Introduction

Treatment-resistant depression in children and adolescents is defined as clinically significant depression that has not responded to treatment with a selective serotonin reuptake (SSRI) for at least 8 weeks by the TORDIA (Treatment of SSRI-Resistant Depression in Adolescents) trial ( ). define treatment-resistant depression in youth as clinically significant depression that fails to respond to an adequate treatment trial of one evidence-based antidepressant or one adequate psychotherapy trial, either in two separate treatment trials or used in combination. They further define treatment-refractory depression as two failed trials of evidence-based antidepressant treatment and a failed adequate psychotherapy trial. A ketamine study in adolescents defined treatment-resistant depression as a failure to respond to at least two antidepressant medications ( ). Based on a systematic review of interventions for treatment-resistant depression in adolescents, proposed that treatment-resistant depression be defined as failure to respond to two evidence based trials of SSRI antidepressants of adequate dose and duration in combination with psychotherapy. The lack of a consistent definition for treatment-resistant depression in children and adolescents is problematic and can be confusing for clinicians trying to critically appraise the evidence and find appropriate treatments for their patients.

Depression in youth

A diagnosis of major depressive disorder in children and adolescents is made using the same Diagnostic and Statistical Manual, 5th edition (DSM 5) diagnostic criteria for adults, with the exception of irritability as a possible mood state in children and teens, even in the absence of depressed mood ( ). The lifetime prevalence of depression in adolescents has been reported to be approximately 12% ( ). National Survey on Drug Use and Health data has shown an increase in the prevalence of major depressive disorder from 8.1% to 13.2% among adolescents aged 12–17 years, from the year 2005 to the year 2017 ( ). This represents an increase of 52% in the prevalence of major depressive disorder in adolescents over the course of 12 years. Of the children and adolescents diagnosed with major depressive disorder, about 40% will not respond to initial treatment ( ).

Even in youth who do respond to the first treatments used, recurrence of depressive symptoms is also a significant problem. In a study examining long term outcomes of 140 adolescents with depression receiving treatment with therapy and/or medication, 93% of youth experienced full remission from their initial depressive episode, but most went on to experience recurrence of depression (53%) and development of a nonmood disorder (79%). Only 15% were reported as having no subsequent depressive episodes or nonmood disorders at follow-up (mean 6 years) ( ).

Depression in adolescence has been shown to be associated with significant psychosocial morbidity, including failure to finish school, unemployment, and pregnancy/parenthood in a metaanalysis of long-term outcome studies ( ). Suicide is a greater risk in patients who experience symptoms of depression prior to adulthood, and about 50% of depressed adults experience symptom onset before the age of 18 ( ). Suicide is now the second leading cause of death in individuals aged 10–34 years ( ). Research examining the association of suicidality in youth with treatment-resistant depression shows that youth with reduced right superior temporal gyrus volume may have a greater risk of suicide attempts, suggesting a possible biological marker to help assess risk in this population ( ).

More severe illness and the presence of suicidality are associated with poor response to treatment in children and adolescents with depression. Children with comorbid psychiatric disorders, increased levels of hopelessness, and increased family stress also have a higher risk of poor treatment response. Children with better global functioning prior to treatment may respond better to treatment, as well as children with depressive episodes that are shorter in duration ( ). Early treatment response has been shown to be a predictor of positive outcomes in pediatric depression and may also predict long-term response and recovery. In a metaanalysis of randomized placebo-controlled studies in children and adolescents, the greatest benefits of SSRIs occurred early in treatment and were minimal after 4 weeks ( ). The Treatment of Adolescents with Depression (TADS) study showed that early responders at 12 weeks of treatment were more likely to achieve full recovery at 3-year follow-up ( ).

Treatment-resistant depression studies

The TORDIA trial examined the efficacy of SSRIs, serotonin-norepinephrine reuptake inhibitors (SNRIs), and cognitive behavior therapy (CBT) in the treatment of pediatric SSRI treatment-resistant depression ( ). The study included 334 patients aged 12–18 years with treatment-resistant depression, defined as no response to at least 8 weeks of treatment with an SSRI at a dosage of at least 40 mg of fluoxetine or its equivalent. The participants were randomized to receive one of four different treatments for the course of the 12-week study. The four treatment groups included switching to a different SSRI, switching to a different SSRI plus CBT, switching to venlafaxine, or switching to venlafaxine plus CBT. Response was defined as a Clinical Global Impression-Improvement score (CGI-I) ≥ 2 ( ) and a reduction in the Children’s Depression Rating Scale-Revised (CDRS-R) ( ) score of at least 50% from baseline. Switching to either medication and adding CBT produced the highest response rate (54.8%). The response rate in the SSRI group (47%) did not differ significantly from the venlafaxine group (48.2%), but more adverse effects were observed in the venlafaxine group. Increased diastolic blood pressure and pulse and skin problems were observed at a significantly higher rate when compared to the SSRI group. Switching to a second SSRI and adding CBT is recommended in youth who have failed their first antidepressant trial, as SSRIs may have fewer adverse effects.

