Book on hepatitis from page 386 to 399

Book on hepatitis from page 386 to 399

386  Hepatology 2012
23. End-stage Liver Disease, HIV Infection
and Liver Transplantation
José M. Miró, Fernando Agüero, Montserrat Laguno, Christian Manzardo,
Montserrat Tuset, Carlos Cervera, Neus Freixa, Asuncion Moreno, Juan-Carlos García-Valdecasas, Antonio Rimola, and the Hospital Clinic OLT in
HIV Working Group
Introduction
Liver disease due to chronic hepatitis B and C is currently one of the leading causes
of morbidity and mortality among HIV-positive patients in the developed world and
its burden is increasing (Joshi 2011). Non-coinfected patients with chronic hepatitis
C tend to progress to end-stage liver disease (ESLD) in 20-30 years, whereas
coinfected patients have higher rates of progression (Mohsen 2003; Poynard 2003).
Furthemore, a higher overall adjusted relative risk (RR) of histological cirrhosis or
decompensated liver disease has been observed in patients coinfected with HIV and
the hepatitis C virus (HCV) when compared to HCV-monoinfected patients
(Graham 2001). End-stage liver disease (ESLD) in coinfected patients has become a
leading clinical state in many of these patients. Medical management is essential in
this scenario and should be considered as a bridge to organ liver transplantation
(OLT), which can play an important role as a therapeutic tool in this subset of
patients. We summarize recent developments in the management of ESLD.
Nevertheless this is an evolving field and many issues remain unclear (Miró 2011).
Epidemiology
Of the approximately 40 million persons infected with HIV globally, 2 to 4 million
are chronically infected with the hepatitis B virus (HBV) and 4 to 5 million have
chronic HCV (Alter 2006).
The prevalence of HCV and/or HBV coinfection is high in developed countries.
Studies performed in European HIV-positive patients showed rates of 33% and 9%,
respectively (Rockstroh 2005, Konopnicki 2005), while in the US figures are very
similar, 28% and 9% (Fung 2004). Other authors have addressed the significance of
HCV as a cause of non–AIDS-related death (Palella 2006, Crum 2006, Lewden
2005). One single-centre study (Martínez 2007) in Spain analysed the cause of 235
End-stage Liver Disease, HIV Infection and Liver Transplantation  387
deaths in 4471 patients (5%) on combination antiretroviral therapy (cART) from
1997 until 2004. The number of patients who died from ESLD increased from 8% in
1997 to 41% in 2004, and in recent years this condition has become the leading
cause of death in HIV-positive patients. In comparison with the general population
of a similar age, deaths due to liver disease were 11 times more frequent in HIV-positive patients. Another prospective multicenter study (Rosenthal 2007) in France
determined mortality due to ESLD in a nationwide population of HIV-positive
patients. The authors followed a total of 21,000 HIV-positive patients, 4000 (19.9%)
of whom were coinfected, and showed that, in 2003,  mortality due to ESLD
represented 23.7% of non–AIDS-related deaths. In this population, ESLD was fatal
in 1.5% of patients in 1995, 6.6% in 1997, 14.3% in 2001, and 12.6% in 2003, and
92.6% of patients who died from ESLD had chronic HCV infection. Another
prospective study of 11 cohorts carried out in Europe, the United States and
Australia (D:A:D Study 2006) included 23,500 HIV-1–infected patients (22.5%
were HCV-positive) and followed them from December 1999 until February 2004.
This study showed that, of the 1250 deaths recorded, those related to AIDS were the
most frequent (31.1%), while liver disease was the most frequent non–AIDS-related
cause (14.5%). HCV infection was shown to be an independent predictor of liver-related death (D:A:D Study 2006). As for hepatocellular carcinoma (HCC), one
study comparing liver-related deaths in HIV-positive patients (Salmon-Ceron 2009)
described an increase in mortality due to HCC from 15% in 2000 to 25% in 2005
(p=0.04).
Clinical features of coinfected patients with ESLD
154 patients with a new diagnosis of Child-Turcotte-Pugh class A compensated
cirrhosis were followed and 36 of them (23.4%) developed a first hepatic
decompensation during follow-up (mean 36 months) (Pineda 2009). The probability
of developing decompensated cirrhosis at 3 and 5 years was 26% and 33%,
respectively. Factors predicting the emergence of an episode of hepatic
decompensation at 5 years were Child-Turcotte-Pugh stage (HR, 3.33 [95% CI,
1.39-7.69]), lack of anti-HCV therapy (HR, 3.38 [95% CI, 1.14-5.04]), and baseline
CD4 cell count below 300 cells/mm3 (HR, 2.40 [95% CI, 1.09-10.53]).
In a retrospective study, the same authors (Pineda 2005) described the frequency
of specific events, such as first decompensation and cause of death in HIV-negative
and HIV/HCV-coinfected subjects. Ascites and jaundice were more frequent among
HIV-positive patients, while upper gastrointestinal bleeding and HCC were more
frequent in monoinfected patients. Hepatic encephalopathy (HE) as both first
decompensation and cause of death was higher in coinfected patients.
The clinical characteristics and outcome of spontaneous bacterial peritonitis
(SBP) were evaluated in an HIV-positive population with cirrhosis (Shaw 2006).
Thirty-five HIV-positive patients with cirrhosis were compared with 70 HIV-negative patients with cirrhosis. An aetiologic diagnosis was made in almost 80% of
the HIV-positive cases and bacteraemia was present in more than 50%, with both
rates being higher than those observed in HIV-negative patients. An important
bacteriological finding in this study was the high incidence of Streptococcus
pneumoniae as the etiologic agent of SBP among HIV-positive patients, second
only to Escherichia coli.
