Most people I talk to who work in Emergency or Acute Medicine hate chest pain. Personally I think they’re crazy but I can see where they’re coming from. Chest pain is the most common reason for emergency admission, making up 1 in 4 acute medical admissions. Most of those patients (at least three quarters, depending on where you come from) will have ACS ruled out after investigation. No wonder we’re disillusioned. And the initial management is hardly challenging either, right? Aspirin, nitrates +/- opiates, clopidogrel and antithrombotics (fondaparinux or low molecular weight heparin), check the ECG and the troponin and take a careful history and examination. Easy peasy. Antiplatelets like clopidogrel or prasugrel (the thienopyridines) have been on the menu for years – ever since the CURE trial, which seemed to show early benefit even in ‘lower risk’ patients – although remember that most patients in that trial either had ECG changes or troponin elevations.
Along came ACCOAST…
This recent trial was published in the NEJM. The abstract is below. If you can access the full paper it’s well worth a read…
[DDET Why is this trial important?] In the ED we often treat patients with suspected ACS with antiplatelet agents empirically, before the diagnosis has been confirmed. Most of these will ultimately have that diagnosis excluded. Our highest risk patients who have dynamic ECG changes or substantial troponin elevations are likely to undergo coronary intervention. We need to know whether these patients derive benefit from being pre-loaded with antiplatelets in the ED, potentially 24 hours or longer before the intervention, or whether we ought to wait until nearer the time. Clearly, antiplatelets aren’t without risk – they can cause major bleeding, so it’s important to know that the benefits outweigh that risk for our patients. What’s more, the newer antiplatelets (prasugrel and ticagrelor) act much quicker than clopidogrel and their action is more reliable. It may be that there is less benefit in pre-treatment when these newer agents are used.
The ACCOAST trial set out to determine whether preloading high risk patients with ACS (who don’t have ST elevation) with the antiplatelet prasugrel is better than giving prasugrel after angiography, and only to those patients who needed intervention. [/DDET]
[DDET Who was studied and what was the intervention?] The patients were quite high risk. They all had confirmed ACS (without ST elevation) and a troponin rise. They were randomised to get prasugrel before angiography (30mg) + 30mg if they underwent PCI (the pre-loaded group) or prasugrel (60mg) only if they underwent PCI – i.e. with no pre-loading. There seems to have been appropriate placebo control and blinding. [/DDET]
[DDET What was the primary outcome?] A composite of the following within 7 days:
* Myocardial infarction
* Urgent coronary revascularisation
Or * The need for rescue therapy with a glycoprotein IIB/IIIA inhibitor
I can follow the rationale for the first 2 outcomes, and I understand the third – which is very commonly used in cardiovascular trials. I’m not sure I’m convinced that the fourth outcome is actually as clinically relevant though – and it seems rather subjective. We do need to bear that in mind when interpreting the results. [/DDET]
[DDET Is this trial appropriately powered?] The trial included 4,403 patients, 398 of whom developed the primary outcome. The authors’ power calculation supposed that 400 patients with the primary outcome (and 4,100 patients in total) would be needed to provide 80% power to detect a relative risk reduction of 24%. We don’t know why they decided that 24% was the minimal clinically significant difference, which is a bit of an issue, but the authors at least did have a rationale for their sample size and achieved that. [/DDET]
[DDET What were the key findings?] There was no difference in the rate of the primary outcome. 10.0% of patients who were pre-treated developed the primary outcome compared to 9.8% of the control group (p=0.81). Similarly, there were no differences in the rates of death, myocardial infarction, urgent revascularisation or stroke – but remember that these are secondary outcomes and the trial was not powered to detect differences in them.
In contrast, more patients in the pretreatment group developed major bleeding than in the control group (2.6% vs. 1.4%). The hazard ratio was 1.90 (95% CI 1.19 – 3.02, p=0.006) meaning that patients who were pre-treated were developing major bleeding at almost twice the rate of those in the control group through the first 7 days of follow up. By day 30 the hazard ratio was 1.97 – meaning that patients who were pre-treated were still developing more major bleeding. [/DDET]
[DDET What does this mean for our practice?] Accepting some of the limitations of this work (e.g. the inclusion of glycoprotein inhibitor bailout therapy in the primary outcome), there’s still plenty we can take away. Pre-treated patients didn’t do better than patients treated selectively after angiography. In fact, they did worse overall – because they were more likely to develop major bleeding.
If patients with confirmed non-ST elevation ACS don’t benefit from pre-loading with antiplatelets, the benefit-harm ratio can only get worse if we use these agents in undifferentiated patients who don’t yet have the diagnosis confirmed. That should make us think twice about using antiplatelets in the ED.
In my opinion, the bottom line is that antiplatelets as a routine treatment for patients with presumed ACS are out. Unless your hospital is still using clopidogrel (which has a slower onset of action) there’s also an argument for reserving prasugrel or ticagrelor for those patients who have actually undergone coronary intervention – and therefore not loading in the ED. [/DDET]
This paper may have a few weaknesses. (What trial doesn’t?) But it has definitely changed my practice!