TOM never expected it to happen to him. He was just 40, with a young family and a good job. He thought he was in good shape too – he played a lot of football, didn’t drink that much, never smoked. And after a medical at work just a year or two back, he had been given the all-clear. His cholesterol levels were normal, he’d been told. So why was he now in intensive care, recovering from a massive heart attack?
Tom’s case is not unusual. About half of all men who have a heart attack seem to have normal amounts of cholesterol in their blood. And over the past decade, other evidence has accumulated to suggest that when it comes to cardiovascular disease, cholesterol is only half the story. Two months ago, results were announced from a huge study that confirms these suspicions: all this time we’ve been overlooking a major cause of heart attacks.
The culprit is one of the most familiar of everyday physiological phenomena, seen in innumerable major and minor illnesses. It’s inflammation – the same process that makes the skin around a pimple red and sore, or a sprained ankle swollen. So how could such a ubiquitous healing response be responsible for the biggest cause of death in many Western nations? The explanation, it turns out, is leading us to rewrite medical textbooks. And with some major new guidelines for doctors due out in the next few weeks, we could have to rethink the way we treat patients.
Advertisement
The conventional view of cardiovascular disease is that it is largely a problem of plumbing. During the 1980s, researchers started to realise that if there is too much cholesterol in the bloodstream, it can build up inside the blood vessel walls in the form of plaques – a process known as atherosclerosis. The plaques can restrict blood flow to the heart, causing the bouts of pain experienced in angina. If the plaque bursts and its contents spill out, clots may form and travel around the body in the bloodstream. Clots that block up the small blood vessels feeding the heart give you a heart attack. If they occlude blood vessels to the brain, you suffer a stroke.
But since the late 1980s there have been several findings that suggest this is not the whole story. The biggest hint is that only half of heart attack victims have blood concentrations of cholesterol over 5.2 millimoles per litre (200 milligrams per decilitre), the level usually seen as a safe limit. “It became apparent that many cases were not really cholesterol-related,” says David Grainger, an expert on inflammation at the University of Cambridge.
Another vital clue came from the fact that statins – drugs that lower blood cholesterol levels – seem to work better than they should. After a few years of widespread use, the figures suggested these drugs were preventing heart attacks by some other way too. Some people who started out with normal cholesterol levels seemed to benefit from the drugs. Scientists started to suspect that statins have a dual role: they can lower cholesterol, and they have some other biological effect that also protects the arteries.
Meanwhile, interest had been growing in a controversial research finding that seemed to turn established notions about heart disease on their head. Finnish researchers announced in the late 1980s that people who have heart attacks are more likely to carry antibodies to the bacterium Chlamydia pneumoniae (Â鶹´«Ă˝, 8 June 1996, p 38). At first the findings were dismissed as erroneous; everyone knew you couldn’t catch a heart attack. But during the 1990s, the evidence mounted. Atherosclerotic plaques were found to contain C. pneumoniae DNA and proteins, and finally live bacteria themselves. Doctors seriously wondered if they might be able to cure atherosclerosis with an antibiotic.
But as more evidence came in, the picture got more complicated. Researchers started to suggest that other infections, viral as well as bacterial, could increase a person’s risk of heart disease. Helicobacter pylori, the bacterium that causes stomach ulcers, was implicated, as were several bugs that cause gum disease. More recently the viruses cytomegalovirus, herpes simplex and hepatitis A have been added to the list. And when scientists started giving antibiotics to people with cardiovascular disease, they got some very strange results.
Take a recent study of 325 heart disease patients by Juan-Carlos Kaski’s team at St George’s Hospital Medical School in London. Antibiotics cut the incidence of further angina attacks by over a third – but this benefit occurred in all patients, even those without antibodies for C. pneumoniae or H. pylori. Perhaps the antibiotics chosen are working against other infections, even ones we haven’t thought to blame yet, says Kaski. Or perhaps, as with statins, the antibiotics are helping the heart by an entirely separate process.
This other process seems to be the key to explaining all these messy findings. Recent animal studies by Grainger, Peter Libby at Harvard Medical School and Alain Tedgui at INSERM in Paris have helped put the pieces together to form a unified theory of cardiovascular disease. It explains all of the anomalies, as well as the role of cholesterol and the well-known risk factors such as obesity, smoking, poor diet and diabetes. The theory says that the cornerstone of the atherosclerotic process is inflammation.
