ON A DARK winter鈥檚 morning earlier this year, 37-year-old Lydia Larkin was in the kitchen of her home getting ready to take her two dogs for a walk, when she suddenly felt light-headed. Then she was overwhelmed by nausea and lost her balance, falling to the floor.
Terrified, she managed to stumble to the bedroom, and woke her husband who called an ambulance. The paramedics thought that Lydia had probably had a stroke. At the hospital, an MRI scan confirmed the unwelcome diagnosis.
Lydia was bewildered. She was healthy, slim, didn鈥檛 smoke and had been told her blood pressure was fine. 鈥淚 had just lost my mum to breast cancer and I was still grieving,鈥� she says. 鈥淚t felt unfair that I should have such bad luck.鈥�
Advertisement
But after some tests, the doctors thought they might have an explanation. Lydia had a hole in her heart.
Astonishingly, 1 in 4 of us is walking around with the same type of heart defect. It has been implicated in strokes, migraines, decompression sickness in divers and perhaps other issues too. Surgeons can fix this kind of hole relatively easily, so anyone at risk should have the operation, right? That is what doctors cannot agree on. Some argue that operating might do more harm than good.
鈥淎stonishingly, 1 in 4 of us is walking around with a hole in the heart. It may be behind strokes, migraines and other issues too鈥�
Lydia, who lives in Dublin in the Republic of Ireland, recovered fairly quickly from her stroke. But she agonised over the decision of whether to have corrective surgery, all the while fearing another stroke might be imminent. 鈥淚 spent my days imagining clots whooshing up to my brain and my nights bolt upright, unable to sleep,鈥� she says.
Like others, Lydia had a hole in the wall that separates the heart鈥檚 two upper chambers. For most people, it probably causes no harm, but there is no way to know for sure if it is causing problems.
Why would such a potentially devastating heart problem be so common? It turns out to be a relic from our time in the womb, when the blood circulation follows a different route. In adults, the right side of the heart sends blood to the lungs, where it picks up oxygen, and the left side sends the oxygenated blood to all the other organs in the body. But an unborn baby doesn鈥檛 use its lungs to breathe, getting all the oxygen it needs through the placenta. So the fetal heart has an opening between the right and left upper chambers, known as the foramen ovale, to bypass the lungs.
When a baby is born and takes its first gasp of air, a flap closes over the hole 鈥� at least in most people. For about 1 in 4 of us, the hole stays open, either all the time or, more commonly, it gapes open when pressure rises in the chest, such as during a coughing fit, or straining to defecate or lift something heavy. This is known as having a patent (open) foramen ovale, or PFO.
Doctors have long suspected this could mean trouble, allowing things to sneak through the gap instead of being filtered out by the fine capillary network of the lungs: blood clots, for instance. Tiny ones that get stuck in the lungs can dissolve harmlessly. But if a blood clot manages to bypass the lungs, it might lodge in a blood vessel elsewhere in the body 鈥� maybe in the brain (see diagram). And blockage of blood flow to the brain is the most common cause of stroke.
Strokes are a major public health problem: about 1 in 5 of us will suffer one in our lifetime and they are the top cause of disability in the West. In 4 out of 10 cases, no one can figure out what triggered the stroke. That is to say the person is relatively young 鈥� that鈥檚 under 55, to a doctor 鈥� and has no obvious risk factors. Could some of these unexplained strokes be caused by a hole in the heart?
Evidence for this idea started to mount in the 1990s, as various studies suggested that the defect is present in 40 to 60 per cent of people who had had an unexplained stroke 鈥� remember, only 25 per cent of the general population has a PFO. 鈥淭hat suggests the PFO plays a role, somehow,鈥� says Anthony Rudd, a stroke physician at Guy鈥檚 and St Thomas鈥� Hospital in London, who is also the 鈥渟troke tsar鈥� of the UK鈥檚 National Health Service.
A PFO is easy to detect once you suspect it might be there; a vein in the subject鈥檚 arm is injected with a solution of tiny bubbles, too small to do any harm. The microbubbles travel to the heart where they can be imaged using ultrasound. Patients are asked to hold their breath and bear down, to force open the flap. If the bubbles can be seen coursing to the lungs, all is well. But if they travel to the left side of the heart, that normally signifies a PFO.
When a hole is discovered in something as important as the heart, the natural response is to find a way to plug it. That used to entail open heart surgery, but then devices were developed that could seal the hole shut without having to open up the chest.
Shaped something like a tiny umbrella, they are pushed into a vein in the groin, threaded up into the heart by a wire, then opened up and left in place while the wire is withdrawn. Eventually, tissue grows over it, sealing the flap.
As confidence grew in the safety of PFO closure, it became offered to increasing numbers of people who had had an unexplained stroke, in the hope of reducing their chance of another stroke. This could be either instead of, or as well as, the standard therapy of blood-thinning medicines or other drug treatment.
Yet unlike the drugs, no trials had been done to demonstrate the operation would achieve that goal, and doctors鈥� opinions differed about how often it should be carried out 鈥� as they still do today. The most enthusiastic tend to be the cardiologists who carry out the procedure, perhaps overly influenced by their close relationships with the device manufacturers. 鈥淚nterventional cardiologists are by definition at the cutting edge of this pioneering new technology,鈥� says Michael Scheck, a neurologist from Loyola University Medical Center in Maywood, Illinois. 鈥淪ome may be a little zealous in fixing a defect without asking all the right questions.鈥�
Soon the closure operation became an option for divers, too, after research revealed that . Divers get the bends when bubbles of nitrogen form in the blood as they ascend, leading to symptoms ranging from itching and joint pain to strokes and paralysis.
