Changes in body mass index, waist circumference and percent body fat rather than their baseline values were significantly associated with incident hypertension after being adjusted for baseline blood pressure in men

Objectives To compare the effects of baseline anthropometric parameters and their changes on incident hypertension. Design A retrospective 5-year follow-up study. Setting Annual general health screenings at a medical check-up center. Participants A general health screening population including 1,221 men and 815 women who were not with hypertension at baseline. Main outcome measures Hazard ratios (HRs) of incident hypertension for baseline body mass index (BMI), waist circumference (WC) and percent body fat (PBF) as well as changes in them after being adjusted for baseline blood pressure. Results The cumulative incidence of hypertension was 23.7% in men and 13.5% in women. The baseline BMI, WC and PBF were not significantly associated with incident hypertension adjusted for baseline blood pressure and other confounders, while the changes in BMI, WC and PBF during follow-up were significantly associated with incident hypertension in men. All the associations with incident hypertension were not significant except for those of each 1 kg/m2 change in BMI during follow-up and for the highest quartile of baseline BMI and PBF compared with the lowest quartile in women. Baseline BMI, WC and PBF were not significantly correlated with changes in blood pressure except for inverse correlations of BMI and PBF with blood pressure changes in men while changes in BMI, WC and PBF were significantly positively correlated with changes in blood pressure. Conclusions Changes in BMI, WC and PBF during follow-up rather than their baseline values were significantly associated with incident hypertension after being adjusted for baseline blood pressure in men.