Specialist Early and Immediate Repair of Post-laparoscopic Cholecystectomy Bile Duct Injuries Is Associated With an Improved Long-term Outcome

Article (PDF Available)inAnnals of surgery 253(3):553-60 · March 2011with122 Reads
DOI: 10.1097/SLA.0b013e318208fad3 · Source: PubMed
Abstract
A majority of bile duct injuries (BDI) sustained during laparoscopic cholecystectomy require formal surgical reconstruction, and traditionally this repair is performed late. We aimed to assess long-term outcomes after repair, focusing on our preferred early approach. A total of 200 BDI patients [age 54(20-83); 64 male], followed up for median 60 (5-212) months were assessed for morbidity. Factors contributing to this were analyzed with a univariate and multivariate analysis. A total of 112 (56%) patients were repaired by specialist hepatobiliary surgeons [timing of repair: immediate, n = 28; early (<21 days), n = 43; and late (>21 days) n = 41], whereas 45 (22%) underwent repair by nonspecialist surgeons before specialist referral [immediate, n = 16; early, n = 26 and late, n = 03]. Outcomes after immediate and early repairs were comparable to late repairs when performed by specialists [recurrent cholangitis:11%, 12%, and 10%; P = 0.96, NS; re-stricture:18%,5%, and 29%; P = 0.01; nonsurgical intervention: 14%, 5%, and 24%; P<0.03; redo surgery: 4%, 2%, and 5%; P = 0.81, NS; overall morbidity: 21%, 14%, and 39%; P<0.02]. On multivariate analysis, immediate and early repairs done by nonspecialist surgeons were independent risk factors (P < 0.05) for recurrent cholangitis [50% and 27%], re-stricturing (75% and 61%), redo reconstructions (31% and 61%), and overall morbidity (75% and 84%). Immediate and early repair after BDI results in comparable, if not better long-term outcomes compared to late repair when performed by specialists.
ORIGINAL ARTICL E
Specialist Early and Immediate Repair of Post-laparoscopic
Cholecystectomy Bile Duct Injuries Is Associated With an
Improved Long-term Outcome
M. Thamara P. R. Perera, MS, FRCS, Michael A. Silva, MD, FRCS, Bassem Hegab, MRCS,
Vijayaragavan Muralidharan, FRACS, Simon R. Bramhall, MD, FRCS,, A. David Mayer, MS, FRCS,
John A. C. Buckels, MD, FRCS, and Darius F. Mirza, MS, FRCS
Introduction: A majority of bile duct injuries (BDI) sustained during laparo-
scopic cholecystectomy require formal surgical reconstruction, and tradition-
ally this repair is performed late. We aimed to assess long-term outcomes after
repair, focusing on our preferred early approach.
Methods: A total of 200 BDI patients [age 54(20–83); 64 male], followed up
for median 60 (5–212) months were assessed for morbidity. Factors contribut-
ing to this were analyzed with a univariate and multivariate analysis.
Results: A total of 112 (56%) patients were repaired by specialist hepato-
biliary surgeons [timing of repair: immediate, n =28; early (<21 days),
n=43; and late (>21 days) n =41], whereas 45 (22%) underwent re-
pair by nonspecialist surgeons before specialist referral [immediate, n =16;
early, n =26 and late, n =03]. Outcomes after immediate and early repairs
were comparable to late repairs when performed by specialists [recurrent
cholangitis:11%, 12%, and 10%; P=0.96, NS; re-stricture:18%,5%, and
29%; P=0.01; nonsurgical intervention: 14%, 5%, and 24%; P<0.03; redo
surgery: 4%, 2%, and 5%; P=0.81, NS; overall morbidity: 21%, 14%, and
39%; P<0.02]. On multivariate analysis, immediate and early repairs done by
nonspecialist surgeons were independent risk factors (P<0.05) for recurrent
cholangitis [50% and 27%], re-stricturing (75% and 61%), redo reconstruc-
tions (31% and 61%), and overall morbidity (75% and 84%).
Conclusion: Immediate and early repair after BDI results in comparable, if
not better long-term outcomes compared to late repair when performed by
specialists.
(Ann Surg 2011;253:553–560)
Laparoscopic cholecystectomy (LC) is one of the most commonly
performed abdominal operations with the benefits of reduction in
surgical morbidity, shorter hospital stay, and a much faster return to
normal activity for patients.1,2–5 However, the incidence of iatrogenic
bile duct injuries (BDI) has remained between 0.3% and 0.6% over
the last decade, well beyond the learning curve of this excellent
surgical procedure.6The pattern of BDI after LC differs from that
occurring at open surgery. Complex BDI are often more common
after laparoscopic surgery, the classical one being an excision injury
of the common bile duct.7In addition, ischaemic stricturing because
of the use of monopolar diathermy and devascularization of bile ducts
are also features of the laparoscopic era.8,9 These complications often
result in long-term physical and psychological morbidity to patients.
Moreover, increased litigation costs are superadded to the health care
costs, exerting pressure on surgeons and health systems.10
From the The Liver Unit, University Hospital Birmingham, Queen Elizabeth Hos-
pital, Birmingham, United Kingdom.
Part of this data was presented at the Association of Surgeons of Great Britain
and Ireland (ASGBI) Annual Scientific Meeting held in Bournemouth, United
Kingdom—May 2008.
Reprints: Darius F. Mirza, MS, FRCS, Queen Elizabeth Hospital, Edgbaston,
Birmingham B15 2TH, United Kingdom. E-mail: darius.mirza@uhb.nhs.uk.
