Objective To study the association of changes in body mass index (BMI) and waist circumference (WC) with cardiovascular risk factors in late adolescence.

Methods As part of a population-based cohort (EPITeen) 1574 adolescents were evaluated when they were 13 y and 17 y. We computed OR and 95% CIs using logistic regression to study the association between changes in BMI z-scores and WC from 13 to 17 years and outcomes at 17 years. The outcomes were define as high systolic blood pressure (SBP) ≥130 mm Hg; high diastolic blood pressure (DBP) ≥85 mm Hg; low high density lipoprotein cholesterol (HDLc) if first quartile and high: triglycerides, low density lipoprotein cholesterol (LDLc), glucose and insulin if fourth quartile.

Results Comparing with adolescents who remained normal weight in both study waves, those who remained overweight/obese presented higher odds of high SBP [females: OR=3.33 (1.78–6.23); males: OR=3.17 (2.00–5.01)], DBP [females: OR=2.47 (1.00–6.10); males: OR=2.37 (1.17–4.79)], triglycerides [males: OR=3.25 (1.88–5.62)], LDLc [males: OR=3.50 (2.03–6.00)], insulin [females: OR=2.18 (1.26–3.79); males: OR=2.32 (1.34–4.03)] and low HDLc [females: OR=2.21 (1.28–3.80); males: OR=1.94 (1.11–3.41)]. The estimates for those who changed to overweight/obese were similar. Among those who changed to normal weight, we found associations in the opposite direction, although statistical significance was not achieved. Regarding changes in WC and considering the 75th percentile as cut-off, results were similar, however, partly explained by BMI.

Conclusion Remaining or changing to higher anthropometric measures in adolescence period was associated with worse cardiovascular risk profile later in adolescence.