Cardiovascular Sherry
L. Morris,
MD
#{149} Paul
F Jaques,
Radiology-assisted Subcutaneous Venous Access’
Two
patients
were
lost
#{149} Matthew
A. Mauro,
Placement Infusion Ports
Implantable infusion port devices are generally placed surgically. A technique for radiology-guided placement in adults is described, and the experience with 103 attempted port placements between June 1989 and October 1991 is analyzed. Placements were successful in 102 attempts (99%). Minor procedural difficulties occurred in six patients (5.9%). One major procedural complication (large hematoma) precluded port placement.
MD
to fol-
after uncomplicated placeThere were four (4.0% of 100 patients) minor late complications. Major late complications requiring port removal occurred in 13 (13.0%): five suspected catheter-related infec-
T
HE need
cess
for
has
MD
of Implantable for Long-term
long-term
venous
escalated
tients
ments.
nous therapy or blood drawing (3). These devices have traditionally required surgical placement, but we now report the successful percutaneous placement of 102 infusion ports an interventional radiology suite.
tions,
four
catheter-related
venous
thromboses refractory to thrombolysis, and one each of wound dehiscence, formation of hematoma near the port, extraluminal migration of the catheter, and poor blood return. With a cumulative follow-up of 15,880
able,
days
(43.5
patient-years)
a rate of major
13.6%.
or 0.86%
per
1,000
access
days,
Index terms: Catheters and catheterization, technology a Interventional procedures, complications #{149} Thrombosis, venous, 9*4422 Veins, US, 9*.1298 1992;
(Davol
attempted
were
reviewed
tying
diagnoses,
ment,
date
the
tients)
184:149-151
Department
Carolina
of Radiology,
Hospitals,
Manning
patient
Univer-
Dr,
Chapel Hill, NC 27510. From the 1991 RSNA scientific assembly. Received December 6, 1991; revision requested January 28, 1992; revision received February 18; accepted February 28. Address reprint requests to P.F.J. 2 9* indicates generalized vein and artery involvement. 0 RSNA, 1992
age,
indications
port
in whom
tions
(93.1%)
(40% 36%
ciency
sex,
in
under-
place-
(1.0%).
diagnosis.
placement therapy (10.7%),
Some
were
16 to 77 50.7 years.
two years,
malignancy
18%
acquired and
patients
Primary
port
wound
was clinical
carried service.
out
removal
(elective
of port
or
patency.
chronic
had
hemato-
immunodefi-
hemophilia anemia
in one
more
indications
than for
of blood
or subclavian
venous
ap-
were made in 101 cases (98%), a translumbar inferior vena caval was used in two cases because
of malignant
superior
vena
one
port
chemotherapy in prod-
in our
cavat
experience,
obstruc-
a previously
throm-
subctavian
nous access was achieved conditions with fluoroscopy described
ye-
under sterile by following
technique
(4).
rently, the distal axillary vein with a 7-cm, 21-gauge needle means of direct ultrasound ther a 5- or 7.5-MHz linear
Cur-
is punctured guided by
(US) with transducer.
eiAf-
ter placement of a 6.3-F transition dilator over an 0.018-inch stiff guide wire, the re-
quired
length
from
the
skin
access
copy
by
of the
distal
intravenous
superior
site, using
catheter,
vena
cava
is measured the
with
guide
wire,
to the fluoros-
and
the
di-
lator is then left capped. A suitable site for the port is chosen on the upper chest watt, centimeters
Guiding
below
principles
the
are
bony allow
support for the for a short but
from
the
access, After
in 94
included antineoptastic in 90 (87.4%), antibiotic administration
including
or evaluation
proaches whereas approach
a few
for pa-
placements
in six (5.9%),
(3.0%),
ad-
placement,
removal, referring
requiring
Axiltary
morbidMedical
received
gynecotogic,
other),
syndrome
in three
port
included
routinely
successful
Technique
records
and reviewed and 23 mate
from being
as much Prophy-
involvement by the interventional service was reserved for situa-
premature),
of place-
in substantial removal.
patients
Ages ranged the mean age
logic,
11
103
Two
not
were
tion and bilateral upper-extremity boses due to Trousseau syndrome.
for port route
records were available 101 patients (78 female
Diagnoses From
Further radiology
parenteral
Some ports purpose.
management,
care and suture by staff of the
RI)
Medical
of placement,
those that resulted ity or premature
were After
routine
ment, procedural complications, and subsequent fate of the port. Minor complications were those that did not result in port removal and resulted in minimal patient morbidity. Major complications were
ports. with I
Cranston,
(Figure). for
total
were corrected the procedure.
antibiotics
ministered.