Copyright C
2011 by Lippincott Williams & Wilkins
ISSN: 0003-4932/11/25303-0553
DOI: 10.1097/SLA.0b013e318208fad3
Early recognition of BDI is essential for optimal treatment to
be initiated. The delay in recognizing such injuries either at the time
of surgery or in the postoperative period is compounded by a further
delay in referral to specialized hepatobiliary centers. Most referring
general surgeons lack either the experience or the multidisciplinary
support to provide adequate assessment, and carry out complex re-
constructive procedures on these patients. Such nonspecialist repairs
have a higher risk of developing subsequent strictures and even sec-
ondary biliary cirrhosis.11
Conventional surgical strategies include early establishment of
drainage and subsequent late repair when inflammation has settled
and awaiting the development of bile duct dilatation before definitive
repair, a process that can take several months.12–14 An early repair by
specialist hepatobiliary surgeons (HBS) has the advantage of short-
ening this process but usually entails carrying out a complex high
reconstruction to a nondilated hepatic duct in the presence of some
inflammation. Furthermore, we have recently reported on perform-
ing specialist “on-table” repairs at referring hospitals as an outreach
service, to facilitate early definitive treatment of injuries recognized
at cholecystectomy.15In this study, we report the long-term morbid-
ity of patients treated with surgical intervention for iatrogenic BDI
sustained at LC, with an emphasis on the long-term outcome of spe-
cialist early and “on-table” repair. We also aimed to identify factors
associated with an adverse outcome. We believe that this is the first
article describing long-term follow-up of patients undergoing early
repair of post-LC BDI.
METHODS
All patients with BDI referred to the Liver Unit at the Queen
Elizabeth Hospital, Birmingham (1991–2007) were prospectively en-
tered onto a database and followed up, and were classified using the
“Strasberg” classification.16 Areviewofcaserecordsandradiology
reports was carried out where necessary. Data collected included the
following: indication for LC, intraoperative findings, time to devel-
opment of symptoms, presenting symptoms, time to recognition of
possible injury and time to specialist referral, type of injury, pres-
ence of associated vascular injuries, and management before and
after specialist referral. Initial recognition of injury was classified as
immediate (recognized at the time of index cholecystectomy), early
(recognized within 21 days), or late (recognized beyond 21 days) de-
pending on the time elapsed since index LC. Similarly delayin refer ral
was defined as “no delay” if the injury was suspected and an imme-
diate referral was made during the LC, “early” referrals (made up to
21 days after LC) and “delayed” referrals (made beyond 21 days).
Patients undergoing surgical intervention with the intention of
repair or reconstruction were also subgrouped on the basis of the
timing of intervention. The “on-table” repair group consisted of in-
juries that were diagnosed and repaired immediately during the index
surgery under same anaesthesia, usually at a hospital remote from the
HBS center. For referrals of BDI recognized during the postoperative
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery rVolume 253, Number 3, March 2011 www.annalsofsurgery.com |553
Per era et al Annals of Surgery rVolume 253, Number 3, March 2011
period, a policy to perform early surgical intervention is usually fol-
lowed. This is possible only if the referral is made early (<21 days)
after LC and provided the patient’s clinical condition is stable to
withstand surgical reconstruction. Depending on the timing of such
intervention, they were further categorized to “early” (surgical inter-
vention up to 21 days) or “late” (surgicalinter vention beyond 21 days).
All patients were followed up long-term by the specialist HBS
team. A morbidity event was recorded for each of the following:
episodes of cholangitis without biliary obstruction, episodes of
cholangitis and/or cholestasis with biliary strictures resulting in either
nonsurgical intervention or revision surgery. An episode of cholangi-
tis was defined as fever with/without rigors, leucocytosis with associ-
ated elevation of liver enzymes. Long-term morbidity from BDI was
defined as the presence during follow-up of 1 or more of the events
described above.
The first part of the data analysis aimed at comparing the out-
come of biliary reconstruction done by specialist HBS with regard
to the timing of surgical intervention, that is, on-table vs early vs
late surgical reconstruction. The second part of the study analyzed
risk factors, and included all those who had had surgical intervention
after BDI, with specialist vs nonspecialist repair and timing of repair
as individual predictive variables of long-term morbidity. Other vari-
ables analyzed were as follows; surgical anatomy described at index
LC (straight forward, difficult anatomy, excessive bleeding at Calot’s
triangle), injury type (type E injury vs others, type E 3–5 vs others, E
3–5 vs E 1 and 2), presence or absence of vascular injury, timing of
initial injury recognition (immediate, early or late) and the timing of
referral to specialist HBS (immediate, early or delayed). Comparison
of results between groups and univariate analysis were done with χ2
test; Yate correction was employed when indicated. Fisher Exact test
was used for comparison of groups involving small numbers. Mul-
tivariate analysis of risk factors responsible for long-term morbidity
was done with Binary logistic regression. Each of the end points
of morbidity, including overall long-term morbidity was assessed in
a separate multivariate model in the regression analysis, with same
variables entered into each model as risk factors. Data analysis was
done with SPSS for Windows software (version 13; Chicago, Illinois).
Significance assigned to 0.05 at 95% confidence interval.
RESULTS
A total of 200 patients, median age 54 years (range 20–83),
male; n =64 (32%), with BDI sustained at LC were referred at a me-
TABLE 1. Description of Laparoscopic Bile Duct Injuries
Strasberg Classification of Bile Duct Injuries (n =200) Strasberg Subgroups of Major Bile Duct Injuries (n =144)
No. Patients No. Patients
Type Description (%) Subgroup Description (%)
Type A Bile leak from cystic duct 19 (9.5%) E 1 Site of CBD division is 20 (14%)
stump or the gallbladder bed >2 cm from the bifurcation
Type B Right segmental duct division 02 (1%) E 2 Site of CBD division is 65 (45%)
where both ends are clipped <2 cm from the bifurcation
Type C Right segmental duct division 09 (4.5%) E 3 Site of CBD division is at 37 (26%)
where the hepatic end remains open the bifurcation
Type D Lateral wall injury to the 26 (13%) E 4 Division or injury to the left, 19 (13%)
common bile duct right or both hepatic ducts
Type E Major CBD division/stricture 144 (72%) E 5 An injury of a right segmental 03 (2%)
with 5 subdivisions duct along with a type E3/E4 injury
Total 200 144
CBD indicates common bile duct.
dian interval of 14 days (0 days– 88months). The median follow-up
after referral was 60 (range 12–212) months. The indication for LC
was chronic symptomatic gallstones in 197 and acute cholecystitis
in 3. The laparoscopic procedure was converted to open in 88 (44%)
patients. The surgical anatomy around Calot’s triangle was described
as straight forward in 52% and difficult in 30%, and was not recorded
in 18%. Overt bleeding was documented in 9 (4%) patients. A total
of 144 (72%) patients sustained serious type E Strasberg (excision)
injuries, followed by type D lateral wall injuries in 27 (13%) patients
(Table 1). Twenty patients (10%) had concomitant vascular injuries
(Table 2), most commonly right hepatic artery ligation/clipping. Mor-
tality as a direct eventual consequence of the bile injury was 4.5%
(n =9) in the entire group. Details of patients died from BDI-related
complications are shown in Table 3. Of these, 7 patients died after
early surgical or nonsurgical intervention after index LC, whereas 2
deaths occurred late including a patient who underwent liver trans-
plantation for secondary biliary cirrhosis.
