A donor heart stopped, restarted, and transplanted

Surgeons at St Vincent's Hospital, Sydney revived donor hearts that had stopped, kept them beating in a perfusion device and transplanted them successfully in 2014.

A rule that excluded most donors

For decades after the first human heart transplant, every donor heart came from the same kind of donor: a person declared dead by neurological criteria whose heart was still beating. The heart was stopped with a cold cardioplegic solution, packed in ice and carried to the recipient inside a window of about four hours. Kidneys, livers and lungs had long been retrieved from a second kind of donor, one whose heart had stopped after life support was withdrawn, a category called donation after circulatory death, or DCD. Hearts were not. A heart that had already stopped, and then sat without oxygen while death was confirmed, was assumed to be too damaged to restart in someone else.

The assumption was never tested in adults until a team in Sydney tested it.

Three hearts, one machine

The recipients were patients at St Vincent's Hospital in Sydney. The donors were Maastricht category III, meaning controlled withdrawal of life support in hospital, under 40 years of age, with a maximum permitted warm ischaemic time of 30 minutes. The retrieval team, led by surgeon Kumud Dhital with Peter Macdonald of the Victor Chang Cardiac Research Institute, protected each heart with a supplemented cardioplegic solution and then placed it in a TransMedics Organ Care System, a portable device that perfuses the heart with warm, oxygenated donor blood and keeps it beating during transport. Four hearts were retrieved. Three were transplanted.

The results were published in The Lancet in 2015. The three recipients, two men and one woman with a mean age of 52, had warm ischaemic times of 28, 25 and 22 minutes and ex-vivo perfusion times of 257, 260 and 245 minutes, meaning each heart beat on the machine for more than four hours before implantation. Lactate, a marker of metabolic distress, fell during perfusion in every heart: from 8.3 to 3.6 millimoles per litre in the first, from 6.79 to 2.8 in the second and from 7.6 to 2.69 in the third. Two recipients needed temporary mechanical support after surgery. All three had normal cardiac function within a week and were recovering at 176, 91 and 77 days when the paper was written.

The Victor Chang Institute's own account of the announcement, dated October 24, 2014, records that the first recipient, Michelle Gribilas, was transplanted in August 2014, that the second, Jan Damen, was 43, and that the preservation solution used to protect the hearts had taken the Institute's laboratory twelve years to develop. The first heart had been stopped for 20 minutes before it was restarted.

The device had been built for a different problem

The Organ Care System was not designed for stopped hearts. It was designed to extend the reach of conventional ones. In the same issue of The Lancet in 2015, Abbas Ardehali and colleagues published PROCEED II, a randomised trial at ten centres in the United States and Europe in which 130 recipients of hearts from brain-dead donors were assigned to the device or to standard cold storage. Thirty-day patient and graft survival was 94 percent with the device and 97 percent with cold storage, a difference within the trial's non-inferiority margin. The device was as good as ice. It was not yet better.

What it offered that ice could not was time and information. A heart on the machine can be transported for longer, and its metabolic state can be read from the perfusate. For a heart that has already stopped, that second property is decisive: the surgeon can watch lactate fall and decide whether the organ is recovering before committing a patient to it.

The series grows

A single centre in the United Kingdom followed. In 2017 Simon Messer, Stephen Large and colleagues reported in the Journal of Heart and Lung Transplantation 28 DCD heart transplants over 25 months, some using direct procurement and machine perfusion as in Sydney and some restoring circulation inside the donor's body first. Ninety-day survival was 92 percent, against 96 percent in a matched group of conventional transplants, and one-year survival was 86 percent against 88 percent. Neither difference was significant.

In 2019 the Sydney group published its own update in the Journal of the American College of Cardiology. From 45 DCD donor referrals between July 2014 and April 2018, 33 hearts were retrieved; 12 donors did not reach circulatory arrest within the permitted window. Eight hearts failed viability criteria on the machine and two were lost to machine malfunction. Twenty-three were transplanted. Nine recipients needed mechanical circulatory support for delayed graft function. One died on the sixth postoperative day, a mortality of 4.4 percent over four years. Every surviving recipient was in New York Heart Association class I with normal biventricular function. The numbers also show the cost of the method: of 45 potential donors, roughly half yielded a transplant.

The randomised trial

Case series from committed centres can flatter a technique. The test that mattered was a randomised comparison, and TransMedics funded one in the United States. Jacob Schroder and colleagues reported it in the New England Journal of Medicine in 2023. Adult candidates were assigned 3:1 to a group that would accept a DCD heart on the device or a conventional heart, whichever came first, or to a group that would receive only a conventional cold-stored heart. In all, 180 patients were transplanted: 90 received a DCD heart and 90 a brain-dead donor heart. In the as-treated analysis of 166 recipients, risk-adjusted six-month survival was 94 percent in the DCD group and 90 percent in the conventional group, meeting the prespecified non-inferiority margin with a p value below 0.001. Serious adverse events related to the graft at 30 days did not differ.

The regulatory sequence followed the evidence. The FDA approved the Organ Care System Heart on September 3, 2021, for brain-dead donor hearts that would otherwise be declined because of long expected ischaemic times. On April 27, 2022, a panel-track supplement extended the indication to the ex vivo reanimation, functional monitoring and beating-heart preservation of DCD hearts.

What it did to the numbers

The claim on this story's card is that the donor pool grew by roughly a third. The primary figure behind it is institutional, not global. The Victor Chang Institute projected at the 2014 announcement that the technique would allow 30 percent more heart transplants per year in Australia, and its current account states that at St Vincent's, DCD donation boosted the number of heart transplants over the following ten years by 32 percent, with close to half of the hospital's transplants now using the device. Messer's paper described DCD as an additive source rather than a replacement. The 2019 Sydney data show why the gain is not larger: a quarter of referred donors never arrest in time, and a further quarter of retrieved hearts fail on the machine.

What the story does and does not show

The 2015 case series showed that three hearts which had stopped for 22 to 28 minutes could be restarted, kept beating for more than four hours and transplanted with normal function at one week. The 2017 and 2019 series showed survival comparable to conventional transplants in 28 and 23 patients respectively. The 2023 randomised trial showed non-inferior six-month survival in 90 DCD recipients. Those are the measured results.

The story does not show equivalence over the long term; the randomised trial's primary endpoint was six months, and the longest single-centre follow-up published here is a few years. It does not show that every stopped heart can be used: donors were young, controlled and selected, and roughly half of referrals did not produce a transplant. The 30 percent figure is one hospital's experience and one institute's projection, not a measurement of the world's donor pool, and no primary source fetched for this story counts how many patients have received a DCD heart worldwide. The randomised trial was funded by the device's manufacturer.

What the sequence does show is that an exclusion rule half a century old rested on an untested assumption, that a laboratory spent twelve years on a preservation solution and a surgical team spent four hours watching lactate fall before anyone knew whether a stopped heart could be given to someone else, and that it could.

The organ pool grew a third because dead hearts came back.