METHODS
Davol,
and
intrave-
AND
Hickman;
were
attempted.
sity of North
tactic
Early
of
(3.9%),
in one (1.0%). more than one
Coagulopathies as possible before
Review of a radiology data base between June 1989 and October 1991 revealed 125 adult patients in whom 128 placements of subcutaneous infusion ports
is comparable to the rates of large surgical series. Radiology-guided placement of infusion ports is safe and may offer advantages over surgical implantation.
Radiology
intermittent
MATERIALS
avail-
complications
nutrition used for
inten-
low-up
in four
ucts
ac-
as new
sive chemotherapeutic regimens for chronic, infectious, and malignant diseases have been formulated. The introduction of tunneled Broviac (1) and Hickman (2) external catheters has aided many patients. More recently available subcutaneous infusion port devices have lower infection rates relative to external catheters and require less intensive maintenance, making them desirable for those pa-
needing
Radiology
clavicle.
to provide
a firm
port during access, gently curved path to the site of venous
reservoir
and avoid administration
mammary
tissue. of intradermat
and
subcutaneous local anesthesia (1 % lidocaine with epinephrine) a 5-cm incision is performed just below the site chosen for the port. This location avoids having the skin incision site overlie the port access dome. A subcutaneous pocket for the port reservoir
skin section. cm
is then incision
by
fashioned using
Overlying
is desirable.
superior minimal
tissue The
size
blunt
thickness of the
pocket
to the dis-
of 0.5-2 is of
149
some
importance;
it should
be just
large
enough to accommodate the port but permit easy closure of the wound. Early in our experience the port was sutured to the deep fascia to prevent port migration or rotation, but we have learned that it is probably unnecessary if the pocket is appropriate in size. The catheter tubing is then tunneled to the venipuncture site and cut to an appropriate length. The incision is closed in two layers
by
using
absorbable
3-0
subcutane-
ous sutures and interrupted nonresorbable 3-0 dermat sutures. The venipuncture site is dilated to 10 F, and the catheter tubing is introduced into the vein through a
10 F peel-away
sheath.
The tip position
is
confirmed with fluoroscopy, and the port device is flushed with 10 mL of heparinized saline solution (100 U/mL) by using a noncoririg Huber needle. Access may be immediate or delayed by at least 7 days (3). Dermal sutures are removed in 10-14 days.
subcutaneous access port (Davol Hickman large venous). The catheter has an outer diameter of 9.6 F and inner diameter of Typical
1.6 mm.
developed RESULTS Placements
were
(99%) dural
of 103 cases. complications
major curred.
procedural A large
ably
successful
in 102
mented
Six minor proce(5.9%) and one
arterial
hemorrhage
after attempted subclavian venipuncture with a 21-gauge needle, prectuded catheter placement in one patient. This major complication occurred before used for venous stantial hematomas
US guidance was access, and no subhave resulted
pneumothorax,
one
small
put-
due to arrhythmias. These resolved spontaneously without sequelae after
25 minutes Two
after
of close
patients
were
uncomplicated
observation. lost
Doppler
after days.
a mean of 88.5 (range, 17-279) Nine devices (9%) were re-
to poor
a trial
to chemother-
blood
of urokinase
return,
developed days after
a periport placement,
and 13 major (13%) tate complications occurred. The minor complications included an incidentally detected spontaneous azygous
moved place
100
ports
were
followed
up
for a cumulative total of 15,880 days, or 43.5 patient-years. Four minor (4%)
vein
tip migration that required
into the transfem-
oral catheter manipulation 75 days after placement. Another complication required that leaking tubing be changed 32 days after placement. This teak was complicated by a 3-cm tip avulsion, with the development of a large groin hematoma, that required transfemoral retrieval. Six patients
150
Radiology
after (range,
a mean 13-197
Thirteen
patients
functioning
(13%)
ports
in place
moved at the completion of therapy after an average of 208 days in place (range, 154-320 days). Sixty-three ports
(63%)
mean
of 187 days
remain
in position
in place
707 days). No deaths able to port placement.
