0691 Long PSG-MSLT Improves Diagnostic Accuracy for Severe Idiopathic Hypersomnia Symptoms Across Multi-center Cohort

Abstract Introduction Diagnosing Idiopathic Hypersomnia (IH) remains challenging due to the limited feasibility and accuracy of objective tests specified in the ICSD-3TR. In the literature, the multiple sleep latency test (MSLT), using a cutoff of mean sleep latency (MSL) ≤ 8 minutes, demonstrates poor validity and reliability. Actigraphy-based determination of total sleep time (TST) ≥ 11 hours is hindered by inaccurate detection of daytime sleep. Twenty-four-hour polysomnography (PSG) can more precisely assess sleep duration but is rarely feasible in routine clinical settings. To address these limitations, we evaluated a modified “long PSG-MSLT,” hypothesizing that aggregated TST across the extended protocol would more accurately identify individuals with severe IH symptoms than MSL alone. Methods The long PSG-MSLT was implemented for clinical IH evaluation across five U.S. sleep centers. Patients aged 15–45 years completed an extended overnight PSG (approximately 9–10 PM to 10 AM) followed by a four-nap MSLT, with the final nap allowing up to 60 minutes of sleep or natural awakening. Diagnostic accuracy for detecting severe IH symptoms—defined as Idiopathic Hypersomnia Severity Scale (IHSS) ≥ 26—was assessed using receiver-operating characteristic (ROC) analysis. Results A total of 112 individuals (mean age 24.7 ± 7.4 years; 62.2% female) completed testing; 84% scored IHSS ≥ 26. MSL demonstrated non-significant diagnostic accuracy (AUC = 0.66; 95% CI 0.53–0.79; p = 0.06), while aggregated TST showed significant accuracy (AUC = 0.79; 95% CI 0.69–0.90; p < 0.0005). An MSL ≤ 8 minutes showed poor sensitivity (23.2%) but high specificity (92.9%) and did not differentiate IHSS severity (Fisher’s exact p = 0.29). Combining MSL ≤ 8 minutes and/or total sleep time ≥ 11 hours metrics improved sensitivity to 51% while maintaining high specificity (92.9%) and significantly distinguished IHSS severity groups (Fisher’s exact p = 0.002). Conclusion The ICSD-3TR criterion of MSL ≤ 8 minutes lacks diagnostic accuracy for identifying individuals with severe IH symptoms and fails to differentiate symptom severity. The modified long PSG-MSLT provides a clinically feasible and more accurate diagnostic approach by capturing increased sleep need—a core feature of many people with IH. Support (if any) NIH NINDS 5R61NS130215-02, PI Kiran Maski MD MPH