When the first children of the pandemic’s newest syndrome needed paediatric ICU beds in Bergamo, Italy, Dr Lorenzo D’Antiga only had a few left.
The Papa Giovanni XXIII Hospital’s paediatric intensive care unit is one of Italy’s largest: 16 highly-equipped beds to support recovery from all types of illness and injury. Children from across the country come to Milan to see their specialists, track their improvements, or receive organ transplants.
By late March, 500 of the hospital’s nearly 1,000 regular beds were given to adults with COVID-19, and all but a couple of the paediatric ICU beds had been reassigned. The paediatric team had been caring for adults with the respiratory illness while continuing interventions on children whose appointments could not be postponed.
Dr D’Antiga, the chief paediatrician at the hospital, had been working tirelessly. He had not seen his family since the beginning of the epidemic. When two, and then three, children presented to the emergency room showing symptoms of an inflammatory syndrome that looked familiar, D’Antiga’s team put the remaining paediatric ICU beds to use immediately.
“We were expecting respiratory symptoms. We were wondering where the children were? Were they also involved [in COVID-19 cases] and to what extent? We were not thinking about Kawasaki disease at all.”
Some of the new children were experiencing a shock. They were presenting with myocarditis, endangering the circulation of their blood; they had high fevers and rashes. Some needed ventilation. But they were not sick with SARS-CoV-2, the novel coronavirus that had spread from China to Italy in February and exploded.
Only two of them tested positive for the virus, yet many had produced antibodies to fight it. Coronavirus had reached them, but they were ill with something new.
“We said, that’s quite unusual. We usually see two or three [Kawasaki disease] patients a year, now we have three in a week. Then over 10 days we’ve seen another seven … this must be related to the epidemic,” Dr D’Antiga recalled.
Kawasaki disease, named after the doctor who examined its symptoms in the 1960s in Japan, is a rare illness that mostly affects children below the age of five. Patients develop a high persistent fever, conjunctivitis and rashes. Their tongue, throat and lips redden, and their hands and feet often swell, with redness in the palms and soles. Some children develop an enlarged lymph node on one side of the neck.
The illness can lead to a coronary artery aneurysm, which is a swelling in the vessel that oxygenates the heart muscle. Kawasaki disease is reported in all races, but Asian children have the highest incidence of illness, suggesting a possible genetic susceptibility. Though frightening for parents and considered serious by physicians, most children do well with aggressive early treatment.
Dr D’Antiga and his team opened their clinical records to recall all incoming cases of Kawasaki disease over the last five years, and compared these to those recorded since COVID-19 hit Bergamo.
The 10 patients they had seen (all of whom were from the region) represented a 30-fold increase in episodes per month. But the new patients were older than the typical Kawasaki patient – one was two, others were between five and nine years old, one was 16 – and their conditions were more severe.
He prepared a study, included the most salient details to best represent the illness and published his team’s results in The Lancet medical journal on May 13. After the study was submitted, another 10 children were identified with Kawasaki-like symptoms – 20 in total, or as many as the five years before COVID-19.
“At this moment, we do not have a database with other countries,” Dr D’Antiga explained. “It has been so fast and we’re so busy … at the moment, that is done at a national level.”
The tip of the iceberg
Dr Michael Levin is the chair of paediatrics and international child health at Imperial College in London, where he has studied malaria, tuberculosis and Kawasaki disease for more than 30 years. By late April, he was following the progress of dozens of children who looked worryingly like the patients in Milan. On May 2, he had enough data to describe 38 of them to a virtual meeting of the World Federation of Pediatric Intensive and Critical Care Societies, held via Zoom.
“I cannot ever remember a time in my 30 years of working in paediatrics where there is so little known about a new spectrum of illness,” Dr Levin said.
It has come at us with unprecedented speed and magnitude, and we have had to learn so much so fast.
The National Health Service (NHS) in the UK had sent its first alert about the unnamed syndrome to caregivers on April 27. Some children presenting in the UK hospitals were very sick, requiring intensive care. One had died. All were connected to SARS-CoV-2 cases; the children had either produced their own antibodies or been near previously ill family members, but only a few were actively infected.
Notable among the NHS’s set of warnings was the likelihood of organ failure. And to the surprise of some, the proportion of black and ethnic minority children was higher than white children in the groups who were showing up in the UK hospitals.
“We think of it as an iceberg: we first noticed the tip of the iceberg, then we saw the Kawasaki group, and now we’re seeing less severely ill children with persistent fever … from the timing of it, it looks like something that follows six weeks from the curve of COVID-19,” Dr Levin said.
Multisystem inflammatory syndrome
The Zoom call was the first to put clinicians from both sides of the Atlantic Ocean in the same virtual room as their colleagues from Boston, London, Barcelona, Genoa and Geneva. There were 1,800 viewers on the call; the conference was so full that some used FaceTime to screen-share Zoom.
Doctors in cities that had just surfaced from the full force of a COVID-19 outbreak addressed doctors in cities that were at the peak of the wave.
They named the new illness paediatric multisystem inflammatory syndrome (PMIS) or multisystem inflammatory syndrome in children (MIS-C) – the name is still evolving – described its symptoms, and prescribed a course of treatment. Each case would require a team of specialists and the facilities of a well-resourced city-based children’s hospital.
“We should know in a couple of weeks how wide the spectrum is. We’ve literally been hearing about cases every day, but we’re not gathering systematic data. It’s all word of mouth from Zoom conferences and webinars,” Dr Levin said on May 12, writing a few days later to say that data-collection processes had already improved.
In Boston, members of a paediatric specialist team had been alerted to the new illness by their colleagues in Europe days before the first cases arrived.
Dr Jane Newburger has studied and treated Kawasaki disease for more than 30 years at Boston Children’s Hospital (BCH), where she is associate cardiologist-in-chief for academic affairs. Before COVID-19 cancelled travel, she regularly attended conferences in Japan and the US.
“I have not dealt with something with the urgency of COVID-19-associated illness,” she said. “It has come at us with unprecedented speed and magnitude, and we have had to learn so much so fast. Everything has sort of burst like a match to a haystack.”
As COVID-19 rolled across Massachusetts, the children’s hospital offered to receive patients from other hospitals, and sent ventilators and clinicians to centres that needed backup. Massachusetts has had in excess of 87,000 cases, more than 5,000 deaths, and 3,000 children have tested positive for the virus (no children have died in the state).
When the paediatric ICU several floors above Dr Newburger’s regular clinic received its first cases of PMIS, few teams were better positioned to treat and study it.
“My interests are how this new inflammatory syndrome affects the heart,” Dr Newburger said. “We know that some of these children have low heart function, some have enlargement of coronaries, or even giant aneurysms. We do not know the extent to which this new inflammatory syndrome … creates the same risk as Kawasaki does.”
Keeping count of cases
Keeping the data organised is another concern.
In the UK, the NHS runs a 34-year-old system called the British Paediatric Surveillance Unit, or BPSU, through which physicians report new illnesses to a central database. This system lifts the rare signals from the noise of daily practice, and eventually shapes health policies (BPSU originally ran on postcards sent from clinics to London). BPSU has been activated to track PMIS.
In the US, multiple reporting systems overlap. There are private and public hospitals, societies of physicians and government databases. Between all of these, Dr Newburger worries, some case descriptions may be duplicated.
“We need to have a registry of registries that allows us to put information from large numbers of PMIS patients together so we can generate reliable statistics about the disease for families and healthcare providers. We have nothing like that right now.”
While caring for the sickest, Dr Adrienne Randolph is keeping such a count. She is a senior associate of critical care medicine and professor of anaesthesia and paediatrics at BCH, working in an ICU on the hospital’s seventh floor. Her days are long. Some PMIS patients require support from machines called ECMO units, a serious intervention that does the work of the heart and lungs via tubes and pumps outside of the body, allowing doctors to deploy countermeasures against the illness.
