Shortening Kleine-Levin Syndrome Episodes With Oral Steroids: Insights From Two Cases

Noor Hajjaj; Danielle Bradshaw; Ty Carter; Mac Manuel; Erin Hauck; Dwayne Henry; Yash Shah

doi:10.13078/jsm.260002

Abstract

Kleine-Levin syndrome (KLS) is a rare sleep disorder marked by recurrent episodes of hypersomnia, cognitive impairment, and behavioral disturbances. Intravenous corticosteroids have been reported to shorten episodes, but evidence for oral steroid therapy remains limited. We describe two patients with KLS whose episodes improved with oral prednisone. Case 1 was a 17-year-old male patient with recurrent hypersomnia and behavioral changes. Steroid-responsive encephalopathy associated with autoimmune thyroiditis was initially considered but later excluded. He demonstrated improvement with intravenous steroids, and subsequent episodes were managed with oral prednisone. Case 2 was a 10-year-old male patient with early-onset KLS who experienced shortened episode duration after oral prednisone following failure of carbamazepine therapy. These cases suggest that oral prednisone may represent an effective alternative to intravenous corticosteroids for reducing episode duration in KLS. Oral therapy provides several advantages, including patient convenience and the feasibility of outpatient care. Therefore, oral steroid therapy warrants further investigation in KLS.

Keywords: Kleine-Levin syndrome, Methylprednisolone, Adolescent

INTRODUCTION

Kleine-Levin syndrome (KLS), also known as recurrent hypersomnia or periodic hypersomnolence, is a rare neurological sleep disorder characterized by recurrent episodes of excessive sleep, cognitive impairment, and behavioral disturbances. These episodes typically last from days to weeks and are often accompanied by confusion, derealization, apathy, hyperphagia, and hypersexuality. KLS predominantly affects adolescents, with a mean age of onset of approximately 15 years, while cases in younger children are exceedingly rare [1].

Due to its episodic nature and overlapping features with other neuropsychiatric conditions, KLS is often challenging to diagnose. Although intravenous methylprednisolone (IV-MP) has been reported to reduce the duration of KLS episodes, its use is associated with hospitalization and increased healthcare costs [2]. There is limited research on the role of oral corticosteroids as a practical alternative to IV therapy. IV-MP has been associated with reduced episode duration in observational studies; however, evidence remains limited and largely non-randomized.

This case series aims to highlight the potential utility of oral corticosteroids in reducing episode duration in patients with suspected KLS, particularly in younger patients and in settings where IV therapy may be impractical. By documenting these cases, this case series aims to contribute to the growing literature on KLS management and explore oral corticosteroid therapy as a feasible and accessible treatment option.

CASE REPORTS

Case 1

A 17-year-old male patient presented with recurrent episodes of excessive somnolence and cognitive disturbances. The patient experienced three episodes following mild illness with the SARS-CoV-2 virus or stress-inducing events such as extended travel. During these episodes, the patient exhibited hypersomnia, sleeping approximately 20 hours per day, and reported experiencing cognitive delay. The patient described significant periods of delayed cognitive recognition, such as a delayed perception of postural changes. Additionally, the patient displayed intermittent regressive behavior, including infantile speech patterns and behavior consistent with that of a four-year-old child. Furthermore, the patient exhibited hypersexual activity during his initial admission.

During the first episode, a lumbar puncture and autoimmune panel were obtained due to concerns of autoimmune encephalitis; the results of both were unremarkable. The electroencephalogram (EEG) was consistent with mild nonspecific generalized cerebral dysfunction. No epileptiform abnormalities or seizures were recorded. The patient received intravenous immunoglobulin (IVIG) treatment, resulting in mild improvement. At that time, brain magnetic resonance imaging (MRI) and computed tomography (CT) were unremarkable. During the second episode, an MRI was obtained with a normal result; however, further workup revealed positive thyroperoxidase antibodies (TPOAb; 16.5 IU/ML), leading to a diagnosis of steroid-responsive encephalopathy associated with autoimmune thyroiditis (SREAT). The patient was initiated on IV-MP therapy (1 g/day for 3 days), resulting in complete resolution of symptoms after two days. On the third presentation, the patient presented similarly to previous episodes. TPOAb was within normal limits with normal thyroid function test, prompting consideration of alternative diagnoses. The patient was started on IV-MP therapy (1 g/day for 3 days), given a previous response; however, he did not respond as quickly as in the second presentation, with symptom resolution after 7 days.

