Is there a role for repeating the multiple sleep latency test across childhood when initially non-diagnostic?

Background: The gold standard investigation for central disorders of hypersomnolence is the Multiple Sleep Latency Test (MSLT). As the clinical features of these disorders of hypersomnolence evolve with time in children, clinicians may consider repeating a previously non-diagnostic MSLT. Currently there are no guidelines available regards the utility and timing of repeating paediatric MSLTs. Methods: Retrospective review of children aged 3-18years with >= 2MSLTs between 2005 and 2022. Narcolepsy was defined as mean sleep latency (MSL) <8min with >= 2 sleep onset REM (SOREM); idiopathic hypersomnia (IH) was defined as MSL <8min with <2 SOREM. MSLTs not meeting these criteria were labelled non-diagnostic. Results: 19 children (9 female) with initial non-diagnostic MSLT underwent repeat MSLT, with 6 proceeding to a 3rd MSLT following 2 non-diagnostic MSLTs. The 2nd MSLT resulted in diagnosis in 6/19 (32 %) (3 narcolepsy, 3 IH); and 2/6 (33 %) 3rd MSLT were diagnostic (2 IH). Median age at initial MSLT was 7.5y (range 3.4-17.8y), with repeat performed after median of 2.9y (range 0.9-8.2y), and 3rd after a further 1.9 years (range 1.2-4.2y). Mean change in MSL on repeat testing was -2min (range -15.5min to +4.9min, p = 0.18). Of the 8 diagnostic repeat MSLTs, in addition to the MSL falling below 8 min, 2 children also developed >= 2 SOREM that had not been previously present. Conclusions: A third of repeat MSLTs became diagnostic, suggesting repeat MSLT should be considered in childhood if clinical suspicion persists. Further work needs to address the ideal interval between MSLTs and diagnostic cut-points specific to the paediatric population.