American
Journal of Kidney Diseases
April 2001 • Volume 37 •
Number 4 • p659 to p676
Viral infections after renal transplantation
Stephen R. Smith, MD, David W. Butterly, MD, Barbara D. Alexander, MD, Arthur Greenberg, MD
Abstract
Viral infections are a leading cause of posttransplantation
morbidity and mortality. A number of recent developments have altered our
understanding and management of these disorders. The pathogenetic roles of
several viruses, including human herpesviruses 6 and 8, have been newly
established. Molecular-based diagnostic tests now make more rapid diagnosis
possible. The licensing of new potent antiviral agents offers a wider choice of
drugs for viral prophylaxis and treatment. The use of more potent
immunosuppressive agents is responsible in part for the increasing incidence of
some viral infections, but this varies among drugs, and individual viruses
differ in their sensitivity to immunosuppressive agents. This review summarizes
the natural history, diagnosis, prevention, and treatment of many common viral
infections after renal transplantation.
Supplemental reference list appears only on the web site electronic pages
(www.ajkd.org)
VIRAL INFECTIONS ARE a major complication of the postoperative course in renal
transplant recipients.1 A number of highly specific
laboratory tests with rapid turnaround times have greatly improved the ability
of clinicians to diagnose viral infections in a timely fashion. Moreover, in
the last several years, a number of new antiviral agents have received approval
from the Food and Drug Administration, leading to changes in the strategy for
treatment and prophylaxis. The use of more potent immunosuppressive therapies
has also changed the spectrum of viral diseases. Concern about hepatitis B and
C occurs before transplantation, when decisions are made to list infected
individuals for transplantation or use organs from infected donors. Excellent
reviews of this controversial area are available elsewhere.
Human herpesviruses
Herpesviruses are DNA viruses that characteristically become latent after
primary infection and may cause disease years after initial exposure. Eight
herpesviruses are known to be pathogenic in humans. Except for human
herpesvirus 8 (HHV-8), all are widespread in industrialized societies.
Serological diagnosis is therefore not useful except in documenting prior
exposure.
HHV-1 and -2: Herpes
simplex 1 and 2
In 50% to 66% of seropositive renal allograft recipients, viral shedding can be
detected within 5 to 14 days after transplantation, but only 15% to 45% develop
symptomatic ulcers or vesicles.2 Cutaneous lesions may persist
for months, and dissemination to internal sites, including the esophagus,
colon, and bladder, as well as ocular involvement with corneal and retinal
infection, may also occur.3-5 Hepatitis from herpes simplex 1
(HSV-1) or HSV-2 is uncommon, but potentially life threatening. Most cases
represent primary infection with donor virus, although fatalities from
reactivation have been reported.6 Typically, HSV hepatitis occurs
very soon after transplantation, with onset within 4 to 20 days. In solid-organ
transplant recipients, the median time of onset was 18 days compared with a
peak time of onset for cytomegalovirus (CMV) hepatitis of 35 days. With
hepatitis, cutaneous manifestations are the exception rather than the rule. The
mortality rate for this disorder is high, up to 67%.6 HSV meningoencephalitis is most
often caused by HSV-1. The virus preferentially affects the temporal lobes, and
hallucinations or focal neurological signs may be present. Magnetic resonance
imaging may show hyperintensity on T2-weighted images in the same distribution.
A single case of interstitial nephritis with allograft failure caused by HSV-1
has been reported.
The definitive diagnosis of active HSV infection relies on culture of vesicular
fluid, mucosal swabs, cerebrospinal fluid (CSF), or urine. Polymerase chain
reaction (PCR) testing can also be used; this technique is preferred for CSF.7 A positive Tzanck smear
signifies herpesvirus infection but is not specific for HSV. Immunofluorescence
staining with specific antisera is preferred.
Treatment regimens for mucocutaneous HSV infections in renal transplant
recipients are similar to those used in immunocompetent individuals (Table 1).
|
Table 1. Antiviral Therapy |
||||
|
Indication |
Normal Renal Function |
GFR 10-50 mL/min |
GFR < 10 mL/min |
HD or PD* |
|
Herpes
simplex |
|
|
|
|
|
Initial
genital† |
ACY, 400 mg 5 × d, or VAL, 1 g 3 ×
d for 10 d |
ACY, no change; VAL, every 12-24 h |
ACY, every 12 h; VAL, 0.5 g every
24 h |
HD: additional dose post-HD |
|
Recurrent
genital† |
ACY, 400 mg 5 × d, or VAL, 1 g 2 ×
d for 5 d |
ACY, no change; VAL, 0.5 g every
24 h |
ACY, every 12 h; VAL, 0.5 g every
24 h |
HD: additional dose post-HD |
|
|
|
|
|
CAPD: every 24 h |
|
Suppression
of recurrent genital |
ACY, 400 mg 2 or 3 × d or VAL, 1
g/d |
ACY, no change; VAL, 0.5 g every
24-48 h |
ACY, 50% every 12 h; VAL, 0.5 g
every 48 h |
HD: additional dose post-HD |
|
|
|
|
|
CAPD: every 24 h |
|
Herpes
labialis† |
PCV, 1% topically every 2 h × 4 d
or ACV, 400 mg 5 × d for 5 d |
Topical, no change; ACY, no change |
Topical, no change; ACY, every 12
h |
HD: additional dose post-HD |
|
Encephalitis |
ACY, 10-15 mg/kg IV every 8 h for
14-21 d |
Every 12-24 h |
Every 24 h |
HD: additional dose post-HD |
|
|
|
|
|
CAPD: 50% every 24 h |
|
Varicella-zoster |
|
|
|
|
|
Varicella‡ |
ACV, 10 mg/kg IV every 8 h for