Our
overall
tions was per 1,000
were
rate
complica-
complications Suspected
catheter-related
rates
of 5.0%,
days,
and
sis rates of access,
a
10-
attribut-
of major
13.6%, or 0.88 days of access.
confirmed
after
(range,
or
infection
or 0.31
per
refractory
1,000
venous
of 4.0%, or 0.25 were calculated.
access
thrombo-
per
1,000
days
thrombol-
ysis. This occurred early in our experience. One cachectic patient had wound dehiscence 15 days after port placement because of premature suhire removal (2 days), resulting in device extrusion and toss. One hemo-
which necessitated device removal. Another catheter tip in an obese patient gradually withdrew into the proximal subclavian vein and became extratuminal 197 days after placement. In atl, 13 devices (13%) were re-
remaining
The
in position,
of venous
owing
philiac patient hematoma 181
to follow-up
placements.
without
253 days
with
quiring removal of the device 13-171 days after placement. Three of these patients had acquired immunodeficiency syndrome, one was immuno-
owing
and
died
ing removal of the port. Five patients had suspected or confirmed catheter-related infections re-
ment
five
respectively.
apy, and the last was a hemophiliac patient who developed a periport infected hematoma. One device was surgically revised 15 days after place-
monary arterial air embolism, and three tip malpositionings, which required a transfemoral manipulation into the superior vena cava. Only the patient with the air embolism needed overnight hospitalization, which was
after
of the docu-
by means
compromised
since. Minor procedural complications included one small hematoma, one small
symptom-
thromboses these were
US examinations 14-50 days after port placement. Four of these cases proved refractory to thrombolysis, necessitat-
complication ochematoma, presum-
reflecting
catheter-related
atic deep venous upper extremity;
of 61 days in days) for major
catheter-related complications, including infection (n = 5), refractory deep venous thrombosis (n = 4), wound dehiscence (n = 1), hematoma formation (n = 1), poor blood return (n = 1), and extratuminal migration of the catheter tip (n = 1). Two ports (2%) were removed at patient request
DISCUSSION While
safe,
cutaneous venous Hickman
been
radiology-guided,
placement access devices central
venous
described
sion
port
placed
either
direct
catheter
has
(4), implantable
devices
been
per-
of tong-term such as the
have
historically
by surgeons jugular
infu-
employing
or cephatic
ye-
nous cutdown or percutaneous intraoperative venous access. We report 102 successful neous port
procedures placement
of percutain 100 patients
in an interventionat radiology suite. Follow-up was available on 100 ports, for
a total
of 15,880
tient-years.
ment
Same-
of these
devices
in the interventionat white operating-room may require longer ated costs.
Our similar
overall
days,
or 43.5
or next-day
is often
possible
radiology scheduling delays and
suite,
complication
to or better
paplace-
than
associ-
rates those
are
re-
July 1992
ported in large surgical series (3,5-10). Brothers et at reported local infection or sepsis in 16.4% and catheter or central venous thrombosis in 9.7% of their series of 300 patients (3). Fiftytwo devices (16%) required premature removal, compared with 13% in the present study. Harvey et at reported on a prospective study of surgically implanted ports in 191 patients (5). Their results of 0.4 infectious and 0.3 thrombotic complications per 1,000 catheter-days are almost identical to our results of 0.31 and 0.25 for similar complications. These data imply that
in most
cases,
the
complication
are patient-specific, lated to the method tation.
