Issue 35 / 10 September 2012

SILENT myocardial infarction (MI) is far more prevalent than previously thought, according to research published last week, which an Australian expert says may have “enormous clinical significance”.

The study, published in the Journal of the American Medical Association, examined a community-based cohort of 936 people aged 67–96 years in Iceland. (1)

It found that 17% had unrecognised or silent MI detected with cardiac magnetic resonance imaging (CMR) compared with 10% who had recognised MI. Electrocardiography (ECG) alone detected only 5% of unrecognised MIs.

The research also found that more than one in five participants with diabetes (21%) had unrecognised MI, and that unrecognised MI detected by CMR was associated with subsequent mortality over 6–7 years.

Professor Richard Harper, emeritus director of cardiology at Monash Medical Centre and adjunct professor of medicine at Monash University, said the study showed a much higher than expected incidence of unrecognised MI using the technique of CMR. He said these findings, particularly if they could be replicated in other patient groups, had enormous clinical significance.

He said it had been known for some time that there was a significant incidence of silent or unrecognised MIs, particularly in patients with diabetes. For example, he said, the Framingham studies, using ECG criteria for recognition, suggested that 15%–20% of MIs were silent. “But in this study using the more sensitive technique of CMR, 63% [157] of the total number of MIs [248] were unrecognised”, Professor Harper said.

He said that while this was just one study in an elderly Icelandic population, it was important. “It brings to attention that many, perhaps most, heart attacks in diabetics and in the older age group are unrecognised, and therefore it is imperative that diabetics are given cardioprotective medications in addition to medications for their diabetes. The only exception would be if you’ve proven that the patient had no coronary atherosclerosis. For practical purposes, it can be assumed that diabetics have  coronary atherosclerosis”, he said.

Professor David Brieger, professor of cardiology at the University of Sydney’s Concord Clinical School, said the paper showed that in an elderly community, magnetic resonance imaging (MRI) scanning was a more effective way to detect silent Ml than 12-lead electrocardiography. “Furthermore, the presence of unrecognised MI detected by MRI is associated with an adverse prognosis, similar to patients with recognised MI”, he said. “This makes sense and confirms our understanding that MRI scanning is a very effective way of detecting small amounts of myocardial necrosis.”

However, Professor Brieger said the authors’ suggestion that MRI may have a role as a screening tool allowing detection of high-risk patients was premature.

“Firstly, it would need to be shown that adding MRI to conventional screening strategies, such as the routine application of the National Heart Foundation absolute risk score, identified an additional population with an adverse prognosis”, he said.

“Secondly, it would be necessary to show that detecting MI in this way and acting on the findings improved prognosis among these patients. Finally, the restricted access to cardiac MRI in our health care environments means that screening will be impractical for the foreseeable future.”

Professor Harper said the most cost-effective screening strategy would be to assume that all males aged over 50 with two or more risk factors and females over 60 with two or more risk factors had coronary atherosclerosis. They could then be provided with appropriate lifestyle advice and started on a cardioprotective pharmacological regimen, comprising an antiplatelet agent such as aspirin, a lipid-lowering agent such as a statin, and an angiotensin-converting enzyme (ACE) inhibitor, he said.

If thought necessary, Professor Harper said coronary atherosclerosis could also be detected with computed tomography (CT) coronary angiogram or a CT calcium score.

“If we really want to reduce the risk of heart attack in the older population, I think we need to screen for coronary atherosclerosis and then start people on a primary preventive regimen, which has been proven to reduce the incidence of subsequent heart attacks”, he said.

– Nicole Mackee

1. JAMA 2012 ; Published online: 5 September

Posted 10 September 2012

7 thoughts on “Silent MI: deadly and common

  1. ex doctor says:

    What action will be taken above and beyond enthusiastic application of already known and validated prevention and treatment measures?
    First prove that the “additional population with an adverse prognosis” identified by MRI over those identified by the extremely cheap NHS absolute risk score can achieve intervention outcomes sufficiently excellent as to justify the massive cost of MRI screening. The health economists will love this one.

  2. Sue Ieraci says:

    Ex-doctor is spot on: if there is no intervention other than risk factor modification, what is the point of making these retrospective diagnoses?

  3. Cardiac nurse says:

    Knowledge is power. We can at least educate diabetics, the elderly,and those over 50 years with a family history or those patients with 1 or more major risk factors for CAD that about 50% people have atypical symptoms prior to or at the time of a heart attack ranging from being suddenly more SOB than usual or sudden absolute exhaustion, cold clammy and unwell.
    Denial and delay to hospital to be avoided.

  4. Sue Ieraci says:

    Rather than increasing imaging, this excerpt seems like a logical response: “”It brings to attention that many, perhaps most, heart attacks in diabetics and in the older age group are unrecognised, and therefore it is imperative that diabetics are given cardioprotective medications in addition to medications for their diabetes. The only exception would be if you’ve proven that the patient had no coronary atherosclerosis. For practical purposes, it can be assumed that diabetics have coronary atherosclerosis”, .

  5. peter says:

    Elderly diabetics are highly likely to have atherosclerosis and further investigation with Coronary CT or calcium score are unlikely to be helpful in determining further management. A functional test such as nuclear perfusion study (sestamibi) or stress Echo may give more clinically relevant information in this group if more than risk factor modification is contemplated.

  6. John Younger says:

    Unrecognised MI is likely to be incredibly important. Secondary prevention measures (including aspirin, ACEI, statin), plus cardiac rehabillatation / investigation, are very effective in preventing death and subsequent events in the post MI population. Therefore identifying these patients is crucial if they are to be treated.
    Risk factor scoring tools are not used once a patient has had a myocardial event – these tools are for primary prevention (and miss 2/3 of MI patients).
    We know (from the DIAD study, JAMA 2009) that routine myocardial perfusion scanning is NOT effective in reducing risk in asymptomatic diabetics. So there is unlikely to be a role for stress nuclear imaging.
    We also know that nuclear scans miss many infarcts that are detected by MRI (Wagner, Lancet 2002). This may be why MRI is superior. Also worth noting that, in unselected cohorts, stress nuclear scans have only a 50% sensitivity for detecting coronary stenoses, (Greenwood, Lancet 2012).
    We know from CT coronary calcium studies that diabetes is NOT a universally associated with coronary artery disease, despite what was previously thought. It can not be assumed that all diabetics have coronary disease.(Perhaps duration and severity of diabetes will be more crucial than a single dicotamous variable; DM or no DM).
    The numbers needed to treat (if there is no coronary disease)are very high and it is very expensive to simply treat all diabetics with cardiac meds. So what should we do?
    Although this study tells us nothing about the cost effectivness of identifying these patients, or if they would benefit from intervention, these would be crucial to know before embracing screening.
    Perhaps using simple echo to identify those patients with LV dysfunction (which is very prevelant – 32% – and under-diagnosed in the elderly – Yousaf – Heart, this week) might be a worthwhile alternative?

  7. Ben Ewald says:

    The discovery of a high prevalence of unrecognised MI makes a big difference to the number of patients with indications for statins. As we swing away from treating cholesterol levels to treating people with established IHD this work demonstrates a new group of patients who should be on maximal preventive treatment. Unfortunately the main diagnostic test we have available in general practice, ECG, identifies a group with unrecognised MI who are not at increased risk. This finding seems counterintuitive and needs replication.

Leave a Reply

Your email address will not be published.