The term immunosenescence is used to describe the decreased function of the immune system with age, and al-so to describe the phenotypic alterations in immune cells and cytokines that develop with age. The most dramatic pheno-typic change is seen in the T cell compartment, where the percentage of cells with an effector memory phenotype increas-es, and the number and diversity of naïve cells decrease. In particular, large, often oligoclonal accumulations of CD28-CD8+ T cells develop. This hallmark change is largely attributable to CMV infection; the accumulating cells are the enormous “inflationary” virus-specific T cell population that are responding to this lifelong, smouldering, subclinical in-fection. In many populations, at least 80% of the elderly carry CMV, so CMV-driven changes in old age were easily as-cribed to aging per se. There is broad agreement that CMV drives this characteristic phenotype of the aged immune sys-tem. There is also considerable evidence that the size of the CD8+CD28- population can be used to describe an “immune risk phenotype” which correlates with an increased inflammatory milieu (“inflammaging”), which may be a predictor of all cause mortality. However, the evidence that CMV contributes causally to the functional failings of the immune system in old age, rather than innocently providing a convenient biomarker, is much less convincing. This important question needs to be addressed with studies including enough CMV seronegative individuals to provide statistically valid data.