A Study of the Diagnostic Utility of HLA Typing, CSF Hypocretin-1 Measurements, and MSLT Testing for the Diagnosis of Narcolepsy in 163 Korean Patients With Unexplained Excessive Daytime Sleepiness

Abstract

To study DQB1*0602 status and hypocretin-1 levels in the cerebrospinal fluid (CSF) in a cohort of patients with hypersomnolence and to test International Classification of Sleep Disorders-2 (ICSD-2) criteria for hypersomnia of central origin. Retrospective case series. One hundred sixty-three consecutive patients with unexplained sleepiness and 282 controls recruited at St. Vincent's Hospital, Korea. The gold standard for diagnosis was ICSD-2 criteria. Patients and controls completed the Stanford Sleep Inventory, and agreed to HLA typing. Polysomnography (87%), Multiple Sleep Latency Test (MSLT) (96%), and CSF hypocretin-1 measurements (53%) were conducted in patients. Most patients (80%) could be classified using the ICSD-2. The 33 patients who could not be classified were without cataplexy (4 with low CSF hypocretin-1). These could not be included because of sleep apnea (apnea-hypopnea index ≥ 5/h, 84%) and/or because sleep prior to MSLT was less than 6 hours (27%). Narcolepsy with cataplexy cases were 92% HLA positive with low hypocretin-1. Cataplexy at interview was predicted by validated Stanford Sleep Inventory questions regarding cataplexy triggers. In contrast, cataplexy-like events were frequently reported in all groups, including controls. Cases with narcolepsy without cataplexy were frequently men (73%) and heterogeneous biologically (36% HLA positive, 40% with low CSF hypocretin-1). None of the controls had low CSF hypocretin-1, whereas 13% were HLA positive. The ICSD-2 was easily applicable in cases with typical cata-plexy. In these cases, the MSLT and further evaluations were almost always positive and may thus not always be needed. Many patients without cataplexy were difficult to classify because of difficulties in interpreting the MSLT in the presence of sleep apnea or reduced sleep.