Read our COVID-19 research and news.

Lambs delivered prematurely are kept afloat—literally—in a sealed, fluid-filled “biobag” that mimics what mom would provide.

Children's Hospital of Philadelphia

Fluid-filled ‘biobag’ allows premature lambs to develop outside the womb

Overcoming engineering, biology, and technology obstacles, a team of researchers has crafted what may be the best artificial womb yet: a fluid-filled bag in which lambs born early can live for up to 4 weeks, before being ushered into the outside world. Although others have designed similar systems that are still in animal testing, this one is notable for its stripped-down simplicity.

The result is a sealed “biobag” with one tube supplying artificial amniotic fluid and another draining it out. Although some lambs experienced complications, and human testing is, at best, several years off, the advance is generating excitement among those who care for pregnant women and their extremely premature babies.

“What they’ve got is a system where the fetus is really existing very much as it would in the mother’s womb,” says Anna David, a maternal-fetal medicine specialist at University College London (UCL). “The fetus knows what to do,” she adds, noting that as best they could, the physicians stepped aside and ceded control.

Fetuses are in a unique environment before birth. The placenta provides nutrients and oxygen, and the fetus’s lungs do not breathe air. They float in amniotic fluid, which is swallowed by the fetus and created by fetal urination; each day, the amniotic fluid “turns over” in this way. For babies that enter the world extremely early—as about 90,000 in the United States and Europe do each year—the prognosis can be grim, with survival ranging from 10% to 50%, and high rates of brain damage, lung disease, and other serious complications.

One of the biggest challenges is that the very technology that saves some of these premature babies, such as mechanical ventilation, also hurts them, for example by damaging their fragile lungs or halting lung development. In the past, researchers have tried to develop artificial wombs that mimic the prebirth environment, but these, too, have been intervention-heavy. In particular, they’ve relied on external pumps to push blood through an oxygenator circuit for gas exchange, which in turn can cause imbalances in fetal blood flow and lead to heart failure.

In the new study, reported today in Nature Communications, researchers at the Children’s Hospital of Philadelphia (CHOP) in Pennsylvania went through several iterations before hitting on technology that largely worked. Their final subjects were eight lambs delivered by cesarean section at about 110 days gestation—the equivalent of 23 or 24 weeks in humans, which is at the edge of viability.

The CHOP team, headed by fetal and pediatric surgeon Alan Flake, began with a high-tech version of a dumpster dive: collecting components doctors at the hospital no longer needed from a system called extracorporeal membrane oxygenation (ECMO). ECMO oxygenates the blood of critically ill infants and children.

For their first try, Flake’s group delivered several lambs shortly before they should have been born. The researchers bathed the lambs in artificial amniotic fluid, with electrolytes that mimicked the real thing, and connected the animals to an oxygenator. The goal was to see whether the system could sustain the lambs. The strategy worked unexpectedly well, with one animal surviving 108 hours, but sepsis and other complications made clear it wasn’t good enough.

Over months, Flake, along with fetal physiologist Marcus Davey, research fellow and surgeon Emily Partridge, and their colleagues kept tinkering. What they ended up with closely mimics biology: an approach that exchanges amniotic fluid rather than recirculating it, a sealed system to keep the outside world at bay, and a pumpless circuit for oxygenation of the blood that’s connected to the lamb fetus through the umbilical cord. The fetus’s heart drives circulation, which keeps blood pressure and other markers at normal levels.

After 4 weeks—the planned length of the experiment—the prematurely delivered lambs were taken off the system and put on artificial ventilators. By and large they did well, though with some modest complications including lung inflammation. Most were sacrificed for further study of how the artificial womb had influenced their development, but some were bottle-weaned, with the longest survivor now more than a year out. Flake is working with U.S. Food and Drug Administration to design an animal trial in accordance with the agency’s standards; he estimates that human testing is at least 3 years off.

If the new technology conjures up visions of The Matrix, with fields of fetuses growing in artificial wombs, Flake says that he’s not in the science fiction business: His goal is to help premature infants already being born, not push the limits of viability even earlier. “You go earlier [and] you’re very likely, in my mind, to open a can of worms,” he says. In part that’s because the system is just not designed to support even younger fetuses whose organs are more underdeveloped. Plus, trying to help them risks creating survivors with a poor quality of life—exactly the outcome Flake wants to avoid.

Experts are encouraged by the study. “This is just a much lower stress environment to support” the growth of premature babies than what’s offered now, says Tippi Mackenzie, a fetal and pediatric surgeon at the University of California, San Francisco, who trained with Flake but was not involved in the work.

There remain many uncertainties. One, Mackenzie says, is how fetuses battling an infection will fare in an artificial womb; infection is a common driver of preterm birth. Another is the long-term impact of an entirely new approach to caring for these babies. A clinical trial should closely track babies for some time, ideally a couple years, to assess how the system compares to the standard of care—not just in terms of survival, but in lingering health problems extremely premature babies can experience, says Neil Marlow, a neonatologist at UCL.

Despite the unknowns, “If it works, we should get on with it,” says Marlow, who like others agrees that the current setup for these children is far from ideal. “I think it’s just going to revolutionize [neonatal] care,” Mackenzie adds. “I don’t think that’s too much to say.”