Content uploaded by J Mark Ryan
Author content
All content in this area was uploaded by J Mark Ryan on Oct 12, 2017
Content may be subject to copyright.
Antibiotic Prophylaxis in Interventional
Radiology
J. Mark Ryan, MD, FRCR, FFRRCSI, Barbara M. Ryan, MD, MRCPI, MSc, and Tony P. Smith, MD
Despite several decades of advances in both minimally invasive techniques and antibiotic therapy, infection remains
one of the more common complications of invasive procedures. Interventional radiology (IR) has traditionally been
believed to be associated with lower infection rates than surgery. However, new interventional techniques, as well as
more aggressive therapeutic interventions, have presented new challenges in relation to pharmacological management
of postprocedural infection and pain. The risk of infection associated with IR procedures can never be completely
eliminated, and the reasons for this are manifold, including more virulent organisms, ongoing and newly emerging
antibiotic resistance, increased numbers of immunocompromised patients, and the adoption into everyday interven-
tional practice of more aggressive interventional techniques such as chemoembolization, uterine fibroid embolization,
and complex biliary intervention. Despite the widespread use of prophylactic antibiotics in IR, and the widely held
belief that they are beneficial and are the standard of care, randomized controlled clinical trials have never validated
the use of antibiotics in this setting. As such, an argument could be made not to use antibiotics at all for prophylaxis
in IR. The purpose of this article is to discuss some of the issues relating to the use of prophylactic antibiotics, and
what choice of antibiotics physicians make when they decide to use prophylaxis for IR procedures.
J Vasc Interv Radiol 2004; 15:547–556
Abbreviations: IR ⫽interventional radiology, RF ⫽radiofrequency, TIPS ⫽transjugular intrahepatic portosystemic shunt, UFE ⫽uterine fibroid embolization
SEVERAL authors (1–3) have reported
on the use of prophylactic antibiotics
in interventional radiology (IR) prac-
tice over the past 15 years. Spies noted
that no randomized controlled trials
had ever been performed in the inter-
ventional setting and as such the role
and effectiveness of antibiotic prophy-
laxis in IR had never been scientifically
established (1). Little has changed in
the past fifteen years and the use of
prophylactic antibiotics for many in-
terventional radiology procedures has
become, for better or for worse, the
perceived standard of care.
In 1998, a survey of members of the
Society of Cardiovascular and Inter-
ventional Radiology (SCVIR), now the
Society of Interventional Radiology
(SIR), was performed to investigate
the use of antibiotic prophylaxis in pa-
tients undergoing vascular and non-
vascular procedures. There were 401
respondents to the survey (3). This
survey showed wide variation among
interventionalists in their perception
of indications for antibiotic prophy-
laxis and in their routine clinical use of
prophylactic antibiotics.
The role of antibiotics in preventing
infections in certain surgical settings
appears to be valid (4). However, it is
interesting to note that some surgical
centers are now revisiting their use of
antibiotic prophylaxis for some proce-
dures, and are reassessing their empir-
ical approach (5–7). This is an impor-
tant development for interventional
radiologists considering many antibi-
otic choices in interventional radiol-
ogy have been derived from surgical
experience (1).
An argument can be made that the
risk associated with prophylactic anti-
biotic administration to an individual
patient is small, with the potential
benefit being great. However the
global impact of this empirical ap-
proach is becoming clear with the
worldwide emergence of more viru-
lent and resistant bacterial strains. In
addition, a small number of patients
have died as a result of antibiotic hy-
persensitivity reactions, raising a crit-
ical question that one is reluctant to
ask in these unfortunate situations:
“Was the use of prophylactic antibiot-
ics necessary or useful?”
The purpose of this article is pri-
marily to discuss some of the emerg-
ing issues regarding the use of antibi-
otics for prophylaxis in interventional
radiology practice.
ANTIBIOTIC RESISTANCE
When choosing an antibiotic for
prophylaxis, one needs to consider the
likely source and type of pathogen,
and the prophylaxis should be specif-
ically directed against these organ-
isms. The routine use of very broad
spectrum antibiotics is inappropriate
and indeed promotes further antibi-
otic resistance (8). However despite
From the Division of Vascular-Interventional Radi-
ology (J.M.R., T.P.S.), Duke University Medical Cen-
ter, Durham, NC; Department of Gastroenterology
(B.M.R.), Manchester Royal Infirmary, Manchester,
United Kingdom. Received September 4; accepted
December 17. Address correspondence to J.M.R.,
Division of Vascular-Interventional Radiology,
Duke University Medical Center, Box 3808, Erwin
Road, Durham, NC 27710; E-mail: mryan98454@
yahoo.com
None of the authors have identified a potential con-
flict of interest.
© SIR, 2004
DOI: 10.1097/01.RVI.000024942.58200.5E
Review Article
547
this being widely known, the antibiot-
ics that tend to be used for prophylaxis
are very often broad-spectrum. This is
a serious and ever increasing problem.
Despite the availability of powerful
new antibiotic therapies, morbidity
and mortality resulting from infec-
tious complications have increased in
the past decade, following several de-
cades of decline. This is in large part
related to the increasing frequency
and spectrum of antibiotic resistant in-
fections that have emerged during the
past two decades (9). Contributory
causes to this increase include a com-
bination of changing microbial charac-
teristics, the selective pressure of in-
appropriate and irresponsible antimi-
crobial use, and social and technical
changes that enhance the transmission
of resistant organisms. Bacterial or-
ganisms have an enviable armamen-
tarium to enhance their ability to resist
antibiotic therapy. Resistance is con-
ferred either by mutational change, or
by the acquisition of resistance-encod-
ing genetic material transferred from
another bacterium. The spread of an-
tibiotic resistance genes is causally re-
lated to the misuse and overuse of an-
tibiotics in human health care and also
to the ubiquitous use of antibiotics in
animal feeds. Other factors including
the increased use of invasive devices, a
greater number of susceptible hosts
(ie, immunocompromised patients), and
lapses in infection control practices. In
the hospital environment, widespread
use of antimicrobials in intensive care
units and for immunocompromised pa-
tients has resulted in the selection of
multiple-drug–resistant organisms. Me-
thicillin-resistant staphylococci, vanco-
mycin-resistant enterococci, and extend-
ed-spectrum

-lactamase–producing
gram-negative bacilli are currently a
major problem in nosocomial infections.
Ongoing surveillance of disease pat-
terns have identified an increasing inci-
dence of more seriously ill patients suf-
fering from multiple-drug–resistant
nosocomial infections (10).
Vancomycin has long been consid-
ered the antibiotic of last resort against
serious and multiple-drug–resistant
infections caused by gram-positive
bacteria. However, increasingly, van-
comycin resistance is being encoun-
tered. It was first seen in enterococci
and, more recently, in Staphylococcus
aureus. In the U.S., vancomycin resis-
tant enterococcus is endemic in many
health care institutions, although colo-
nization of healthy individuals is vir-
tually unknown. In contradistinction,
colonization of healthy individuals
and animals is endemic in Europe, re-
lated to the use of avoparcin (a vanco-
mycin-like glycopeptide) in the agri-
cultural industry. The use of
avoparcin was abandoned in Germany
in 1994, and during the following 3
years the prevalence of vancomycin
resistant enterococci in the gastrointes-
tinal tract of healthy individuals fell
from 12% to 3% (11). Unfortunately,
this approach of discontinuing the use
of certain drugs to combat antimicro-
bial resistance is not always success-
ful. Twenty years after ceasing the use
of streptomycin, enterobacteriaceae re-
main resistant to it (12). Strategies un-
der investigation to overcome infec-
tions caused by vancomycin-resistant
strains include glycopeptide deriva-
tives with higher potency than vanco-
mycin; the use of small molecules that
re-sensitize bacteria to antibiotics; and
the introduction of novel non-glyco-
peptide antibiotics. These new agents,
including daptomycin, quinupristin-
dalfopristin, and linezolid, specifically
interfere with protein or peptidogly-
can synthesis or with membrane per-
meability. Most of these agents are still
being developed, and currently repre-
sent the only realistic pharmaceutical
option for treating vancomycin-resis-
tant enterococci (13).
