For all the hype that surrounds them, probiotics—products that contain supposedly beneficial bacteria—have rarely proven their worth in large, rigorous studies. There are good reasons for this disappointing performance. The strains in most commercially produced probiotics were chosen for historical reasons, because they were easy to grow and manufacture, and not because they are well-adapted to the human body. When they enter our gut, they fail to colonize. As I wrote in my recent book, they’re like a breeze that blows between two open windows.
But even though probiotic products might be underwhelming, the probiotic concept is sound. Bacteria can beneficially tune our immune systems and protect us from disease. It’s just a matter of finding the right strains, and helping them to establish themselves. Many scientists are now trying to do just that, and one such team, led by Pinaki Panigrahi at the University of Nebraska Medical Center, has just scored a big win.
Since 2008, Panigrahi’s team has been running a large clinical trial in rural India, where they gave a probiotic of their own devising to thousands of randomly selected newborn babies. Their product contained a strain of Lactobacillus plantarum, chosen for its ability to attach to gut cells. The team also added a sugar, chosen to nourish the microbe and give it a foothold when it enters a baby’s gut. Together, this combination is called a synbiotic. And it was strikingly effective.
The team found that babies who took this concoction had a significantly lower risk of developing sepsis—a life-threatening condition where infections trigger body-wide inflammation, restricted blood flow, and organ failure. Sepsis is one of the biggest killers of newborn babies, ending around 600,000 lives every year when they’ve barely begun. Some proportion of these cases begin in the gut, and probiotics might be able to prevent them by ousting harmful microbes, or by stopping benign ones from crossing into the bloodstream and causing infections.
Sure enough, in Panigrahi’s trial, just 5.4 percent of the infants who took the synbiotic developed sepsis in their first two months of life, compared to 9 percent of those who received a placebo. That’s a reduction of 40 percent. Such estimates always come with a margin of error, but the team calculate that the reduction in risk should still be somewhere between 25 and 50 percent.
The effect was twice as large as what the team expected, especially since the infants took daily doses of the synbiotic for just one week. And given the clear evidence of benefits, independent experts who were monitoring the study decided to stop the trial early: It would have been unethical to continue depriving half the newborns of the treatment. Panigrahi originally planned to enroll 8,000 babies into the study. He stopped at 4,557.
Which is still a huge number! Probiotics trials have been criticized in the past for being small and statistically underpowered. Those that looked at sepsis, for example, usually involved just 100 to 200 babies, making it hard to know whether any beneficial effects were the result of random chance. The biggest trial to date included 1,315 infants; Panigrahi’s study is over three times bigger. “[It] exemplifies how intervention research should be done,” writes Daniel Tancredi from the University of California, Davis, in a commentary that accompanies the paper.
“In most studies, people take the probiotics that are available on the shelf without asking why that probiotic should work in the disease they’re interested in. And they think they’ll stumble onto something good,” says Panigrahi. “It’s counter-intuitive, but we did the same thing.”
At first, his team tested Lactobacillus GG and Lactobacillus sporogenes—the most commonly used probiotics in India—in small pilot studies. Both strains are claimed to colonize the gut. “We did the trial and the colonization was almost zero,” says Panigrahi. To find more suitable strains, the team collected stool from healthy volunteers and screened the microbes within for those that could stick to human cells, and could prevent disease-causing bacteria from doing so. They ended up with a strain called Lactobacillus plantarum ATCC strain 202195, which not only colonized infant guts successfully, but stayed there for up to four months. That’s when they launched the big trial.
Aside from preventing sepsis, it also reduced the risk of infections by both the major groups of bacteria: the Gram-positives, by 82 percent; and the Gram-negatives, which are harder to treat with antibiotics, by 75 percent. It even reduced the risk of pneumonia and other infections of the airways by 34 percent. That was “completely unexpected,” says Panigrahi, and it’s the result he’s especially excited about. It suggests that the synbiotic isn’t just acting within the gut, but also giving the infants’ immune systems a body-wide boost.
Probiotics are not without risk. There have been rare cases where the bacteria in these products have caused sepsis in newborn or preterm infants. But Panigrahi saw no signs of that in his study: His synbiotic didn’t seem to cause any harmful side effects.
Beyond protecting infants, Panigrahi says that this approach would also reduce the use of antibiotics, and slow the spread of drug-resistant infections. And perhaps best of all, it can be done cheaply. You’d need to treat 27 infants to prevent one case of sepsis, and each week-long course costs just one U.S. dollar.
“It’s a very important study,” says Marie-Claire Arrieta from the University of Calgary. “It not only shows an effective and low-cost way to prevent a horrible infant disease that kills millions worldwide, but provides important clues on how to improve strategies to change the infant-gut microbiome.”
Two earlier trials tested off-the-shelf probiotics on 1,099 and 1,315 premature infants respectively. Neither found any benefits for sepsis. Nor did an Indian trial involving 668 babies born with a low birth weight. In retrospect, such failures were to be expected. Sepsis is a varied and complicated condition. The microbiome is also incredibly varied in early life, and changes in ways we barely understand. “It’s not surprising that a one-size-fits-all approach hasn’t worked thus far,” says Arrieta. Success probably depends on choosing the right strain, administering it at the right time, and feeding it appropriately.
Then again, Panigrahi’s trial only included healthy newborns of normal weight, whose mothers had begun to breastfeed them. They already had the best odds of fighting off infections, so it’s unclear if his synbiotic would work equally well with weaker or smaller babies, who are more prone to sepsis. It’s also unclear exactly why the synbiotic worked, or what effect it might have on the infants’ microbiomes in the long run.
“We may need to test this in different settings and we’re working with the government to do so,” says Panigrahi. “But this should be the standard of care. The money involved is very small. The synbiotic can be manufactured anywhere without fancy technology. And it can do so much good.”