Hsu Chun-Hsien, Chang Jin-Biou, Liu I-Chao, Lau Shu Chuen, Yu Shu-Man, Hsieh Chang-Hsun, Lin Jiunn-Diann, Liang Yao-Jen, Pei Dee, Chen Yen-Lin
Department of Family Medicine, Cardinal Tien Hospital, School of Medicine, College of Medicine, Fu Jen Catholic University, Taipei, Taiwan.
Department of Pathology, National Defense Medical Center, Division of Clinical Pathology, Tri-Service General Hospital, Taipei, Taiwan.
Prev Med. 2015 Mar;72:76-82. doi: 10.1016/j.ypmed.2014.12.036. Epub 2015 Jan 7.
To compare four different blood pressure (BP) measurements-systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP), and pulse pressure (PP)-in predicting future metabolic syndrome (MetS) among the normotensive elderly population, and to estimate the optimal cutoff value of the best single measurement for clinical practice.
A total of 2782 non-medicated participants aged ≥ 60 years were enrolled in a standard health examination program in Taiwan from January 2004 to December 2013. Two thirds of the participants were randomly designated as the training group (n=1855) and the other one third as the validation group (n=927). The mean follow-up time was 3.60 years for both the training and validation groups. MAP and PP were calculated from SBP and DBP.
SBP, DBP, and MAP were associated with future MetS, whereas PP was not. MAP had the largest hazard ratio in Cox regression (men 1.342 [95% CI 1.158-1.555] and women 1.348 [95% CI 1.185-1.534] in the training group; men 1.640 [95% CI 1.317-2.041] and women 1.485 [95% CI 1.230-1.794] in the validation group) and the largest area under the receiver operating characteristic curve (men 0.598 ± 0.021 and women 0.602 ± 0.021 in the training group). Multivariable Cox regression further indicated that a higher MAP level was independently associated with the future occurrence of MetS. Participants with MAP above the cutoff value (84.0mm Hg for men, 83.3mm Hg for women) had a higher cumulative incidence of MetS than did their counterparts after four years' follow-up in both the training and validation groups. The results derived from the training data could be replicated in the validation data, indicating that the results were generalizable across distinct samples.
MAP is more accurate than SBP, DBP, and PP in predicting future MetS among the normotensive geriatric population. Calculation of MAP is recommended when dealing with normotensive patients aged ≥ 60 years in clinical practice.
比较四种不同的血压测量值——收缩压（SBP）、舒张压（DBP）、平均动脉压（MAP）和脉压（PP）——在预测正常血压老年人群未来代谢综合征（MetS）方面的效果，并估计临床实践中最佳单一测量值的最佳临界值。
2004年1月至2013年12月，共有2782名年龄≥60岁的未接受药物治疗的参与者参加了台湾的一项标准健康检查项目。三分之二的参与者被随机指定为训练组（n = 1855），另外三分之一为验证组（n = 927）。训练组和验证组的平均随访时间均为3.60年。MAP和PP由SBP和DBP计算得出。
SBP、DBP和MAP与未来的MetS相关，而PP则无关。在Cox回归中，MAP的风险比最大（训练组男性为1.342[95%CI 1.158 - 1.555]，女性为1.348[95%CI 1.185 - 1.534]；验证组男性为1.640[95%CI 1.317 - 2.041]，女性为1.485[95%CI 1.230 - 1.794]），且在受试者工作特征曲线下的面积最大（训练组男性为0.598±0.021，女性为0.602±0.021）。多变量Cox回归进一步表明，较高的MAP水平与未来MetS的发生独立相关。在训练组和验证组中，随访四年后，MAP高于临界值（男性为84.0mmHg，女性为83.3mmHg）的参与者MetS的累积发病率高于其对应人群。从训练数据得出的结果可以在验证数据中复制，表明这些结果在不同样本中具有可推广性。
在预测正常血压老年人群未来的MetS方面，MAP比SBP、DBP和PP更准确。在临床实践中，当处理年龄≥60岁的正常血压患者时，建议计算MAP。
