatlon In Cystic Fibrosis’
Severe hemoptysis in cystic fibrosis can be life-threatening because of acute blood loss or because it interferes with sustaining physical therapy of the chest. Hemoptysis was controlled in 19 of 20 cystic fibrosis patients by means of embolization with a combination of 250-590-aim particles of polyvinyl alcohol foam and absorbable gelatin pledgets. Repeat embolization was needed in eight patients to achieve or maintam effective hemostasis. One or more aberrant bronchial arteries were found in seven patients (35%), and a spinal artery branching from a vessel that also supplied bronchial circulation was present in 11 patients
MD
data
suggest
that
although applicability
embolization has a wider than previously reported, care should be taken during angiography to evaluate the bronchial circulation because aberrant bronchial vessels and spinal arteries arising from bronchial circulation are common.
S
hemoptysis
EVERE
to Control
threatens
the
life and well-being of patients with cystic fibrosis. Although consenvative treatment effectively controls minor bleeding, bronchial artery embolization or surgery has often been required to stop massive hemoptysis (1-3). This report examines our experience in 20 cystic fibrosis patients who underwent bronchial artery embolization
to control
data suggest en applicability
hemoptysis.
embolization in cystic
previously
The
has a widfibrosis than
AND
Approximately
425 cystic
pa-
20 cystic
fibrosis
patients
that warranted
giogmaphic
evaluation.
Nineteen
patients
were
treated
with
of the
20
embolization
the patients at their first angiographic procedure was 23 years (range, 6-43
Radiology
procedures. lized more Patients
Some arteries were embothan once. were selected for angiography
after evaluation by pulmonary (C.F.D. and/or R.C.S.) and radiologic (A.M.C.) consultants. In 1 1 instances, hemoptysis was first treated with exhaustive conservative methods, including high-dose intravenous antibiotics. Patients who failed conservative I From
the
Cystic
Rainbow
Babies
Western
Reserve
Fibrosis
Research
Center,
and Children’s Hospital, Case University, Cleveland. Received March 20, 1989; revision requested May 3; revision received January 26, 1990; accepted February 1. Supported in part by National Institutes of Health grant DK27651, the Cystic Fibrosis Foundation, and the United Way Services of Cleveland. Address reprint requests to A.M.C., Chief Diagnostic Radiologist, Cleveland Metropolitan General Hospital, 3395 Scranton Rd. Cleveland, OH 44109. c RSNA, 1990
for angiography na: mL by or
treatment
were
on the basis
considered
of four
femoral
ducer
sheath
nique.
The
(Cook, most
prior
of 5.0-6.5
the
F were
with
crite-
(a) one hemorrhage greater than 300 in 24 hours (“massive”), accompanied ongoing hemoptysis at a lower daily near daily rate; (b) three or more 100mL hemorrhages within 1 week, accompanied by ongoing hemoptysis at a lower daily or near daily rate; (c) chronic or slowly increasing hemoptysis interfering with life-style; or (d) hemoptysis preventing effective postural drainage or home management.
con-
introduced
through
an intro-
the Seldinger
RC 1, and
Bloomington, useful
in-
Informed
to angiogmaphy.
artery
Hi,
tech-
RC 2 models
Ind) were
for cannulating
found
bronchial
vessels. Occasionally, Sos open-end guide wires with steerable cores (USC! Division of C.R. Bard, Billemica, Mass) or 4-F catheor coaxial
3-F Teflon
catheters
the descending
near
thoracic
Most
to-
vessels
(T-4-T-6).
suspected
bronchial
were
evaluated
projection
administration
meglumine
All
angiographi-
in the anteroposterior after
arose
of the carina
containing
cut film
amate
aor-
searched However,
was directed
of hemoptysis.
the level
vessels
cally
(Cook)
patients who localize bleeding to one side of the claiming to experience a “guror “funny feeling” prior to each
circulation
of one or more bronchial arteries during the acquisition of the original diagnostic or subsequent angiogram(s). Mean age of
years) (Table 1). One to three angiograms per patient were obtained, for a total of 36
175:401-405
into
episode
an-
Index terms: Arteries, bronchial, 943.1299 #{149} Arteries, therapeutic blockade, 943.1299 #{149} Fibrosis, cystic, 60.252 #{149} Lung, hemorrhage, 60.252 1990;
obtained
Catheters
could chest, gling”
(12 male)
hemoptysis
was
fama de-
procedure,
and risks.
angiographic evaluation ward one lung in seven
bow Babies and Children’s Hospital, Cleveland. From June 1982 to December 1986,
sent
of the
benefits
ta and intercostal arteries were for bronchial artery circulation.
tients are followed up on a regular basis by the Division of Pediatric Pulmonary Medicine at University Hospitals, Rain-
developed
explanation
cluding
were used. In general,
METHODS fibrosis
tailed
ters
reported.
PATIENTS
All patients and their immediate ilies and/or guardians were given
on
of iothal-
(Conray
60%; Mallin-
ckrodt, St Louis). When branches of a yessel were seen to overlie the midline and run cephalad and/or caudad, an oblique
view
was obtained
tionship
to establish
to the spinal
Embolization
their
rela-
circulation.
was
directed
vessel(s) considered source of hemoptysis
toward
the
the most likely on the basis of the
following criteria: 1 . Vessels having a diameter greater than 2.5 mm as measured on cut-film
giogmams
were
defined
as enlarged
an-
and
suspect.
2. Localizing symptoms gling or a “funny feeling”
such as gumin one lung
confined the embolization vessels feeding that lung.
to all enlarged This was the
situation in seven 3. In the other
an attempt
patients. patients
made to embolize all significantly larged vessels feeding both lungs. ly the
right
lung
vessels
were
was
enUsual-
larger
and
were embolized first (2). 4. Vessels smaller than 2.5 mm in diameter or vessels in either lung in which catheter
placement
was
insecure
were
not
401
embolized 5.
regardless
Presence
was
a relative
zation
the
exception).
spinal
zation
and
to emboliwas
taken
before
at appropriate
Vessels
circulation
Care
circulation
embolization
to
emboli-
times
during
procedure.
were
59O-Mm (Ivalon;
artery
contraindication
(one
identify
of symptoms.
of spinal
first
particles Pacific
embolized
with
of polyvinyl alcohol Medical Industries,
250foam La
Mesa, Calif) until flow markedly slowed. The Ivalon was followed by increasing sizes of absorbable gelatin pledgets (Gelfoam; Upjohn, Kalamazoo, Mich) from 1mm3 cubes to 2 X 2 X 20-mm strips, until forward
flow
was
pletely
stopped.
turco
coil (Cook)
minimal
of a large vessel Ivalon /Gelfoam chial trunk strips
but
not
com-
two patients a Gianwas seated at the origin
In
after completion embolization.
