91av

IT WAS rare for schoolchildren in post-Edwardian England to misbehave, at least by today’s standards. But at the turn of the 1920s there were noticeable outbreaks of bad behaviour among boys and girls of all ages – so serious in fact that their parents consulted doctors. Formerly obedient and accommodating children either turned into hyperactive rebels, in a syndrome people labelled “Apache”, or became morose and somnolent.

Gradually it emerged that the affected children had neurological problems. They went on to develop violent tremors similar to those of Parkinson’s disease, and became increasingly rigid and paralysed.

This phenomenon was not just confined to England, nor just to children. It soon emerged that all over the world people were succumbing to the mysterious “sleepy sickness” or “sleeping sickness” (unrelated to the tropical disease of the same name). About 25 per cent died from respiratory problems or other causes, while about 33 per cent recovered. Many of the rest remained in semi-comatose states and were committed to asylums or special hospitals.

The incidence of the disease peaked in 1924, then petered out with few new cases by the end of the decade. No cure emerged, so the surviving patients – about 4000 in the UK, and an estimated 100,000 to 200,000 worldwide – stayed in care all their lives.

The disease’s cause was a mystery. Because of its timing, much suspicion fell on the great flu pandemic of 1918, which had travelled the globe. Other suspects included the Borna virus (which normally infects horses), polio, and other viruses.

The mystery of sleepy sickness remains unsolved to this day, but recent advances in molecular biology have provided new ways to investigate it. Scientists around the world have begun looking for traces of flu virus in victims’ brains. Among those engaged in this molecular detective work are researchers from my lab at St Bartholomew’s and the Royal London Hospital, along with K. C. Lo, Alex Elliot and Rod Daniels at the National Institute for Medical Research in Mill Hill, London. Despite encountering several setbacks, we hope to solve one of the greatest medical mysteries of the 20th century.

The first recorded description of sleepy sickness was in 1917. It is a rare and precious moment for a doctor to report a new clinical entity, but Constantin von Economo was a rather rare person himself. A baron from Romania, he was a balloonist and first world war pilot before becoming a doctor in Vienna. He began his first report of the condition with the words “We are dealing with a kind of sleeping sickness”, and in a few lines laid bare the disease’s hallmarks: the somnolence and stupor, and the later onset of “shaking palsy”. His reputation was established overnight, and sleepy sickness was later called von Economo’s disease, as well as encephalitis lethargica.

In the early 20th century, there was little doctors could do other than admit patients to hospitals and care for them physically. In addition the British government made sleepy sickness a notifiable disease, obliging doctors to report all their cases. These records are now a source of reliable statistics.

In the 1960s there was an interlude of hope when neurologist Oliver Sachs gave patients in New York levodopa, the new drug for Parkinson’s disease. Some began to walk and talk haltingly. But for unknown reasons the medicine became less and less effective: it had to be given in increasingly large doses, which patients could not tolerate, and they sank back into somnolence and inactivity. Sachs wrote about them in his book, Awakenings, which was later made into a film of the same name.

So what progress have we made in establishing the cause of sleepy sickness? The chief suspect has continued to be the 1918 flu pandemic. In 1982, epidemiologists Ray Ravenholt and William Foege from the US Centers for Disease Control in Atlanta, Georgia, published a review of the evidence, setting out 10 factors linking the two diseases. While not establishing proof, they made a compelling case for the flu virus.

Ravenholt and Foege pointed out that numerous regions and nations had suffered an outbreak of pandemic flu, followed closely by sleepy sickness. In Seattle, Washington state, there were several waves of flu, each followed by what appeared to be a surge of sleepy sickness cases about 12 months later (see Graphic).

Flu reached Western Samoa in 1918, the authors pointed out, and one year later there was a rise of what was called locally fa-aniniva – interpreted as “fatal disease of the head”. In contrast, American Samoa’s strict quarantine procedures stopped pandemic flu hitting those islands until 1926, the first year that sleepy sickness occurred there.

One criticism of the flu hypothesis is that the first cases of sleepy sickness were noticed in Vienna and France in 1917 – before the accepted start-date of the flu pandemic. But last year my team published research on the pandemic’s origins, which showed there were flu cases in the army camps of northern France in the winter of 1916-1917 that had all the classic hallmarks of the great pandemic strain (Lancet Infectious Diseases, vol 2, p 111).

There have been other flu pandemics. The Russian flu of 1889 was followed by reports of a strange neurological condition called “nona”. But why were there no outbreaks after the Asian flu pandemics of 1957 and 1968? It might be because each virus had a different genetic makeup. And the 1918 virus was exceptionally contagious and virulent, which might have somehow raised the risk of sleepy sickness.

Sporadic cases of sleepy sickness still occur, as you might expect if some strains of flu were the cause. In the 1980s, neurologist Michael Swash at the Royal London Hospital reported the cases of eight patients who had acquired the disease in the preceding 20 years. The exact mechanism by which the virus could cause sleepy sickness is unclear. But we do know that it can reach the brain, as the pathogen’s genes have been found in the cerebrospinal fluid of children suffering fits after a bout of flu. To meet the estimated incidence of sleepy sickness, of half a million cases worldwide after the first world war, about one case of flu in a thousand would have had to trigger sleepy sickness.

And there is at least one precedent for a respiratory virus triggering a neurological disease. A rare complication of measles is subacute sclerosing panencephalitis, a fatal type of brain inflammation. The measles virus spreads from the epithelial cells in the skin and throat to the brain via the blood. There it slowly and quietly multiplies, with symptoms becoming apparent only four to five years after the initial measles infection.