388  Hepatology 2012
HCC has a faster and worse outcome in HIV/HCV-coinfected patients than in
HCV-monoinfected patients (Puoti 2004, Bruno 2006, Berretta 2011). Of note, this
disorder is expected to rise significantly in patients with HIV infection and chronic
VHC/VHB hepatitis.
The findings of HCC in 41 HIV-positive and 2384 HIV-negative patients were
compared in an Italian study (Puoti 2004). The authors found a more aggressive
course of HCC in HIV-positive patients, with an independent association between
HIV infection and a more advanced stage of HCC at clinical presentation, in
addition to a higher rate of infiltrating neoplasm and extrahepatic-extranodal
metastasis. Furthermore, portal vein thrombosis was more frequent among HIV-positive patients with HCC.
A retrospective study from Canada and the US compared 63 HIV-positive
patients with HCC and 226 HIV-negative patients with HCC (Bräu 2007) and
revealed that HIV+ patients were younger and more frequently symptomatic. In this
cohort, median survival was similar between HIV-positive (7 months) and HIV-negative patients (7.5 months, p=0.44, log-rank), as was tumour stage.
Recently, these findings were confirmed grouping a comparison of 104 HIV-infected patients who had HCC with 484 HIV-uninfected patients with HCC
(Berretta 2011). The group of patients with HIV infection was younger, had better
BCLC stage at diagnosis and the median survival time was significantly shorter.
One of the factors independently associated with survival was the HCC diagnosis
via a screening program. However, there are no data available on the cost-effectiveness of screening for HCC in cirrhotic patients with HIV infection (Joshi
2011).
Prognosis after decompensation
The survival rate of HIV-positive patients with decompensated cirrhosis is much
lower than that of HIV-negative patients – approximately 50% at 1 year (Pineda
2005, Merchante 2006, Murillas 2009). In a multicentre case-control study (Pineda
2005), the outcome of cirrhosis after the first decompensation in coinfected patients
was much worse than in the monoinfected population. Survival at 1, 2, and 5 years
for the coinfected/monoinfected population was 54%/74%, 40%/61%, and
25%/44%, respectively. In another study (Merchante 2006) severity of liver disease
(Child-Turcotte-Pugh score or HE as the first hepatic decompensation) and the level
of cellular immunosuppression (<100 CD4 cells) were identified as independent
predictors of poor outcome in coinfected patients. On the other hand, HAART was
associated with a reduced mortality rate.
104 HIV-positive patients with HCC or cirrhosis after their first hepatic
decompensation were analysed (Murillas 2009). The median survival time of this
cohort was 14 months, similar to that observed by Merchante (13 months). This
study included HCV and non–HCV-infected patients, and it did not find significant
differences in survival based on the aetiology of cirrhosis, suggesting that HIV-positive patients have an overall poor outcome regardless of the nature of their liver
disease. Furthermore, the MELD score and the inability to reach an undetectable
plasma HIV-1 viral load at any time during follow-up were the only variables
independently associated with the risk of death (p<0.001). This is particularly
relevant because the MELD score has been increasingly used to establish the
End-stage Liver Disease, HIV Infection and Liver Transplantation  389
prognosis of patients with cirrhosis and, consequently, to indicate liver
transplantation.
Recently, a Spanish study (Lopez-Diéguez 2011) showed that the mortality rate
for patients with decompensated cirrhosis was 27.1 deaths/100 person years and 4.0
deaths/100 person years for patients with compensated cirrhosis. The risk of first
hepatic decompensation was relatively low but the time from the first liver
decompensation to the next decompensation was critically reduced. HIV-positive
patients with cirrhosis have a poor prognosis after the development of SBP (Shaw
2006). HIV infection was associated with a more than 6-fold increase in the
probability of dying within a month of the first episode of SBP. Impaired renal
function at diagnosis and severity of liver disease were identified as predictors of
death. HIV-positive patients also had a dramatically shorter survival time than HIV-negative patients: only 50% of patients were still alive 3 months after the first
episode of SBP and only 23% were alive after 1 year. Death was mostly related to
complications of advanced liver disease rather than to AIDS-related conditions.
High mortality rates among coinfected patients with ESLD waiting for liver
transplantation have also been reported in observational studies (Maida 2005, Prieto
2008, Murillas 2009). One study (Prieto 2008) analysed 18 patients who were on the
waiting list for OLT. Eight (44%) received a transplant, 8 (44%) died while on the
waiting list, and 2 (12%) were still on the waiting list at the end of the study (Prieto
2008). 10 (67%) out of 15 patients on the transplant waiting list died after a median
follow-up of 5 months, and 5 (33%) underwent liver transplantation (Murillas
2009).
Two case-control studies have analysed mortality rates among coinfected patients
with ESLD waiting for liver transplantation. In the first (Ragni 2005), mortality
rates during the pre-transplant evaluation in HIV-positive (N=58) and HIV-negative
(N=1359) patients were 36% and 15%, respectively (p<0.001), although these data
were not confirmed by the second (Subramanian 2009). Waiting list mortality was
14.4% in patients with HIV infection (N=167) and 11.1% in the control group
(N=792) (p=0.30). In a multivariate analysis, a MELD score higher than 25 was the
only variable related to death on the waiting list (Subramanian 2009). Recently, it
was observed that the survival rate at 1 and 3 years after listing were 81% and 55%
in HIV infected vs. 91% and 82% in patients without HIV infection, respectively
(p=0.005) (Vibert 2011).
