Validation of Multiple Sleep Latency Test for the diagnosis of pediatric narcolepsy type 1

Objective To validate polysomnographic markers (sleep latency and sleep-onset REM periods [SOREMPs] at the Multiple Sleep Latency Test [MSLT] and nocturnal polysomnography [PSG]) for pediatric narcolepsy type 1 (NT1) against CSF hypocretin-1 (hcrt-1) deficiency and presence of cataplexy, as no criteria are currently validated in children. Methods Clinical, neurophysiologic, and, when available, biological data (HLA-DQB1*06:02 positivity, CSF hcrt-1 levels) of 357 consecutive children below 18 years of age evaluated for suspected narcolepsy were collected. Best MSLT cutoffs were obtained by receiver operating characteristic (ROC) curve analysis by contrasting among patients with available CSF hcrt-1 assay (n = 228) with vs without CSF hcrt-1 deficiency, and further validated in patients without available CSF hcrt-1 against cataplexy (n = 129). Results Patients with CSF hcrt-1 deficiency were best recognized using a mean MSLT sleep latency <= 8.2 minutes (area under the ROC curve of 0.985), or by at least 2 SOREMPs at the MSLT (area under the ROC curve of 0.975), or the combined PSG + MSLT (area under the ROC curve of 0.977). Although specificity and sensitivity of reference MSLT sleep latency <= 8 minutes and >= 2 SOREMPs (nocturnal SOREMP included) was 100% and 94.87%, the combination of MSLT sleep latency and SOREMP counts did not improve diagnostic accuracy. Age or sex also did not significantly influence these results in our pediatric population. Conclusions At least 2 SOREMPs or a mean sleep latency <= 8.2 minutes at the MSLT are valid and reliable markers for pediatric NT1 diagnosis, a result contrasting with adult NT1 criteria. Classification of evidence This study provides Class III evidence that for children with suspected narcolepsy, polysomnographic and MSLT markers accurately identify those with narcolepsy type 1.