Editorial Commentary:Persistent Clinical Impact of Cytomegalovirus in Organ Transplantation

In 1979, Henry Balfour termed cytomegalovirus (CMV) the “troll of transplantation” for the way CMV unwittingly sneaks up on a successful transplant recipient and either precipitates one of its many indirect effects or directly harms the patient [1]. Although we have decreased, rather dramatically, the clinical impact of CMV on organ transplantation [2, 3], with dramatic improvement in patient and graft survival, we know that CMV infection and disease still occur, as is nicely illustrated in the article by San Juan et al. [4] in this issue of the Clinical Infectious Diseases. This is the second CMV study published in the journal by the Spanish Network for Research on Infection in Transplantation, which is a cohort of transplant recipients from many institutions in Spain [5]. The strength of the study is that data are collected from >1400 prospectively observed patients, with standardized data collection and case definitions [5]. The authors confirm well-known and long-appreciated risk factors of CMV infection and disease in kidney and kidney-pancreas transplant recipients—namely, donor and recipient serostatus (i.e., mismatch), treatment with antilymphocyte antibody therapy, treatment for rejection and chronic graft malfunction [6–8]. Because of the sample size, San Juan et al. [4] discover some additional interesting insights that deserve additional analysis and study. Namely, sirolimus use is associated with deceased risk of CMV disease, whereas donor age >60 years and cyclosporine use are associated with an increased risk of CMV disease. They suggest that donors >60 years of age may be more likely to be CMV seropositive and thus transmit donor-derived virus, but they do not provide data to support this hypothesis [4]. The findings involving sirolimus are quite interesting with respect to risk of CMV infection. There is a 73% reduction in CMV disease with sirolimus use, despite the fact that sirolimus-treated patients in this cohort had an increase in the need for treatment for rejection, which one would think would increase the risk of CMV disease. Sirolimus has also been shown to be associated with a reduced rate of polyoma virus infection in kidney transplant recipients [9]. The mechanism for the effect of sirolimus on CMV infection may be as simple as a reduced level of immunosuppression or may be specific to a sirolimus mechanism of action at the cellular level that interacts with CMV replication [10]. Interestingly, among the 6% of patients with CMV disease in this cohort, the median time of onset of CMV infection was 62 days after transplantation, which suggests that many of the patients with kidney or kidney-pancreas transplants underwent a preemptive strategy to prevent CMV infection or disease [11]. However, San Juan et al. [4] note that late-onset CMV disease did occur in ∼25% of patients; presumably, these patients did receive antiviral prophylaxis. Surprisingly, long-term outcome, measured at a mean duration of 1 year, was good in the CMV-infected group. This stands in contrast to some other studies that suggest that late-onset disease is associated with a worse outcome, despite our ability to treat CMV infection with intravenous ganciclovir or oral valganciclovir [12, 13]. When CMV organ involvement does occur, it is now most commonly clinically manifest in the gastrointestinal tract. The Spanish Network for Research on Infection in Transplantation authors confirm this finding rather dramatically by reporting that 78% of all CMV disease occurred in the gastrointestinal tract. The reason we are seeing this clinical manifestation of CMV disease in organ transplantation is still not clear. Very few cases of CMV disease in this cohort occurred outside the gut. There are some unanswered questions from this study. First, what was the mortality rate among patients in the cohort who developed CMV infection or disease? The category CMV disease–attributable mortality, as reported by San Juan et al. [4], fails to take into account the known associations of CMV infection with invasive fungal disease, graft loss, and the other indirect effects that can impact organ transplantation [14–16]. It would be of interest to know whether the population with CMV infection or disease had worse overall outcome compared with that of the total cohort, including graft loss and survival. Second, San Juan et al. [4] do not provide specific insight into the late-onset versus early-onset CMV disease groups. Were there different risk factors for occurrence? Were the late-onset cases more difficult to treat? The occurrence of late CMV disease after prophylaxis is the subject of significant investigation. Several approaches are being investigated. One approach is a longer duration of prophylaxis. Preliminary data suggest that this approach might be successful [17], and a randomized trial is under way. The second approach is to understand the immune mechanisms related to late-onset cases and, presumably, boost immunity to prevent such cases [18]. Although it has been almost 30 years since Hank Balfour coined the term “troll of transplantation,” we find that CMV still has the ability to subvert successful transplantation. Potential conflicts of interest. D.R.S.: no conflicts.