The TORDIA trial showed that remission at week 24 was associated with early response ( ). When comparing the participants who were in remission at 24 weeks with those who were not, the remitted group had significantly less severe symptoms by week 6 of the study. Early response to treatment may be predictive of remission at a much earlier stage than the recommended acute treatment duration of 12 weeks. An analysis of the TORDIA data using a Bayesian hierarchical model showed that switching to an SSRI vs venlafaxine produced a greater response and quicker time to response ( ). When examining these results in youth with significant comorbid anxiety, however, venlafaxine performed as well as the SSRIs, suggesting clinicians may consider switching to an SNRI earlier in treatment after a failed response to an initial SSRI in children with treatment-resistant depression and comorbid anxiety.

Long-term follow-up in the TORDIA trial showed that by 72 weeks, about 61% of the participants had achieved remission, with no difference between treatment groups in remission rate or time to remission ( ). Certain factors were associated with a lower chance of remitting, including substance use, severity of depression, and more impairment in functioning at baseline. Anhedonia was shown to be a predictor of poor recovery with SSRIs, with longer time to remission and fewer depression-free days when compared to other depressive symptoms ( ). Treatment nonadherence was found to be a common cause of treatment nonresponse ( ). A history of nonsuicidal self-injury and hopelessness were significant predictors for suicide attempts throughout the study ( ). Venlafaxine was associated with more treatment-emergent suicidality when compared to the SSRI group, a finding which further strengthened the recommendation to try a second SSRI prior to switching to an SNRI ( ).

Substance use was fairly common in the TORDIA trial, with about 28% of participants reporting repeated substance use (between three and nine times in the previous month), and youth who had lower substance use at baseline or who decreased their substance use during the study showed better response to treatment during the study ( ). Parent conflict was an important factor in response to treatment in the TORDIA trial ( ). Higher parent-reported conflict predicted lower rates of remission, though there was no effect seen on treatment response. Youth who received more than nine CBT sessions had a higher rate of response than those who had nine or less sessions, emphasizing the importance of an adequate number of sessions for treatment-resistant depression in youth ( ). Overall, the TORDIA trial showed a response rate of about 50% when switching to a second SSRI. With approximately 40% of depressed youth not responding to initial treatment (from acute depression treatment studies), this means that approximately 20% of depressed youth (half of initial nonresponders) will not respond to the first two treatments tried. There is inadequate evidence guiding clinicians in treatment decisions for these remaining youth.

Augmentation of antidepressants with antipsychotic medications has shown some benefit in adult depression studies, but there is currently very little evidence examining their efficacy in children and adolescents with depression, and no randomized controlled trials in this population. A case series examining the efficacy of quetiapine (median dose 200 mg/day) augmentation in 10 adolescents with treatment-resistant depression showed a 70% response rate (CGI-I ≥ 2) ( ). A case report described improvement in depressive symptoms with adjunctive clozapine for an adolescent with treatment-resistant depression ( ). Large, randomized controlled treatment studies are needed before recommending this as an evidence-based augmentation strategy in children and adolescents with depression. The potential side effects with antipsychotics are concerning, and children and adolescents seem to be at an increased risk of experiencing cardiometabolic side effects with the second-generation antipsychotics ( ). Other augmentation strategies using medications may include lithium, bupropion, and mirtazapine, but are based on adult studies, with no randomized controlled trials showing efficacy or safety of these strategies in the pediatric population ( ).