If it turns out that flu is the cause of sleepy sickness, the ramifications would be huge. In the 20th century there were three global flu pandemics, and it is widely agreed that another could happen any day if a virulent new strain emerges (91av, 14 December 2002, p 36). I do not think this time will be too far ahead.

The immediate impact of the next pandemic will be bad enough, but if a global outbreak of sleepy sickness is likely to occur 12 months later, we should be prepared. On an optimistic note, there is a new class of anti-flu drugs, the neuraminidase inhibitors, and at least one of these can get into the brain. Such medicines might help prevent sleepy sickness.

This speculation is irrelevant unless we know for sure if flu is indeed the cause of sleepy sickness. But new avenues for investigating this 80-year-old murder mystery have opened up thanks to modern molecular techniques such as PCR. This can amplify minute traces of DNA or RNA by up to a million-fold, allowing us to look for flu genes in the brains of sleepy sickness victims.

My team has made use of a tissue archive at the Royal London Hospital. By painstakingly combing through the ancient handwritten records, we found eight patients who had died at the hospital from acute encephalitis lethargica in the early 1920s.

As a control experiment we first tested whether PCR could detect flu genes (which are composed of RNA) in formalin-fixed brain slices from recently infected mice. That worked. The next step was to check whether RNA would survive eighty years of storage, by looking in the archived brain sections for RNA we knew should be present – messenger RNA for the beta-actin gene, commonly switched on in brain cells. That, too, was successful. Then we looked for sequences from three flu genes. In June we announced that those tests came back negative (Virchows Archiv, vol 442, p 591).

That was disappointing, but not completely surprising, as it echoed recent findings on similar historical archives by Jeffery Taubenberger’s team at the Armed Forces Institute of Pathology in Washington DC. But we think that neither ours nor Taubenberger’s studies disprove that flu was the cause of sleepy sickness. It is possible that viral RNA is less likely to survive 80 years of storage than beta-actin mRNA. Or perhaps the PCR tests simply weren’t sensitive enough to detect the tiny quantities of flu RNA likely to be present.

By then another line of enquiry had opened up. Seven years ago, I was contacted by a lady called Jean Price. Her brother Philip Leather had been a victim of sleepy sickness in 1928, and he was still living, staying in care facilities in Birmingham.

It is worth dwelling on Philip’s case, because it helps us to grasp the human impact of this heart-rending disease. Jean cannot recall if Philip was one of the flu pandemic’s victims. But she remembers him as a quiet, creative 10-year-old who invented stories about a fairyland called Magton, where children would always be safe. In one of his notebooks he wrote: “The fairy is beautiful. The heather smells fresh in Magton Heath… remember that nothing can hurt you in fairyland.”

But in the real world, Philip was already starting to suffer the ravages of sleepy sickness, becoming apathetic and badly behaved. On one occasion he hid under the bedcovers for three days. The family called in a medical expert on sleepy sickness. The doctor confirmed Philip’s diagnosis, but nothing could be done for the boy, and at 11 he was admitted to Hollymoor Mental Hospital. Philip spent the rest of his life in care.

I visited Philip in 1996. He had not spoken for decades, but he seemed to react to a photograph of himself as a young child (above). I wondered then whether deeper inside, he was still the quiet 10-year-old he had been before illness put his life on hold.

Last December, at the age of 84, Philip died from a respiratory infection. Jean had kindly granted us permission to use Philip’s remains for our research. So on a cold wintry day, I travelled up to the Queen Elizabeth Hospital in Birmingham and collected Philip’s brain from the post-mortem room.

We began our task by taking 20 thin slices from the brain, then bathed them in enzymes that digest proteins and fat molecules. Then the Mill Hill team incubated the samples with PCR primers for the virus’s matrix gene, which encodes a structural protein and is known to vary little between modern flu strains and the pandemic virus.

Last month we got the results. Once more they were negative. It was a disappointing moment for us, but we still do not feel the flu hypothesis has been disproved. Again, perhaps the PCR tests were insufficiently sensitive. Perhaps the flu virus triggered the neurological problems, then was virtually eradicated from the brain – a case of “hit and run”.

We won’t give up on the flu hypothesis yet. Last month we spent six hours searching through the archives at the Royal London and found another three samples, on which we will repeat the PCR tests. In addition, another relative of a still-living patient with sleepy sickness has contacted me, and we are looking for more individuals. Meanwhile, we will store Philip’s brain for the future to await the arrival of more sensitive PCR techniques.

And there are different avenues to explore. We intend to search for other viruses in the archived tissue samples, and in recently deceased sleepy sickness victims. The old suspects of polio and the Borna virus are the obvious ones for which to look.

More recently the herpes virus (which causes cold sores and genital herpes) has been proposed as the cause of sleepy sickness. And last month virologist Ernie Gould at the Centre for Ecology and Hydrology in Oxford, UK, suggested to me that the trigger could have been the West Nile virus (or a relative), which was first recognised in humans around the turn of the 20th century. It would be alarming if Gould were correct, as a particularly virulent strain of West Nile spread to North America in 1999, where it is infecting rising numbers of people every year, and it has also recently reached Europe.

So investigating the cause of sleepy sickness has never been more important. If Philip’s story tells us anything, it is that we must learn all we can from the past. It was in this spirit that his sister allowed us to remove his brain after death, and gave us the wonderful chance to study it. Only by solving the mystery of sleepy sickness can we stop this debilitating illness from wreaking havoc on the world once more.