‘My whole skull was throbbing’: How it feels to survive bird flu

A colorised transmission electron micrograph of avian influenza A H5N1 virus particles
H5N1 first emerged in China in the late 1990s - CDC and NIAID/AFP via Getty Images

It sounds like something out of a science fiction film but clade 2.3.4.4b, as the latest major strain of H5N1 bird flu circulating is known, is very real. And while scientists are wary of saying so in public, many are increasingly worried about its potential to trigger a pandemic – one that could prove deadlier than Covid-19.

The difference – were that to happen – is that this one will have been a slow burn. Unlike SARS-CoV-2, which emerged suddenly in Wuhan in December 2019 and was already well-adapted to humans – H5N1 first emerged in China in the late 1990s. Then it affected wild waterfowl. But by 1997 it was turning up in ducks and domestic chickens in Hong Kong – and that same year, it claimed its first human fatality, a three-year-old boy.

More recently, the virus has been isolated from a wide range of mammals, including foxes, mink, penguins, seals and cows. And this week a dozen mice tested positive for the virus in New Mexico – the first time the rodents have been infected with H5N1 outside of a laboratory.

But the bigger worry is that the virus is now rampant in US dairy herds – since March, the virus has been detected in cattle on farms in nine states and three dairy workers have been infected.

Until now, the symptoms have been mild. But should the virus mutate further that could change: the latest dairy worker reportedly developed respiratory symptoms – the first time this has happened in a human case in US and an indication that the virus could be evolving to become transmissible between people in the same way as seasonal influenza.

That is a worry as H5N1 has a human case fatality rate of 50 per cent – many times higher than Covid-19. However, since the first human case in 1997, just 463 deaths have been recorded worldwide, with most confined to individuals closely related, suggesting the virus has trouble spreading outside of “family clusters”.

The death reported in Mexico on Wednesday of a person confirmed to have been infected with an H5N2 virus is also a concern but is not thought to be linked to the much wider H5N1 outbreak.

So, how concerned should we be? And what is it like to be infected with H5N1?

‘It felt like I was going colour blind’

Nineteen years ago, during the outbreak of H5N1 that swept Southeast Asia in 2005, I travelled to Hanoi, in Vietnam, to speak to a man who had been  hospitalised with the virus and had lived to tell the tale.

At the time, Vietnam had recorded 29 human cases, of which 20 had resulted in deaths, and with millions of chickens infected at battery farms across the country, the government had begun extensive poultry culls. With the help of a local photographer, I tracked Nguyen Thanh Hung down to a tiny apartment in a backstreet of Hanoi.

A wiry cement salesman with a passion for long-distance running, Hung, who was then 42, had fallen ill after nursing his sick brother in hospital. Hung’s ordeal began when he had attended the funeral of his brother’s son the previous December. The funeral coincided with Tet – Vietnam’s version of Thanksgiving – and after the service Hung and his brother had sat down to a meal of duck’s blood soup, a local delicacy.

Hung hardly touched the soup, which is made from raw, congealed duck’s blood. However, two days later his brother developed a headache and breathing difficulties and was taken to Hanoi’s tropical diseases hospital. Learning his brother was ill, Hung rushed to his bedside where he sat vigil for the next five days, watching as his brother’s breathing became steadily more laboured.

Nguyen Thanh Hung, who survived bird flu
Nguyen Thanh Hung, who survived bird flu

In a scene that would foreshadow the experiences of Covid-19 patients fifteen years later, Hung’s brother was intubated and placed on a respirator. But he found the procedure uncomfortable and was administered pressurised oxygen via a face mask instead. However, Hung told me, his brother kept on removing the mask, forcing Hung to lean across his body to replace it. “I was about 30cm from his face,” he told me.

Hung’s brother died on January 9, prompting Hung to return to their village for a second funeral. Then, two days later, Hung also began to feel unwell. X-rays revealed an ominous white shadow on his lung. Initially, doctors suspected tuberculosis but when Hung’s temperature soared he was also taken to Hanoi’s tropical diseases hospital.

“My whole skull was throbbing,” he told me. “I felt like I was going colour blind and the walls were pressing in on me.”

Ten days later, a chronically backlogged lab sent back word that his brother had died of H5N1. Soon after, it was confirmed Hung had the virus, too.

“I was quite scared at first, but by then I was starting to feel better. The fever and pain were over and I thought to myself, ‘I’m through the worst’.” Nonetheless, it was another ten days before Hung was well enough to be discharged.

To this day, it is not known how Hung contracted the virus. Was it through imbibing the raw duck’s blood or was he infected at his brother’s bedside? The latter, of course, would be extremely worrying, suggesting that H5N1 can be transmitted as an aerosol from person to person. Although scientists took swabs from Hung to culture the virus and conduct more precise molecular studies, the tests were inconclusive.

However, Peter Horby, then the WHO’s representative in Hanoi, now director of the Pandemic Sciences Institute at the University of Oxford, could not rule of the possibility of “human-to-human transmission”. His colleague, Jeremy Farrar, now Chief Scientist at the World Health Organization but then director of the Oxford University Clinical Research Institute at the Hospital for Tropical Diseases in Ho Chi Minh City, told me it had been “a very frightening time” for everyone, clinicians included.

In light of the latest positive case in the US, the question is: could human-to-human transmission be happening now without our realising? We know from genomic analysis that the new clade of H5N1 began circulating in dairy cows in the US as early as December 2023 – in other words four months before it was detected in March 2024. But to date only around 30 people have been tested for the virus, the majority of them dairy workers.

The worry is that if we wait until patients present at hospitals with severe symptoms of bird flu, it will be too late to contain the outbreak. The genie will already have escaped the bottle, just as it did in Wuhan in the last months of 2019.

As two eminent flu experts warned in an article in the British Medical Journal this week, “the hazard and risk of a major outbreak of H5N1 are  large, plausible and imminent”.

That would be tragedy that everyone, including Hung, should be keen to avoid.

Protect yourself and your family by learning more about Global Health Security