For these reasons, physicians attending  HIV-positive patients with cirrhosis
should follow patients prospectively and evaluate them early for OLT after the first
clinical decompensation of liver disease. Similarly, patients whose cirrhosis is
associated with HCC should also be evaluated (Llovet 2004). Both prevention and
effective treatment of these complications may improve the likelihood of survival
until OLT, and this should be performed also with the HIV-negative patients
(Agüero 2007, Spengler 2011).
Management of cirrhosis complications
Management of the complications of cirrhosis (portal hypertension, ascites,
gastrointestinal bleeding, encephalopathy, SBP, HCC, and hepatorenal syndrome)
must be planned, just as in the HIV-negative population (Arroyo 2008, EASL 2011,
Spengler 2011,  Bruix 2011). Medical management also includes prevention of
390  Hepatology 2012
infection. In view of the short survival associated with the development of SBP,
primary antibiotic prophylaxis with quinolones or trimethoprim-sulfamethoxazole
should be considered (Fernández 2007).
It has been observed that transient elastometry could be used to select HIV/HCV-coinfected patients undergoing screening with upper gastrointestinal endoscopy for
oesophageal varices (Pineda 2009). This study found that HIV/HCV-coinfected
patients with cirrhosis who harbour oesophageal varices requiring preventive
therapy for bleeding had liver stiffness values higher than those who did not require
treatment. Liver stiffness values lower than 21 kPa were highly predictive of varices
not at risk for bleeding.
As far as HCC is concerned, patients may benefit from more frequent imaging,
i.e., every 3 months (Bräu 2007). Treatment of HCC may not be successful,
depending on the stage. Other issues that may delay the progression of liver disease,
such as avoidance of hepatotoxic drugs (e.g., didanosine) and vaccination for
hepatitis A and B, should be kept in mind.
Substance abuse
Smoking has been linked to HCC (Kuper 2000) and increased hepatic fibrosis
(Pessione 2001). It may also increase histological activity in chronic HCV patients
irrespective of alcohol consumption (Hezode 2003).
According to one study (Rosenthal 2007), alcohol consumption was more
frequent among coinfected patients who died from ESLD (92%), and another study
suggested that excess alcohol consumption increases HCV RNA levels (Cooper
2005).
In addition, daily cannabis smoking was significantly associated with the presence
of moderate to severe fibrosis in patients with chronic HCV infection and those with
hepatitis C cirrhosis should abstain from or reduce cannabis use (Ishida 2008).
HCV/HBV management
Specific treatment for infection with HBV or HCV is possible, although more
difficult, in patients with advanced cirrhosis, especially for HCV infection
(Rockstroh 2008, GESIDA 2010).
One of the objectives when treating HCV-monoinfected patients with advanced
liver cirrhosis using pegylated-interferon plus ribavirin is to obtain undetectable
plasma HCV RNA levels at the time of OLT in order to reduce the risk of HCV
recurrence posttransplant. One study (Everson 2005) using a low accelerating
dosage regimen (LADR) of anti-HCV therapy in monoinfected patients on the OLT
waiting list showed that 30 (24%) of 124 patients achieved a sustained virologic
response (SVR) and 12 (80%) of 15 patients who were HCV RNA-negative before
OLT remained HCV RNA-negative 6 months or more after transplantation. This
approach has not yet been addressed in the HIV infection setting. Safety data is also
available from the APRICOT substudy (Mauss 2004). Hepatic decompensation was
observed only in HIV/HCV-coinfected patients with markers of advanced cirrhosis,
and its incidence was 10.4% (14/134). However, 6 (43%) of the 14 patients died as
a result of hepatic decompensation. Antiretroviral treatment with didanosine was an
associated risk factor. In contrast, no hepatic decompensation was noted in
HIV/HCV-coinfected patients without cirrhosis. Therefore, anti-HCV treatment
End-stage Liver Disease, HIV Infection and Liver Transplantation  391
during the pretransplant evaluation or while patients are on the waiting list should
be individualized (e.g., patients with Child-Turcotte-Pugh class A and HCC or
genotypes 2/3) and patients must be monitored closely because of their high risk of
hepatic decompensation and death. In this sense, the AADSL Practice Guidelines
for HCV infection (Ghany 2009), state that HIV-infected patients with
decompensated liver disease (CTP class B or C) should not be treated with peg-interferon + ribavirin, and should be considered candidates for liver transplantation
(Grading IIa, C).
Promising results are available related to the new HCV protease inhibitors
(telaprevir and boceprevir) in HCV-monoinfected patients. However, data from
larger trials regarding the HIV/HCV-coinfected patient is lacking. Their role in this
subset of patients is unknown.
Since HBV replication is a contraindication for OLT and only patients without
HBV viraemia are accepted for OLT, treatment of this infection should be a priority.
HIV-positive patients who require antiretroviral therapy and have chronic HBV
infection can be treated with lamivudine (or emtricitabine) and tenofovir as part of
their triple antiretroviral therapy (Rockstroh 2008, Soriano 2008, GESIDA 2010).
Adefovir and tenofovir have proven useful against HBV and could be used in cases
of resistance to lamivudine (Rockstroh 2008,  Soriano 2008,  GESIDA 2010).
Recently, one study (Heathcote 2011) showed that tenofovir disoproxil fumarate
(TDF) is safe and effective in the long-term management of patients with chronic
hepatitis B.