Normally the nitrogen is gradually breathed out through the lungs, as long as divers ascend slowly enough. In people with a PFO, the bubbles might bypass the lungs, making the bends more likely, or even cause a stroke as the diver struggles up a ladder laden with wet kit, for instance. 鈥淭he more bubbles that cross over, the more damage the divers suffer,鈥� says Peter Wilmshurst, a cardiologist at the Royal Shrewsbury Hospitals, UK, who led the research (Clinical Science, vol 99, p 65).
PFOs were also suspected of causing migraines, specifically the kind accompanied by strange visual disturbances known as aura. People who get migraine with aura are twice as likely to have a PFO as the rest of the population and about six times as likely to have a large one.
Although the cause of migraine is unknown, the theory goes that a PFO allows something in the blood to bypass the lungs and reach the brain. Doctors found that migraines with aura tended to disappear in people who had their PFO closed for other reasons. So a trial in people with migraine was set up, funded by a maker of a new closure device, called Starflex. It was the first large trial to test PFO closure against people having a sham operation, where patients got a general anaesthetic and an incision made, but no device implanted.
In fact it was one of the first large trials ever to involve such a placebo group, where both patients and the doctors who assessed the outcomes were kept in the dark about whether patients had had the real operation or not. It was seen as ahead of its time in terms of being a true test of an operation.
But the results were not as hoped for, to say the least. Not only did the device fail to stop people getting migraines, but it also caused a higher rate of heart problems than you would expect with this procedure; in two cases the device came loose in the heart, which could have been fatal.
The trial was also the centre of a libel case against Wilmshurst, who was one of the principle investigators. At a cardiology conference, Wilmshurst said to a reporter that in some patients, the device may not have completely closed their PFO, and also that there were flaws in the way the trial was designed. Starflex鈥檚 manufacturer NMT Medical sued, in a case that helped to spur the ongoing campaign to reform UK libel law.
Fortunately for Wilmshurst, the company went bankrupt before the case got to court. For him, the debacle highlights an important principle: 鈥淲e need to take into account both the benefit of closure and the risk,鈥� he says. 鈥淥ften there is little or no benefit and a significant risk.鈥�
After all, no surgical procedure is completely safe. During PFO closure there is an estimated 1 in 100 chance of a complication. That could mean anything from bruising or damage to blood vessels or the heart, or the device falling out. There is even a risk that the operation could trigger another stroke, if a blood clot or air bubble travels through the PFO as the device is pushed through.
鈥淣o operation is completely safe. There is even a risk that it could trigger another stroke as the hole is forced open鈥�
Off label
The migraine trial also highlights wider problems within the medical device industry. Unlike drug companies, device manufacturers do not have to prove their products are effective before marketing them in Europe 鈥� merely that they pass basic safety tests.
In the US, medical devices are regulated similarly to drugs, yet are commonly used 鈥渙ff label鈥�, without having gone through the official approval process for that particular kind of operation. And doctors in the US tend to be more gung ho about medical interventions than their UK counterparts in general, which could help explain why the US carries out some 8000 PFO closures a year, more than 10 times the UK鈥檚 figure despite having only five times the population.
The best way to resolve the uncertainties over the procedure鈥檚 risks and benefits is with further randomised trials. At the time Lydia collapsed in her kitchen, back in January, only one trial in stroke patients had been published, involving Starflex. That had found no significant difference between , although there were criticisms of the study鈥檚 design that meant of intervention.
Lydia鈥檚 neurologist told her that he was eagerly awaiting the results of trials using other devices, but in the meantime he recommended that she have the operation. 鈥淚 didn鈥檛 know what to think,鈥� she says.
Since then the results of two further trials have been published, but they haven鈥檛 helped. Even the most favourable could be taken as evidence both in support and against the procedure, . 鈥淚f the PFO closure device were a drug we would just walk away,鈥� says neurologist Clay Johnston of the University of California, San Francisco.
Doctors have voiced frustrations that bigger trials were not set up from the start 鈥� there would by now be data from thousands of operations, rather than hundreds. 鈥淲e would be in a better position to judge whether PFO closure is the correct decision and if so in which patients,鈥� says Brian Clapp, a cardiologist at Guy鈥檚 and St Thomas鈥� Hospital.
It will be a few more years before the results of two further trials are in. In the meantime, all doctors can do is try to ensure that the procedure is only performed on those who most need it, by questioning patients closely to see if they have any other risk factors that could have caused their stroke. 鈥淚t takes me two or three times as long to decide whether to close a PFO as it does to perform the procedure,鈥� says Wilmshurst.
In the UK, official guidelines now say that the operation should take place only at a few specialist centres that keep good records, so all data can be pooled. 鈥淲hile it won鈥檛 be a randomised trial, it has the potential to tell us more about the sort of individuals who are having the procedure and their outcomes,鈥� says Rob Henderson, a cardiologist at Nottingham University Hospitals.
Lydia certainly would have welcomed more information. But in the end she decided to have the operation. 鈥淚 was scared witless, so I felt I had no choice but to go ahead and get it closed,鈥� she says. Fortunately in her case there were no complications and the operation was a success. Lydia now goes for daily 5-kilometre walks along the beach without problems, and is convinced she did the right thing.
But one case cannot teach us any broader lessons 鈥� for that we need definitive evidence from bigger and better trials. 鈥淲e don鈥檛 want to start closing PFOs simply because they are there,鈥� says Clapp. 鈥淚 think it is generally better to be thoughtful about this.鈥�
This article appeared in print under the headline 鈥淗eart-stopping鈥�
