Post-Transplant Infections, Malignancies and Surgical Complications: What's New in 2004
M. Kay Washington
Vanderbilt University Medical Center
Division of the post-transplant period into early (0-7 days), middle (1 week to 2 months)
and late (more than 2 months) intervals provides a useful framework for formulation of differential
diagnoses of transplant dysfunction at different times during the patient's clinical course. These
divisions are, of course, inexact, but do provide a timeframe for particular complications in the
|Time after Transplantation ||Complication ||Key Histologic Features|
|Early (0-7 days)||Preservation/reperfusion injury ||Centrilobular pallor, ballooning degeneration|
|Functional bile flow impairment ||Centrilobular cholestasis|
|HAT thrombosis ||Centrilobular ballooning, necrosis; biliary complications late|
|Middle (1 week to 2 months) ||Acute cellular rejection ||Mixed portal inflammatory infiltrate; bile duct damage, endothelialitis|
|Infections ||Viral inclusions|
|PTLD ||Variable; EBV hepatitis to dense "lymphoma" infiltrate|
|Drug reactions ||Variable|
|Late (> 2 months) ||Recurrent disease ||Features of original disease|
|Chronic rejection ||Bile duct loss, foam cell arteriopathy|
Preservation/Reperfusion Injury and Utility of
Although advances in organ preservation through use of University of Wisconsin (UW) solution have
reduced preservation injury, when storage time exceeds 10-12 hours, post-transplant complications due to
preservation and reperfusion become more common. Ischemic injury to the graft may be divided into two
types: cold ischemia, from prolonged preservation, which injures sinusoidal-lining cells, mostly
endothelial cells, and warm ischemia, which occurs during organ implantation and results in hepatocyte
injury. In cold ischemic injury, sinusoidal lining cells show cellular swelling, nuclear pyknosis, and
detachment, with sinusoidal accumulation of leukocytes and platelets. Kupffer cell activation occurs in
some circumstances, inducing release of IL8, E-selectin, and other compounds that initiate and perpetuate
injury and impair hepatic microcirculation. In warm ischemia, hepatocytes show cytoplasmic
vacuolization, mitochondrial swelling, fragmentation of rough endoplasmic reticulum, and intranuclear
chromatin clumping. The complement pathway is activated by reperfusion, leading to direct cytotoxic
effects of the complement membrane attack complex.
Contributors to preservation/reperfusion injury include:
|Hepatocyte factors: cytochrome p450, mitochondrial dysfunction. Reperfusion injury of parenchymal cells, as judged by serum transaminase levels, generally lasts from several hours to 2 days.|
|Sinusoidal factors: platelet, PMN, Kupffer cell activation with endothelial injury and apoptosis.|
|Bile duct factors: Bile duct cells are more sensitive to reperfusion injury than hepatocytes. Bile duct epithelial cells detach from the basement membrane; greater numbers of cholangiocytes are found in bile with longer preservation times. Injury to bile ducts is associated with increased bilirubin levels, and underlies the long-lasting phase of reperfusion graft injury, with elevated GGT levels persisting up to 17 days and longer. Bile and its components amplify preservation injury, with hydrophobic bile salts injuring hepatocytes as well as biliary epithelium. Hydrophilic bile salts may offer a protective effect to hepatocytes.|
The role of Kupffer cells in graft injury is controversial and may depend on the model chosen for
study. Contribution of Kupffer cells may be most pronounced when there is an additional insult such as
endotoxemia, or impaired microcirculation in fatty liver.
While even grafts considered to function well may show severe impairment of hepatocellular secretory
function, fatty livers are highly sensitive to both warm and cold ischemic injury. Sinusoidal congestion
and endothelial cell injury are the key events.
Morphologic changes of preservation/reperfusion injury in the early post-transplant period are
characterized by centrilobular hepatocyte swelling, which imparts a distinctive pale appearance to
central areas, with cholestasis seen as an additional finding in more severe cases. These changes may
persist for 10-14 days, but should show improvement on successive biopsies. Cholestasis may be
particularly persistent. In severe preservation injury, hepatocyte necrosis in centrilobular areas may
be seen, and periportal injury with bile ductular reaction and bile plugs in periportal cholangioles,
similar to that seen in cholestasis due to sepsis.
In many centers, post-perfusion biopsies at the time of graft implantation are routinely obtained.