The FDA has approved repetitive transcranial magnetic stimulation (rTMS) as a treatment for treatment-resistant depression in adults, but it is not approved for use in the pediatric population. A randomized controlled trial examining the efficacy of TMS for treatment-resistant depression in adolescents included a double-blind design using a sham TMS treatment for the control group ( ). The trial included 103 adolescents, aged 12–21 years, with treatment-resistant depression. This was a 6-week study, with a total of 30 treatments given as monotherapy. Participants were randomized to receive left prefrontal 10 Hz TMS or sham TMS treatment (including a sham coil that appeared identical to the active coil, an acoustically matched profile, and a mild percussive sensation to mimic active treatment). The primary outcome measure was change in the Hamilton Depression Rating Scale (HAM-D-24) ( ) from baseline to endpoint. There was no significant difference between the active TMS treatment group and sham TMS treatment group on the primary outcome measure, with change in scores being − 11.1 and − 10.6, respectively. There were also no significant differences between TMS and sham TMS in response (41.7% and 36.4%, respectively) or remission rates (29.2% and 29.0%, respectively). Participants in the active treatment group reported side effects, including headaches, eye pain, nausea, and facial twitching, which were observed significantly more than in the sham treatment group.

A recent systematic review showed 14 studies examining the efficacy of TMS in adolescents with depression, eight of which were open-label, multisubject trials ( ). The other six studies were post hoc analyses of the open label studies and one 3-year follow-up of an open trial. The eight open-label studies included a total of 142 adolescents with depression, all of which showed TMS having some effect on reducing symptoms of depression. The review concluded the studies were of either poor or fair quality, with a high risk of bias. A recent randomized sham-controlled trial compared active TMS and sham TMS in 103 adolescents aged 12–21 with treatment-resistant depression. Each group received 30 daily treatments over 6 weeks. There was no significant difference between the two groups on the primary outcome measure, a change in the Hamilton Depression Rating Scale from baseline to endpoint ( ). More large randomized controlled trials are needed before routinely recommending TMS for adolescents with treatment-resistant depression.

There are no randomized controlled trials examining the efficacy of electroconvulsive therapy (ECT) in children and adolescents with depression. A retrospective chart review examined the efficacy of ECT in 54 adolescents with treatment-resistant mood disorders, including unipolar and bipolar depression ( ). The youth received their first ECT treatment course prior to the age of 18. Response was defined as a Clinical Global Impressions (CGI) score ≤ 2 and was assessed after initial treatment and at 1-year follow-up. The mean number of treatments was 13.7 for initial treatment and the response rate was 52.8%. The study showed that 82% of the youth were considered responders at 1-year follow-up, with a reduction in suicidal ideation and self-injurious behaviors. Prolonged seizures were common during ECT (74%) and long-term side effects were memory loss limited to the time surrounding ECT treatment. A retrospective observational study including 190 adolescents and young adults, aged 16–25 years, examined the efficacy of ECT for multiple psychiatric diagnoses, including 128 subjects with treatment-resistant depression and showed significant improvement on self-reported depressive symptoms as a group (the study did not report on percent of responders separately) ( ). A case series of 13 adolescents aged 15–18 years with treatment-resistant depression showed a mean of 14 ECT sessions to be effective for reducing depressive symptoms in 10 of the 13 (77%) participants ( ).

A retrospective study of ECT in adolescents with treatment-resistant depression showed an average improvement of 46% in the participants’ symptoms, with a third of participants experiencing at least 60% improvement in symptoms ( ). A retrospective case report showed significant improvement in 83% of adolescents (aged 14–17 years) with treatment-resistant depression receiving ECT. The study included six subjects, and ECT was well-tolerated by the adolescents in the study ( ). Side effects of ECT can include transient cognitive impairment, headache, prolonged seizures, and nausea. The American Academy of Child and Adolescent Psychiatry developed practice parameters for the use of ECT in adolescents ( ). The recommendation is that ECT should be considered for severe, persistent, and significantly disabling depression in adolescents who have failed at least two adequate trials of appropriate psychopharmacological agents along with other appropriate treatments.

Ketamine is another treatment being used in adults with treatment-resistant depression, but there are no randomized controlled studies in children and adolescents. An open-label study examined the efficacy of intravenous ketamine in 13 adolescents aged 12–18 years with treatment-resistant depression ( ). The participants were administered six ketamine (0.5 mg/kg) infusions over 2 weeks. Response (≥ 50% reduction in CDRS-R score from baseline to endpoint) was seen in five (38%) of the adolescents, with relapse occurring within 2 weeks in two of these responders. Adverse effects reported included transient blood pressure changes, transient (within 1 h) dissociative symptoms, dysphoria, and nausea. A systematic review on ketamine treatments for use in adolescents included the above study and two case reports in adolescents with treatment-resistant depression ( ). Ketamine was shown to improve depressive symptoms in some patients. The authors concluded that ketamine may have potential for treatment-resistant depression in youth but considering ketamine in treatment guidelines for adolescents with treatment-resistant depression is premature, given the limited evidence of efficacy and safety in this patient population. A commentary on esketamine use in adolescents raised concerns about repetitive and chronic exposures to esketamine in this patient population ( ). Animal studies of esketamine on developing brains show permanent negative effects on the neurodevelopmental process in the prefrontal cortex, which is a risk that could be heightened in children and adolescents with genetic risk factors for schizophrenia. Ketamine and esketamine lack sufficient efficacy and safety data to recommend for use in children and adolescents with treatment-resistant depression.