As our
experience
rates
rather than reof device imptan-
has
grown,
we
have modified our technique to reduce the duration of the procedure and the procedural complication rate. The use of tidocaine with epinephrine aids hemostasis, and excessive bleeding from the incision site has not been a problem. Even patients with severe hemophilia or other coagulopathies have tolerated the procedure after appropriate factor replacement. USdirected axiltary venipuncture has virtually eliminated the risks of pneumothorax and arterial hemorrhage, unlike typical surgical techniques used for venous access. Air embolism was a problem early in our experience. This problem occurred when the catheter section of the port was being passed through the peel-away sheath and the patient failed to control her or his respiration.
We have
eliminated
therefore holding
no longer techniques,
Volume
184
Number
this
problem,
rely on breathby manually
1
and
squeezing shut the lumen of the sheath until a sufficient length of catheter is inserted. Tip malposition (eg, in the internal jugular, brachiocephalic, or azygos veins) may occur during placement and is immediately detected at fluoroscopy. This problem can usually be corrected by means of direct manipulation of the catheter into the distal superior vena cava, but refractory cases may require relocation with techniques for retrieval of the transfemoral catheter.
In terms
of late
complications,
References 1.
2.
3.
4.
the
risk of catheter-related venous thrombosis can be reduced with low-dose warfarin therapy (1-2.5 mg daily) (11). Three of the six patients with suspected catheter-related infection requiring device removal had acquired immunodeficiency syndrome. This finding is not unexpected, since these patients are at higher risk for catheter-related sepsis relative to other immunosuppressed or immunocompetent patients (12). Use of prophylactic antibiotics does not seem warranted, given our low rates of infection without their routine administration. The steady increase in the number of patients referred for port placement from the various clinical services highlights the institutional need for this service to at least supplement the surgical alternative. After port placement, the clinical services have not had any difficulty in gaining access to or otherwise maintaining most ports, and further radiotogic intervention is rarely required. We conclude that radiologyguided, percutaneous placement of subcutaneous infusion ports is safe and efficacious, and may offer advantages over traditional surgical placement. N
5.
6.
7.
8.
9.
10.
11.
12.
Broviac 1W, Cole JJ, Scribner BH. A silicone rubber atrial catheter for prolonged parenteral alimentation. Surg Gynecol Obstet 1973; 136:602-606. Hickman RO, Buckner CD, Clift RA, et al. A modified right atrial catheter for access to the venous system in marrow transplant recipients. Surg Gynecol Obstet 1979; 148: 871-875. Brothers TE, Von Moll LK, NiederhuberJE, et al. Experience with subcutaneous infusion ports in three hundred patients. Surg Gynecol Obstet 1988; 166:295-301. Robertson U, Mauro MA, Jaques PF. Radiologic placement of Hickman catheters. Radiology 1989; 170:1007-1009. Harvey WH, Pick TE, Reed K, et al. A prospective evaluation of the Port-a-cath implantable venous access system in chronically ill adults and children. Surg Gynecol Obstet 1989; 169:495-500. Guenier C, Ferreira J, Pector JC. Prolonged venous access in cancer patients. EurJ Surg Oncol 1989; 15:553-555. Greene FL, Moore W, Strickland G, et al. Comparison of a totally implantable access device for chemotherapy (Port-a-cath) and long-term percutaneous catheterization (Broviac). South Med J 1988; 81:580-583. Morris JB, Occhionere ME, Gauderer MWL, et al. Totally implantable vascular access devices in cystic fibrosis: a four-year experience with fifty-eight patients. J Pediatr 1990; 117:82-85. Reed WP, Newman KA, Wade JC. Choosing an appropriate implantable device for long-term venous access. Eur J Cancer Clin Oncol 1989; 25:1383-1391. Hayward SR. Ledgerwood AM, Lucas CE. The fate of 100 prolonged venous access devices. Am Surg 1990; 56:515-519. Bern MN, Lokich JJ, Wallach SR. et al. Very low doses of warfarin can prevent thrombosis in central venous catheters: a randomized prospective trial. Ann Intern Med 1990; 112:423-428. Skoutelis AT, Murphy RL, MacDonell KB, et al. Indwelling central venous catheter infections in patients with acquired immune deficiency syndrome. J Acquir Immune Defic Syndr 1990; 3:335-341.
