biliary complications and outcomes of liver transplantation from donors after cardiac death

ORIGINAL ARTICLE

Biliary Complications and Outcomes of LiverTransplantation from Donors after CardiacDeathAnurag Maheshwari,1 Warren Maley,2 Zhiping Li,1 and Paul J. Thuluvath1

1Section of Hepatology and 2Division of Transplant Surgery, The Johns Hopkins University School ofMedicine, Baltimore, MD

Biliary complications after liver transplantation (LT) using organs retrieved from donors after cardiac death are not wellcharacterized. The aim of this study was to evaluate the severity of biliary complications and outcomes after donation aftercardiac death liver transplantation (DCD-LT). A retrospective evaluation of 20 DCD-LTs from 1997-2006 was performed. Therecipient age was 53 � 8.7, and the donor age was 35 � 11 years. The warm ischemia time, cold ischemia time, peak alanineaminotransferase level, and peak aspartate aminotransferase level were 33 � 12 minutes, 8.7 � 2.7 hours, 1757 � 1477 U/L,and 4020 � 3693 U/L, respectively. The bilirubin and alkaline phosphatase levels at hospital discharge after LT were 3.2 � 5.4mg/dL and 248 � 200 U/L, respectively. During a median follow-up of 7.5 months (range: 1-73), 5 patients (25%; 1 death afterre-LT) died (3 from sepsis, 1 from recurrent hepatocellular carcinoma at 4 months, and 1 from a cardiac event at 46 months),and additionally, 4 patients (20%) required re-LT (1 because of hepatic artery thrombosis, 1 because of primary graftnonfunction, and 2 because of biliary strictures). Twelve (60%) developed biliary complications, and of these, 11 (55%) hadserious biliary complications. The biliary complications were as follows: a major bile leak for 2 patients (10%; both eventuallyunderwent retransplantation), anastomotic strictures for 5 patients (25%), hilar strictures for 7 patients (35%), extrahepaticdonor duct strictures for 9 patients (45%), intrahepatic strictures for 10 patients (50%), stones for 1 patients (5%), casts for 7patients (35%), and debris for 2 patients (10%). More than 1 biliary complication was seen in most patients, and these wereunpredictable and required multiple diagnostic or therapeutic procedures. Serious biliary complications are common afterDCD-LT, and research should focus on identifying donor and recipient factors that predict and prevent serious biliarycomplications. Liver Transpl 13:1645-1653, 2007. © 2007 AASLD.

Received January 29, 2007; accepted March 28, 2007.

See Editorial on 633

Donation after cardiac death (DCD) donors are a signif-icant segment of the extended-criteria donors beingused for liver transplantation (LT) because of the ongo-ing shortage of donors.1,2 An increasing number of or-gan procurement organizations have developed proto-cols for the use of DCD organs. Analyses of the United

Network for Organ Sharing data set have shown in-creased graft loss and retransplantation rates after do-nation after cardiac death liver transplantation (DCD-LT) in comparison with conventional donation afterbrain death liver transplantation (DBD-LT).1,3,4 The in-creased graft failure is unrelated to modifiable donor orrecipient factors.4 However, a recent analysis has sug-gested that the graft survival of DCD-LT is similar tothat of DBD-LT when the donor age is greater than 60

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CIT, cold ischemia time; DBD, donation after braindeath; DBD-LT, donation after brain death liver transplantation; DCD, donation after cardiac death; DCD-LT, donation after cardiacdeath liver transplantation; ERC, endoscopic retrograde cholangiography; ERCP, endoscopic retrograde cholangiopancreatography;HAT, hepatic artery thrombosis; HCC, hepatocellular carcinoma; HTK, histidine-trytophan-ketoglutarate; LT, liver transplantation;MELD, model for end-stage liver disease; MRC, magnetic resonance cholangiography; MRCP, magnetic resonance cholangiopancre-atography; PTC, percutaneous transhepatic cholangiography; UW, University of Wisconsin; VRE, vancomycin-resistant Enterococcus;WIT, warm ischemia time.Address reprint requests to Paul. J. Thuluvath, M.D., F.R.C.P., Division of Gastroenterology and Hepatology, The Johns Hopkins University Schoolof Medicine, 1830 E. Monument Street, Suite 430, Baltimore, MD 21205. Telephone: 410 614 5389; FAX: 410 614 9612; E-mail: [email protected]

DOI 10.1002/lt.21212Published online in Wiley InterScience (www.interscience.wiley.com).