Treatment of Bile Duct Injuries
Overall, 157 patients (78%) were managed with surgical inter-
vention after BDI; the majority of them underwent biliary reconstruc-
tion by specialist HBS after referral (112 of 200; 56%) whereas 45
patients (22%) were operated with the aim of repair or reconstruction
at primary centers by non-HBS before referral (Fig. 1). The majority
(106 patients; 95%) who had their first surgical intervention at the
Liver Unit by HBS were treated with hepaticojejunostomy with a 50
to 60 cm jejunal Roux loop, whereas the remainder [6 of 112 (5%);
type D injuries, n =4 and nonexcisional type E injuries, n =2] had
bile duct repair over T-tube. Right hemihepatectomy was performed
in 3 patients by HBS in those who sustained concomitant vascular in-
juries; 2 simultaneously with biliary reconstruction, whereas another
as a delayed procedure after uncontrolled biliary sepsis. In compar-
ison, more patients had repair over T-tube (13 of 45, 29%; type D
injuries, n =8 type E injuries, n =5) when primary surgical in-
tervention was done at the referring centers by non-HBS (Table 4).
Forty-three (22%) patients with BDI were managed with endoscopic
stenting or percutaneous drainage of collections only and considered
nonsurgical management.
Overall Outcomes After Surgical Intervention
The outcomes of BDI managed by surgical intervention
is summarized in Table 5. Surgical interventions carried out by
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
554 |www.annalsofsurgery.com C
2011 Lippincott Williams & Wilkins
Annals of Surgery rVolume 253, Number 3, March 2011 Specialist Repair of Bile Duct Injury
TABLE 2. Vascular Injuries Associated with BDI
Description of Vascular Injury Number of Patients (%)
Right hepatic artery injury 16 (80%)
Right portal vein injury 01 (5%)
Both right hepatic artery and portal vein 02 (10%)
Cystic artery 01 (5%)
Total 20
non-HBS seem to be significantly associated with each of the
morbidity end points assessed in this study. However, this comparison
is not truly comparable to HBS repairs, as only those non-HBS repairs
that subsequently develop complications are referred for specialist
HBS care.
Of the different surgical approaches employed in the 2 groups,
the outcome of bile duct repair over a T-tube seemed less successful
in the hands of non-HBS, but this did not show statistical significance
(P=0.09, Fisher Exact test). Seven patients (7 of 13; 54%) managed
by non-HBS with T-tube repair developed biliary strictures during
follow-up, and underwent biliary reconstruction (n =6) by HBS
at median 18 (4.5–132) months or percutaneous dilatation (n =1).
In comparison, only in 1 of 6 patients operated by HBS with repair
over T-tube developed stricture requiring percutaneous dilatation. The
biliary stricture rate was not different after repair over T-tube and
hepaticojejunostomy, in patients managed by either surgeon group
[strictures after T-tube repair vs hepaticojejunostomy: 7 of 13 vs 24
of 32; P=0.28 (non-HBS) and 1 of 6 vs 18 of 106; P=1.00 (HBS)].
Specialist HBS Repair of BDI
On-table repairs were carried out by the specialist HBS “out-
reach” team on 28 of 112 (25%) patients, whereas an equal propor-
tion of early and late repairs were done during the study period (n
=43;38% and n =41;37%, respectively) (Table 6). In the majority,
early repair was dependent on the referral pattern with the exception
of 2 patients who were referred within 21 days, yet the surgical inter-
TABLE 3. Mortality Consequent to the BDI Sustained at Laparoscopic Cholecystectomy
Strasberg Type Delay Duration
No. of Injury and Referral in of Survival
Patient Additional Injuries from LC Prereferral Management Postreferral Management and Remarks from LC
1E2+RPV,RHA 2 days none Ischaemic right lobe; RHH and biliary reconstruction; 3 days
died from sepsis and multiorgan failure
2A+cystic artery 18 days Percutaneous drainage ERC and stenting, drainage; death from sepsis 26 days
3 A 19 days Percutaneous drainage ERC and stenting; died from sepsis 36 days
4E2+duodenal injury 27 days Drainage of biloma Laparotomy, biliary reconstruction was not done due 72 days
to oedema; repair over T-tube. Died of sepsis
5 E2 27 days Percutaneous drainage PTBD; severe sepsis and died of multiorgan failure 127 days
6 E2 27 days Drainage of biloma Laparotomy, biliry reconstruction was not done due 72 days
to oedema; repair over T-tube. Died of sepsis
7E4+RHA 5 days none Early biliary reconstruction day 6. Later developed 135 days
cholangitic abscess & failed control with
percutaneous drainage. Subsequently underwent
RHH and died of sepsis
8 E2 125 days nil Delayed biliary reconstruction by HBS on day 180. 69 months
Recurrent cholangitis related morbidity
9 E4 212 days Early biliary reconstruction Secondary biliary cirrhosis, underwent orthotopic 129 months
by non-HBS liver transplantation at 121 months
ERCP indicates endoscopic retrograde cholangiography; HBS, hepatobiliary surgeons; PTBD, percutaneous transhepatic biliary drainage; RPV, right por tal vein; RHA, right
hepatic artery; RHH, right hemihepatectomy; non-HBS, nonhepatobiliary surgeons.
vention was postponed to achieve control of associated systemic sep-
sis. Percutaneous transhepatic cholangiography and drainage (PTCD)
was done in 5 patients in the early repair group, whereas transhep-
atic dilatation was carried out only in 8 patients before surgery in
the late repair group. Patients in the late repair group had a longer
follow-up (months; range) compared with on-table and early repair
groups; “late” repair (96; 94–196 months) vs “on-table” (55; 10–186)
vs “early”: (96; 94–196 months). Table 6 also shows the demographic
and injury characteristics of patients operated on by HBS.
