|
|
||||||||
Articles |
Department of Chemical Pathology, Queensland Liver Transplant Service,
1
Princess Alexandra Hospital, Woolloongabba, Queensland, Australia 4102, and
2
Royal Brisbane Hospital, Herston, Queensland, Australia 4029.
3
Department of Pathology and Laboratory Medicine, College
of Medicine, University of Cincinnati, Cincinnati, OH 45267-0714.
a Author for correspondence. Fax 61 7 3240 7070;
| Abstract |
|---|
|
|
|---|
Key Words: indexing terms: liver transplantation liver function tests laboratory costs graft ischemia rejection therapeutic drug monitoring
| Introduction |
|---|
|
|
|---|
After the patient is wheeled from the operating theater to the
intensive care ward, several important problems may arise, for which
the laboratory may be central to the diagnosis and management. The most
important of these are listed in Table 1
. For convenience these problems are listed at the time at which
they most commonly present, although many of them can occur at any
time. Some of these problems are considered individually below.
|
All laboratories are being forced to exercise rigorous cost containment. Thus assessment of the liver transplant recipient requires the choice of tests that provide adequate diagnostic information, at the minimum cost. Whereas many tests have been proposed as being potentially useful in assessing the posttransplant recipient, only tests that can be turned around in a time course that makes them of diagnostic use are of real value in routine clinical practice on the transplant ward. This generally means within 2 h.
Most of the routine tests that are used in assessing the liver
transplant recipient are individually nonspecific. However, because of
the high prevalence of disease, the use of tests in combination
increases their diagnostic efficacy. In our experience, most problems
can be satisfactorily assessed with a routine panel of liver function
tests (LFTs)1
generated quickly and cheaply on the laboratory analyzer,
which operates 24 h each day. Our routine profile of LFTs is shown
in Table 2
. In themselves they are useful in that they identify the
presence of a problem, but not the problem itself. Abnormal test
results can be meaningful only when used with other data, e.g.,
coagulation results, and when the clinical status of the patient is
considered.
|
Tests of liver function need not only be available from the laboratory. Some liver transplant units leave in place a T-tube draining the bile duct after surgery, whenever technically feasible. With such patients, an excellent general indicator of a successful operation comes independently of the laboratory: the flow of black bile from the draining T-tube.
| Immediate Problems |
|---|
|
|
|---|
All transplanted livers show some degree of injury posttransplant. This
may be a result of the period of ischemic storage between the time of
harvest and transplant or reperfusion injury (5). Fig. 1
shows the changes in the routine LFTs in various clinical
states after transplantation. Even in uncomplicated cases, on day 1
plasma transaminase activity will be increased at least 45 times the
upper limit of the reference range. The prothrombin time (PT) settles
quickly to near-normal values. The initial bilirubin concentration is
dependent upon the nature of the pretransplant liver disease, often
with a small rise during the first week as blood products are
metabolized. Its concentration may increase posttransplant as a
consequence of tissue injury during surgery and blood transfusion. In
uncomplicated cases, the biliary markers alkaline phosphatase (ALP) and
-glutamyltransferase (GGT) usually remain within the reference
range.
|
Two important complications may arise immediately on arrival in intensive care: primary nonfunction and hepatic artery thrombosis. The latter may present weeks after the surgery but is considered here for convenience.
graft ischemia and primary nonfunction
There may be extended periods of ischemic storage between harvest
and revascularization. Immediately after the blood supply is
reconnected, there will be evidence of hepatocyte injury, with a rise
in plasma transaminase activity that is approximately proportional to
the degree of injury suffered. The change of storage solution for donor
livers from EuroCollins to University of Wisconsin solution
dramatically reduced the degree of ischemic injury. Signs of severe
injury are poor bile flow, high transaminase activities, and
coagulopathy. Fig. 1
shows the changes in LFTs from a case with severe
ischemic liver damage. The transaminase activity on day 1 was nearly 80
times the reference range. PT remained elevated to at least twice the
normal value for 10 days. The bilirubin concentration rose
progressively over the fortnight posttransplant, poor liver function,
tissue injury, and blood transfusion all contributing. Cholestasis is
common after severe ischemia, and activities of both ALP and GGT
increased.