Pandemic emergence of multiple-
drug–resistant bacteria and the spread
of resistance genes must lead to the
stricter application of effective preven-
tion strategies. These include changes
in antibiotic treatment regimens, more
appropriate antibiotic usage including
more specific organism targeting, bet-
ter hygiene measures, more effective
infection prevention, and control of
horizontal nosocomial transmission of
organisms (14). Failure to address the
problem of antibiotic resistance will
mean ever-increasing problems in the
coming decades.
ANTIBIOTIC
HYPERSENSITIVITY
Patients frequently state that they
have a penicillin allergy, thus present-
ing a physician with a therapeutic di-
lemma. Penicillin and

-lactam aller-
gic reactions should be determined by
a careful history. Many patients who
say they have a penicillin allergy in
fact do not. If a patient has a penicillin
allergy, then the clinician should de-
termine whether it is of an anaphylac-
tic or non-anaphylactic variety. Two
percent of the population has some
degree of penicillin hypersensitivity
but most reactions to

-lactams are of
the non-anaphylactic variety and usu-
ally manifest clinically as a mild mac-
ulopapular rash or fever. Uncom-
monly, penicillin allergies manifest as
anaphylactic reactions (eg, broncho-
spasm, laryngospasm, hypotension, or
hives). Hypersensitivity reactions tend
to be similar when re-challenged with
an offending antigen. Patients who
have a questionable penicillin allergy,
or who have had only fever or rash,
may be safely given

-lactam antibiot-
ics without fear of anaphylaxis. If a
patient has had an anaphylactoid re-
action to penicillins, they should never
receive penicillin. In addition, use of
cephalosporins in these patients is
contraindicated because of the 15%
cross-hypersensitivity. Patients with a
documented history of anaphylactic
reactions should receive non–

-lactam
antibiotics, such as vancomycin. Al-
though monobactams (aztreonam)
and carbapenems (imipenem, mero-
penem) are structurally related to

-lactams, they are unrelated in terms
of allergic potential. There is no cross-
reactivity between monobactams or
carbapenems with

-lactams, and
these drugs may be used safely in pa-
tients with anaphylactic reactions to

-lactams (15).
Timing of Prophylaxis
Better understanding regarding the
timing of antibiotic administration be-
gan with the experimental work of
Burke (16) in the early 1960s. Other
studies during the following decade
confirmed what Burke (16) suggested
and led to the current widespread
practice of administering the antibiot-
ics just as the procedure is about to
begin or at least less than 2 hours be-
fore the procedure (17). If antibiotic
prophylaxis is administered more
than 3 hours before the procedure, the
incidence of adverse infectious events
increases fivefold (18). Whereas the
timing of administration of the antibi-
otic has become fairly standard, dura-
tion of administration remains highly
variable, despite the fact that numer-
548 •Antibiotic Prophylaxis for the Interventional Radiologist June 2004 JVIR
ous studies in surgical patients have
demonstrated that a single appropri-
ately timed administration of antibi-
otic prophylaxis is as effective, and in
some cases is more effective, than a
multiple-dose protocol (19–22).
Often the antibiotic is administered
by the IR nurse when the patient ar-
rives in the interventional suite. This is
done because if there is a delay in the
patient’s arrival, the timing of the dose
will remain optimal. A thorough re-
view of the patient’s current medica-
tions should be performed when the
patient is first referred to the IR service
and appropriate antibiotic coverage
should be chosen by a practitioner ex-
perienced in interventional proce-
dures. Difficulties arise when the pa-
tient is already on antibiotics. Several
things need to be checked in this situ-
ation, including the timing of the last
dose of antibiotics and the appropri-
ateness of the antibiotic in relation to
the procedure and the likely patho-
gens. If one is not satisfied with either
of these situations, an additional dose
may be administered. If a procedure is
prolonged, a supplemental dose of an-
tibiotic is often administered. In cer-
tain procedures, typically where a pa-
tient has a blocked viscus, such as
biliary or kidney obstruction, the risk
of puncture site infection is far out-
weighed by the much more serious
risk of postprocedural bacteremia
caused by intravasation of organisms
into the bloodstream. This risk re-
mains present until the organ is ade-
quately drained, and in situations
such as this appropriate antibiotic
treatment continues until satisfactory
drainage of the viscus is achieved.
SPECIFIC IR PROCEDURES
Antibiotic choices for specific IR
procedures are described in the Table.
Uterine Fibroid Embolization
Uterine fibroid embolization (UFE)
for treatment of symptomatic fibroid
disease has become a preferred alter-
native to surgery for many patients.
The demand for UFE has risen expo-
nentially, with women themselves be-
ing the driving force behind the de-
mand for this therapy. As with any
minimally invasive therapy, UFE is
not completely without risk of infec-
tion; however, this appears to be very
uncommon (23,24).
In some reports on UFE, the admin-
istration of preprocedural antibiotics
is routine (25–27). Although evidence
from randomized controlled trials is
lacking, the current antibiotic of choice
for many interventionalists is1gof
cefazolin, based on the most likely
source of pathogens during solid or-
gan embolization, which is contamina-
tion by skin pathogens (staphylococ-
cus or streptococcus). Cefazolin is
inexpensive and has activity against
many clinically important pathogens
including most staphylococci, strepto-
cocci and Escherichia coli. Some inter-
ventionalists do not administer pro-
phylactic antibiotics routinely for UFE,
and some use a multiple-day, multi-
ple-drug approach to prophylaxis. In
the prospective Ontario Uterine Fi-
broid Embolization Trial, antibiotic
prophylaxis use was routine at four
hospitals and reserved only for pa-
tients at higher infection risk at a fur-
ther four hospitals. Routine antibiotic
coverage comprised1gofcefazolin
administered intravenously 1 hour be-
fore the procedure. Patients who were
allergic to penicillin were adminis-
tered 500 mg vancomycin intrave-
nously. There were only two infection-
related hysterectomies required in 570
UFE procedures, one in each group
(23,28).
In contrast to the low infection rate
observed by many interventionalists,
there has been a single study from the
United Kingdom reporting a 17% re-
admission rate (seven out of 42 pa-
tients) for infection after UFE. There
Commonly Used Antibiotics for Prophylaxis in IR
Procedure Suggested Prophylaxis
Angiography Not routine
Angioplasty Not routine
Stent placement Not routine
Peripheral or dialysis access
thrombolysis
Not routine
Uterine fibroid embolization 1 g cefazolin intravenously
Hepatic chemoembolization 1.5–3 g ampicillin/sulbactam intravenously
Splenic embolization 1 g ceftriaxone IV
Renal embolization 1 g ceftriaxone IV
TIPS 3 g ampicillin/sulbactam IV
Tunneled central venous access 1 g cefazolin IV
Endograft 1 g cefazolin IV
Biliary drainage 3 g ampicillin/sulbactam IV
Cholecystostomy 1 g ceftriaxone IV
Gastrostomy 1 g cefazolin IV
Nephrostomy 1 g ceftriaxone IV
RF ablation 1 g cefazolin IV
Solid organ embolization: Many interventional radiologists use prophylaxis
primarily against contamination by skin pathogens. For this cefazolin is satisfactory.
