For decades, cases were rare. A virus called enterovirus D68 (EV-D68) popped up only sporadically, causing mild respiratory symptoms, and was identified less than 700 times in total. Then, in 2014, something changed.

Large outbreaks of the virus spread across the US, Canada, Europe and Asia, with thousands of people, mostly children, succumbing to severe respiratory symptoms.

At the same time, a mysterious polio-like illness called acute flaccid myelitis (AFM) began paralysing children in small clusters across the US – and ever since, scientists have sought to understand whether or how the two were linked.

Now, we've got our strongest evidence yet that EV-D68 is what's behind these AFM clusters, with a new analysis by researchers from the University of New South Wales (UNSW) in Australia supporting a causal relationship between EV-D68 and AFM.

"More than 120 children developed the condition, known as acute flaccid myelitis, in the US alone," explains infectious disease epidemiologist Raina MacIntyre, "but experts were baffled as to the cause."

One of the reasons for the confusion was that up until that point, EV-D68 had never been known to cause the strange symptoms seen in children with AFM.

The condition affects the nervous system, causing weakness in limbs and other disruptions to nerves, such as difficulty in swallowing, drooping eyes, some pains, and other polio-like symptoms.

In the 2014 outbreak, 120 such cases were identified in the US, representing about 10 percent of the EV-D68 cases seen at the same time. All up, the virus was registered in patients across 49 US states.

To determine what level of association existed between the two outbreaks, MacIntyre's team analysed the scientific literature on acute flaccid myelitis, filtering through more than 100 previous studies on the condition, alongside other relevant non-peer-reviewed data found on the internet.

Against this database, the researchers applied the Bradfield Hill criteria – a group of nine principles developed to determine causality, established by English epidemiologist Sir Austin Bradford Hill, who demonstrated the connection between smoking and lung cancer in the 1950s.

"We are first to use his approach to analyse the relationship between EV-D68 and acute flaccid myelitis," MacIntyre says.

"Our results show that it is very likely that EV-D68 is the cause of the mystery illness and the paralysis of children."

As for what can explain the emergence of these polio-like symptoms, given EV-D68 never gave rise to them before, the researchers say a couple of things are possible.

"We know from genetic analysis that the virus has mutated," MacIntyre told ScienceAlert.

"The original strain does not appear to cause AFM, but strains that have evolved more recently [called clades A, B and B1] do cause AFM in mice, and have been associated with the US and European outbreaks."

Another relevant issue is surveillance. As far as viruses go, EV-D68 is still a relatively recent discovery (and rare), and AFM is an extremely new condition that scientists are still only starting to understand.

Notwithstanding that animal studies reveal certain clades of the enterovirus are more pathogenic, we still need a lot more research to gauge what's really going on here – and how dangerous the risks of EV-D68 could ultimately be.

While incidence of the virus may have been quieter since the initial 2014 outbreaks, MacIntyre counsels against any complacency.

" Epidemic diseases tend to be cyclical," she says. "Good surveillance (which was not present in the past) is needed to get a good handle on this."

As for whether the weight of the new findings makes the case for the development of a vaccine to protect against EV-D68 and AFM more probable, MacIntyre says it's unlikely.

While there is no treatment for the paralysing, polio-like illness, the condition's rareness to date so far count against the chances of a vaccine being engineered.

"Vaccine development can be a long process from concept to licensure, sometimes 10-20 years," MacIntyre explains.

"Sometimes in serious epidemics (such as Ebola in West Africa), vaccine development is fast-tracked. However, with cases of AFM being in the hundreds, this is unlikely for EV-D68."

The findings are reported in Eurosurveillance.