IN THE 1890s one of the biggest pandemics in history, known at the time as “Russian flu”, swept the world. It left 1m people dead. Russian flu is now thought to have been misnamed. It was probably not influenza, but rather a coronavirus ancestral to one that now just causes symptoms described by sufferers as “a cold”. When it was new, however, few people had immunity to it, so it was often lethal. And not only that. For, as the pandemic receded, it left in its wake a wave of nervous disorders. A similar wave followed the next big pandemic, the “Spanish” flu of 1918 (which, though nothing much to do with Spain, really was influenza). One common symptom was lethargy so bad that in Tanganyika (modern-day Tanzania) it helped cause a famine because so many people were too debilitated to pick the harvest.
Something similar is happening now, with the covid-19 pandemic. A wave of what has become known as “long covid” is emerging in countries where acute cases have been falling. Formally, the condition is called “post-covid syndrome” (PCS). But even the official definition of its symptoms is fluid, because knowledge of its details is still evolving. Britain’s National Institute for Health and Care Excellence, for example, defines PCS as “signs and symptoms that develop during or after an infection consistent with covid-19, continue for more than 12 weeks and are not explained by an alternative diagnosis”. It does not, though, specify a list of such symptoms.
There are, indeed, many of them. A survey of almost 3,800 people around the world reported 205. A sufferer typically has several at a time, with the most debilitating usually being one of three: severe breathlessness, fatigue or “brain fog”.
Britain’s Office for National Statistics (ONS) estimates that 14% of people who have tested positive for covid-19 have symptoms which subsequently linger for more than three months (see chart 1). In more than 90% of those cases the original symptoms were not severe enough to warrant admission to hospital. According to the ONS, in the four weeks from February 6th nearly half a million people in Britain reported they had had long covid for more than six months—and this will not include any of those infected towards the end of 2020 in the country’s second wave.
At the time when the ONS collected those data, at least 1.1% of Britain’s population, including 1.5% of working-age adults, reported symptoms dragging on for three months or longer. Multiply that by the hundreds of millions around the world who have been infected at some point by SARS–CoV-2, the virus that causes covid-19, and a public-health catastrophe may be in the making. In the short term, it was only right that effort focused on dealing with the acute disease. Today covid-19’s chronic after-effects also need to be considered.
It ain’t over ’til it’s over
Not all of the suffering badged as long covid is actually caused by SARS–CoV-2. Even before the virus came along lots of young and healthy people would develop similarly debilitating symptoms for medically unexplained reasons. The classic example of such a mystery illness is chronic-fatigue syndrome (CFS), which often seems to follow a viral or bacterial infection. Chronic migraines and other symptoms often seen in long covid would, in normal years, also strike lots of people out of the blue. The data do, nevertheless, suggest that the effects of long covid are swamping this symptomatic background. Researchers in Britain compared the persistence of a dozen typical long-covid symptoms in nearly 22,000 people who had tested positive for SARS–CoV-2 with the rates of these symptoms in a similar group with no record or likelihood of having been infected. In both, many people got better as time passed (see chart 2). But after 12 weeks the rate of symptoms in the covid-19 group was eight times higher than in the uninfected group.
Who should be diagnosed with PCS is still being worked out. Many of those with long-covid symptoms have tested positive neither for SARS–CoV-2 nor for antibodies against it—perhaps because tests were not available when they were ill or those tests were not sensitive enough to pick up the relevant antibodies before they disappeared (a problem with several of the first generation of antibody tests). Studies comparing symptomatic individuals with and without a positive viral or antibody test generally find the same patterns of symptoms in both. Yet many doctors are brushing off individuals with no laboratory proof of past infection.
Those showing up at long-covid clinics in America and Europe are predominantly middle-aged and mostly women. Ethnic minorities are under-represented, even though they have higher acute infection rates. Many doctors suspect this is because white people in these parts of the world are often in a better position than others to seek care, and are more demanding about doing so. Some see parallels with CFS, known sceptically in the past as “yuppie flu” because of the demographic profile of those who spoke out about it.
A study by King’s College London found the median age of those with self-reported long covid to be 45, echoing the message of the clinic-attendance data. But the ONS found, contrary to what those data seem to suggest, that women were only slightly more likely than men to develop the condition—though it is unclear whether the types of symptoms experienced by women may be more debilitating.
Broadly speaking, there are three types of long-covid patients, says Avindra Nath of America’s National Institutes of Health. The first are characterised by “exercise intolerance”, meaning they feel out of breath and exhausted from even small tasks involving physical activity. The second are characterised by cognitive complaints in the form of brain fog and memory problems. The third are characterised by problems with the autonomic nervous system, a set of nerves that control things like heartbeat, breathing and digestion. Patients in this group suffer from symptoms such as heart palpitations and dizziness.
Impairments of the autonomic nervous system are known as dysautonomia, an umbrella term for a variety of syndromes. Igor Koralnik of Northwestern Memorial Hospital, in Chicago, who has been treating long-covid patients with neurological symptoms, says there has been a marked increase in dysautonomia since the pandemic began. David Putrino, director of rehabilitation innovation at Mount Sinai Hospital, in New York, says that roughly 80% of people who show up at his long-covid clinic have symptoms that are “dysautonomia-like”, regardless of the underlying cause. “And by far these symptoms are the most debilitating, so if we rehabilitate them we can often make the biggest impact in people’s lives.”
Based on these patterns of symptoms, and various laboratory tests of long-covid patients, doctors are focusing on three possible biological explanations. One is that long covid is a persistent viral infection. A second is that it is an autoimmune disorder. The third is that it is a consequence of tissue damage caused by inflammation during the initial, acute infection.
According to the first of these hypotheses, some patients never clear the virus completely. They are not infectious, says Dr Nath, so it could be that they harbour some altered form of the pathogen which is not replicating and is thus undetectable by the standard test for SARS–CoV-2, but is nevertheless making some viral product that their bodies are trying to fight off. This sort of thing is known to occur with other viruses, including measles, dengue and Ebola. RNA viruses, of which SARS–CoV-2 is an example, are particularly prone to this phenomenon, says Dr Nath.
Proof of this hypothesis is lacking, but there are pertinent clues. Researchers are looking for SARS–CoV-2 or its products in all sorts of fluids and tissues from people with prior infection. There is already evidence that the virus can persist in the body, though the data are predominantly from those who did not develop long covid. A study published recently in Nature showed that some people had traces of SARS–CoV-2 proteins in their intestines four months after they had recovered from acute covid-19. Viral products from SARS–CoV-2 have also been found in people’s urine several months after their recovery. Dr Putrino says viral material has been detected in stool samples from some patients in his long-covid clinic, but not all.
The second hypothesised mechanism for long covid, that it is an autoimmune disease, holds that the virus, though gone, has caused something to go awry with the immune system—which now attacks some of the body’s own tissues. A growing body of evidence backs this idea, too.
The immune system is a complex machine, with many cellular and molecular components, any of which might break and cause symptoms. Some of those suffering from long covid have badly behaving macrophages, the cells responsible for detecting and engulfing harmful invaders. Others exhibit abnormal activation of their B-cells—white blood cells which churn out custom-made antibodies to gum up specific pathogens. In these cases, their B-cells seem to make an unusual quantity and variety of “auto-antibodies”, which attack the body’s own cells instead of invaders. Others still have low levels of interferons, a group of molecules involved in fighting off viral infections. And some have problems with their T-cells, which are parts of the immune system that have the jobs of destroying infected cells and alerting B-cells to the presence of pathogens, so that appropriate antibodies can be made.
Several studies have found reduced T-cell counts in people who have had acute covid-19, and also that their surviving T-cells are “exhausted”—meaning they mount only a weak response to infections. Laboratory studies by Dr Koralnik’s team have found that long-covid patients with brain fog have different T-cell responses from those of people who were once infected but are now asymptomatic.
All of this suggests that some individuals cannot fight the virus off completely, or that parts of their immune systems act in ways that may be detrimental to their bodies. Some doctors think people who are already vulnerable to developing an autoimmune condition are pushed further in that direction by the stress which covid-19 puts on their bodies. Such disorders are typically diagnosed in middle age, which is consistent with the age-peak found by King’s College, and are more common in women—as is, albeit to a lesser extent, long covid.
The third hypothesis about the cause of long covid, inflammation, holds that the fight put up by the body against the acute illness causes irreparable collateral damage. This often happens during a viral infection, but it could be particularly likely with covid-19. Out-of-control inflammation, caused by cytokines (molecules that drum up inflammation) is a hallmark of the illness.
One guess is that the inflammation which happens when people are ill somehow damages parts of their autonomic nervous systems. Another suggestion, made by Dr Koralnik, is that in some patients SARS–CoV-2 may damage the cells that line blood vessels, either by infecting them directly or via inflammation. This would change the way blood flows to the brain, and may thus explain the brain fog.
Whys and wherefores
Studies intended to investigate each of these possibilities are under way. But the three theories are not mutually exclusive. Indeed, most researchers agree that long covid is probably a term which embraces several conditions with different causes.
Determining these will help both with the development of treatments and with their prescription. If persistent viral infection turns out to be a cause, the search will be on for suitable antiviral drugs. Treatment would consist either of a defined course of medication that clears the virus completely (as is now possible for hepatitis C, for example) or of drugs that people take routinely to keep the virus at bay, the approach taken with HIV/AIDS.
Treatments for immune disorders already exist, and some may work for long covid. “As soon as we define the immune abnormality in these patients, then it will become very clear how to treat them,” says Dr Nath. “It is quite possible that we may need multiple treatments for different types of immune response—and we should be able to figure that out as well.”
Some of those with long covid have felt dramatically better after a covid-19 vaccination. But the relief tends to be temporary. Doctors have seen this before. People with CFS, for example, sometimes feel temporarily better after a flu shot or other vaccination. Nobody knows why. One possibility is that the revved-up immune system alleviates their symptoms for a time. A placebo effect may also be involved. Akiko Iwasaki, an immunologist at Yale University, has proposed clinical trials of covid-19 vaccines for long covid. She argues that seeing which work, even if only for a short time, may unmask the specific immune abnormality involved—and show what sorts of drugs could work as well.
At the moment, the only treatment is rehabilitation. To design protocols for long covid, Dr Putrino’s team have been working with experts on disorders with similar symptoms, including dysautonomia, CFS and Lyme disease. “We’ve tried to be extremely symptom-centric,” he says. “We try to dig through a person’s life and understand what is causing the biggest triggers [of symptoms] that interfere most with their daily life.”
He describes some typical examples. Many patients come in having lost a lot of weight because if they have a full meal “their symptoms just wash over them and that’s it for the day.” That is common in dysautonomia, whereby stretching of the stomach causes an autonomic-nervous-system reaction. These patients are advised by nutritionists on how to eat smaller, nutritious meals and to find out what foods are easiest on them. Some patients experience a drop in blood pressure when they move about, and feel dizzy—another hallmark of dysautonomia. Simply wearing compressive stockings to prevent blood pooling in their legs can help these people a lot. So can avoiding going outdoors in hot and humid weather. Those with extreme fatigue are taught how to watch for “energy windows”, in which to do the most important tasks of the day.
Dr Putrino’s team have identified another common problem. They tested 25 of their long-covid patients and found that all had carbon-dioxide levels which were too low. This may sound surprising, given that CO2 is a waste product derived from respiration, and is harmful if present in too high a concentration. But it also helps regulate acidity, and incorrect acidity can disrupt all sorts of metabolic processes. Low CO2 levels are also often seen in dysautonomia and CFS. The solution is breathing exercises to help with CO2 retention. (Elsewhere, opera singers are teaching long-covid patients helpful breathing techniques.)
At Dr Koralnik’s neurology clinic, the approach is similar. Long-covid patients are first assessed to see whether their specific problem is memory, attention, fluency in word finding, “or whatever they may have that could be different than someone else who also has brain fog”. Cognitive rehabilitation is then tailored to their needs.
It is painstaking work. After an average of 150 days of rehab, which includes two half-hour sessions each week with a therapist, plus remote follow-up, Dr Putrino’s patients report a 30-40% improvement in fatigue levels. Such improvement was not seen in comparable patients who were not undergoing rehab, so his team are confident that the effect is real. But out of about 100 patients whose outcomes are being monitored for research purposes, only three say they have recovered fully.
This means that even with appropriate health care, many of those who have long covid will continue to struggle in their daily lives. A survey in Britain, albeit of a self-selected group of people who responded, found that the illness affected the ability to work of 80% of those suffering from it, and about 40% said it affected their ability to care for others (see chart 3).
All this suggests that, even when the pandemic of acute covid-19 has been dealt with, a big problem will remain. Post-viral syndromes on this scale affect not only those who are experiencing them directly. They also have serious consequences for everyone else. ■
All our stories relating to the pandemic and the vaccines can be found on our coronavirus hub. You can listen to The Jab, our podcast on the race between injections and infections, and find trackers showing the global roll-out of vaccines, excess deaths by country and the virus’s spread across Europe and America. And you can also sign up to our new weekly newsletter, “Simply Science”, which showcases the best of The Economist’s reporting and analysis on science.
A version of this article was published online on April 28th, 2021
This article appeared in the Science & technology section of the print edition under the headline “The sting in the tail”