Combination antiretroviral therapy (HAART)
The role of HAART in the progression of liver disease and in overall mortality in
HCV/HIV-coinfected patients remains controversial (Tedaldi 2003, Qurishi 2003).
Nevertheless, permanent discontinuation of HAART was independently associated
with risk of first hepatic decompensation and a poorer survival rate (Lopez-Diéguez
2011) and to increased risk of fibrosis progression (Thorpe 2011). Antiretroviral
drug regimens should be carefully planned in persons with HIV and ESLD. These
patients should follow general recommendations (GESIDA 2011, DHHS 2011) and
their liver function must be closely monitored for signs of hepatotoxicity. Careful
consideration of drug prescriptions and possible interactions is essential.
Furthermore, some antiretroviral drugs may be contraindicated in cirrhotic patients
(e.g., didanosine, nevirapine, full-dose ritonavir) and their dosing should be adjusted
according to the degree of hepatic impairment (Wyles 2005, Back 2011, Tuset
2011).
Therapeutic drug monitoring (TDM) may be useful for efavirenz and protease
inhibitors. Indinavir and atazanavir can increase unconjugated bilirubin levels by
inhibiting UDP-glucuronosyltransferase. As total bilirubin is a component of both
the Child-Turcotte-Pugh and MELD scores, results in patients taking these drugs
should be interpreted with caution.
It is noteworthy that the newer antiretroviral drug raltegravir, which is not a
substrate of CYP450, can be used in HIV-1 OLT recipients. One study (Tricot
2009) enrolled 13 patients with HIV-1 infection who underwent solid organ
transplantation (8 liver and 5 kidney) and received raltegravir. The authors found a
lack of significant interaction between raltegravir and calcineurin inhibitors that
392  Hepatology 2012
allowed simplified management of immunosuppressive treatment, excellent
tolerability, and no events related to outcome (acute rejection) or HIV infection.
Therefore, the combination of two nucleos(t)ide reverse transcriptase inhibitors
(tenofovir + emtricitabine or abacavir + lamivudine) + raltegravir is the
antiretroviral regimen of choice in HIV-infected liver transplant recipients.
Finally, given the speed with which new antiretrovirals appear and thus new
interactions, physicians should consult updated databases on drug interactions (Back
2011, Tuset 2011).
Orthotopic liver transplant (OLT)
OLT is the only therapeutic option for patients with ESLD. HIV infection is not a
contraindication for liver transplantation (Miró 2007, Stock 2007, Samuel 2008).
There are 3 different classes of criteria for including HIV-positive patients on the
liver transplant waiting list: liver disease, HIV infection, and other criteria.
Liver disease criteria
These are the same as for the non–HIV-infected population; the main indication for
OLT in HIV-positive patients is ESLD caused by HCV coinfection. Less frequent
indications are HBV coinfection (either acute or ESLD) and liver cancer.
In the UK guidelines (O’Grady 2005), indications for liver transplantation include
acute liver failure, decompensated liver disease - with ascites, encephalopathy (it is
important to exclude HIV-related dementia), variceal bleeding that is difficult to
manage with standard therapy, and poor liver function (albumin <30 g/l, INR >1.5,
and elevated serum bilirubin >450 mmol/l) - and HCC detected during regular
tumour surveillance. In the Eurotransplant region these criteria have been replaced
by the MELD score. The criteria for liver transplantation in patients with HCC are
as follows: no more than 3 tumour nodules, no nodule greater than 5 cm in diameter,
absence of macroscopic portal vein invasion, and absence of recognizable
extrahepatic disease.
A new indication for liver transplant in HIV+ patients has been described in a
recent study (Tateo 2009) in which 3 patients underwent liver transplantation and
the cause of ESLD was nodular regenerative hyperplasia (NRH). OLT is the only
therapeutic option in cases of severe portal hypertension such as that observed in
these patients.
HIV infection criteria
Most liver transplant groups from Europe and North America use similar HIV
criteria. These are summarized in Table 1 (O’Grady 2005, Grossi 2005, Miró 2007,
Anonymous 2004).
Clinical criteria
Some authors are in favour of withdrawing exclusion criteria for some opportunistic
infections that can be effectively treated and prevented, such as tuberculosis,
candidiasis, and Pneumocystis jiroveci pneumonia  (Roland 2003,  Neff 2004,
Radecke 2005). In fact, the NIH-sponsored study has updated the inclusion criteria
for opportunistic complications and only untreatable diseases continue to be an
exclusion criteria for liver transplantation (e.g., progressive multifocal
End-stage Liver Disease, HIV Infection and Liver Transplantation  393
leukoencephalopathy, chronic cryptosporidiosis, multidrug-resistant systemic fungal
infections, primary CNS lymphoma, and visceral Kaposi’s sarcoma) (Roland 2006).
Table 1. HIV criteria for OLT in some European countries and the US.
Spain
(Miró 2005)
Italy
(Grossi 2005)
UK
(O’Grady 2005)
US
(Anon 2004)
Previous C events     
Opportunistic
infections
Some*  None in the
previous year.
None after ART-induced
immunological
reconstitution.