One
of the bron-
artery arising from a thyrocervical was embolized only with Gelfoam because of precarious catheter posiA postembolization angiogram was
tion.
always obtained. hemoptysis and
The criteria for control a successful embolization were hemoptysis that (a) was nonexistent or minimal (returning to the patient’s (b) allowed full and drainage, and (c) did
baseline),
postural with
was terminated whenevof injected contrast mate-
exceeded
350
angiography later until being
mL.
In
hemoptysis
controlled
repeat
was
for the
reevaluation
were when
they
early
bleeding
(n 5) and blood cells
in our were
failed
with identical
study. to locate
red
procedures
in the
procedure,
embolization
attempted these
This
procedure
for repeat angiograsame criteria enu-
to those of the primary Ancillary methods such as bronchoscopy tium-99m-labeled
48 hours,
performed.
primary
and
hours If after
than
as a distinct
results. The rationale phy was based on the
but
instances,
reoccurred
for more
embolization
counted
menated
those
was continued 24-48 completion, as described.
significant
was
effective not interfere
life-style.
Angiography em the quantity nial
of
to help
techne(n 2),
experience, abandoned
define
the site
of
hemoptysis.
Figure
1. Patient
sented
with
1. Angiograms of a 14-year-old boy with cystic fibrosis who initially premassive hemoptysis, at times more than 1,000 mL/d. All hemorrhages were accompanied by a right-side “gurgling.” (a) A large right bronchial artery was found and embolized with polyvinyl alcohol foam (2SO-59O-Mm particles) and absorbable gelatin strips on July 23, 1982. (b) Postembolization angiogram. The patient bled again in September 1982. The original vessel had become partially recanalized and was reembolized on September 13, 1982. That evening, the patient bled massively. Panangiography demonstrated an aberrant bronchial artery arising from the right thyrocervical trunk (c). This artery was embolized with a solitary pledget of absorbable gelatin (d). Reembolization of the same branches was required 6 months later, in March 1983. Bleeding was controlled by these procedures until the patient died in November 1984. Note extravasion of contrast material.
RESULTS In 19 of the 20 patients, hemoptysis was successfully treated with embolization of one or more bronchial amteries. Significant hemoptysis was controlled in each, although four patients died of respiratory failure and! or pulmonary infection during the follow-up period. Usual therapy, including antibiotics and sometimes transfusions,
was
patients underwent tion procedure. group died 18, spectively, after eight surviving lowed up for a with a range of None of the
402
.
Radiology
continued.
one
Eleven
embolizaThree patients in this 23, and 34 months, meembolization. The patients were folmean of 37 months, 17-55 months. six patients who me-
quired
two distinct embolization prodied during the follow-up which covered a range of 17-
cedures period, 54
the
months
(mean,
initial
raphy
and
formed ter the months). of
months
embolizations
study, significant
after
Repeat
angiog-
were
from 3 days initial study Measured
giogmaphic free
36
procedure).
per-
to 35 months (mean, 9.5 from the last the
patients
hemoptysis
tion
procedure. procedures
Repeat were
for
in
16-
after
12
(Fig
pro-
1) and
examination.
he
underwent
three
embolization after
not
did
giographic
at
9
the
undergo original
an-
Although angiogmaphic
this patient underwent only twice, at 16 and the initial angiographic
24
study.
A total at
2)
during
months
the
primary
1 (Fig
embolization
procedures,
1)
the
patient
and 22 months in patient 4. The sumviving patient was followed up for 62 months, dating from his initial pmocedume and 40 months, dating from his last procedure (Table 1). Note that
embolizaperformed
8 months
cedure
patient
an-
were
49 months (mean, 26 months). One of the two patients (patient who required three embolization procedures died 26 months after initial
af-
and
2
the
20
of 63 arteries
patients,
and
35
were
found
were
success-
May
in
1990
a-
p
#{182}. *,
I:’’
4 a.
d. Figure
Patient
b.
c.
e.
f.
Angiognams of man with cystic fibrosis who presented at age 22 years with a 5-year history of slowly increasing hemoptysis. His job required lifting of heavy objects, which precipitated episodes of hemoptysis. (a, b) Selective anteropostenior (a) and oblique (b) angiograms obtained in August 1984 show a large bronchial artery feeding the right lung. A midline branch vessel (arrows) was thought to represent a spinal artery. Embolization was not attempted at this time because conservative treatment had not been maximized. (c) A second, smaller bronchial artery feeding both lungs was also not embolized because catheter seating was marginal. The patient’s hemoptysis increased during the following year, preventing him from working. (d) Repeat angiogram demonstrates placement of catheter distal to the spinal artery branch seen on a and b. Careful embolization with polyvinyl alcohol foam (250-590-sam particles) and absorbable gelatm pledgets yielded nearly total occlusion of the artery (e). The presumed spinal artery is more cleanly visible (arrows) than in a and b. Hemoptysis was controlled for 9 months, allowing the patient to return to full activity. He bled again in September and October 1986. (f) A repeat angiogram obtained in October showed necanalization of the embolized antery. (g) The mecanalized vessel was reembolized with polyvinyl alcohol foam and absorbable gelatin pledgets. (h) The smaller bronchial artery was embolized during the same procedure. Hemoptysis was controlled. 2.
12.
fully embolized. Table 2 gives the distribution of the number of bronchial arteries found per patient. When reembolization was required, mecanalization had occurred in a previously embolized vessel(s). Signifi-
cant control One marginal
Volume
hemoptysis
was
brought
under
in each patient. patient considered to have criteria for embolization 175
Number
#{149}
2
g.
h.
Radiology
403
#{149}
underwent
angiogmaphy
undergo nal
but
embolization
artery
was
the
from
found
largest
Conservative
did
because that
not a spi-
originated
bronchial
artery.
treatment
finally
con-
trolled his hemoptysis, with no major recurrences during 32 months of follow-up. Twenty-eight bronchial arteries found angiogmaphically were not embolized for the following reasons: (a) The artery was a small vessel feeding the left lung only (n 8), (b) the amtery was a small vessel feeding the right lung only (i = 4), (c) a spinal artery arose from a bronchial artery (n 6), and (d) the artery was in a dangerous location (small head and neck vessels) (n 10). One or more aberrant bronchial amtenies were detected in seven of 20 patients (35%) (Fig 1). Fifteen aberrant arteries these seven patients originated from the following sites: (a) thyrocervical trunk (n 9), (b) internal mammary artery (n 4), (c) costocervical trunk (n 1), and (d) lower intercostal artery (n 1).