If coverage is required against additional microorganisms, the antibiotic regimen
needs to be adjusted accordingly.
Benefits of prophylaxis has never been validated for IR procedures in randomized
controlled clinical trials.
For high risk cases, ie, Procedure duration ⬎2 hrs; Arterial sheath left in ⬎24 hrs;
Closure device used ⬍1 week-Single Dose 1 g cefazolin IV.
Alternative to cefazolin is Vancomycin1gIVforserious anaphylaxis to penicillin or
cephalosporin allergy; or for patients infected with methicillin resistant
Staphylococcus aureus.
Cefazolin is active against S aureus and S epidermidis.
Ceftriaxone, cefoperazone or cefoxitin are active against E coli, Klebsiella.
Enterobacter, pseudomonas, and clostridia.
Ampicillin/sulbactam has enhanced activity against enterococcus over
cephalosporins.
Cefotetan (or ampicillin, gentamicin and metronidazole) are active against gram-
negative bacteria, anaerobes, enterococci, B fragilis.
Vancomycin, cefazolin, cefoxitin, ceftazidine, cefuroxime need adjustment if
continued in patients with renal dysfunction.
Ryan et al •549
Volume 15 Number 6
was one infection-related hysterec-
tomy. Three of the patients had uri-
nary tract infection, probably related
to bladder catheterization; in three pa-
tients, no source of infection was iden-
tified; and in one patient, there was
spontaneous extrusion of a fibroid.
The authors approach to antibiotic
prophylaxis was co-amoxiclav TID
and metronidazole BID on the day of
the procedure, followed by a further
48 hours administration of the same
antibiotics post-procedurally. They
admitted their patients to the hospital
for 3 days, which is uncommon in the
U.S. (29). It has been suggested that
polypharmacy and prolonged antibi-
otic administration can in fact be det-
rimental when prophylaxis is the goal.
This occurs because normal gram-pos-
itive organisms are destroyed, which
allows harmful gram-negative organ-
isms to proliferate (30). Most studies of
prophylaxis have indicated that sin-
gle-drug use is most effective and that
combinations are unnecessary (22).
There have been anecdotal case re-
ports in the literature of fatal sepsis
after UFE (31). In the case report by
Vashisht (31), the patient did not re-
ceive prophylactic antibiotics and de-
veloped an E coli urinary tract infec-
tion 3 days after the procedure for
which she received antibiotic therapy.
The patient returned 1 week later with
E coli sepsis, to which she succumbed.
In a recent article reporting fatal sepsis
after UFE, again prophylactic antibiot-
ics were not administered (32). The
patient presented 5 days after the pro-
cedure with S pyogenes sepsis.
Whether or not prophylactic antibiotic
administration would have prevented
infection in these cases is unknown.
Options for UFE: (i) no prophylaxis;
(ii) cefazolin 1 g intravenously;
(iii) 1.5–3 g ampicillin/sulbactam
intravenously.
Transjugular Intrahepatic
Portosystemic Shunt
Dravid et al (3) reported that 69% of
respondents to their survey used anti-
biotic prophylaxis in patients under-
going transjugular intrahepatic porto-
systemic shunt (TIPS). An infection
rate of 13% was reported. Patients
who undergo TIPS are usually ill pa-
tients with multisystem disease and
infection in these patients is poorly tol-
erated. Dravid et al (3) concluded that
prophylaxis is necessary for TIPS, and
suggested the use of cefoxitin1gin-
travenously every 6 hours for 48
hours.
Some interventionalists use ceftri-
axone 1 g intravenously once daily for
48 hours. It has superior activity over
cefoxitin against E coli, enterobacter,
gram-negative bacteria, anaerobes and
enterococci, and has the advantage of
a once daily dosing regimen. Ampicil-
lin/sulbactam (Unasyn, Roerig, NY)
gives superior coverage against en-
terococcus and is also a suitable
choice. However there is no evidence
from clinical trials that the use of pro-
phylactic antibiotics in TIPS is
beneficial.
Infectious complications from TIPS
can be divided into two groups: (i)
periprocedural sepsis not involving
infection of the stent, and (ii) infection
of the TIPS stent. Periprocedural sep-
sis, which appears to be uncommon,
has been recognized since the early
days of TIPS (33); however, infection
of the TIPS stent itself has only re-
cently been recognized, reportedly oc-
curring in 1.7%–5.1% of cases (34–37).
Brown et al (38) studied a series of
patients with Enterococcus faecium in-
fection after TIPS. A review of 314 pa-
tients who underwent TIPS at their
institution identified four patients
with enterococcal infection (1.3%). All
four patients had thrombosis of their
TIPS at the time of diagnosis. All were
treated with prolonged courses of in-
travenous antibiotics. Two of four pa-
tients survived and underwent a suc-
cessful repeat TIPS (38).
Sanyal et al (34) were the first to
describe the clinical picture of infec-
tion of the TIPS stent. The diagnosis
was based on the occurrence of fever
with positive blood cultures, and ei-
ther a thrombus or vegetations on the
stent, or persistent bacteremia in a pa-
tient with a TIPS and no other detect-
able source of infection. The clinical
features of endotipsitis included fever,
tender hepatomegaly, hypoxemia,
septic pulmonary emboli, septic
shock, neutrophilia, and subsequent
development of necrotizing fasciitis.
Organisms included oral and enteric
aerobic gram-negative bacteria, and
Candida. Most patients responded to
antibiotic therapy.
DeSimone et al (36) studied 99 TIPS
during an 8-year period, and identi-
fied five patients with no other alter-
native source of bacteremia despite
rigorous evaluation who were pre-
sumed to have TIPS infections. Pa-
tients developed bacteremia a median
of 100 days after TIPS placement
(range, 6–732 days), well beyond the
effective period of antibiotic prophy-
laxis. Bacteremia resolved in all after
treatment with intravenous antibiotics
(36).
Deibert et al (39) studied the effec-
tiveness of a single dose of a second-
generation cephalosporin to prevent
post-TIPS infection. One hundred-five
transjugular interventions were ran-
domized to either receive no antibiotic
treatment (46 interventions) or2gce-
fotiam (56 interventions) given at the
beginning of the procedure. Post-TIPS
infection was defined as an increase in
WBC count (⬎15,000), fever (⬎38.5 de-
grees C), or a positive blood culture.
Patients who did not receive cefotiam
had an infection rate of 20%, versus
14% of patients treated with cefotiam
(39). The difference was not statisti-
cally significant. Some would say that
Deibert et al’s (39) findings argue
against the routine use of prophylactic
antibiotics for TIPS. Some may argue
that cefotiam was a suboptimal antibi-
otic choice in the first place, and an
agent with better coverage against en-
terococcus would have been more ap-
propriate, and would presumably
have had an effect on outcome.
Acute infection related to TIPS
placement appears to be uncommon,
whether or not prophylactic anti-
biotics are of value remains
undetermined.
Options for TIPS: (i) no prophylax-
is; (ii) 1 g ceftriaxone intravenously;
(iii) 1.5–3 g ampicillin/sulbactam
intravenously.