Such biopsies may be helpful in predicting development of clinically evident preservation injury. In
particular, sinusoidal neutrophilic infiltrate and hepatocyte necrosis have been shown to predict
development of preservation injury and initial poor function
. Livers with steatosis on T0
biopsies are also more likely to develop preservation injury  and graft loss  . Preservation
injury may also be associated with subsequent development of biliary complications  . In experimental
models, hepatocyte injury, rather than endothelial cell damage, may be more predictive of subsequent
graft function, an observation that supports the correlation of hepatocyte necrosis with poor graft
function in retrospective studies of T0 biopsies.
Diagnosis of vascular complications (hepatic artery thrombosis or stenosis, portal vein
thrombosis, and hepatic vein thrombosis) following OLT may prove challenging for both the clinician and
the pathologist, and a high index of suspicion is necessary to make such a diagnosis at an early stage.
Hepatic artery thrombosis (HAT) occurs in 1.6%
to 10.5% of adults and 10-25% of children and is the most common vascular complication in OLT. HAT
stenosis is less common, occurring in up to 5% of patients, with median time to diagnosis 100 days; no
particular risk factors are identified.
Hepatic artery thrombosis may be classified as early, defined as within 30 days of OLT, or
late. HAT early after OLT is associated with severe graft dysfunction and carries a high mortality
rate. Late HAT may have a milder clinical course. In some centers, late HAT is more common than early
HAT, perhaps because it is detected by protocol Doppler exams when clinically silent. Biliary
complications in general are more frequent in late HAT, and severe organ dysfunction in early HAT  .
Portal vein thrombosis/stenosis is lesscommon
that HAT. Hepatic vein thrombosis is unusual,
except in patients undergoing OLT for Budd-Chiari syndrome, but may be more common in living related
| ||Early HAT ||Late HAT ||Portal vein thrombosis|
|Incidence ||1.6 to 10.5% (adults); up to 25% (children) ||1.7  to 2.8 %  ||Rare|
|Presentation ||Fever, increased ALT/AST, Fulminant hepatic failure, relapsing bacteremia, biliary tract necrosis ||Asymptomatic; recurrent cholangitis, biliary strictures, abscess, bile leak, fever ||Refractory ascites, variceal bleeding, encephalopathy|
|Associated Conditions ||Complex anatomy, small HA, extension graft, dehydration, prolonged ischemic time, PSC,  CMV, small donor liver  ||Complex anatomy, cigarette smoking ; CMV  ||PVT in native liver, interposition graft, previous portocaval shunt, hypercoagulable state|
|Outcome ||Mortality up to 73% without Retransplantation, otherwise 27-58% ||Unknown for asymptomatic; poor outcome for retransplantation if septic ||Early PVT: Loss of graft if not revascularizedLate PVT: similar to non-transplant PVT|
|Biopsy Findings ||Centrilobular coagulative necrosis or ballooning degeneration; infarcts;? increased apoptosis with minimal inflammation ||Not well described; bile duct necrosis; hepatic abscess; bile ductular reaction ||Portal fibrosis; rarely, sclerosis of PV branches; hepatocyte atrophy and sinusoidal dilatation; rarely, infarcts|
Morphology and differential diagnosis of vascular
Morphologic changes due to hepatic artery thrombosis may be quite variable and irregularly
distributed in the graft. Severe injury is characterized by coagulative necrosis, which may become
secondarily infected, leading to hepatic abscesses. Injury to the biliary tree may be manifested as
biliary necrosis. Canalicular cholestasis and bile ductular reaction are non-specific findings, but may
be seen in parenchyma adjacent to infarcts. Some investigators have suggested that increased numbers of
apoptotic hepatocytes may be seen in mild ischemia prior to severe necrotic damage. By using TUNEL assay
to compare apoptosis in acute rejection with vascular occlusion  , Sedivy and colleagues found that
the number of apoptotic bodies was much higher in vascular occlusion than in acute rejection; apoptotic
bodies were found in periportal areas as well as centrilobular areas.
Distinguishing ischemic injury from acute cellular rejection generally does not present
diagnostic problems. The mixed portal inflammatory infiltrate seen in acute rejection is not a feature
of ischemic injury. The isolated centrilobular pattern of rejection may be confused with ischemic
injury, but generally shows a more pronounced lymphocytic inflammatory infiltrate with endothelialitis.
In the setting of recurrent hepatitis C, increased apoptotic bodies are an expected finding and cannot be
used as an indication of ischemia. In addition, apoptotic bodies in biopsies in the early transplant
period may be a manifestation of the expected modulation of the size of the graft.
