Most doctors were expecting Covid-19 to resemble a severe lung disease. What they encountered was a virus that wreaked devastation on the entire body
When the pandemic reached London, Simon Ashworth thought he knew what to expect. An intensive care specialist since 1996, Ashworth has worked as a consultant at St. Mary’s Hospital in Paddington for nearly 20 years. He had seen reports from China and Italy and was preparing himself to treat patients with such severe Covid-19 that they had developed a lethal complication: acute respiratory distress syndrome (ARDS), a condition where the lungs grow so inflamed that vital organs are slowly starved of oxygen. Of those that develop ARDS, around a third die.
Ashworth was used to treating these patients in his ICU. He explains that the microscopic air sacs in their lungs fill up with “cheese” – white blood cells, mucus and bits of dead lung – and they become stiff and fibrous. The patient begins to drown. To keep them breathing, he snakes a tube into their airway and pumps air into their lungs mechanically. But ventilating patients with ARDS is a high wire act. Too little pressure, and the lungs won’t inflate; too much, and they can become even more damaged.
This is what Ashworth was expecting when he intubated his first Covid-19 patients. Yet to his surprise, their lungs weren’t like those of ARDS patients at all. They stretched easily. They didn’t require the high pressure needed to inflate a typical ARDS lung. “They could get stiff lungs,” he says. “It’s the final common pathway as they get worse and worse and worse – the lungs get consolidated and damaged and inflamed. But that wasn’t where they were starting from.”
Even more strangely, while a typical ARDS patient may be unconscious or struggling to breathe, the Covid-19 patients on Ashworth’s ward often seem comparatively normal. Doctors from around the world have reported patients who arrive at hospital breathing fine and holding conversations normally despite having catastrophically low levels of oxygen in their red blood cells, a phenomenon known as silent hypoxia.
Covid-19 has confounded the expectations of doctors. Patients suffer from a bewildering variety of complications. They urinate blood, complain of heartburn and lose their sense of smell and taste. A 56-year-old man in a Beijing Hospital developed brain inflammation; his face began to twitch and he hiccuped uncontrollably. A 71-year-old woman, returning to the US from Egypt, developed back pain, vomiting and bloody diarrhoea. Only on her fourth day in hospital did she begin to cough, and was subsequently found to have Covid-19. A neurosurgeon at Mount Sinai Health System in New York treated five patients with sudden strokes. All were under 50, and had either mild symptoms of Covid-19 or no symptoms at all.
Ashworth has seen bodies laced with blood clots and patients hit with heart attacks, strokes and kidney failure. “I think that in the scientific and political discourse that took place, there was a failure to recognise that we just don’t know very much about this disease,” says Ashworth. “Everybody said, ‘well this is just ARDS and it’s viral’ and actually, what we’ve got is a disease, which if four million people didn’t have it, it would be fascinating. It is fascinating. It’s a fascinating disease. It’s creating avenues for research that will probably help us treat things like flu. It does all sorts of things that we’ve never really thought about before.”
For most people, Covid-19 is mild – a slight fever and dry cough. But for a small number of patients who become critically ill – around six per cent of confirmed cases – Covid 19 warps into a frighteningly lethal condition. In the UK, a third of patients taken to hospital with the disease end up being killed by it.
For the doctors who treat it at this advanced stage, Covid-19 is an enigma. Not only does it damage the lungs in unexpected ways, but it may also invade the heart, gut, blood vessels, kidneys and brain. “We have come back to the virus as the starting point and recognize that it triggers a whole series of different processes,” says Ashworth. HIV is the archetype of a complex disease, he says. “Covid-19 is not far off it.”
When the outbreak began, Ali Rezaei Haddad was nearing the end of his eight-month placement in neurosurgical intensive care at St George’s Hospital in Tooting, London. Like many younger doctors, he agreed to stay on, and in the space of a day his ICU was emptied in anticipation of a wave of Covid-19 patients. Although St George’s can now take up to 500 Covid-19 patients, the lockdown has meant that it has never reached that number. Haddad has now been treating Covid-19 patients for more than a month. He has found that the lungs were not the only organ in distress. “In the acute phase, at least, your body’s not responding the way it does to other lung pathologies,” he says.
Almost immediately, his team had to grapple with kidney failure. According to a preliminary study out of Wuhan, 20 per cent of patients suffer this complication. More than half of the people hospitalised because of Covid-19 have blood or protein in their urine, indicating kidney damage. People with acute kidney injury may be five times as likely to die from Covid-19 than patients without it. And while intubating a patient can damage their kidneys on its own, Covid-19 seems to do more than that.
“There’s a theory out there that the virus directly affects the kidneys, but we don’t have direct evidence for that,” says Haddad. Regardless, the team had to adapt. In the first few days, they tried to ‘dry out’ their patients. In crude terms, the patient must urinate more than they drink. Lungs drained of fluid are typically easier to ventilate, so the patients receive diuretics. But weak kidneys made this treatment dangerous, and Haddad’s team had to abandon it.
This kidney dysfunction led to more discoveries. Haddad had to hook up his patients to dialysis machines – a doctor inserts a line into a vein and a machine does the kidney’s work of purifying the blood. But the machines kept clotting up – during filtration any clot means the filter and system must be replaced. Simultaneously, patients were giving out an abnormally high reading of D-dimer, a protein fragment in the blood left behind when a clot dissolves. Haddad’s team was suspicious. They scanned the patients and discovered a now well-known indicator of Covid-19 – lungs and veins laced with blood clots.
This pattern has repeated across the world. There have been reports of purple rashes, swollen legs and clogged catheters. Studies from the Netherlands and France suggest that clots appear in 20 to 30 per cent of critically ill Covid-19 patients. A group in Hamburg carried out autopsies on 12 Covid-19 patients and found evidence for blood clotting problems, including clots in the lungs and the legs. In all 12 cases, the cause of death was found within the lungs or the heart.
One way to respond to this threat is to pump patients full of blood-thinning medication. “My career has been short-lived, but I haven’t seen anything like this. I don’t think many people have,” Haddad says. A review of records of more than 2,700 patients, published in the Journal of the American College of Cardiology, indicates this treatment may help the most seriously ill.
Many patients still don’t recover. A team from UCL Queen Square Institute of Neurology has shown that Covid-19 may form clots within the brain, leading to strokes. Clots in the lungs make ventilators less useful – reduced blood flow into the lungs means oxygen from the ventilator can’t make its way into the bloodstream.
Faced with these puzzling symptoms, doctors must turn to unusual forms of treatment. Flipping Covid-19 patients onto their stomachs – a technique known as proning – appears to ease chest congestion. The logistics of this practise are intense; it takes a whole team to flip a patient, who is usually unconscious and hooked to life support machines. “What’s interesting about Covid-19 is that patients seem to derive enormously more benefit than I’ve ever seen in any other population,” says Ashworth. “The changes it causes are quite dramatic.”
Damage to one area of the body can lead to damage in another. On top of the collateral damage a clotted-lung does to the heart – it must beat with self-destructive force to deliver its blood – case studies show clots in the coronary artery. A paper in JAMA Cardiology documented heart damage in nearly 20 per cent of hospitalised patients. Another Wuhan study found 44 per cent of patients admitted to the ICU had irregular heartbeats.
This might explain why people with heart disease die so frequently from Covid-19. Heart disease was found to be the most common pre-existing health condition in people who died from Covid-19 during March in England and Wales, according to data from the Office of National Statistics. “It’s all quite unclear at this point,” says Graham Cooke, professor of infectious diseases at Imperial College London. “We see that people with ischemic heart disease have a higher mortality rate when they get Covid-19 – it’s possible that part of that reason may be because they have small clots in the coronary arteries that are causing death.”
In an extensive analysis he conducted on patients, Cooke was surprised at how few had chronic lung disease, despite initial warnings that asthmatics were at higher risk of severe illness. In a similar analysis, asthma showed up in only about five per cent of New York State’s fatal Covid-19 cases. This might be because this group has self-isolated more diligently. “But it might be because they don’t have an underlying problem with their arteries,” says Cooke.
Covid-19 lays waste to its victims’ organs, but sometimes the worst damage is done by their own bodies. A “cytokine storm” is an immune system overreaction. The storm takes its name from the protein released by the body to coordinate its response to an infection. Traces of these “cytokines” can be measured in the blood after the fact. When a pathogen like Covid-19 enters the body, cytokines attract immune cells to the point of infection to destroy the intruder. This response is essential – it brings on inflammation, the process by which the body heals.
Yet for some, this response grows out of control; excess cytokines swarm on the location, and excess immune cells follow. The body is destroyed along with the virus. Patients get better, says Haddad, then their blood pressure drops and they die. This includes young people, mostly men, disproportionality from BAME backgrounds. “You start to get 35-year-olds and 40-year-olds dying – that was something we weren’t used to,” he says. “Next to all the PPE piles, we had death certificates ready to go.” Wrecked organs and excess clotting could all theoretically be explained by this response.
Cytokine storms are not new. The H5N1 influenza virus, for instance, also had a high fatality rate linked to an out-of-control cytokine response. And the storms may also be one reason why young people may have less severe infections – their immune systems are less developed and so produce less cytokines. “A cytokine storm has been understood for some decades as an unfortunate but potential response to severe infection – the determining factors are not well characterised,” says John Wilson, president-elect of the Royal Australasian College of Physicians and a respiratory physician. “Why it occurs in coronavirus infection is a question that is still being investigated. And whether or not the inflammatory effect can be dealt with by drugs is also still under investigation.”
This reaction can potentially explain the damage to multiple organs. But it isn’t the whole story, says Ashworth. “I still suspect that this isn’t just a cytokine storm,” he says. “I’m sure some of them have pulmonary emboli blood clots going to the lung, or bacterial infections – I suspect it covers several different things that are potentially going on. Again, I hate saying it, but we don’t properly understand it.”
Another reason for Covid-19’s wide devastation may lie in the virus that causes the disease: Sars-CoV-2. We know that the virus enters human cells by binding to a particular receptor called the Ace-2 – usually responsible for regulating blood pressure – then replicating itself. These receptors are common in the upper respiratory tract, where the virus first makes it home, but also found in the lungs, heart, kidney, and intestine. The virus may migrate from this initial binding to these organs’ receptors – a study in The Lancet showed that the endothelial cells that line the blood vessels, which also have Ace-2 receptors, showed evidence of an attack. This process could explain excess clotting, and why patients with hypertension and diabetes are at such risk.
The way these receptors differ may even account for the body’s wide range of reactions to Covid-19. “The expression of Ace-2 in different tissues is variable and may be dependent on the individual,” says Wilson. “It may also depend on the particular underlying condition of that organ.”
Ace-2 receptors may help the virus reach as far as the brain. Traces of the Sars-Cov-2 virus have been detected in patient’s spinal fluid. In a study of 214 patients with coronavirus, neurologic symptoms were seen in 36.4 per cent of patients and were even more common in those with severe infection. Some of Covid-19’s more bizarre symptoms like strokes, seizures, loss of taste and smell, erratic breathing or heart rhythm and Guillain Barre syndrome – where the body’s immune system attacks the nerves – could be explained by the virus attacking the brain. In a review article published by Jilin University in China, researchers propose that if Sars-CoV-2 infects nerve cells, particularly neurons in the medulla oblongata – the brain stem that controls the heart and the lungs – the damage could contribute to “acute respiratory failure of patients with Covid-19.” Some of Haddad’s patients, when they came round from the sedative that lets them tolerate intubation, were still unable to breathe on their own. As with Sars, which could cause brain inflammation, Haddad is convinced that the virus invades the brain’s Ace-2 receptors, knocking out its breathing centres.
Cooke is cautious about this relationship. “I haven’t seen enough data to comment on that,” he says. “There is some suggestion, I think it’s unlikely, but there’s some suggestion that the virus could infect the part of the brain that detects that you have low oxygen, and therefore, that’s why they’re not responding. But I wouldn’t put too much weight on that.”
The bodies of those killed by Covid-19 could yield vital clues about why the disease is so ferocious. Michael Osborn has been carrying out post-mortems during the pandemic. Postmortems are not usually conducted when the cause of death is known, but Osborn and his team at the Imperial College Trust in London are attempting to model how Covid-19 kills, and in time find the best treatments. “It is very worthwhile doing a select number of consented post-mortems where lots of tissue can be taken and sent and examined in great detail,” he says. There is still a lot to be discovered. He is carrying out examinations of the brain stem respiratory sensor, to see if it might play a part in patients’ breathing problems, and investigating the “activation of complement” in the immune system – the defence mechanism that controls clotting.
While he acknowledges that Covid-19 behaves bizarrely from a clinical standpoint, from his perspective after the disease has wreaked its damage, it is similar to most severe respiratory infections. The significant transformation takes place within the lungs. This isn’t surprising, he explains, because, in death, there’s only a limited number of ways the body can react. “What we’re not doing is looking at the lungs and seeing a change that we’ve never seen before and has never been described,” he says. “What we’re seeing is a selection of changes that you see in a variety of other diseases.”
Osborn thinks that the complex impact that Covid-19 has on people may have less to do with the virus itself and more to do with our genetics.
“The way you react to colon cancer, you’d react differently to me, because you’re a different person to me,” he says. The consumer genetics giant 23andMe plans to expand a study it launched last month, analysing hospital data to try to identify genetic differences that may help this confounding array of symptoms. But at this point why Covid-19 patients fall into such diverse clinical groups is unclear, he says. “It is likely to be idiosyncrasies of the patient, their genetic makeup, maybe other comorbidities, and also the severity of the illness and so forth,” he says.
For Osborn, Covid-19 has differed from other respiratory infections in one frightening way – its severity. In a less severe infection like pneumonia, the lung might be patchy, showing the telltale ‘ground glass opacities’ revealed on X-rays while the patient is alive – white smears in the lung scan where black should be – “cheese” instead of air. Instead, Covid-19 destroys lungs, transforming them into something closer to the consistency of the liver.
This severity reflects a blunt truth – we do not have a treatment for Covid-19 yet. Doctors can support failing organs with ventilators and dialysis machines, but against the virus itself they are empty-handed. “You’ve got a situation where this virus gets a free run at you until your immune system works out how to deal with it,” says Ashworth. And part of how it deals with it is it to kill the cells which have been infected by Covid-19, which are part of your body.”
Hopes for a Covid-19 treatment fall into two camps – antivirals and anti-inflammatories. The latter quell the body’s inflammatory response; the former tackle the virus itself. Amongst antivirals, the leading option is remdesivir, developed by the pharmaceutical giant Gilead. Results so far have been mixed, for both a National Institute of Allergy and Infectious Diseases trial and a trial conducted in China. “That’s not surprising, because by the point where people have got severe disease, it’s not really the virus driving their illness,” says Cooke. “The virus is probably at quite low levels by that point – it’s the hyper inflammation driving their illness. So we shouldn’t be surprised that an antiviral drug wouldn’t be very effective by the time people are in intensive care.” If we could deliver the drug earlier, it might have a more positive response, but remdesivir has to be injected directly into the vein, making its early delivery difficult.
The other group of drugs target inflammation – they halt the hyperinflammatory state and reduce its damage. According to a recent study in the British medical journal The Lancet, anti-inflammatory drugs may help quell this response. Probably most high profile of these has been tocilizumab, says Cooke, which is typically used to treat rheumatoid arthritis. A trial, featuring nearly every NHS trust in the UK, was carried out last month. These drugs may halt the progression to a cytokine storm. Any successful drug will help us flatten the number of patients who reach a critical level. “It may be that you can treat people with these drugs around day seven of their illness, to prevent them getting sick enough to need to go to ICU,” says Cooke.
But until we have a treatment, doctors and nurses must find other ways to manage the brutal complexity of Covid-19. “It’s so important that we have the time to conduct trials, work out what works and out what does harm,” says Ashworth. “That can have a quite big impact on the mortality and the consequences of this illness going forward.” Until then, Covid-19 will remain an enemy we do not fully understand.