Some**
Neoplasms  No  No  No
CD4 cell
count/mm
3
>100***  >200 or >100 if
decompensated
cirrhosis
>200 or >100 if
portal
hypertension
>100***
Plasma HIV-1 RNA
viral load BLD on
HAART****
Yes Yes Yes Yes
* In Spain, patients with previous tuberculosis, Pneumocystis jiroveci pneumonia (PCP) or
esophageal candidiasis can be evaluated for OLT; ** In the US, PCP and esophageal
candidiasis were not exclusion criteria; *** Patients with previous OIs should have >200 CD4
cells/mm
3
; **** If PVL was detectable, post-OLT supression with HAART should be predicted in
all patients. BLD, Below the level of detection
Immunological criteria
All groups agree that the CD4+ lymphocyte count should be above 100 cells/mm
3
for OLT (Roland 2003, Neff 2004). This figure is lower than that for kidney
transplantation (CD4 >200 cells/mm
3
), because patients with cirrhosis often have
lymphopenia due to hypersplenism, which leads to a lower absolute CD4+ count,
despite high CD4 percentages and good virologic control of HIV. In Spain and the
US, the CD4+ count must be greater than 200 cells/mm
3
in patients with previous
opportunistic infections (Miró 2005, Anonymous 2004).
In Italy (Grossi 2005) and the UK (O’Grady 2005) the CD4+ cut-off is 200 cells/
mm
3
, unless patients have decompensated cirrhosis or portal hypertension. In these
scenarios, they use the same CD4+ cell threshold as in Spain and the US (100
cells/mm
3
).
Virologic criteria
The essential criterion for OLT is that the patient must be able to have effective,
safe and long-lasting HAART during the post-transplant period (Neff 2004, Fung
2003). The ideal situation is one in which the patient tolerates HAART before
transplant and is ready for the transplant with undetectable HIV viral load by ultra-sensitive techniques (<50 copies/ml). Some patients do not have an indication for
HAART, as they are long-term non-progressors with no immunological criteria
(CD4+ lymphocyte count above 350 cells/mm
3
) or clinical criteria to start HAART
and a detectable plasma viral load. In this setting, it is unknown whether and when
(pre-transplant or post-transplant) it would be beneficial to initiate HAART in order
to reach an undetectable plasma viral load.
394  Hepatology 2012
Other criteria
To be included on the OLT waiting list, an HIV-infected patient must have a
favourable psychiatric evaluation. One observational prospective study found that
HIV-1-infected patients with ESLD improved on all the items of a psychometric
score (MADRS) at the follow-up evaluation (Barbanti 2009). In this study, the score
variation was 10.20 at baseline and 4.09 at follow-up (p<0.001).
Patients who actively consume drugs should not be placed on the waiting list. In
Spain, patients must undergo a 2-year consumption-free period for heroin and
cocaine (Miró 2005), and 6 months with no consumption of other drugs (e.g.,
alcohol). Patients who are on stable methadone maintenance programmes can be
included and can continue on the maintenance programme after the procedure (Liu
2003). Finally, as is the case with any transplant candidate, HIV-positive patients
must show an appropriate degree of social stability in order to ensure adequate care
in the post-transplant period.
Outcome of OLT in HIV-positive patients
Overall short-term survival rates of HIV-positive patients who undergo OLT have
been reported to be similar to those of HIV-negative patients when there is no HCV
coinfection (Ragni 2003,  Neff 2003,  Fung 2004,  Norris 2004,  De Vera 2006,
Schreibman 2007, Vennarecci 2007, Duclos-Vallee 2008, Tateo 2009, Sioutis 2010,
Coffin 2010, Antonini 2011, Cherian 2011) (Table 2).
HIV+ patients have not been shown to have an increased risk of post-operative
complications or a higher incidence of opportunistic infections or tumours than
HIV-negative patients (Samuel 2008). However, some concerns have lately arisen.
Findings from a case-control study (81 HIV/HCV-coinfected liver transplant
recipients vs. 213 control patients) found that coinfected individuals were about
twice as likely to have treated acute rejection than HCV-monoinfected patients
(35% vs 18%, p=0.001) (Terrault, 2009). Moreover, a recent retrospective study
(Cherian 2011) has observed a 12% (3/24) incidence of post-transplant hepatic
artery thrombosis (HAT) while in people without HIV infection that number is
around 4.4 % (Bekker 2009). The prothrombotic state associated with HIV and liver
disease could be the underlying factor involved.
Regarding infections, it has been observed that bacterial infections are common in
liver (43%) and kidney recipients (35%), whereas HCV infection was the only
factor associated with an increased risk of bacterial infection (liver recipients only)
(Blumberg 2008). A high rate of severe (43%) and opportunistic (11%) infections in
a cohort of 84 HIV/HCV-coinfected patients who underwent liver transplantation
has also been noted (Moreno 2011). Bacterial infections occurred in 38 patients
(45%), CMV infection in 21 (25%), uncomplicated herpes virus infection in 13
(15%), and fungal infections in 16 patients (19%, 7 invasive cases). A pretransplant
MELD score >15, history of category C AIDS-defining event and non-tacrolimus
based immunosupression regimes were factors independently associated with severe
infections. Further studies are needed in order to obtain more robust conclusions.
End-stage Liver Disease, HIV Infection and Liver Transplantation  395
Table 2. Liver transplantation in HIV-infected patients: main cohorts of cases (≥10)
in the late HAART era (2003-2011).