Adjunctive l -methylfolate was examined as a potential treatment for treatment-resistant depression in 10 adolescents (mean age 14) in a case series ( ). Eighty percent of patients had a single mutation among two methylene tetrahydrofolate reductase (MTHFR) gene variants, indicative of reduced MTHFR activity. The dose of l -methylfolate ranged from 2 to 15 mg/day. In the study, 80% of participants showed improvement in symptoms of depression, irritability, and anxiety. l -Methylfolate was well-tolerated. More rigorous studies are required in this patient population before l -methylfolate augmentation can be recommended for adolescent treatment-resistant depression.

A medication consensus treatment guideline was developed for major depressive disorder in youth ( ) but has not been updated. The recommendation was initial treatment with an SSRI, and then switching to a second SSRI if no response. This recommendation is now consistent with the findings of the TORDIA study. If there is a partial response to treatment with an SSRI, augmentation with a second antidepressant may be considered, although this has not been studied in children and adolescents. If there is no response to the second SSRI tried, an alternate antidepressant from a different class (venlafaxine, bupropion, mirtazapine, duloxetine) is recommended. Desvenlafaxine and vilazodone have now been studied in youth with depression and may be considered in the list of alternate antidepressants. If there is still no response, the clinician should reassess symptoms. Newer antidepressants should be used with caution due to limited safety and efficacy data in children and adolescents.

Antidepressant studies for acute treatment

It is important to use evidence-based treatments for acute depressive episodes in children and adolescents in order to decrease the risk of treatment-resistant cases. SSRIs are the medications with the most evidence supporting their use in the treatment of depression in children and adolescents. Studies examining the efficacy of SSRIs in acute pediatric depressive episodes led to Food and Drug Administration (FDA) approval for two SSRIs in this population. Fluoxetine is approved for children with major depression aged 8 years and older, and escitalopram is approved for adolescents with major depression aged 12 years and older. Three randomized controlled trials examining the efficacy of fluoxetine in the acute treatment of major depressive disorder showed it to be superior to placebo in children and adolescents ( ; ). Doses used in these studies ranged from 10 to 40 mg/day, with two studies using fixed dosing of 20 mg/day. Escitalopram has one positive randomized, placebo-controlled trial in adolescents, aged 12–17 ( ), which contributed to its FDA approval for adolescent depression. Dosing ranged from 10 to 20 mg/day. Sertraline and citalopram are not FDA approved for the treatment of pediatric depression, but they have shown benefit in randomized trials and may be considered in this patient population ( ).

Other antidepressants have been studied in children and adolescents with major depressive disorder, with negative results. Escitalopram was not found to be superior to placebo in a randomized controlled trial that included younger children, aged 6–17 years ( ). Paroxetine was not found to be superior to placebo on most outcome measures in the treatment of major depressive disorder in children and adolescents in three randomized controlled trials ( ; ; ). Citalopram did not separate from placebo in a randomized controlled trial in adolescent inpatients with major depressive disorder ( ).

Venlafaxine ER was not superior to placebo in the treatment of pediatric major depressive disorder in two placebo-controlled trials ( ). In two 10-week randomized controlled comparison trials of duloxetine in the treatment of pediatric major depressive disorder, duloxetine and fluoxetine were not superior to placebo ( ; ). The efficacy of desvenlafaxine versus fluoxetine for major depressive disorder in children and adolescents aged 7–18 was examined in an 8-week randomized controlled trial ( ). Neither the desvenlafaxine nor the fluoxetine treatment groups were shown to be superior to placebo. A metaanalysis of five trials, including 973 patients, found that SNRIs (duloxetine and venlafaxine) were not superior to placebo in the acute treatment of major depressive disorder in children and adolescents, aged 7–18 years ( ). Adverse effects reported in these studies included abdominal pain and dizziness in the venlafaxine studies and nausea, headache, and nasopharyngitis in the duloxetine studies, but showed the treatments to be generally well-tolerated.