The patient demonstrated complete return to baseline between episodes, fulfilling a key KLS diagnostic criterion. Transient elevation of TPOAb was not sustained in subsequent episodes, supporting KLS over SREAT. The clinical discussion addressed the possibility of KLS, given the compelling symptomatology related to excessive sleep and altered cognitive states. Additionally, the patient’s sibling was diagnosed with narcolepsy following the patient’s first two episodes, suggesting a potential genetic predisposition that supports the diagnosis of a sleep disorder.

A few months later, the patient experienced another episode with the same clinical presentation as previous episodes. The symptoms persisted for five days without improvement. At that point, oral prednisone was initiated at a dose of 1 g/day for three days with no tapering. The patient showed significant improvement and returned to baseline after the second dose of oral steroids.

During a subsequent episode, the patient exhibited similar symptoms, though less severe than before. This time, a lower dose of oral prednisone (500 mg/day) was administered for three days. Improvement was observed, but the higher dose produced more rapid resolution, indicating a dose-dependent effect. Lithium therapy was considered but not initiated due to family concerns regarding potential side effects.

Case 2

A 10-year-old male patient with a history of attention-deficit/hyperactivity disorder (ADHD) presented with three episodes of hypersomnia and abnormal behavior. During his initial episode at the age of nine, following a cruise trip, he exhibited hypersomnia, hyperphagia, and confusion. Laboratory workup ruled out infection and thyroid abnormalities. MRI results were normal. The patient was taking stimulants and clonidine; however, stimulants were discontinued during episodes, and clonidine (0.1 mg) was stopped without improvement, confirming these medications did not account for hypersomnia. The episode persisted for 27 days. The patient was diagnosed with stress-related sleep disturbances. Subsequently, he returned to his baseline level of function.

One month later, the patient exhibited similar symptoms, besides inappropriate physical contact with his mother and profane language use. This episode lasted for 27 days. Carbamazepine was initiated based on a clinical suspicion of KLS.

Three months later, prior to the start of the academic year, he experienced a similar episode but without hypersexuality. He displayed regressive behavior, including difficulty with self-care tasks and disorientation. No significant improvement was observed with carbamazepine. This episode persisted for 29 days.

Right-sided migraine headaches with photophobia and vomiting accompanied all episodes. Although persistent headaches with vomiting are atypical for KLS, these symptoms occurred consistently during episodes and are considered part of the clinical presentation rather than a separate primary headache disorder. The patient reported no recollection of events during these episodes but acknowledged increased food consumption.

EEG was not performed during symptomatic episodes. The decision was made to start lithium therapy as the patient met the International Classification of Sleep Disorders, Third Edition (ICSD-3) criteria for KLS. Lithium was not initiated due to parental concerns, so no clinical response to lithium was available.

Several months later, the patient experienced another episode with a similar clinical presentation. At the time, the patient was not receiving lithium, as the family was hesitant to initiate treatment due to concerns about potential side effects. Oral prednisone was administered at a dose of 1 g/day for three days, starting on day 10 of symptoms. Oral corticosteroids were selected over IV therapy to facilitate outpatient management and accommodate family preference. The patient demonstrated complete recovery following the third dose of oral steroids. No tapering regimen was implemented, given the short duration of therapy and rapid clinical response.

DISCUSSION

KLS, also known as recurrent hypersomnia or periodic hypersomnolence, is a rare sleep disorder marked by recurrent episodes of intense hypersomnia. These episodes are often accompanied by cognitive and behavioral disturbances such as confusion, derealization, apathy, hyperphagia, and hypersexuality. The similarities in cognitive and behavioral manifestations between SREAT and KLS can complicate the diagnostic process, as in the first case. A critical distinguishing feature of SREAT is the elevation of serum antithyroid antibodies, particularly TPOAb and antithyroglobulin antibody (TgAb). However, the presence of these antibodies alone does not confirm a diagnosis, as they can be found in 2%–20% of the healthy population; a compatible clinical presentation is also necessary for diagnosis. In our 17-year-old male patient, the presentation of hypersomnia, hyperphagia, and hypersexuality, along with initially positive TPOAb, made the diagnosis challenging. The patient’s responsiveness to steroids further complicated the clinical picture, as steroid responsiveness is seen in the majority of SREAT cases.

