What happens when we don’t publish clinical trials

The last blog I posted emphasised the importance of publishing all clinical trials. The story of Lorcainide is a stark warning of what happens when we don’t.

In 1980 a cardiologist in Nottingham named Alan Cowley carried out a small clinical trial of a drug called Lorcainide. It was known at the time that heart attacks could cause irregular heartbeats in patients (known as arrhythmia), and these arrhythmias often lead to early death. Lorcainide had been shown to suppress arrhythmia, so it made sense that patients who came to hospital with a heart attack should be treated with the drug. Cowley and his colleagues carried out a small trial with 95 patients, and tested them to see whether they were getting more or fewer arrhythmias. The drug worked, lowering the frequency of serious arrhythmia.

The doctors noticed something else however. Of the 48 patients on the drug, 9 had died, compared to only 1 patient on the placebo. This was a very small trial, so the doctors weren’t overly alarmed. It’s not surprising that 10 patients died in the study; these are patients who are presenting with heart attacks after all. It was just worrying that there was such an imbalance between the groups. The doctors chalked it up to bad luck, and viewed their trial as a success.

At the time, this was a perfectly valid opinion. The study had been designed to analyse arrhythmias, not look at mortality. Furthermore, it was a tiny study, so they were justified in assuming the increased death was down to chance. Unfortunately however, what happened next ensured that the importance of this study would not be recognized.

The doctors wrote an article describing their findings and tried to get it published. They submitted it to three different journals, but without success.  At the same time, the company that made Lorcainide decided to discontinue it (for unrelated commercial reasons), so the doctors lost interest and decided not to publish their results.

To be clear, they were trying to publish the study as a success. Lorcainide was able to decrease serious arrhythmias after a heart attack. But within the paper was the information about increased mortality, and this would have been noticed. If it had been published, the study may not have prevented prescription, but it would certainly have suggested the need for further study.

Although Lorcainide was never brought to market, other similar anti-arrhythmia treatments were prescribed to heart attack patients throughout the 80s. However, in 1983 there was a review of the available literature that proposed that there was no benefit in using these drugs. The authors of that study actually suggested that there might even be increased death following treatment, but this harmful effect was too small to be sure it was real.

In hindsight it is clear that this small effect on mortality was in fact bigger than was realised. That study looked at published data to come to their conclusion. However, they were missing an important clinical trial, one that was in fact sitting unpublished on a hospital desk. If they had access to this data, they may have come to a different conclusion, flagging up the danger years earlier than it was.

Towards the end of the decade, after more trials were published, two studies were carried out, both of which suggested that these drugs were doing more harm than good. At this point the danger was realised, and prescriptions dropped. However, it is estimated that 20,000 to 75,000 people died every year because of the use of anti-arrhythmia medication.

In 1993, 13 years after it was originally carried out, the clinical trial on Lorcainide was published. The authors pointed out that it was perfectly reasonable to assume the increased death was a matter of chance, and they are probably correct in that. Unfortunately, when they decided not to publish and leave their study to gather dust, they contributed to an unfolding tragedy. At the time, the need to publish all trials regardless of their results wasn’t appreciated, so whether they can be blamed for what happened is a difficult question. What is certain however is that as a result of not publishing hundreds of thousands of people died.

Well carried out clinical trials are the bedrock of modern medicine, and unbelievably, we are still in a ridiculous situation where reporting and publishing of trials is patchy at best. Until this situation is corrected we are at risk of another catastrophe like Lorcainide. Sign the AllTrials petition here to register your support for reform.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s