LIVER TRANSPLANTATION 13:1645-1653, 2007

© 2007 American Association for the Study of Liver Diseases.

years in both groups, and furthermore, the graft sur-vival of a DCD liver is similar to that of a split liver afteran adjustment for age.1 Markov modeling would sug-gest that DCD-LT may be life-saving for many recipientswhen their model for end-stage liver disease (MELD)score is more than 24, despite a hypothetically lower1-year patient survival rate (73% versus 65%).5 Tworecent studies have attempted to identify recipient anddonor combinations that would benefit from the use ofDCD organs and achieve outcomes similar to those ofDBD-LT. Mateo et al.6 defined a recipient cumulativerelative risk, using risk factors identified from a Coxregression analysis. They hypothesized that low-riskrecipients who underwent transplantation with low-risk DCD livers [cold ischemia time (CIT) � 10 hoursand warm ischemia time (WIT) � 30 minutes] wouldachieve 1-year and 3-year graft survival rates similar tothose of recipients of donation after brain death (DBD)allografts. Lee et al.7 found that a combination of favor-able donor characteristics (donor age � 45 years, CIT �10 hours, and WIT � 15 minutes) could achieve 1-year,3-year, and 5-year patient and graft survival rates com-parable to those for DBD-LT, regardless of the recipi-ent’s condition. On the basis of these data, it has beensuggested that DCD organs could be used judiciouslyamong stable recipients with a minimization of CIT.Although these strategies remain unproven, there isincreasing recognition that DCD-LT could play a signif-icant role in expanding the donor pool with acceptableoutcomes in selected recipients.

Several studies have shown a higher incidence ofserious biliary complications (6%-19%) after DCD-LTversus DBD-LT (Figure 1).8,9 The incidence of biliarycomplications for recipients after DBD-LT ranges from8%-20%.10-12 However, these complications have min-imal impact on graft or patient survival, except in thosewho develop diffuse ischemic intrahepatic stric-tures.12,13 The impact of biliary complications afterDCD-LT on graft and patient survival, morbidity, andhealth-care utilization has not been well characterized.In this study, our objective was to carefully analyze thenature of biliary complications and its impact on graftand patient survival after DCD-LT.

PATIENTS AND METHODS

We performed a retrospective review of all liver trans-plants performed at the Johns Hopkins Hospital, usingDCD donor organs from March 1997 to March 2006.During this period, there were 20 LTs performed withDCD livers, with 17 in the post-MELD era (i.e., 2002onward). The etiology of donor death was anoxic en-cephalopathy in a majority of cases, with 4 cases ofstroke and intracerebral hemorrhage as the cause ofirreversible neurological injury.

Organ Procurement and Preservation.

All donor organ retrievals were performed in a con-trolled fashion, with the extubation of donors and sub-sequent death in the operating room followed by rapidorgan recovery to minimize WIT. Where possible, do-

nors were brought to the operating room for the with-drawal of life support, and a nontransplant physiciandeclared the time of death. Heparin was infused intra-venously to prevent microemboli, and our protocol didnot involve the routine use of phentolamine. Organrecovery began 5 minutes after the declaration of car-diac death to ensure irreversible brain death. The ces-sation of spontaneous respirations and the lack of apulse (measured by arterial line insertion) were used ascriteria for the cessation of cardiopulmonary function.If death did not occur within 1 hour of the withdrawal oflife support, the organ donation was cancelled, and thepatient was returned to the floor for end-of-life care.

A midline sternotomy and laparotomy were made,and the aorta and inferior mesenteric vein were cannu-lated for the infusion of the cold preservation fluid. Thefirst 11 donors received the University of Wisconsin(UW) solution for cold preservation, whereas the histi-dine-trytophan-ketoglutarate (HTK) solution was usedfor the last 9 organs recovered. The crystalloid HTKsolution was felt to offer more rapid organ cooling, giventhe higher rate of the cold perfusate flow in comparisonwith the colloidal UW solution. Once the cold flush wasinitiated, the aorta was cross-clamped in the chest justabove the diaphragm, whereas the organs to be recov-ered were surface-cooled with ice. Upon the completionof the cold perfusate infusion, the organs were rapidlyremoved from the abdomen and stored in the cold pres-ervation fluid until transplantation.

RESULTS

From March 1997 to March 2006, there were 20 LTsperformed with DCD donors, of which 3 were performedin the pre-MELD era. The median follow-up time was7.5 months (range: 1-73). Table 1 shows the donor and

TABLE 1. Baseline Characteristics of 20 DCD-LT

Recipients

Recipient age (years) 53 � 8

Donor age (years) 35 � 11MELD score at LT 19 � 10WIT (minutes) 33 � 12CIT (hours) 8.7 � 2.7Peak ALT after LT (U/L) 1757 � 1477Peak AST after LT (U/L) 4020 � 3693Bilirubin at hospital discharge

(mg/dL) 3.2 � 5.4Alkaline phosphatase at hospital

discharge (U/L) 248 � 200Mortality 5 (25%)Retransplantation 4 (20%)Serious biliary complications 11 (55%)

Abbreviations: ALT, alanine aminotransferase; AST,aspartate aminotransferase; CIT, cold ischemia time;DCD-LT, donation after cardiac death livertransplantation; LT, liver transplantation; MELD, modelfor end-stage liver disease; WIT, warm ischemia time.

1646 MAHESHWARI ET AL.

LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases

recipient characteristics. The mean donor age was 35 �11 years, 6 (30%) were female donors, and the meantime spent in the intensive care unit before death was4.2 � 3.7 days. The cause of death was irreversiblebrain injury in all cases. Liver biopsies were not rou-tinely performed on all donor livers unless this wasclinically indicated, and only 4 donor livers (20%) un-derwent a biopsy. The mean WIT was 33 � 12 minutes,whereas the mean CIT was 8.7 � 2.7 hours.

