Percutaneous transhepatic biliary drainage in patients with postsurgical bile leakage and nondilated intrahepatic bile ducts.

TL;DR: Patients with nondilated intrahepatic bile ducts who underwent a PTBD procedure for the treatment of bile leakage between January 2000 and August 2012 were retrospectively assessed and PTBD was an effective treatment with low complication rates.

Abstract: Objective and Background: Bile leakage is a serious postoperative complication and percutaneous transhepatic biliary drainage (PTBD) may be an option when endoscopic treatment is not feasible. In this retrospective study, we established technical and clinical success rates as well as the complication rates of PTBD in a large group of patients with postoperative bile leakage. Methods: Data on all patients with nondilated intrahepatic bile ducts who underwent a PTBD procedure for the treatment of bile leakage between January 2000 and August 2012 were retrospectively assessed. Data included type of surgery, site of bile leak, previous attempts of bile leak repair, interval between surgery and PTBD placement. Outcome measures were the technical and clinical success rates, the procedure-related complications, and mortality rate. Results: A total of 63 patients were identified; PTBD placement was technically successful in 90.5% (57/63) after one to three attempts. The clinical success rate was 69.8% (44/63). Four major complications were documented (4/63; 6.3%): liver laceration, pneumothorax, pleural empyema, and prolonged hemobilia. One minor complication involved pain. Conclusions: PTBD is an effective treatment with low complication rates for the management of postsurgical bile leaks in patients with nondilated bile ducts.

Summary

Introduction

• Postoperative bile leakage is a rare but serious complication of several surgical procedures, such as cholecystectomy, pancreatoduodenectomy, liver resection, and liver transplantation.
• Endoscopic and percutaneous transhepatic approaches are both described as effective to resolve biliary obstruction [10] [11] [12] [13] .
• They are also performed in patients with bile leakage and nondilated bile ducts.
• To their knowledge, the authors report the largest study on only PTBD procedures for biliary leakage in a group of exclusively postoperative patients.
• The aim of this retrospective study was primarily to investigate the technical success and also determine the clinical success and complication rates of this procedure.

Patients

• Eligible patients for this study were all subjects who underwent a PTBD procedure for the treatment of postsurgical bile leakage at the Department of Surgery of the Erasmus University Medical Centre, Rotterdam, the Netherlands, between January 2000 and August 2012.
• Only those patients with nondilated intrahepatic bile ducts on preinterventional ultrasound were included.
• Patients were subjected to PTBD when there was evidence of persisting bile leakage that could not be treated conservatively or with ERCP.
• As this study was retrospectively performed, the local ethics committee waived written informed consent.

Variables and Definitions

• The authors retrieved the following information from the clinical records of the patients included: date of the PTBD procedure, time interval between the initial surgical procedure and the PTBD procedure, type of surgery, site of bile leak, previous attempts of bile leak repair, technical and clinical success rates, procedure-related complications, and patient's survival.
• The initial diagnosis of bile leakage was based on either the presence of bile in a surgical drain or in a percutaneous drain which was placed in a fluid collection on imaging studies performed on clinical findings (e.g. fever, abdominal pain, peritonitis and sepsis).
• The diagnosis was mostly confirmed by PTC, occasionally by ERCP.
• The PTBD procedure was considered to be technically successful if either an internal-external drainage or external drainage only was accomplished.
• Complications were classified as major or minor according to the Society of Cardiovascular Interventional Radiology guidelines [29] .

Technique of Percutaneous Biliary Drainage

• All PTBD procedures were performed under local anesthesia (except when patients were already anesthetized) and patients received intravenous medication for conscious sedation.
• Under the supervision of an interventional radiologist, fractionated intravenous injection was titrated with steps of 0.05 and 2.5 mg for fentanyl and midazolam, respectively, until effective analgesia and sedation were obtained.
• Oxygen saturation and heart rate were monitored during the course of the procedure.
• After puncturing the bile duct, contrast material was injected and the correct position of the needle was indicated by opacification of the bile duct.
• Finally, the sheath and catheter were exchanged for a 10-Fr biliary multi-sidehole drainage catheter (Cook Medical, Inc.) over a stiff guidewire for internal-external drainage.

