The last couple of weeks have taught me a lot about how clinical research gets published and disseminated. On 1st February, our article, titled, “Familial History of Stroke Is Associated with Acute Coronary Syndromes in Women”, was published online by Circulation Cardiovascular Genetics prior to print. This is one of the subspeciality journals of “Circulation”, a publication aimed at scientists generally interested in cardiovascular medicine and research. Over the last few years, several of the major journals have increasingly multiplied into “sub-speciality journals”. For example, “Lancet” has spawned “Lancet Neurology”, “Lancet Infectious Diseases” and “Lancet Oncology”. Not only are the journals able to publish more specialised research that might have been rejected by the parent, general journal; they can also charge more for subscriptions, reprints and so on.

Interestingly, even as an author, I did not receive a copy of my manuscript and do not have access to the journal article, unless I pay for access or buy a reprint. So I will tell you what the gist of the research is. Basically, within the Oxford Vascular Study, a much studied cohort of 90 000 patients from the Oxfordshire general practice population, we looked at about 1000 patients with stroke and about 1000 patients with heart attacks.

Previous analyses from the same study have shown that women with stroke are twice as likely to have female relatives with stroke as male relatives with stroke. In addition, young women with heart attacks are twice as likely as young men with heart attacks to have mothers with heart attacks. Therefore, mother-to-daughter transmission seems to be important. For this reason we looked at family history of stroke in detail among patients with acute coronary syndromes (heart attacks and “unstable angina”).

Firstly, we found that family history of stroke is as common in heart attack patients as stroke patients. Secondly, women with heart attacks were twice as likely to have history of stroke in their mother as in their father. Thirdly, when we looked at the coronary arteries (which supply the heart) directly using coronary angiography, family history did not predict the location of the coronary artery disease or how severe it was. We concluded that family history of stroke needs to be studied in more detail and may well be important in better identifying women most at risk of heart attack, since women are less likely to be picked up by current risk prediction tools used by doctors. Also family history probably has its effect via influences on clotting rather than on arteries directly, given our lack of correlation with disease on angiography.

I spoke with only 1 freelance American journalist and helped write 2 press releases in the week before the article went online. On 1st February, I received an e-mail from the University that the article had been picked up by news brokering websites from Reuters to Yahoo, newspapers and TV from Canada and the Phillippines to India, South Africa and the UK tabloids. I don’t mind telling you I was surprised by all the interest! Even BBC Radio 4 contacted me to go on Woman’s Hour.

I have taken three lessons from this experience. First, a journal article will probably be read by almost nobody, primarily because it is published in a journal, and secondly because access to that journal requires money. Second, although research is published in journals, the immense speed and penetration of the global media/internet machine (based in this case on 3 interviews or press releases!) have led to the devolution of the dissemination of research findings away from journals, even though journals may be the trusted source of the original research. Third, as scientists, if we want our research to be understood by the broader public, then we need to do more than publish articles in journals, we must engage with the media and with the public. Both doctors and patients are more likely to use internet search engines than journals so we have to make sure that Google is well-informed, otherwise a great opportunity for health communication will be missed.

Evidence-based guidelines recommend that after a heart attack, the blocked coronary artery needs to be reopened quickly by either fibrinolysis (or “clot-busting” drugs) or primary percutaneous coronary intervention (primary PCI), which aims to open the artery using balloons and stents. However, there are several reasons for delay in these treatments.

Firstly, “patient delay” is the delay from the onset of chest pain or symptoms to when a call is made to emergency medical services, and can only really be reduced by better public education about heart attacks. Secondly, “system delay” is a combination of “transportation delay” (the time taken for the patient to get to the hospital) and “door-to-balloon delay” (the time taken for the patient to receive the artery-opening therapy once they are in the hospital). In terms of training of doctors and measurement of outcomes within hospitals and across health systems, there has been a huge focus on the “door-to-balloon” delay. However, to know the effect of delaying therapy on outcome, we need to look at “system delay”, which is what a Danish study does in this week’s JAMA.

Due to excellent public medical databases in Denmark, the authors were able to study over 6000 patients with the particular form of heart attack (“STEMI”) which is best treated by primary PCI, and obtain estimates for the various types of delay outlined above. The authors excluded patients with a treatment delay greater than 12 hours or a system delay greater than 6 hours.

Interestingly, treatment delay and patient delay were not associated with mortality, but the authors are quick to assert that “should not deter encouraging patients to seek medical help as soon as possible after the onset of symptoms”. On the other hand, system delay predicted mortality, with a hazard ratio of 1.10 per 1-hour delay. In other words, for every one hour of system delay, there is a 10% increase in mortality. When the authors analysed further, they found that a 1-hour transportation delay led to 10% increase in mortality, whereas a 1-hour door-to-balloon delay led to a 14% increase in mortality. In other words, time does really mean muscle (and life) when it comes to the heart.

Most medical students will recognize the quote:

‘Half of what you'll learn in medical school will be shown to be either dead wrong or out of date within five years of your graduation; the trouble is that nobody can tell you which half—so the most important thing to learn is how to learn on your own.’

Dave Sackett: “Old fart from the frozen north” “Father of EBM”

The rapid assessment and treatment of a patient with a heart attack is drummed into most medical students very early on in their training. ABC: airway, breathing, circulation. Part of that resuscitation is the delivery of Oxygen to patients with a heart attack, mainly due to the fact the flow of oxygenated blood in the heart is stopped for a period of time.

The idea for providing oxygen in a heart attack is it may improve the amount of oxygen of the cells in the heart that are dying mainly due to the lack of oxygen, ultimately reducing pain and the size of the dead heart muscle. To most this will make sense in terms of pathophysiological reasoning.

Today a Cochrane review by Cabello and Burls on Oxygen therapy for acute myocardial infarction looks at the evidence from randomised controlled trials to establish whether routine use of inhaled oxygen in acute heart attack infarction improves patient-centred outcomes, in particular pain and death.

Now, here is the half of what is learnt learn that may eventually be out of date:

Three trials involving 387 patients were included and 14 deaths occurred. The pooled relative risk of death was 2.88 (95% confidence interval 0.88 to 9.39) in an intention-to-treat analysis and 3.03 (95% confidence interval 0.93 to 9.83) in patients with confirmed heart attack.

While suggestive of harm, the small number of deaths recorded meant that this could be a chance occurrence. Basically, there is no conclusive evidence from randomised controlled trials to support the routine use of inhaled oxygen in patients with acute heart attack.

The neat thing about EBM is you are never really sure of which half is out of date; this review adds to that half. As the reviewers rightly state, we need an urgent large scale trial to unpick the uncertainty.