The average recipient age was 53 � 8.7 years, and 18recipients (90%) were males. The mean MELD score attransplantation was 19.6 � 10. The etiology of end-stage liver disease in the recipients was chronic hepa-titis C (n � 11), alcoholic cirrhosis (n � 2), primarysclerosing cholangitis (n � 3), cryptogenic cirrhosis (n �2), autoimmune hepatitis (n � 1), or sickle cell hepa-topathy (n � 1). Five patients (25%) had concomitanthepatocellular cancer at the time of transplantation. Allpatients with hepatocellular carcinoma (HCC) fulfilledthe Milan criteria for LT. The peak alanine aminotrans-ferase (ALT) and aspartate aminotransferase (AST) lev-els after LT were 1757 � 1477 and 4020 � 3693 IU/L,respectively. The mean bilirubin and alkaline phospha-tase levels at the time of discharge from the hospitalwere 3.2 � 5.4 mg/dL and 248 � 200 U/L, respectively.

During follow-up, there were 5 deaths (25%) due tothe following causes: sepsis in the immediate postoper-ative period (n � 3), recurrent HCC 4 months after LT(n � 1), and cardiovascular complications 46 monthsafter LT (n � 1). In addition, 4 recipients (20%) requiredre-LT, the mortality in this subgroup being 25% (1/4).In the immediate postoperative period, 2 re-LTs wererequired because of hepatic artery thrombosis (HAT;n � 1) and primary graft nonfunction (n � 1), whereasan additional 2 re-LTs were performed within 6 monthsof the initial LT because of biliary complications (severeductopenia with anastomotic dehiscence and necrosisof the extrahepatic biliary tree requiring closure of thedistal bile duct and percutaneous drainage with asso-ciated biliary sepsis in 1 patient and severe biliary stric-turing with recurrent biliary sepsis in the other). Thetotal number of grafts lost in the entire cohort was 9/20(45%), and this included 2 deaths with functioning al-lografts free of biliary complications. Figure 2A,B showsthe Kaplan-Meier survival analysis of the graft and pa-tient survival rates at 72 months.

Biliary Complications

Sixteen patients (80%) underwent cholangiographicevaluation by endoscopic retrograde cholangiography(ERC; n � 10), percutaneous transhepatic cholangiog-raphy (PTC; n � 2), magnetic resonance cholangiogra-phy (MRC; n � 2), or a T-tube cholangiogram (n � 2). Of

Figure 1. (A) Cholangiogram showing hilar stricturing withcast formation within the bile ducts at the hilum. (B) Cholan-giogram showing a filling defect (cast) above the anastomosisand the cast that was removed from the duct. (C) Moderatelysevere anastomotic stricture and diffuse intrahepatic involve-ment mimicking sclerosing cholangitis.

BILIARY COMPLICATIONS FROM DONORS AFTER CARDIAC DEATH 1647


the 4 patients that did not undergo cholangiographicevaluation, 1 underwent re-LT at 4 weeks because ofHAT, 1 died on postoperative day 12 from sepsis, 1 waslost to follow-up 14 months after transplantation, and 1did not show clinical or biochemical evidence of biliarycomplications requiring cholangiography. Twelve pa-tients (12/16, or 75%) who underwent cholangiographydemonstrated abnormalities, which included stones,strictures, or bile leaks; of these, 11 patients had seri-ous complications, whereas 1 patient had a small soli-tary stone. Table 2 summarizes the nature of the biliarycomplications observed in this cohort. More than 1complication was noted in a majority of the patients.Table 3 describes the nature and extent of the biliarycomplications of each individual patient. The biliarycomplication rates were similar when UW and HTK so-lutions were used for cold preservation (55% versus45%, P � not significant).

Biliary Strictures

Ten patients (50%) developed biliary strictures. All pa-tients demonstrated intrahepatic strictures of various

severities, whereas 9 patients (45%) also had evidenceof stricturing in the left or right hepatic ducts, and 7(35%) showed evidence of hilar strictures and abnor-malities of the donor duct (strictures or duct irregular-ity) proximal to the anastomosis. One patient was notedto have mild stricturing of the recipient duct. Two pa-tients with severe intrahepatic strictures required re-transplantation within 6 months of the initial trans-plant because of the development of recurrent biliarysepsis in 1 patient and necrosis of the extrahepatic bileduct associated with severe ductopenia in the other.Neither patient was noted to have identifiable abnor-malities of the hepatic artery. Of the 6 patients (30%) inthis cohort that were noted to have hepatic artery ste-nosis (n � 4) or thrombosis (n � 2), 3 patients werenoted to have biliary strictures and 2 patients were not,whereas 1 was not evaluated before his re-LT due toHAT.