Data Analysis

• Nominal data are presented as numbers and percentages.
• Normally distributed variables are presented as mean (SD).
• Non-normally distributed variables are presented as median (interquartile range).
• The Mann-Whitney U test was used to compare time between surgery and PTBD for successful procedures versus failed procedures.
• Fisher exact tests were applied to compare nominal outcomes between groups.

Results

• More than 900 patients underwent a PTBD procedure for various indications between 2000 and 2012, of whom 63 patients with nondilated bile ducts.
• The background characteristics of the patients are shown in table 1 .
• The most frequent bile leakage was seen at the hepaticojejunostomy after liver transplantation or pancreaticoduodenectomy.
• The PTBD was performed a median of 17 days after the initial surgical intervention, range 2-664 days (1 patient was treated conservatively for almost 2 years after the initial surgery with multiple ERCP procedures, before undergoing a PTBD procedure).
• In all these patients a PTBD placement was indicated because the leak could not be found or because the re-exploration did not heal the leak.

Technical Success

• In table 2 the technical success rate is shown for the three different sites of bile leakage.
• 41 patients received an internal-external drainage catheter, 16 patients an external one.
• One of these 2 patients was re-operated and recovered uneventfully.
• Technical success or failure was not statistically significantly related to time between surgery and first PTBD procedure (Mann-Whitney test, p = 0.13).
• In the present study the authors used ultrasound guidance either alone or, when bile ducts were not visible or could not be punctured directly, combined with fluoroscopy.

Clinical Success

• The overall clinical success rate was 69.8%, as the bile leak was completely resolved in 44 of the 63 patients.
• In 40 of the 51 patients (78.4%) in whom the PTBD procedure was successful at the first attempt, the bile leak resolved (an example is shown in figure 2 ).
• 1 All Fisher exact tests comparing the four groups on the outcomes of table 2 give a p value of 1.0.
• The difference can be explained by the complexity and the severity of the underlying diseases of their patients.
• The increased clinical success rate, from 60% in the first period to 84.4% in the second study period, was mainly due to the technical progress made by their radiologist, which led to a higher percentage of patients with a successful PTBD procedure.

Complications

• Only one minor complication (1.6%) was documented, namely pain after the procedure.
• This concerned a 50-year-old woman with persisting bile leakage at the resection plane following a left hepatectomy.
• Still, the major complication rates of most studies are below the 10% threshold recommended by the Society of Cardiovascular and Interventional Radiology [29] .
• Several limitations of the study need to be addressed.

Discussion

• This study showed that PTBD is an effective treatment for the management of postsurgical bile leaks, as the bile leak resolved in 69.8% (44/63) of cases.
• The technical success rate was high at 90.5% and the complication rate was low at 7.9%.
• To the best of their knowledge, this study encompassed the largest time span and sample size of its kind.

Figures

Table 3. Outcome measures by type of surgery

Table 1. Patient characteristics (n = 63)

Fig. 1. Flow diagram showing the attempts, successes and failures of the PTBD procedures.

Table 2. Outcome measures by the site of bile leakage, n (%)

Fig. 2. A case where the bile leak resolved after the PTBD procedure was successful at the first attempt.