Biliary Interventions
Bile is contaminated in one-third of
patients with malignant biliary ob-
struction and in greater than two-
thirds of patients with benign stricture
or stones (40). Therefore in the setting
of biliary obstruction, the differentia-
tion between prophylaxis and treat-
ment is blurred. The frequency of in-
fectious complication for biliary
drainage procedure has been reported
to be 24%–46% (41,42). Dravid et al (3)
reported in their survey that 89% of
interventionalists routinely use antibi-
otics for biliary drainage. Brody et al
550 •Antibiotic Prophylaxis for the Interventional Radiologist June 2004 JVIR
(43) evaluated the use of routine bile
cultures to determine the risk factors
for bacterial colonization in patients
with biliary obstruction undergoing
percutaneous biliary drainage. During
a 2-year period bile cultures were pro-
spectively obtained in 76 patients un-
dergoing 86 biliary drainage proce-
dures. They found that pyrexia,
previous biliary instrumentation, and
bilioenteric anastomosis were signifi-
cant predictors of a positive bile culture.
Enterococcus species was the most fre-
quent organism isolated, with yeast,
gram-negative aerobic bacilli, and Strep-
tococcus viridans also reported. This
study concluded that preprocedural bile
cultures provide valuable information
for planning antibiotic prophylaxis.
Clark et al (44) performed a pro-
spective study of 480 biliary proce-
dures. The complexity of the proce-
dures ranged from simple tube
injection to de novo biliary drainage
tube placement. All patients received
prophylaxis of 1 g cefotetan intrave-
nously. Forty-two patients developed
elevated white cell count or fever, and
seven (2%) developed overt sepsis.
They found that patients who had pre-
vious biliary intervention were at
greater risk for infection. They sug-
gested that in patients with a recent
history of biliary intervention, that ce-
fotetan alone was inadequate, and that
4 g mezlocillin intravenously should
also be administered. The use of third-
generation cephalosporins (eg, ceftri-
axone 1 g intravenously daily) may be
more effective than second generation
because they have enhanced biliary
excretion and thus reach higher con-
centrations in the bile.
Ampicillin/sulfabactam (Unasyn)
has greater activity against enterococ-
cus and is considered by many to be
the most appropriate choice in the set-
ting of biliary obstruction. Antibiotic
treatment should be continued until
the system is effectively drained.
One study of antibiotic prophylaxis
before endoscopic retrograde cholan-
giopancreatography in patients with
obstructive jaundice concluded that
prophylaxis resulted in fewer cases of
cholangitis and was cost effective com-
pared to a strategy of no prophylaxis
(45). Both the American and European
societies of gastrointestinal endoscopy
currently recommend the use of pre-
ERCP antibiotic prophylaxis in pa-
tients with a history of previous biliary
sepsis, in those with biliary obstruc-
tion, and in patients with pancreatic
pseudocysts.
Hepatolithiasis can lead to liver ab-
scess formation, secondary biliary cir-
rhosis, portal hypertension, and death
from sepsis or hepatic failure. In addi-
tion to clearance of the stones and re-
lief of bile stasis either by surgery or
by IR manipulation, effective antimi-
crobial therapy plays a crucial role in
the treatment of these patients. A
study by Chen et al (46) demonstrated
the presence of bacteria in the bile of
all patients with hepatolithiasis, most
frequently gram-negative bacteria
such as klebsiella, E coli, and Pseudo-
monas sp, and the gram-positive en-
terococcus. Bacteroides were the most
frequently found anaerobes. Ceftazi-
dime and cefoperazone are effective
against pseudomonas. Antibiotics
should be adjusted according to the
results of bacteriologic cultures (46).
Options for biliary intervention: (i)
1 g ceftriaxone intravenously; (ii) 1.5–3
g ampicillin/sulbactam; (iii)1gce-
fotetan intravenously ⫾4 g mezlocil-
lin intravenously; (iv) 1 g ceftazidine
intravenously.
Genitourinary Procedures
Septicemia is less common after
genitourinary intervention than after
biliary procedures. An incidence of
bacteremia of 17% in patients under-
going nephrostomy tube exchange
was reported by Cronan et al (47). The
incidence was similar in patients who
received no periprocedural antibiotics
to those who did. Factors which pre-
dispose to genitourinary tract con-
tamination include advanced age, dia-
betes, indwelling catheter, ureteroin-
testinal conduit, stones, and bacteri-
uria. Therefore in presence of these
factors, the differentiation between
prophylaxis and treatment is blurred.
The incidence of septic shock in the
setting of pyonephrosis has been re-
ported to be as high as 7% (48). Organ-
isms that commonly infect the genito-
urinary tract include E coli, proteus,
klebsiella, and enterococcus. The value
of prophylaxis for urologic procedures
has not been demonstrated in clinical
trials, however patients who have
signs of infection, or who fall into a
high risk group as described above,
should be treated with an appropriate
antibiotic before manipulation. Urine
cultures should be performed and
therapy tailored appropriately.
Options for urologic procedures: (i)
no prophylaxis; (ii) 1 g ceftriaxone in-
travenously; (iii) 1.5–3 g ampicillin/
sulbactam; (iv) 1 g ampicillin ⫹genta-
micin 120 mg intravenously.
Gastrostomy
Fluoroscopically placed gastros-
tomy and gastrojejunostomy tubes un-
commonly cause infectious complica-
tions. When infections do occur they
are usually minor skin infections at the
site of insertion or at the site of t-fas-
tener placement (49). Local treatment
with topical preparations is usually
adequate.
Options for gastrostomy: (i) no pro-
phylaxis; (ii) 1 g cefazolin intravenously.
Chemoembolization
Several studies have shown that he-
patic abscesses are likely to occur after
hepatic chemoembolization in patients
with history of biliary reconstructive
surgery, or biliary-enteric anastomosis
(50–52). An explanation for this is that
biliary reconstructive surgery causes
bile duct ischemia, which is exacer-
bated by chemoembolization as the
peribiliary capillary plexus is supplied
by branches of the hepatic artery. Bile
duct ischemia, combined with the
presence of enteric pathogens (second-
ary to bilio-enteric anastomosis), pro-
motes formation of intrahepatic ab-
scesses. Kim et al reported an overall
incidence of liver abscess of 4.5% in
157 patients who underwent chemo-
embolization. Eighty-six percent of
these patients had previously under-
gone a Whipple procedure, and in-
deed only one patient who had had a
Whipple procedure did not get a
postembolization abscess. All patients
had received cefazolin 1 g intrave-
nously and metronidazole 500 mg in-
travenously 1 hour before the proce-
dure, followed by qid 8 cefazolin and
qid 12 metronidazole postprocedure
until discharge. Oral amoxicillin/cla-
vulanate (or ciprofloxacin in penicil-
lin-sensitive patients), was given for 5
days after discharge.
Song et al (53) reported 2,439 pa-
tients with hepatic tumors who had
undergone a total of 6,255 hepatic che-
moembolization procedures. Fifteen
liver abscesses occurred in 14 patients
Ryan et al •551
Volume 15 Number 6
(0.2%). Thirteen liver abscesses were
successfully treated with parenteral
antibiotics and percutaneous catheter
drainage. Irreversible deterioration of
liver function occurred in two pa-
tients. Statistically they found that a
biliary anomaly (particularly a biliary-
enteric anastomosis) prone to ascend-
ing biliary infection was the most
important risk factor for the develop-
ment of liver abscess after hepatic che-
moembolization (53).
A study by Geschwind et al (54)
followed eight patients who under-
went chemoembolization after biliary
surgery. These patients were divided
into two groups. Four patients were
administered intravenous cephalexin
for prophylaxis, and four were given
bowel preparation and tazobactam/
piperacillin (Zosyn, Lederle). Tazobac-
tam/piperacillin provides extensive
coverage against gram-positive and
gram-negative aerobic and anerobic
organisms. All patients in the first
group developed hepatic abscesses,
which required treatment with percu-
taneous catheter drainage and antibi-
otics. None of the patients in the sec-
ond group developed abscesses (54).
This study is too small to draw any
conclusions about whether a broad-
spectrum antibiotic prophylaxis com-
bined with bowel preparation pro-
vides protection against abscess
formation after chemoembolization in
patients who have a history of biliary
reconstructive surgery. If in fact bowel
preparation is effective in the setting
of previous biliary surgery, there may
be a case for investigating the poten-
tial benefit of selective bowel decon-
tamination with non-absorbed antibi-
otics such as neomycin. Some
operators empirically mix antibiotics
with the embolic materials to be intro-
duced, though the advantages of this
approach remain unproven (55).
The data set regarding the use of
antibiotic prophylaxis; for chemoem-
bolization is small and conflicting. In
the absence of randomized clinical tri-
als, the effectiveness of prophylaxis in
this setting is unproven. It would ap-
pear from several studies that bilioen-
teric anastomosis is a predisposing
factor for abscess formation.
Options for chemoembolization: (i)
no prophylaxis; (ii) 1.5–3 g ampicillin/
sulbactam intravenously; (iii) 1 g cefa-
zolin ⫾500 mg metronidazole intrave-
nously; (iv) 1 g ceftriaxone intravenously.
Radiofrequency Ablation
There have been no randomized
controlled trials on antibiotic use in
patients undergoing radiofrequency
(RF) ablation. However a number of
groups have reported on their experi-
ence and preferred protocols. Gold-
berg and colleagues use cefazolin 1 g
single dose before RF ablation of liver
lesions. Dupuy and colleagues do not
use prophylactic antibiotics unless the
patient has a biliary catheter or stent in
situ, or unless the patient has a biliary-
enteric bypass. In our institution pa-
tients undergoing liver RF are given
ampicillin/sulbactam 1.5 g intrave-
nously before an RF procedure.
There is no consensus on the effec-
tiveness of prophylactic antibiotics for
patients undergoing RF ablation of
liver, lung, adrenal, renal, or other
solid lesions. In the absence of defini-
tive scientific evidence, some practitio-
ners continue to empirically use pro-
phylaxis (56,57).
Options for RF procedures: (i) no
prophylaxis; (ii) 1.5–3 g ampicillin/
sulbactam intravenously; (iii) 1 g cefa-
zolin intravenously.
Angiography, Stent, and Endograft
Placement
Shawker et al (58) performed a pro-
spective study on the incidence of bac-
teremia after angiography, reporting
an incidence of 4%. However this
study found that the phenomenon was
transient and not associated with sep-
sis in any of the affected patients (58).
As a result of this study, along with
years of clinical experience with an-
giography, a consensus has evolved
that prophylaxis is unnecessary for an-
giography. Dravid et al (3) reported
that few respondents to their survey
ever used prophylactic antibiotic for
routine angiographic examinations or
stent placement. In contrast to Shawk-
ers findings, Meyer et al (59) reported
a higher incidence of transient bacte-
remia following arterial intervention.
In a prospective study of 45 patients
undergoing a therapeutic angio-
graphic procedure (all were emboliza-
tions), 32% of patients who received
no antibiotics (n⫽25) developed bac-
teremia, although none developed
clinical sepsis. The organisms isolated
included S epidermidis, streptococcus,
and cornyebacterium, all of which are
normal mucosal or skin flora. By con-
trast, in the group who received anti-
biotic prophylaxis (n⫽20), none had
positive blood cultures (59). This
study concluded that bacteremia was
more common than previously recog-
nized, especially in procedures of ex-
tended duration, such as emboliza-
tion. It is in part as a result of the
findings of this study that many inter-
ventional radiologists give prophy-
laxis targeted against skin pathogens
before performing solid organ emboli-
zation. Meyer et al’s (59) findings were
consistent with the findings of
Shawker et al (58), that bacteremia was
transient, did not appear to predispose
to sepsis, and the authors stopped
short of recommending antibiotic pro-
phylaxis. Given the nature and source
of the organisms isolated, the authors
did stress the critical importance of
scrupulous sterile technique when
performing embolization.
Stent infection is an uncommon but
serious complication of endovascular
treatment. A pig-model of iliac artery
stent infection with S aureus demon-
strated that there was a statistically
significant increase in infection of the
stent/artery complex over the angio-
plastied artery alone. This is likely to
be a result of a combination of factors
including the presence of a foreign
body, and the presence of an inflam-
matory reaction at the site of stent im-
plantation (60). Stent infection has
been reported in the aorta, iliac, renal,
coronary, and subclavian arteries (61–
68), and is usually accompanied by
arteritis with pseudoaneurysm forma-
tion. There are often contributory fac-
tors such as re-puncturing the same
vessel during a short time interval, or
using a vascular sheath that has been
in place for more than 24 hours (69).
Malek et al (62) suggested that pro-
phylactic antibiotics should be admin-
istered if an arterial sheath is left in
overnight after stent placement, or in
patients undergoing multiple endo-
vascular interventions. Infection has
also been associated with the use of
arterial closure devices, especially
where there has been re-intervention
within 1 week (70). One approach is to
administer prophylactic antibiotics be-
fore the second procedure (1 g cefazo-
lin) to patients in whom re-interven-
tion is repeated within 7 days of using
an arterial closure device.
Symptoms of stent infection typi-
552 •Antibiotic Prophylaxis for the Interventional Radiologist June 2004 JVIR
cally present days to weeks after the
intervention. Patients develop fever,
leucocytosis and pain, although some
patients may be asymptomatic. The
low risk of stent infection, despite the
serious outcome associated, means
that at the present time prophylactic
antibiotics are uncommonly used for
routine arterial stent placement. How-
ever, for those deemed to be at high-
risk (ie, reintervention ⬍7 days, pro-
longed indwelling arterial sheath,
prolonged duration of procedure), an-
tibiotics may be given (62).
Prophylactic antibiotics are given
routinely for aortic endograft therapy
at many institutions (71), although
there is no scientific evidence to sup-
port this approach. Cefazolin 1 g intra-
venously given at the commencement
of the procedure gives coverage
against skin pathogens. Prosthetic
graft infection is an uncommon event,
though is associated with a high mor-
tality rate (72,73).
Eliason et al (74) reported an infec-
tion in a patient after coil embolization
of an endoleak. The patient developed
abdominal pain, fever, and leukocyto-
sis. An abdominal CT and indium scan
revealed an infected endovascular
graft. Interventional radiologists
should remain aware that re-interven-
tion, which is required in up to 20% of
patients, is an important source of en-
doluminal graft infection (74).
The use of covered stents in an in-
fected field remains controversial. It is
generally recommended that infected
aneurysms be treated with autografts
or allografts. Kurimoto et al (75) de-
scribed an infected pseudoaneurysm
that was successfully treated by endo-
vascular stent-graft with adjunctive
antibiotic administration.
Options for angiography, stent
placement, endograft: (i) no prophy-
laxis; (ii) 1 g cefazolin intravenously
(for high-risk patients).
Central Venous Access
Placement of central venous cathe-
ters has become an increasing part of
the workload of an interventional ra-
diology service. Patients requiring
central venous access are frequently
immuno-compromised and thus at in-
creased risk of infectious complica-
tions. The infection rate has in fact
been extremely low, comparable to the
infection rates of surgically placed
catheters (76). A standard prophylaxis
is 1 g cefazolin intravenously before
starting the procedure, and the adop-
tion of scrupulous sterile technique
and appropriate presurgical prepara-
tion is mandatory.