A spinal vessel artery of the
artery
that also circulation 20 patients
Only
one
spinal
branching
from
a
supplied bronchial was present in 11 (55%) (Fig 2a, 2b). artery
demonstrated large yessels feeding the lung, which were embolized. Massive hemoptysis greater than 500 mL recurred within 24 hours in one patient, prompting angiography
immediate tion
study
foam
controlled
other
patient,
hemoptysis.
restudy. rocervical
trunk
vessel
this
series
In all
but
two
patients,
developed
often
after
requiring
the
of
parentemal
pain.
is strikingly
in previous
occurred
Except Fellows
different
predominated
moptysis
for bronchial vessels seven of 20 patients
Radiology
#{149}
within
54 53 23 (died) 52 42
11/19/F 12/22/M
1 2
41 40
2
28
2
26
15/6/F 16/32/M
0 1
26 22
17/33/M
1
22
18/23/F 19/43/M 20/l6/M
1 2 1
7 17 17
Arteries Arteries)
old
24
of Number of Bronchial Identified (20 Patients, 63
In
our
other
prima-
(1,2,4-7). bleeding
often
stops
in
with
to the 2% reported in autopsy studies (10). Although new anastomotic channels may develop along pleural surfaces as the result of inflammatory processes, we found no bronchial an-
cystic
conservative
therapy small
(8). However, relatively amounts of hemoptysis cause disastrous decreases in pulmonary function
and
infections.
decreased crease the Postural
and the
of
increases effective
penicillins hemostasis and risk of hemoptysis
semisynthetic
tate
activity Some
drainage
may
in
can cause also in(3,9).
also
precipi-
noted may
We
function, Aberrant
hemoptysis this
the embolizain this study
suggested in
preserving
life-style, origins,
was
pulmonary and
life.
often
multiple,
were (35%)
found in as opposed
by
process
fibrosis
work
in passing
consider
criteria
important
have
embolization,
mentioned
(1,2,5,7).
tion
investigators
nonmassive
require
usually
enlarged
tory
several
that
fed by lateral chest wall vessels (5,7,11). Instead, all aberrant vessels reached the lung through the central vascular pedicle, suggesting these were usual pathways that were mereteries
ly
hemoptysis. While
2 6 6 2 2 1 1
was
which
experience,
fibrosis
of
in which
embolization in
No. of Patients
1 2 3 4 5 6 7
from
series
my diseases
404
of
for the study populations et al, this cohort of 20 pa-
but
initial
2 2 1 1 1
DISCUSSION
artery
patients,
55
6/32/M 7/20/M 8/26/F 9/28/F 10/19/M
No. of Bronchial Arteries Present
performed
two
62
I
embolization.
bronchial
other
3
5/14/M
Table 2 Distribution
was
hemoptysis
giogmaphically after embolization. In one patient, a bronchial artery was embolized distal to a spinal artery branch identified on the diagnostic angiogram (Fig 2). No spinal cord injury was found, although it was actively sought by means of physical examination. Seven patients required panangiogmaphy. This included aortography of the descending thoracic aorta distal to the aortic arch, bilateral subclavian angiography, and bilateral intemnal mammary angiogmaphy. Five patients underwent panangiogmaphy as part of the initial angiogmaphic evaluation. In these five patients the bronchial circulation arising from the descending thomacic aorta was small and thought to be insufficient to explain significant hemoptysis. Vessels feeding the lungs from all identifiable sources were embolized, and in each instance control of hethe
fever
Minimal
those
4/24/M
for
Postembolization
tients
In
substantial embolization,
narcotics
26(died) 34 (died) 18(died)
Note-Patient 15 did not undergo embolization because a spinal artery branched from the main bronchial artery. The first angiographic procedure in patient 12 was diagnostic angiography; the second and third procedures included embolization.
adminisalleviation. The discomfort was located in the chest, back, or neck and lasted 2-7 days. Dysphagia often accompanied tration
3 1 1
13/14/F
a Gian-
up. pain
Follow-up Time (mo)
No. of Embolizations
14/28/F
complication when
to properly seat in a bronchial artery. The coil was maneuvemed into the right internal iliac artery, where it has created no problems after almost 4 years of follow-
a spinal
resulted.
Ivalon
failed
of
an-
thy-
hemoptysis
occurred
coil
turco
and
1/14/M 2/26/M 3/12/F
the
a right
with
Gelfoam controlled in this patient. The only significant
always
seen
of
and
hours
was
In
remained prompting
Embolization
blood
branch
Patient/ Age (y)/ Sex
Emboliza-
hemoptysis for 96 hours,
unchanged
in
of Embolizations
Follow-up
a large
from a bronchial artery per patient. All of those spinal arteries branched from might-side bronchial arteries (n = 8) or right-side head and neck yessels (n = 3). In seven patients, the spinal artery was identified branching from a small bronchial artery that was not embolized. In three patients,
artery
1).
1
Number
bronchial vessel from the might thyrocerviwith one pledget of Gel-
of
branching cal trunk
common.
branched
(Fig
Table
the
chronic
inflamma-
accompanying
(1).
A
rich
exists
cystic
anastomotic
connecting tinal structures to the ies (1,4,6,11,12). These
net-
many bronchial channels
mediasartemin-
terlace with spinal, head, and neck vessels. Spinal arteries arising as branches from bronchial arteries were present in 1 1 of 20 patients (55%), which is more than 10 times the prevalence found
in
other
series
(13).
This
May
fact
1990
further supports the anastomotic theory, which explains our abundance of aberrant bronchial arteries. In this series the bronchial artery component of four arteries with a spinal
branch
was
complications. stances,
embolic
ed proximal nal
occluded
in-
were
inject-
origin
since
was
three
particles
to the
circulation,
artery
without
In at least
of the
the
spi-
spinal
identifiable
only
at post-
gesics) days).
and
long-lasting
Adequate
(up
to 5-7
analgesia,
in addi-
tion to providing relief from pain, a!lowed prompt return to effective postuna! drainage. Many patients develop recurrent hemoptysis months to years later. Repeat embolization of the original yessels controls the problem. This suggests that Ivalon is a long-duration but
not
permanent
occluding
and that necanalization of vessels occluded by Ivalon may occur (25). It is unknown if a proximal “penmanent” occluden could decrease the
must be incomplete because each bronchial artery was embolized until little flow remained. More likely, the
Proximal embolization was performed with Gelfoam because that material is easy to use. However, was thought that a nonpermanent cluder could organize and fibrose placed proximal to a truly permanent distal embolizing material. When became apparent that permanent
Ivalon particle size (250-590 zm) played a role. Neuronadiologists often use particles larger than 250 m to embolize spinal arteriovenous malformations
(15,16).
This
particle
size effectively occludes the malformation but is too large to allow partides to enter the small spinal feeders, sparing the spinal cord from injury. Complications have occurred when small
particles
as alcohol used
on liquid
on bucrylate
(4,5,17,18).
