International Centre for Circulatory Health, Imperial College London, London, UK.
J Hypertens. 2011 Oct;29(10):2004-13. doi: 10.1097/HJH.0b013e32834a8a42.
Resistant hypertension is a well recognized clinical entity, which has been inadequately researched to date.
A multivariable Cox model was developed to identify baseline predictors of developing resistant hypertension among 3666 previously untreated Anglo-Scandinavian Cardiac Outcome Trial (ASCOT) patients and construct a risk score to identify those at high risk. Secondary analyses included evaluations among all 19 257 randomized patients.
One-third (1258) of previously untreated, and one-half (9333) of all randomized patients (incidence rates 75.2 and 129.7 per 1000 person-years, respectively) developed resistant hypertension during a median follow-up of 5.3 and 4.8 years, respectively. Increasing strata of baseline SBP (151-160, 161-170, 171-180, and >180 mmHg) were associated with increased risk of developing resistant hypertension [hazard ratio 1.24 (95% confidence interval, CI 0.81-1.88), 1.50 (1.03-2.20), 2.15 (1.47-3.16), and 4.43 (3.04-6.45), respectively]. Diabetes, left ventricular hypertrophy, male sex, and raised BMI, fasting glucose, and alcohol intake were other significant determinants of resistant hypertension. Randomization to amlodipine ± perindopril vs. atenolol ± thiazide [0.57 (0.50-0.60)], previous use of aspirin [0.78 (0.62-0.98)], and randomization to atorvastatin vs. placebo [0.87 (0.76-1.00)] significantly reduced the risk of resistant hypertension. Secondary analysis results were similar. The risk score developed allows accurate risk allocation (Harrell's C-statistic 0.71), with excellent calibration (Hosmer-Lemeshow χ statistics, P = 0.99). A 12-fold (8.4-17.4) increased risk among those in the highest vs. lowest risk deciles was apparent.
Baseline SBP and choice of subsequent antihypertensive therapy were the two most important determinants of resistant hypertension in the ASCOT population. Individuals at high risk of developing resistant hypertension can be easily identified using an integer-based risk score.
耐药性高血压是一种公认的临床病症，但迄今为止研究不足。
我们建立了多变量 Cox 模型，以确定 3666 例先前未经治疗的 Anglo-Scandinavian 心脏结局试验（ASCOT）患者中出现耐药性高血压的基线预测因素，并构建风险评分以识别高危患者。次要分析包括对所有 19257 例随机患者的评估。
先前未经治疗的患者中有三分之一（1258 例）和所有随机患者中有一半（9333 例）在中位随访 5.3 年和 4.8 年后分别发生耐药性高血压（发生率分别为每 1000 人年 75.2 和 129.7 例）。基线收缩压（151-160、161-170、171-180 和 >180mmHg）较高的分层与发生耐药性高血压的风险增加相关[风险比 1.24（95%置信区间，0.81-1.88）、1.50（1.03-2.20）、2.15（1.47-3.16）和 4.43（3.04-6.45）]。糖尿病、左心室肥厚、男性和升高的 BMI、空腹血糖和饮酒量也是耐药性高血压的其他重要决定因素。与氨氯地平+培哚普利相比，随机分配至阿替洛尔+噻嗪类药物[0.57（0.50-0.60）]、以前使用过阿司匹林[0.78（0.62-0.98）]以及随机分配至阿托伐他汀与安慰剂[0.87（0.76-1.00）]显著降低了耐药性高血压的风险。次要分析结果相似。所开发的风险评分能够准确地进行风险分配（哈雷尔 C 统计量 0.71），并且具有良好的校准度（Hosmer-Lemeshow χ 统计量，P=0.99）。在风险最高和最低的十分位数之间，风险增加了 12 倍（8.4-17.4）。
在 ASCOT 人群中，基线收缩压和随后选择的降压治疗是耐药性高血压的两个最重要决定因素。使用基于整数的风险评分可以轻松识别发生耐药性高血压风险较高的个体。