Author  Year  Country  Nº
cases
Virus  Follow-up
(months)
Survival
Ragni  2003  International  24  HCV-62%; HBV-29%   17  18 (75%)
Neff  2003  US  16  HCV or HVB  12  14 (87%)
Fung  2004  US  29  HCV-90%  18  20 (69%)
Norris  2004  UK   14  HCV-50%; HBV/OH-50%
12
19
2 (29%)
7 (100%)
De Vera  2006  US  27  HCV-100%  27  13 (48%)
Schreibman  2007   US  15  HCV-40%;HBV-33%  74  10 (67%)
Vennarecci  2007  Italy  12  HCV-91%  26  6 (50%)
Duclos-Vallée
2008  France  35  HCV-100%  44  22 (63%)
Roland  2008  US  11  HCV-55%; VHB-45%  36  7 (64%)
Tateo  2009  France  13  HBV-100%  32  13
(100%)
Sioutis  2010  Germany  26  HCV-58%; HBV-34%  30  17 (65%)
Coffin  2010  US  22  HBV-100%  42  19 (86%)
Antonini  2011  France  59  HCV-100%  Not
reported
34 (58%)
Di
Benedetto
2011  Italy  23  HCV-87%  24  13 (57%)
Baccarani  2011  Italy  27  HCV-78%  26  23 (85%)
Cherian  2011  UK  24  HCV-50%; HBV-33%  88  15 (71%)
Spanish
study
2011  Spain   248  HCV-96%   50  193
(72%)
HIV/HCV coinfection
Mid-term survival is affected by recurrent hepatitis C (de Vera 2006). After OLT,
recurrence of HCV infection is universal, regardless of whether the patient is
infected by HIV or not. In fact, it is currently the number one cause of death. Some
studies have suggested that recurrence of HCV in coinfected patients tends to be
more severe and occurs earlier (de Vera 2006, Castells 2006, Antonini 2011). The
outcomes of 27 coinfected patients were compared to 54 HCV-monoinfected
patients who underwent OLT (de Vera 2006). The researchers found that HIV-positive patients had a higher likelihood of developing cirrhosis or dying of an
HCV-related complication than HIV-negative patients (RR=2.6; 95%CI, 1.06-6.35).
Cumulative 1-, 3- and 5-year survival for coinfected and monoinfected patients was
67% vs. 76%, 56% vs. 72%, and 33% vs. 72%, respectively (p=0.07).
In a retrospective study (Mindikoglu 2008) in the US that enrolled 138 HIV-positive patients who underwent liver transplant during the HAART era (1996-2006), the rate of survival at years 2 and 3 was significantly lower in HIV-positive
patients (70% and 60%) than in the general population (n=30,520) (81% and 77%),
although this difference was observed only in the HCV/HIV-  and HBV/HIV-coinfected groups. None of the 24 HIV-monoinfected recipients died. Therefore,
liver transplant in HIV-positive patients does not have higher short-term mortality
(1-2 years). Nevertheless, the management and outcome of HCV reinfection could
affect survival in the medium term (3-5 years) and long term (5-7 years).
396  Hepatology 2012
In France,  data from 35 HIV/HCV-coinfected patients were analysed and
compared with those of 44 HCV-monoinfected patients. Survival rates at 2 and 5
years were 81%/91% and 51%/73% in HIV/HCV-coinfected patients/HCV-monoinfected patients, respectively (p=0.004) (Duclos-Vallée 2008).
In Spain, data from a multicentre case-control study show that survival of
HIV/HCV-coinfected patients (N=84) at 1 year was similar to that of HCV-monoinfected patients (N=252) - 88% vs. 89% (NS) - but it was significantly lower
at 3 and 5 years: 62% vs. 77% and 48% vs. 75%, respectively (p<0.01). The
variables independently associated with mortality were HCV genotype 1 infection,
non-traumatic donor death, number of units of blood transfused during surgery, and
development of invasive fungal infection after transplant (Miró 2009). However, a
recent Italian case-control study (Baccarani 2011) that included 27 HIV-positive
and 27 HIV-negative recipients found that patient survival at 1, 2, and 5 years were
88%, 83%, and 83% for HIV-positive patients vs. 100%, 73%, and 73% for HIV-negative patients (p=0.95). The estimated graft survival rate at 1, 3 and 5 years was
92%, 87%, and 87% for HIV-positive patients vs. 95%, 88%, and 82% for HIV-negative patients (p=0.59), respectively. The median follow-up was only 26 months
(range 1-64) and 27 months (range 1-48) for HIV+ recipients and HIV-negative
recipients (p=0.85), respectively, and the aetiology of ESLD was HCV in most
cases (78% vs. 67%, respectively).
In order to definitively answer the survival question, additional cohort studies
analysing donor and recipient characteristics, issues related to the activity of both
viruses and the efficacy and safety of antiviral therapies are necessary for
determining the long-term prognosis of this procedure.
Rapid progression of HCV-related liver disease in HIV-positive recipients would
represent a major drawback and would shorten life expectancy in this group of
patients. In fact, it is currently the primary cause of death. A French study observed
that progression to fibrosis (≥F2) was significantly higher in HIV-positive patients
(p<0.0001) (Duclos-Vallée 2008) and MELD was the only significant predictor of
mortality, although donor age was of borderline significance (p=0.06). 11 (19%) out
of 59 patients who underwent OLT developed fibrosing cholestatic hepatitis (FCH)
(Antonini 2011). Nine of them (82%) died of liver failure after developing FCH.
