Primary Cytomegalovirus Disease after Five Years of Antiviral Prophylaxis

The optimal duration of anti-cytomegalovirus (CMV) prophylaxis remains a major conundrum in transplantation. The practice of administering antiviral prophylaxis for 100 days has shifted the epidemiology of CMV (1, 2), so that it occurs at a delayed period, often during months 4–6 posttransplantation. To prevent this complication, a longer duration of prophylaxis has been proposed (1–4). Nonetheless, the optimal duration of antiviral prophylaxis is not defined. Herein, we report the occurrence of primary CMV disease in a lung transplant patient despite receiving more than five years of antiviral prophylaxis. The patient is a 37-year-old woman who received a single lung transplant in 1996 for the treatment of lymphangioleiomyomatosis. She was CMV-seronegative and the donor was CMV-seropositive. She received oral ganciclovir prophylaxis (1 g three times daily); for unclear reasons, she remained on prophylaxis until 5.5 years later. She did not develop acute rejection on an immunosuppressive regimen consisting of mycophenolate mofetil, cyclosporine, and prednisone. In June 2002, oral ganciclovir prophylaxis was discontinued; at that time, she remained CMV-seronegative. Six weeks after discontinuing prophylaxis, she developed fever and constitutional symptoms. CMV DNA load was 63,400 copies/ml of whole blood. She received intravenous ganciclovir for 4 weeks followed by valganciclovir for 8 weeks. CMV DNA load declined to undetectable level by week 4 of treatment and CMV IgM and IgG were positive. There was no clinical relapse or recurrent CMV DNAemia, as measured by CMV PCR every 2–4 weeks for 3 months. This case is reminiscent of late-onset CMV disease that we currently observe upon discontinuation of standard 100-day antiviral prophylaxis, wherein the majority of CMV disease cases occur during the first 3 months after stopping prophylaxis. This case report therefore challenges the proposed prolongation of antiviral prophylaxis as key to complete prevention of CMV disease. Indeed, a duration of 180 days of prophylaxis is suggested for lung recipients (3, 4). As illustrated by this report, prolongation of antiviral prophylaxis, even for more than 5 years, may not completely prevent CMV disease in some at-risk patients. Although CMV replication is triggered by many factors such as allograft rejection, none of these were observed in our patient. It is certainly possible that the occurrence of CMV disease in our patient is a harbinger of an overimmunosuppressed state. Additionally, we hypothesize that antiviral prophylaxis may have prevented immunosensitization and thus the lack of CMV-specific immune effectors during and soon after ganciclovir use (5). This hypothesis is supported by (1) the lack of CMV-seroconversion during prophylaxis and (2) the onset of CMV reactivation as soon as prophylaxis was discontinued. Hence, in addition to asking what is the optimal duration of prophylaxis, the more important question should be asked: why do some patients fail to develop effective CMV-specific immunity while on antiviral prophylaxis? This report should stimulate investigation into the pathogenesis of posttransplant CMV disease, including the interaction between the host and CMV. We believe that a better understanding of CMV disease pathogenesis could eventually be translated into clinical tools that can be used in our ongoing efforts of CMV prevention in transplantation. Supha Kijpittayarit Paul Deziel Albert J. Eid Raymund R. Razonable Division of Infectious Diseases Department of Internal Medicine and William J. von Liebig Transplant Center Mayo Clinic College of Medicine Rochester, MN