Full text

E-Mail karger@karger.com
Original Paper
Dig Surg 2013;30:444–450
DOI: 10.1159/000356711
Percutaneous Transhepatic Biliary Drainage
in Patients with Postsurgical Bile Leakage
and Nondilated Intrahepatic Bile Ducts
E.A.deJong 
a
A.Moelker 
b
T.Leertouwer 
b
S.Spronk 
c
M.VanDijk 
d
C.H.J.vanEijck 
a
Departments of
a
 Surgery,
b
 Radiology, and
c
 Epidemiology and Radiology, Erasmus University Medical Centre, and
d
 Departments of Pediatric Surgery and Pediatrics, Erasmus MC-Sophia Children’s Hospital, Rotterdam , The Netherlands
volved pain. Conclusions: PTBD is an effective treatment
with low complication rates for the management of postsur-
gical bile leaks in patients with nondilated bile ducts.
© 2014 S. Karger AG, Basel
Introduction
Postoperative bile leakage is a rare but serious compli-
cation of several surgical procedures, such as cholecystec-
tomy, pancreatoduodenectomy, liver resection, and liver
transplantation. The incidence ranges from 0.9 to 9.0%
depending on the type of surgical procedure
[1–7] . It is
associated with significant mortality rates (8.7–39.0%)
and morbidity (22–44%)
[1, 3, 8, 9] .
Endoscopic and percutaneous transhepatic approach-
es are both described as effective to resolve biliary ob-
struction
[10–13] . The endoscopic approach via endo-
scopic retrograde cholangiopancreatography (ERCP) is
the first choice
[6, 13–16] . When ERCP is not successful
or not feasible, percutaneous transhepatic biliary drain-
age (PTBD) becomes an alternative. PTBD procedures
are most often performed in patients with biliary obstruc-
tion and dilated bile ducts, and therefore biliary access
can easily be achieved. However, they are also performed
in patients with bile leakage and nondilated bile ducts.
Key Words
Percutaneous transhepatic biliary drainage · Bile leakage ·
Nondilated bile ducts
Abstract
Objective and Background: Bile leakage is a serious postop-
erative complication and percutaneous transhepatic biliary
drainage (PTBD) may be an option when endoscopic treat-
ment is not feasible. In this retrospective study, we estab-
lished technical and clinical success rates as well as the com-
plication rates of PTBD in a large group of patients with post-
operative bile leakage. Methods: Data on all patients with
nondilated intrahepatic bile ducts who underwent a PTBD
procedure for the treatment of bile leakage between January
2000 and August 2012 were retrospectively assessed. Data
included type of surgery, site of bile leak, previous attempts
of bile leak repair, interval between surgery and PTBD place-
ment. Outcome measures were the technical and clinical
success rates, the procedure-related complications, and
mortality rate. Results: A total of 63 patients were identified;
PTBD placement was technically successful in 90.5% (57/63)
after one to three attempts. The clinical success rate was
69.8% (44/63). Four major complications were documented
(4/63; 6.3%): liver laceration, pneumothorax, pleural empy-
ema, and prolonged hemobilia. One minor complication in-
Received: June 17, 2013
Accepted after revision: October 22, 2013
Published online: January 14, 2014
Prof. C.H.J. van Eijck
Department of Surgery
‘s-Gravendijkwal 230
NL–3015 CE, Rotterdam (The Netherlands)
E-Mail c.vaneijck
 @  erasmusmc.nl
© 2014 S. Karger AG, Basel
0253–4886/14/0306–0444$39.50/0
www.karger.com/dsu
Downloaded by:
Erasmus Univ.of Rotterdam Medical Library
149.126.75.65 - 1/22/2016 4:37:19 PM

