作者简介：
Michael Bursztyn: 以色列哈达萨大学医院高血压部，曾组织过以色列高血压学会会议，第34届欧洲肾病协会高血压学术委员会成员，美国心脏协会（AHA）高血压研究委员成员，哈达萨大学医院决策委员会成员。
很多研究表明动态血压监测（ABPM）显示的血压降低幅度要低很多，但是ABPM 也显示有很多患者的血压已经降至正常或者比诊室测量的血压要低很多。如ONTARGET研究中ABPM的平均水平约为124 mm Hg，而HYVET研究中诊室血压为174/91 mm Hg，醒时ABPM的平均收缩压为134 mm Hg。约相差40 mm Hg。如果大多数患者的动态血压相似，其中肯定有很多已经达到正常水平。这些研究表明，比较极端一点想的话，（ABPM显示）血压“正常”的老老年患者可能是有益的。
如果这些研究中很多患者的血压是正常的，那么其实很多时候我们并没有“治疗高血压”。对于这一结论，我并不能确定。但我想这些研究的结果并不应该受到质疑，而是探究造成这种推论的原因。如HOPE研究纳入1万例血管疾病或其他的高危的患者，在其他治疗基础上加用雷米普利10 mg或安慰剂。治疗组主要终点的相对风险降低22%。这是为类似患者推荐10 mg雷米普利的根据。两组的诊室血压无显著差别，所以研究者总结说研究结果与血压无关。但是，ABPM亚组表明两组间的血压存在很大差别，约为9 mm Hg，这打破了获益与血压无关的结论。
但ACCOMPLISH研究中诊室血压与ABPM不存在差异。所以对这些研究进行解读的时候应该谨慎。HOPE试验不是一项高血压研究，由于其假想与血压无关，研究者甚至没有公布测量血压的方法。他们建议并不需要测量血压，只要应用能够带来获益的药物即可。这在一定程度是一个自我应验预言——他们没有认真测量血压，所以他们得到的结果是与血压无关。ACCOMPLISH是一个血压研究，所以，研究这认真地对血压进行测量，且诊室血压与ABPM一致。另一种可能性是入选者的性质不同。
International Circulation: You showed in your slides that the blood pressure decreases presented by many studies would be a lot less if you looked at the ABPM information. Is that right?
Prof. Bursztyn: Yes. But ABPM also shows that many of the subjects in the studies may have been either normotensive or had much lower blood pressure than that was estimated from the clinic blood pressure measurements. You have studies like ONTARGET， where the ABPM average was something like 124 mmHg systolic， or studies like HYVET， where the clinic blood pressure was 174/91 mmHg and the mean awake ABPM was 134 mmHg systolic. There was about 40 mmHg difference. If the majority of participants had similar ambulatory blood pressures， many of them must have been normotensive. Given the results of the study， that suggests， if you take it to an extreme， that it may be useful to reduce the blood pressure of normotensive subjects in the very elderly.
International Circulation: But if a lot of these patients were possibly normotensive， then it is possible that， in some of these trials， they were not treating high blood pressure.
Prof. Bursztyn: This is a legitimate conclusion. I do not know if it is true， though. I think the results of the studies should not necessarily be questioned. The inferences behind what is causing these results are to be questioned. Take， for instance， the HOPE study: This study enrolled almost 10，000 patients with vascular disease or some other high risk. On top of other treatments， the subjects received ramapril 10 mg or placebo. The treatment arm had a relative risk reduction of 22% in the primary endpoint. This was the basis for the recommendation of 10 mg ramapril in similar patients. There was no significant difference in the clinic blood pressure between the two arms. So the investigators concluded that the results were independent of blood pressure. However， the small ABPM sub-study suggested there is a big difference in blood pressure， notably at 9 mmHg. This undermines the conclusion that the results are independent of blood pressure. In the ACCOMPLISH study， however， there was no difference between the clinical blood pressure and the ABPM. In some studies there is no difference. In other studies there is marked difference. I am saying that you have to be cautious with the interpretation of these studies and possibly for other studies. The HOPE study was not a blood pressure study. They did not even publish the method of blood pressure measurement in their study. Essentially， their hypothesis was unrelated to blood pressure. They suggested that we should not measure blood pressure， just take beneficial medication. This is， to some extent， a self-fulfilling prophecy – they did not measure blood pressure carefully and so they got results that were independent of blood pressure. ACCOMPLISH was a blood pressure study. Accordingly， they made an effort to measure blood pressure carefully and there was concordance between the clinic measurements and the ABPM. Another possibility is that the nature of the populations was different， but the populations seem to be quite similar. I have to assume that it has to do with the proficiency and care with which blood pressure was measured during the trial.