A revised version of The Center for
Disease Control Guidelines for the
Prevention of Intravascular Catheter-
related Infections was released in Au-
gust 2002 (77). The emphasis in the
guidelines was on sterile technique,
the use of 2% chlorhexidine skin prep-
aration, avoidance of routine catheter
change as a means of preventing infec-
tion, and the use of antiseptic or anti-
biotic impregnated catheters if infec-
tion rates are high. The guidelines did
not specifically address antibiotic pro-
phylaxis; indeed the benefit of antibi-
otic prophylaxis for central venous ac-
cess remains unproven. Coagulase-
negative staphylococci are the most
common cause of catheter-related in-
fections, followed by enterococci. The
percentage of enterococci resistant to
vancomycin has risen from 0.5% in
1989 to 25.9% in 1999. Migration of
skin organisms into the catheter tract
is the most common route of infection,
with contamination of the catheter
hub contributing to the intraluminal
colonization of long-term catheters.
Recommendations include the use of
catheters with the fewest number of
lumens necessary for management of
the patient, so as to reduce portals for
colonization. For patients requiring in-
termittent long-term venous access, to-
tally implantable devices are more ap-
propriate and more resistant to
contamination. Such devices are now
in use for dialysis access management
in an attempt to reduce infection rates.
Whigham et al (78) investigated the
catheter infection rate in a total of 391
patients, including patients with and
without HIV who had implantable ve-
nous access devices placed by an in-
terventional radiologist. These re-
ported a significant difference in the
infectious complication rate encoun-
tered in HIV-positive patients com-
pared with the general population.
Most infections in oncology patients
with long-term venous access are
caused by gram-positive bacteria. Van
et al (79) looked at the efficacy of ad-
ministering antibiotics before insertion
of a central venous catheter with or
without vancomycin/heparin flush
technique in the first 45 days after in-
sertion of the catheter to prevent
gram-positive catheter-related infec-
tions in oncology patients. They re-
ported that administration of an anti-
biotic before catheter insertion
followed by vancomycin/heparin
flush decreases the number of gram-
positive infections. However, they did
not endorse this strategy for general
use, only in cases of recurrent infec-
tions (79–81).
Management of infections associ-
ated with implantable ports can be dif-
ficult to treat and are potentially fatal,
thus prevention is a priority. Preven-
tive strategies include systemic
periprocedural administration of anti-
biotics in addition to local application
of antimicrobial agents (antibiotics or
antiseptics) at the time of implant.
There is little in the way of scientific
evidence to support such techniques
as lavage of the port pocket although
many interventionalists continue to do
this empirically (82).
Options for central venous access:
(i) no prophylaxis; (ii) 1 g cefazolin
intravenously.
CONCLUSION
The benefits of the use of antibiotic
prophylaxis in IR practice have never
been determined in randomized con-
trolled trials. However, it has become
the perceived standard of care for
many IR procedures. Appropriate pro-
phylaxis for IR procedures requires a
sound knowledge of the likely patho-
gens and procedure-specific infection
issues. Ideally this would lead to the
choice of an agent that was narrow in
its spectrum of activity. Unfortunately
this has not been the tendency, with
the use of more broad-spectrum anti-
biotics becoming increasingly evident.
This contributes to ever-evolving and
increasingly problematic antibiotic re-
sistance of microorganisms (83). The
medical community is already seeing
the result of this in everyday practice,
and it is a problem that is unlikely to
improve in the foreseeable future. One
obvious approach to combat this prob-
lem would be to stop giving prophy-
lactic antibiotics (especially broad-
spectrum ones) for procedures where
they have no proven effectiveness,
which would in fact be most IR and
surgical procedures. However this is
extremely difficult to do when pro-
phylactic antibiotics are considered
Ryan et al •553
Volume 15 Number 6
standard of care for many interven-
tions. It may expose individual physi-
cians who did not give antibiotic pro-
phylaxis to legal challenges if a patient
became infected after a procedure. Ac-
tion by an individual physician is not
likely to be effective in countering the
emergence of antibiotic resistance.
This is something that has to be done
collectively, on a community, national,
and global scale. It all comes back to
performing randomized controlled
clinical trials, and as Dr Spies noted in
1988, these are increasingly unlikely to
ever be done (1).
References
1. Spies JB, Rosen RJ, Lebovitz AS. An-
tibiotic prophylaxis in vascular and in-
terventional radiology: a rational ap-
proach. Radiology 1988; 166:381–387.
2. McDermott VG, Schuster MG, Smith
TP. Antibiotic prophylaxis in vascu-
lar and interventional radiology. AJR
Am J Roentgenol 1997; 169:31–38.
3. Dravid VS, Gupta A, Zegel HG, Mo-
rales AV, Rabinowitz B, Freiman DB.
Investigation of antibiotic prophylaxis
usage for vascular and nonvascular in-
terventional procedures. J Vasc Interv
Radiol 1998; 9:401–406.
4. Wenzel PR. Preoperative antibiotic
prophylaxis. N Engl J Med 1992; 326:
337–339.
5. Sanchez-Manuel FJ, Seco-Gil JL. An-
tibiotic prophylaxis for hernia repair.
Cochrane Database System Review
2003; 2:CD0033769.
6. Koc M, Zulfikaroglu B, Kece C, Ozalp
N. A prospective randomized study
of prophylactic antibiotics in elective
laparoscopic cholecystectomy. Surg
Endosc 2003; 17:1716–1718.
7. Leone M, Albanese J, Tod M, et al.
Ceftriaxone (1g intravenously) pene-
tration into abdominal tissues when
administered as antibiotic prophylaxis
during nephrectomy. J Chemother
2003; 15:139–142.
8. Goldman DA, Weinstein RA, Wenzel
RP, et al. Strategies to prevent and
control the emergence and spread of
antimicrobial-resistant organisms in
hospitals: a challenge to hospital lead-
ership. JAMA 1996; 275:234–240.
9. Jones RN, Phaller MA. Bacterial resis-
tance; a worldwide problem. Diag Mi-
crobiol Infect Dis 1998; 31:379–388.
10. National nosocomial infections surveil-
lance (NNIS) system report. Data
summary from January 1992–April
2000. Am J Infect Control 2000; 28:429–
448.
11. Klare I, Badstubner D, Konstabel C,
Bohme G, Claus H, Witte W. De-
creased incidence of VanA-type VRE
isolated from poultry meat and from
fecal samples in humans in the com-
munity after discontinuation of
avoparcin usage in animal husbandry.
Microb Drug Resist 1999; 5:45–52.
12. Chiew YF. Can susceptibility to an
antimicrobial be restored by halting its
use. The case of streptomycin versus
Enterobacteriaceae. J Antimicrobial
Chemother 1998; 41:247–251.
13. Boneca IG, Chiosis G. Vancomycin
resistance: occurrence, mechanisms
and strategies to combat it. Expert
Opin Ther Targets 2003; 7:311–328.
14. Dzidic S, Bedekovic V. Horizontal
gene transfer-emerging multidrug re-
sistance in hospital bacteria. Acta Phar-
macol Sin 2003; 24:519–526.
15. Cunha BA. Antimicrobial selection in
the penicillin-allergic patient. Drugs
Today 2001; 37:377–383.
16. Burke JF. The effective period of pre-
ventive antibiotic action in experimen-
tal incisions and dermal lesions. Sur-
gery 1961; 50:161–168.
17. Stone HH, Hooper CA, Kolb LD, Ge-
heber CE, Dawkins CE. Antibiotic
prophylaxis in gastric, biliary, and co-
lonic surgery. Ann Surg 1976; 184:443–
452.