Radiology
e
151
L. Morris,
MD
#{149} Paul
F Jaques,
Radiology-assisted Subcutaneous Venous Access’
Two
patients
were
lost
#{149} Matthew
A. Mauro,
Placement Infusion Ports
Implantable infusion port devices are generally placed surgically. A technique for radiology-guided placement in adults is described, and the experience with 103 attempted port placements between June 1989 and October 1991 is analyzed. Placements were successful in 102 attempts (99%). Minor procedural difficulties occurred in six patients (5.9%). One major procedural complication (large hematoma) precluded port placement.
MD
to fol-
after uncomplicated placeThere were four (4.0% of 100 patients) minor late complications. Major late complications requiring port removal occurred in 13 (13.0%): five suspected catheter-related infec-
T
HE need
cess
for
has
MD
of Implantable for Long-term
long-term
venous
escalated
tients
ments.
nous therapy or blood drawing (3). These devices have traditionally required surgical placement, but we now report the successful percutaneous placement of 102 infusion ports an interventional radiology suite.
tions,
four
catheter-related
venous
thromboses refractory to thrombolysis, and one each of wound dehiscence, formation of hematoma near the port, extraluminal migration of the catheter, and poor blood return. With a cumulative follow-up of 15,880
able,
days
(43.5
patient-years)
a rate of major
13.6%.
or 0.86%
per
1,000
access
days,
Index terms: Catheters and catheterization, technology a Interventional procedures, complications #{149} Thrombosis, venous, 9*4422 Veins, US, 9*.1298 1992;
(Davol
attempted
were
reviewed
tying
diagnoses,
ment,
date
the
tients)
184:149-151
Department
Carolina
of Radiology,
Hospitals,
Manning
patient
Univer-
Dr,
Chapel Hill, NC 27510. From the 1991 RSNA scientific assembly. Received December 6, 1991; revision requested January 28, 1992; revision received February 18; accepted February 28. Address reprint requests to P.F.J. 2 9* indicates generalized vein and artery involvement. 0 RSNA, 1992
age,
indications
port
in whom
tions
(93.1%)
(40% 36%
ciency
sex,
in
under-
place-
(1.0%).
diagnosis.
placement therapy (10.7%),
Some
were
16 to 77 50.7 years.
two years,
malignancy
18%
acquired and
patients
Primary
port
wound
was clinical
carried service.
out
removal
(elective
of port
or
patency.
chronic
had
hemato-
immunodefi-
hemophilia anemia
in one
more
indications
than for
of blood
or subclavian
venous
ap-
were made in 101 cases (98%), a translumbar inferior vena caval was used in two cases because
of malignant
superior
vena
one
port
chemotherapy in prod-
in our
cavat
experience,
obstruc-
a previously
throm-
subctavian
nous access was achieved conditions with fluoroscopy described
ye-
under sterile by following
technique
(4).
rently, the distal axillary vein with a 7-cm, 21-gauge needle means of direct ultrasound ther a 5- or 7.5-MHz linear
Cur-
is punctured guided by
(US) with transducer.
eiAf-
ter placement of a 6.3-F transition dilator over an 0.018-inch stiff guide wire, the re-
quired
length
from
the
skin
access
copy
by
of the
distal
intravenous
superior
site, using
catheter,
vena
cava
is measured the
with
guide
wire,
to the fluoros-
and
the
di-
lator is then left capped. A suitable site for the port is chosen on the upper chest watt, centimeters
Guiding
below
principles
the
are
bony allow
support for the for a short but
from
the
access, After
in 94
included antineoptastic in 90 (87.4%), antibiotic administration
including
or evaluation
proaches whereas approach
a few
for pa-
placements
in six (5.9%),
(3.0%),
ad-
placement,
removal, referring
requiring
Axiltary
morbidMedical
received
gynecotogic,
other),
syndrome
in three
port
included
routinely
successful
Technique
records
and reviewed and 23 mate
from being
as much Prophy-
involvement by the interventional service was reserved for situa-
premature),
of place-
in substantial removal.
patients
Ages ranged the mean age
logic,
11
103
Two
not
were
tion and bilateral upper-extremity boses due to Trousseau syndrome.
for port route
records were available 101 patients (78 female
Diagnoses From
Further radiology
parenteral
Some ports purpose.
management,
care and suture by staff of the
RI)
Medical
of placement,
those that resulted ity or premature
were After
routine
ment, procedural complications, and subsequent fate of the port. Minor complications were those that did not result in port removal and resulted in minimal patient morbidity. Major complications were
ports. with I
Cranston,
(Figure). for
total
were corrected the procedure.
antibiotics
ministered.