Long-term Results of Biliary Reconstruction by
Hepatobiliary Surgeons
The outcome of biliary reconstruction by HBS, in terms of
specific events and overall morbidity, are shown in Table 7. More
patients in the late reconstruction group (12 of 41; 29%) developed
biliary re-stricturing after reconstruction, compared to on-table and
early repairs (p =0.01). As a consequence, more patients in the late
repair group required PTCD with or without stenting (P=0.03).
However, the incidence of recurrent cholangitis and redo biliary re-
constructions was not different in the 3 treatment groups. Overall, 16
(16 of 41; 39%) patients in late reconstruction group had at least 1
complication after the initial BDI during the follow-up, compared to
6 (21%) and 6 (14%) in on-table and early repair groups, respectively
(P=0.02). The effect of length of follow-up on long-term morbidity
was further analyzed. The proportion patients with at least 1 morbid-
ity event at lesser than 5 years follow-up in each treatment group by
HBS (on-table or early and late) was similar [4 of 6 (66%), 4 of 6
(66%), and 12 of 16 (75%), respectively; P=ns].
Risk of Long-term Morbidity After Surgical
Repair/Reconstruction
Univariate Analysis
Age at LC, gender, and time to suspicion of BDI were not
associated with poor long-term outcomes. Adverse overall morbidity
was seen in patients with difficult surgical anatomy at LC (P=0.003)
and those with straight forward dissections at LC (P=0.005)
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
C
2011 Lippincott Williams & Wilkins www.annalsofsurgery.com |555
Per era et al Annals of Surgery rVolume 253, Number 3, March 2011
FIGURE 1. Referrals and treatment pattern of the laparoscopic bile duct injuries.
TABLE 4. Surgical Treatment of Bile Duct Injuries (n =157)
On-Table Repair Early (<21 days) Late (>21 days) Repair
or Reconstruction Repair or Reconstruction or Reconstruction Total
Patients operated by HBS 28 43 41 112
Bile duct repair (01) (05)
Reconstruction (27) (38)(41)
Patients operated by non-HBS 16 26 03 45
Bile duct repair (04) (09)
Reconstruction (12) (17) (03)
Total 44 69 44 157
HBS indicates hepatobiliary surgeons; non-HBS, nonhepatobiliary surgeons.
TABLE 5. Summary of Outcomes After Surgical Intervention
to BDI; Results by Surgeon Group
Non-HBS HBS Significance
(n =45) (n =112) (Fisher Exact Test)
Stricture (%) 31 (69%) 19 (17%) <0.001
Recurrent cholangitis (%) 15 (33%) 12 (11%) <0.001
Intervention/dilatation (%) 23 (51%) 16 (14%) <0.001
Redo reconstruction (%) 24 (53%) 4 (3%) <0.001
Overall morbidity (%) 37 (82%) 28 (25%) <0.001
on univariate analysis. The extent of the injury according to the Stras-
burg classification was also not an independent risk factor for long-
term morbidity (Table 8). Both early refer rals (<21 days) and delayed
referrals (>21 days) were associated with significant long-term mor-
bidity on univariate analysis (24 of 84; 29%: P=0.001 and 53 of 86;
62%:P=0.001, respectively). Furthermore, delayed referrals (n =
86) were significantly associated with biliary strictures (n =41 of 86,
47%; P=0.001) comprising the following: 23 patients presenting to
HBS with biliary strictures having been operated on by non-HBS be-
fore referral, 6 injuries managed conservatively, 12 strictures during
follow-up after definitive late repairs by HBS). Nonsurgical interven-
tions were also seen more often after delayed referral (n =35 of 86,
41%; P=0.007) comprising the following: 9 diagnostic PTCD after
delayed referral to HBS, 2 ERCPs, and 24 therapeutic PTCD of which
15 were after non-HBS repair and 9 were done after specialist HBS
repair.
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
556 |www.annalsofsurgery.com C
2011 Lippincott Williams & Wilkins
Annals of Surgery rVolume 253, Number 3, March 2011 Specialist Repair of Bile Duct Injury
TABLE 6. Specialist Repair of Bile Duct Injuries—Group Characteristics
On-Table Repair (Group A) Early Repair (Group B) Late Repair (Group C)
Number of patients (%) 28 (25%) 43 (38%) 41 (37%)
Median age (range) 60 (22–74) years 55 (29–80) years 55 (20–74) years
No. Type E injuries (%) 26/28 (93%) 36/43 (84%) 36/41 (88%)
Presentation
Bile leak - 28/43 19/41
Jaundice - 15/43 25/41
Median time to referral (days) 0 7 (1–20) 127 (14–2267)
Endoscopic interventions (ERC) +/stenting - 23/43 32/41
Percuataneous interventions (PTC/D) - 5/43 8/43
Additional drainage procedures -
Surgical (laparotomy) - 7/43 5/41
Image guided - 2/43 5/41
Median follow-up 55 (10–186) months 55 (5–194) months 96 (94–196) months
ERC indicates endoscopic retrograde cholangiography; PTC/D, percutaneous transhepatic cholangiography/drainage.
TABLE 7. Results of Biliary Reconstruction by HBS
On-Table Repair (n =28) Early Repair (n =43) Late Repair (n =41) Significance
Stricture (%) 5 (18%) 2 (5%) 12 (29%) 0.01
Recurrent cholangitis (%) 3 (11%) 5 (12%) 4 (10%) 0.96
Intervention/dilatation (%) 4 (14%) 2 (5%) 10 (24%) 0.03
Redo reconstruction (%) 1 (4%) 1 (2%) 2 (5%) 0.81
Overall morbidity (%) 6 (21%) 6 (14%) 16 (39%) 0.02
Significant at 95% confidence interval level; χ2test.
Timing of repair (on-table vs early vs late) by different surgeon
groups (specialist HBS vs non-HBS) predicted adverse long-term out-
come on univariate analysis; on-table and early surgical repairs done
by non-HBS carried higher incidence of overall morbidity [overall
morbidity: 12 of 16 (75%); P=0.007 and 22 of 26 (85%); P=0.001,
respectively]. In contrast, only late repairs performed by HBS pre-
dicted unfavorable outcome on univariate model [overall morbidity:
19 of 41 (46%); P=0.011).
