Cytomegalovirus (CMV) has been implicated in vascular pathologies and may warrant inclusion in cardiovascular predictive algorithms. We addressed this in healthy older adults and renal transplant recipients (RTR) as they retain a high burden of CMV. RTR (n = 45) stable more than 2 years after transplantation and 58 age‐matched healthy adults were assessed. Plasma inflammatory biomarkers (soluble isoform of the interferon‐β receptor [sIFNAR2], soluble tumour necrosis factorreceptor‐1 [sTNFR1], soluble cluster of differentiation 14 [sCD14], C reactive protein, P‐selectin, intracellular cell adhesion molecule‐1, vascular cell adhesion molecule‐1), and measures of CMV burden (antibodies, saliva CMV DNA, and interferon γ responses to CMV) were assessed in 2014 and evaluated in 2017 as predictors of vascular health—defined using flow‐mediated dilatation (FMD), pulse wave velocity (PWV), and augmentation indices (Aix@ 75). Linear regression models adjusted for age, sex, and body mass index (BMI) were optimized to identify risk factors. In 2017, RTR had inferior vascular health marked by impaired FMD and PWV. Detectable CMV DNA (P = .02) was associated with impaired FMD, whilst CMV glycoprotein B (gB) antibody attenuated this effect (P = .03) (adjusted R2 = .42). In healthy adults, the optimal model for predicting FMD (R2 =.22) incorporated high P‐selectin (P = .03) and low ICAM‐1 (P = .03) levels with no significant impact of CMV. Elevated sIFNAR2 (P = .04) and gB antibody (P = .06) levels predicted increasing Aix@ 75 (poor vascular health) in healthy adults (R2 = .4), whilst optimal models for RTR (R2 = .37) linked low sIFNAR2 and CMV IE‐1 antibody levels with lower Aix@ 75 (better vascular health). CMV IE‐1 antibody was also protective in relation to PWV in healthy adults (R2 = .55). Overall, measures of active CMV replication were more predictive of impaired FMD in RTR than standard biomarkers, but increased CMV gB antibodies may be protective.