18. Classen DC, Evans RS, Pestotnik SL.
The timing of prophylactic administra-
tion of antibiotics and the risk of sur-
gical wound infection. N Engl J Med
1992; 326:281–286.
19. Stone HH, Haney BB, Kolb LD, Gehe-
ber CE, Hooper CA. Prophylactic and
preventive antibiotic therapy, timing,
duration and economics. Ann Surg
1979; 189:691–699.
20. Nelson CL, Green TG, Porter RA.
One day versus seven days of preven-
tive antibiotic therapy in orthopedic
surgery. Clin Orthop 1983; 176:258–
263.
21. DiPiro JT, Cheung RP, Bowden TA,
Mansberger JA. Single dose systemic
antibiotic prophylaxis of surgical
wound infections. Am J Surg 1986; 152:
552–559.
22. Thadepalli H, Mandal AK. Antibiotic
prophylaxis in the surgical patient. In-
fez Med 1998; 6:71–80.
23. Pron G, Mocarski E, Cohen M, et al.
Hysterectomy for complications after
uterine artery embolization for leiomy-
oma: results of a Canadian multicenter
clinical trial. J Am Assoc Gynecol
Laparosc 2003; 10:99–106.
24. Walker WJ, Pelage JP. Uterine artery
embolization in symptomatic fibroids:
clinical results in 400 women with im-
aging follow up. BJOG 2002; 109:1262–
1272.
25. Ryan JM, Gainey M, Glasson J, Doherty
J, Smith TP. Simplified pain protocol
after uterine artery embolization. Radi-
ology 2002; 224:610–612.
26. Siskin GP, Stainken BF, Dowling K,
Meo P, Ahn J, Dolen EG. Outpatient
uterine artery embolization for symp-
tomatic uterine fibroids: experience in
49 patients. J Vasc Interv Radiol 2000;
11:305–311.
27. Katsumori T, Nakajima K, Mihara T,
Tokuhiro M. Uterine artery emboli-
zation using gelatin sponge particles
alone for symptomatic fibroids. AJR
Am J Roentgenol 2002; 178:135–139.
28. Pron G, Bennett J, Common A, et al.
Technical results and effects of opera-
tor experience on uterine artery embo-
lization for fibroids: The Ontario Uter-
ine Fibroid Embolization Trial. The
Ontario UFE Collaborative Group. J
Vascular and Interventional Radiology
2003; 14:545–554.
29. Mehta H, Sandhu C, Matson M, Belli
A-M. Review of re-admissions due to
complications form uterine fibroid em-
bolization. Clin Rad 2002; 57:1122–
1124.
30. Walker W, Green A, Sutton C. Bilat-
eral uterine artery embolization for
myomata: results, complications and
failures. Min Invas Ther Allied Technol
1999; 8:449–454.
31. Vashisht A, Studd J, Carey A, Burn P.
Fatal septicemia after fibroid emboliza-
tion. Lancet 1999; 354;307–308.
32. de Blok S, de Vries C, Prinssen HM,
Blaauwgeers HL, Jorna-Meijer LB.
Fatal sepsis after uterine artery embo-
lization with microspheres. J Vasc In-
terv Radiol 2003; 14:779–783.
33. Freedman AM, Sanyal AJ, Tisnado J, et
al. Complications of transjugular in-
trahepatic portosystemic shunts: a
comprehensive review. Radiographics
1993; 13:1185–1210.
34. Sanyal AJ, Reddy KR. Vegetative in-
fection of transjugular intrahepatic
portosystemic shunts. Gastroenterol-
ogy 1998; 115:110–115.
35. Schiano TD, Atillasoy E, Fiel MI, et al.
Fatal fungaemia resulting from an in-
fected transjugular intrahepatic porto-
systemic shunt stent. Am J Gastroen-
terol 1997; 92:709–710.
36. DeSimone JA, Beavis KG, Eschelman
DJ, Henning KJ. Sustained bactere-
mia associated with transjugular intra-
hepatic portosystemic shunt (TIPS).
Clin Infect Dis 2000; 30:384–386.
37. Armstrong PK, MacLeod C. Infection
of transjugular intrahepatic portosys-
temic shunt devices: three cases and a
review of the literature. Clin Infect Dis
2003; 36:407–412.
38. Brown RS, Brumage L, Yee HF, et al.
Enterococcal bacteremia after trans-
jugular intrahepatic portosystemic
shunts (TIPS). Am J Gastroenterol
1998; 93:636–639.
39. Deibert P, Schwarz S, Olschewski M,
Siegerstetter V, Blum HE, Rossle M.
Risk factors and prevention of early
554 •Antibiotic Prophylaxis for the Interventional Radiologist June 2004 JVIR
infection after implantation or revision
of transjugular intrahepatic portosys-
temic shunts: results of a randomized
study. Dig Dis Sci 1998; 43:1708–1713.
40. Yee AC, Ho C-S. Complications of
percutaneous drainage: benign versus
malignant diseases. AJR Am J Roentge-
nol 1987; 148:1207–1209.
41. Hamlin JA, Friedman M, Stein MG,
Bray JF. Percutaneous biliary drain-
age: complications of 118 consecutive
catheterizations. Radiology 1986; 158:
199–202.
42. Funaki B, Zaleski GX, Straus CA, et al.
Percutaneous biliary drainage in pa-
tients with nondilated intrahepatic
ducts. AJR Am J Roentgenol 1999; 173:
1541–1544.
43. Brody LA, Brown KT, Getrajdman GI,
et al. Clinical factors associated with
positive bile cultures during primary
percutaneous biliary drainage. J Vasc
Interv Radiol 1998; 9:572–578.
44. Clark CD, Picus D, Dunagan WC.
Bloodstream infections after interven-
tional procedures in the biliary tract.
Radiology 1994; 191:495–499.
45. Thompson BF, Arguedas MR, Wilcox
CM. Antibiotic prophylaxis prior to
endoscopic retrograde cholangiopan-
creatography in patients with ob-
structive jaundice: is it worth the
cost? Aliment Pharmacol Ther 2002;
16:727–734.
46. Sheen-Chen S, Chen W, Eng H, et al.
Bacteriology and antimicrobial choice
in hepatolithiasis. Am J Infect Control
2000; 28:298–301.
47. Cronan JJ, Horn DL, Marcello et al.
Antibiotics and nephrostomy tube
care: preliminary observations. Radiol-
ogy 1989; 172:1043–1045.
48. Larsen E, Gasser T, Madsen P. Anti-
biotic prophylaxis in urologic surgery.
Urol Clin North Am 1986; 13:591–604.
49. Ryan JM, Hahn PF, Boland GW, Mc-
Dowell RK, Saini S, Mueller PR. Per-
cutaneous gastrostomy with t-fastener
gastropexy: results of 316 consecutive
procedures. Radiology 1997; 203:496–
500.
50. Chen C, Chen PJ, Yang PM. Clinical
and microbiological features of liver
abscess after transarterial embolization
for hepatocellular carcinoma. Am J
Gastroenterology 1997; 92:2257–2259.
51. de Baere T, Roche A, Amenabar JA, et
al. Liver abscess formation after local
treatment of liver tumors. Hepatology
1996; 23:1436–1440.
52. Kim W, Clark TW, Baum RA, Soulen
MC. Risk factors for liver abscess for-
mation after hepatic chemoemboliza-
tion. J Vasc Interv Radiol 2001; 12:965–
968.
53. Song SY, Chung JW, Han JK, et al.
Liver abscess after transcatheter oily
chemoembolization for hepatic tumors:
incidence, predisposing factors, and
clinical outcome. J Vasc Interv Radiol
2001; 12:313–320.