We
Although
such
were
believe
these agents should whenever possible risk.
agents
cement that
be avoided to minimize
every
this
attempt
should
be made to not embolize the spinal circulation, the presence of a spinal artery branch is not an absolute contraindication
to embolization
Clinical
history
hemoptysis
may
in cystic
in one sels
help
the
In
a prior
to be
located
embolization
feeding
(3).
experienced feeling”
believed
lung,
locate
fibrosis
seven patients who gurgling on “unusual to hemoptysis
(5-7).
orifices
suspect
lung
always
or swamped
embolization
developed
moptysis
Number
#{149}
may
not
fibrosis.
2
11.
12.
ocif 14.
ex-
the meocwere of colandif-
16.
17.
is an
he-
However,
18.
zation
4.
5.
KE, Taik Khaw H.
in cystic
K, Schuster
Bronchial
fibrosis:
artery
technique
S,
Uflacker
20.
R, Kaemmerer
Picon
A,
of bronchial
ogy 1977; 122:33-37. Wholey MH, Chamomro
Miller
WH.
arteries. HA,
Bronchial
PD, et al.
Radiol-
Rao G, Ford
artery
22.
of massive
hemoptysis
of intercostal
1980;
137:617-620.
Ferris
EJ.
arteries.
Pulmonary
by emboliRadiology
hemorrhage:
vascu-
therapy.
Rabkin JE, Astafjev VI, Gothman LN, Grigonjev YG. Tnanscatheter embolization in the management of pulmonary hemorrhage. Radiology 1987; 163:361-365. Winzelberg GG, Wholey MH. Scintigraphic detection of pulmonary hemonnhage using Tc-99m-sulfur colloid. Clin Winzelberg
GG,
Wholey
MH,
Sachs
M.
Scintigraphic localization of pulmonary bleeding using technetium Tc-99m sulfur colloid: a preliminary report. Radiology 1982; 143:757-762.
and
Bronchial artery embolization in the management of hemoptysis: technical aspects and long-term results. Radiology 1985; 157:637-644. Remy J, Annaud A, Fardou H, Giraud R, Voisin C. Treatment of hemoptysis by
WB,
140:249.
lan evaluation and interventional Chest 1981; 80:710-714.
emboli-
long-term results. J Pediatr 1979; 95:959963. Schuster SR, Fellows KE. Management of major hemoptysis in patients with cystic fibrosis. J Pediatn Sung 1977; 12:889-896. Holsclaw DS, Grand RJ, Shwachman H. Massive hemoptysis in cystic fibrosis. Pediatr 1970; 76:829-838.
embolization 6.
1981;
J, Cosgnove R, Melanson 0, Ethier R. Spinal arteriovenous malformations: advances in therapeutic embolization. Radiology 1986; 163-169. Kandjiev V, Symeonov A, Chankov I. Etiology, pathogenesis and prevention of spinal cord lesions in selective angiography of the bronchial and intercostal artenies. Radiology 1974; 112:81-83. Ivanick MJ, Thorwarth W, Donohue J, Mandell V, Delaney 0, Jaques PF. Infanction of the left main-stem bronchus: a complication of bronchial artery embolization. AJR 1983; 141:535-537. Vujic I, Pyle R, Parker E, Mithoefen J.
NucI Med 1981; 6:537-540.
Fellows Shwachman
3.
Radiology
Theron
zation 19.
TF, Doenshuk CF. State of fibrosis. Am Rev Respir Dis
1976; 113:833-878. Cauldwell EW, Siekert RG, Linninger RE, Anson BJ. The bronchial arteries: an anatomic study of 150 human cadavers. Sung Gynecol Obstet 1948; 86:395-412. Vujic I, Pyle R, Hungenford GD, Griffith CN. Angiognaphy and therapeutic blockade in the control of hemoptysis. Radiology 1982; 143:19-23. Vujic I. Reply to “Control of massive hemoptysis by embolization of intercostal arteries” (letter). Radiology 1981; 140:250. Kadir 5, Kaufman SL, Barth KH, White RI. Clinical applications of embolotherapy. In: Selected techniques in interventional radiology. Philadelphia: Saunders, 1982; 73-75. Fellows KE. Control of massive hemoptysis by embolization of intercostal arteries
Control
References 1.
RE, Boat
(letter). 15.
21.
7.
175
in cystic
10.
13.
it dis-
Wood
the art: cystic
it
all significant vessels must be occluded for effective hemostasis. This procedure necessitates detailed know!edge of vascular anatomy because anatomic variability is much higher than previously believed and spinal artery branches are present in at least half the patients. U
sig-
nificant pain, usually accompanied by dysphagia and fever. Other investigators have tended to minimize the amount and duration of discomfort (1,5). In our experience the pain was often severe (requiring narcotic anal-
Volume
occludens
ist and that retaining access to vessel was important to control bleeding, proximal permanent cluders such as Gianturco coils no longer used. Embolization lateral vessels feeding bronchial teries could prove exceptionally ficult. Bronchial artery embolization effective method for controlling
2.
the
bronchoscopes, making localization impossible (1,6,7). Radionuclide scans with Tc-99m-labeled red blood cells in two massively bleeding patients (more than 1,000 mL) were also futile (21-24). Seventeen of 19 patients who underwent
particulate
hemoptysis.
of all yes-
stopped the bleeding. In contrast to the experience in most other series, bronchoscopy-even by very experienced observers-was useless (19,20). Lange volumes of blood filled all bronchial
tal
of recurrent
177:825-828.
9.
agent
embolization angiography. Preferential flow of blood, carrying partides toward a faster-flowing bronchial artery, may have spared the spinal circulation (14). This explanation
frequency
8.
by bronchial artery embolization. Radiology 1983; 146:627-634. Stern RC, Wood RE, Boat TF, Matthews LW, Tucker AS, Doershuk CF. Treatment and prognosis of massive hemoptysis in cystic fibrosis. Am Rev Respir Dis 1978;
23.
Winzelberg GG, Wholey MH, Janmolowski CA, Sachs M, Weinberg JH. Patients with hemoptysis examined by TC99m sulfur colloid and Tc-99m-labeled red blood cells: a preliminary appraisal.
24.
Haponik
Radiology ER.
pulmonary 25.
1984;
153:523-526.
ER, Rothfeld B, Britt Radionuclide localization
hemorrhage.
EJ, Bleecker of massive
Chest
1984;
86:208-212. Tomashefski JF Jr. Cohen AM, CF. Long-term histopathologic
Doershuk follow-up of bronchial arteries after therapeutic embolization with polyvinyl alcohol (Ivalon) in patients with cystic fibrosis. Hum Pathol 1988; 19:555-561.
emboli-
zation for massive hemoptysis. JAMA 1976; 236:2501-2504. Uflacker R, Kaemmerer A, Neves C, Picon PD. Management of massive hemoptysis
Radiology
405
#{149}
M. Cohen,
MD
#{149} Carl
Bronchial Hemoptysis
F. Doershuk,
(55%).
