Of the 14 antidepressants studied, only fluoxetine (Prozac) was more effective in treating depression than an inactive placebo in children and teens, the review found.
And Effexor (venlafaxine) was linked to a higher risk of suicidal thoughts and attempts compared to a placebo and five other antidepressants, the researchers reported.
"In the clinical care of young people with major depressive disorder, clinical guidelines recommend psychotherapy -- especially cognitive behavioral therapy or interpersonal therapy -- as the first-line treatment," said study author Dr. Andrea Cipriani. He is an associate professor in the department of psychiatry at the University of Oxford, in England.
Major depression affects about 3 percent of children aged 6 to 12, and 6 percent of teens aged 13 to 18, the researchers noted.
"Prozac should be considered only for patients who do not have access to psychotherapy or have not responded to non-pharmacological interventions," Cipriani said.
Because the available studies of antidepressant use among children and adolescents are limited and of questionable quality, "we should not underestimate these potential risks," he added.
"Children and adolescents taking antidepressant drugs should be closely monitored regardless of the treatment chosen, particularly at the beginning of treatment," Cipriani advised.
In 2004, the U.S. Food and Drug Administration issued a "black box warning" about antidepressant use among teens and children because of fears that these drugs might increase the risk of suicidal thoughts and suicide attempts.
Despite the FDA warning, the use of antidepressants in these age groups crept up between 2005 and 2012. For example, the proportion of kids and teens taking an antidepressant rose from more than 1 percent to nearly 2 percent in the United States, Cipriani said.
One expert was not surprised by the findings.
This study shows what has been known -- that these "medicines look less effective and riskier in children and adolescents than they do in adults," said Dr. Peter Kramer. He is a clinical professor emeritus of psychiatry and human behavior at Brown University, in Providence, R.I.
"Among them, Prozac has always stood out as relatively more effective," Kramer said.
But it isn't clear how antidepressants work in children, Kramer noted. "We just know so little about these medicines and what they do in children with developing brains," he pointed out.
In general, treatment should begin with psychotherapy, but for some patients Prozac might be an option, he suggested.
"There are a lot of desperate cases where the illness is so destructive that doing something that's highly imperfect may seem like the right move," Kramer added.
For the study, Cipriani and his colleagues reviewed 34 studies that included more than 5,200 children and teens. This kind of study, called a meta-analysis, tries to find common ground among numerous trials. Its limitations are that the conclusions rely on how well the studies that are included were done.
Moreover, most of the trials (65 percent) were financed by drug companies. And 90 percent had a risk of being biased in favor of the medication, Cipriani said.
The investigators found that only with Prozac did the benefits outweigh the risks in terms of relieving symptoms with few side effects.
The antidepressants Effexor, imipramine (Tofranil) and duloxetine (Cymbalta) had the worst side effects, causing more patients to stop taking them than those taking a placebo, the review authors found.
The review was published online June 8 in The Lancet.
"This study gives us real concern about the usefulness of antidepressants," said the author of an accompanying journal editorial, Dr. Jon Jureidini.
With a meta-analysis, the benefits of antidepressants may be overstated and the harms understated, said Jureidini. He is a research leader at the Robinson Research Institute of the University of Adelaide in Australia.
"Unfortunately, we cannot rely on the truthfulness of how the information has been processed," Jureidini said. "So we can't even be confident about using Prozac in children and adolescents."
Jureidini said that before prescribing an antidepressant, a doctor must be sure the benefits will outweigh the harms.
"We should shift our practice away from prescribing and toward other treatments, because drugs are being oversold to us," he said.
Young children should never be given antidepressants and they should be given sparingly to teens, Jureidini said. "I would almost never use them," he said.
Jureidini said that treatment starts with understanding what is troubling the child or teen, and why they have withdrawn from social activities, school and friends.
"We need to help young people get back into their lives, and their depression will get better," he said. "They need help and support to re-engage with the world."
Visit the U.S. National Institute of Mental Health for more on teens and depression.
Copyright © 2016 HealthDay. All rights reserved.