Weeks later the $100,000 device, Australia's first portable life-support machine able to fully take over the function of both heart and lungs, was at the centre of a statewide rescue mission as dozens of previously healthy young people suddenly became disastrously ill - victims of the H1N1 swine flu epidemic that swept the country that winter.

''It was incredible. We'd only just landed the [machine] and the fear was we'd be sitting on our hands,'' says Forrest, a clinical associate professor and head of cardiothoracic anaesthesia and perfusion at Royal Prince Alfred Hospital. He had lobbied NSW Health for the equipment with the intensive care unit head, Robert Herkes, and colleagues at St Vincent's Hospital. ''But from the first week of June the phone didn't stop ringing. We'd budgeted to do one case a month for the whole state. We did 24 in the first two months.''

In the process, the NSW experience has rewritten the rule book on how to care for people who become critically ill far from a major hospital - achieving life-saving transfers across what had until then been considered unfeasible distances and narrowing the survival gap between country and city people.

But 2½ years ago that pattern had not yet begun to emerge from the engulfing chaos. Forrest vividly recalls the patients. Many were obese, putting extra strain on their lungs. But their weight also appeared to give them extra reserves. ''It's the scrawny middle-aged men who don't make it,'' Forrest says.

With no resistance in the community, except among older people exposed to a similar flu strain before 1950, the infection rampaged through the bodies of susceptible people. It caused overwhelming pneumonia, flooding lungs with, ''a very watery infiltrate … you could pull about 100 millilitres of greenish fluid'', Forrest says.

The extracorporeal membrane oxygenation (ECMO) machine takes over the workload of over-stressed heart and lungs artificially, giving them the chance to recover.

Only two Sydney hospitals, RPA and St Vincent's, have ECMO facilities, often used for lung and heart transplant patients treated there.

But if people fall critically ill somewhere else within NSW's 800,000 square kilometres, they may not survive a conventional transfer. Mechanical ventilation can further injure damaged lungs; air pressure changes in flight can further reduce already dangerous blood oxygen saturation.

Portable ECMO, in which a team of at least three doctors and nurses flies to a country hospital and stabilises the patient on the machine before bringing them back to one of the Sydney centres, may circumvent those difficulties.

NSW now has five of the machines, from the German company Maquet. They are rotated between the two hospitals and the airbase of the Ambulance Service of NSW at Bankstown, where they can be anchored to a bracket in a helicopter or fixed wing aircraft, or a road ambulance.

The ambulance service is responsible for determining which patients qualify for ECMO retrieval, which is proposed to go national, first to Queensland and then Victoria.

But in the case of the 2009 swine flu patients, it was not even clear whether trying to save those most critically ill was in their best interests.

''One of the last patients had a very intense inflammatory response of her airway,'' Forrest says, making it impossible to get standard breathing equipment into her swollen, bleeding throat and lungs already compromised by a life-threatening clot; blood-thinning medication was not an option as it would have triggered other complications.

The woman's body was so full of infection, ''you couldn't even see her trachea on a chest X-ray'', Forrest says, and death looked certain. But he and colleagues decided nevertheless to attach her to the ECMO machine, ''to see what happened … the blood clot broke up, and she went home to seven kids. It was unbelievable.''

''You have these fantastic saves and it's brilliant, but there are others [who] would have died weeks ago if you hadn't intervened but they're still going to die anyway. That's the downside,'' Forrest says.

Still, the net result is convincing. Among what Forrest calls the ''high quality survivors'' of swine flu who were retrieved on ECMO, ''some of them would have survived. [But] we think on current estimates half of them would have died'', if they had been transported using conventional ventilation or heart support.

In their analysis of the epidemic, published this year in the journal Intensive Care Medicine, Forrest and colleagues found 86 per cent of adults retrieved on ECMO left hospital alive, versus 38 to 47 per cent of people with similar severity of illness treated with standard techniques.

The lives saved come at a price. Transport costs vary but are typically several thousand dollars. Each patient needs a $7000 dedicated ''circuit''of tubes and polymer blood filter, plus $2000 cannulae to attach it. ECMO support after the patient gets to hospital is thought to amount to about double the $2500-a-day cost of regular intensive care, and the swine flu patients remained on the equipment for an average of 10 days.

But the patients are young - the average age of swine flu cases was 34 - and stand every chance of returning to complete health and productive lives.

Forrest argues ECMO retrieval supports basic social justice, spreading the chance of recovery evenly across NSW. ''Before this, if you got sick you had a better chance of surviving in one hospital than another 20 kilometres down the road,'' he says.

''It has been a lottery in the past. This provides equity of access for people in regional centres, and peripheral hospitals in Sydney, to the most advanced cardiac and respiratory support that's available.''