At its worst, ischemic injury becomes primary nonfunction, which is a catastrophic although infrequent complication of liver transplantation. The reported incidence in most series averages ~6% (6). The cause of primary nonfunction is gross graft ischemia with massive infarction of the liver secondary to various pathologies such as hepatic artery or portal vein thrombosis, preservation injury or hypoperfusion of the donor liver due to prolonged hypotension, or cardiac arrest before or during donor hepatectomy, although it appears that some preexisting pathologies in the donor liver, e.g., steatosis, may contribute. The presentation is usually dramatic and resembles fulminant hepatic failure with worsening coagulopathy refractory to clotting factor infusions, gross increase of transaminases (>50x), hypoglycemia, acidosis, hyperkalemia, hypophosphatemia, and the clinical features of oliguria, hypotension, adult respiratory distress syndrome, and progressive cerebral edema. Some patients with primary nonfunction may have massive ascitic fluid loss, which can confound the diagnosis. Because of the loss in the ascitic fluid the rise in transaminases may be much smaller than usually seen (7). Clotting factors may be lost in ascitic fluid and if the ascitic fluid loss is due to reasons other than primary nonfunction, a similar prolongation in PT may still result. However, if due to dilutional causes, there is a good response to infusion of fresh frozen plasma.
hepatic artery thrombosis
Hepatic artery thrombosis is an important potential complication
postoperatively. In established programs, the incidence is
characteristically on the order of 510% (8). Although
ligation of the hepatic artery as a treatment of liver metastases is
usually well tolerated, in the liver transplant patient hepatic artery
thrombosis is associated with considerable morbidity and mortality,
presumably because the graft is devoid of the normal collateral blood
supply delivered through the hepatic ligaments. While thrombosis may
occur at any time after transplantation, most cases occur within the
first 2 weeks, and the complication is often fatal without
retransplantation. Late hepatic artery thrombosis may be asymptomatic,
although a more distinct presentation usually occurs. This may be
relatively subtle and present as relapsing septicemia following bile
duct ischemia, or at the other extreme it may present as fulminant
hepatic failure with an abrupt clinical deterioration accompanied by
gross prolongation of the PT and massive increase of the transaminases.
In the latter circumstance, without prompt retransplantation, the
mortality is 100%. Because of the variable clinical presentation, a
high index of suspicion is required. Dynamic, real-time tests such as
the monoethylglycine xylidide test (9) may be of value in
identifying the presence of problems such as the less obvious hepatic
artery thrombosis; we emphasize, however, that the tests are not
specific and serve simply to indicate that a problem exists.
In our experience we have found it necessary to provide a 24-h stat service for the routine LFT profile and the monoethylglycine xylidide test, and our hematology service offers a coagulation service including PT.
| Intermediate Problems |
|---|
|
|
|---|
Diagnosis of rejection on clinical grounds alone is unreliable, and laboratory data are also required. Liver biopsy is still generally considered to be the best means of objectively identifying developing liver transplant rejection (13). However, biopsy is a procedure that is time-consuming and not without morbidity, and by itself is not an absolutely reliable index of clinically significant rejection with a substantial number of false positive results (14)(15). The search continues for other reliable indices of rejection.
To identify possible useful indices of rejection, some understanding of the rejection process is necessary. Briefly, it is generally agreed that during acute rejection expression of both Class I and Class II antigens, primarily involving the biliary epithelium and vascular endothelium, is greatly increased. The major target for the inflammatory response is thus the biliary tract and to a lesser extent the hepatocytes (10). Thus indices of Class I and Class II antigen expression, the inflammatory response directed against these foreign antigens or markers associated with the biliary system, might be anticipated to provide the greatest information in the early stages of acute rejection. As the initial target in the rejection process is the biliary tree, it might be anticipated that bile should reflect any pathological changes. An indwelling T-tube provides ready access to bile, and samples can be collected directly from the site of rejection, which theoretically should provide more useful information than in peripheral blood.
nonspecific indices associated with rejection
Because of the nature of the rejection process, it is of
particular value to look at specific markers. However, such markers are
not readily available in all centers, the assays usually have a longer
turnaround time, and costs are usually substantially more than the
routine, simple LFTs. Although standard LFTs are nonspecific, how
informative they are with regard to liver transplant rejection should
be considered.