Bile Leaks

Two patients (2/20, or 10%) were noted to have largebile leaks and eventually required re-LT. One patienthad an early bile leak on day 2 after LT and requiredre-LT for primary graft nonfunction, whereas the otherhad an anastomotic leak 30 days after LT due to dehis-cence of anastomosis and necrosis of extrahepatic bileducts and underwent re-LT soon after. Both leaks weretreated with stenting by ERC, although the second pa-tient eventually required operative closure of the distalbile duct and percutaneous drainage while awaitingre-LT.

Filling Defects

Eight patients (40%) were noted to have filling defectson cholangiography. Although 1 patient had an isolated6-mm stone in the recipient duct, the other 7 had evi-dence of cast formation proximal to the anastomosiswith evidence of ductal strictures. Two patients werealso noted to have significant debris in the intrahepaticducts requiring long-term percutaneous drainage. Allfilling defects were successfully managed initially byendoscopy with biliary sphincterotomy and by en-doscopic extraction after the dilatation of biliarystrictures. One patient who underwent a hepaticojeju-nostomy at transplantation required multiple percuta-neous drainage procedures.

length of fu in months847260483624120

Cum

ulat

ive

Surv

ival

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

Graft Loss K-M analysis

length of fu in months847260483624120

Cum

ulat

ive

Surv

ival

1.0

0.8

0.6

0.4

0.2

0.0

Patient Survival K-M analysis

A

B

Figure 2. (A) Kaplan-Meier graft survival. (B) Kaplan-Meierpatient survival.

TABLE 2. Nature of the Biliary Complications

Major bile leak 2 (10%)

Anastomotic strictures 5 (25%)Hilar strictures 7 (35%)Intrahepatic strictures 10 (50%)Donor duct strictures 9 (45%)Stones 1 (5%)Casts 7 (35%)Debris 2 (10%)



TA

BL

E3

.B

ilia

ryC

om

pli

cati

on

sA

mo

ng

Ind

ivid

ual

Rec

ipie

nts

of

DC

D-L

T

Pati

ent

CIT

(hou

rs)

WIT

(min

ute

s)

Don

orD

uct

Rec

ipie

nt

Du

ct

An

ast

omot

icS

ite

Bilia

ryPro

cedu

res

Ou

tcom

ePro

xim

alto

An

ast

omos

isH

ilu

mR

igh

tD

uct

Lef

tD

uct

Intr

ah

epati

cB

ran

ches

Dis

talto

An

ast

omos

isF

illin

gD

efec

ts

115

—N

orm

al

Nor

mal

Nor

mal

Nor

mal

Diffu

sen

arr

owin

gN

orm

al

Non

eN

orm

al

PTC

�3

wit

hex

tern

al

dra

inage

Dea

thdu

eto

bilia

ryse

psi

s

28

21

Irre

gula

rity

wit

hca

stM

oder

ate

stri

ctu

reM

oder

ate

stri

ctu

reM

oder

ate

stri

ctu

rePoo

rly

visu

alize

dN

orm

al

Non

eS

ever

est

rict

ure

ER

CP

�6

wit

hsh

ort-

term

sten

tin

g

Alive

wit

hou

tst

ents

312

37

1.5

-cm

foca

ldilata

tion

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Non

eN

orm

al

MR

CP

Alive

,n

ost

ents

nec

essa

ry4

11

38

Diffu

seir

regu

lari

tyw

ith

cast

Sev

ere

stri

ctu

reS

ever

est

rict

ure

Sev

ere

stri

ctu

rew

ith

cast

Sev

ere

stri

ctu

reN

orm

al

Non

eM

oder

ate

stri

ctu

reE

RC

P�

11

wit

hst

ent,

PTC

�6

wit

hex

tern

al

dra

inage

.