54. Geschwind JF, Kaushik S, Ramsey DE,
Choti MA, Fishman EK, Kobeiter H.
Influence of a new prophylactic antibi-
otic therapy on the incidence of liver
abscesses after chemoembolization
treatment of liver tumors. J Vasc Interv
Radiol 2002; 13:1163–1166.
55. Spigos DG, Jonasson O, Mozes M, et al.
Partial splenic embolization in the
treatment of hypersplenism. AJR Am J
Roentgenol 1979; 132:777–782.
56. Mayo-Smith WW, Dupuy DE, Parikh
PM, Pezzullo JA, Cronan JJ. Imaging-
guided percutaneous radio-frequency
ablation of solid renal masses: tech-
niques and outcomes of 38 treatment
sessions in 32 consecutive patients.
AJR Am J Roentgenol 2003; 180:1503–
1508.
57. Dupuy DE, Goldberg SN. Image-
guided radiofrequency tumor ablation:
challenges and opportunities–part II. J
Vasc Interv Radiol 2001; 12:1135–1148.
58. Shawker TH, Kluge RM, Ayella RJ.
Bacteremia associated with angiogra-
phy. JAMA 1974; 229:1180–1182.
59. Meyer P, Reizine D, Aymard A, Guerin
JM, Merland JJ, Habib Y. Septic com-
plications in interventional radiology:
evaluation of risk and preventive mea-
sures. Preliminary studies. J Intervent
Rad 1988; 3:73–75.
60. Hearn AT, James KV, Lohr JM, et al.
Endovascular stent infection with de-
layed bacterial challenge. Am J Surg
1997; 174:157–159.
61. Dieter RS. Coronary artery stent in-
fection. Clin Cardiol 2000; 23:808–810.
62. Malek AM, Higashida RT, Reilly LM,
et al. Cardiovasc Intervent Radiol
2000; 23:57–60.
63. Leroy O, Martin E, Prat A, et al. Fatal
infection of coronary stent implanta-
tion. Cathet Cardiovasc Diagn 1996; 39:
168–170.
64. Deitch JS, Hansen KJ, Regan JD,
Burkhart JM, Ligush J. Infected renal
artery pseudoaneurysm and mycotic
aortic aneurysm after percutaneous
transluminal renal artery angioplasty
and stent placement in a patient with a
solitary kidney. J Vasc Surg 1998; 28:
340–344.
65. Chalmers N, Eadington DW, Gandan-
hamo D, Gillespie IN, Ruckley CV.
Case report: infected false aneurysm at
the site of an iliac stent. Br J Radiol
1993; 66:946–948.
66. Liu P, Dravid V, Freiman D, Zegel H,
Weinberg D. Persistent iliac endarte-
ritis with pseudoaneurysm formation
following balloon-expandable stent
placement. Cardiovasc Intervent Ra-
diol 1995; 18:39–42.
67. Deiparine MK, Ballard JL, Taylor FC,
Chase DR. Endovascular stent infec-
tion. J Vasc Surg 1996; 23:529–533.
68. Bunt TJ, Gill HK, Smith DC, Taylor FC.
Infection of a chronically implanted il-
iac artery stent. Ann Vasc Surg 1997;
11:529–532.
69. McCready RA, Siderys H, Pittman JN,
et al. Septic complications after car-
diac catheterization and percutaneous
transluminal coronary angioplasty. J
Vasc Surg 1991; 14:170–174.
70. Carey D, Martin JR, Moore CA, Valen-
tine MC, Nygaard TW. Complica-
tions of femoral artery closure devices.
Catheter Cardiovasc Interv 2001; 52:
3–7.
71. Becker GJ, Kovacs M, Mathison MN, et
al. Risk stratification and outcomes of
transluminal endografting for abdomi-
nal aortic aneurysm: 7-year experience
and long-term follow-up. J Vasc Interv
Radiol 2001; 12:1033–1046.
72. Heikkinen L, Valtonen M, Lepantalo
M, Saimanen E, Jarvinen A. Infrare-
nal endoluminal bifurcated stent graft
infected with Listeria monocytogenes. J
Vasc Surg 1999; 29:554–556.
73. Katzen BT, Becker GJ, Mascioli CA, et
al. Creation of a modified angiogra-
phy (endovascular) suite for translumi-
nal endograft placement and combined
interventional-surgical procedures. J
Vasc Interv Radiol 1996; 7:161–167.
74. Eliason JL, Guzman RJ, Passman MA,
Naslund TC. Infected endovascular
graft secondary to coil embolization of
endoleak: a demonstration of the im-
portance of operative sterility. Ann
Vasc Surg 2002; 16:562–565.
75. Kurimoto Y, Tsuchida Y, Saito J, Yama
N, Narimatsu E, Asai Y. Emergency
endovascular stent-grafting for in-
fected pseudoaneurysm of brachial ar-
tery. Infection 2003; 31:186–188.
76. Openshaw KL, Picus D, Hicks ME,
Darcy MD, Vesely TM, Picus J. Inter-
ventional radiologic placement of
Hohn central venous catheters: results
and complications in 100 consecutive
patients. J Vasc Interv Radiol 1994;
5:111–115.
77. Miller DL, O’Grady NP. Guidelines
for the prevention of intravascular
catheter-related infections: recommen-
dations relevant to interventional radi-
ology. J Vasc Interv Radiol 2003; 14:
133–136.
78. Whigham CJ, Goodman CJ, Fisher RG,
Greenbaum MC, Thornby JI, Thomas
JW. Infectious complications of 393
peripherally implantable venous ac-
cess devices in HIV-positive and HIV-
negative patients. J Vasc Interv Radiol
1999; 10:71–77.
79. Van DE, Van Woensel JB. Prophylac-
tic antibiotics for preventing early cen-
tral venous catheter Gram positive
infections in oncology patients (Co-
Ryan et al •555
Volume 15 Number 6
chrane Review). Cochrane Database
System Review 2003; 2:CD003295.
80. Brodwater BK, Silber JS, Smith TP, et
al. Conversion of indwelling chest
port catheters to tunneled central ve-
nous catheters. J Vasc Interv Radiol
2000; 11:1137–1142.
81. Philipneri M, Al Aly Z, Amin K,
Gellens ME, Bastani B. Routine re-
placement of tunneled, cuffed, hemo-
dialysis catheters eliminates paraspi-
nal/vertebral infections in patients
with catheter-associated bacteremia.
Am J Nephrol 2003; 23:202–207.
82. Darouiche RO. Antimicrobial ap-
proaches for preventing infections as-
sociated with surgical implants. Clin
Infect Dis 2003; 36:1284–1289.
83. Centers for Disease Control and Pre-
vention (CDC) Hospital Infection Con-
trol Practices Advisory Committee.
Guideline for prevention of surgical
site infection, 1999. Am J Infect Control
1999; 27:97–132.
Call for Editor
The Society of Interventional Radiology is seeking applications for the position of JVIR Editor-in-
Chief. The position will become available in January 2006 when the current editor, Dr. Karim Valji,
concludes his term. The ideal candidate will be a practicing interventional radiologist and an active
member of SIR with considerable experience in writing, editing, or reviewing scientific articles for
publication.
Interested individuals should contact Cathy Mendelsohn, JVIR, 1020 N. Haddow, Arlington
Heights, IL 60004. Phone 847-222-1708; Fax 847-342-0901; email cathy@sirweb.org for a complete
position description and application materials. Deadline for receipt of completed applications is
August 1, 2004.
556 •Antibiotic Prophylaxis for the Interventional Radiologist June 2004 JVIR