Survival rate was significantly lower in the FCH group when compared to non-FCH
patients: 26 vs 76 months (p=0.004)
There is insufficient experience on the efficacy and safety of therapy with
pegylated-interferon and ribavirin in coinfected transplant patients. One study (Miró
2007) summarized the reports evaluating the effectiveness of treatment of HCV
reinfection in OLT with pegylated-interferon + ribavirin (Fung 2004, de Vera 2006,
Vennarecci 2007, Castells 2007, Duclos-Vallée 2008, Di Benedetto 2011, Miró
2011). These patients were treated when they had histological criteria. Only 29
(21.3%) out of 136 HCV/HIV-coinfected patients achieved an SVR. SVR-associated factors were investigated in 23 HIV/HCV-coinfected liver recipients and
found that donor age <60 years (p=0.02), genotype other than 1 (p=0.001), and use
of cyclosporin A (p=0.002) were independently associated with SVR (Krishnan
2008). New strategies are necessary to improve the outcome of HCV recurrence in
this setting. In this sense, a German study showed that SVR was obtained in 6 out of
7 patients treated within the first 3 months after OLT (Emmelkamp 2007). The low
accelerating dosage regimen (LADR) approach has recently been addressed in 10
End-stage Liver Disease, HIV Infection and Liver Transplantation  397
coinfected patients who developed fibrosing cholestatic hepatitis after OLT
(Antonini 2011). None of these patients displayed a sustained virological response
while adverse effects were reported in all patients. Anemia was present in the vast
majority of cases and 4 died in the context of the anti-HCV virus. Finally, 2 cases of
spontaneous clearance of HCV RNA after OLT have been described. This
phenomenon is very infrequent and its mechanism is not known (Bhagat 2008).
Table 3. Summary of studies evaluating the effectiveness of the treatment of HCV
reinfection in OLT with pegylated interferon + ribavirin.
Author + Year of
Publication
HIV/HCV coinfected patients  Non-HIV HCV-monoinfected
patients (Control Group)
No. of cases  SVR
No. (%)
No. of cases  SVR
No. (%)
Fung 2004   12  2 (17)  -  -De Vera 2006
b
15  4 (27)  27  7 (28)
c
Vennarecci 2007
d
9  1 (11)  -  -Castells 2007
e
5  1 (20)  9  1 (11)
Duclos-Vallée 2008  19  3 (16)  20  15 (75)
Di Benedetto 2011  9  4 (44) 
Spanish study 2011  67  14 (21)  -  -Total  136  29 (21)  -  -b,Most cases were genotype 1. Three patients were treated with classical interferon plus
ribavirin; c,Rate of sustained virological response was not specified. Data show the rate of
virologic response (clearance of HCV RNA from serum); d,The authors did not specify the type
of interferon used; e,These patients were included in the Spanish study and were not taken into
account for the overall response rate; f,3/27 (11%) genotypes1-4 and 6/13 (46%) genotypes 2-3. SVR: sustained virological response (modified from Miró 2007).
On the other hand, in two recent genome-wide association studies (Ge 2009,
Thomas 2009), a single nucleotide polymorphism (rs12979860) 3 kilobases
upstream of the IL28B gene, which encodes the type III interferon-l, was shown to
be associated with natural clearance of HCV among HIV-negative individuals of
both European and African ancestry and with more than a twofold difference in
response to anti-HCV drug treatment with pegylated-interferon and ribavirin in
HCV-monoinfected patients. In a Swiss study (Rauch 2010), this antiviral effect
was stronger in patients with HCV genotypes 1 or 4. Similar results have been
recently communicated in HCV/HIV-coinfected patients (Rallon 2011). Moreover,
IL28B polymorphisms predicted treatment response in HIV/HCV-coinfected
patients with prior relapse while only predicted response in prior nonresponders
carrying HCV genotypes 1 or 4 (Labarga 2011). The role of IL28B polymorphysms
of the donor and their impact on the natural history of HCV recurrence and response
to antiviral therapy in liver transplant recipients is not yet known.
HIV/HVB coinfection
Cohorts of HIV/HBV-coinfected patients are not as large as those of HIV/HCV-infected patients. Nevertheless, the outcome of HBV infection after OLT is much
better (Tateo 2009, Coffin 2010). The survival rate in the short and medium term in
HBV/HIV-coinfected patients is high and similar to that observed in HBV-monoinfected patients, probably due to the low incidence of HBV reinfection. A
French study (Tateo 2009) that included 13 HIV/HBV-coinfected patients (3 out of
398  Hepatology 2012
6 patients with positive anti-HCV serology had HCV RNA detectable before OLT),
revealed 100% graft and patient survival after a mean follow-up of 32 months. They
compared outcomes of 22 HBV/HIV patients and 20 HBV monoinfected patients
who underwent OLT (Coffin 2010). Cumulative patient and graft survival at one
and three years was 85% in the coinfected patients and 100% in the monoinfected
group (p=0.08).
Hepatocellular carcinoma
Preliminary Italian data show good results in 7 HIV-1–infected patients with HCC
who underwent OLT. They observed an 86% overall patient and graft survival rate
after a mean follow-up of 8 months, and recommend OLT in HIV-infected patients
with early stage HCC (Di Benedetto 2006, Di Benedetto 2008). The same authors
found no recurrence in 7 HIV-1 OLT recipients after a median follow-up of 13
months (Di Benedetto 2008). Recently, data have published from a case (VIH+)
control (VIH-) study of patients with HCC who underwent OLT (Vibert 2011).
They observed that the proportion of patients who exceeded the Milan criteria were
similar in both groups (4/21 (19%) vs. 17/65 (26%), respectively, p=0.5) while no
differences were seen in number of nodules, maximum diameter and microvascular
invasion grade. Time on waiting list was similar (6.4 vs. 4.1 months, respectively
p=0.2). 16 out of 21 HIV-infected patients underwent OLT vs. 58 out of 65 in the
control group. The survival rate at 1 and 3 years was 81% and 55% vs. 91% y 82%
respectively (p=0.005). These authors found a high rate of HCC recurrence in the
case group 5/16 (30%) vs. 9/58 (15%) in the control group. Further studies with
more patients and longer follow-up are needed to precisely define the recurrence
rate in this setting.
