TO THE EDITOR—We read with great interest the recent article by Freeman et al describing a link between cytomegalovirus (CMV) infection, CD8+ T-cell expansion, and inflammation in treated human immunodeficiency virus (HIV) infection [1]. A large longitudinal study establishing CMV infection as a risk factor for severe, non-AIDS adverse clinical events in HIV infection highlights the relevance of this research [2]. Although multiple factors acting across a broad pathophysiological landscape bring age-related morbidities early to prominence in HIV infection, the potential influence of CMV crystallizes around inflammation and immune senescence [1, 3]. Despite highly prevalent CMV coinfection throughout most HIV-infected cohorts, overt effects on inflammation and immune senescence concentrate within subsets heretofore not distinguished by either immunological or virological aspects of CMV infection. There could be significant benefit to identifying those at greatest risk for adverse outcomes and intervening against immunological or virological factors at play. To this end, several studies detected

CMV in saliva or semen of a substantial fraction of HIV-infected men having sex with men [4, 5]. Thus, there is potential for CMV replication to directly drive immune activation in this setting, a prospect supported by reduction in T-cell activation in HIV-infected individuals following valganciclovir treatment [6]. From an immunological perspective, some middle aged and even younger HIV-infected individuals have T-cell responses against CMV inflated to a degree that has been associated with immunological deterioration and increased risk for all cause mortality in old elderly individuals [7]. Because CMV memory inflation has an established relationship with immune senescence in the aging general population, its development may also relate to immunological deterioration in the antiretroviral treated HIV-infected population, a group often characterized as undergoing accelerated aging. Rapid inflation of CMV-specific T-cell responses in HIV infection may relate to the higher likelihood of detectable CMV in semen or saliva and, therefore, increased CMV replication overall, but there is no conclusive evidence for this relationship [8]. Given the propensity for CMV infection to induce exceptionally strong specific T-cell responses,