Author + information
- S0735109716004812-e8dfd59453c89b8534480955776bc30eAltan Onat, MD∗ (, )
- S0735109716004812-6c0346b102b9985a12eb88145521832fBarış Şimşek, MD and
- S0735109716004812-097b3df7e5f447271d8e4ab51ab4a4d8Tuğba Akbaş, MD
- ↵∗Department of Cardiology, Cerrahpaşa Medical Faculty, Istanbul University, Nisbetiye cad. 59/24, 34335 Etiler, Istanbul, Turkey
The Dutch PREVEND (Prevention of Renal and Vascular End-Stage Disease) cohort study (1) reported both the risk factors for atrial fibrillation (AF) and the association of AF with cardiovascular events, heart failure, and all-cause mortality. At a mean 9.7 years of follow-up of more than 8,000 middle-aged men and women, 265 cases of AF were confirmed. A greater focus was given to associations with outcome than to determinants of AF, perhaps because no unexpected factors were perceived by the investigators to emerge. However, one of the main reported findings was related to the use of antihypertensive drugs. Age- and sex-adjusted Cox regression models indicated that antihypertensive drug use was a prominent and highly significant determining variable of incident AF, with a more than 2-fold hazard ratio (HR). At 32%, antihypertensive drug use led the population-attributable risk estimates of reversible or treatable factors (i.e., beyond age and male sex), and this percentage was greater than the aggregate of the next 3 factors (previous myocardial infarction, previous stroke, and obesity). Nonetheless, authors noted that body mass index was strongly associated with AF, thereby overlooking the potential and seemingly paradoxical role of antihypertensive drug use.
This phenomenon may well not be paradoxical. On the basis of numerous longitudinal observations in the Turkish Adult Risk Factor study (2), we proposed that an enhanced proinflammatory state and (novel) autoimmune activation were common mechanisms underlying chronic diseases. Such a state may also be induced by long-term use of certain drugs, especially statins. We documented epidemiological evidence of increased development of coronary heart disease in users of statin drugs in a primary prevention setting, independent of traditional risk factors (3), thus extending the findings obtained in meta-analyses that statins may increase the risk of type 2 diabetes. Nearly 3,000 participants free of coronary heart disease at baseline were analyzed with Cox models at 7.9 years of follow-up. Excess risk was attributed to a likely modifying effect of statins on lipoprotein(a). Turkish adults are recognized to be prone to enhanced low-grade inflammation, a feature they share with the PREVEND cohort.
The PREVEND study, consisting of a nondiabetic cohort and aiming to monitor and prevent the development of diabetes, predominantly comprises subjects with microalbuminuria. As such, people susceptible to impaired glucose tolerance are likely overrepresented. It had previously been reported in that cohort that apolipoprotein E levels, anticipated to be cardioprotective, predicted cardiovascular disease and mortality independent of apolipoprotein E genotype (4). The investigators reasonably expressed the opinion that elevated apolipoprotein E concentrations reflected either an adverse lipoprotein profile or a proinflammatory response.
A further, seemingly “paradoxical” determinant of incident AF has been reported in several studies, the largest of which was the Women’s Health Study (5). This concerns the inverse prediction of incident AF by low-density lipoprotein (LDL) cholesterol (HR: 0.72) so that a 1-SD increment in LDL cholesterol has a 28% rate of protection against the risk of AF. The driver of this process was the number of small LDL and small very LDL particles. In line with these findings are our as yet unpublished observations that may be summarized as being highly consistent with an autoimmune process in which “reduced” assays of apolipoprotein B secondary to a proinflammatory state emerged as a basic mechanism in the development of AF.
With this letter, we urge the medical community to undertake prospective epidemiological research, with the aim of confirming (or negating) whether antihypertensive medication may be a predictor of AF or other adverse outcomes in cohorts formed by appropriate participants, that is, in persons prone to metabolic syndrome or impaired glucose tolerance.
Please note: The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Vermond R.A.,
- Geelhoed B.H.,
- Verweij N.,
- et al.
- Onat A.,
- Aydın M.,
- Köroğlu B.,
- et al.
- Mora S.,
- Akinkuolie A.O.,
- Sandhu R.K.,
- et al.