Multivariate Analysis
LC’s described as difficult dissections carried an increased
risk of overall morbidity (P=0.01) and carried a much higher risk
of subsequent re-stricturing of biliary system (P=0.007) (Table 9).
Concurrent vascular injuries, though not significant as an independent
predictor on univariate analysis, seem to have a combined effect with
other variables on morbidity and increased the risk of biliary strictures
(8 of 20; P=0.006) contributing to increased overall morbidity (11
of 20; P=0.018). Although recognition delay of BDI was not a
predictor of adverse outcome, late referrals beyond 21 days of injury
were associated with an increased likelihood of having to undergo
redo biliary reconstruction (14 of 86, 16%; P=0.008) and increased
overall morbidity (53 of 86, 62%; P=0.001).
On-table repair done by non-HBS was an independent risk
factor for recurrent cholangitis (8 of 16, 50%: P=0.001), biliary
strictures (12 of 16, 75%; P=0.001), redo reconstructions (5 of
16, 31%; P=0.001), and overall morbidity (12 of 16, 75%; P=
0.014). Similar prognostic outcome was observed in those who un-
derwent early (<21 days) surgical intervention at the non-HBS centers
[restricture (16 of 26, 61%; P=0.001), redo reconstructions (16 of
26, 61%; P=0.001), and overall morbidity (22 of 26, 84%; P=
0.001)].
In contrast, on-table and early reconstructions carried out by
specialist HBS were not associated with significant long-term mor-
bidity. However, late specialist biliary reconstruction was associated
with adverse outcomes on multivariate analysis, with an increased
likelihood of strictures (12 of 41, 29%; P=0.001), nonsurgical in-
terventions (9/41, 22%; p =0.001), and redo reconstructions (2 of
41, 4%; P=0.024)
DISCUSSION
The early high rate of biliary injuries after LC was initially
attributed to the learning curve of a new procedure. However, the
incidence of such injuries has remained unchanged, well over 15
years after the widespread acceptance of the procedure.2,5,17,18 The
“critical window of safety” and “common error traps” during the
index LC have been well documented during the last decade, that
should reduce the incidence of such injuries; part of this study predates
this practice and it was not possible to report on the use of these
approaches by referring surgeons.19,20 The complexity of injuries
sustained at LC often poses challenges for subsequent management,
and are better managed at specialist HBS centers.21 Performing a
biliary enteric anastomosis is technically demanding in the absence
of dilated bile ducts. Although injuries seem fresh in the immediate
and early postinjury phase, a technically sound anastomosis is not
feasible in all situations especially if the surgeons do not have the
required expertise and facilities.
Recent interest has centered on the timing of definitive surgi-
cal intervention after laparoscopic BDI.22 Immediate recognition of
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
C
2011 Lippincott Williams & Wilkins www.annalsofsurgery.com |557
Per era et al Annals of Surgery rVolume 253, Number 3, March 2011
TABLE 8. Univariate Analysis of Risk Factors (χ2)
Cholangitis Strictures Nonsurgical Interventions Redo Reconstructions Overall Morbidity
Variable (no. patients) No. Patients PNo. patients PNo. Patients PNo. Patients PNo. Patients P
Intraoperative findings
Straight forward (104) 13 ns 25 0.046 22 ns 09 ns 33 0.005
Difficult anatomy (59) 13 0.044 26 0.011 23 0.028 13 0.02 34 0.003
Bleeding (09) 02 ns 02 ns 02 ns 01 ns 04 ns
Injury type
E (144) vs Non-E (56) 23 vs 5 ns 48 vs 13 ns 43 vs 14 ns 21 vs 5 ns 64 vs 20 ns
E 3,4,5 (59) vs E 1,2 (85) 11 vs 12 ns 21 vs 27 ns 20 vs 23 ns 8 vs 13 ns 29 vs 35 ns
E 3,4,5 (59) vs Other (141) 11 vs 17 ns 21 vs 40 ns 20 vs 37 ns 8 vs 18 ns 29 vs 55 ns
Associated vascular injury
Yes (20) 2 ns 08 ns 08 ns 3 ns 11 ns
No (180) 26 ns 53 ns 49 ns 23 ns 73 ns
Initial recognition of Injury
Immediate (49) 11 ns 18 ns 19 ns 06 ns 20 ns
Early (125) 15 ns 34 ns 32 ns 19 ns 49 ns
Late (26) 02 ns 09 ns 06 ns 01 ns 15 ns
Referral delay
No delay (30) 3 ns 03 ns 04 ns 02 ns 07 ns
Early referral (84) 10 ns 17 ns 18 ns 10 ns 24 0.001
Late referral (86) 15 ns 41 0.001 35 0.007 14 ns 53 0.001
Repair at HBS unit
On-table (28) 03 ns 05 ns 04 ns 01 ns 06 ns
Early (43) 05 ns 02 ns 02 ns 01 ns 06 ns
Late (41) 04 ns 12 0.018 10 ns 02 ns 16 0.011
Repair at Non-HBS unit
On-table (16) 08 0.001 12 0.001 11 0.001 05 0.04 12 0.007
Early (26) 07 ns 16 0.001 10 ns 16 0.001 22 0.001
Late (03) 00 ns 03 0.027 02 ns 01 ns 03 ns
ns indicates not significant; HBS, hepatobiliary surgeons; non-HBS, nonhepatobiliary surgeons.
such injuries at the time of index laparoscopy provides the option
of “on-table” repair and a better overall outcome is expected.22 Al-
though the definition of “early repair” remains unclear, an increasing
number of surgeons tend to advocate this where the definitive surgi-
cal intervention is carried out in the absence of sepsis and minimal
peritoneal contamination.10,22 In the literature, early repairs have been
referred to as operations carried out within many hours to weeks, and
results of “on-table” versus early and late repair have been reported in
these settings. However, limited data exists in the literature on
long-term outcome after on-table or early repair of laparoscopic BDI,
and reported results are contrasting.22–25
In the first part of the outcome analysis, we compared the
results of biliary reconstruction at a specialist center with the focus on
timing of such repair. We have shown that on-table and early repairs
done by HBS have a better overall outcome at 5 years, compared
to late repairs. The delay in referral may have initiated a process
of ongoing inflammation and scarring of the injured bile ducts and
includes a significant number of patients with failed non-HBS repairs,
both of which may have contributed to the inferior outcome in the
late repair group. Those in the late repair group had comparatively
longer follow-up and it is possible that this may have an effect on
overall results. Interestingly, a majority of patients with morbidity in
all 3 treatment groups had a follow-up beyond 5 years after repair,
recognizing the need for even longer surveillance, at least up to 7 to
10 years.
