In this prospective cohort study involving 300 patients with angiographically defined CAD at baseline, homozygousity for allele 2 of the IL-1 RA and seropositivity to CMV alone and in combination were not associated with an increased risk for cardiovascular events during follow-up; in addition, combination of the CMV-seropositivity and IL-1RN*2 allele were not associated with a proinflammatory response. Therefore, these data do not support the hypothesis that IL-1RN*2 genotype in combination with CMV seropositivity might be a strong risk factor for secondary cardiovascular events in patients with already prevalent cardiovascular disease.
These results are in contrast to several reports in the literature suggesting a positive association between CMV seropositivity and progression of atherosclerosis . However, the earlier evidence was mostly based on small case-control studies with rather vague definitions of clinical endpoints and with little or no adjustment for potential confounders. Meanwhile, several large prospective seroepidemiological studies have shown no independent association for CMV and coronary heart disease [9, 10]. But clearly more carefully conducted prospective studies in different patient populations are needed before definite conclusions can be drawn [11, 12].
Evidence for a possible influence of the IL-1 RA polymorphism on CAD has been inconclusive so far . One study reported a positive association with risk of CAD , another reported an positive association for the IL1RN*2 genotype with risk of restenosis in patients with CAD, if the population was restricted to a subgroup of patients with single vessel disease . We found no association of CMV-serpositivity and IL1RN*2 allele with risk of CVE during follow-up, even if both factors were combined.
It has been described that subjects with the IL1RN*2 genotype might be more resistant against some infections . Patients with the IL1RN*2 genotype indeed showed a lower CMV-seroprevalence than others in our population (55.2% vs. 46.6%), although this difference was not statistically significant.
Recently, it has been suggested that the detrimental effects of CMV-seroprevalence may be limited to subjects with an increased inflammatory response as characterized by high levels of CRP [14, 15] or high IL-6 levels . We did not find a positive association of CMV-seroprevalence with the occurrence of CVE confined to subjects with high levels (above the median) of IL-6 or CRP, although baseline IL-6 and CRP values (the latter only tentatively) showed a positive predictive association with the occurrence of cardiovascular events during follow-up. However, this association was not modified by CMV-serostatus (data not shown). As CMV-seroprevalence may be related to various adverse factors associated with CAD itself, issues of confounding have to be considered and may explain part of the discrepancies in the literature so far. There may be indeed a complicated interplay among established risk factors and exposure to infectious agents as recently suggested by our group .
When looking at the results of this study the following limitations should be considered: seroprevalence may not be a good marker for recurrent active infection with CMV, or even more relevant in this context, reactivation of CMV; cytomegalovirus-IgG does not diagnose active virus infection after primary infection. Furthermore, the current study cannot exclude a weak association between seroprevalence to CMV and IL-1RN*2 genotype and risk of CVE. However, the study had a power of 80 % to detect an RR of at least 1.6 and more associated with CMV-seropositivity alone, and an RR of 2.8 and more associated with combined CMV-seropositivity and presence of the IL-1RN*2 allele, respectively. In addition, lack of association of the CMV-seropositivity and IL-1RN*2 allele with established inflammatory marker levels, which play a key role in atherogenesis, are further reassuring the essentially negative findings of this study.