There are very limited case reports on KLS in individuals <12 years old. In an analysis of 475 case reports published in 2024, only 23 patients presented with onset of symptoms at <10 years old [3], accounting for 4.8% of cases, which is extremely rare. Moreover, clinical history, including episodes of hypersomnia and atypical behavior, underscores the complexity of diagnosing KLS in younger populations, as seen in our 10-year-old patient, where the typical presentation may be obscured by coexisting conditions such as ADHD.

KLS is diagnosed clinically with no confirmatory test. KLS etiology is still unknown, but it has been suggested that genetic and autoimmune factors may play a role [4]. A multiplex family with six affected members, and the increased prevalence of KLS in the Ashkenazi Jewish population, provides further support to the former. Besides genetic and immunemediated theories, there has been evidence suggesting infectious processes as precipitating factors of KLS episodes, without the identification of a single specific infection. Vaccination, substance use, and other stress-associated processes have also been reported in association with the precipitation of KLS episodes [5]. In one case report, a 19-year-old girl experienced her first episode of hypersomnia following alcohol intoxication; she reportedly complained of cognitive slowing, visual disturbances, and amnestic periods [6].

Although KLS is considered a central hypersomnolence disorder, the connection between KLS and specific human leukocyte antigen (HLA) types remains understudied in the literature. Unlike narcolepsy, which is strongly associated with genetic factors and specific HLA alleles, the role of HLA in KLS is not well defined. Whether HLA plays a direct causal role in KLS development or if it serves merely as a marker of susceptibility remains unclear. Further research is needed to determine whether HLA is a causative factor or if it only contributes to the syndrome onset.

To our knowledge, there is no definitive treatment for KLS. Lithium, antiepileptics, antidepressants, and antipsychotics have been studied, with lithium demonstrating efficacy in reducing the duration of episodes, improving symptoms, and decreasing relapse rates [7]. Additionally, the use of IV steroids during the first 10 days of episodic symptom onset has been shown to decrease the duration of the episode course [2].

The mechanism by which corticosteroids may shorten KLS episodes remains uncertain. Episodes are frequently triggered by infections or stress, suggesting a transient immune-mediated component. Corticosteroids may reduce dysfunction in thalamic and hypothalamic networks involved in sleep–wake regulation through anti-inflammatory and immunomodulatory effects. Dysregulation of gamma-aminobutyric acid (GABAergic) and glutamatergic neurotransmission has also been implicated in KLS and may be indirectly modulated by steroid treatment. Supporting the role of neuroinflammation, cases with positive anti-NMDA receptor antibodies have been reported during acute exacerbations. These observations provide a biologically plausible explanation for the temporal association between corticosteroid administration and symptom improvement, although causality cannot be definitively established.

In a case report of a KLS episode precipitated by excess exposure to a GABA-metabolite, it was hypothesized that KLS pathophysiology is directly related to an imbalance in the GABA and glutamate neurotransmitters [8]. It was further suggested that IV steroids limit KLS episode duration by re-establishing GABAergic/glutamatergic homeostasis. In another case of KLS, a 14-year-old boy was found to have positive anti-NMDA type glutamate receptor antibodies in his cerebrospinal fluid9; further supporting the theory that neurotransmitter imbalances play a key role in KLS. This proposed mechanism would suggest new therapeutic considerations and ultimately requires further research efforts aimed at KLS pathophysiology.

The present cases differ from prior reports in several important ways. Most existing literature has focused on IV-MP, whereas our cases highlight the potential utility of high-dose oral corticosteroids as an alternative. Additionally, one of our patients represents early-onset KLS, which is rarely described in the literature. Finally, a possible dose–response relationship was observed in one case, which has not been well characterized in prior reports.