Infectious complications of liver transplantation are generally manifested one week or longer after
transplantation. Whereas bacterial and fungal infections often involve other organ systems and do not
involve liver biopsy for diagnosis, viral infections may directly involve the liver allograft and require
biopsy to differentiate them from acute rejection. Cytomegalovirus, in particular, is not uncommon in the posttransplant
period, with some centers reporting incidences of over 20%. Mean time to occurrence is 38 days after OLT
(8-290 days), although late presentations are reported (up to almost 10 years after OLT), with an
incidence of 8.5% in the total cohort at one transplantation center  . In this study, all patients
with late CMV infections were women, and all were seropositive before OLT. Clinical presentation in all
cases was fever and hepatitis; one patient died of bacterial sepsis. The most important risk factor is
CMV-positive serology of the donor, regardless of recipient serologic status.
CMV may be a co-factor in complications following OLT and is sometimes associated with bacterial and
fungal infections, acute and chronic rejection, and hepatic artery thrombosis. Contribution from CMV to
development of allograft rejection may be by upregulation of MHC Class I and II via release of
interferons. However, short term CMV viremia does not appear to worsen hepatitis C recurrence  .
Persistent CMV infection has been demonstrated in patient with chronic rejection using in situ
hybridization, which proved more sensitive than immunoperoxidase techniques in this study  . 12/29
patients were CMV positive by immunoperoxidase or ISH. However, CMV was found in 8/15 explanted grafts
but was not associated with etiology of graft loss, which included chronic rejection (8 cases), and
hepatic artery thrombosis (3 cases).
Cytomegalovirus infection in the transplanted liver is usually manifested as scatted clusters of
necrotic hepatocytes with associated microabscesses. Kupffer cells may be prominent and aggregated into
small nodules, or more rarely, granulomas. Characteristic nuclear inclusions may be sparse (only 1-2 per
biopsy core) and identifiable only on immunoperoxidase stain. The inclusions are rarely seen in biliary
epithelium. The most common finding in one study describing the morphology of CMV hepatitis in OLT
biopsies was mild to moderate portal inflammation with lymphocyte predominance, found in 29/30 biopsies
 . Other common findings were hepatocyte ballooning, slight cholangitis, Kupffer cell prominence,
parenchymal and sinusoidal inflammation, and microabscesses (26/30)
. Viral inclusions were found in only
3/30 cases with use of immunoperoxidase methods on frozen sections.
Smaller collections of lobular neutrophils do not seen to be related to CMV infection and have no
influence on graft survival. MacDonald and colleagues studied 57 OLT biopsies with microabscesses and
compared these to 19 biopsies from OLT patients with known CMV infection  . CMV infection occurred
at a median of 32 days, while the "minimicrobscess syndrome" occurred at a median of 91 days. These
patients usually presented with marked elevation of serum transaminases on routine blood work. The
investigators could not detect CMV in the biopsies with nested PCR methodology, and proposed that the
microabscesses may be a manifestation of a viral infection. We also studied our liver transplant
biopsies with microabscesses, and found that numerous microabscesses in the biopsy (>9) correlated
with CMV infection, while the size of the microabscesses did not  . In our study, 43/97 patients had
microabscesses at some point on liver biopsy, with the following associated conditions:
|19% had CMV|
|27% had other infections|
|10% had graft ischemia|
|15% had biliary obstruction/cholangitis|
|3% had sepsis and ischemia|
|26%- no etiology found|
Recently, human herpes virus-6 infections have
been reported in solid organ transplant patients. This virus, like other herpes viruses, can establish a
latent or persistent infection. The clinical picture varies from asymptomatic, to mild skin rash, to
severe disseminated disease. The frequency of active HHV-6 infections has been reported as high as
38-82% in renal transplant patients, and an association with allograft rejection has been suggested.
Post-OLT infection has been described in 8 patients  , with diagnosis based on serology and antigen
detection in liver biopsies. The liver biopsies were remarkable for a lymphocytic infiltrate. Most
antigen-positive cells were lymphocytes, and the virus was not identified in parenchymal cells. Only two
of the reported cases were not associated with rejection or with CMV infection; all were adults with
reactivation, not primary infection, and had mild clinical symptoms. In addition to the association with
increased risk of CMV disease and opportunistic infections
, HHV-6 infection in pediatric liver
transplant patients may also be associated with development of acute rejection  .
Fungal infections remain a serious problem in
management of liver transplant patients, occurring in 5-42% of OLT recipients, with associated mortality
of 25 to 67%. Liver transplant patients are particularly prone to invasive candidiasis, which comprises
62-91% of fungal infections in this population. This propensity for candidal infection is influenced by
surgical factors, including technical complexity, prolonged operation time, greater transfusion
requirements and bleeding complications requiring re-operation, and type of biliary anastomosis. Length
of post-OLT ICU stay is also associated with increased risk. However, one study of evolving trends in
fungal infection in OLT patients over a 10 year period found a significant decline in Candida infections
from 9% to 1.7%, while overall incidence of fungal infections remained unchanged  . This change may
be related to improved surgical techniques and post-operative management, with decrease in OR time, blood
loss, cold ischemic time, roux-en-Y anastomosis, rate of retransplantation and rejection episodes.