Liver retransplantation
The experience of liver retransplantation (re-OLT) in the HIV-infected population is
scarce and most of the cases are mentioned in articles published with patients from a
single center (de Vera 2006, Vogel 2005, Polard 2005). The incidence and outcome
of liver retransplantation in patients with HIV infection is unknown. Currently, in
people without HIV infection, re-OLT accounts for approximately 10% of all liver
transplants (Pfitzmann 2007,  Reese 2009) and overall post-retransplant patient
survival rate is between 15 and 20 percentage points lower than the primary OLT
survival rate (Carrión 2010), which is a concern due to the significant disparity
between the number of patients waiting for their first OLT and the shortage of
available organs. A recent report (Gastaca 2011) observed a 6.7% of re-OLT in a
cohort of 227 patients with HIV infection. Overall survival (95% confidence
interval) at 30 days, 1 year, and 3 years after re-LT for HIV-positive and HIV-negative patients was 93% (59%-99%) vs. 85% (78%-90%), 40% (13%-66%) vs.
72% (64%-78%), and 40% (13%-66%) vs. 64% (55%-71%), respectively
(p=0.231). More patients at risk are needed to provide more conclusive results.
End-stage Liver Disease, HIV Infection and Liver Transplantation  399
Conclusions
ESLD is an increasingly frequent clinical scenario in the setting of HIV/HCV-HBV
coinfection, and its burden is expected to continue to increase.
Early diagnosis of ESLD complications is particularly important and should be
actively monitored and treated. In general terms, the management of ESLD in HIV-positive patients should be the same as in those without HIV infection.
Physicians taking care of ESLD patients should follow them prospectively and
evaluate them for OLT after the first clinical decompensation of liver disease.
OLT is a life-saving procedure in this population, and is safe and effective in
patients with HBV infection. However, recurrence of HCV infection in coinfected
patients can affect both graft and patient survival in the medium- and long-term.
Prospective and larger studies with a longer follow-up must be carried out in order
to determine the benefit of OLT in this setting.
The members of the Hospital Clinic OLT in HIV Working Group are: JM Miró, A. Moreno, C.
Manzardo, M. Laguno, JL Blanco, J Mallolas, C. Cervera, M. Tuset, M. Monras, N. Freixa, J.
Blanch, I.Perez, D. Paredes, J. Fuster, C. Fontdevila, JC García-Valdecasas, JM Gatell, A.
Rimola, (Hospital Clinic – IDIBAPS. University of Barcelona, Barcelona); C. Tural and A. Jou
(Hospital Germans Trías i Pujol, Badalona, Barcelona, Spain); and J. Murillas and  M.
Peñaranda (Hospital Son Dureta, Palma de Mallorca, Spain).
References
Agüero F, Laguno M, Moreno A, et al. Management of end-stage liver disease in HIV-infected
patients. Curr Op HIV/AIDS 2007;2:474-81. (Abstract)
Alter M. Epidemiology of viral hepatitis and HIV co-infection. J Hepatol 2006;44:S6-S9.
(Abstract)
Anonymous. Solid organ transplantation in the HIV-infected patient. Am J Transplant
2004;4:Suppl 10:83-8 (Abstract)
Antonini TM, Sebagh M, Roque-Afonso AM, et al. Fibrosing cholestatic hepatitis in HIV/HCV
co-infected transplant patients-usefulness of early markers after liver transplantation.
Am J Transplant 2011;11:1686-95. (Abstract)
Baccarani U, Adani GL, Bragantini F, et al. Long-term outcomes of orthotopic liver
transplantation in human immunodeficiency virus-infected patients and comparison
with human immunodeficiency virus-negative cases. Transplant Proc 2011;43:1119-22. (Abstract)
Back D, Gibbons S. The University of Liverpool HIV drug interactions website. (Accessed
November 15, 2011 at http://www.hiv-druginteractions.org.)
Barbanti Silva V, Ferrara M, Mazzi F, et al. What is the biopsychosocial role of human
immunodeficiency virus positivity in patients with end-stage liver disease who
undergo orthotopic liver transplantation? Transplant Proc 2008;40:1910-2. (Abstract)
Berretta M, Garlassi E, Cacopardo B, et al. Hepatocellular carcinoma in HIV-infected patients:
check early, treat hard. Oncologist 2011;16:1258-69. (Abstract)
Bekker J, Ploem S, de Jong KP. Early hepatic artery thrombosis after liver transplantation: a
systematic review of the incidence, outcome and risk factors. Am J Transplant
2009;9:746-57. (Abstract)
Benhamou Y, Di Martino V, Bochet M, et al. Factors affecting liver fibrosis in human
immunodeficiency virus-and hepatitis C virus-coinfected patients: impact of protease
inhibitor therapy. Hepatology 2001;34:283-7. (Abstract)
Bhagat V, Foont J, Schiff E, et al. Spontaneous Clearance of Hepatitis C Virus After Liver
Transplantation in Two Patients Coinfected with Hepatitis C Virus and Human
Immunodeficiency Virus. Liver Transpl 2008;14:92-5. (Abstract)
Blumberg E, Barin B, Olthoff K, et al. Bacterial Infections in HIV Positive Transplant Recipients.
American Transplant Congress. May 31- June 4 2008. Toronto Canada. Abstract#
17.