There was also a small increase in problems such as poor eyesight or hearing.
Following the report, British government's chief medical officer, Sir Liam Donaldson, has written to GPs, obstetricians and other medical staff.
Sir Liam said in the letter that antibiotics should continue to be given to women in premature labour where there was evidence of infection, or risk of it because their waters had broken.
"Antibiotics save lives and pregnant women with possible or obvious infections must be considered for treatment with antibiotics," the Independent quoted the letter, as stating.
But it added that the drugs should "not routinely be given" to women in premature labour where there was no sign of infection and whose waters had not broken, in line with "existing good clinical practice".
A clinical trial, which was published in 2001, involving 12,000 women in the UK and around the world, found antibiotics delayed labour and improved outcomes for mother and babies at risk of giving birth prematurely.
The latest study, which was a follow-up study seven years later of more than 8,000 of the women in the UK, published in The Lancet, has found a puzzling difference between two groups in the original study.
Among women whose waters had broken, giving antibiotics caused no long-term benefit or harm to their children, despite the improved outcome at birth revealed in the first study.
But among women whose waters had not broken, giving antibiotics was associated with up to a threefold rise in cerebral palsy, from 1.6 per cent in those given placebo to 4.4 per cent in those on two antibiotics, and a small increase in "mild" functional impairment such as poor co-ordination and poor eyesight.