The guideline committee has made the following recommendations for research.
What is the most clinically and cost-effective dual antiplatelet therapy for people aged 75 and over with an acute coronary syndrome, who are having percutaneous coronary intervention (PCI)? 
The evidence reviewed for this guideline found that prasugrel is the most clinically and cost-effective antiplatelet therapy when used with aspirin for the general acute coronary syndrome population having PCI, particularly for people with ST‑segment elevation myocardial infarction (STEMI). However, the summary of product characteristics for prasugrel states that its use in people aged 75 and over 'is generally not recommended and should only be undertaken with caution after a careful individual benefit/risk evaluation by the prescribing physician indicates that benefits in terms of prevention of ischaemic events outweigh the risk of serious bleedings'. There were not enough data available for this guideline to determine whether prasugrel is less effective, or even harmful, in people aged 75 and over. Further research is needed to determine the optimal dual antiplatelet therapy for this group of older people.
2 Primary PCI and fibrinolysis in people with acute STEMI who have a long anticipated transfer time for primary PCI
In people with acute STEMI who present more than 1 hour after the onset of symptoms, is a primary PCI-related delay of 120 to 180 minutes associated with outcomes similar to, better or worse than pre-hospital administered fibrinolysis? 
Primary PCI is the preferred coronary reperfusion therapy provided it can be delivered 'in a timely fashion'. It is suggested that primary PCI is the preferred reperfusion strategy for primary PCI-related delays of at least up to 2 hours. However, there is inadequate evidence to conclude whether primary PCI is still preferable at primary PCI-related time delays of more than 2 hours.
No specifically designed randomised controlled trial (RCT) or observational study has addressed the issue of the extent to which primary PCI-related time delay (and other factors such as presentation delay and a person's risk profile) diminishes the advantages of primary PCI over fibrinolysis. For example, in more geographically remote areas, a short presentation delay together with an anticipated long primary PCI-related delay could favour a strategy of pre-hospital fibrinolysis.
To answer this question, an RCT of pre-hospital fibrinolysis versus primary PCI in people with acute STEMI who have a primary PCI-related time delay of 2 hours or more is needed. Primary endpoints would include cardiovascular and all-cause mortality and other major adverse cardiovascular events.
What is the role of ischaemia testing in people after an acute coronary syndrome and what is the comparative efficacy and cost effectiveness of the different non-invasive tests (for example, stress electrocardiogram [ECG], echocardiography, radionuclide scanning and MRI)? 
An increasing number of non-invasive tests are now available for the investigation of suspected myocardial ischaemia. These tests need different equipment, different clinical expertise, come at different costs and may differ in their ability to detect and quantify myocardial ischaemia. Their place in the routine investigation of patients admitted with unstable angina and non-ST‑segment elevation myocardial infarction (NSTEMI) (particularly those who have not undergone angiography), as opposed to their selective use, is not clear. Management of unstable angina and NSTEMI would be enhanced if the relative place of these investigations were better understood and an assessment of their cost effectiveness made.
What is the relationship between hospital volume of primary PCI procedures and optimal outcomes in people with acute STEMI? 
There is a suggestion that outcomes may be better in larger-volume primary PCI units, and some retrospective registries have reported data to support this. However, the quality of the data is poor and still leaves the question open. In the UK, primary PCI is provided by units that vary greatly in the number of cases per year. The development of services has been ad hoc and not designed specifically around the provision of primary PCI. If it was possible to conclusively show that people were or were not better off having treatment in larger-volume units, then it would have important implications for the national provision of primary PCI.
What is the clinical and cost effectiveness of the systematic use of risk scoring systems (in addition to clinical assessment) for ischaemic outcomes and bleeding complications in the management of unstable angina and NSTEMI (at all levels of risk) compared with clinical assessment alone? 
Most risk scoring systems currently predict the likelihood of mortality or ischaemic cardiovascular events at various times after a patient's admission to hospital with an acute coronary syndrome. A number of interventions (such as drugs and revascularisation procedures) have been shown to reduce these adverse outcomes. This effect tends to be greatest in patients at highest risk. However, as a broad generalisation, patients who are at highest ischaemic risk are also those who are at higher risk of bleeding complications associated with the use of multiple antiplatelet and antithrombin agents. There are fewer scoring systems that predict bleeding risk, but we know that bleeding complications are associated with a significantly worse outcome. Using a combination of scoring systems assessing ischaemic and bleeding risk when evaluating data from randomised trials and registries may help to determine where the net clinical benefit (reduction in ischaemic risk minus any increase in bleeding risk) lies.
For patients with unstable angina and NSTEMI (at differing levels of risk), how do clinical outcome data (adverse cardiovascular events and bleeding complications) collected in cardiac registries compare with data derived from RCTs? 
Patients recruited to participate in clinical trials are often highly selected; trials tend not to include patients who are very elderly, are at high risk, or have significant comorbidity. On the other hand, good registry data include information on all patients, but are observational and not randomised. Often there is uncertainty about how the outcome data from RCTs can be applied to the much larger unselected population of patients admitted to UK hospitals with unstable angina or NSTEMI. A greater understanding of the differences between RCT and registry populations, and their levels of ischaemic and bleeding risk would help inform future management. Collection of well-validated registry data is essential if conclusions from RCTs are to be applied appropriately to all patients with unstable angina and NSTEMI, not just to patients who are comparable to trial populations.
What is the optimal management of hyperglycaemia in people with acute coronary syndrome who have diagnosed or previously undiagnosed diabetes? 
Existing studies on the optimal management of hyperglycaemia in people who have acute coronary syndrome and diagnosed or previously undiagnosed diabetes are generally of poor quality.
It is recommended that a large RCT is conducted for people with acute coronary syndrome and hyperglycaemia (blood glucose 11 mmol/litre and over) stratified by NSTEMI and STEMI and by known diabetes and without a previous diagnosis of diabetes.
The interventions for the trial should be intravenous insulin or subcutaneous insulin administered within 4 hours of presentation to hospital. The aim is to achieve blood glucose between 6 and 11 mmol/litre for at least 24 hours. The comparator should be standard care.
Does continuing beta-blocker treatment beyond 1 year after a myocardial infarction (MI) improve outcomes for people with normal left ventricular systolic function? 
Recent cohort studies have suggested that continuing treatment with a beta-blocker beyond a year after an acute MI may not confer any benefit to the person in terms of reduced morbidity or mortality. This is particularly relevant given recent changes in acute management strategies. While beta-blockers are valuable in reducing mortality and morbidity for up to a year after an MI, they have side effects and represent an additional treatment burden to people who are already taking many other medications. However, there is also some suggestion that there are risks associated with withdrawal of beta-blockers in this population. The balance of risks and benefits of long-term beta blockade has not been clearly determined, particularly in the context of modern acute treatment of MI.