7-10 d |
Every 12-24 h |
Every 24 h |
HD: additional dose post-HD |
|
|
|
|
|
CAPD: every 24 h |
|
Zoster§ |
VAL, 1 g 3 × d, or ACY, 800 mg 5 ×
d for 7 d, or ACY, 10 mg/kg IV every 8 h for 7 d |
VAL 1 g every 12-24 h; ACY, 800 mg
every 8 h or IV every 12-24 h |
VAL, 0.5 g every 24 h; ACY, 800 mg
every 12 h or IV every 24 h |
HD: additional dose post-HD |
|
CMV |
|
|
|
|
|
Prophylaxis|| |
GAN, 5 mg/kg IV 2 × d (see note
for duration), then 1 g 3 × d to complete 3 mon; VAL, 2 g 4 × d for 3 mon |
GAN, 25%-50% IV every 24 h; GAN,
0.5-1 g every 24 h; VAL, 1.5 g 2-3 × d |
GAN, 25% IV every 24 h; GAN, 0.5 g
every 24 h; VAL, 1.5 g every 24 h |
HD: administer daily dose post-HD |
|
Treatment |
GAN, 5 mg/kg IV 2 × d for 14 d,
then 1 g 3 × d to complete 6 wk |
25%-50% IV every 24 h; 0.5-1 g
every 24 h |
25% IV every 24 h; 0.5 g every 24
h |
HD: administer daily dose post-HD |
|
Influenza |
|
|
|
|
|
Prophylaxis
or treatment¶ |
Amantadine, 100 mg 2 × d |
100 mg every 48 h |
200 mg every 7 d |
No supplement |
|
|
Rimantadine, 100 mg 2 × d |
100 mg 2 × d or every 24 h |
100 mg every 24 h |
No data |
|
Treatment¶ |
Zanamivir, 10 mg by inhalation 2 ×
d × 5 d |
No adjustment, limited data 75 mg
every 24 h |
No adjustment, limited data |
No adjustment, limited data |
|
|
Oseltamivir, 75 mg 2 × d |
|
No data |
No data |
|
Parvovirus |
|
|
|
|
|
Treatment |
IV IgG, 0.4 mg/kg, every 24 h ×
5-10 d |
No adjustment |
No adjustment |
No adjustment |
|
|
||||
Valacyclovir
and famciclovir have recently been approved for the treatment of HSV. Their
increased oral bioavailability is an important advance over acyclovir, but the
benefit of less frequent dosing is tempered for many patients by the
significantly greater cost.
Primary oral or genital herpes and recurrences should be treated. Suppressive
therapy is useful for frequent recurrences. For oral herpes prophylaxis,
once-daily valacyclovir is efficacious in immunocompetent adults. No data are
available in renal transplant recipients, and some caution is necessary because
of several reports of thrombotic thrombocytopenia and/or hemolytic uremic
syndrome in renal transplant recipients administered valacyclovir prophylaxis.8 Data on
famciclovir use in transplant recipients are also lacking. Penciclovir is
available for topical use; it has no apparent advantage over acyclovir. In
transplant recipients, prophylaxis with acyclovir, 200 mg, three times daily or
even once daily can prevent, delay, or lessen the severity of oral or genital
HSV reactivation.9
Given these data, we use acyclovir or valacyclovir for treatment of acute
episodes, but we prefer acyclovir for prophylaxis.
HHV-3: Varicella-zoster
virus
Before effective antiviral therapy, varicella was frequently associated with an
adverse outcome. In one study, 12% of 160 pediatric renal allograft recipients
developed varicella after renal transplantation. Of these patients, 16% had
reinfections and 42% had severe disease, defined as either persistence of fever
and new vesicle development lasting longer than 6 days or mucosal involvement.
There was one death. Complications in the severe-disease group included urinary
retention caused by bladder nerve involvement, transaminase level elevations,
and thrombocytopenia.10 Among renal transplant
recipients, primary varicella is more severe in adults than children, but
available studies are small and subject to reporting bias. Complications in
adults include disseminated intravascular coagulation, hepatitis, and secondary
bacterial or fungal sepsis. In one series, four of five adults died. However,
only one patient was administered timely acyclovir treatment.11 The outcome is more favorable
with acyclovir.12
After primary infection, varicella-zoster virus (VZV) becomes latent in spinal
dorsal root or cranial nerve sensory ganglia. Reactivation leads to herpes
zoster. In healthy individuals, only the primary dermatome is involved, and the
major morbidity is postherpetic neuralgia. Renal transplant recipients may
develop localized dermatomal herpes zoster involving one to three dermatomes
(primary and two contiguous dermatomes), disseminated cutaneous zoster that
crosses the midline or involves more than three dermatomes, or visceral zoster
with lung or liver involvement. Up to 33% of untreated cases disseminate.13
Besides VZV hepatitis, Reye’s syndrome should also be considered in the differential
diagnosis of hepatic dysfunction after varicella, particularly if
hyperbilirubinemia is absent. Varicella pneumonia presents with diffuse
pulmonary infiltrates and hypoxemia. However, aspiration and other forms of
pneumonia may also occur in these critically ill patients.12,14 Cerebellar ataxia is the most
common neurological complication, but VZV may also cause myelitis or
meningoencephalitis. Seizures occur in 25% to 50% of the patients. Anogenital
zoster can lead to transient bladder paralysis with urinary retention.15
VZV infections are usually diagnosed clinically; however, virus can be cultured
from vesicular fluid or bronchial washings, and immunohistochemical staining of
biopsy material can provide laboratory confirmation. VZV DNA can be identified
in liver biopsy specimens in cases of VZV hepatitis. CSF findings in VZV
meningoencephalitis resemble those in HSV encephalitis. PCR or the measurement
of CSF antibody directed against VZV permits rapid diagnosis.16 CSF pleocytosis may be
observed in uncomplicated zoster, presumably as a response to nerve root
inflammation.
Vesicular fluid contains infectious viral particles, and patients should be
isolated until after all cutaneous lesions have crusted. However, this will not
completely limit exposure because varicella spreads through the respiratory
route before the rash develops. The period of contagion extends from 2 to 3
days before through 5 days after onset of the rash. If administered within 72
hours of exposure, varicella-zoster immune globulin (VZIG) provides effective
postexposure prophylaxis in renal allograft recipients because it prevents or
lessens the severity of primary varicella.17 VZIG may extend the incubation
period of varicella from the normal 2 weeks up to 4 weeks. Occasional cases of
varicella have been reported in transplant recipients in whom the natural
antibody titer has declined.10 Some authorities recommend
VZIG as postexposure prophylaxis for all transplant recipients unless immune
titers are positive.14 This approach has not been
widely adopted. Titers often are unavailable at the time of exposure. We
withhold VZIG if a patient gives a reliable history of prior varicella.
Children awaiting transplantation, like all other children, should be
vaccinated. Immunization is highly efficacious in preventing varicella in renal
transplant patients. At 1 and 10 years, 62% and 42% of vaccinated children
still had protective antibody titers, respectively. Among vaccine recipients,
protective antibodies were lost in 7.4% within 1 year after transplantation and
24% after 5 years. This contrasts with loss rates of 0.4% and 4.5% at 4 years
in transplant recipients with natural immunity after varicella. Although 45% of
varicella-naive children developed varicella, no child who became and remained
seropositive after vaccination developed varicella after transplantation. The
attack rate was intermediate (25%) in vaccinees without protective antibodies.
Immunization also reduced the severity of varicella and the incidence of
zoster.18
Vaccination uses the live attenuated Oka strain. In a small trial, 16 of 17
individuals vaccinated after renal transplantation developed and retained
antibodies for at least 1 year. One patient developed mild varicella from the
vaccine strain.19 However, the vaccine is not
licensed for use in immunosuppressed patients, and vaccination after
transplantation is contraindicated. Among healthy individuals, transmission of
the vaccine strain to household contacts has been described, but VZIG is not
recommended, even for immunosuppressed siblings of vaccinees.
Because it has been shown to lessen the severity of disease, acyclovir is
indicated for renal transplant recipients with VZV infection. Controlled
studies are mainly limited to comparative trials of immunocompromised cancer
patients, but one report included a small subset of patients administered
azathioprine and prednisone after renal transplantation.20 Compared with placebo,
acyclovir markedly limited the rate of cutaneous or visceral dissemination.13,21 Uncontrolled descriptions of
marked improvement after initiation of acyclovir therapy also support
treatment.11 In a retrospective study, only
one death occurred among 66 acyclovir-treated pediatric renal transplant
recipients with varicella. Nine patients (13%) had severe disease with
pneumonitis, encephalitis, or prolonged fever and rash. Azathioprine therapy
was discontinued at the time of diagnosis, although cyclosporine and prednisone
were continued.22 In an earlier study, the
subgroup of individuals in whom azathioprine therapy was discontinued had a
milder course.10 Because an increased incidence
of varicella reactivation has been reported in pediatric renal allograft
recipients treated with mycophenolate, it also seems prudent to discontinue
this agent in patients with varicella.23
VZV is less sensitive to acyclovir than HSV; therefore, bigger doses are
required. Patients with varicella and disseminated zoster should be
administered high-dose intravenous acyclovir with hospital monitoring (Table 1
). After the fever resolves, patients without visceral involvement may safely
be switched to oral acyclovir therapy.24 Localized zoster can be
managed with oral therapy without hospitalization. Prednisone and acyclovir
have not eliminated or shortened the course of postherpetic neuralgia
consistently.25 In immunocompetent
individuals, valacyclovir modestly but significantly shortened the time to
cessation of postherpetic pain and reduced the percentage of patients with
persistent pain at 6 months. Thus, it is probably the preferred drug.8 Famciclovir may also have a
similar advantage, but experience is more limited. Gabapentin, tricyclic
antidepressants, carbamazepine, and capsaicin can be tried in patients with
established postherpetic neuralgia.
HHV-4: Epstein-Barr
virus
Epstein-Barr virus (EBV) is responsible for a number of disorders, but the
major concern in renal transplant recipients is posttransplant
lymphoproliferative disorder (PTLD).26 Acute infection with EBV leads
to polyclonal activation of B cells with expansion of lymphoid tissues. Because
the cellular immune responses provided by natural killer cells and major
histocompatibility complex (MHC)-restricted cytotoxic T cells are critical to
clearing the virus, the T-cell–targeted immunosuppression used in organ
transplantation puts allograft recipients at risk for PTLD. Specific agents
vary in their propensity to cause PTLD. Antilymphocyte globulins are by far the
principal risk factor (discussed later).
Four disease patterns for PTLD have been identified: (1) uncomplicated
infectious mononucleosis; (2) benign, polyclonal, polymorphic B-cell
hyperplasia; (3) early malignant transformation of polyclonal polymorphic
B-cell lymphoma; and (4) monoclonal polymorphic B-cell lymphoma.27,28 In the Cincinnati registry,
PTLD comprised 22% of posttransplantation tumors.29 European data showed a high
risk during the first year (224 per 100,000, or 37 times the expected
incidence). The presentation is dichotomous. Younger patients present earlier
with polyclonal disease at an average 9 months after transplantation or intensification
of immunosuppression. Older patients present later at an average 6 years, with
monoclonal extranodal disease.
Extranodal involvement is present in 70% of the patients, and the most frequent
sites are the central nervous system, allograft, and gastrointestinal tract.
Presenting features can include fever and lymphadenopathy or isolated
gastrointestinal symptoms, including anorexia, abdominal pain, diarrhea,
obstruction, or bowel perforation. The usual sites of digestive tract disease
are the distal ileum and right colon.
Allograft involvement accounts for approximately 20% of patients and is
greatest in lung transplant recipients. In approximately one third of the
patients with allograft involvement, disease is confined to that site.
Occasionally, a biopsy for presumed rejection instead shows PTLD, and it is
critically important to differentiate these two clinical entities.
Immunohistochemical analysis shows renal interstitial infiltrates related to
EBV stain for B-cell markers rather than the standard T-cell infiltrates seen
in acute rejection. Demonstration of EBV genome with in situ hybridization or
immunohistochemical stains is diagnostic.
A number of modalities have been tried to facilitate PTLD diagnosis. Neither an
increase in antibody titers nor an increase in viral shedding is useful.
EBV-positive B cells increase 100- to 1,000-fold after transplantation, 50% to
60% of transplant recipients shed EBV asymptomatically, and the rate increases
with increasing immunosuppression and treatment of rejection. The isolation of
EBV DNA in serum rather than peripheral-blood lymphocytes, which signifies a
high viral load, has recently been shown to separate transplant recipients with
PTLD from those without. However, these results have not yet been confirmed,
and their clinical utility remains to be shown.30 Because of these limitations,
the diagnosis of PTLD rests on finding immunohistochemical evidence of EBV in
tumor tissue.
Specific treatment recommendations for primary EBV with infectious
mononucleosis are lacking. Temporary reduction of immunosuppression and
avoidance of antilymphocyte globulin therapy seem appropriate. Hairy cell
leukoplakia responds to acyclovir therapy. Beyond reduction or discontinuation
of immunosuppression, there is no consensus on the treatment of PTLD. Patients
presenting early in the postoperative course, as well as patients with
polymorphic disease, are more likely to have a favorable course.28,29 Central nervous system
lymphomas initially respond to radiation, although recurrences are common.
Localized gastrointestinal malignancies can be managed with resection and
immunosuppression reduction. Antiviral therapy with ganciclovir, acyclovir, or
interferon has been used with limited utility; interferon can promote
rejection. For more advanced cases and monomorphic disease, radiation therapy
and cytotoxic chemotherapy are used. Treatment options, including a variety of
novel immunotherapeutic approaches, have been extensively reviewed.26,31 Ganciclovir or acyclovir
administered for CMV prophylaxis reduces the later occurrence of PTLD.32 However, preventive strategies
have not been widely applied, and their cost-effectiveness is uncertain.
HHV-5: CMV
CMV is named for the characteristic swelling of infected cells, which often
contain inclusions. The virus is highly associated with cells. Because there
are many genetically different CMV strains, seropositivity does not imply
immunity, and reinfection with a different strain can cause symptomatic CMV
disease.
Seropositivity rates for CMV among different populations range from 40% to 80%.
Clinically apparent CMV disease can arise from the recipient’s latent strain or
from primary or superinfection with virus that originates in a transplanted
organ or blood product from a seropositive donor. Seronegative recipients of a
kidney from a seropositive donor (D+/R–) usually seroconvert within the first 6
months after transplantation. Without prophylaxis and in the absence of
antilymphocyte antibody therapy, approximately half the D+/R– individuals
develop symptomatic CMV disease. Antilymphocyte therapy with OKT3 is associated
with a greater risk for primary symptomatic CMV infection and a 2- to 10-fold
greater risk for reactivation in seropositive recipients.33 Seronegative patients
administered a blood transfusion from a seropositive donor are at similar risk.
Symptomatic CMV infection occurs in less than 10% of seropositive recipients,
independent of donor serological status.34
In renal allograft recipients, the clinical course of CMV disease from a
primary infection does not differ from that with reactivation or
superinfection. Typically, a syndrome characterized by fever and fatigue
beginning 1 to 6 months after transplantation and accompanied by leukopenia or
thrombocytopenia occurs. Fever spikes may be dramatic and can be interspersed
with periods without fever during which the patient feels well. This syndrome
may be accompanied by signs and symptoms of tissue-invasive disease, most
commonly including hepatitis, pneumonitis, and gastrointestinal ulceration.35 Gastrointestinal involvement
may occur without other signs of CMV disease. Any part of the gastrointestinal
tract may be involved; thus, patients may present with odynophagia, substernal
or epigastric pain, diarrhea, and upper or lower gastrointestinal bleeding. The
biliary tract and appendix are rare sites of involvement. Normal-appearing
cells surrounding areas of erosion, hemorrhage, or ulceration may show CMV
cytopatholytic effects on histological examination. Therefore, random biopsy
specimens are sometimes helpful, but infected tissue usually has an abnormal
gross appearance. The incidence of CMV gastrointestinal ulceration appears to
be increased in patients administered mycophenolate.35 CMV can also cause migratory
abdominal pain in association with mesenteric lymphadenitis.
Rare consequences of CMV infection in renal transplant recipients include
adrenalitis, encephalitis, and, as a late event, chorioretinitis. CMV infection
in the transplanted kidney is also uncommon. Inclusions typical of
herpesviruses may be seen in tubular epithelial cells and glomerular cells, and
they are sometimes but not always accompanied by an inflammatory response.36 CMV has been associated with
glomerulopathy characterized by endothelial swelling with narrowing of
capillary lumens, mononuclear cell infiltration, and mild segmental
hypercellularity, but without inclusions. Some patients experienced improvement
in renal function after a reduction in immunosuppression. The causal
relationship has been questioned, and this controversy has been reviewed in
detail.37
CMV is associated with an increased subsequent risk for acute rejection.38,39 Whether the risk is conferred
by the reduction of immunosuppressive medication used to treat CMV infection or
is a direct immune-modulating effect of the virus is not known. Cytokines
produced by CMV infection increase class II MHC expression on endothelial
cells, as well as endothelial and tubular cell intercellular adhesion molecule
1 expression. CMV also encodes a glycoprotein homologous to class I MHC
antigens.
A causal link between CMV infection and chronic allograft rejection has not
been established. However, the vascular changes of chronic rejection mirror
changes in cardiac allograft coronary arteries, vanishing bile duct syndrome in
liver transplants, and bronchiolitis obliterans in lung transplant recipients
in whom the link to prior CMV infection is stronger but still unproved. CMV
infection predisposes to subsequent opportunistic infection, particularly Pneumocystis
carinii pneumonia and fungal infection. CMV and VZV may occur
simultaneously.
Two approaches to the diagnosis and management of CMV disease are available.
The conventional approach is to perform a diagnostic test when symptoms arise,
confirm the diagnosis, and treat accordingly. In the preemptive approach,
high-risk patients are screened serially in an attempt to identify individuals
destined to become symptomatic with CMV disease, and this subgroup is treated
to prevent disease altogether. The diagnosis of CMV disease is made by
detection of the virus in the peripheral blood or when characteristic findings
are noted on histological examination of infected tissue and confirmed by
immunohistochemical staining. Urinary shedding alone does not imply clinically
significant CMV disease. Likewise, bronchoalveolar lavage fluid cultures are
not sensitive or specific for CMV pneumonitis. However, viral load in
bronchoalveolar lavage fluid correlates with disease.40
Despite their drawbacks, the introduction of tests detecting CMV viremia in
blood (Table 2) has greatly refined the diagnosis of CMV disease and simplified
its clinical management.
|
Table 2. Peripheral-Blood Tests for CMV Viremia |
||||
|
|
Sensitivity |
Specificity |
Time Course of Positive Test |
Other |
|
Shell
vial culture |
Low |
Medium |
Poor correlation with symptoms |
Rapid loss of specimen viability;
long turnaround time |
|
pp 65
antigen |
Medium to high |
High |
Correlates well with symptoms, but
can be falsely negative early in the course or with leukopenia |
Rapid loss of specimen viability |
|
DNA
hybrid capture |
High |
High |
Correlates well with symptoms, but
can be falsely negative early in the course or with leukopenia |
Can be run on stored sample; may
be too sensitive, with detection of clinically insignificant viremia |
|
PCR
assays |
High |
Medium |
Positive before symptom onset;
persists after symptom resolution |
Leukocyte tests more sensitive
than plasma test; prone to specimen contamination |
PCR tests
may be too sensitive; when qualitative PCR is used as part of a prospective
surveillance program, many patients with positive test results do not go on to
have CMV disease and the test remains positive long after symptoms are gone.41 Quantitative PCR
may be more helpful in following up response to therapy.42 Currently, we
favor a direct test for CMV DNA by hybrid capture. Like the pp65 antigen test,
it tracks well with symptoms, but it may be run on stored samples.40 Quantitative
assessment of the viral load using these molecular tests is being refined and
standardized and will likely prove useful in the preemptive approach to CMV
disease.40
Several prevention strategies can be used to reduce the likelihood of CMV
disease. When blood transfusion is needed in a seronegative renal transplant
recipient, use of blood from a seronegative donor is optimal, but seronegative
blood is in short supply and better reserved for other patient populations. The
likelihood of CMV disease can be reduced with a leukocyte filter. It is
estimated that allocation of kidneys to maximize the number of donor and
recipient CMV-negative pairs would increase the complexity of the matching
algorithm with minimal effect on graft and patient survival.43 In the case of
evaluation of more than one potential living related donor for a seronegative
recipient, CMV status should be considered as one factor, equivalent in
magnitude to one HLA-DR match.44
Active immunization with the Towne strain of attenuated CMV affords protection
against severe infection in D+/R– patients. However, mild CMV disease was not
eliminated, and seropositive recipients showed no benefit. Currently, no
vaccine is approved by the Food and Drug Administration for CMV. Passive
immunization with CMV immune globulin reduced the rate of CMV disease in D+/R–
patients from 60% in the placebo group to 21% in a randomized trial.
Seroconversion rates were unaffected. When CMV immune globulin was used in
combination with oral acyclovir, 600 mg/d, for 3 months in D+/R– patients, the
rate of CMV disease was reduced to 10%.45 CMV immune
globulin is expensive and cumbersome to use; it requires seven intravenous
doses over 16 weeks. Polyimmune globulin has modest efficacy in the prevention
of CMV disease in seropositive recipients.
Antiviral therapy has also been studied. High-dose oral acyclovir was initially
reported to be useful in renal allograft recipients. However, three subsequent
studies failed to confirm its benefit.46 Ganciclovir is
much more active against CMV than acyclovir, owing in part to its much longer
intracellular half-life.47 Intravenous
ganciclovir administered coincidently with antilymphocyte therapy for induction
or rejection reduces but does not eliminate the risk for CMV disease. In a
randomized trial, CMV disease occurred within 6 months in 14% of seropositive
transplant recipients treated with this therapy versus 33% of patients administered
antilymphocyte therapy alone.48 The preemptive
approach using weekly screening followed by intravenous ganciclovir in the
event of a positive test result reduced the incidence of CMV disease only
slightly.49
Prolonged treatment with oral ganciclovir, despite less than 10%
bioavailability, is potent prophylactic therapy for CMV disease. Several
randomized trials have documented that the rate of CMV disease during treatment
with 3 to 4 months of oral ganciclovir in doses as low as 250 mg twice daily is
less than 5%. Ganciclovir has been shown to be more effective than acyclovir in
preventing CMV infection and disease. A strategy using intravenous ganciclovir
during antilymphocyte therapy followed by 3 to 4 months of oral ganciclovir
virtually eliminated CMV disease in these patients. As indicated in Table 1
, we use this strategy. An evidence-based set of clinical practice guidelines
for CMV prophylaxis after renal transplantation is available.50
Valacyclovir, a prodrug of acyclovir, was recently shown to be effective
compared with placebo. The rate of acute rejection in valacyclovir-treated
D+/R– patients was half that of the placebo group.39 Oral ganciclovir
has not been compared directly with valacyclovir. Valganciclovir, a ganciclovir
prodrug with increased oral bioavailability, is currently under development.
No reliable measures can indicate whether CMV disease will resolve
spontaneously or after reduction of immunosuppression alone. Thus, most or all
patients with laboratory-confirmed CMV syndrome should be treated. Dosage
recommendations for ganciclovir are listed in Table 1
. Other schemes are also available.51 Intravenous
ganciclovir induces remission of CMV disease in renal transplant recipients in
all but rare cases that are ganciclovir resistant.47 Ganciclovir
alone is seldom the cause of leukopenia during CMV treatment; mycophenolate or
azathioprine dosage should be reduced or discontinued when leukopenia develops.
Occasionally, severe leukopenia requires brief treatment with granulocyte
colony-stimulating factor. Fever and leukopenia typically improve or resolve
within 3 to 4 days, and quantitative markers of viremia decrease markedly
during the first 2 weeks of treatment.42 Qualitative
tests for viremia (pp65 antigen or CMV DNA by hybrid capture) may still be
positive after 2 weeks of therapy.40 Recurrence of CMV
disease after treatment with intravenous ganciclovir occurs frequently, especially
after primary infection. Therefore, oral prophylaxis is recommended for at
least 3 months thereafter.1 CMV immune
globulin is sometimes administered for severe CMV infection in addition to an
antiviral agent, although controlled studies of kidney transplant recipients
are lacking. Recurrence is only rarely caused by the development of resistance,
but ganciclovir-resistant strains of CMV are generally sensitive to foscarnet,
which is otherwise to be avoided because of nephrotoxicity.47 Combined use of
low-dose ganciclovir and low-dose foscarnet may be a useful strategy to lessen
foscarnet toxicity.
HHV-6
HHV-6 causes exanthema subitum, a febrile disease of infancy that may be
associated with rash. Immunoglobulin G (IgG) seroprevalence for HHV-6 in adults
in industrialized societies approaches 90%. The cell in which the latent virus
resides has not been established.
Our understanding of HHV-6 as a cause of disease in immunocompromised patients
is still evolving. HHV-6 has been isolated from circulating monocytes of renal
transplant recipients. Based on serological criteria or virus isolation,
reactivation of HHV-6 is observed in 20% to 55% of renal allograft recipients.
However, a clear relationship to a clinical disorder is lacking.52 Sixty percent of renal
transplant biopsy specimens contain HHV-6 antigen in distal tubular epithelial
cells, but the prevalence is the same whether the biopsy diagnosis is rejection
or cyclosporine nephrotoxicity.53
Several forms of clinical involvement have been noted. In bone marrow
transplant (BMT) recipients, HHV-6 DNA has been localized to lung tissue of
inpatients with pneumonitis, but its causative role is not definite. The virus
probably can cause encephalitis. HHV-6 has the potential to suppress bone
marrow. Leukopenia is most common, but thrombocytopenia and even aplastic
anemia also may occur.54 HHV-6 was cultured from the
blood of six patients who developed leukopenia and fever 17 to 90 days after
liver transplantation. CMV cultures were negative and patients responded to
ganciclovir or foscarnet.55 Similar findings have been
reported after renal transplantation, in which HHV-6 infection was associated
with fever, leukopenia, thrombocytopenia, and elevated serum alanine
transaminase and 
-glutamyl
transpeptidase levels.56 This syndrome tends to occur
earlier than CMV disease. Simultaneous isolation of HHV-6 and CMV was also
observed.56 Interstitial nephritis with
allograft dysfunction was attributed to HHV-6 in a patient with fever,
leukopenia, newly detected HHV-6 IgM, and negative CMV antigen and cultures who
had a clinical response to ganciclovir.57
HHV-6 should be suspected in patients with typical CMV syndrome and negative CMV
viral studies. A shell vial early antigen assay is highly specific and
sensitive. Cultures are also specific, but the issue of asymptomatic shedding
arises. Because of its ubiquity, attribution of clinical significance to HHV-6
genome recovered from tissue specimens by PCR is problematic. Immunostaining
methods for tissue specimens that detect structural proteins present only when
the virus is replicating are more specific.54
HHV-6 is sensitive to ganciclovir. Thus, patients presenting with seronegative CMV
or combined CMV and HHV-6 infection who undergo therapy for CMV also receive
appropriate coverage for HHV-6.
HHV-7
HHV-7 may cause a syndrome similar to exanthema subitum, with rash,
splenomegaly, leukopenia, neutropenia, and relative lymphocytosis. Infectious
mononucleosis syndrome with lymphadenopathy and seizures may also occur.
Primary infection occurs somewhat later than with HHV-6.58
The significance of reactivation of HHV-7 in renal transplant recipients is
only beginning to emerge. In one study, CMV was detected by PCR in 58% of renal
transplant recipients, HHV-7 in 46%, and HHV-6 in 23%, with frequent overlap.
Detection of HHV-7 replication was associated with an increase in rejection
episodes and CMV disease. These findings await confirmation.59
HHV-8
HHV-8 is the etiologic agent for Kaposi’s sarcoma (KS). This virus was first
identified when a search for foreign DNA sequences in the genome of KS tissue
isolated from patients with human immunodeficiency virus (HIV) disclosed
sequences homologous to but distinct from known herpesviruses. The DNA
sequences were subsequently found in patients with KS not associated with HIV
and up to 96% of KS tissue obtained from organ transplant recipients.
Seroprevalence for HHV-8 varies greatly with the population examined, ranging
from nil in US blood donors to 50% in individuals from sub-Saharan Africa,
where KS is endemic, and 100% in patients with HIV and KS. End-stage renal
disease per se is not a risk factor for HHV-8 exposure because the
seropositivity rate is similar in hemodialysis patients and blood donors.60 The incidence of KS in renal
transplant recipients varies in parallel to the prevalence in the donor and
recipient populations. In two high-prevalence areas, Saudi Arabia and Italy,
the KS incidence rates among renal transplant recipients were 5.3% and 1.6%,
respectively. KS accounted for 87.5% of secondary malignancies in Saudi Arabia
in contrast to the 3.7% rate in the Cincinnati registry.61,62 Either the donor or recipient
can serve as the source of HHV-8.63,64 The use of antilymphocyte
globulin may contribute to subsequent development of KS.65
Both limited and invasive variants of KS occur in renal allograft recipients.
The classic limited form has an indolent course characterized by the
development of violaceous nodules predominantly localized to the skin of the
lower extremities. Nodules may ulcerate or infiltrate the lymphatic system and
cause lower-extremity edema. The aggressive form involves the viscera,
including oropharynx, lung, and gastrointestinal tract, and may present with
pulmonary symptoms or gastrointestinal hemorrhage or perforation. Nine Saudi
Arabian patients (64%) and five Italian patients (38%) had limited KS. The
remainder had multiple sites of cutaneous involvement or visceral involvement.
In the Cincinnati Transplant Tumor Registry, the average age of presentation
was 43 years and mean time of onset was 21 months (range, 1 to 225 months).
Nearly half the cases (46%) occurred within 1 year after transplantation; most
patients were renal transplant recipients. The male-female ratio was 3:1 in
contrast to the 17:1 ratio observed in classic KS.66
Detection of anti–HHV-8 antibodies provides evidence for viral latency.
However, diagnosis of KS is based on a biopsy or excisional specimen showing
the typical histological findings. The tumor originates in the dermis and is
characterized histologically by malignant spindle cells with vascular
proliferation.
Initial treatment consists of reduction or cessation of immunosuppression.
Among the 27 patients reported in the Saudi Arabian and Italian studies, 59%
had complete remissions, 15% had partial remissions, and 26% died. A variety of
treatments, including local irradiation, intralesional chemotherapy, and
systemic chemotherapy, were used in patients who did not respond to reduction
of immunosuppression alone.61,62 Interferon has been used in
the nontransplantation setting, but its propensity to promote rejection is
problematic. Patients administered ganciclovir as treatment for other viruses
had a lower incidence of KS, but it is doubtful that long-term prophylaxis is a
suitable approach.
Respiratory viruses
A number of respiratory viruses, including influenza, parainfluenza,
respiratory syncytial virus (RSV), and adenoviruses, can affect renal
transplant recipients. These viruses share a common presentation with upper
respiratory symptoms and cough, but more serious complications can occur.67 Typically, clinicians are
presented with a patient with a viral syndrome. If recovery is slow or
pneumonia develops, identification of the virus responsible is necessary to
determine whether specific therapy is available and limit evaluation for other
causes.
Although definitive, virus cultures require several days, as well as
specialized laboratories. More recently, rapid identification has become
possible using direct enzyme-linked immunosorbent assay or immunofluorescence
techniques that identify viral antigens in respiratory secretions. Thus, the
diagnosis will most often be made by analysis of nasopharyngeal, bronchial, or
bronchoalveolar fluid washing specimens. Nasal and throat swab specimens
provide inferior results. At many hospitals, a rapid-turnaround respiratory
virus panel can provide a quick diagnosis from these specimens.
Patients suspected of having one of these disorders should be placed in
respiratory isolation.
Influenza
Influenza A and B are RNA viruses that cause respiratory infections from
December to March. Major potential complications include the development of
primary influenza or secondary bacterial pneumonia.
Although influenza causes substantial morbidity and mortality in the elderly
and patients with significant comorbidities, data on outcome in immunocompromised
patients are lacking because there have been few longitudinal studies. Several
large series of infections involving more than 500 transplant recipients
identified only rare cases of these viruses as transplant pathogens.67 One study spanning two
influenza seasons identified 12 renal transplant recipients with influenza A.
Complications included one case each of pneumonitis and bronchitis, but most
individuals had mild and self-limited disease.68 Similarly, three renal
transplant recipients with influenza B recovered within 5 days without
treatment or complications.69
Annual vaccination remains the primary method of preventing and controlling
influenza. Like the elderly, renal transplant recipients have lower response
rates and develop lower antibody titers after vaccination than young healthy
controls. However, the vaccine still lessens the complications and severity of
infection.70 Therefore, all renal
transplant recipients should be vaccinated annually, as should all persons in
close contact, including household members and health care providers.
Chemoprophylaxis with amantadine or rimantadine is effective in other high-risk
groups and can also be used. However, it has never been studied in the
transplant population.
In established disease, amantadine and rimantadine and the new neuraminidase
inhibitors, zanamivir and oseltamivir, shorten symptom duration in healthy
adults by approximately 24 hours if started within 48 hours of symptom onset.
The possible benefits of amantadine or rimantadine in transplant recipients are
unproven because no trials have been conducted in this group. Nevertheless,
most authorities support their use in the treatment of complicated influenza A
infection.71 It remains to be seen whether
the new agents will be of value.
Parainfluenza
Prior infection with these ubiquitous RNA viruses does not confer immunity,
although the severity of infection diminishes with age and exposure.
Bronchiolitis or croup are more likely to occur in children, but adults have
upper respiratory tract disease exclusively. Parainfluenza viruses are
responsible for significant morbidity and mortality caused by pneumonia in BMT
recipients, but disease is limited to pharyngitis and cough in adult renal
transplant recipients.72 Infection is associated with
an increase in the frequency of acute rejection episodes, but graft survival at
6 months is not affected. Prolonged shedding of parainfluenza viruses in the
immunocompromised host has been documented.
The effectiveness of specific antiviral agents for parainfluenza virus
infection has not been established. Clinical experience with inhaled ribavirin
in the transplant population is limited to case reports and retrospective
reviews, none of renal transplant recipients.
Adenoviruses
These double-stranded DNA viruses have been recovered from virtually every
organ system of humans and associated with many clinical syndromes, including
enteric infections, urinary tract infections, upper and lower respiratory tract
disease, and conjunctivitis.
Evidence of adenovirus as a cause of serious disease in solid-organ transplant
recipients is accumulating. A comprehensive review of more than 300 cases of
adenovirus infection in the immunocompromised population showed that 11% of all
transplant recipients become infected with adenoviruses. The infection
frequently involves the organ transplanted, and infections are often more
severe and of longer duration than in healthy hosts.73 Type of immunosuppression,
patient age, and infecting virus serotype influence fatality rates. In renal
transplant recipients, adenovirus infections are associated with a 17% fatality
rate. The majority of infections in renal transplant recipients involve the
urinary (hemorrhagic cystitis) or respiratory tract (pneumonia).
Diagnosis of acute infections can be accomplished by direct antigen detection,
as well as culture. In addition, viral inclusions may be observed in tissue
specimens. There are no well-defined treatment options for adenovirus
infections. Case reports suggest the possible efficacy of ribavirin,
vidarabine, immunoglobulin, and ganciclovir.74 None of these has proven
efficacious during a clinical trial.
RSV
RSV is a single-stranded RNA virus that causes yearly outbreaks from November
through April. Epithelial necrosis and airway inflammation may be sufficient to
impede airflow, particularly during expiration, resulting in characteristic
wheezing and air trapping. Patients with cellular immune deficiencies tend to
have worse infection and prolonged viral shedding.
In BMT recipients, RSV is a significant pathogen with 45% to 100% mortality.67 Data for renal transplant
recipients are sparse. In one report of two renal transplant recipients with
RSV infection, only one patient developed pneumonia, and both recovered. In
another report, one of four patients required mechanical ventilation, and all
recovered. None of these patients was administered ribavirin.
Results of retrospective reviews suggest that treatment with aerosolized
ribavirin decreases the mortality of RSV pneumonia in BMT recipients if they
are identified and treated early. However, controlled studies are lacking.
Considering the mild nature of RSV infections in renal transplant recipients
and the expense and potential side effects of ribavirin, routine use of this
drug cannot be recommended. Immunoglobulin also has been used to treat RSV, but
its value is equally uncertain.67
Other viral
infections
Polyomaviruses
Polyomaviruses are small DNA viruses. Oral or respiratory transmission occurs
early in life; 60% to 80% of adults are seropositive. However, polyomavirus
causes significant symptomatic disease only in immunocompromised patients.75,76
BK
polyomavirus
Although the BK subtype of polyomavirus was first isolated from the urine of a
renal transplant recipient in 1971, recognized infection was rare until
recently. The use of more potent combination regimens, including mycophenolate
and tacrolimus, is probably responsible. In a recent series, the incidence of
clinically significant polyomavirus infection was 2.5%, with most occurrences
in the first year posttransplantation. The virus is trophic for the urinary
tract, including the epithelium of the bladder, ureter, and renal tubules.
Clinical manifestations include hemorrhagic cystitis, ureteral strictures, and
interstitial nephritis. Suspicion of polyomavirus infection should arise when a
renal transplant recipient presents with gross hematuria, urinary tract
obstruction, or a slowly increasing serum creatinine level.75,76
Diagnosis can be made by histological examination, immunoperoxidase staining of
biopsy tissue, detection of viral inclusions containing decoy cells on urine
cytology, or urine electron microscopy. Renal biopsy findings include viral
cytopathic changes in tubular cells and a dense, variable, inflammatory
infiltrate frequently containing numerous plasma cells. Affected areas are
often sharply demarcated from surrounding tissue.75,76 Because such processes as
acute rejection and CMV infection can also cause reactive tubular cell changes,
the diagnosis requires confirmation by one of the methods listed. Electron
microscopy of negatively stained urine provides a simple, rapid, and relatively
inexpensive method to follow up the disease.76 Asymptomatic shedding of
polyomavirus in the urine has been reported. However, in a recent series, no
virus was detected in the urine of 23 patients with renal dysfunction for
reasons other than polyoma infection.76 Recently, the combination of
screening by urine cytology with diagnostic confirmation by plasma PCR has been
proposed as an alternative to biopsy.77
No pharmacological agents currently available have activity against
polyomavirus, and no known prophylactic strategies offer demonstrated efficacy.
Because the histological picture is similar to that in acute rejection and
polyomavirus interstitial nephritis can coexist with acute rejection, some
patients have undergone treatment for acute rejection along with intensification
of their maintenance immunosuppression. These individuals have done poorly,
with graft loss in the short term in more than half. In two recent series,
reduction in maintenance immunosuppression resulted in stabilization or
improvement in graft function in the ma