Antidepressants Do Not Increase Risk of Mania in Children and Adolescents

Key Highlights

New research led by Suvi Virtanen, PhD, suggests that antidepressants do not induce mania or hypomania in pediatric patients with unipolar depression.
Prior to this research, there were few studies on the effects of antidepressants on children and adolescents with unipolar depression. Investigating these rare outcomes, such as incident mania, has been challenging in randomized clinical trials (RCTs).
The investigators used nationwide data from Sweden to examine incident mania or hypomania in pediatric patients. They conducted an observational study with a 12-week time frame, extending it by 4 weeks to account for potential slower medication titration in children.
Participants in the study were aged 4-17 and diagnosed with unipolar depression. They underwent a "washout period" where they didn't take antidepressants for a year before the depression diagnosis, except for those with a prior mania diagnosis.
The treatment group included participants who took antidepressants within 90 days of diagnosis, while the control group did not take antidepressants during that period.
At 12 weeks, the cumulative incidence of mania/hypomania was 0.26% in the treatment group and 0.20% in the control group. By 52 weeks, these figures increased to 0.79% and 0.52%, respectively. Boys in the treatment group had a slightly higher risk of mania.

Antidepressants do not induce mania or hypomania for pediatric patients with unipolar depression, according to new research.¹

In a new study led by Suvi Virtanen, PhD, of the department of medical epidemiology and biostatistics at Karolinska Institutet in Stockholm, an international team of investigators wanted to see if antidepressants cause mania or hypomania in pediatric patients with unipolar depression.

Previously, there had been few studies on children and adolescents on this topic. “It is critical to understand potential risks related to anti- depressants in pediatric patients, but rare outcomes, such as incident mania, are difficult to investigate in randomized clinical trials (RCTs),” investigators wrote. “RCTs often have insufficient sample sizes and strict inclusion criteria, which correspond poorly with the characteristics of the population actually receiving the medication in clinical practice.”

The investigators used nationwide data registered in Sweden from July 2006 - December 2019 to investigate incident mania or hypomania in pediatric patients. The investigators also drew their research from other sources, such as the National Patient Register for diagnoses research and the Prescribed Drug Register for prescribed drugs.

The investigators ran an observational study that had a 12-week time frame—they added an extra 4 weeks to The International Society for Bipolar Disorders Task Force’s customary 9 weeks to “account for a potentially slower titration schedule in pediatric patients.”

For the study, the investigators identified participants who had an increased risk of developing mania. The study cohort had participants aged 4-17 years old, all diagnosed with unipolar depression. Participants went through a “washout period” where they did not take any antidepressants for a year prior to their depression diagnosis—although this excluded participants with a prior mania diagnosis.

While the treatment group consisted of participants who took antidepressants within 90 days of the diagnosis, the control group did not take any antidepressants within 90 days of the diagnosis.

By 12 weeks, the estimated cumulative incidence of mania/ hypomania was 0.26% in the treatment group and 0.20% in the control group.

Yet, by 52 weeks, the treatment group rose to 0.79% and the control group went up to 0.52%. Virtanen and colleagues found boys in the treatment group had a higher risk of mania than the girls. Investigators found that there was no evidence that proved antidepressants increased the risk of mania.

“The risk was similar in treatment and control groups by 12 weeks, which is the time window during which treatment-emergent mania is expected to occur,” investigators wrote. “We found a small risk difference only in the longer follow-up analysis, suggesting that the risk increase may be attributable to factors other than the medication.”

In an editorial accompanying the study, authors Krista Huybrechts, MS, PhD, of the division of pharmacoepidemiology and pharmacoeconomics in the department of medicine at Brigham and Women’s Hospital and Harvard Medical School, and Mark Olfson, MD, MPH, of the department of psychiatry in the College of Physicians and Surgeons at Columbia University and New York State Psychiatric Institute, provided additional context to the meaning of Virtanen and colleagues’ findings.²

“The risk difference…provides important additional context from a public health point of view: the 29% increased risk on the relative scale corresponds to the 0.06 excess cases per 100 patients treated,” Huybrechts and Olfson wrote. “While this is not trivial considering the large number of children being treated with antidepressants, it nevertheless provides an important context when considering the risk-benefit trade-off.”

Huybrechts and Olfson found the study “reassuring” since it confirmed that manic episodes in children and adolescents occur at an equal frequency, whether they take antidepressants or not.

The study’s results conflict with a previous register-based study, which found that pediatric patients with unipolar depression, who were treated with antidepressants, had a substantially increased risk of mania.¹

This new study, however, used a target trial emulation framework to minimize common biases in observational cohort studies. While the previous studies followed up with patients up to 5 years and excluded outcomes within the first 2 months, the new study used a shorter follow-up period of 12 weeks.

The investigators said the study’s strength was its large size and target trial emulation framework, while the limitations were that the findings apply primarily to selective serotonin reuptake inhibitors, and therefore they did not have information on medication dosing.

“Similarly to RCTs, our cohort study found incident mania to be rare in the clinical practice setting,” investigators concluded. “Furthermore, RCTs have found no clear evidence of antidepressants increasing the risk of mania compared with placebo.”