METHODS
Davol,
and
intrave-
AND
Hickman;
were
attempted.
sity of North
tactic
Early
of
(3.9%),
in one (1.0%). more than one
Coagulopathies as possible before
Review of a radiology data base between June 1989 and October 1991 revealed 125 adult patients in whom 128 placements of subcutaneous infusion ports
is comparable to the rates of large surgical series. Radiology-guided placement of infusion ports is safe and may offer advantages over surgical implantation.
Radiology
intermittent
MATERIALS
avail-
complications
nutrition used for
inten-
low-up
in four
ucts
ac-
as new
sive chemotherapeutic regimens for chronic, infectious, and malignant diseases have been formulated. The introduction of tunneled Broviac (1) and Hickman (2) external catheters has aided many patients. More recently available subcutaneous infusion port devices have lower infection rates relative to external catheters and require less intensive maintenance, making them desirable for those pa-
needing
Radiology
clavicle.
to provide
a firm
port during access, gently curved path to the site of venous
reservoir
and avoid administration
mammary
tissue. of intradermat
and
subcutaneous local anesthesia (1 % lidocaine with epinephrine) a 5-cm incision is performed just below the site chosen for the port. This location avoids having the skin incision site overlie the port access dome. A subcutaneous pocket for the port reservoir
skin section. cm
is then incision
by
fashioned using
Overlying
is desirable.
superior minimal
tissue The
size
blunt
thickness of the
to the dis-
of 0.5-2 is of
149
some
importance;
it should
be just
large
enough to accommodate the port but permit easy closure of the wound. Early in our experience the port was sutured to the deep fascia to prevent port migration or rotation, but we have learned that it is probably unnecessary if the pocket is appropriate in size. The catheter tubing is then tunneled to the venipuncture site and cut to an appropriate length. The incision is closed in two layers
by
using
absorbable
3-0
subcutane-
ous sutures and interrupted nonresorbable 3-0 dermat sutures. The venipuncture site is dilated to 10 F, and the catheter tubing is introduced into the vein through a
10 F peel-away
sheath.
The tip position
is
confirmed with fluoroscopy, and the port device is flushed with 10 mL of heparinized saline solution (100 U/mL) by using a noncoririg Huber needle. Access may be immediate or delayed by at least 7 days (3). Dermal sutures are removed in 10-14 days.
subcutaneous access port (Davol Hickman large venous). The catheter has an outer diameter of 9.6 F and inner diameter of Typical
1.6 mm.
developed RESULTS Placements
were
(99%) dural
of 103 cases. complications
major curred.
procedural A large
ably
successful
in 102
mented
Six minor proce(5.9%) and one
arterial
hemorrhage
after attempted subclavian venipuncture with a 21-gauge needle, prectuded catheter placement in one patient. This major complication occurred before used for venous stantial hematomas
US guidance was access, and no subhave resulted
pneumothorax,
one
small
put-
due to arrhythmias. These resolved spontaneously without sequelae after
25 minutes Two
after
of close
patients
were
uncomplicated
observation. lost
Doppler
after days.
a mean of 88.5 (range, 17-279) Nine devices (9%) were re-
to poor
a trial
to chemother-
blood
of urokinase
return,
developed days after
a periport placement,
and 13 major (13%) tate complications occurred. The minor complications included an incidentally detected spontaneous azygous
moved place
100
ports
were
followed
up
for a cumulative total of 15,880 days, or 43.5 patient-years. Four minor (4%)
vein
tip migration that required
into the transfem-
oral catheter manipulation 75 days after placement. Another complication required that leaking tubing be changed 32 days after placement. This teak was complicated by a 3-cm tip avulsion, with the development of a large groin hematoma, that required transfemoral retrieval. Six patients
150
Radiology
after (range,
a mean 13-197
Thirteen
patients
functioning
(13%)
ports
in place
moved at the completion of therapy after an average of 208 days in place (range, 154-320 days). Sixty-three ports
(63%)
mean
of 187 days
remain
in position
in place
707 days). No deaths able to port placement.
