We thank Jain et al. for their interest and correspondence regarding our recently published study. 1 The authors note that the incidence of cytomegalovirus (CMV) disease in our retrospective study was much lower than that reported by others. As acknowledged in our article, we did not use routine surveillance to detect CMV viremia and the true rate of CMV infection could have been higher than what we reported. However, it is unlikely that we did not account for most or all cases of clinically significant or symptomatic CMV infection due to the frequent follow-up schedule in our liver transplant program.
The overall incidence of CMV disease in our high to moderate risk liver transplant recipients at 12 months posttransplantation of 3% and 4% with valganciclovir and ganciclovir, respectively, is similar to that previously reported by Gane et al. 2 (4.8% with ganciclovir at 6 months). In addition, the incidence of CMV disease in our high risk group of 7% and 22% with valganciclovir and ganciclovir, respectively, was similar to that reported by others using ganciclovir (9.3-21.4%). [2][3][4] Finally, the valganciclovir pivotal study reported a 12% incidence of CMV disease at 6 months in their high-risk control group using the same dosing regimen of ganciclovir as ours. 5 Therefore, we feel the overall low rate of CMV disease in our study was similar to that reported in other large studies which used prospective surveillance methods. [2][3][4][5][6] In a previously published study by Jain et al., 7 the incidence of CMV infection was high, at 17% overall and 25.9% in high risk patients using a valganciclovir regimen of 900 mg/day or 450 mg every other day depending on renal function for 3 to 6 months posttransplantation. The authors concluded that valganciclovir prophylaxis was ineffective in liver transplant recipients. However, their study did not include a contemporary or historical control group. In addition, the clinical characteristics of patients assigned to the different dosing regimens and how the medication was dosed in patients with renal insufficiency were not provided. It is possible that some of their patients received an inade-