Case Studies : Transplantation

Journal Logo


Case Studies

Rubin, Robert

Author Information
doi: 10.1097/
  • Free

Case 1

A 46-year-old woman underwent orthotopic liver transplantation from a 57-year-old cadaveric donor (cause of death was an intracranial hemorrhage) for end-stage liver disease caused by biliary cirrhosis. The biliary anastomosis was accomplished with a choledochojejunostomy. The transplant operation was technically difficult, with the requirement of 29 units of blood (all blood used was from cytomegalovirus seronegative donors). Because of continued bleeding, we reexplored the patient the second day after transplant to drain a hematoma and to assure future hemostasis. After this procedure, the patient became quite stable.

Serologic studies revealed the donor to be Epstein-Barr virus (EBV)-, cytomegalovirus (CMV)-, varicella-zoster virus (VZV)-seropositive, and hepatitis B and C, and human immunodeficiency virus (HIV-seronegative. Similar testing on the recipient revealed EBV and VZV seropositivity and CMV, hepatitis B and C, and HIV seronegativity. The patient was discharged 14 days after transplant on an immunosuppressive regimen that included tacrolimus, mycophenolate, and prednisone after an induction course of antilymphocyte globulin. Antimicrobial prophylaxis included trimethoprim- sulfamethoxazole, acyclovir, and fluconazole.

The patient remained well until 7 weeks after transplant, when she presented with fever (without rigors), malaise, anorexia, and night sweats. Physical examination revealed a swollen, tender liver. Pertinent laboratory data included the following: hematocrit 34, white blood cells 2.8/mm3, platelet count of 94,000/mm3, SGOT 357 mg/dL, alanine transferase 390 mg/dL, and bilirubin 3.7/4.2 mg/dL. A liver biopsy revealed an inflammatory cellular infiltrate and the presence of active CMV replication by immunofluorescence. The CMV pp65 antigenemia assay was strongly positive with 1,166 positive cells; a similar positive result was observed in the CMV polymerase chain reaction (PCR) assay.

Intravenous ganciclovir at a dose of 5 mg/kg twice daily was initiated and continued for 7 days, and then switched to oral valganciclovir 900 mg/day for an additional 2 weeks (a total of 3 weeks of therapy). At this time the patient felt well, her liver function tests (LFTs) had normalized, and the CMV antigen test, although still positive, had fallen to a level of 257 positive cells. Over the next month, the diagnosis of probable rejection was made on liver biopsy (no CMV-positive cells were noted), and the patient received three pulse doses of intravenous methyprednisolone, with striking improvement in the LFT abnormalities. Three weeks later, the patient presented again with fever and malaise, and evaluation revealed an increase in her LFTs, a further increase in the antigenemia assay (from 257 to 734 positive cells), and an SGOT value of 502 mg/dL. A diagnosis of relapsing CMV was made, and intravenous ganciclovir was restarted at a dose of 5 mg/kg twice daily. After 6 days of this therapy, no improvement in symptoms had been noted; the antigenemia assay was now at 950 positive cells. A possible diagnosis of ganciclovir-resistant infection was made, and therapy was instituted with intravenous foscarnet, and the immunosuppressive program was decreased by >50%. During a 14-day period, the antigenemia assay improved significantly, with the foscarnet course continued until <100 positive cells were present, at which time the foscarnet dose was decreased and prophylactic doses of valganciclovir were administered and continued for 3 months. The patient continues to do well >3 years after transplant.


This patient was at high risk for clinically important CMV disease. Primary CMV infection, derived from the allograft, was highly likely because the liver donor was seropositive (thus connoting latent infection capable of being reactivated) and the recipient was seronegative (donor +, recipient −, D+R−). Just on this basis, the risk of clinical disease was approximately 50%. The administration of ATG and the need for relatively high levels of immunosuppression increased the risk of CMV disease further. The optimal approach to this patient would have been ∼4 months of oral valgancyclovir as prophylaxis. Prevention is to be preferred over treatment.

The diagnosis of CMV hepatitis and systemic CMV disease were clear cut. An extremely high viral load was documented. The exact duration of therapy needed was a problem in this patient. Although some centers use fixed durations of therapy (for example, 2–3 weeks), we advocate a more flexible approach based upon an assessment of the viral burden, level of immunosuppression, need for acute rejection therapy, the presence or absence of other immunomodulating processes such as hepatitis C, and the response to initial therapy. A classical question: “How long to treat?” is best answered: “Long enough!” We think in terms of the therapeutic prescription, which has two components: an immunosuppressive regimen to prevent and treat rejection, and an antimicrobial program to make it safe.