The
#{149}Robert
MD
C. Stern,
Artery Embolizatlon In Cystic Fibrosis’
Severe hemoptysis in cystic fibrosis can be life-threatening because of acute blood loss or because it interferes with sustaining physical therapy of the chest. Hemoptysis was controlled in 19 of 20 cystic fibrosis patients by means of embolization with a combination of 250-590-aim particles of polyvinyl alcohol foam and absorbable gelatin pledgets. Repeat embolization was needed in eight patients to achieve or maintam effective hemostasis. One or more aberrant bronchial arteries were found in seven patients (35%), and a spinal artery branching from a vessel that also supplied bronchial circulation was present in 11 patients
MD
data
suggest
that
although applicability
embolization has a wider than previously reported, care should be taken during angiography to evaluate the bronchial circulation because aberrant bronchial vessels and spinal arteries arising from bronchial circulation are common.
S
hemoptysis
EVERE
to Control
threatens
the
life and well-being of patients with cystic fibrosis. Although consenvative treatment effectively controls minor bleeding, bronchial artery embolization or surgery has often been required to stop massive hemoptysis (1-3). This report examines our experience in 20 cystic fibrosis patients who underwent bronchial artery embolization
to control
data suggest en applicability
hemoptysis.
embolization in cystic
previously
The
has a widfibrosis than
AND
Approximately
425 cystic
pa-
20 cystic
fibrosis
patients
that warranted
giogmaphic
evaluation.
Nineteen
patients
were
treated
with
of the
20
embolization
the patients at their first angiographic procedure was 23 years (range, 6-43
Radiology
procedures. lized more Patients
Some arteries were embothan once. were selected for angiography
after evaluation by pulmonary (C.F.D. and/or R.C.S.) and radiologic (A.M.C.) consultants. In 1 1 instances, hemoptysis was first treated with exhaustive conservative methods, including high-dose intravenous antibiotics. Patients who failed conservative I From
the
Cystic
Rainbow
Babies
Western
Reserve
Fibrosis
Research
Center,
and Children’s Hospital, Case University, Cleveland. Received March 20, 1989; revision requested May 3; revision received January 26, 1990; accepted February 1. Supported in part by National Institutes of Health grant DK27651, the Cystic Fibrosis Foundation, and the United Way Services of Cleveland. Address reprint requests to A.M.C., Chief Diagnostic Radiologist, Cleveland Metropolitan General Hospital, 3395 Scranton Rd. Cleveland, OH 44109. c RSNA, 1990
for angiography na: mL by or
treatment
were
on the basis
considered
of four
femoral
ducer
sheath
nique.
The
(Cook, most
prior
of 5.0-6.5
the
F were
with
crite-
(a) one hemorrhage greater than 300 in 24 hours (“massive”), accompanied ongoing hemoptysis at a lower daily near daily rate; (b) three or more 100mL hemorrhages within 1 week, accompanied by ongoing hemoptysis at a lower daily or near daily rate; (c) chronic or slowly increasing hemoptysis interfering with life-style; or (d) hemoptysis preventing effective postural drainage or home management.
con-
introduced
through
an intro-
the Seldinger
RC 1, and
Bloomington, useful
in-
Informed
to angiogmaphy.
artery
Hi,
tech-
RC 2 models
Ind) were
for cannulating
found
bronchial
vessels. Occasionally, Sos open-end guide wires with steerable cores (USC! Division of C.R. Bard, Billemica, Mass) or 4-F catheor coaxial
3-F Teflon
catheters
the descending
near
thoracic
Most
to-
vessels
(T-4-T-6).
suspected
bronchial
were
evaluated
projection
administration
meglumine
All
angiographi-
in the anteroposterior after
arose
of the carina
containing
cut film
amate
aor-
searched However,
was directed
of hemoptysis.
the level
vessels
cally
(Cook)
patients who localize bleeding to one side of the claiming to experience a “guror “funny feeling” prior to each
circulation
of one or more bronchial arteries during the acquisition of the original diagnostic or subsequent angiogram(s). Mean age of
years) (Table 1). One to three angiograms per patient were obtained, for a total of 36
175:401-405
into
episode
an-
Index terms: Arteries, bronchial, 943.1299 #{149} Arteries, therapeutic blockade, 943.1299 #{149} Fibrosis, cystic, 60.252 #{149} Lung, hemorrhage, 60.252 1990;
obtained
Catheters
could chest, gling”
(12 male)
hemoptysis
was
fama de-
procedure,
and risks.
angiographic evaluation ward one lung in seven
bow Babies and Children’s Hospital, Cleveland. From June 1982 to December 1986,
sent
of the
benefits
ta and intercostal arteries were for bronchial artery circulation.
tients are followed up on a regular basis by the Division of Pediatric Pulmonary Medicine at University Hospitals, Rain-
developed
explanation
cluding
were used. In general,
METHODS fibrosis
tailed
ters
reported.
PATIENTS
All patients and their immediate ilies and/or guardians were given
on
of iothal-
(Conray
60%; Mallin-
ckrodt, St Louis). When branches of a yessel were seen to overlie the midline and run cephalad and/or caudad, an oblique
view
was obtained
tionship
to establish
to the spinal
Embolization
their
rela-
circulation.
was
directed
vessel(s) considered source of hemoptysis
toward
the
the most likely on the basis of the
following criteria: 1 . Vessels having a diameter greater than 2.5 mm as measured on cut-film
giogmams
were
defined
as enlarged
an-
and
suspect.
2. Localizing symptoms gling or a “funny feeling”
such as gumin one lung
confined the embolization vessels feeding that lung.
to all enlarged This was the
situation in seven 3. In the other
an attempt
patients. patients
made to embolize all significantly larged vessels feeding both lungs. ly the
right
lung
vessels
were
was
enUsual-
larger
and
were embolized first (2). 4. Vessels smaller than 2.5 mm in diameter or vessels in either lung in which catheter
placement
was
insecure
were
not
401
embolized 5.
regardless
Presence
was
a relative
zation
the
exception).
spinal
zation
and
to emboliwas
taken
before
at appropriate
Vessels
circulation
Care
circulation
embolization
to
emboli-
times
during
procedure.
were
59O-Mm (Ivalon;
artery
contraindication
(one
identify
of symptoms.
of spinal
first
particles Pacific
embolized
with
of polyvinyl alcohol Medical Industries,
250foam La
Mesa, Calif) until flow markedly slowed. The Ivalon was followed by increasing sizes of absorbable gelatin pledgets (Gelfoam; Upjohn, Kalamazoo, Mich) from 1mm3 cubes to 2 X 2 X 20-mm strips, until forward
flow
was
pletely
stopped.
turco
coil (Cook)
minimal
of a large vessel Ivalon /Gelfoam chial trunk strips
but
not
com-
two patients a Gianwas seated at the origin
In
after completion embolization.