De Novo Malignancies
Contribution of a dysfunctional immune system to risk for developing malignancy is currently an area
of interest at many transplant centers. As survival following OLT improves, ascertainment of late causes
of death becomes more important in improving patient survival. In most studies, the incidence of de novo
malignancies in the OLT population is 3-15%, roughly twice that in the general population
one study examining causes of death in 38 liver transplant patients who survived at least three years
(total cohort 299 patients), de novo malignancies and cardiovascular disease accounted for more than half
of the late deaths.
Development of malignancy in the setting of solid organ transplantation is multifactorial,
and depends upon individual and regional predispositions to malignancy, pretransplantation disease
states, recipient viral status, and use and intensity of immunosuppression regimens. Risk is cumulative,
and increases with time from transplantation, from 13% after 5 years, to 26% after 8 years in one study
 . By 10 years, 30% had developed de novo cancers. Other investigators estimate risk from 4.1 to
16%, depending on population, type of transplant, and follow up. Standardized incidence ratio is 3.94
for non-skin cancers, 3.14 for nonmelanoma skin cancer (>1 indicates excess risk) in some series 
. Much of the data on de novo malignancies comes from registries, with the limitation of voluntary
reporting and lack of uniformity in reporting.
In general, the pattern of malignancies in OLT patients is similar to that in the general population.
Lymphomas are more common in liver recipients, compared to kidney, and develop in a shorter time (27 v 72
months)  . Skin , cervical, renal, and vulvar carcinomas are more common in kidney transplant
recipients. In a study from
Pittsburgh, 1000 patients were followed for mean of 93 months prospectively. In this group,
44 developed post-transplant lymphoproliferative disorder, and 81 developed
|35 skin cancers (2 melanomas, 2 Kaposi's sarcoma)|
|11 GI cancers|
|8 lung cancers|
|7 oropharyngeal cancers|
|3 breast cancers|
|2 carcinoma of unknown primary|
|2 thyroid cancers|
|1 each: brain, hepatocellular carcinoma, eye|
The incidence of breast cancer and GU cancer was less than matched cohorts. The rate for
GI malignancies was not significantly greater than in general population; risk factors for development of
colon cancer in this group were age >45, primary sclerosing cholangitis, length of time with
ulcerative colitis, and presence of colon polyps. Enhanced cancer susceptibility for patients with
alcoholic liver disease, mainly for oropharyngeal and lung cancer, was noted.
Of note, Kaposi's sarcoma (KS) occurs in up to 1% of adult recipients, depending upon the
patient population. One center reported KS in 5/459 adult recipients  . Multivisceral involvement
was common, and in 3, the graft was involved. Involvement of the hepatic hilum caused chronic biliary
obstruction in one case. Patient survival is poor.
- Busquets J, Figueras J, Serrano T, Torras J, Ramos E, Rafecas A, et al. Postreperfusion biopsies are useful in predicting complications after liver transplantation. Liver Transplantation. 2001;7(5):432-435.
- Gaffey MJ, Boyd JC, Traweek ST, Ali MA, Rezeig M, Caldwell SH, et al. Predictive value of intraoperative biopsies and liver function tests for preservation injury in orthotopic liver transplantation. Hepatology. 1997;25:184-189.
- Washington MK, Pinson CW, Raiford DS, Chapman WC. Pre- and post-perfusion biopsies of the hepatic allograft: how useful is histopathologic assessment? Laboratory Investigation 2002;82:296A.
- Busquets J, Figueras J, Serrano T, Torras J, Ramos E, Rafecas A, et al. Postreperfusion biopsy changes predict biliary complications after liver transplantation. Transplantation Proceedings. 2002;34(1):256-8.
- Stange BJ, Glanemann M, Nuessler NC, Settmacher U, Steinmuller T, Neuhaus P. Hepatic artery thrombosis after adult liver transplantation. Liver Transplantation. 2003;9(6):612-620.
- Gunsar F, Rolando N, Pastacaldi S, Patch D, Raimondo ML, Davidson B, et al. Late hepatic artery thrombosis after orthotopic liver transplantation. Liver Transplantation. 2003;9(6):605-611.
- Bhattacharjya S, Gunson BK, Mirza DF, Mayer DA, Buckels JA, McMaster P, et al. Delayed hepatic artery thrombosis in adult orthotopic liver transplantation-a 12-year experience. Transplantation. 2001;71(11):1592-6.
