OS02.6 risk factors for hypersomnia and narcolepsy and response to therapy in survivors of childhood brain tumors

Abstract

INTRODUCTION: According to international headache society guidelines, narcolepsy is diagnosed when mean sleep onset latency on multiple sleep latency test (MSLT) is <8 minutes and two or more sleep onset REM sleep episodes are present. Hypersomnia has similar criteria except only one or less sleep onset REM episodes are present on MSLT. A negative overnight polysomnogram is also required the night before the MSLT. Daytime sleepiness is recognized in brain tumor survivors but few studies have reported results of overnight polysomnogram and MSLT, risk factors for development of hypersomnia and narcolepsy, or response to therapy. Our objectives for this study were to understand prevalence of hypersomnia and narcolepsy in childhood brain tumor patients, describe the risk factors for its development, and response to stimulant therapy. MATERIALS AND METHODS: After obtaining institutional review board approval, medical records were reviewed of 156 brain tumor survivors coded as having sleep disorder, hypersomnia or narcolepsy. Hypersomnia or narcolepsy was diagnosed according to the international headache society classification based on overnight polysomnogram and MSLT. Each affected survivor was matched by gender, and age and time from tumor diagnosis to 2-3 survivors without reported daytime sleepiness. Study variables included tumor location, radiation dose, number of surgeries, ventricular shunt, and body mass index. Response to pharmacologic therapy was assessed via parent report. RESULTS: The 39 survivors with hypersomnia (n=13) and narcolepsy (n=26) were matched with 110 controls. A total of 2336 brain tumor patients were treated during the study period, translating to an estimated hypersomnia and narcolepsy prevalence of 1670/100,000. Median time from tumor diagnosis to hypersomnia/narcolepsy was 6.1 years (range 0.4-13.2), and 4.7 years (range-1.5-10.4) from radiation. Of the many variables studied, only midline tumor location (OR 5.3, CI 2.2-12.7, p = <0.001) and radiation dose of >30 Gray (p = <0.01) correlated with hypersomnia and narcolepsy, while posterior fossa tumor location was protective (OR 0.1, CI 0.03-0.35, p=<0.001). Most of 37 survivors treated with stimulants showed improvement in wakefulness and school performance (response rate CI 0.97 [0.86-0.99] and 0.83 [0.65-0.94]); two were treated with strategic naps. CONCLUSION: Hypersomnia and narcolepsy prevalence of 1670/100,000 among childhood brain tumor survivors is much higher than 20-50/100,000 reported in general population. Midline tumor location and >30 gray radiation to the brain strongly correlated with hypersomnia and narcolepsy, while posterior fossa tumor location may be protective. Stimulant therapy is beneficial and is generally well tolerated in this population.