One
of the bron-
artery arising from a thyrocervical was embolized only with Gelfoam because of precarious catheter posiA postembolization angiogram was
tion.
always obtained. hemoptysis and
The criteria for control a successful embolization were hemoptysis that (a) was nonexistent or minimal (returning to the patient’s (b) allowed full and drainage, and (c) did
baseline),
postural with
was terminated whenevof injected contrast mate-
exceeded
350
angiography later until being
mL.
In
hemoptysis
controlled
repeat
was
for the
reevaluation
were when
they
early
bleeding
(n 5) and blood cells
in our were
failed
with identical
study. to locate
red
procedures
in the
procedure,
embolization
attempted these
This
procedure
for repeat angiograsame criteria enu-
to those of the primary Ancillary methods such as bronchoscopy tium-99m-labeled
48 hours,
performed.
primary
and
hours If after
than
as a distinct
results. The rationale phy was based on the
but
instances,
reoccurred
for more
embolization
counted
menated
those
was continued 24-48 completion, as described.
significant
was
effective not interfere
life-style.
Angiography em the quantity nial
of
to help
techne(n 2),
experience, abandoned
define
the site
of
hemoptysis.
Figure
1. Patient
sented
with
1. Angiograms of a 14-year-old boy with cystic fibrosis who initially premassive hemoptysis, at times more than 1,000 mL/d. All hemorrhages were accompanied by a right-side “gurgling.” (a) A large right bronchial artery was found and embolized with polyvinyl alcohol foam (2SO-59O-Mm particles) and absorbable gelatin strips on July 23, 1982. (b) Postembolization angiogram. The patient bled again in September 1982. The original vessel had become partially recanalized and was reembolized on September 13, 1982. That evening, the patient bled massively. Panangiography demonstrated an aberrant bronchial artery arising from the right thyrocervical trunk (c). This artery was embolized with a solitary pledget of absorbable gelatin (d). Reembolization of the same branches was required 6 months later, in March 1983. Bleeding was controlled by these procedures until the patient died in November 1984. Note extravasion of contrast material.
RESULTS In 19 of the 20 patients, hemoptysis was successfully treated with embolization of one or more bronchial amteries. Significant hemoptysis was controlled in each, although four patients died of respiratory failure and! or pulmonary infection during the follow-up period. Usual therapy, including antibiotics and sometimes transfusions,
was
patients underwent tion procedure. group died 18, spectively, after eight surviving lowed up for a with a range of None of the
402
.
Radiology
continued.
one
Eleven
embolizaThree patients in this 23, and 34 months, meembolization. The patients were folmean of 37 months, 17-55 months. six patients who me-
quired
two distinct embolization prodied during the follow-up which covered a range of 17-
cedures period, 54
the
months
(mean,
initial
raphy
and
formed ter the months). of
months
embolizations
study, significant
after
Repeat
angiog-
were
from 3 days initial study Measured
giogmaphic free
36
procedure).
per-
to 35 months (mean, 9.5 from the last the
patients
hemoptysis
tion
procedure. procedures
Repeat were
for
in
16-
after
12
(Fig
pro-
1) and
examination.
he
underwent
three
embolization after
not
did
giographic
at
9
the
undergo original
an-
Although angiogmaphic
this patient underwent only twice, at 16 and the initial angiographic
24
study.
A total at
2)
during
months
the
primary
1 (Fig
embolization
procedures,
1)
the
patient
and 22 months in patient 4. The sumviving patient was followed up for 62 months, dating from his initial pmocedume and 40 months, dating from his last procedure (Table 1). Note that
embolizaperformed
8 months
cedure
patient
an-
were
49 months (mean, 26 months). One of the two patients (patient who required three embolization procedures died 26 months after initial
af-
and
2
the
20
of 63 arteries
patients,
and
35
were
found
were
success-
May
in
1990
a-
p
#{182}. *,
I:’’
4 a.
d. Figure
Patient
b.
c.
e.
f.
Angiognams of man with cystic fibrosis who presented at age 22 years with a 5-year history of slowly increasing hemoptysis. His job required lifting of heavy objects, which precipitated episodes of hemoptysis. (a, b) Selective anteropostenior (a) and oblique (b) angiograms obtained in August 1984 show a large bronchial artery feeding the right lung. A midline branch vessel (arrows) was thought to represent a spinal artery. Embolization was not attempted at this time because conservative treatment had not been maximized. (c) A second, smaller bronchial artery feeding both lungs was also not embolized because catheter seating was marginal. The patient’s hemoptysis increased during the following year, preventing him from working. (d) Repeat angiogram demonstrates placement of catheter distal to the spinal artery branch seen on a and b. Careful embolization with polyvinyl alcohol foam (250-590-sam particles) and absorbable gelatm pledgets yielded nearly total occlusion of the artery (e). The presumed spinal artery is more cleanly visible (arrows) than in a and b. Hemoptysis was controlled for 9 months, allowing the patient to return to full activity. He bled again in September and October 1986. (f) A repeat angiogram obtained in October showed necanalization of the embolized antery. (g) The mecanalized vessel was reembolized with polyvinyl alcohol foam and absorbable gelatin pledgets. (h) The smaller bronchial artery was embolized during the same procedure. Hemoptysis was controlled. 2.
12.
fully embolized. Table 2 gives the distribution of the number of bronchial arteries found per patient. When reembolization was required, mecanalization had occurred in a previously embolized vessel(s). Signifi-
cant control One marginal
Volume
hemoptysis
was
brought
under
in each patient. patient considered to have criteria for embolization 175
Number
#{149}
2
g.
h.
Radiology
403
#{149}
underwent
angiogmaphy
undergo nal
but
embolization
artery
was
the
from
found
largest
Conservative
did
because that
not a spi-
originated
bronchial
artery.
treatment
finally
con-
trolled his hemoptysis, with no major recurrences during 32 months of follow-up. Twenty-eight bronchial arteries found angiogmaphically were not embolized for the following reasons: (a) The artery was a small vessel feeding the left lung only (n 8), (b) the amtery was a small vessel feeding the right lung only (i = 4), (c) a spinal artery arose from a bronchial artery (n 6), and (d) the artery was in a dangerous location (small head and neck vessels) (n 10). One or more aberrant bronchial amtenies were detected in seven of 20 patients (35%) (Fig 1). Fifteen aberrant arteries these seven patients originated from the following sites: (a) thyrocervical trunk (n 9), (b) internal mammary artery (n 4), (c) costocervical trunk (n 1), and (d) lower intercostal artery (n 1).