Because the rejection process involves an attack upon the grafted
liver, and particularly upon the biliary system, that LFTs should
change is to be expected, and many papers have reported attempts to use
these comparatively simple measures to identify developing rejection.
That results are quickly available is an advantage of these tests. The
disadvantage is that they are nonspecific. For example, after severe
ischemia, cholestasis is common, and the changes seen may be similar in
cholestasis and rejection, thus indicating the importance of
considering results only in the context of the clinical history. Fig. 1
shows the changes in LFTs in a case with mild to moderate rejection,
confirmed by liver biopsy on day 10, and treated with
methylprednisolone. The transaminase activity was initially at 20 times
the reference range, indicating a moderate degree of ischemic injury,
independent of the rejection episode. The PT fell steadily towards the
reference range. The bilirubin concentration rose over the first 8 days
and started to fall before the rejection was treated. GGT rose to twice
the reference range on day 34 and rose progressively until the
rejection was treated. ALP did not rise above the reference range until
day 9.
That an increase in bilirubin concentration is an early sign of
developing rejection is a widely held belief
(16)(17). However, in our experience bilirubin
changes are of little help in this context. Fig. 2
shows the changes in bilirubin concentration in nine patients
with biopsy-diagnosed steroid-sensitive rejection, over the period
before and after treatment with methylprednisolone (1 g intravenously
on 3 consecutive days). Whereas almost all cases show a decrease in
bilirubin concentration after treatment, the pattern before treatment
is very variable, and no confident prediction about developing
rejection could be made on the basis of the bilirubin concentration
alone. Application of receiver-operating characteristic analysis
confirms the poor diagnostic value for bilirubin (18).
Acute rejection is common within the first few days after
transplantation, at a time when the tissue damage associated with the
major surgery and occasional large blood transfusions intraoperatively
make it likely that bilirubin concentration will be increased without
any diagnostic importance being attached.
|
Because the initial attack in a rejection episode is specifically directed against the biliary system, it might be anticipated that biliary markers would be the most informative. ALP is the usual marker used in this capacity (13)(19) although it is not a particularly good marker of rejection. Despite periodic reports of the value of GGT as a marker of rejection (20)(21), this enzyme has been undervalued in this context. We have recently studied the relative utility of GGT and ALP in the diagnosis of rejection and found GGT to be very much more useful with a diagnostic sensitivity for clinically significant rejection of 91.0% compared with ALP's diagnostic sensitivity of 68.7%. The positive predictive value for GGT was 70.9% and for ALP 67.6% (22). We believe that GGT estimation is an essential cost-effective component of the posttransplant LFT panel used for patient monitoring.
Because the initial target in rejection is the biliary system, transaminases that reflect hepatocyte damage should not rise until rejection is well established. A rise in transaminase activity should therefore not be used as a principal index of rejection. It has been reported that an early rise in transaminase activity associated with rejection is associated with more severe disease (23), although this assertion is disputed (24).
A variety of other nonspecific markers, e.g.,
-glutathione
S-transferase (
-GST) (25), have been
proposed as useful indices of rejection. For the reasons outlined
above,
-GST, an intrahepatocyte enzyme, should rise later in the
rejection process than the biliary markers, although it is an excellent
index of hepatocyte injury. More-specific markers have been proposed,
and the performance of some of these is considered below.
specific products related to lymphocyte activation
When a foreign graft is inserted into a recipient, recipient
leukocytes respond to the foreign antigens by producing soluble factors
that induce both lymphocyte and antigen proliferation. Thus cytokines
such as the interleukins (IL) IL-1, IL-2, IL-5, IL-6, tumor necrosis
factor (TNF), and
-interferon (
-IFN) may be expected to rise in
response to an episode of acute rejection. Substances such as
ß2-microglobulin (ß2-M), intercellular adhesion
molecule-1 (ICAM-1) and neopterin, which are induced by cytokines, may
also be expected to be of potential use. However, secretion of these
substances is a reflection of leukocyte activation and is not specific
for rejection. Below we present an overview of the extensive literature
on potential markers of liver transplant rejection.
Measurement of IL-1, TNF, and
-IFN has proven disappointing, with
increases in all agents seen in rejection, infection, and other
complications (26). Likewise, increases in IL-6 in bile
are seen in both rejection and inflammation (27). Serum
concentrations of IL-5, the T cell-derived eosinophil-activating
cytokine, may be increased in both rejection and cholangitis, but
biliary concentrations appear to have high specificity with relatively
low diagnostic sensitivity (28). Whereas serum
concentrations of soluble IL-2 receptor are higher in rejection,
substantial overlap occurs with amounts seen in infection
(29). However, biliary IL-2 receptor is reported to have a
diagnostic sensitivity of 94% and a specificity of 84% for acute
rejection (30).
Serum concentrations of ß2-M rise in response to any substantial
inflammation (26) and are not specific for rejection.
However, if the ratio of bile:serum ß2-M is used, then a very high
specificity for rejection has been reported with little overlap between
rejection and infection, and a diagnostic sensitivity of 96% and
specificity of 87% (31). ICAM-1 is an adhesion molecule
induced by cytokines that is found on the surface of several different
cells and appears important in the rejection process. Serum
concentrations rise in response to a variety of inflammatory processes,
but it is reported that ICAM-1 in bile increases only in response to
developing rejection (32). Neopterin is released by
stimulated leukocytes under the control of
-IFN. Serum
concentrations rise in response to various inflammatory stimuli
(33), but in bile a rise in neopterin concentration is
reported to be specific for rejection (34). Our own
experience with cytokines and neopterin in bile, however, is that
substantial intraindividual variation exists and that concentrations
may be increased in conditions other than rejection.
Based upon our practical experience and from our reading of the
literature, Table 3
compares the relative diagnostic efficacy of some of these
proposed markers.
|
| Longer-Term Monitoring of the Liver Transplant Patient |
|---|
|
|
|---|
Measurement of blood azathioprine and prednisolone concentrations is performed infrequently. Occasionally, area under the curve measurement of prednisolone may be useful in determining dosage (36). Azathioprine's major toxic effects are upon the bone marrow. The most important side effects of treatment with prednisolone include glucose intolerance and diabetes mellitus (37), hyperlipidemia (38), and bone loss (39).
The mainline drug for immunosuppression in liver transplantation is CsA, with tacrolimus being used increasingly either as first-line immunosuppression or as rescue therapy in rejection. Both drugs act primarily by blocking synthesis of cytokines, including IL-2 and TNF (40). CsA in particular is poorly absorbed, although the new formulation of CsA, known as Neoral (Novartus), has much greater bioavailability. Both CsA and tacrolimus are toxic in high concentrations, and low blood concentrations are associated with rejection (41)(42). Thus the first requirement from the laboratory is drug measurement. Because of the extreme lipophilicity and temperature-dependent partitioning into erythrocytes of these drugs, whole blood is required for analysis, and the best clinical correlation is found when trough concentrations are used (43). Target blood concentrations fall with time posttransplant. CsA may be assayed by either HPLC or immunoassay. Because of the potential problem of metabolites cross-reacting in immunoassays, some laboratories still use HPLC for analysis, despite the increase in time and operator expertise that is required. Tacrolimus is more potent than CsA, thus blood concentrations are lower. Assay is principally by ELISA, although liquid chromatography-mass spectrometry-mass spectrometry analysis has been used (44). Special preparation is required for these assays; by negotiation with our transplant unit these assays are not available on a stat basis, but CsA is assayed every day of the week.