In this case, an extremely high viral load was documented, and a relatively short course of therapy was prescribed. The occurrence of ganciclovir resistance at first presentation is highly unusual. Factors leading to the development of resistance include the following: high viral loads, too low a dose of ganciclovir for too short a period of time, acute antirejection activity, and the failure to clear the virus from the blood at the time of cessation of antiviral therapy. In this patient, 1 week of therapy did not result in a measurable response. Although proof of resistance developing is not present, switching to foscarnet is reasonable because the need to control this infection as quickly as possible was becoming evident.

Case 2

A 60-year-old man had received a kidney transplant 10 years previously for the treatment of end-stage kidney disease because of polycystic kidney disease. The donor was an accident victim. Donor and recipient were CMV seropositive at the time of transplant, but there was no evidence of clinical disease from CMV up to the present. The allograft had functioned well up to this point (mean serum creatinine, 1.6 mg/dL). Approximately 1 month ago, the patient had the onset of a flu-like illness with fever, malaise, muscle aches, headache, and a nonproductive cough. Nasal swabs were tested by immunofluorescence for a battery of respiratory viruses, and the diagnosis of influenza A2 was made. Supportive therapy (not including the neuraminidase inhibitors) was administered and, after 8 days of illness the patient appeared to be improving. On the ninth day of illness, the patient's course changed acutely: a rigor, followed by a temperature spike to 103°F occurred, with a change in sputum (from minimal sputum to a grossly purulent sputum). Blood cultures and a sputum sample all grew Streptococcus pneumoniae, and therapy with vancomycin and ampicillin was initiated, with a prompt response to intravenous and then oral therapy. The patient received a total of 10 days of therapy.

Three weeks after completion of the antibacterial therapy, the patient developed malaise, moderate temperature elevations (Tmax=101°F), and easy fatigability. Computed tomography scans failed to reveal any lymph node pathology, and flow cytometry failed to show evidence of a monoclonal abnormality. CMV viremia was demonstrated by antigenemia and PCR assay. The patient was treated with 5 days of intravenous ganciclovir followed by oral valganciclovir at a dose of 900 mg/day (full treatment doses). The patient became afebrile by day 5 of therapy, and cleared the viremia shortly thereafter. An additional 3 months of oral therapy was administered at “prophylactic” doses, and the patient stayed well with no further CMV activity.


In the era when CMV diagnosis was difficult (requiring tissue biopsy for pathology or traditional tissue culture to demonstrate viremia) and no effective antiviral therapy was available, “late” clinical disease was essentially unknown; that is, the direct clinical effects of CMV occurred 1–4 months after transplant, or not at all. Today, there are at least two forms of late CMV (clinical disease beginning >6 months after transplant):

  1. In patients receiving a partially effective prophylactic regimen, the incubation period is prolonged, but typical CMV disease is noted more than 6 months after transplant. This form of disease is most common in patients at risk for primary disease (D+R–), those whose immunosuppressive program was particularly intensive, or whose prophylactic regimen was curtailed.
  2. In this patient, a different mechanism was operative. The essential first step in the pathogenesis of CMV disease is reactivation of latent virus. The level of viral replication is then greatly amplified by high-dose steroids and the use of the calcineurin antagonists, cyclosporine and tacrolimus. A major factor in the reactivation process is the presence of tumor necrosis factor and, perhaps, other proinflammatory cytokines. Essentially any condition in which tumor necrosis factor is released will result in the reactivation of CMV, with the possibility of subsequent CMV disease (depending on the level of immunosuppression present). Among the conditions associated with these events are pneumococcal and Gram-negative sepsis, acute rejection, significant hepatitis, and other inflammatory conditions. Our term for this is “cytokine storm” followed by a “second wave” due to CMV.

Case 3

A 56-year-old man received an orthotopic cardiac transplant in treatment for end-stage cardiac disease because of an ischemic cardiomyopathy. Both donor and recipient were CMV seropositive. Standard immunosuppression was administered with tacrolimus, mycophenolate, and prednisone. Two episodes of acute rejection were diagnosed and treated, first with steroids and then with OKT3. The patient received antimicrobial prophylaxis with trimethoprim-sulfamethoxazole and 3 months of oral ganciclovir. During the first year posttransplant occasional urine and blood tests revealed low levels of replicating CMV.

A 1-year posttransplant detailed assessment was conducted: coronary angiography was unremarkable, but intravascular ultrasound was quite abnormal, leading to the diagnosis of cardiac allograft vasculopathy. The patient was treated with a prolonged course of valganciclovir, an antilipid program and exercise training. Unfortunately, progressive cardiac dysfunction occurred, and the patient died 21 months after transplant.


This is a typical example of an indirect effect of CMV replication: the occurrence of a progressive allograft vasculopathy. Although the patient had no clear evidence of direct manifestations of CMV replication, the CMV effects played a major role in the patient's demise. Optimal management of this entity is unclear, but is the subject of intense study. It is of great interest that CMV-associated vasculopathy is also well recognized in other forms of solid organ transplantation.

© 2007 Lippincott Williams & Wilkins, Inc.