Telomere length and cardiovascular disease precursors: a 7-year follow-up from childhood to early adolescence European Journal of Preventive Cardiology

Abstract

Leucocyte telomere attrition is considered a marker for cellular age-ing caused by cumulative exposure to oxidative stress, inflammation, and endocrine dysfunction. Apart from a reflection of biological age-ing (and thus telomeres being an outcome), telomere attrition is potentially involved in the onset of cardiovascular disease as cellular senescence reduces proliferative potential of cardiovascular systems limiting the regenerative capacity of aged and injured myocardium and vasculature. 1 In a meta-analysis of prospective studies, short telo-mere length or attrition was a predictor for future coronary heart disease. 2 A recent critical mechanistic interpretation tried to distinguish telomeres as atherosclerosis cause vs. atherosclerosis consequence. 3 Even though cardiovascular diseases are more common in older age, the origins can be tracked back to childhood. Indeed, a recent study showed that telomere dynamics during early life might be important in cardiovascular disease development as short telomere length in atherosclerosis is largely determined before the clinical manifestations. 4 On the contrary, telomere length was cross-sectionally associated with blood pressure and vascular elasticity in midlife adults but not in 11-12 years old children. 5 As the identification of cardiovascular disease predictors from childhood onwards is important, we used a childhood/adolescence cohort. The first objective was to investigate the cross-sectional association of children's telomere length with cardiovascular disease precursors, i.e. body mass index (BMI), blood pressure, blood lipid levels, insulin/glucose and retinal microvasculature. Secondly, the longitudinal relation between telomere length and these cardiovascular disease precursors was tested over three measurement waves. Herein, (bi)directionality could be tested via cross-lagged modelling. Belgian children were followed-up for 7 years in spring 2008, 2010, and 2015. From 242 participants at follow-up, 181 had complete data for biological measures (9-15 years, 4.4% overweight), as part of different study projects approved by the Ethics Committee following the Declaration of Helsinki guidelines. 6 The average relative telomere length was measured by a quantitative real-time polymerase chain reaction protocol 7 and calculated using qBase software expressed as the ratio of telomere copy number to single-copy gene number (T/S) and normalized to the average T/S ratio of the entire sample set. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured using an electronic sphygmomanometer (Welch Allyn 4200B-E2) in triplicate. Retinal photographs were taken from both eyes in 2015 only (Canon 45 6.3-megapixel nonmydriatic ret-inal camera; retinal vessel measurement system IVAN). Arterioles and venules in an area 0.5-1 disc diameter from the optic disc margin were summarized in the central retinal arteriolar and venular equivalent (CRAE and CRVE) and arteriolar-to-venular ratio. Fasting blood samples were assessed on glucose, insulin, leptin, high density lipo-protein (HDL), total cholesterol, and triglyceride. The homeostasis model assessment estimating insulin resistance (HOMA-IR) was calculated. Cross-lagged models were performed in Mplus (version 5.1) using maximum-likelihood estimation with robust standard errors. The cross-lagged models allow to consider interdependencies within repeatedly measured data, while testing bidirectionality. Each cross-lagged model included correlations within waves (e.g. cross-sectional: telomere Time 1 to cardiovascular precursor Time 2), cross-lagged paths (e.g. longitudinal: from telomere Time 1 to cardiovascular precursor Time 2) while also adjusting for autoregressive paths (e.g. telomere Time 1 to telomere Time 2) and the confounders age, sex and parental socioeconomic status (see Figure 1). Additional adjustment for BMI did not change the P-value level. For descriptive purposes, telomere attrition d-score was calculated to avoid regression to the mean. 8