Previous authors have suggested surgical revision of failed
repairs to be more likely when primary intervention was carried out
either intraoperatively (immediate repair) or at an earlier time.25–27
Most of these conclusions have been centered on a case mix where
surgical interventions were done byboth HBS and non-HBS surgeons,
and this may have resulted in skewed results. In our series, we have
demonstrated better overall outcome in long-term when immediate
and early surgical repair was done by specialist HBS. Opponents
of early bile duct repair argue that the presence of inflammation
and the absence of duct dilatation make it difficult to assess the
extent of ischaemic damage and perform a high biliary anastomosis,
resulting in an increased subsequent late ischaemic stricture rate.23
This argument may be true in the case of immediate repair, as even
in the expert hands the stricture rate of this treatment group was
relatively higher in our study. In case of early repairs, our results
albeit with a median follow-up of 5 years contradict this argument,
and support the approach to repair these injuries early rather than
waiting till bile duct dilatation develops.
We have demonstrated, through univariate and multivariate
analysis that on-table and early surgical intervention carried out by
non-HBS stands out as an independent predictor of recurrent cholan-
gitis and biliary strictures leading to increased incidence of non-
surgical interventions and revision surgery.28,29 Therefore, results of
on-table and early repair of BDI at these centers cannot be similar to
those performed at specialist hepatobiliary centers. The provision of
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
558 |www.annalsofsurgery.com C
2011 Lippincott Williams & Wilkins
Annals of Surgery rVolume 253, Number 3, March 2011 Specialist Repair of Bile Duct Injury
TABLE 9. Multivariate Analysis of Risk Factors (Binary Logistic Regression)
Cholangitis Strictures Nonsurgical Interventions Redo Reconstructions Overall Morbidity
Variable (No. Patients) No. Patients PNo. Patients PNo. Patients PNo. Patients PNo. Patients P
Intraoperative findings
Straight forward (104) 13 ns 25 ns 22 ns 09 ns 33 ns
Difficult Anatomy (59) 13 ns 26 0.007 23 ns 13 ns 34 0.01
Bleeding (09) 02 ns 02 ns 02 ns 01 ns 04 ns
Injury type
E (144) vs Non-E (56) 23 vs 5 ns 48 vs 13 ns 43 vs 14 0.01 21 vs 5 ns 64 vs 20 ns
E 3,4,5 (59) vs E 1,2 (85) 11 vs 12 ns 21 vs 27 ns 20 vs 23 ns 8 vs 13 ns 29 vs 35 ns
E 3,4,5 (59) vs other (141) 11 vs 17 ns 21 vs 40 ns 20 vs 37 ns 8 vs 18 ns 29 vs 55 ns
Associated vascular injury
Yes (20) 2 ns 08 0.006 08 0.015 3 ns 11 0.018
No (180) 26 ns 53 ns 49 23 ns 73 ns
Initial recognition of injury
Immediate (49) 11 ns 18 ns 19 ns 06 ns 20 ns
Early (125) 15 ns 34 ns 32 ns 19 ns 49 ns
Late (26) 02 ns 09 ns 06 ns 01 ns 15 ns
Referral delay
No delay (30) 3 ns 03 ns 04 ns 02 ns 07 ns
Early referral (84) 10 ns 17 ns 18 ns 10 ns 24 ns
Late referral (86) 15 ns 41 ns 35 ns 14 0.008 53 0.001
Repair at HBS unit
On-table (28) 03 ns 05 ns 04 ns 01 ns 06 ns
Early (43) 05 ns 02 ns 02 ns 01 ns 06 ns
Late (41) 04 ns 12 0.001 10 0.001 02 0.024 16 ns
Repair at non-HBS unit
On-table (16) 08 0.001 12 0.001 11 ns 05 0.001 12 0.014
Early (26) 07 0.07 16 0.001 10 ns 16 0.001 22 0.001
Late (03) 00 03 - 02 ns 01 03
ns indicates not significant; HBS, hepatobiliary surgeons; non-HBS, nonhepatobiliary surgeons.
specialist HBS outreach services could address most of the problems
encountered with regards to surgical expertise and access, allowing
the patient to be managed at the referring hospital, and preventing the
need for an additional procedure.15
Most late complications postrepair can be managed by a com-
bined multidisciplinary approach using interventional radiology (eg,
percutaneous transhepatic biliary dilatation/PTD).30,31 Overall, 16
(14%) patients who had biliary reconstructions by HBS in this study
needed at least 1 episode of nonsurgical intervention after initial bile
duct repair, although the reported incidence of such interventions is
10% to 15% in the literature.26,32 The majority of patients (63%) who
required a nonsurgical intervention in our series had late surgical re-
pair after BDI. This multidisciplinary service can only be provided
at tertiary referral centers, which further emphasize the need for spe-
cialist intervention in complex BDI sustained at LC.
CONCLUSIONS
Early and immediate biliary reconstruction after laparoscopic
BDI have an equally good or better long-term outcome when per-
formed at a specialist HBS center compared to injuries that are re-
paired late. A delay in referral to a specialist team and the presence of
associated vascular injuries contribute to adverse overall morbidity.
Immediate assistance sought from HBS or early referral to such cen-
ters may reduce the long-term morbidity of this debilitating injury
that continues to be the drawback of laparoscopic cholecystectomy.
REFERENCES
1. Sicklick JK, Camp MS, Lillemoe KD, et al. Surgical management of bile duct
injuries sustained during laparoscopic cholecystectomy: perioperative results
in 200 patients. Ann Surg. 2005;241:786–792.
2. Soper NJ, Brunt LM, Kerbl K. Laparoscopic general surgery. NEngJMed.
1994;330:409–419.
