La pharmacologie dans les troubles bipolaires chez l’enfant et l’adolescent bénéficie d’un intérêt croissant mais peu d’études concernent le lithium. L’efficacité du lithium en monothérapie est modérée, avec des taux de réponse de l’ordre de 35 à 63 %. Le bénéfice des combinaisons thérapeutiques (association au divalproate de sodium, à la carbamazépine ou à la rispéridone) est net avec des taux de réponse atteignant 70 à 90 %. Le taux de rechute sous lithium en monothérapie est élevé (37 à 56 %) mais peu étudié. Si le profil de tolérance du lithium est rassurant, les effets secondaires rapportés semblent plus fréquents que chez l’adulte, et l’on ne dispose d’aucune donnée à long terme. Enfin, la lisibilité de tous ces résultats souffre d’une limite majeure liée à l’hétérogénéité clinique avec l’usage extensif du diagnostic de bipolarité chez l’enfant prépubère. Les études futures bénéficieront du cadre diagnostique distinguant les diagnostics de « Dysrégulation Émotionnelle Sévère » (ou « Temper Dysregulation Disorder with Dysphoria » dans le DSM-5) chez l’enfant et de trouble bipolaire de type I chez l’adolescent.

Bipolar disorder in youths is a severe psychiatric disorder that potentially results in high-relapse rates and disrupts the normal psychosocial development. Many adults with bipolar disorder identify the onset of their symptoms in childhood and adolescence, indicating the importance of early accurate diagnosis and treatment. The lifetime prevalence is estimated in adolescence at 0.1% for bipolar I disorder and 1% for all bipolar spectrum disorders. “Classic” bipolar disorder with well-delineated episodes is rare among prepubertal children. The treatment of bipolar disorder in youths raises challenges and difficulties. Lithium is recommended as one of the first-line treatment options of bipolar disorder in children and adolescents with FDA approval as early as 12years. Lithium prescription is approved in France for bipolar disorder only for adolescents older than 16years old. This comprehensive literature review incorporates research studies evaluating the effectiveness of lithium in children and adolescents with pediatric bipolar disorder (randomized controlled trials, open-label studies and retrospective chart reviews). Lithium was evaluated as a monotherapy, as well as part of a combination treatment, in the acute and maintenance phases. The effectiveness of a lithium monotherapy is moderate with response rates of approximately 35 to 63%. Its efficacy is similar to valproate and carbamazepine; however, lithium appears to be superior in case of psychotic symptoms. A recent study reported the superiority of risperidone over lithium in bipolar mania with a comorbid attention-deficit/hyperactivity disorder. While it is ideal to use the lowest possible dose of monotherapy medication to decrease adverse side effects, most patients require adjunctive medication treatment for the stabilization of bipolar symptoms. Data suggest that a combined treatment with lithium (using valproate, carbamazepine or risperidone) may provide a better remission of symptoms, with response rates between up to 70 and 90%. The relapse rate in lithium monotherapy is high (37 to 56%); nevertheless, very few studies have measured the long-term effects of the medication in pediatric bipolar disorder. The data from the available pharmacological studies need to be interpreted cautiously because of the controversy surrounding the definition and diagnosis of bipolar disorder in youths. The continuity between childhood-onset bipolar disorder and the current adult-oriented DSM-IV criteria in adolescence is currently debated. The diagnostic framework has evolved to distinguish the diagnoses of “Severe Emotional Dysregulation” (or “Temper Dysregulation Disorder with Dysphoria” in DSM-V) in children from bipolar I disorder in adolescents. Research samples often include both of these clinical pictures. This heterogeneity in the patient population introduces a major caveat towards the interpretation of the available literature addressing treatment strategies. The distinction between these two forms is even more relevant given the fact that preliminary studies have shown various clinical courses, family psychiatric histories, different prognoses, as well as therapeutic responses. Even if the safety profile of lithium is reassuring, the reported side effects seem to be a more frequent in youths than in adults, especially if the child is young. The most commonly reported (>20%) adverse events were nausea/vomiting, diarrhea, abdominal pain, headache, dizziness, tremor and pollakiuria. Weight gain and acne are side effects that are perceived to be particularly shameful during adolescence. No data on long-term side effects are available. The limited studies examining the pharmacokinetics of lithium may indicate the need for vigilance in the variability induced by the weight in pediatric patients. Very few studies have examined predictors and moderators in the treatment of bipolar disorder in children and adolescents. Being younger or suffering from comorbid attention-deficit/hyperactivity disorder have been associated with a poorer response to lithium. The aims of future researches in bipolar disorder in youths may establish some evidence-based strategies for lithium especially: (i) by examining the acute efficacy in mania and also in bipolar depression with a particular attention to age-specific aspects and diagnosis at inclusion, (ii) by investigating the long-term effectiveness of lithium treatment, (iii) by characterizing the long-term safety of lithium, (iv) by identifying pretreatment variables associated with the effectiveness and tolerability of lithium, (v) by studying adherence concerns in adolescents.