Mirtazapine has had negative unpublished randomized controlled trials in children and adolescents with major depressive disorder ( ). A randomized controlled trial examined the efficacy of selegiline transdermal system in the treatment of major depressive disorder in 308 adolescents and did not show a difference between active treatment and placebo ( ). Vilazodone was examined as a potential treatment for 473 children and adolescents (aged 7–17 years) with major depressive disorder in an 8-week randomized, double-blind, placebo-controlled trial ( ). Treatment groups included vilazodone (15 or 30 mg/day), fluoxetine (20 mg/day), or placebo. Neither vilazodone nor fluoxetine were superior to placebo, but vilazodone was well-tolerated. Vilazodone was similarly not superior to placebo for the acute treatment of depression in another randomized controlled trial including adolescents with major depressive disorder ( ). A small, open-label trial of bupropion showed positive results in children with comorbid ADHD, but no large, controlled studies have been conducted in children or adolescents with major depressive disorder ( ). Vortioxetine has shown promise as a possible treatment for pediatric depression in an open-label study including youth with depressive or anxiety disorders, but no randomized-controlled trials have been completed with this medication ( ).

A recent metaanalysis of antidepressants in children and adolescents showed that fluoxetine compared to placebo is more efficacious than other antidepressants for major depressive disorder ( ). It also showed venlafaxine to be associated with an increased risk of treatment-emergent suicidality in depression studies. A metaanalysis of antidepressants in the treatment of pediatric depression showed a placebo response rate of 50% in treatment studies, which is high, especially when considering the antidepressant treatment response rates of 60% ( ). This makes it very difficult for active treatments to show efficacy in treatment studies.

Despite the number of negative trials of antidepressants, these studies provide important safety data which clinicians may find helpful when considering treatment with antidepressants in children and adolescents that are off label. When treating with medication, clinicians should ensure patients have an adequate trial of medication with adequate dosing. Doses of medications should be safely increased to maximum recommended doses in youth depending upon clinical response. In the Treatment for Adolescents with Depression (TADS) study, response and remission rates improved significantly between 12 weeks (acute treatment) and 36 weeks with fluoxetine treatment ( ). A discontinuation study including 102 youth with acute response to fluoxetine at 12 weeks, showed continuation treatment with fluoxetine to be superior to placebo in the 6-month follow-up study. Relapse rates were significantly higher in the placebo group (69.2%) compared to the fluoxetine continuation group (42%), with a shorter time to relapse in the placebo group ( ). A metaanalysis including 882 depressed youth showed that antidepressants significantly reduced the risk of relapse and recurrence of depressive episodes in children and adolescents, with a relapse/recurrence rate of 40.9% in the youth on antidepressants versus 66% in the youth on placebo ( ). Recommendations include keeping patients on treatments for at least a year after response is seen in order to avoid relapse ( ). Youth with higher number of recurrences and longer recovery periods should have a longer maintenance period of treatment, as should those with chronic depressive episodes, with recommendations to continue treatment longer than a year ( ).

Common side effects associated with SSRIs include gastrointestinal symptoms, headache, insomnia, somnolence, and dry mouth ( ; ; ). The risk of QT interval prolongation on EKG is potentially increased with some antidepressants. A metaanalysis examining this association found a modest but significant effect of SSRI use on QT interval prolongation, with citalopram causing more prolongation than other SSRIs ( ). The FDA recommends citalopram dosing of 40 mg/day or less to mitigate this risk ( ). The risk of QT interval prolongation was not observed in other studies examining this side effect with SSRIs, including citalopram ( ; ). A study examining the connection between QT interval prolongation and psychotropic medications found minimal evidence connecting clinically meaningful QT prolongation with most classes of psychotropics and recommended clinicians focus on other risk factors for QT prolongation when prescribing psychiatric medications ( ). Less common but possible side effects with SSRIs include, but are not limited to, decreased bone density, increased risk of bleeding, metabolic effects, liver toxicity, extrapyramidal symptoms, serotonin syndrome, tics, and myoclonus ( ; ; ; ).

