Antibiotics normally don’t increase the risk of autoimmune disorders

Children exposed to antibiotics in the womb or early in life don’t generally have an increased risk of developing autoimmune conditions in adolescence, according to a study of more than 6 million children. But the relationship is complicated.

The idea that antibiotics might contribute to autoimmunity can be traced back to the 1980s, when David Strachan, then at the London School of Hygiene and Tropical Medicine, suggested that fewer infections and cleaner environments may explain rising rates of childhood allergies.

This led to the development of the hygiene hypothesis. This proposes that early exposure to certain microbes trains the immune system, preventing it from overreacting to harmless substances and causing allergic reactions or autoimmune conditions. These conditions, such as type 1 diabetes, inflammatory bowel disease and lupus, occur when immune cells inadvertently attack the body’s own tissues.

Since then, numerous studies have shown that many microbes – especially those in our guts – shape our immune system. For instance, some produce compounds that are essential for the development of specialised immune cells, called regulatory T cells, which help prevent autoimmunity. This has led to questions about whether antibiotics, which significantly alter the gut microbiome, may contribute to the development of autoimmune conditions.

“Over the years, there has been a lot of pre-clinical work done, mostly in animal models, that strongly supports this notion that antibiotics, or modulation of the gut microbiome, have a major impact on our immunity,” says Martin Kriegel at the University of Münster in Germany.

For example, a 2016 study in mice showed that repeated use of antibiotics early in life raises the risk of type 1 diabetes. Mice genetically susceptible to the condition were exposed to an antibiotic three times – once through their mother’s breastmilk and again at 4 and 5 weeks old. Around 50 per cent of male mice and 80 per cent of female mice in this group developed type 1 diabetes by 30 weeks of age. By comparison, only about 25 per cent of male mice and 50 per cent of female mice that weren’t exposed to antibiotics developed the condition.

Human studies have also hinted at a link. An analysis of more than 10 million people published earlier this year found that those prescribed an antibiotic had a 40 per cent greater risk, on average, of later being diagnosed with inflammatory bowel disease. Meanwhile, a 2019 study of more than 110,000 people found that antibiotic prescriptions were associated with a 60 per cent higher chance of developing rheumatoid arthritis.

But other research has shown the opposite. For instance, a 2017 study of more than 15,000 children tested for type 1 diabetes or coeliac disease found no association between the development of these conditions and the use of antibiotics before 4 years of age.

Most recently, Eun-young Choi at Sungkyunkwan University in South Korea and her colleagues tracked incidences of six autoimmune conditions – type 1 diabetes, chronic childhood arthritis, ulcerative colitis, Crohn’s disease, lupus and Hashimoto’s disease – in more than 2.6 million children whose mothers had an infection during pregnancy. About 1.5 million of their mothers were prescribed antibiotics while pregnant. The researchers then did the same in a separate group of 3.4 million children who had an infection within six months of being born, 1.9 million of whom were treated with an antibiotic.

After accounting for factors such as infection type, socioeconomic status and sex, the researchers found no overall association between children’s exposure to antibiotics in utero or early childhood and later risk of developing an autoimmune condition in adolescence.

So why is there so much discrepancy between studies? The problem is, the gut microbiome is extremely complex. A wide variety of factors influence it, making it nearly impossible for researchers to account for them all. For instance, none of the studies mentioned above adjusted for diet, which can significantly alter gut microbiota.

Different antibiotics may also have different effects. For instance, Choi and her team did find an association between the use of broad-spectrum antibiotics during pregnancy and children later developing Crohn’s disease. Participants’ age at exposure also seemed to matter. Those prescribed antibiotics within two months of birth had a 30 per cent greater risk of Hashimoto’s disease.

This doesn’t mean we should forgo antibiotics. “When antibiotic use in pregnancy is recommended, it is because evidence suggests that the benefits outweigh the possible risks,” Christopher Zahn at the American College of Obstetricians and Gynecologists in Washington DC. For example, urinary tract infections raise the risk of premature birth, low birth weight and serious complications, such as sepsis, he says.

Some antibiotics may even protect against autoimmunity. A 2018 study found that infection with the pathogenic bacteria Enterococcus gallinarum triggered autoimmunity in mice predisposed to such conditions. Treating them with antibiotics not only prevented them from dying but also stopped their immune cells from attacking their body.

“So it is all just terribly complex,” says Kriegel.

Still, the most recent findings should provide reassurance about antibiotics for those who are pregnant or have young children, says Zahn.