Giving all women who need a forceps or ventouse delivery a single dose of antibiotics almost cuts the risk of infection in half, new research shows.
Experts said current guidelines from the World Health Organisation (WHO) should be updated, so that women needing an assisted delivery are given a preventative dose of antibiotics.
Such a move would cut infections in new mothers by more than 7,000 in the UK every year, including life-threatening sepsis, they said.
The study found that only about one in 10 women (11%) suffered an infection if given antibiotics, compared with about one in five (19%) of those given a placebo drug.
Overall, the use of antibiotics actually dropped, as fewer women went on to need them for infections.
The new research, published in The Lancet medical journal, was based on a clinical trial carried out at 27 UK maternity units.
Women were randomly split into two groups, with the first group of 1,715 women given a single dose of intravenous amoxicillin (a type of penicillin) and clavulanic acid as soon as possible and no more than six hours after giving birth.
The second group of 1,705 women were given a placebo (saline).
Overall, 65% of the babies were delivered by forceps and 35% by ventouse, with 89% of all women needing an episiotomy (a cut to make the vaginal opening bigger).
Researchers then looked at data on whether women had a confirmed or suspected infection within six weeks of having their baby.
Of those women with complete data at the end of the study, 180 women in the amoxicillin and clavulanic acid group had a confirmed or suspected infection (11% of 1,619 women).
This compared with 306 women in the placebo group (19% of 1,606 women) – showing antibiotics cut the risk of infection almost in half.
From the infections, there was a 56% reduction in cases of sepsis, with 11 cases in the antibiotic group compared with 25 cases in the placebo group.
The study also found that for every additional 100 doses of antibiotic given as a preventative (prophylactic) measure, 168 doses could be avoided due to fewer post-delivery infections.
Adopting a policy of giving all women in this group antibiotics could actually reduce overall antibiotic use by 17%, researchers said.
Professor Marian Knight from the University of Oxford, who led the research, said: “These findings highlight the urgent need to change current WHO antibiotic guidelines and other guidance from organisations in the UK, North America, and Australasia, that do not recommend routine antibiotic prophylaxis for assisted childbirth.
“Pregnancy-associated infection is a major cause of death and serious illness.
“Almost one in five women develop an infection after assisted vaginal delivery and our results show that this could be reduced by almost half by a single dose of prophylactic antibiotic.”
The study is the largest of its kind to date.
Women having a caesarean section are already routinely given antibiotics as they are shown to reduce wound infection, inflammation of the womb and serious maternal infection by 60% to 70%.
The new study found that women given antibiotics had less chance of stitches getting infected or bursting, and had less pain and less chance of needing intimate care in that area.
Those women given antibiotics were also much less likely to need a visit from a GP, nurse or midwife, or to go to hospital due to the way their wound was healing.
The researchers said the total average NHS costs six weeks after birth were £52.60 less per women in the antibiotic group compared with women given placebo (£102.50 compared with £155.10).
One woman in the placebo group reported a skin rash and two women in the antibiotic group reported other allergic reactions, one of which was reported as a serious adverse event.
In 2016, an estimated 19,500 women died because of pregnancy-related infections around the world, the researchers noted.
Even in high-income countries, infections account for one in 20 maternal deaths.
And for every woman who dies from pregnancy-related infection, another 70 women develop a severe infection that can cause long-term problems.