The black box warning on all antidepressants regarding the risk of suicidality applies to children, adolescents, and young adults ( ). After the FDA warning was posted, a decrease in antidepressant prescriptions was seen among children and adolescents without the expected increase in psychotherapy treatments, and suicide rates increased ( ; ; ). A study using validated death data in the United States from 1990 to 2017 showed a downward trend of suicide deaths in adolescents the decade preceding the FDA warning ( ). After the FDA warning, declines were seen in both diagnosis of depression and in depression treatment. The suicide trend reversed from 4.4 suicide deaths per 100,000 adolescents in 2005 to 7.2 in 2017, a change that was seen almost immediately after the warning. The study estimates that this may have resulted in 2365 excess suicide deaths in adolescents in the United States from 2005 to 2010. While these effects cannot be directly attributed to the FDA warning, they are concerning. A metaanalysis of studies concluded that the benefits of antidepressants in the pediatric population outweigh the risks. This study showed a number needed to treat of 10 and a number needed to harm of 112 ( ). Though the TORDIA trial found a possible increased risk of treatment emergent side effects with venlafaxine compared to SSRIs, another study found no difference for risk of suicide attempts between the different antidepressants in children and adolescents ( ). Clinicians should follow dosing guidelines when starting youth on antidepressants and titrate carefully, as evidence shows there may be a connection between self-injurious behaviors in youth starting at a higher than modal dose of these medications ( ). Any child or adolescent started on an antidepressant should be followed closely by their treating physician, especially early in treatment, for response and possible adverse effects.

Psychotherapy studies for acute treatment

CBT is a treatment option clinicians should consider at any point in treatment with depressed youth. The TORDIA trial showed that combination treatment with medication and CBT was most effective for treatment-resistant depression ( ). There are not any other studies examining the efficacy of CBT in youth with treatment-resistant depression, but there is ample evidence supporting its use in acute depressive episodes. In a randomized trial examining the efficacy of CBT in 175 adolescents with major depressive disorder, CBT was shown to be superior to systematic behavior family therapy and individual nondirective supportive therapy in both response and remission rates ( ). There have been multiple studies showing similar benefits of CBT in the treatment of child and adolescent depression ( ; ; ; ). The TADS study showed no difference between the CBT group and the placebo group in acute treatment; however, long-term follow-up results showed response and remission rates both increased significantly over time in the CBT group ( ; ). Response rates went from 43% to 65% to 81% at 12, 18, and 36 weeks, respectively. Remission rates also increased with time, going from 19% (12 weeks) to 27% (18 weeks) to 64% (36 weeks). This suggests CBT monotherapy may require more time to see improvement in depressive symptoms.

Interpersonal therapy has evidence supporting its use in children and adolescents with depressive disorders. A 16-week randomized clinical trial of 63 adolescents with depression examined the efficacy of IPT modified for depressed adolescents (IPT-A) and was found to be superior to treatment as usual ( ). Several other studies have shown similar efficacy of IPT-A in the treatment of adolescent depression ( ; ; ). IPT is a good option to consider at any point in the treatment of depressed youth.

A multicenter, observer-blinded, randomized controlled comparison trial (IMPACT) examined the efficacy of CBT, short-term psychoanalytical therapy, and brief psychological intervention in 470 adolescents (aged 11–17 years) with major depressive disorder ( ). All three treatment groups showed improvement, with no significant difference on outcome measures, which included self-reported symptoms on the Mood and Feelings Questionnaire (MFQ) ( ). These results suggest even brief psychological interventions may have some benefit in depressed adolescents. Family-based interventions have been shown to be effective for depressive symptoms in adolescents. Attachment-based family therapy (ABFT) includes weekly therapy sessions focusing on the parent and child relationship. It was shown to be helpful for decreasing depressive symptoms and suicidal ideation in a randomized controlled trial of 66 adolescents ( ).

A recent metaanalysis of 55 randomized psychotherapy trials for youth aged 4–18 years with depression showed an overall modest effect size of 0.36 at posttreatment and 0.21 at follow-up (average 42 weeks after posttreatment) with psychotherapy ( ). Effects were significantly larger for IPT-A than for CBT and for youth-reported outcomes than parent-reported. A metaanalysis examining treatment effects with psychotherapy in depression across the age span showed significantly lower effect sizes in children and adolescents when compared to adults, with effect sizes of 0.35 in children, 0.55 in adolescents, and 0.98 in young adults ( ). A literature review of psychotherapy studies showed similar findings, with CBT being classified as possibly efficacious in children with depression and CBT and IPT-A being classified as well-established, effective treatments in adolescents with depression ( ).