- Oh CK, Pelletier SJ, Sawyer RG, Dacus AR, McCullough CS, Pruett TL, et al. Uni- and multi-variate analysis of risk factors for early and late hepatic artery thrombosis after liver transplantation. Transplantation. 2001;71(6):767-72.
- Sedivy R, Gollackner B, Casati B, Mittlbock M, Kaserer K, Steininger R, et al. Apoptotic hepatocytes in rejection and vascular occlusion in liver allograft specimens. Histopathology. 1998;32(6):503-7.
- Shibolet O, Ilan Y, Kalish Y, Safadi R, Ashur Y, Eid A, et al. Late cytomegalovirus infection occurring two or more years following liver transplantation: a report of seven cases and review of the literature. Transplantation Proceedings. 2003;35:663-664.
- Teixeira R, Pastacaldi S, Davies S, Dhillon AP, Emery VC, Rolles K, et al. The influence of cytomegalovirus viraemia on the outcome of recurrent hepatitis C after liver transplantation. Transplantation. 2000;70(10):1454-8.
- Evans PC, Coleman N, Wreghitt TG, Wight DG, Alexander GJ. Cytomegalovirus infection of bile duct epithelial cells, hepatic artery and portal venous endothelium in relation to chronic rejection of liver grafts. Journal of Hepatology. 1999;31(5):913-20.
- Lautenschlager I, Hockerstedt K, Taskinen E. Histologic findings associated with CMV infection in liver transplantation. Transplantation Proceedings. 2003;35(2):819.
- MacDonald GA, Greenson JK, DelBuono EA, Grady WM, Merion RM, Frank TS, et al. Mini-microabscess syndrome in liver transplant recipients. Hepatology. 1997;26(1):192-7.
- Lamps LW, Pinson CW, Raiford DS, Shyr Y, Scott MA, Washington MK. The significance of microabscesses in liver transplant biopsies: a clinicopathological study. Hepatology. 1998;28(6):1532-7.
- Lautenschlager I, Hockerstedt K, Linnavuori K, Taskinen E. Human herpesvirus-6 infection after liver transplantation. Clinical Infectious Diseases. 1998;26(3):702-7.
- Humar A, Kumar D, Caliendo AM, Moussa G, Ashi-Sulaiman A, Levy G, et al. Clinical impact of human herpesvirus 6 infection after liver transplantation. Transplantation. 2002;73(4):599-604.
- Razonable RR, Paya CV. The impact of human herpesvirus-6 and -7 infection on the outcome of liver transplantation. Liver Transplantation. 2002;8(8):651-8.
- Feldstein AE, Razonable RR, Boyce TG, Freese DK, El-Youssef M, Perrault J, et al. Prevalence and clinical significance of human herpesviruses 6 and 7 active infection in pediatric liver transplant patients. Pediatric Transplantation. 2003;7(2):125-9.
- Singh N, Wagener MM, Marino IR, Gayowski T. Trends in invasive fungal infections in liver transplant recipients: correlation with evolution in transplantation practices. Transplantation. 2002;73(1):63-7.
- Planas JMM, Gonzalez ER, de Tejada AH, Renedo FP, Sanchez-Turrion V, Jimenez BV, et al. Late mortality in patients with liver transplantation: causes and risk factors. Transplantation Proceedings. 2003;35:707-708.
- Sanchez EQ, Marubashi S, Jung G, Levy MF, Goldstein RM, Molmenti EP, et al. De novo tumors after liver transplantation: a single-institution experience. Liver Transplantation. 2002;8(3):285-291.
- Schmilovitz-Weiss H, Mor E, Sulkes J, Bar-Nathan N, Shaharabani E, Melzer E, et al. De novo tumors after liver transplantation: a single-center experience. Transplantation Proceedings. 2003;35:665-6.
- Xiol X, Guarfiola J, Menendez S, Lama C, Figueras J, Marcoval J, et al. Risk factors for development of de novo neoplasia after liver transplantation. Liver Transplantation. 2001;7(11):971-5.
- Fung JJ, Jain A, Kwak EJ, Kusne S, Dvorchik I, Eghtesad B. De novo malignancies after liver transplantation: a major cause of death. Laboratory Investigation 2001;7(11, Suppl 1):S109-S118.
- Aseni P, Vertemati M, Minola E, Arcieri K, Bonacina E, Carmozzi M, et al. Kaposi's sarcoma in liver transplant recipients: morphological and clinical description. Liver Transplantation. 2001;7(9):816-823.