Our
overall
tions was per 1,000
were
rate
complica-
complications Suspected
catheter-related
rates
of 5.0%,
days,
and
sis rates of access,
a
10-
attribut-
of major
13.6%, or 0.88 days of access.
confirmed
after
(range,
or
infection
or 0.31
per
refractory
1,000
venous
of 4.0%, or 0.25 were calculated.
access
thrombo-
per
1,000
days
thrombol-
ysis. This occurred early in our experience. One cachectic patient had wound dehiscence 15 days after port placement because of premature suhire removal (2 days), resulting in device extrusion and toss. One hemo-
which necessitated device removal. Another catheter tip in an obese patient gradually withdrew into the proximal subclavian vein and became extratuminal 197 days after placement. In atl, 13 devices (13%) were re-
remaining
The
in position,
of venous
owing
philiac patient hematoma 181
to follow-up
placements.
without
253 days
with
quiring removal of the device 13-171 days after placement. Three of these patients had acquired immunodeficiency syndrome, one was immuno-
owing
and
died
ing removal of the port. Five patients had suspected or confirmed catheter-related infections re-
ment
five
respectively.
apy, and the last was a hemophiliac patient who developed a periport infected hematoma. One device was surgically revised 15 days after place-
monary arterial air embolism, and three tip malpositionings, which required a transfemoral manipulation into the superior vena cava. Only the patient with the air embolism needed overnight hospitalization, which was
after
of the docu-
by means
compromised
since. Minor procedural complications included one small hematoma, one small
symptom-
thromboses these were
US examinations 14-50 days after port placement. Four of these cases proved refractory to thrombolysis, necessitat-
complication ochematoma, presum-
reflecting
catheter-related
atic deep venous upper extremity;
of 61 days in days) for major
catheter-related complications, including infection (n = 5), refractory deep venous thrombosis (n = 4), wound dehiscence (n = 1), hematoma formation (n = 1), poor blood return (n = 1), and extratuminal migration of the catheter tip (n = 1). Two ports (2%) were removed at patient request
DISCUSSION While
safe,
cutaneous venous Hickman
been
radiology-guided,
placement access devices central
venous
described
sion
port
placed
either
direct
catheter
has
(4), implantable
devices
been
per-
of tong-term such as the
have
historically
by surgeons jugular
infu-
employing
or cephatic
ye-
nous cutdown or percutaneous intraoperative venous access. We report 102 successful neous port
procedures placement
of percutain 100 patients
in an interventionat radiology suite. Follow-up was available on 100 ports, for
a total
of 15,880
tient-years.
ment
Same-
of these
devices
in the interventionat white operating-room may require longer ated costs.
Our similar
overall
days,
or 43.5
or next-day
is often
possible
radiology scheduling delays and
suite,
complication
to or better
paplace-
than
associ-
rates those
are
re-
July 1992
ported in large surgical series (3,5-10). Brothers et at reported local infection or sepsis in 16.4% and catheter or central venous thrombosis in 9.7% of their series of 300 patients (3). Fiftytwo devices (16%) required premature removal, compared with 13% in the present study. Harvey et at reported on a prospective study of surgically implanted ports in 191 patients (5). Their results of 0.4 infectious and 0.3 thrombotic complications per 1,000 catheter-days are almost identical to our results of 0.31 and 0.25 for similar complications. These data imply that
in most
cases,
the
complication
are patient-specific, lated to the method tation.