Die

dat

46

mon

ths

from

card

iova

scu

lar

dis

ease

510

60

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Non

eN

orm

al

T-t

ube

chol

an

giog

ram

�2

Alive

wit

hou

tbilia

ryco

mplica

tion

s6

818

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Non

eN

orm

al

MR

CP

�3

Alive

wit

hou

tbilia

ryco

mplica

tion

s7

827

Not

evalu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dD

ied

onpos

toper

ati

veday

12

from

VR

Ese

psi

s8

12

29

Not

evalu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dA

live

thou

ghlo

stto

FU

at

14

mon

ths

99

22

Nor

mal

Nor

mal

Mod

erate

stri

ctu

rew

ith

deb

ris

Mod

erate

stri

ctu

rew

ith

deb

ris

Mod

erate

stri

ctu

rew

ith

deb

ris

Not

applica

ble

bec

au

seof

H-J

an

ast

omos

is

Not

applica

ble

bec

au

seof

H-J

an

ast

omos

is

PTC

�22

Alive

wit

hlo

ng-

term

exte

rnal

dra

inage

for

26

mon

ths

10

12

15

Diffu

seir

regu

lari

tyM

oder

ate

stri

ctu

reM

oder

ate

stri

ctu

reM

oder

ate

stri

ctu

reM

oder

ate

stri

ctu

reN

orm

al

Non

eN

orm

al

ER

CP

�3

wit

hdilata

tion

Alive

wit

hper

sist

ent

alk

alin

eph

osph

ata

seel

evati

on11

7.5

31

Nor

mal

Nor

mal

Mild

stri

ctu

reM

ild

stri

ctu

reM

oder

ate

stri

ctu

reN

orm

al

Non

eN

orm

al

ER

CP

�2

Die

dat

4m

onth

sbec

au

seof

met

ast

ati

cH

CC

12

658

Not

evalu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

one

Alive

wit

hou

tbilia

ryco

mplica

tion

s13

932

Sev

ere

stri

ctu

rean

dn

ecro

sis

Sev

ere

stri

ctu

reS

ever

est

rict

ure

wit

hdeb

ris

Sev

ere

stri

ctu

rew

ith

deb

ris

Sev

ere

stri

ctu

rew

ith

deb

ris

Nor

mal

Non

eLarg

ebile

leak,

2/2

deh

isce

nce

ER

C�

2w

ith

sten

tin

g,PTC

�3

wit

hex

tern

al

dra

inage

Ret

ran

spla

nt

at

3m

onth

sdu

eto

deh

isce

nce

ofan

ast

omos

isan

dn

ecro

sis

ofdon

ordu

cts

14

552

Irre

gula

rity

wit

hou

tst

rict

ure

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Non

eN

orm

al

T-t

ube

chol

an

giog

ram

No

sten

tsre

qu

ired



TA

BL

E3

.C

on

tin

ued

Pati

ent

CIT

(hou

rs)

WIT

(min

ute

s)

Don

orD

uct

Rec

ipie

nt

Du

ct

An

ast

omot

icS

ite

Bilia

ryPro

cedu

res

Ou

tcom

ePro

xim

alto

An

ast

omos

isH

ilu

mR

igh

tD

uct

Lef

tD

uct

Intr

ah

epati

cB

ran

ches

Dis

talto

An

ast

omos

isF

illin

gD

efec

ts

15

833

Nor

mal

Nor

mal

Nor

mal

Not

seen

Nor

mal

Nor

mal

Non

eLarg

ebile

leak

ER

Cw

ith

sten

tR

etra

nsp

lan

tat

pos

toper

ati

veday

2du

eto

PN

Fw

ith

subse

qu

ent

dea

thfr

omH

AT

an

dse

psi

s16

734

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Nor

mal

Sin

gle

6-m

mst

one

Nor

mal

ER

C�

2A

live

wit

hou

tbilia

ryco

mplica

tion

s17

833

Irre

gula

rity

wit

hca

stM

ild

stri

ctu

reM

ild

stri

ctu

rew

ith

cast

Mild

stri

ctu

rew

ith

cast

Mild

stri

ctu

res

Nor

mal

Non

eM

oder

ate

stri

ctu

reE

RC

�3

wit

hst

ents

Alive

3m

onth

saft

erLT

18

10-m

mst

rict

ure

wit

hla

rge

cast

Sev

ere

stri

ctu

reS

ever

est

rict

ure

Sev

ere

stri

ctu

reS

ever

est

rict

ure

Nor

mal

Non

eM

oder

ate

stri

ctu

reE

RC

�2

wit

hst

ents

Alive

at

3m

onth

saft

er LT

19

636

Not

evalu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dN

otev

alu

ate

dR

etra

nsp

lan

tat

4w

eeks

du

eto

HA

T20

525

Mild

stri

ctu

rew

ith

cast

Mild

stri

ctu

rew

ith

cast

Sev

ere

stri

ctu

rew

ith

cast

Sev

ere

stri

ctu

rew

ith

cast

Sev

ere

diffu

sest

rict

ure

sw

ith

du

ctal

dilata

tion

Mild

stri

ctu

reN

one

Mild

stri

ctu

reE

RC

�4

wit

hst

ent,

PTC

�10

wit

hex

tern

al

dra

inage

Ret

ran

spla

nt

at

5m

onth

sdu

eto

recu

rren

tbilia

ryse

psi

s

NO

TE

:Th

edon

ors

ofpati

ents

1-1

3h

ad

the

UW

solu

tion

for

cold

pre

serv

ati

on,

an

dth

edon

ors

ofpati

ents

14-2

0h

ad

the

HTK

solu

tion

.A

bbre

viat

ion

s:C

IT,

cold

isch

emia

tim

e;D

CD

-LT,

don

ati

onaft

erca

rdia

cdea

thlive

rtr

an

spla

nta

tion

;E

RC

,en

dos

copic

retr

ogra

de

chol

an

giog

raph

y;E

RC

P,

endos

copic

retr

ogra

de

chol

an

giop

an

crea

togr

aph

y;H

AT,h

epati

cart

ery

thro

mbos

is;H

CC

,hep

ato

cellu

lar

carc

inom

a;H

TK

,his

tidin

e-tr

ytop

han

-ket

oglu

tara

te;L

T,l

iver

tran

spla

nta

tion

;M

RC

P,

magn

etic

reso

nan

cech

olan

giop

an

crea

togr

aph

y;PTC

,per

cuta

neo

us

tran

shep

ati

cch

olan

giog

raph

y;U

W,

Un

iver

sity

ofW

isco

nsi

n;

VR

E,

van

com

ycin

-res

ista

nt

En

tero

cocc

us;

WIT

,w

arm

isch

emia

tim

e;H

-J,

hep

ati

co-j

eju

nos

tom

y;FU

,fo

llow

up;

PN

F,

pri

mary

non

-fu

nct

ion

.



