Abstract: Background　Body mass index（BMI）is the most sensitive physical measurement index for predicting hypertension in adults.At present，most studies focus on the impact of BMI measuring on a single time point or the variation of BMI during a specific period on hypertension.But there is limited research on the association between the trajectory of BMI based on age and the occurrence of hypertension.Objective　To explore the relationship between trajectory of BMI and the incidence of hypertension in adults aged 20 to 59 years.Methods　People who had participated in health checkups for at least three times at a tertiary class hospital in Qingdao during 2015 to 2019 were selected as the research objects.And physical examination results〔height，weight，systolic blood pressure（SBP），diastolic blood pressure（DBP）〕，laboratory test indicators〔fasting blood glucose（FPG），serum total cholesterol（TC），triacylglycerol（TG），low density lipoprotein cholesterol（LDL-C），high density lipoprotein cholesterol（HDL-C）〕 were collected.The age，BMI，SBP and DBP at the first physical examination and the age at the diagnosis of hypertension were recorded.Latent class linear mixed model（LCLMM）was used to identify the dynamic variations of BMI trajectories relating to age and then the subjects were grouped according to its trajectory.Then Cox regression model was used to analyze the association between BMI trajectories and the incidence of hypertension in adults aged 20 to 59 years.Hazard ratios （HR）and 95% confidence interval（CI）were reported.Results　A total of 2 683 samples were included and four distinct BMI trajectories，which were classified to the medium-slow increase group（1 681 cases，62.7%）and the medium-level-rapid increase group（639 cases，23.8%），high level-rapid increase group （272 cases，10.1%），low level-rapid increase group（91 cases，3.4%）were identified through LCLMM.The incidence of hypertension，age，BMI，SBP，DBP and FPG level，TC level，TG level，LDL-C level，HDL-C level 〔on first physical examination intermediate level-slow rise group，medium level-rapid rise group，high level-rapid rise group，low level-rapid rise group were compared，the difference was statistically significant（P<0.05）.Taking the medium-slow rise group as a reference，after adjusting for confounding factors，the risk of hypertension in medium-rapid rise group，high-level-rapid rise group and low-level-rapid rise group 〔HR（95%CI）〕were respectively 1.47（91.09，1.98），2.56（1.85，3.55），and 6.35（3.36，12.01）.Conclusion　Adults with faster BMI increments along with age have a higher risk of hypertension.In the future，adults BMI trajectories should be focused on and especially the population with faster BMI increments，so as to identify high-risk groups as soon as possible and control population obesity in time，achieving the purpose of early prevention of hypertension.

Key words: Hypertension, Body mass index, Adult, Correlation

摘要： 背景　体质指数（BMI）是预测成年人高血压最为敏感的身体测量指标，目前绝大多数研究仅关注于单一时点的BMI或BMI在一定时期的变化值对高血压发病的影响，但有关BMI随年龄的变化轨迹与高血压发病之间关系的研究较少。目的　探究20~59岁成年人BMI随年龄变化轨迹与高血压发病之间的关系。方法　选取2015—2019年在青岛市某三甲医院参与过至少3次健康体检的人群作为研究对象，收集体格检查〔身高、体质量、收缩压（SBP）、舒张压（DBP）〕、实验室检测指标〔空腹血糖（FPG）、血清总胆固醇（TC）、三酰甘油（TG）、低密度脂蛋白胆固醇（LDL-C）、高密度脂蛋白胆固醇（HDL-C）〕，记录研究对象首次体检时年龄、首次体检时BMI、首次体检时SBP、首次体检时DBP、确诊高血压时年龄。采用潜在类线性混合模型（LCLMM）分析BMI随年龄的动态变化并根据其轨迹对研究对象进行分组，随后应用Cox回归模型分析20~59岁成年人BMI变化轨迹与高血压发病的关系，并报告风险比（HR）及95%可信区间（CI）。结果　共纳入2 683例作为研究对象，通过LCLMM共识别出4条BMI随年龄变化轨迹，分别为中水平-缓慢上升组（1 681例，62.7%）、中水平-快速上升组（639例，23.8%）、高水平-快速上升组（272例，10.1%）、低水平-快速上升组（91例，3.4%）。中水平-缓慢上升组、中水平-快速上升组、高水平-快速上升组、低水平-快速上升组高血压发病率、首次体检时年龄、首次体检时BMI、首次体检时SBP、首次体检时DBP、FPG水平、TC水平、TG水平、LDL-C水平、HDL-C水平比较，差异有统计学意义（P<0.05）。以中水平-缓慢上升组为参照，调整混杂因素后，中水平-快速上升组、高水平-快速上升组和低水平-快速上升组出现高血压的风险〔HR（95%CI）〕分别为1.47（1.09，1.98）、2.56（1.85，3.55）、6.35（3.36，12.01）。结论　BMI随年龄增长上升越快的成年人罹患高血压的风险越高，今后应当关注成年人BMI的变化轨迹，尤其要关注BMI上升较快的群体，从而尽早识别高危人群，以及时控制人群肥胖并达到早期预防高血压的目的。

关键词: 高血压, 体质指数, 成年人, 相关关系