PTBD for post-surgical bile leakage
Dig Surg 2013;30:444–450
DOI: 10.1159/000356711
445
But when the bile ducts are decompressed, many passes
are often required to gain access
[17] .
The first clinical studies on accessing nondilated bile
ducts for percutaneous transhepatic cholangiography
(PTC) reported a 25% technical success rate and a 21%
complication rate [15, 18] . More recent small series have
shown that PTC and subsequent drainage for a variety
of indications can be feasible in patients with nondilated
bile ducts as well; the technical success rates were similar
to those in patients with dilated bile ducts (91–100%)
[17, 19–26] . Several studies have described the manage-
ment of biliary leakage both after hepaticojejunostomy
or post-hepatectomy, however a PTBD was performed
in the minority of patients
[27, 28] . To our knowledge,
we report the largest study on only PTBD procedures for
biliary leakage in a group of exclusively postoperative
patients. The aim of this retrospective study was primar-
ily to investigate the technical success and also deter-
mine the clinical success and complication rates of this
procedure.
Materials and Methods
Patients
Eligible patients for this study were all subjects who underwent
a PTBD procedure for the treatment of postsurgical bile leakage at
the Department of Surgery of the Erasmus University Medical
Centre, Rotterdam, the Netherlands, between January 2000 and
August 2012. Only those patients with nondilated intrahepatic bile
ducts on preinterventional ultrasound were included. Patients
were subjected to PTBD when there was evidence of persisting bile
leakage that could not be treated conservatively or with ERCP. As
this study was retrospectively performed, the local ethics commit-
tee waived written informed consent.
Variables and Definitions
We retrieved the following information from the clinical re-
cords of the patients included: date of the PTBD procedure, time
interval between the initial surgical procedure and the PTBD pro-
cedure, type of surgery, site of bile leak, previous attempts of bile
leak repair, technical and clinical success rates, procedure-related
complications, and patient’s survival.
On the preinterventional ultrasound, nondilated ducts were
defined as peripheral bile ducts measuring <2 mm in diameter or
by visualization of ducts smaller than the adjacent portal vein
[17] .
The initial diagnosis of bile leakage was based on either the
presence of bile in a surgical drain or in a percutaneous drain
which was placed in a fluid collection on imaging studies per-
formed on clinical findings (e.g. fever, abdominal pain, peritonitis
and sepsis). The diagnosis was mostly confirmed by PTC, occa-
sionally by ERCP.
The PTBD procedure was considered to be technically success-
ful if either an internal-external drainage or external drainage only
was accomplished. The bile leak was considered clinically success-
fully resolved if repeat cholangiograms, performed after bile drain
production had stopped, showed absence of contrast extravasation
outside the bile ducts (definitive closure of the leak). Complica-
tions were classified as major or minor according to the Society of
Cardiovascular Interventional Radiology guidelines
[29] . Major
complications are those resulting in hospitalization (for outpa-
tients), unplanned increase in the level of care, prolonged hospi-
talization, permanent adverse sequelae, or death. Minor complica-
tions are those requiring nominal therapy, or a short hospital stay
for observation as needed, exclusive of sequelae.
Technique of Percutaneous Biliary Drainage
Preprocedural antibiotic prophylaxis consisted of 1 g of Aug-
mentin
®
administered intravenously. All PTBD procedures were
performed under local anesthesia (except when patients were al-
ready anesthetized) and patients received intravenous medication
for conscious sedation. Under the supervision of an intervention-
al radiologist, fractionated intravenous injection was titrated with
steps of 0.05 and 2.5 mg for fentanyl and midazolam, respectively,
until effective analgesia and sedation were obtained.
Oxygen saturation and heart rate were monitored during the
course of the procedure. The puncture site was chosen under ul-
trasound guidance and the operation field was disinfected
(chlorhexidine, 70% alcohol) and anaesthetized (10 ml of 2% lido-
caine). Next, percutaneous puncture of the intrahepatic bile duct
was performed using of a 21-gauge needle (Chiba Tip Needle;
Cook Medical, Inc., Bloomington, Ind., USA) under ultrasound
guidance. After puncturing the bile duct, contrast material was
injected and the correct position of the needle was indicated by
opacification of the bile duct. In some cases, if the bile ducts were
not visible with ultrasound, the bile duct was punctured under
fluoroscopic guidance. The needle was then advanced towards a
peripheral portal vein and after removing the needle stylet, con-
trast was gently injected while withdrawing the needle until a
bileduct opacified. When the contrast injection showed that the
needle tip was successfully placed in a bile duct, a 0.018-in guide-
wire was advanced. The needle was subsequently exchanged for a
small coaxial catheter and a cholangiography was made. Next, the
small coaxial catheter was replaced by a sheath with a distal mark-
er and a 4-Fr inner diameter using the 0.018-in guidewire. A cath-
eter with hydrophilic guidewire was then introduced through the
4-Fr sheath and attempts were made to enter the small bowel. Fi-
nally, the sheath and catheter were exchanged for a 10-Fr biliary
multi-sidehole drainage catheter (Cook Medical, Inc.) over a stiff
guidewire for internal-external drainage. If the common bile duct
could not be cannulated and no drainage tube could be placed in
the small bowel, an 8.5-Fr external drainage catheter was posi-
tioned in the common bile duct. After positioning of the PTBD,
healing of the bile leak was monitored with repeat cholangio-
grams.
D a t a A n a l y s i s
Nominal data are presented as numbers and percentages. Nor-
mally distributed variables are presented as mean (SD). Non-nor-
mally distributed variables are presented as median (interquartile
range). The Mann-Whitney U test was used to compare time be-
tween surgery and PTBD for successful procedures versus failed
procedures. Fisher exact tests were applied to compare nominal
outcomes between groups.
Downloaded by:
Erasmus Univ.of Rotterdam Medical Library
149.126.75.65 - 1/22/2016 4:37:19 PM

