Prognostic interactions between cardiovascular risk factors.

BACKGROUND Cardiovascular disease (CVD) still remains the leading cause of death worldwide, especially in Europe where the prevalence of hypertension is 60% higher compared with the United States and Canada and the clustering of hypertension and the metabolic disorders central adiposity, dyslipidemia and dysglycemia, known as the metabolic syndrome (MetS), affects 25% of the population. Despite the great initiatives of many primary prevention strategies, risk factor control is still poor. In an attempt to optimize risk factor control, two issues among others have been of great debate in the past decade: (1) the superiority of systolic blood pressure (SBP) as a risk factor in the elderly; and (2) the clinical relevance of MetS. However, in order to further elucidate these issues, we need to get a deeper understanding of how the cardiovascular risk factors interact with one another. Thus, prognostic interactions were used in the present PhD thesis to test the following hypotheses: Primary hypotheses: (1) The superiority of SBP over diastolic blood pressure (DBP) as a risk factor occurs at an earlier age if an individual presents with other cardiovascular risk factors. (2) The prevalence and prognostic significance of MetS differ according to age and gender. The first hypothesis is explored in paper 1 (for the endpoint fatal and nonfatal (total) stroke) and paper II (for mortality from coronary heart disease (CHD), stroke, and all-causes), while the second hypothesis is explored in paper III (for total CHD, total stroke, and CVD mortality). METHODS Using 34-42 cohorts from the MORGAM Project with baseline between 1982-1997, approximately 68 000-86 000 apparently healthy men and women aged 19-78 years, without CVD (papers I-III) and not receiving antihypertensive treatment (papers I-II) were included. During 12-13 years of follow-up, the incident events of total stroke were up to 1957, of total CHD were 4368, and of all-cause mortality were 7903. In papers I-II, event risk was analyzed by multivariate-adjusted Cox regressions including SBP and DBP simultaneously, as well as other cardiovascular risk factors and any significant interactions between variables. In paper III, MetS prevalence and prognostic significance was considered according to modified definitions of the International Diabetes Federation (IDF) and the revised National Cholesterol Education Program - Adult Treatment Panel (NCEP-ATP III), and the influence of possible interactions between age and gender on MetS prevalence and prognostic significance was explored using logistic as well as multivariate-adjusted Cox regressions. MetS was analyzed separately for men and women in various age-groups. RESULTS Taking into account the significant interactions between cardiovascular risk factors, the results were as follows: Papers I-II: Age-related shifts were shown for the independent relative importance of SBP and DBP as risk factors for stroke (both total and fatal) and all-cause mortality, but not for CHD mortality where SBP remained significant in all ages. The prognostic shift to the superiority of SBP was significantly established in the 6th decade, and only for stroke mortality was this shift influenced by other cardiovascular risk factors, such that it occurred at an earlier age in men from high-risk countries and with a higher cholesterol level. However, from mid-age and onwards, a potential harmful effect of low DBP for the risk of total stroke and all-cause mortality was present. Paper III: The prevalence and prognostic significance of MetS showed great variations among countries and were influenced by both age and gender. With older age, the prevalence of MetS increased 5-fold in women from ages 19-39 years to 60-78 years and 2-fold in men. The CVD risk associated with MetS was (1) higher in women than in men especially when using the NCEP-ATP III criteria, and (2) independently of age in men whereas in women total CHD risk decreased significantly and the total stroke risk tended to increase (although not significant) with older age. CONCLUSION The present thesis elucidates through prognostic interactions the complex interplay between cardiovascular risk factors. Our results indicate the independent prognostic superiority of SBP in elderly Europeans, and only for stroke mortality risk this prognostic superiority of SBP was influenced by other cardiovascular risk factors such that it was established at an earlier age. The prevalence and prognostic significance of MetS differed according to both age and gender. In women, MetS was associated with higher relative event risks and the MetS associated relative CHD risk decreased with advancing age.