Complementary alternative medicine

Complementary alternative treatments have been examined as possible alternatives to traditional antidepressants in children and adolescents. Studies examining the efficacy of omega-3 fatty acids in the treatment of acute pediatric depression have mixed results. A 10-week randomized controlled trial in adolescents aged 12–19 years with major depressive disorder did not find any significant difference between omega-3 fatty acids (eicosatetraenoic acid (EPA) and docosahexaenoic acid (DHA)) and placebo on any of the outcome measures ( ). A 16-week, double-blind randomized controlled study included 28 children with major depressive disorder, aged 6–12 years, randomized to either omega-3 (EPA and DHA) or placebo ( ). Response and remission rates were found to be significantly higher in the omega-3 treatment group as compared to placebo. Omega-3 fatty acids appeared to be well-tolerated in both studies and may be considered as adjunctive treatment in children and adolescents with depression, but more evidence examining their efficacy in this population is needed. There are no randomized controlled trials examining the efficacy of augmentation of antidepressants with Omega-3 fatty acids in pediatric depression or in treatment-resistant pediatric depression.

Vitamin C was studied as an adjunctive treatment to antidepressants for major depressive disorder in 24 youth aged 7–14 years in a 6-month, double-blind, placebo-controlled trial ( ). The treatment groups included fluoxetine (10–20 mg/day) plus placebo and fluoxetine (10–20 mg/day) plus vitamin C (1000 mg/day). The group receiving combined treatment with fluoxetine and vitamin C showed significant improvement when compared to the group receiving fluoxetine and placebo. It was well-tolerated with no major adverse effects reported in the study. Larger studies are needed to examine the efficacy of adjunctive vitamin C in the treatment of pediatric depression. The efficacy of vitamin D supplementation on depressive symptoms was evaluated in a randomized controlled trial of 113 youth with vitamin D deficiency and depressive symptoms ( ). No significant difference was found between the vitamin D and placebo groups on depressive symptoms at endpoint. The efficacy of zinc supplementation (10 mg of zinc oxide) for acute pediatric depression was examined in a 6-month, double-blind, placebo-controlled trial in 674 school-aged children, and was not found to be superior to placebo on any of the outcome measures ( ). Other complementary alternative treatments have been studied in small, open-label studies or in case reports but lack randomized controlled trials examining their efficacy in pediatric depression.

Combination treatment

In cases where depressed patients may not be fully responding to medication or psychotherapy alone, combination treatment with medication and psychotherapy may be considered. Studies have shown benefit with combination treatment in children and adolescents with depression. The TADS study showed combination treatment (fluoxetine plus CBT) to be most effective compared to fluoxetine alone, CBT alone, and placebo in a 12-week study ( ). A greater decrease in suicidal ideation, acceleration in treatment response, and a 1.5-fold increased probability of sustained early response to treatment was also found in the combination treatment group ( ). A recent metaanalysis of 71 trials showed that fluoxetine alone or in combination with CBT was superior to CBT alone and psychotherapy for moderate to severe depressive disorder in children and adolescents ( ). The addition of CBT may be effective in preventing relapse when added to antidepressant medication. Two sequential treatment studies included youth who responded to acute treatment with fluoxetine who were then randomized to receive continuation treatment for 6 months with fluoxetine alone or fluoxetine plus CBT. The youth who had the addition of CBT to their treatment had a significantly lower risk of relapse by the end of the studies ( ). These findings further support the recommendation to add CBT at any point in treatment to improve outcomes in depressed youth.

The ADAPT (Adolescent Depression and Psychotherapy Trial) examined the efficacy of CBT alone compared with CBT plus SSRI in adolescents who had failed to respond to brief psychotherapy ( ). Youth in both treatment groups showed response (much or very much improved on CGI-I), but the response rates did not differ significantly between the groups at weeks 12 or 28. Suicidality decreased in both treatment groups, with no significant difference between the groups, suggesting CBT may have a protective effect for depressed youth with suicidality.

A systematic review showed CBT in combination with antidepressants to be effective for depressive symptoms and associated with lower rates of relapse in children and adolescents ( ). The addition of CBT to treatment with an SSRI may increase response rates ( ; ). The benefit of CBT in improving response rates as well as preventing relapse supports the use of CBT at any phase of treatment in youth with treatment-resistant depression.