deJong/Moelker/Leertouwer/Spronk/
VanDijk/vanEijck
Dig Surg 2013;30:444–450
DOI: 10.1159/000356711
446
R e s u l t s
More than 900 patients underwent a PTBD procedure
for various indications between 2000 and 2012, of whom
63 patients with nondilated bile ducts. The number of
PTBD procedures annually varied from 2 to 10 with a
median of 4. The background characteristics of the pa-
tients are shown in table1 . The most frequent bile leakage
was seen at the hepaticojejunostomy after liver transplan-
tation or pancreaticoduodenectomy.
The PTBD was performed a median of 17 days after
the initial surgical intervention, range 2–664 days (1 pa-
tient was treated conservatively for almost 2 years after
the initial surgery with multiple ERCP procedures, before
undergoing a PTBD procedure). Only 6 patients had a
PTBD procedure within 1 week after surgery. In 43 pa-
tients (68.3%) an ERCP was not possible because of post-
surgical inaccessibility of the bile ducts. In the remaining
20 patients, PTBD was performed after reported failure
of ERCP.
Of the 63 patients, 27 (42.9%) had undergone a surgi-
cal re-exploration before the PTBD. In all these patients
a PTBD placement was indicated because the leak could
not be found or because the re-exploration did not heal
the leak. Of the 63 patients, 8 (12.7%) underwent the pro-
cedure fully anesthetized.
Technical Success
The overall technical success rate was 90.5% (57/63).
In 51 patients (81.0%) the first attempt was successful. In
6 of 12 patients (50.0%) in whom the first attempt failed,
a second or third attempt at a later stage was successful
( fig.1 ). In table2 the technical success rate is shown for
the three different sites of bile leakage.
The main outcome measures of the PTBD procedure
analyzed per type of surgery are shown in table 3 . The
catheter was placed at the right site in 44 patients, at the
left site in 11 patients, at both the right and left site in 1
patient, and in 1 patient a right-sided catheter was placed
when it failed at the left site.
41 patients received an internal-external drainage
catheter, 16 patients an external one. In 5 cases the ex-
ternal drainage catheter was later converted to an inter-
nal-external drainage catheter, a mean 12 days (range
7–18) after the initial procedure. In 2 of the 6 patients in
whom the PTBD failed, the second attempt failed as well.
One of these 2 patients was re-operated and recovered
uneventfully. The other was treated with an ERCP and
subsequent stent placement, after which the bile leak
healed.
Technical success or failure was not statistically sig-
nificantly related to time between surgery and first PTBD
procedure (Mann-Whitney test, p= 0.13). The technical
success rate increased from 82.1% in the first study period
(2000–2006) to 97.1% in the last study period (2007–
2012) (Fisher exact test, p= 0.08).
Table 1. Patient characteristics (n= 63)
Characteristic n (%)
Age, years, mean (SD) 57 (15)
Range 21–87
Male 31 (49)
Initial surgical intervention
Cholecystectomy 22 (35)
Biliodigestive anastomosis 19 (30)
Liver transplantation 12 (19)
Liver resection 10 (16)
Site of bile leak
Bile ducts
1
29 (46)
Hepatico- or choledochojejunostomy 30 (48)
Plane of resection 4 (6)
1
Leaks at the bile ducts were at the cystic duct stump (n= 4),
ductus choledochus (n= 14), and common left or right hepatic
duct (n= 11).
Successful first
attempt (n = 51)
Technical failure
(n = 6)
PTBD procedures in
patients with bile leaks
(n = 63)
Technical success
(n = 57)
Successful second
attempt (n = 5)
Successful third
attempt (n = 1)
Clinical success
(n = 44)
Complications
(n = 4)
Clinical failure
(n = 13)
Complications
(n = 1)
Fig. 1. Flow diagram showing the attempts, successes and failures
of the PTBD procedures.
Downloaded by:
Erasmus Univ.of Rotterdam Medical Library
149.126.75.65 - 1/22/2016 4:37:19 PM

PTBD for post-surgical bile leakage
Dig Surg 2013;30:444–450
DOI: 10.1159/000356711
447
Clinical Success
The overall clinical success rate was 69.8%, as the bile
leak was completely resolved in 44 of the 63 patients. Af-
ter a technically successful PTBD, the bile leakage healed
in 44 of the 57 (77.2%) patients. In 40 of the 51 patients
(78.4%) in whom the PTBD procedure was successful at
the first attempt, the bile leak resolved (an example is
shown in figure 2 ).
Bile leakage at the biliodigestive anastomosis was re-
solved in 80.0% (24/30) after PTBD placement, leakage
at the bile ducts in 65.5% (19/29), and leakage at the
plane of resection in 25.0% (1/4) ( table2 ). However, in
13 patients the bile leak did not resolve; 8 of these pa-
tients died before healing of the bile leak could be dem-
onstrated. In the 4 other patients the bile leak was re-
solved after (a) a new, successful, ERCP procedure with
subsequent stent placement (n= 2), (b) an additional
left-sided drainage catheter to heal the leak, next to the
right-sided catheter that was already present, and (c) by
using coils.
Table 3. Outcome measures by type of surgery
Outcome
1
Total sample
(n= 63)
Cholecystectomy
(n= 22)
Biliodigestive anastomosis
(n= 19)
Liver transplantation
(n= 12)
Liver resection
(n= 10)
Technically successful 57 (91) 21 (95) 15 (79) 11 (92) 10 (100)
Clinically successful 44 (70) 17 (77) 13 (68) 6 (50) 8 (80)
Total complications 5 (8) 1 (5) 3 (16) – 1 (10)
Major complications 4 (6) 1 (5) 2 (11) – 1 (10)
30-Day mortality 4 (6) – 1 (5) 2 (17) 1 (10)
1
All Fisher exact tests comparing the four groups on the outcomes of table 2 give a p value of 1.0.
Table 2. Outcome measures by the site of bile leakage, n (%)
Outcome
1
Total sample
(n= 63)
Leakage at the bile
ducts (n= 29)
Leakage at the hepatico- or
choledochojejunostomy (n= 30)
Leakage at plane of
resection (n= 4)
Technically successful 57 (91) 25 (86) 28 (93) 4 (100)
Clinically successful 44 (70) 19 (66) 24 (80) 1 (25)
Total complications 5 (8) 1 (3) 3 (10) 1 (25)
Major complications 4 (6) 1 (3) 2 (7) 1 (25)
30-Day mortality 4 (6) 1 (3) 2 (7) 1 (25)
1
All Fisher exact tests comparing the four groups on the outcomes of table 2 give a p value of 1.0.
Fig. 2. A case where the bile leak resolved
after the PTBD procedure was successful at
the first attempt.
Color version available online
Downloaded by:
Erasmus Univ.of Rotterdam Medical Library
149.126.75.65 - 1/22/2016 4:37:19 PM

deJong/Moelker/Leertouwer/Spronk/
VanDijk/vanEijck
Dig Surg 2013;30:444–450
DOI: 10.1159/000356711
448
Nine patients with iatrogenic bile duct injury required
definitive surgical reconstruction on the biliary tract.
They all had clinically successfully resolved bile leaks after
PTBD placement. Hepaticojejunostomy was performed
in all 9 patients a mean 106 days after PTBD (range 43–
156). Time between surgery and first PTBD was not sta-
tistically significantly different between clinical successes
and clinical failures with medians (IQR) of 18 (11–27)
and 14 (11–24), respectively (Mann-Whitney U test, p=
0.52). The clinical success rate increased statistically sig-
nificantly from 60% in the first study period (2000–2006)
to 84.8% in the second study period (2007–2012) (Fisher
exact test, p= 0.04).
C o m p l i c a t i o n s
The overall complication rate was 7.9% (5/63). Major
complications (6.3%) documented were: pneumothorax
(n= 1), pleural empyema (n= 1), liver laceration (n= 1),
and prolonged hemobilia (n= 1). Only one minor com-
plication (1.6%) was documented, namely pain after the
procedure.
The case of prolonged hemobilia concerned a 75-year-
old man with bile leakage after a complicated cholecys-
tectomy after an uncomplicated PTBD procedure. The
hemobilia was based on a fistula between the right
hepatic artery and the common bile duct. Emboliza-
tions and covered stent placements in the hepatic artery
successfully stopped the bleeding. All other complica-
tions were treated successfully without any clinical se-
quelae.
The 30-day mortality rate of all patients was 6.3%
(4/63); death occurred 1, 2, 4, and 28 days, respectively,
after the PTBD procedure. In-hospital mortality was
11.1% (7/63). These 7 patients were already in a poor con-
dition before the procedure. Six died of underlying prob-
lems, i.e. sepsis (n= 4), irresectable cholangiocarcinoma
(n= 1), and aspiration pneumonia (n= 1). All of them
had a technical successful PTBD placement at the first at-
tempt but all, except the patient that died due to an irre-
sectable cholangiocarcinoma, were clinical failures. In the
patients who died from sepsis, the PTBD could not con-
trol the source of sepsis. No evidence of bile leak resolu-
tion was seen at the time of death in the patient who died
from aspiration pneumonia.
One patient died of procedure-related complications.
This concerned a 50-year-old woman with persisting bile
leakage at the resection plane following a left hepatecto-
my. Multiple passes were needed before the bile duct was
correctly punctured and an internal-external drainage
catheter could be placed. Within 1 h after placement, a
liver laceration led to severe hypovolemic, hemorrhagic
shock. She was re-operated and packed but the shock
could not be corrected and she died the next day.
D i s c u s s i o n
This study showed that PTBD is an effective treatment
for the management of postsurgical bile leaks, as the bile
leak resolved in 69.8% (44/63) of cases. The technical suc-
cess rate was high at 90.5% and the complication rate was
low at 7.9%. To the best of our knowledge, this study en-
compassed the largest time span and sample size of its
kind. Over the last 20 years a number of small studies in
at least 10 patients also reported high technical success
rates (91–100%), high clinical success rates (70–100%),
and low major complication rates (0–13%)
[8, 19, 21, 22,
25, 26, 30]
.
Technical Success
The above high technical success rates were achieved
despite nondilated bile ducts. In the present study we
used ultrasound guidance either alone or, when bile
ducts were not visible or could not be punctured direct-
ly, combined with fluoroscopy. In a study from Kuhn
etal.
[19] the fluoroscopy time in patients with nondi-
lated bile ducts was significantly longer than that in pa-
tients with dilated bile ducts. These authors also applied
supplementary techniques for the opacification of the
intrahepatic bile system: in 16 of 21 patients they used
either CT-guided percutaneous puncture, T-drainage,
or a temporary gallbladder drainage to aid the PTBD
placement, resulting in a 100% technical success rate.
Aytekin et al.
[20] achieved access to dilated bile ducts
with percutaneous puncture of a peripheral duct under
ultrasonographic guidance. In patients with nondilated
peripheral bile ducts at ultrasound, the puncture was
performed under fluoroscopic guidance using cholangi-
ography (obtained via the drainage catheter near the
leak site or surgical T-tube). Furthermore, in some stud-
ies the puncture site was usually more central when the
bile ducts were nondilated, as this facilitates the punc-
ture
[21, 22, 26] . Also, Funaki et al. [17] reported a high-
er number of passes in patients with nondilated bile
ducts. In the present study, too, the puncture site was
occasionally more central, and in some cases multiple
passes were needed before the bile ducts were correctly
punctured. Since the radiologist performing these pro-
cedures remained the same over the study period, the
increase of the success rate (97.1%) in the last study pe-
Downloaded by:
Erasmus Univ.of Rotterdam Medical Library
149.126.75.65 - 1/22/2016 4:37:19 PM

Frequently asked questions

Q1. What are the contributions in "Dsu356711_bw.indd" ?

In this retrospective study, the authors established technical and clinical success rates as well as the complication rates of PTBD in a large group of patients with postoperative bile leakage.