A study performed by Léotard et al. [2], involving 26 patients with KLS, examined the effects of IV-MP (1 g/day for 3 days) across 43 sessions. The duration of episodes treated with IVMP was compared to that of untreated patients matched by age, sex, and disease characteristics. Results showed that 42.3% of treated patients had episodes at least one week shorter than their previous ones, while only 10.4% of untreated patients saw such a reduction. The benefit was more pronounced when IV-MP was given within the first 10 days of an episode, with 65.5% of patients responding positively. The study concluded that early IV-MP treatment had a favorable benefit/risk ratio, particularly for patients with long episodes, as it led to earlier cessation in half of the cases (Table 1) [2].

There is very limited research on the use of oral steroids in cases of KLS. In one documented case, a 10-year-old boy received pulsed doses of MP, followed by a 6-week course of oral steroid therapy with symptom improvement, followed by a 4-week asymptomatic period [10]. Details regarding dosing, timing, and comparative episode duration were limited. This case demonstrates the need for further study into the use of oral steroids in management and relapse prevention in pediatric patients (Table 1).

Both patients met the diagnostic criteria for KLS as defined by the ICSD-3. These criteria include: 1) recurrent episodes of hypersomnia lasting from 2 days to several weeks, 2) normal or near-normal functioning between episodes, 3) cognitive dysfunction, derealization, or behavioral disturbances during episodes, and 4) exclusion of other sleep, neurological, or psychiatric disorders. In both cases, patients demonstrated recurrent hypersomnia, associated neurobehavioral symptoms, and complete return to baseline between episodes, with no alternative diagnosis identified after appropriate evaluation.

Both our cases demonstrate the potential benefit of oral prednisone in reducing episode duration in KLS. Prior studies have primarily focused on IV-MP, with one study reporting that early administration of IV steroids resulted in a significant reduction in episode length. However, IV therapy presents logistical and financial challenges, making oral steroids a more practical alternative.

The mechanism by which corticosteroids impact KLS remains unclear, but their anti-inflammatory and immunomodulatory effects may play a role in mitigating episodic hypersomnia and neuropsychiatric symptoms. While these findings suggest oral prednisone is effective in shortening episodes, further research is needed to establish standardized protocols and dosing regimens with respect to long-term efficacy.

A key distinguishing feature of KLS is the presence of recurrent episodes with a complete return to baseline between episodes, which was observed in both patients described in this report.

In Case 1, the transient elevation of antithyroid antibodies, absence of persistent encephalopathy, and normalization of laboratory findings in subsequent episodes support an episodic disorder rather than a progressive autoimmune encephalopathy such as SREAT.

Although rare, early-onset cases of KLS have been reported. The presence of hypersomnia, behavioral changes, and recurrent episodes in this patient supports a KLS-like presentation despite the atypical age of onset.

Memory impairment is commonly reported in KLS; however, the degree of recall may vary, and marked impairment of recall represents a potential area of diagnostic uncertainty.

Although lithium has demonstrated benefit in some patients with KLS, it was not initiated in these cases, limiting assessment of its therapeutic effect.

While corticosteroids may reduce episode duration, their use in KLS remains investigational, and the observed improvement may reflect the natural course of the disorder.

Limitations

This study has some limitations, including its small sample size and observational design. Diagnostic uncertainty remains, particularly in distinguishing KLS from other neurological and psychiatric conditions. In Case 2, an EEG was not performed during symptomatic periods, limiting the exclusion of alternative diagnoses such as epileptic phenomena. Additionally, the observed response to corticosteroids may partly reflect the natural course of the disorder rather than a definitive treatment effect.

Conclusion

This case series highlights that oral corticosteroids, specifically prednisone, may reduce episode duration and symptom severity in KLS and could represent a practical alternative to IV therapy, with advantages in accessibility and reduced need for hospitalization. Early initiation may also contribute to improved outcomes.

However, evidence for oral steroid use in KLS remains limited, and optimal dosing, timing, and long-term efficacy are not yet established. Further larger case series and controlled studies are needed before standardized treatment recommendations can be made, and corticosteroid use should currently be considered with caution and individualized to the patient.