Factors associated with treatment nonresponse

If an adequate evidence-based treatment trial is not effective in treating an acute depressive episode in a child or adolescent patient, the clinician should consider possible diagnostic, medical, or psychosocial contributing factors. Revisiting diagnostic criteria may be necessary to ensure a different diagnosis, such as bipolar disorder, has not been missed is an important step ( ). Clinicians should also discuss with parents the importance of monitoring their children to ensure they are compliant with the prescribed treatments, as noncompliance can be a contributor to treatment failure. The possible effects of comorbid disorders on the depression diagnosis and treatment may need to be considered. Medical comorbidities can also affect treatment response or contribute to presenting depressive symptoms ( ; ). Psychosocial factors, such as peer relationships, bullying, abuse, or parental mental illness can affect response and should always be assessed as part of the complete evaluation when a child is in treatment. A population-based study including 1443 children followed from 5 months to 15 years showed that maternal depression in the early childhood years (prior to age 5) was associated with higher levels of adolescent depression, generalized anxiety, and social phobia symptoms ( ). Bullying from peers during childhood (ages 6–12) was a significant mediator in this study, accounting for a 35.9% association with major depression.

A study examining the connection between peer victimization and mental health disorders also examined a school-based treatment program targeting bullying ( ). This study found that youth experiencing traditional or cyberbullying victimization had higher levels of mental health problems. A nationwide school-based antibullying program was associated with a 30% decrease in traditional victimization and an increase in child-reported bullying to adults. This emphasizes the importance of school and parental involvement in decreasing peer victimization and improving mental health outcomes in youth. In another study, peer and parental support was associated with maintained improvement in depressive symptoms in children who received targeted cognitive-behavioral family interventions ( ). The study found that children with depressive symptoms 6 months after the intervention were more likely to have lower levels of peer support and facilitative parenting immediately following the intervention and increased social victimization as reported by teachers. Certain stressors in early childhood seem to increase the risk for depressive symptoms. A metaanalysis of 62 articles on early life stressors found that sexual abuse, physical abuse, death of a family member, domestic violence, and emotional abuse were associated with a significantly increased risk of a major depression diagnosis before age 18 ( ).

Genetic differences in metabolism of medications (P450 polymorphisms) may be considered when children do not respond to initial evidence-based treatments ( ). With the advances in genetic testing, pharmacogenetic testing is a topic parents may bring up with the clinician in appointments, especially in treatment-resistant cases. While these tests are becoming easier to access, it is important to discuss with parents that the tests do not predict response to treatment, and caution should be taken when interpreting these test results. The FDA recently released a statement about these pharmacogenetic tests which stated that claims of genetic tests to predict response to medications may not have evidence to support this use for most medications, including antidepressants ( ). The FDA cautioned that the relationship between genetic variations and effectiveness of antidepressant medications has not been established, and that clinicians should not make changes to a patient’s medication regimen based on the results of these tests. The American Academy of Child and Adolescent Psychiatry (AACAP) policy statement on pharmacogenetic testing recommends clinicians avoid using testing results to select treatments for children and adolescents due to lack of evidence supporting its use in this population ( ). The statement notes that many factors affect treatment response in children, and genetic variations are best addressed by the clinician with careful medication management and monitoring.

Conclusion

Depression is a common psychiatric diagnosis in youth, with lifetime prevalence in adolescence similar to that seen in adults. It can lead to a wide array of impairment in social, academic, and family functioning. A diagnosis of depression increases the risk of suicide, the second leading cause of death in patients aged 10–34, highlighting the importance of finding effective treatments for youth with depression. Using evidence-based treatment in the acute management of depression in children and adolescents may lower the risk of the development of treatment-resistant depression. Based on consensus guidelines and the evidence-base, SSRIs are recommended as initial medication treatment for children and adolescents with major depressive disorder. If there is no treatment response after an adequate dose and duration of an SSRI, switch to another evidence-based SSRI is indicated. If response is still inadequate, switching to a non-SSRI (SNRIs, for example) is recommended. Augmentation with a second medication or combination treatment with two antidepressant medications may be needed if there is a partial response to the antidepressant, though there is limited evidence in pediatric depression. Psychotherapy, particularly CBT or IPT-A, may be considered at any time in treatment to improve response, and there is evidence suggesting that CBT prevents relapse in youth with acute response to treatment. Other somatic treatments, such as rTMS and ECT have been used for treatment of youth with major depression who have failed to respond to medications and psychotherapy. Ketamine is currently being studied for treatment-resistant depression in youth. Though the body of research in pediatric depression treatment is growing, the evidence examining effective treatments for treatment-resistant depression in youth remains limited, and more studies are needed to help guide clinicians in managing these challenging cases.