3. McMahon AJ, Russell IT, Baxter JN, et al. Laparoscopic versus minilaparotomy
cholecystectomy: a randomised trial. Lancet. 1994;343:135–138.
4. Dunn D, Nair RFS, McCloy R. Laparoscopic cholecystectomy in England and
Wales: results of an audit by the Royal College of Surgeons of England. Ann
R Coll Surg Eng. 1994;76:269–275.
5. Russell RCG. General surgery: biliary surgery. BMJ. 1993;307:1266–1269.
6. de Reuver PR, Rauws EA, Bruno MJ, et al. Survival in bile duct injury patients
after laparoscopic cholecystectomy: a multidisciplinary approach of gastroen-
terologists, radiologists, and surgeons. Surgery. 2007;142:1–9.
7. de Santiba˜
nes E, Palavecino M, Ardiles V, . Bile duct injuries: management of
late complications. Surg Endosc. 2006;20:1648–1653.
8. Schiano Di Visconte M. Analysis of pathogenetic mechanisms of common bile
duct iatrogenic lesion during laparoscopic cholecystectomy. A review of the
literature. Minerva Chir. 2002;57:663–667.
9. Chaudhary A, Manisegran M, Chandra A, et al. How do bile duct injuries
sustained during laparoscopic cholecystectomy differ from those during open
cholecystectomy? J Laparoendosc Adv Surg Tech. 2001;11:187–191.
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
C
2011 Lippincott Williams & Wilkins www.annalsofsurgery.com |559
Per era et al Annals of Surgery rVolume 253, Number 3, March 2011
10. McLean TR. Monetary lessons from litigation involving laparoscopic chole-
cystectomy. Am Surg. 2005;71:606–612
11. Thomson BN, Parks RW, Madhavan KK, et al. Liver resection and trans-
plantation in the management of iatrogenic biliary injury. World J Surg.
2007;31:2363–2369.
12. Kapoor VK. Bile duct injury repair: when? what? who? J Hepatobiliary Pan-
creat Surg. 2007;14:476–479.
13. Walsh RM, Henderson JM, Vogt DP, et al. Long-term outcome of biliar y recon-
struction for bile duct injuries from laparoscopic cholecystectomies. Surgery.
2007;142:450–456.
14. Lillemoe KD, Pitt HA, Cameron JL. Postoperative bile duct strictures. Surg
Clin North Am. 1990;70:1355–1380.
15. Silva MA, Coldham C, Mayer AD, et al. Specialist outreach service for on-
table repair of iatrogenic bile duct injuries–a new kind of ‘travelling surgeon.’
Ann R Coll Surg Engl. 2008;90:243–246.
16. Strasberg SM, Hertt M, Soper NJ. An analysis of the problems of the bil-
iary injury during laparoscopic cholecystectomy. J Am College Surgeons.
1995;180:101–125.
17. Savader SJ, Lillemoe KD, Prescott CA, et al. Laparoscopic cholecystectomy
related bile duct injuries: a health and financial disaster. Ann Surg. 1997;225:
268–273.
18. Walsh RM, Henderson JM, Vogt DP, et al. Trends in bile duct injuries during
laparoscopic cholecystectomy. J Gastrointest Surg. 1998;2:458–462.
19. Strasberg SM. Avoidance of biliary injury during laparoscopic cholecystec-
tomy. J Hepatobiliary Pancreat Surg. 2002;9:543–547.
20. Strasberg SM. Error traps and vasculo-biliary injury in laparoscopic and open
cholecystectomy. J Hepatobiliary Pancreat Surg. 2008;15:284–292.
21. Goykhman Y, Kory I, Small R, et al. Long-term outcome and risk fac-
tors of failure after bile duct injury repair. J Gastrointest Surg. 2008;12:
1412–1417.
22. Thomson BN, Parks RW, Madhavan KK, et al. Early specialist repair of biliary
injury. Br J Surg. 2006;93:216–220.
23. Kohneh Shahri N, Lasnier C, Paineau J. Bile duct injuries at laparoscopic
cholecystectomy: early repair results [Article in French]. Ann Chir. 2005;130:
218–223.
24. Mercado MA, Chan C, Orozco H, et al. Acute bile duct injury. Surg Endosc.
2003;17:1351–1355.
25. Mercado MA. Early versus late repair of bile duct injuries. Surg Endosc.
2006;20:1644–1647.
26. Walsh RM, Vogt DP, Ponsky JL, et al. Management of failed biliary repairs for
major bile duct injuries after laparoscopic cholecystectomy. J Am Coll Surg.
2004;199:192–197.
27. de Reuver PR, Grossmann I, Busch OR, et al. Referral pattern and timing of
repair are risk factors for complications after reconstructive surgery for bile
duct injury. Ann Surg. 2007;245:763–770.
28. Wudel LJ Jr, Wright JK, Pinson CW, et al. Bile duct injury following
laparoscopic cholecystectomy: a cause for continued concern. Am Surg.
2001;67:557–563.
29. Mirza DF, Narsimhan KL, Ferraz Neto BH, et al. Bile duct injury following
laparoscopic cholecystectomy: referral pattern and management. Br J Surg.
1997;84:786–790.
30. Lillemoe KD, Martin SA, Cameron JL, et al. Major bile duct injuries during
laparoscopic cholecystectomy. Follow-up after combined surgical and radio-
logical management. Ann Surg. 1997;225:459–468.
31. Misra S, Melton GB, Geschwind JF, et al. Percutaneous management
of bile duct strictures and injuries associated with laparoscopic cholecys-
tectomy: a decade of experience. J Am Coll Surg. 2004;198:218–226.
32. Shah SR, Mirza DF, Afonso R, et al. Changing referral pattern of biliary
injuries sustained during laparoscopic cholecystectomy. Br J Surg. 2000;87:
890–891.
