Lithium treatment of Kleine-Levin syndrome: An advance for a disorder of hypersomnolence.

In 1925, Willi Kleine described a series of 5 patients with “periodic sleepiness” of diverse etiology. In 1936, Levin published additional cases with “a syndrome of periodic somnolence and morbid hunger.” The diagnosis of Kleine-Levin syndrome (KLS), according to the International Classification of Sleep Disorders–3, requires (1) recurring episodes, usually more than once a year and at least once every 18 months; (2) normal alertness, cognitive function, behavior, and mood between episodes; and (3) at least one of the following during episodes: cognitive dysfunction, altered perception, eating disorder (anorexia or hyperphagia), and disinhibited behavior (such as hypersexuality). Recent evidence suggests that KLS can be a lifelong disease. Early manifestation, combined with hypersexuality during symptomatic phases, seems to be a predictor for a long course of the disease, which can last from 1 to 27 years. Due to the lack of biological markers, diagnosis at first manifestation (mainly in the mid-20s) is difficult. KLS is a rare disease, with an estimated 1–5 cases/ million population, which might therefore be underdiagnosed. Due to its scarcity, it has rarely been studied systematically, and pathophysiologic insights and treatment recommendations are based on reports of few patients. Arnulf et al. have compiled the largest group of KLS cases worldwide. Their analysis has contributed greatly to the understanding of the symptomatology, course, and treatment of the disease. In this issue of Neurology®, Leu-Semenescu et al. report on lithium therapy in KLS. This single-center, open-label, prospective study is of immense importance, as data on therapy efficacy in patients with KLS is uncommon and mainly includes small samples. So far, case reports on KLS treatments have been published on stimulants (methylphenidate, modafinil, pemoline, piracetam, ephedrine) and neuroleptics (haloperidol, chlorpromazine, levomepromazine, trifluoperazine, thioridazine, risperidone) during the symptomatic episodes, and antidepressants (imipramine, monoamine oxidase inhibitors, moclobemide, fluvoxamine, sertraline, trazodone), antiepileptic medication (valproic acid, carbamazepine), and other medications (lithium, hydrocortisone, melatonin, benzodiazepines, levodopa-benserazide, acetazolamide, clarithromycin) for the prevention of episodes. A recently republished Cochrane analysis on the treatment of KLS could not identify one single study among 31 that met the evidence-based selection criteria of randomized controlled studies. However, evaluation of the studies showed that amphetamines improved sleepiness, but no other symptoms. Antidepressants had no effect on preventing relapses and antiepileptic medication improved abnormal behavior in only one case (carbamazepine). By contrast, lithium improved abnormal behavior and reduced duration of episodes and relapses. In an earlier study, Arnulf et al. interviewed 108 patients with KLS by questionnaire, 30 of whom had been treated with lithium, and found recovery in 7% and decreased episode frequency in 17%. In contrast to the present study, these results reflected the opinions of the patients, and no information was given on the dose, duration of treatment, or serum levels of lithium. The present study compares patients with KLS treated with lithium (n 5 71) with untreated patients (n5 59) in a prospective study over up to 49 months. The long-term comparison of the 2 groups provides excellent additional insights into the natural course of KLS. This is the first study ever to document length and duration of episodes by having the patients report every new episode by telephone. Notably, 36.6% of patients with KLS in the lithium-treated group were free of episodes, compared to 3.4% in the untreated group. The other 63.4% of patients had a mean decrease of 2.6 episodes per year and 7.6 days per episode on average, decreases of 18.1 days for the longest episode and of 36.9 days spent incapacitated per year of treatment. It is interesting that the authors documented mini-episodes of 24 hours under lithium treatment, which has never been reported before. Besides the effect of lithium on the course of KLS, this article has practical implications, showing that lithium plasma levels .0.8 mmol/L seem to be important for sustained efficacy, since lower levels resulted in relapses. However, this needs to be confirmed bymore data. What makes lithium such a powerful medication in KLS? Lithium has been used for more than 60