Procedures

Sixteen patients in this cohort underwent multiple di-agnostic and therapeutic biliary procedures, includingMRC (4), ERC (36), and PTC (47). The mean number ofERCs required for the group was 3.6 � 2.5 (range:1-10), whereas the mean number of PTCs required inthis cohort was 6.7 � 3 (range: 1-21).

DISCUSSION

We previously reported a 15.5% incidence of biliarycomplications in our center after conventional LT(DBD).12 In contrast, the incidence of all biliary abnor-malities in this cohort of recipients who received DCDlivers was 60% (12/20), and of these, 55% (11/50)showed evidence of serious complications. In those whounderwent a cholangiographic evaluation, 75% (16/20)had biliary abnormalities. Of those who did not undergocholangiography, 50% (2/4) experienced graft losswithin 30 days of transplantation, whereas the entirecohort reported a 40% graft loss within a year after LT.Our findings suggest that a majority of patients whoreceive DCD-LT may develop 1 or more biliary compli-cations.

Biliary strictures proximal to the anastomosis werethe most common complication observed and werenoted in 50% of the patients in this series. Nonanasto-motic proximal duct strictures are traditionally attrib-uted to ischemic preservation injury or vascular com-promise secondary to HAT or stenosis. However, inDCD-LT, the ischemic injury may occur before organretrieval in the absence of hepatic artery stenosis or

thrombosis. In fact, a significant proportion of patientswith biliary strictures after LT do not have demonstra-ble evidence of hepatic artery narrowing or thrombosis.In our study, we identified hepatic artery abnormalitiesin 6 patients, but only 3 were noted to have biliarystrictures. Additionally, 2 patients that required re-LTfor biliary complications did not have evidence of com-promised hepatic artery circulation.

Many single-center studies had previously reported ahigher rate of biliary complications among recipients ofDCD livers, and our study corroborates these observa-tions (Table 4). Abt et al.9 reported a significantly higherincidence of biliary complications at their center amongthe recipients of DCD-LT (33% versus 9.5%) in compar-ison with a cohort of 221 recipients of DBD-LT duringthe same period. Foley et al.8 also reported a signifi-cantly higher rate of biliary strictures in the DCD groupat 1 year (33% versus 10%) and at 3 years (37% versus12%) in their cohort of 36 recipients of DCD-LT in com-parison with their center’s experience of 553 DBD-LTs.This study also reported a significantly higher inci-dence of hepatic artery stenosis in their DCD organrecipients (16% versus 5%). WIT in our series washigher than that in other reports.8,9,14,15 Although onecould argue that the higher incidence of biliary compli-cations and morbidity in our series was perhaps due tohigher WIT, 3 patients (Table 3) with WIT values of 52,60, and 58 minutes showed minimal or no biliary com-plications. This may suggest that many other factorsare involved in the development of biliary complica-tions. However, on the basis of other reports, our centerhas decided to use organs only if WIT is less than 30

TABLE 4. Single-Center Experience with DCD-LT

University of

Wisconsin,

Wisconsin

(n � 36)*

Albert Einstein

Medical Center,

Pennsylvania

(n � 19)†

University of

Pennsylvania,

Pennsylvania

(n � 30)‡

Kings College,

London,

United Kingdom

(n � 32)§

Johns Hopkins

University,

Maryland

(n � 20)

DemographicsDonor age (years) 34 30 36 35Recipient age (years) 38 53WIT (minutes) 17 20 20 14 33CIT (hours) 8.2 9.5 6.1 8.6 8.7MELD score at orthotopic LT 19.6Peak ALT (IU/mL) 141 1757

ResultsFollow-up time (months)� 36 (1–80) 16 (1.5–37) 27 (1–46) 15 (1–40) 14 (1–73)Patient survival 68% at 3 years 74% 79% at 3 years 89% at 1 year 78% at 1 yearGraft survival 50% 72% at 3 years 86% at 1 year 62% at 1 yearRetransplantation 19% 11% 6% 3% 20%Biliary complications 37% 11% 33% 9% 55%Hepatic artery complications 22% 16% 0 6% 30%Primary nonfunction 5% 5% 6% 3% 5%

NOTE: The data are shown as means. Unless it is stated otherwise, the patient survived at the end of the follow-up period. Themissing data were not reported in the publications. Abbreviations: ALT, alanine aminotransferase; CIT, cold ischemia time;DCD-LT, donation after cardiac death liver transplantation; LT, liver transplantation; MELD, model for end-stage liver disease;WIT, warm ischemia time. *Data from Foley et al.8 †Data from Yuan et al.14 ‡Data from Abt et al.9 §Data from Muiesan et al.15

�The ranges are shown in parentheses.



minutes and CIT is less than 8 hours.6,7 We believe thatthese arbitrary optimal WIT and CIT values need to bevalidated in prospective multicenter studies.

