围手术期使用阿司匹林或可乐定均不能预防心肌梗死

在美国心脏病学会2014年年会上，哈密尔顿(安大略省)总医院David Braley心脏、血管与卒中研究所人群健康研究中心的P. J. Devereaux博士及其同事报告了POISE-2 (围手术期缺血评估2)临床试验的结果：对于接受非心脏手术并且存在大血管并发症风险的患者，围手术期使用阿司匹林或可乐定均不能预防死亡或心肌梗死(MI)。

事实上，这两种预防性策略不仅没有保护作用，而且还会产生有害效应：阿司匹林会增加患者出现大出血的风险，可乐定则会增加患者出现具有重要临床意义的低血压、心动过缓或非致命性MI的风险。POISE-2试验的结果于2014年3月31日同步在线发表在《新英格兰医学杂志》(New England Journal of Medicine)上[doi:10/1056. NEJMoa1401105] 和 [doi:10/1056. NEJMoa1401106])。

MI是与非心脏手术相关的最常见的大血管并发症，过去认为围手术期使用阿司匹林可以通过抑制血栓形成来预防MI。目前，存在血管并发症风险的患者中有1/3都接受了围手术期阿司匹林治疗，即便这种预防性策略的风险和效益并不确定。

同样地，小规模试验表明对于接受非心脏手术的风险患者，抗高血压药物可乐定(α2肾上腺素能激动剂)能降低心肌缺血的风险并且不会诱发血流动力学不稳定，从而有可能预防MI和死亡。此外，可乐定还具有额外的镇痛、抗焦虑、抗寒战、抗炎等可能有益的效应。

POISE-2试验旨在确定这两种方法在预防术后30天内MI或死亡的复合终点方面是否比安慰剂更有效。

该试验总共纳入了来自23个国家135家医院的10,010例患者，按是否已经开始每日预防性使用阿司匹林对受试者进行分层。在双盲状态下将受试者随机分配至接受围手术期阿司匹林(4,998例)或安慰剂(5,012 例)，以及围手术期可乐定(5,009 例)或安慰剂(5,001例)。受试者的平均年龄为68.6岁，男性占52.8%。由于血管病疾病病史；年龄大；需要透析；吸烟；或者合并了糖尿病、心衰、一过性脑缺血发作或高血压，大部分受试者都存在出现大血管并发症的风险。受试者接受的手术类型包括普外科、骨科、泌尿外科、妇科、血管外科和胸外科手术。

阿司匹林组7%的患者出现了主要结局死亡或MI，安慰剂组为7.1%，差异无统计学意义。两组在其他不良结局，包括卒中、心脏血运重建、肺栓塞、急性肾损害和深静脉血栓形成等方面的风险也没有统计学差异。

此外，阿司匹林组和安慰剂组在中位住院时间、ICU和CCU停留时间等方面也都没有统计学差异。不过，阿司匹林组出现大出血的风险(4.6%)高于安慰剂组(3.8%)，危险比为1.23。最常见的出血部位是手术部位和消化道。

上述效应在所有患者亚组中都是一致的。特别是无论患者是否已经开始接受每日阿司匹林预防性治疗，结果都是一样的。

与安慰剂相比，可乐定也不能预防术后30天内出现死亡或MI；两组的发生率分别为6.8%和7.3%。两组在其他不良结局风险、住院时间、ICU和CCU停留时间等方面也都没有统计学差异。不过，可乐定使患者出现具有重要临床意义的低血压(47.6%比37.1%)、具有重要临床意义的心动过缓(12%比8.1%)和非致命性心跳骤停(0.3%比0.1%)的风险增加。

上述效应在所有患者亚组中也都是一致的。特别是在接受血管手术的患者亚组中，可乐定并不优于安慰剂，而既往小规模研究表明将该药用于这类患者可以预防MI和死亡。

虽然设计开展POISE-2试验并不是为了确定阿司匹林为何不能有效预防围手术期MI，但Devereaux博士及其同事也给出了三种可能的解释。第一，MI与大出血相关，而阿司匹林会增加患者发生大出血并发症的风险。研究者说：“阿司匹林有可能通过抑制血栓形成预防了部分围手术期MI，但同时也通过诱导出血以及之后的心肌血氧供需失衡导致了部分MI，因此在这项研究中总体呈现出中性效应。”

