LAUSR

Objective: Despite accumulated evidence of strong relationship between obesity and hypertension, risk for hypertension according to the change of body weight (BW) is not clearly identified. Therefore, this study was to evaluate the incidental risk of hypertension according to the change of BW. Design and method: 26,483 normotensive Korean men had been followed up from 2005 to 2010. Based on baseline BW in 2005, the changes of BW [(BW at censoring time – BW at baseline)/follow-up period (person-years)] were categorized into 5 groups according to their change levels from the lowest to the highest quintile (1st – 5thquintile). On the base of 3rd quintile, 1st and 2nd quintile had negative changes of BW, and 4th and 5th quintile had positive changes of BW. Cox proportional hazard models and spline-smoothing method were used to evaluate the effect of BW change on the incidental risk of hypertension. Results: During follow-up, 4,445 (16.8 %) cases of hypertension newly developed (quintile 1: 19.0%, quintile 2: 12.9%, quintile 3: 12.4%, quintile 4: 14.8%, quintile 5: 24.8%). When quintile 3 was set as a reference in adjusted model, the hazard ratios (HRs) for incidental hypertension exhibited a J-shaped relationship with the BW changes (quintile 1: 1.66 [95% confidence interval (CI): 1.41–1.95], quintile 2: 0.96 [95% CI: 0.80–1.14], quintile 3: 1.00 [reference], quintile 4: 1.30 [95% CI: 1.10–1.54], and quintile 5: 3.39 [95% CI: 2.91–3.96], respectively). Conclusions: The incidental risk of hypertension increased in weight loss as well as weight gain, which demonstrated J-shaped relationship. This finding warrants further studies to investigate the incidental relationship between BW changes and hypertension.