Many of the children have spent more than a week in intensive care. Parents need reassurance. No stranger to trauma, Dr Randolph is hopeful. “I’m happy to say that a great majority of these kids are recovering. With good supportive care, and intervention, there is recovery,” she said.
A few years into the H1N1 flu outbreak almost 10 years ago, Dr Randolph established a network of surveillance sites across the US to report emerging diseases, research that was converted into a Centers for Disease Control and Prevention (CDC)-run tool. In 2016, she led a study to count and characterise every paediatric ICU bed in the US.
“We were anticipating that the next pandemic would be a novel influenza, so we set up a surveillance registry. In 2013 we had over 36 sites,” she explained, “ready to collect data if a public health authority declared an outbreak of a novel pathogen.”
During the first months of COVID-19, there was little money available for a surveillance initiative on children. Then the first PMIS patients arrived, and the CDC funded her to launch it. With them, her team produced a 45-page questionnaire for clinicians to record every symptom, lab test and intervention in every new case in the US. She is regularly on the phone with a handful of government health agencies. The work has just begun.
“My goal with this first and foremost is to get the data to the people who can make decisions about how to act on this, how to get the right resources, how to do the right education and how to improve health outcomes,” she explained.
Who is most at risk?
By early May, New York, New Jersey, Massachusetts and Connecticut reported clusters of children with PMIS and three deaths. Dr Randolph already knows there are more.
“I have [seen reports of] cases in over 22 different states. It’s growing every day. It’s not just a thing that’s been happening on the East Coast,” she said on May 16. Dr Randolph is extremely cautious, not wanting to amplify parents’ anxieties. It is the diversity among the cases that keeps her focused on the data.
“People are, as we speak, working on this across over 65 sites. They’re spending long hours filling out the data and asking us questions to ensure their cases meet the criteria for reporting.”
If Kawasaki disease can provide a frame for understanding PMIS, it is a limited view. Kawasaki disease has no known culprit, but many suspects. PMIS patients are all linked directly to coronavirus infection or proximity to someone who was sick. Some children with PMIS are showing a spectrum of features of Kawasaki disease, but many of the new patients are older and their conditions are more severe. There have been no recorded cases in Japan, South Korea and China, where Kawasaki is usually seen in higher numbers, though all were hit hard by SARS-CoV-2.
PMIS could teach these doctors about a disease whose cause they have pursued for half a century. It could inform drug companies rushing to develop a SARS-CoV-2 vaccine. If a reaction to the body’s own antibodies is triggering PMIS in some children – an idea that is being discussed, but remains far from proven – it could influence discussions about how to deploy vaccines, or how to reopen schools and countries.
These are similar to the questions Dr Randolph asked during the H1N1 pandemic. “When the vaccine comes out, it’ll be in short supply and we need to know who’s at risk. Who gets the vaccine first? Who has the most risk factors? Usually, we can’t give the vaccine to everyone immediately.”
Beware of assumptions
Dr Michael Levin is able to study the syndrome from one of the best-resourced children’s facilities in the world, but COVID-19, blind to poverty and prosperity, has struck communities across 188 countries. He says any centre can be part of the effort to better understand the illness.
“The main thing we need to do is get the patients into studies right now. As long as [caregivers] can set up a data and a sample collection system, they might not have the resources to do detailed genomics and proteomics and so on, but they can collaborate.”
The World Health Organization (WHO) launched a working surveillance group led by doctors from Canada, Spain, the US, the UK and India, hoping to coordinate the vast and diverse troves of incoming data.
Over the weekend of May 15, as teams in the US poured over their datasets, a nine-year-old boy in France died of complications likely from PMIS.
He is one of at least six children reported to have died from the illness so far. Over the past month, doctors across the United Kingdom and the European Union say they have seen more than 200 cases of the new syndrome, while New York state reported more than 140.
Dr Mary Beth Son, a rheumatology programme director at BCH who works with the Kawasaki team, cautions against bias.
“I think the danger here is that you make assumptions and assume it’s like what we’ve seen before,” she said. “We need to keep our minds open.”