We are all familiar with this process as a local reaction to a wound, irritant or infection. Extra blood is pumped to a trouble spot in the body so that circulating immune cells can reach the affected tissues, ready to do battle with bacteria. The local blood vessel walls become lined with molecules that are “sticky”, so that immune cells will adhere to them. Then the immune cells squeeze through into the surrounding tissue, where they do their work. These events are orchestrated by numerous chemical signalling molecules, some of which promote the process of inflammation, and some of which are antagonistic to it. Their balance dictates when inflammation should flare up and when it should subside.
The reworked theory of heart disease says that the whole process of atherosclerosis is actually a runaway inflammatory reaction. It may begin when parts of the blood vessel wall, particularly at junctions, become mildly damaged by constant turbulent flow (Â鶹´«Ă˝, 6 February 1999, p 32). It may be that natural anti-inflammatory safeguards that normally protect our blood vessels fail to work properly. Or it may be an inevitable side effect of inflammation aimed at infections, irritating the blood vessel walls as it protects us. Either way, the endothelial cells that line the vessel walls start producing different surface molecules – the ones that make them sticky.
Immune cells called macrophages and T cells adhere to the vessel walls, squeeze through them, and lodge between the endothelial lining cells and the adjacent muscle cell layer (see Diagram). Chemicals that normally promote inflammation attract vessel wall muscle cells into the developing plaque. Other inflammatory signals encourage the macrophages and muscle cells to scavenge cholesterol from the blood and so develop into a new type of cell called foam cells. And as the atherosclerosis progresses, inflammation continues to play a role. It’s now thought that it is highly inflamed plaques that tend to finally give way, producing the fatal clots that cause heart attacks or strokes.
So have we been giving people the wrong health advice all these years? Thankfully not. The inflammatory model actually fits extremely well with the cardiovascular risk factors we already know about. High blood pressure and free radicals in the blood from smoking put endothelial cells under stress, making them more prone to damage. Fat cells themselves produce inflammatory chemicals, and people with diabetes suffer chronic inflammation. And as the immune and muscle cells must take up lipids to become foam cells, there is still an important place for cholesterol in the inflammatory model – but only as half the picture.
Significantly, the inflammatory theory answers an awful lot of the puzzling questions that the simple “plumbing” theory of heart disease can’t. Why do half the people who get heart attacks have normal cholesterol levels? They could have more of the factors that promote inflammation instead. Statins work better than they should do? The drugs are now known to reduce various markers of inflammation in the blood, and it has been suggested that they have a general anti-inflammatory effect. People who have more infections have more heart attacks? Infections increase levels of pro-inflammatory chemicals in the bloodstream. Antibiotics reduce the risk of heart attacks even in people clear of infections? The antibiotics effective against C. pneumoniae also have a general anti-inflammatory effect. It’s almost too good to be true. “We’ve crossed the Rubicon in establishing a causal link,” says Grainger.
So what happens now? The most obvious consequence of discovering a new disease process is that it gives drug developers a host of new biochemical pathways on which medicines could act. They have already started trying to exploit the new knowledge. Drug companies have several dozen cardiovascular drugs in development they hope will work by dampening down the inflammatory process in the blood vessel walls.
One, Eli Lilly’s raloxifene, is already used to treat osteoporosis, and the firm is investigating whether it could also prevent heart disease, in a 10,000-patient trial due to finish in two or three years. Until the menopause, women have less heart disease than men, and it is widely believed this is something to do with the female reproductive hormone, oestrogen. Grainger says: “There’s a tantalising possibility that the anti-inflammatory and oestrogenic activity are linked.” Raloxifene has some oestrogen-like effects, and is already known to reduce markers of inflammation in the blood.
Grainger’s team is investigating other agents that either suppress pro-inflammatory chemicals or promote anti-inflammatory ones. He has identified a family of peptides and small molecules that inhibit immune cell migration, and these look promising in animal models of heart disease plus other inflammatory disorders such as asthma.
Another tactic is to target the cells of the blood vessel walls to try to stop them becoming sticky in the first place. One of the key surface molecules here is called vascular cell adhesion molecule-1 (VCAM-1). AtheroGenics, a biotech firm based in Alpharetta, Georgia, has an oral drug in clinical trials that blocks production of VCAM-1. The compound, currently named AGI-1067, is being tested as a means of preventing the re-occlusion of blood vessels after they have been cleared by surgery.