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
560 |www.annalsofsurgery.com C
2011 Lippincott Williams & Wilkins
    • "Schmidt et al. found on univariate and mutivariate analyses that associated arterial injury and repair in the presence of active peritonitis were associated with increased complications after BDI [9]. However the delayed repair view is not universal, with many authors supporting early repairs [1,10,11] . A recent paper has emphasised that early repair by a specialist Hepatobiliary surgeon is associated with equivalent stricture rates to delayed repair but improved quality of life, return to normal activities and lower cost [12]. "
    [Show abstract] [Hide abstract] ABSTRACT: Right hepatic arterial injury (RHAI) is the most common vascular injury sustained during laparoscopic cholecystectomy, occurring in up to 7% of cholecystectomies. RHAI is also the most common vascular injury associated with a bile duct injury (BDI) and is reported to occur in up to 41 - 61% of cases when routine angiography is employed following a BDI. We present an unusual case of erosion of vascular coils from a previously embolised right hepatic artery into bilio-enteric anastomoses causing biliary obstruction. This is on a background of biliary reconstruction following a major BDI. A 37-year old man underwent a bile duct reconstruction following a major BDI (Strasberg-Bismuth E4 injury) sustained at laparoscopic cholecystectomy. He had two separate bilio-enteric anastomoses of the right and left hepatic ducts and had a modified Terblanche Roux-en-Y access limb formed. Approximately three weeks later he was admitted for significant gastrointestinal bleeding and was hypotensive and anaemic. Selective computed tomography angiography revealed a 2 x 2 centimetre right hepatic artery pseudoaneurysm, which was urgently embolised with radiological coils. Two months later he developed intermittent fevers, rigors, jaundice, and right upper quadrant pain with evidence of intrahepatic biliary dilatation on magnetic resonance cholangiopancreatography. The degree of intrahepatic biliary dilatation progressively increased on subsequent imaging over several months, suggesting stricturing of the bilio-enteric anastomoses. Several attempts to traverse these strictures with a percutaneous transhepatic approach had failed. Then, approximately ten months after the initial BDI repair, choledochoscopy through the Terblanche access limb revealed multiple radiological coils within the bilio-enteric anastomoses, which had eroded from the previously embolised right hepatic artery. A laparotomy was performed to remove the coils, take down the existing obstructed bilio-enteric anastomoses and revise this. Following this the patient recovered uneventfully. Obstructive jaundice and cholangitis secondary to erosion of angiographically placed embolisation coils is a rarely described complication. In view of the relative frequency of arterial injury and complications following major bile duct injury, we suggest that these patients be formally assessed for associated arterial injury following a major BDI.
    Full-text · Article · Apr 2015
    • "With recent advances in endoscopic techniques, endoscopic stenting has emerged as an effective treatment option for postoperative biliary stricture789. There are some ineffective cases, however, in which percutaneous transhepatic biliary drainage (PTBD) catheter dilatation [10, 11] or surgical interventions are required121314151617181920212223. The aim of this study was to review our experiences with PTBD catheter dilatation and/or surgery for biliary stricture following Lap-C, and to discuss its feasibility and clinical efficacy. "
    [Show abstract] [Hide abstract] ABSTRACT: Background Biliary strictures following laparoscopic cholecystectomy (Lap-C), which are often associated with vascular injuries, remain a serious problem to manage. The aim of this study was to review our experiences with postoperative biliary stricture.Methods This study involved 14 consecutive patients with biliary strictures that resulted from bile duct injuries during Lap-C between 1997 and 2013. Their medical records were retrospectively analyzed.ResultsPercutaneous transhepatic biliary drainage (PTBD) catheter dilatation was first attempted in eight patients, and five patients were successfully treated. Biliary re-stricture recurred in one patient after 34-month follow-up period. This patient underwent repeated catheter dilatations, which led to recurrent stricture resolution. All five patients maintained biliary tract patency over 72-month follow-up period. The remaining nine patients underwent surgical procedures, including hepaticojejunostomy in two patients, re-hepaticojejunostomy in two patients, and the remaining five patients, with biliary strictures involving the secondary biliary branch and concomitant vascular injuries underwent right hemihepatectomy with cholangiojejunostomy. There were no major postoperative complications. After 80-month follow-up period, all nine patients were alive without biliary stricture.ConclusionsPTBD catheter dilatation is recommended first for postoperative Lap-C-associated biliary strictures. In complicated injury patients with vascular injuries, right hemihepatectomy with cholangiojejunostomy should be indicated.
    Article · Dec 2014
    • "Vascular evaluation should be performed routinely because the presence and type of vascular injury can modify the surgical technique and management; • repair in centers of expertise: management of patients in a referral (or tertiary) center is a factor of success in BDI repair [10] [37] [56] [57]; • early vs. late repair: in the review by Reuver et al. published in 2007 [58], immediate repair of BDI was identified as an independent factor for failure of repair. Conversely, Stewart et al. [58] and Perera et al. [59] concluded that immediate repair provided identical if not superior results, compared to delayed repair, if it was performed by a specialist and under good conditions (i.e. no infec- tion). "
    [Show abstract] [Hide abstract] ABSTRACT: Late complications arising after bile duct injury (BDI) include biliary strictures, hepatic atrophy, cholangitis and intra-hepatic lithiasis. Later, fibrosis or even secondary biliary cirrhosis and portal hypertension can develop, enhanced by prolonged biliary obstruction associated with recurrent cholangitis. Secondary biliary cirrhosis resulting in associated hepatic failure or digestive tract bleeding due to portal hypertension is a substantial risk factor for morbidity and mortality after bile duct repair. Parameters that determine the management of late complications of BDI include the type of biliary injury, associated vascular injury, hepatic atrophy, the presence of intra-hepatic strictures or lithiasis, repetitive infectious complications, the quality of underlying parenchyma (fibrosis, secondary biliary cirrhosis) and the presence of portal hypertension. Endoscopic drainage is indicated for patients with uncontrolled acute sepsis, patients at high operative risk, patients with cirrhosis who are not eligible for liver transplantation and patients who have previously undergone several attempts at repair. Roux-en-Y hepaticojejunostomy, whether de novo or as an iterative repair, is the technique of reference for post-cholecystectomy BDI. Hepatic resection is indicated in only rare instances, mainly in case of extended hilar stricture, multiple stone retention in one sector of the liver or in patients for whom the repair is deemed technically difficult. Liver transplantation is indicated only in exceptional circumstances, when secondary biliary cirrhosis is associated with liver failure and portal hypertension.
    Full-text · Article · Jun 2014
Show more