Sara van den Berg

188 Chapter 7 index tended to be higher in recent CMV seroconverters (i.e. converted after DCS round 5), although the effect was not significant (p=0.11, Figure 5F ; n=4 participants recently seroconverted). When restricting this analysis to women aged 60-65 years, because 3 out of 4 recent CMV seroconverters were women in this age category, we observed that women who seroconverted recently had a higher increase in frailty index than their CMV- (p=0.006) ( Figure 5G ). Although the sample size of ST CMV+ individuals was very small, these unique human data suggest a relationship between becoming infected with CMV at older age and frailty. Together, these results suggest that there is no significant association between CMV- seropositivity, duration of CMV infection or CMV-specific antibody levels and frailty, but age of seroconversion might be associated with frailty, and recent CMV seroconversion with an increase in frailty, in older individuals. CMV- CMV+ 9.2% (n = 11) 12.7% (n = 21) 3.4% (n = 4) 9.1% (n = 15) 1.7% (n = 2) 1.8% (n = 3) Total 14.3% (n = 17) 23.6% (n = 39) Serostatus is serostatus at endpoint. Cardiovascular intervention: participant undergone either bypass surgery, Cardiac valve dilation surgery, Cardiac catheterization, Pacemaker placement, or peripheral vascular surgery. Table 2: Cardiovascular disease prevalence CMV-specific antibody levels are increased in CMV+ individuals with cardiovascular disease A more specific health outcome that has been associated with CMV infection is cardiovascular disease [45]. Of all participants in this cohort, 20.9% (n=56) had any form of cardiovascular disease ( Table 2 ), most of them men (n=38, χ 2=8.9, p=0.003). The percentage of individuals with any form of cardiovascular disease tended to be higher among CMV+ compared to CMV- individuals (23.6% versus 14.3%, Table 2) ( χ 2  =5.3, p=0.02, analysis stratified by sex as confounder). Furthermore, the occurrence of cardiovascular disease was positively associated with CMV-specific antibody levels at endpoint (p=0.04 Figure 5H ). Although in the group with CVD the median increase in CMV-specific antibodies seemed higher, no significant difference was observed (p=0.12 Figure 5I ). Within the ST CMV+ group, 3 out of 19 individuals had any form of cardiovascular disease, a number too low to investigate a possible association between cardiovascular disease and CMV-duration. In conclusion, CMV-specific antibody levels in CMV+ individuals might be related to the risk of cardiovascular disease. DISCUSSION We investigated how duration of CMV infection was related to CMV-specific antibody levels, T-cell numbers, CMV-specific T-cell responses, frailty and cardiovascular disease prevalence. We demonstrated that within individuals, CMV-specific antibody levels increased over time, albeit only slightly. Nevertheless, duration of CMV infection was not the major determinant