Another sticky molecule called E-selectin is also under investigation, though it is still a long way from clinical trials. But rather than attempting to prevent the body from producing E-selectin, John Hallenbeck, chief of the stroke branch at the US National Institute of Neurological Disorders and Stroke, is giving rats the substance itself. The extra supply, administered in the form of a nasal spray, somehow trains the rats’ immune system to ignore the molecule on their vessel walls.
It will be many years before some of these drugs can be used in the clinic. But perhaps we won’t have to wait that long to start putting the new theories about inflammation into practice. After all, statins, which are already used to treat or prevent heart attacks and stroke, are now known to have anti-inflammatory effects. So it may be time to reappraise just who should be receiving them.
A big step towards answering this question came in November, when a team led by Paul Ridker of the Brigham and Women’s Hospital in Boston announced the results of a 28,000-woman study (The New England Journal of Medicine, vol 347, p 1557). They had investigated the importance of a marker of inflammation in the blood called C-reactive protein (CRP), made by the liver in response to inflammation. One of its jobs is to bind to the surface of bacteria, helping flag them for destruction by macrophages.
Ridker’s group measured CRP and cholesterol levels in blood samples from the women and looked to see how well they predicted who would have heart attacks over eight years. High cholesterol raised the women’s risk of a heart attack by up to 50 per cent – but using this measure still left out of the high-risk group three-quarters of women who had an attack. High CRP, on the other hand, increased the risk by 130 per cent. And the two tests identified two quite different groups of women: not many had high CRP as well as high cholesterol.
The researchers concluded that CRP is a better predictor of risk than cholesterol, and a combination of the two tests is better still. Ridker reckons that 1 in 15 people could have high CRP and low cholesterol. At present they are not being treated and may be oblivious to their higher-than-normal risk of a heart attack. “Women who were outside current guidelines might well be at higher risk than those who are inside current guidelines,” he said at a meeting of the American Heart Association in Chicago, shortly after the work was published.
Ridker’s study triggered a spate of headlines in American newspapers about a blood test that was “better than cholesterol”. Even before the study was published, CRP testing had started to creep into some clinics, though the way it is used and interpreted is hugely variable. So the big question is: should we all be getting our CRP levels checked, just as we now monitor cholesterol, and use that information to decide who should be taking statins? Some researchers and cardiologists say we should. Others are more sceptical.
One problem the sceptics point to is that the precise relationship between CRP and inflammation is unclear. Lori Mosca, a cardiologist from Columbia University in New York City, argues that CRP is a good predictor for the likelihood of heart disease in large groups of people, but not for individuals. For a population study, small errors don’t change the picture. “But they can be disastrous at a therapeutic level,” she says. “With CRP we don’t know whether it is cause, correlate or consequence.”
The problem is that CRP levels don’t just reflect what is going on in the blood vessels. Factors as diverse as infections, colds, gum disease, cancer or exercise make them fluctuate from day to day. To get round this problem, doctors may have to check CRP levels regularly, so they could identify and ignore brief spikes.
Mosca also believes that CRP levels and inflammation would be irrelevant if we revised our cholesterol guidelines. The current norms are simply far too high, she says. If we didn’t have so much cholesterol in our blood, we could tolerate long-term inflammation.
The debate could soon be settled by a report from the American Heart Association and the US Centers for Disease Control and Prevention on the role of inflammation in heart disease. Written by 50 leading American researchers, it is due to be published in the AHA journal Circulation in the next few weeks.
On a cautionary note, however, George Mensah, head of the CDC’s cardiovascular section, says he personally wants the report to contain the familiar advice on smoking, exercise, diet and weight. “This is still the important health message to get across,” he says. “No one suggests that a pill will ever replace a healthy lifestyle.”
But whatever the report says, the pro-inflammation camp have no doubt the tide of opinion is now turning in their favour. In the same way that evidence for the dangers of cholesterol built up throughout the 1980s, the case for inflammation being a risk is now overwhelming, Grainger says. And the rethink could have an equally significant effect. “The widespread use of cholesterol-lowering drugs led to a 30 to 40 per cent reduction of heart disease across the board,” he says. “We could help another 30 to 40 per cent with anti-inflammatory medication.”
And that’s got to be worth thinking about.