A spinal vessel artery of the
artery
that also circulation 20 patients
Only
one
spinal
branching
from
a
supplied bronchial was present in 11 (55%) (Fig 2a, 2b). artery
demonstrated large yessels feeding the lung, which were embolized. Massive hemoptysis greater than 500 mL recurred within 24 hours in one patient, prompting angiography
immediate tion
study
foam
controlled
other
patient,
hemoptysis.
restudy. rocervical
trunk
vessel
this
series
In all
but
two
patients,
developed
often
after
requiring
the
of
parentemal
pain.
is strikingly
in previous
occurred
Except Fellows
different
predominated
moptysis
for bronchial vessels seven of 20 patients
Radiology
#{149}
within
54 53 23 (died) 52 42
11/19/F 12/22/M
1 2
41 40
2
28
2
26
15/6/F 16/32/M
0 1
26 22
17/33/M
1
22
18/23/F 19/43/M 20/l6/M
1 2 1
7 17 17
Arteries Arteries)
old
24
of Number of Bronchial Identified (20 Patients, 63
In
our
other
prima-
(1,2,4-7). bleeding
often
stops
in
with
to the 2% reported in autopsy studies (10). Although new anastomotic channels may develop along pleural surfaces as the result of inflammatory processes, we found no bronchial an-
cystic
conservative
therapy small
(8). However, relatively amounts of hemoptysis cause disastrous decreases in pulmonary function
and
infections.
decreased crease the Postural
and the
of
increases effective
penicillins hemostasis and risk of hemoptysis
semisynthetic
tate
activity Some
drainage
may
in
can cause also in(3,9).
also
precipi-
noted may
We
function, Aberrant
hemoptysis this
the embolizain this study
suggested in
preserving
life-style, origins,
was
pulmonary and
life.
often
multiple,
were (35%)
found in as opposed
by
process
fibrosis
work
in passing
consider
criteria
important
have
embolization,
mentioned
(1,2,5,7).
tion
investigators
nonmassive
require
usually
enlarged
tory
several
that
fed by lateral chest wall vessels (5,7,11). Instead, all aberrant vessels reached the lung through the central vascular pedicle, suggesting these were usual pathways that were mereteries
ly
hemoptysis. While
2 6 6 2 2 1 1
was
which
experience,
fibrosis
of
in which
embolization in
No. of Patients
1 2 3 4 5 6 7
from
series
my diseases
404
of
for the study populations et al, this cohort of 20 pa-
but
initial
2 2 1 1 1
DISCUSSION
artery
patients,
55
6/32/M 7/20/M 8/26/F 9/28/F 10/19/M
No. of Bronchial Arteries Present
performed
two
62
I
embolization.
bronchial
other
3
5/14/M
Table 2 Distribution
was
hemoptysis
giogmaphically after embolization. In one patient, a bronchial artery was embolized distal to a spinal artery branch identified on the diagnostic angiogram (Fig 2). No spinal cord injury was found, although it was actively sought by means of physical examination. Seven patients required panangiogmaphy. This included aortography of the descending thoracic aorta distal to the aortic arch, bilateral subclavian angiography, and bilateral intemnal mammary angiogmaphy. Five patients underwent panangiogmaphy as part of the initial angiogmaphic evaluation. In these five patients the bronchial circulation arising from the descending thomacic aorta was small and thought to be insufficient to explain significant hemoptysis. Vessels feeding the lungs from all identifiable sources were embolized, and in each instance control of hethe
fever
Minimal
those
4/24/M
for
Postembolization
tients
In
substantial embolization,
narcotics
26(died) 34 (died) 18(died)
Note-Patient 15 did not undergo embolization because a spinal artery branched from the main bronchial artery. The first angiographic procedure in patient 12 was diagnostic angiography; the second and third procedures included embolization.
adminisalleviation. The discomfort was located in the chest, back, or neck and lasted 2-7 days. Dysphagia often accompanied tration
3 1 1
13/14/F
a Gian-
up. pain
Follow-up Time (mo)
No. of Embolizations
14/28/F
complication when
to properly seat in a bronchial artery. The coil was maneuvemed into the right internal iliac artery, where it has created no problems after almost 4 years of follow-
a spinal
resulted.
Ivalon
failed
of
an-
thy-
hemoptysis
occurred
coil
turco
and
1/14/M 2/26/M 3/12/F
the
a right
with
Gelfoam controlled in this patient. The only significant
always
seen
of
and
hours
was
In
remained prompting
Embolization
blood
branch
Patient/ Age (y)/ Sex
Emboliza-
hemoptysis for 96 hours,
unchanged
in
of Embolizations
Follow-up
a large
from a bronchial artery per patient. All of those spinal arteries branched from might-side bronchial arteries (n = 8) or right-side head and neck yessels (n = 3). In seven patients, the spinal artery was identified branching from a small bronchial artery that was not embolized. In three patients,
artery
1).
1
Number
bronchial vessel from the might thyrocerviwith one pledget of Gel-
of
branching cal trunk
common.
branched
(Fig
Table
the
chronic
inflamma-
accompanying
(1).
A
rich
exists
cystic
anastomotic
connecting tinal structures to the ies (1,4,6,11,12). These
net-
many bronchial channels
mediasartemin-
terlace with spinal, head, and neck vessels. Spinal arteries arising as branches from bronchial arteries were present in 1 1 of 20 patients (55%), which is more than 10 times the prevalence found
in
other
series
(13).
This
May
fact
1990
further supports the anastomotic theory, which explains our abundance of aberrant bronchial arteries. In this series the bronchial artery component of four arteries with a spinal
branch
was
complications. stances,
embolic
ed proximal nal
occluded
in-
were
inject-
origin
since
was
three
particles
to the
circulation,
artery
without
In at least
of the
the
spi-
spinal
identifiable
only
at post-
gesics) days).
and
long-lasting
Adequate
(up
to 5-7
analgesia,
in addi-
tion to providing relief from pain, a!lowed prompt return to effective postuna! drainage. Many patients develop recurrent hemoptysis months to years later. Repeat embolization of the original yessels controls the problem. This suggests that Ivalon is a long-duration but
not
permanent
occluding
and that necanalization of vessels occluded by Ivalon may occur (25). It is unknown if a proximal “penmanent” occluden could decrease the
must be incomplete because each bronchial artery was embolized until little flow remained. More likely, the
Proximal embolization was performed with Gelfoam because that material is easy to use. However, was thought that a nonpermanent cluder could organize and fibrose placed proximal to a truly permanent distal embolizing material. When became apparent that permanent
Ivalon particle size (250-590 zm) played a role. Neuronadiologists often use particles larger than 250 m to embolize spinal arteriovenous malformations
(15,16).
This
particle
size effectively occludes the malformation but is too large to allow partides to enter the small spinal feeders, sparing the spinal cord from injury. Complications have occurred when small
particles
as alcohol used
on liquid
on bucrylate
(4,5,17,18).