Therapeutic drug monitoring is not only therapeutic drug measurement. All immunosuppressant drugs have important side effects that require laboratory support for assessment. CsA is nephrotoxic, particularly at higher concentrations (45). With multiple immunosuppressant drugs being used simultaneously, it is sometimes difficult to identify which drug is causing which effect. However, CsA by itself certainly worsens glucose tolerance (46), causes hyperlipidemia (47) and hypomagnesemia (48), and increases bone turnover (49). Hepatic and pancreatic toxicity (50) have both been reported. Tacrolimus has a similar toxicity profile, causes more hepatotoxicity (51), and is more diabetogenic (52), but causes less hyperlipidemia (53).
Mycophenolate mofetil has been developed essentially as an azathioprine substitute. Early trials suggest that it is more efficacious than azathioprine (54) and that it has a relatively low toxicity. Currently it is being promoted as a drug for which therapeutic drug monitoring is not helpful. This claim requires further analysis when more clinical experience with the drug accumulates (55).
Rapamycin (Sirolimus) is structurally related to tacrolimus, although its mechanism of action differs. Few data are available regarding its human toxicity, but animal studies demonstrate much less renal toxicity than either CsA or tacrolimus (56).
OKT3 is used by many units, either as induction therapy or, as with our own unit, for treatment of steroid-resistant rejection. Because OKT3 (a monoclonal antibody) is the product of a murine-derived line, use of OKT3 may result in the formation of anti-mouse antibodies. These have the potential to interfere in vivo with possible future treatment with OKT3 (57) or in vitro with other assays using murine antibodies, i.e., heterophilic antibody interference. Companies producing diagnostic kits add blocking agents to overcome this potential interference, but if antibodies are present in very high concentration, they may overwhelm the added blockers. Identification of heterophilic antibody interference in routine laboratory assays has the potential to be used as a quick, cheap screen for substantial OKT3 antibody presence (58).
disease recurrence
Disease recurrence is a major problem in the liver transplant
recipient. Hepatitis B (59), hepatitis C
(60), and malignancy (61) are especially
prone to recur, to the extent that many units have changed their
indications for accepting patients onto a liver transplant waiting list
(62). Debate continues about whether conditions such as
primary biliary cirrhosis recur (63). The changes seen
with laboratory tests will vary depending upon the original condition.
The changes seen with hepatitis and malignancy may initially be
confused with sepsis and rejection or cholestasis, and usually liver
biopsy is required to resolve the etiology of the changes.
malignancy
The incidence of malignancy of both solid organs and hematological
tissues is higher among transplant patients than in the general
population (64), and it is well recognized that organ
transplant recipients are at greatly increased risk of developing
EpsteinBarr virus-associated lymphoproliferative disorders
(65), which may regress if immunosuppression is reduced
(66). Monitoring the development of such lesions is
primarily clinical, but hematological indices will change and there may
also be alterations of LFTs.
| Costs |
|---|
|
|
|---|
In conclusion, the laboratory is an essential component of the liver transplant program. Its role is to provide high-quality, cost-effective, and time-effective results. Staff need to be aware of potential confounding factors, such as heterophile antibodies, that may be outside the experience of even experienced clinical staff. Inexperienced junior medical staff may be rotated into the Transplant Unit. Although they are responsible for the day-to-day care of patients, they are not familiar with the special problems of this area. The laboratory must ensure that clinical staff are aware of these particular concerns. In defining a potential test menu, it should be clear which tests are essential and which are merely interesting. Clinical staff are dealing with acute problems and unless results are available within a time frame that enables them to be used in "real time" to aid in clinical decisionmaking, they will not use the information. Better use can be made of some of the older routine tests, which are substantially cheaper and faster to produce.
| Acknowledgments |
|---|
| Footnotes |
|---|
-GST,
-glutathione S-transferase; IL, interleukin; TNF, tumor necrosis factor;
-IFN,
-interferon; ICAM-1, intercellular adhesion molecule-1; ß2-M, ß2-microglobulin; ALP, alkaline phosphatase; GGT,
-glutamyltransferase. | References |
|---|
|
|
|---|
-glutathione S-transferasea sensitive marker of hepatocellular damage associated with acute liver allograft rejection. Transplantation 1994;58:1345-1351.
[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
![]() |
E. R. Black Diagnostic strategies and test algorithms in liver disease Clin. Chem., August 1, 1997; 43(8): 1555 - 1560. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |