Babies whose mothers use a type of anti-depressant during pregnancy are at risk of being born with withdrawal symptoms, research suggests.
Babies can be affected by their mother's drugs
Spanish research found use of selective serotonin reuptake inhibitors (SSRIs) was associated with neonatal withdrawal syndrome, the Lancet reports.
The researchers say doctors should avoid prescribing SSRIs to pregnant women, or use them cautiously.
SSRIs have become a standard treatment for depression in the past seven years.
However, it is known that coming off the drugs can cause symptoms of withdrawal in the user.
The latest study by a team at the University of La Laguna in Spain suggests that unborn babies could also be at risk from their mothers' use of the drugs.
Symptoms of neonatal withdrawal syndrome include convulsions, irritability, abnormal crying and tremor.
The researchers screened the World Health Organization database of adverse drug reactions for cases of neonatal convulsions and neonatal withdrawal syndrome associated with the use of SSRIs.
They found that by November 2003, a total of 93 cases of SSRI use associated with either neonatal convulsions or withdrawal syndrome had been reported.
Of these cases 64 were associated with Seroxat (paroxetine), 14 with Prozac (fluoxetine), nine with sertraline and seven with citalopram.
The dose was only reported in 13 of the cases associated with paroxetine and ranged from 10mg to 50mg per day.
The duration of treatment was reported only in eight cases. These ranged from four months to five years.
Lead researcher Professor Emilio Sanz said: "The results suggest that symptoms of withdrawal might be a greater problem for paroxetine than for other drugs.
"Paroxetine should not be used in pregnancy, or if used, it should be given at the lowest effective dose.
"With the other SSRIs, especially citalopram and venlafaxine, their use should be carefully monitored and new cases promptly communicated to drug vigilance systems."
More work needed
Also writing in the Lancet, researchers at Yale University School of Medicine said it would be wrong to assume that neonatal withdrawal syndrome was only associated with paroxetine use.
Dr Vladislav Ruchkin said: "It remains to be seen whether Sanz and colleagues' report ultimately reflects a minor problem for a particular antidepressant, or further evidence of a larger set of serious problems for SSRI use in young people.
"From a pessimistic extreme, these reports might jointly herald the beginning of the end for the uncontested SSRI hegemony of the past decade."
He said that until further research replicated or refuted the findings, it would be best to focus on non-drug therapies.
Dr Patrick O'Brien, a consultant obstetrician at University College London Hospital, told BBC News the paper was interesting.
"There is an increasing tendency to prescribe these drugs in pregnancy because of the perception they are completely safe," he said.
"The strength of this study is that it is has uncovered so many cases that one would have to accept that it suggests a link.
"But the weakness is that nobody has any idea how common these problems are in women taking these drugs in pregnancy. Are we talking about a risk of one in a million, or one in 10?"
Dr O'Brien said doctors had to weigh up the potential risk of prescribing the drugs, with the need to treat women suffering from depression.
"My advice is that if you can avoid medication in pregnancy do, but without a doubt there will be women who should be taking medication for depression during their pregnancy," he said.
In December 2003 regulators told doctors to stop prescribing the majority of SSRIs to children amid fears that they could make young patients suicidal.