<International Circulation>: The International Carotid Stenting Study (ICSS) is a multicenter randomized controlled study of carotid artery stenting (CAS) compared with carotid endarterectomy (CEA) in patients with symptomatic carotid stenosis. It has received a lot of attention since first published in the Lancet in 2010. Interim analysis of the study showed that CEA is still the main treatment for patients suitable for surgery. As the chief investigator of ICSS， could you talk about the influence of these results on clinical practice?

　　Prof. Brown: Yes I can. I think it was worth pointing out that these are only interim analyses and we have recently completed a more complete analysis of the data long-term follow up which was presented at the Stroke conference but has not yet been published. I’ve drafted the paper and it will be coming out. In the interim it has led to caution among neurologists recommending stenting as an alternative to endarterectomy and if a patientis suitable for endarterectomy.It has also led people to think carefully about who should receive stenting and who should receive endarterectomy at least in Britain and neurology based centers throughout the world share this caution. It may have had less interest in North American Cardiologists who still find stenting a good alternative to surgery.

　　《国际循环》：国际颈动脉支架研究（ICSS）是一项多中心随机对照研究，于症状性颈动脉狭窄患者中对颈动脉内膜切除术（CEA）与颈动脉支架置入术（CAS）实施了对比分析。自其结果首次于2010年在《柳叶刀》发表以来，该研究备受关注。其中期分析结果显示，CEA仍是手术患者的主要治疗方法。作为ICSS研究的首席研究员，您认为这些结果会给临床实践带来怎样的影响？

　　Brown教授：我认为这些仅是中期分析的结果。我们最近已经完成并在卒中大会上发布了对其长期随访数据的更完整的分析，但相关结果尚未公开发表。我正在撰写相关文章，不久就将写完。在此期间，中期分析结果会使神经学家在对无动脉内膜切除术禁忌证的患者作出用支架置入术替代动脉内膜切除术的推荐时更为谨慎，使人们更仔细的思考哪些患者应该应用支架、哪些患者应该选择动脉内膜切除术。至少在英国以及其他以神经病学为基础的中心中，会出现上述谨慎的态度。但是，对北美的心脏病学家而言则影响不大，他们仍然认为支架置入术是动脉内膜切除术的良好选择。

　　<International Circulation>: I came across CREST (North American) study. Can you discuss the differences a bit?

　　Prof. Brown: That’s right and in fact， it’s quite interesting that CREST has been interpreted as showing different results but in fact it doesn’t. It shows the same results in terms of symptomatic patients in terms of our primary analysis. It depends on what you find to be the most important analysis. We presented in the interim paper 120-day outcomeevent， which was a combination of stroke， myocardial infarction (MI) and death.

　　In our analysis， we defined MI as a clinically diagnosed event and there were very few of those and therefore of little interest. We found the main influence to be stroke. The same was the case in CREST. There was an excess of stroke found in stenting compared with endarterectomy. Actually our results were extremely similar when you look at stroke and death rates. However， the difference is that the CREST trials had many more MI but they used a different definition of MI.

　　They screened for them using enzyme measures 48 hours after treatment. Many patients had a rise in cardiac enzymes and a change in ECG results， both of which were required for their definition， which differs from our trial.

　　I think there were two reasons for this. One was the screening. We said that we are interested only in the clinically relevant diagnosis of MI. The other probable reason is the use of selection criteria for patients. Because a lot of their investigators are cardiologists， they were more likely to include patients with ischemic heart disease than we were. That’s evident when you look through the baseline data. In their analysis， they equated MI with stroke in that they had similar numbers， which led to stroke rates with CASin their study being canceled out by MI with CEA. So they stated that the two treatments were equivalent.

　　However， most neurologists will say that a stroke is a more important outcome event than MI. I think it’s worth looking into the raw data if you would like to know more about it.

　　Subsequently， we’ve looked at the impact of all events in ICSS on disability. In other words， we looked to see if MI had an impact on daily living and to see if stenting/stroke events affected daily living.  And I do have to say they do cancel each other out. ICSS defines very similar measures of disability in the long-term follow-up paper presented at the European Stroke conference.We found that the modified distribution in rank and scores (distribution from 0 to 6: 0 being no symptoms at all， 6 being death) in the two arms of our trial (stenting vs endarterectomy) was identical.

　　So although there was an excess of stroke， it didn’t translate into long-term disability after stenting. One thing to note is that one design feature of CREST is that their main they combined patients who never had carotid artery symptoms with those who did. In ICSS， we only included patients who only had recent symptoms. We know that the risks of treatment are higher for patients with recent symptoms. Therefore， ICSS applies only to symptomatic patients.

　　《国际循环》：在北美进行的CREST研究与您的研究有哪些不同之处？

　　Brown教授：非常有趣的是，CREST研究被认为取得了不同的结果，但事实上并非如此。在有症状的患者中，其结果与我们的最初分析结果是一样的。这取决于分析的关注重点。在我们的中期报告中，我们关注的是120天结局事件即卒中、心肌梗死及死亡的复合终点。在我们的分析中，我们将心肌梗死定义为临床诊断事件，其较少发生因而不太受关注。我们发现受影响最大的是卒中，这与CREST研究的结果（与动脉内膜切除术相比，支架置入术组卒中发生率更高）相一致。其实，就卒中发生率及死亡率而言，我们的结果是非常相似的。所不同的是，CREST试验心肌梗死发生率更高些，这可能与其所应用的心肌梗死定义与我们不同有关。他们是通过测定治疗48小时候的酶含量，根据心肌酶增高伴心电图改变来定义心肌梗死的。很多患者都会有上述两种情况。这与我们的试验是有所不同的。我认为，这样做的原因有两种。其一是就筛查而言，我们只对临床诊断的心肌梗死有兴趣；其二是所用的患者入选标准不同。因为CREST研究中很多研究者都是心脏病学家，所以与我们相比他们可能入选了更多的缺血性心脏病患者。通过基线数据，我们可以很容易地证实这一点。同时在CREST研究中，心肌梗死与卒中患者的数量相当，因此他们研究中行CEA时的心肌梗死发生率超过了行CAS时的卒中发生率，故其认为两种治疗是等效的。然而，大多数神经病学家会认为，与心肌梗死相比，卒中是更为重要的结局事件。通过寻找和查看原始数据，我们发现事实确实如此。随后，我们观察了ICSS研究中所有事件结局对患者日常生活的影响。换句话说，我们探讨了心肌梗死及支架植入/卒中是否会影响患者的日常生活。在欧洲卒中大会上我们公布了长期随访的相关结果，即两者对日常生活的影响是相当的，试验中支架置入术及内膜切除术两组患者的MRS评分相同。因此，虽然卒中发生率增多了，但并不影响支架置入术后患者长期的日常生活能力。需要强调的是，CREST研究的设计特点是其同时入选了伴有及不伴有颈动脉症状的患者，而在ICSS研究中我们仅入选了有近期症状的患者。伴有近期症状的患者，存在更好的治疗风险，故ICSS的结论仅适用于有症状的患者。