Hierarchical modelling of blood lipids’ profile and 10-year (2002–2012) all cause mortality and incidence of cardiovascular disease: the ATTICA study

Nomikos et al. Lipids in Health and Disease (2015) 14:108 DOI 10.1186/s12944-015-0101-7 RESEARCH Open Access Hierarchical modelling of blood lipids’ profile and 10-year (2002–2012) all cause mortality and incidence of cardiovascular disease: the ATTICA study Tzortzis Nomikos1, Demosthenes Panagiotakos1,3*, Ekavi Georgousopoulou1, Vassiliki Metaxa2, Christina Chrysohoou2, Ioannis Skoumas2, Smaragdi Antonopoulou1, Dimitrios Tousoulis2, Christodoulos Stefanadis2, Christos Pitsavos2 and the ATTICA Study group Abstract Background: The traditional view on the relationship between lipid biomarkers and CVD risk has changed during the last decade. However, it is not clear whether novel lipid biomarkers are able to confer a better predictability of CVD risk, compared to traditional ones.Under this perspective, the aim of the present work was to evaluate the predictive ability of blood lipids’ profile on all cause mortality as well as 10-year incidence of CVD, in a sample of apparently healthy adults of the ATTICA epidemiological study. Methods: From May 2001 to December 2002, 1514 men and 1528 women (>18 y) without any clinical evidence of any other chronic disease, at baseline, were enrolled. In 2011–12, the 10-year follow-up was performed in 2583 participants (85 % follow-up participation rate). Incidence of fatal or non-fatal CVD was defined according to WHO-ICD-10 criteria. Baseline serum blood lipids’ profile (Total-C, HDL-, non HDL-, LDL-cholesterol, triglycerides (TG), apolipoprotein (Apo)A1 and B, and lipoprotein–(a) levels were also measured. Results: The 10-year all-cause mortality rate was 5.7 % for men and 2.0 % for women (p = 0.55). The, 10-year CVD incidence was 19.7 % in men and 11.7 % in women (p < 0.001). Multi-adjusted analysis revealed that TC, non-HDL-C, TG and TG/HDL-C ratio, were independent predictors of all cause mortality (RR per 1 mg/dL or unit (95 % CI): 1.006 (1.000–1.013), 1.006 (1.000–1.013), 1.002 (1.000–1.004), 1.038 (1.001–1.077), respectively). Moreover, TC, HDL-, LDL-, non-HDL-cholesterol, TG, apoA1, TC/HDL-C and TG/HDL-C were independently associated with CVD risk. Among all lipid indices the ratio of apoB/apoA1 demonstrated the best correct reclassification ability, followed by non-HDL-C and TC/HDL-C ratio (continuous Net Reclassification Index 26.1 and 21.2 %, respectively). Conclusion: Elevated levels of lipid biomarkers are independently associated with all-cause mortality, as well as CVD risk. The ratio of apoB/apoA1, followed by non-HDL-C, demonstrated the best correct classification ability of the developed CVD risk models. Keywords: Cardiovascular disease, All cause mortality, Lipids, Lipoproteins, Apolipoproteins, Epidemiology * Correspondence: 1 Department of Nutrition and Dietetics, School of Health Science and Education, Harokopio University, Athens, Greece 3 46 Paleon Polemiston St., Glyfada, Attica 166 74, Greece Full list of author information is available at the end of the article © 2015 Nomikos et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Nomikos et al. Lipids in Health and Disease (2015) 14:108 Background Cardiovascular disease (CVD) remains one of the major causes of death world-wide, despite the huge efforts that have been undertaken the past decades for the clarification of its pathogenesis and treatment, as well as its prevention at population and individual level [1]. The identification of the risk factors that lead to atherosclerosis and their subsequent modification by lifestyle interventions and pharmaceutical treatment is the cornerstone of the prevention policies [2]. Large cohort studies have shown that smoking and dyslipidemia are the two most important risk factors for myocardial infarction, followed by diabetes, hypertension and obesity [3]; and most scoring systems utilize age, gender, systolic blood pressure, smoking status, and total cholesterol (TC) or Low Density Lipoprotein cholesterol (LDL-C) concentration as the main variables of their algorithms [4–6]. Nowadays, the reduction of LDL cholesterol levels remains the primary target for the primary prevention of CVD [7, 8] and this is supported by a strong body of evidence showing that it is an important marker of coronary heart disease (CHD) [9]. However, it has been acknowledged that a high residual CVD risk still exists, with a considerable proportion of potential CVD “candidates” being underestimated, and therefore, the need for the identification of novel biomarkers, especially of lipid/lipoprotein metabolism, is emerged [10]. Moreover, a high residual risk characterizes people with obesity or metabolic syndrome, where LDL-C is less predictive for developing CVD [11]. These people are characterized by low levels of HDL-C, elevated levels of triglycerides and a high content of small dense pro-atherogenic apo B lipoprotein particles [12]. Small dense lipoproteins (which is estimated by the apoB or non-HDL cholesterol concentration), penetrate more easily the arterial wall and it seems that their number rather than their cholesterol content drives foam cell formation [13]. It is therefore possible that people with a high content of elevated small dense lipoprotein particles have near normal LDL-C values due to the discordance between the apoB particle number and their cholesterol content. These people will have an underestimated risk prediction score [14]. Moreover, accumulating evidence from recent epidemiological, genetic and biochemical studies changed the traditional view for the role of HDL-C which seems to serve as a strong predictor rather than a causative factor of CVDs [15, 16]. Finally, recent cohort studies have demonstrated the predictive power of Lp(a) since its elevated levels correlate to CVD risk in a continuous and independent manner [17]. It is therefore obvious that the traditional view on the relationship between lipid biomarkers and CVD risk has changed during the last decade. However, it is not clear whether novel lipid biomarkers (e.g., apoB, apoA1, Lp(a), non-HDL-C) are able to confer a better predictability of CVD risk, Page 2 of 9 compared to the more traditional ones (LDL-C, HDL-C and TGs). Based on the existing literature, studies evaluating the predictive ability of a variety of blood lipids/lipoproteins for CVD incidence are lacking. Under this perspective, the aim of the present work was to evaluate the predictive ability of blood lipids’ profile, i.e., TC, LDL-C, HDL-C non-H (...truncated)