第二，从统计学上讲，心率结果并不能排除阿司匹林治疗可能具有的中度有益效应。第三，冠脉血栓形成有可能并不是围手术期MI的主要机制，所以阿司匹林的抗血栓形成效应不能解决这一未知的主要机制。

同样，也不清楚可乐定为何没能发挥保护效应，但研究者给出了两种可能的原因。第一，可乐定会诱导低血压，这会增加患者出现围手术期MI的风险。第二，可乐定还会诱导心动过缓，这可能提示了该药对心率控制的总体负面效应；而且这也会增加患者出现围手术期MI的风险。

POISE-2试验由加拿大卫生研究院、澳大利亚国家卫生与医学研究理事会、西班牙卫生与社会政策部共同资助。拜耳制药提供了该试验使用的阿司匹林，勃林格殷格翰提供了可乐定和部分研究经费。Devereaux博士声明与雅培、拜耳制药、勃林格殷格翰、Covidien、罗氏和Stryker公司之间存在利益关系；其他一些作者也声明与多家企业存在利益关系。

随刊评论

围手术期MI的发生机制很复杂

Prashant Vaishnava博士和Kim A. Eagle博士说，围手术期MI可能存在多种机制，有的机制可能还相互对立，包括出血过多、体液大幅转移、顽固性心动过速、固定性冠脉阻塞引起的心肌压力、严重低血压或高血压、冠脉斑块破裂以及冠脉痉挛等等。

因此，针对其中一种机制的医学治疗有可能反而诱发了另一种不同的机制，从而最终导致MI风险增加，这并不令人感到意外。他们指出：“阿司匹林在降低冠脉血栓形成风险的同时可能也增加了出血过多的风险；可乐定可能只是减少了抵消具有重要临床意义的低血压的高血压性波动。

总的来说，POISE-2试验的结果提供了令人信服的证据，不支持围手术期使用阿司匹林或可乐定。

Prashant Vaishnava博士和Kim A. Eagle博士来自美国密歇根大学卫生系统Samuel与Jean A. Frankel心血管中心。他们声明无潜在利益冲突。上述内容摘自POISE-2试验报告的随刊评论(New Engl. J. Med. 2014 March 31 [doi:10.1056/NEJMe1402976])。

 

Neither perioperative aspirin therapy nor perioperative clonidine prevented death or MI in patients undergoing noncardiac surgery who were at risk for major vascular complications, according to data reported at the annual meeting of the American College of Cardiology.

Far from being protective, both preventive strategies exerted harmful effects: Aspirin raised the risk of major bleeding, and clonidine increased the risks of clinically important hypotension, bradycardia, and nonfatal MI, said Dr. P. J. Devereaux of the Population Health Research Institute, David Braley Cardiac, Vascular, and Stroke Research Institute, Hamilton (Ont.) General Hospital, and his associates in the POISE-2 (Perioperative Ischemic Evaluation 2) clinical trial.

The POISE-2 findings were simultaneously reported online in the New England Journal of Medicine 2014 March 31 [doi:10/1056. NEJMoa1401105] and [doi:10/1056.
NEJMoa1401106]).

MI is the most common major vascular complication related to noncardiac surgery, and perioperative aspirin is thought to prevent it by inhibiting thrombus formation. At present, one-third of patients at risk for vascular complications receive perioperative aspirin even though the risks and benefits of this preventive strategy are uncertain.

Similarly, small trials have indicated that the antihypertensive agent clonidine, an alpha₂ adrenergic agonist, reduces the risk of myocardial ischemia without inducing hemodynamic instability when given to at-risk patients undergoing noncardiac surgery, which in turn may prevent MI and death. Clonidine has additional analgesic, anxiolytic, antishivering, and anti-inflammatory effects that may be helpful.

The POISE-2 trial was designed to determine whether either of these approaches was more effective than placebo at preventing the composite endpoint of MI or death within 30 days of surgery.

A total of 10,010 patients were enrolled at 135 hospitals in 23 countries, stratified by whether they were already taking daily aspirin prophylaxis, and randomly assigned in a double-blind fashion to receive either perioperative aspirin (4,998 subjects) or placebo (5,012 subjects), and to receive either perioperative clonidine (5,009 subjects) or placebo (5,001 subjects). The mean age of these participants was 68.6 years, and 52.8% were men. Most were at risk because of their history of vascular disease; advanced age; need for dialysis; smoking status; or comorbid diabetes, heart failure, transient ischemic attack, or hypertension.