Our patients required multiple interventions for theirbiliary complications. Unlike extrahepatic strictures,intrahepatic strictures are less amenable to endoscopicmanagement and are associated with an increased in-cidence of biliary sludge or cast formation, recurrentcholangitis, and biliary fibrosis. Biliary complicationsafter DCD-LT are associated with significant morbidityand graft loss. In this study, we did not perform a costanalysis of biliary complications, but a study is inprogress at our institution to assess the relative cost ofDCD-LT versus that of matched DBD-LT.

The pathophysiology of biliary stricturing after LT ispoorly understood. Although it is widely accepted thatthe initial injury to the bile duct is most likely ischemicin nature, the mechanisms leading to stricture forma-tion are poorly understood. Injury to the biliary epithe-lial cells seems to be central to the eventual formation ofstrictures,16 and this injury may occur before or afterorgan retrieval because of the depletion of energystores.17 Hydrophobic bile salts may cause furtherdamage to the biliary epithelium,18 and flushing thebiliary tree free of bile during organ harvesting couldminimize this injury. In addition, the infusion of a coldperfusate into the biliary tree through a nasobiliarytube during the organ retrieval may further reduceischemic injury by reducing the metabolic require-ments of biliary epithelia. Ischemic injury to the micro-vascular endothelium in the preservation phase alsoleads to cell disruption and may contribute to micro-vascular thrombosis, and this may prevent effectiverevascularization and worsen ischemic injury to thebiliary epithelium.18,19 Furthermore, the activation ofneutrophils that promote the release of free radicalsalso causes oxidative stress, further damaging suscep-tible epithelial cells. The end result is the loss of epithe-lial lining in the biliary tree leading to ulcer formation,which may serve as a nidus for biliary sludge forma-tion.20 The ulcerations in the epithelial lining are asso-ciated with granulation tissue on the underlyingstroma, and the progressive fibrosis associated withwound healing promotes the formation of biliary stric-tures.

Specific strategies need to be developed to decreasethe significantly higher rates of biliary complicationsnoted among the recipients of DCD-LT. Although somestrategies such as limiting CIT and WIT are universallyaccepted, others such as flushing the biliary tree free ofbile and alternate aortic flushes with low-viscosity pres-ervation solutions such as HTK21 or Marshall’s solu-tion22 deserve further consideration. In our smallstudy, biliary complications were similar with UW andHTK solutions (55% versus 45%, P � not significant),but our study did not have adequate power to detect adifference. Laboratory studies have suggested that thepretreatment of donor organs with interleukin-6 in theaortic flush significantly reduces the ischemic preser-vation injury to the liver and may help restore the in-tegrity of the biliary tree after transplantation. The post-

transplant use of agents such as tranilast23 andcaptopril24 may exert an antifibrotic response that mayhelp reduce the incidence of biliary strictures. Simi-larly, antioxidants such as vitamin E and peroxisomeproliferator-activated receptor � ligands25 such as thethiazolidinedione drugs14 may prevent myofibroblastactivation and the development of fibrosis. Only care-fully performed, randomized controlled studies couldprove the efficacy of these strategies, and the livertransplant community should focus on this area toimprove graft survival and reduce biliary complicationsafter DCD-LT.

An important question that needs to be addressed bythe transplant community is whether DCD-LT shouldbe routinely used or selectively used in a certain recip-ient population. It is unlikely that we will have a con-sensus, but this issue needs to be debated. On thebasis of the current data, it is premature to abandonDCD-LT. However, DCD-LT is associated with signifi-cant morbidity, cost, and graft loss. We believe thatDCD-LT should be used in patients who would notreceive DBD-LT in an expedited manner (e.g., the dis-ease severity is not reflected by the MELD score) or inpatients with cancers who do not qualify for additionalMELD points (HCC outside the Milan criteria but fallingwithin the San Francisco criteria or small cholangiocar-cinoma). We understand that these recommendationswill be controversial, yet we believe that our findingsmerit a serious discussion of this topic.

REFERENCES

1. Doshi MD, Hunsicker LG. Short- and long-term outcomeswith the use of kidneys and livers donated after cardiacdeath. Am J Transplant 2007;7:122-129.

2. Tector AJ, Mangus RS, Chestovich P, Vianna R, Fridell JA,Milgrom ML, et al. Use of extended criteria livers decreaseswait time for liver transplantation without adversely im-pacting post transplant survival. Ann Surg 2006;244:439-450.

3. Abt PL, Desai NM, Crawford MD, Forman LM, MarkmannJW, Olthoff KM, et al. Survival following liver transplanta-tion from non-heart-beating donors. Ann Surg 2004;239:87-92.