As our
experience
rates
rather than reof device imptan-
has
grown,
we
have modified our technique to reduce the duration of the procedure and the procedural complication rate. The use of tidocaine with epinephrine aids hemostasis, and excessive bleeding from the incision site has not been a problem. Even patients with severe hemophilia or other coagulopathies have tolerated the procedure after appropriate factor replacement. USdirected axiltary venipuncture has virtually eliminated the risks of pneumothorax and arterial hemorrhage, unlike typical surgical techniques used for venous access. Air embolism was a problem early in our experience. This problem occurred when the catheter section of the port was being passed through the peel-away sheath and the patient failed to control her or his respiration.
We have
eliminated
therefore holding
no longer techniques,
Volume
184
Number
this
problem,
rely on breathby manually
1
and
squeezing shut the lumen of the sheath until a sufficient length of catheter is inserted. Tip malposition (eg, in the internal jugular, brachiocephalic, or azygos veins) may occur during placement and is immediately detected at fluoroscopy. This problem can usually be corrected by means of direct manipulation of the catheter into the distal superior vena cava, but refractory cases may require relocation with techniques for retrieval of the transfemoral catheter.
In terms
of late
complications,
References 1.
2.
3.
4.
the
risk of catheter-related venous thrombosis can be reduced with low-dose warfarin therapy (1-2.5 mg daily) (11). Three of the six patients with suspected catheter-related infection requiring device removal had acquired immunodeficiency syndrome. This finding is not unexpected, since these patients are at higher risk for catheter-related sepsis relative to other immunosuppressed or immunocompetent patients (12). Use of prophylactic antibiotics does not seem warranted, given our low rates of infection without their routine administration. The steady increase in the number of patients referred for port placement from the various clinical services highlights the institutional need for this service to at least supplement the surgical alternative. After port placement, the clinical services have not had any difficulty in gaining access to or otherwise maintaining most ports, and further radiotogic intervention is rarely required. We conclude that radiologyguided, percutaneous placement of subcutaneous infusion ports is safe and efficacious, and may offer advantages over traditional surgical placement. N
5.
6.
7.
8.
9.
10.
11.
12.
Broviac 1W, Cole JJ, Scribner BH. A silicone rubber atrial catheter for prolonged parenteral alimentation. Surg Gynecol Obstet 1973; 136:602-606. Hickman RO, Buckner CD, Clift RA, et al. A modified right atrial catheter for access to the venous system in marrow transplant recipients. Surg Gynecol Obstet 1979; 148: 871-875. Brothers TE, Von Moll LK, NiederhuberJE, et al. Experience with subcutaneous infusion ports in three hundred patients. Surg Gynecol Obstet 1988; 166:295-301. Robertson U, Mauro MA, Jaques PF. Radiologic placement of Hickman catheters. Radiology 1989; 170:1007-1009. Harvey WH, Pick TE, Reed K, et al. A prospective evaluation of the Port-a-cath implantable venous access system in chronically ill adults and children. Surg Gynecol Obstet 1989; 169:495-500. Guenier C, Ferreira J, Pector JC. Prolonged venous access in cancer patients. EurJ Surg Oncol 1989; 15:553-555. Greene FL, Moore W, Strickland G, et al. Comparison of a totally implantable access device for chemotherapy (Port-a-cath) and long-term percutaneous catheterization (Broviac). South Med J 1988; 81:580-583. Morris JB, Occhionere ME, Gauderer MWL, et al. Totally implantable vascular access devices in cystic fibrosis: a four-year experience with fifty-eight patients. J Pediatr 1990; 117:82-85. Reed WP, Newman KA, Wade JC. Choosing an appropriate implantable device for long-term venous access. Eur J Cancer Clin Oncol 1989; 25:1383-1391. Hayward SR. Ledgerwood AM, Lucas CE. The fate of 100 prolonged venous access devices. Am Surg 1990; 56:515-519. Bern MN, Lokich JJ, Wallach SR. et al. Very low doses of warfarin can prevent thrombosis in central venous catheters: a randomized prospective trial. Ann Intern Med 1990; 112:423-428. Skoutelis AT, Murphy RL, MacDonell KB, et al. Indwelling central venous catheter infections in patients with acquired immune deficiency syndrome. J Acquir Immune Defic Syndr 1990; 3:335-341.
Radiology
e
151