We
Although
such
were
believe
these agents should whenever possible risk.
agents
cement that
be avoided to minimize
every
this
attempt
should
be made to not embolize the spinal circulation, the presence of a spinal artery branch is not an absolute contraindication
to embolization
Clinical
history
hemoptysis
may
in cystic
in one sels
help
the
In
a prior
to be
located
embolization
feeding
(3).
experienced feeling”
believed
lung,
locate
fibrosis
seven patients who gurgling on “unusual to hemoptysis
(5-7).
orifices
suspect
lung
always
or swamped
embolization
developed
moptysis
Number
#{149}
may
not
fibrosis.
2
11.
12.
ocif 14.
ex-
the meocwere of colandif-
16.
17.
is an
he-
However,
18.
zation
4.
5.
KE, Taik Khaw H.
in cystic
K, Schuster
Bronchial
fibrosis:
artery
technique
S,
Uflacker
20.
R, Kaemmerer
Picon
A,
of bronchial
ogy 1977; 122:33-37. Wholey MH, Chamomro
Miller
WH.
arteries. HA,
Bronchial
PD, et al.
Radiol-
Rao G, Ford
artery
22.
of massive
hemoptysis
of intercostal
1980;
137:617-620.
Ferris
EJ.
arteries.
Pulmonary
by emboliRadiology
hemorrhage:
vascu-
therapy.
Rabkin JE, Astafjev VI, Gothman LN, Grigonjev YG. Tnanscatheter embolization in the management of pulmonary hemorrhage. Radiology 1987; 163:361-365. Winzelberg GG, Wholey MH. Scintigraphic detection of pulmonary hemonnhage using Tc-99m-sulfur colloid. Clin Winzelberg
GG,
Wholey
MH,
Sachs
M.
Scintigraphic localization of pulmonary bleeding using technetium Tc-99m sulfur colloid: a preliminary report. Radiology 1982; 143:757-762.
and
Bronchial artery embolization in the management of hemoptysis: technical aspects and long-term results. Radiology 1985; 157:637-644. Remy J, Annaud A, Fardou H, Giraud R, Voisin C. Treatment of hemoptysis by
WB,
140:249.
lan evaluation and interventional Chest 1981; 80:710-714.
emboli-
long-term results. J Pediatr 1979; 95:959963. Schuster SR, Fellows KE. Management of major hemoptysis in patients with cystic fibrosis. J Pediatn Sung 1977; 12:889-896. Holsclaw DS, Grand RJ, Shwachman H. Massive hemoptysis in cystic fibrosis. Pediatr 1970; 76:829-838.
embolization 6.
1981;
J, Cosgnove R, Melanson 0, Ethier R. Spinal arteriovenous malformations: advances in therapeutic embolization. Radiology 1986; 163-169. Kandjiev V, Symeonov A, Chankov I. Etiology, pathogenesis and prevention of spinal cord lesions in selective angiography of the bronchial and intercostal artenies. Radiology 1974; 112:81-83. Ivanick MJ, Thorwarth W, Donohue J, Mandell V, Delaney 0, Jaques PF. Infanction of the left main-stem bronchus: a complication of bronchial artery embolization. AJR 1983; 141:535-537. Vujic I, Pyle R, Parker E, Mithoefen J.
NucI Med 1981; 6:537-540.
Fellows Shwachman
3.
Radiology
Theron
zation 19.
TF, Doenshuk CF. State of fibrosis. Am Rev Respir Dis
1976; 113:833-878. Cauldwell EW, Siekert RG, Linninger RE, Anson BJ. The bronchial arteries: an anatomic study of 150 human cadavers. Sung Gynecol Obstet 1948; 86:395-412. Vujic I, Pyle R, Hungenford GD, Griffith CN. Angiognaphy and therapeutic blockade in the control of hemoptysis. Radiology 1982; 143:19-23. Vujic I. Reply to “Control of massive hemoptysis by embolization of intercostal arteries” (letter). Radiology 1981; 140:250. Kadir 5, Kaufman SL, Barth KH, White RI. Clinical applications of embolotherapy. In: Selected techniques in interventional radiology. Philadelphia: Saunders, 1982; 73-75. Fellows KE. Control of massive hemoptysis by embolization of intercostal arteries
Control
References 1.
RE, Boat
(letter). 15.
21.
7.
175
in cystic
10.
13.
it dis-
Wood
the art: cystic
it
all significant vessels must be occluded for effective hemostasis. This procedure necessitates detailed know!edge of vascular anatomy because anatomic variability is much higher than previously believed and spinal artery branches are present in at least half the patients. U
sig-
nificant pain, usually accompanied by dysphagia and fever. Other investigators have tended to minimize the amount and duration of discomfort (1,5). In our experience the pain was often severe (requiring narcotic anal-
Volume
occludens
ist and that retaining access to vessel was important to control bleeding, proximal permanent cluders such as Gianturco coils no longer used. Embolization lateral vessels feeding bronchial teries could prove exceptionally ficult. Bronchial artery embolization effective method for controlling
2.
the
bronchoscopes, making localization impossible (1,6,7). Radionuclide scans with Tc-99m-labeled red blood cells in two massively bleeding patients (more than 1,000 mL) were also futile (21-24). Seventeen of 19 patients who underwent
particulate
hemoptysis.
of all yes-
stopped the bleeding. In contrast to the experience in most other series, bronchoscopy-even by very experienced observers-was useless (19,20). Lange volumes of blood filled all bronchial
tal
of recurrent
177:825-828.
9.
agent
embolization angiography. Preferential flow of blood, carrying partides toward a faster-flowing bronchial artery, may have spared the spinal circulation (14). This explanation
frequency
8.
by bronchial artery embolization. Radiology 1983; 146:627-634. Stern RC, Wood RE, Boat TF, Matthews LW, Tucker AS, Doershuk CF. Treatment and prognosis of massive hemoptysis in cystic fibrosis. Am Rev Respir Dis 1978;
23.
Winzelberg GG, Wholey MH, Janmolowski CA, Sachs M, Weinberg JH. Patients with hemoptysis examined by TC99m sulfur colloid and Tc-99m-labeled red blood cells: a preliminary appraisal.
24.
Haponik
Radiology ER.
pulmonary 25.
1984;
153:523-526.
ER, Rothfeld B, Britt Radionuclide localization
hemorrhage.
EJ, Bleecker of massive
Chest
1984;
86:208-212. Tomashefski JF Jr. Cohen AM, CF. Long-term histopathologic
Doershuk follow-up of bronchial arteries after therapeutic embolization with polyvinyl alcohol (Ivalon) in patients with cystic fibrosis. Hum Pathol 1988; 19:555-561.
emboli-
zation for massive hemoptysis. JAMA 1976; 236:2501-2504. Uflacker R, Kaemmerer A, Neves C, Picon PD. Management of massive hemoptysis
Radiology
405
#{149}