4. Merion RM, Pelletier SJ, Goodrich N, Englesbe MJ, Del-monico FL. Donation after cardiac death as a strategy toincrease deceased donor liver availability. Ann Surg 2006;244:555-562.

5. Barshes NR, Shah BB, Sondhi M, Freeman RB, Wong JB,Vierling JM, et al. Is accepting a liver graft from a donorafter cardiac death (dcd) better than remaining on the waitlist and preferring a liver graft from a donor after braindeath (dbd)? Results of a decision analysis. Transplanta-tion 2006;82:272.

6. Mateo R, Cho Y, Singh G, Stapfer M, Donovan J, Kahn J,et al. Risk factors for graft survival after liver transplanta-tion from donation after cardiac death donors: an analysisof OPTN/UNOS data. Am J Transplant 2006;6:791-796.

7. Lee KW, Simpkins CE, Montgomery RA, Locke JE, SegevDL, Maley WR. Factors affecting graft survival after livertransplantation from donation after cardiac death donors.Transplantation 2006;82:1683-1688.

8. Foley DP, Fernandez LA, Leverson G, Chin LT, Krieger N,Cooper JT, et al. Donation after cardiac death: the Univer-sity of Wisconsin experience with liver transplantation.Ann Surg 2005;242:724-731.



9. Abt P, Crawford M, Desai N, Markmann J, Olthoff K,Shaked A. Liver transplantation from controlled non-heart-beating donors: an increased incidence of biliarycomplications. Transplantation 2003;75:1659-1663.

10. Greif F, Bronsther OL, Van Thiel DH, Casavilla A, IwatsukiS, Tzakis A, et al. The incidence, timing, and managementof biliary tract complications after orthotopic liver trans-plantation. Ann Surg 1994;219:40-45.

11. Pfau PR, Kochman ML, Lewis JD, Long WB, Lucey MR,Olthoff K, et al. Endoscopic management of postoperativebiliary complications in orthotopic liver transplantation.Gastrointest Endosc 2000;52:55-63.

12. Thuluvath PJ, Atassi T, Lee J. An endoscopic approach tobiliary complications following orthotopic liver transplan-tation. Liver Int 2003;23:156-162.

13. Thuluvath PJ, Pfau PR, Kimmey MB, Ginsberg GG. Biliarycomplications after liver transplantation: the role of en-doscopy. Endoscopy 2005;37:857-863.

14. Yuan GJ, Zhang ML, Gong ZJ. Effects of PPARg agonistpioglitazone on rat hepatic fibrosis. World J Gastroenterol2004;10:1047-1051.

15. Muiesan P, Girlanda R, Jassem W, Melendez HV, O’GradyJ, Bowles M, et al. Single-center experience with livertransplantation from controlled non-heartbeating donors:a viable source of grafts. Ann Surg 2005;242:732-738.

16. Alpini G, McGill JM, Larusso NF. The pathobiology ofbiliary epithelia. Hepatology 2002;35:1256-1268.

17. Noack K, Bronk SF, Kato A, Gores GJ. The greater vulner-ability of bile duct cells to reoxygenation injury than toanoxia. Implications for the pathogenesis of biliary stric-tures after liver transplantation. Transplantation 1993;56:495-500.

18. Kukan M, Haddad PS. Role of hepatocytes and bile ductcells in preservation-reperfusion injury of liver grafts.Liver Transpl 2001;7:381-400.

19. Teoh NC, Farrell GC. Hepatic ischemia reperfusion injury:pathogenic mechanisms and basis for hepatoprotection. JGastroenterol Hepatol 2003;18:891-902.

20. Demetris AJ, Fontes P, Lunz JG III, Specht S, Murase N,Marcos A. Wound healing in the biliary tree of liver allo-grafts. Cell Transplant 2006;15:S57-S65.

21. Minor T, Hachenberg A, Tolba R, Pauleit D, Akbar S. Fi-brinolytic preflush upon liver retrieval from non-heartbeating donors to enhance postpreservation viability andenergetic recovery upon reperfusion. Transplantation2001;71:1792-1796.

22. Pirenne J, Van Gelder F, Coosemans W, Aerts R, GunsonB, Koshiba T, et al. Type of donor aortic preservationsolution and not cold ischemia time is a major determi-nant of biliary strictures after liver transplantation. LiverTranspl 2001;7:540-545.

23. Ikeda H, Inao M, Fujiwara K. Inhibitory effect of tranilaston activation and transforming growth factor beta 1 ex-pression in cultured rat stellate cells. Biochem BiophysRes Commun 1996;227:322-327.

24. Wengrower D, Zannineli G, Pappo O, Latella G, Sestieri M,Villanova A, et al. Prevention of fibrosis in experimentalcolitis by captopril: the role of tgf-beta1. Inflamm BowelDis 2004;10:536-545.

25. Miyahara T, Schrum L, Rippe R, Xiong S, Yee HF Jr,Motomura K, et al. Peroxisome proliferator-activated re-ceptors and hepatic stellate cell activation. J Biol Chem2000;275:35715-35722.



biliary complications and outcomes of liver transplantation from donors after cardiac death

Documents