A Bayesian meta-analysis on early tobacco exposure and vascular health: From childhood to early adulthood

Abstract

Background: Smoking has been consistently associated with increased cardiovascular risk in adults. Although exposure to tobacco products often starts in early life, evidence for the possible adverse effects on the cardiovascular system of the young is scarce. We sought to derive pooled estimates of smoking effects on indices of early vascular damage in children and adolescents.

Design and methods: We performed a systematic review and meta-analysis of clinical studies involving young individuals up to 21 years old that provided data on smoking exposure (active or passive) and flow-mediated dilatation, carotid to femoral pulse wave velocity and maximum carotid intima-media thickness. We employed three distinct methodologies of random-effects data synthesis, including the Sidik-Jonkman estimator, the Hartung and Knapp correction and a Bayesian method with a well-informed prior on the level of between-study variance.

Results: In 12 studies and 5279 individuals in total, smoking exposure was related to deterioration in all three outcomes (mean adjusted flow-mediated dilatation decrease: -0.77%, 95% confidence interval -1.38--0.15, mean adjusted pulse wave velocity increase: 0.1 m/s, 95% confidence interval 0.02-0.17 and mean adjusted carotid intima-media thickness increase: 0.35 mm, 95% confidence interval 0.16-0.55, for the Sidik-Jonkman estimator). No difference was established between active and passive smoking on associations with arterial damage.

Conclusions: Exposure to tobacco products is associated with subclinical vascular damage early in life, even from childhood. Public health initiatives should target these very young age groups to prevent early smoking exposure and associated arterial damage and its sequelae.

Keywords: Smoking; adolescence; childhood; vascular disease.