Young children who are given repeated courses of antibiotics are at greater risk than those who use fewer drugs of becoming obese, US researchers say.
The JAMA Pediatrics report found children who had had four or more courses by the age of two were at a 10% higher risk of being obese.
But scientists warn this does not show antibiotics cause obesity directly and recommend children continue using them.
Many more studies are needed to explain the reasons behind the link, they say.
US researchers from the University of Pennsylvania and Bloomberg School of Public Health reviewed the health records of more than 64,500 American children between 2001 and 2013.
The children were followed up until they reached five years of age.
Almost 70% of them had been prescribed two courses of antibiotics by the time they were 24 months old.
But those who had four or more courses in this time were at a 10% higher risk of being obese at the age of five than children who had been given fewer drugs.
And the type of antibiotics they were prescribed appeared to make a difference too - those given drugs targeted at a particular bug were less likely to put on weight.
But those given a broad-spectrum antibiotic - that can kill several types of bacteria indiscriminately - were more likely to have a higher body mass.
Prof Charles Bailey at the University of Pennsylvania, said: "We think after antibiotics some of the normal bacteria in our gut that are more efficient at nudging our weight in the right direction may be killed off and bacteria that nudge the metabolism in the wrong direction may be more active."
And researchers say the study highlights that over prescribing inappropriate antibiotics could have a negative impact on child growth.
Prof Nigel Brown, president of the Society for General Microbiology in the UK, said: "This study adds further evidence that the use of antibiotics early in life has a role to play in obesity.
"While antibiotic use is only one factor that may predispose children to be obese, the study emphasises the importance of rapid diagnostic tests that allow precise targeting of antibiotics, which will kill the disease-causing bacteria and cause minimum disruption to the normal gut flora."
And Prof Bailey acknowledged his study had limitations as they were not able to look at the children's weight or exercise regimes.
He says the team will now start to explore what influence lifestyle factors has on these findings.
But Dr Graham Brudge, at the University of Southampton, said: "The design of the study did not allow testing as to whether antibiotic use during infancy causes obesity in childhood, only that there may be an association.
"It would be a concern if parents took from this that they ought to be reluctant to allow antibiotic use in their children.
"The key risk factors for childhood obesity are over-consumption of high energy, nutrient-poor foods and lack of exercise."
Meanwhile in a separate study, scientists reporting in the journal of the American Society for Microbiology found that a species of gut bacteria - called Clostridium ramosum - could promote weight gain in mice.
Mice with these bacteria present in their guts became obese when fed a high-fat diet, while those that did not have the bacteria put on less weight despite being given high-calorie meals.
The scientists, from the German Institute of Human Nutrition, in Nuthetal, are now trying to understand how the bacteria interact with digestion.