The types of surgery they underwent included general, orthopedic, urologic, gynecologic, vascular, and thoracic procedures.

The primary outcome of death or MI occurred in 7% of the aspirin group and 7.1% of the placebo group, a nonsignificant difference. The risks of other adverse outcomes, including stroke, cardiac revascularization, pulmonary embolism, acute kidney injury, and deep vein thrombosis, also were not significantly different between the two groups.

Median length of hospital stay and length of ICU and CCU stays also were not significantly different between patients who received aspirin and those who received placebo. However, aspirin raised the risk of major bleeding (4.6%), compared with placebo (3.8%), for a hazard ratio of 1.23. The most common sites of bleeding were the surgical site and the GI tract.

These effects were consistent across all subgroups of patients. In particular, they were the same whether the patients were already taking daily prophylactic aspirin therapy.

Clonidine also did not prevent death or MI within 30 days, compared with placebo; the rates were 7.3% and 6.8%, respectively. The risks of other adverse outcomes also were not significantly different between the two groups, nor were lengths of hospital, ICU, or CCU stays. However, clonidine raised the risk of clinically important hypotension (47.6% vs. 37.1%), clinically important bradycardia (12% vs. 8.1%), and nonfatal cardiac arrest (0.3% vs. 0.1%).

These effects were consistent across all subgroups of patients. In particular, clonidine was no more beneficial than was placebo in patients who underwent vascular surgery, a subgroup in whom previous, smaller studies found the drug to protect against both MI and mortality.

POISE-2 was not designed to determine why aspirin wasn’t effective at preventing perioperative MI, but Dr. Devereaux and his associates offered three possible explanations. First, MI was associated with major bleeding, and aspirin raises the risk of this complication. "It is possible that aspirin prevented some perioperative MI through thrombus inhibition [but] caused some MIs through bleeding and the subsequent mismatch between the supply of and the demand for myocardial oxygen, thus resulting in the overall neutral effect in our study," they said.

Second, the heart rate findings couldn’t statistically rule out a possible moderate beneficial effect of aspirin therapy. And third, it is possible that coronary-artery thrombus isn’t the dominant mechanism of perioperative MI, and aspirin’s antithrombotic effect didn’t address this unknown dominant mechanism.

Similarly, it is not known why clonidine failed to be protective, but the investigators offered two possible reasons. First, the drug induced hypotension, which raises the risk of perioperative MI. And second, it also induced bradycardia, which may be a proxy for an overall adverse effect on heart rate control; this also can increase the risk of perioperative MI, they said.

The POISE-2 trial was supported by the Canadian Institutes of Health Research, the National Health and Medical Research Council of Australia, and the Spanish Ministry of Health and Social Policy. Bayer Pharma provided the aspirin used in the study, and Boehringer Ingelheim provided the clonidine and some funding. Dr. Devereaux reported ties to Abbott, Bayer Pharma, Boehringer Ingelheim, Covidien, Roche, and Stryker and; some of his associates reported ties to several industry sources.

View on the News

Complex mechanisms underlie perioperative MI

Myriad, and sometimes opposing, mechanisms contribute to perioperative MI, including excess bleeding, dramatic fluid shifts, unrelenting tachycardia, myocardial stress with fixed coronary obstruction, profound hypo- or hypertension, coronary plaque rupture, and coronary spasm, said Dr. Prashant Vaishnava and Dr. Kim A. Eagle.

So it shouldn’t be surprising that a medical therapy aimed at one of these mechanisms may actually augment a different mechanism, and end up raising MI risk. "Aspirin may reduce coronary thrombosis at the expense of excess bleeding; clonidine may reduce hypertensive swings only to be countered by clinically important hypotension," they noted.

On balance, the POISE-2 results provide cogent evidence against the use of either perioperative aspirin or clonidine.

Dr. Vaishnava and Dr. Eagle are at the Samuel and Jean A. Frankel Cardiovascular Center at the University of Michigan Health System, Ann Arbor. They reported no potential conflicts of interest. These remarks were taken from their editorial accompanying the POISE-2 trial reports (New Engl. J. Med. 2014 March 31 [doi:10.1056/NEJMe1402976]).